Content uploaded by FC Prinsloo
Author content
All content in this area was uploaded by FC Prinsloo on May 19, 2023
Content may be subject to copyright.
A geo-informatics approach to sustainability assessments of floatovoltaic
technology in South African agricultural applications
by
FREDERIK CHRISTOFFEL PRINSLOO
DOCTOR OF PHILOSOPHY
in
GEOGRAPHY
at the
COLLEGE OF AGRICULTURE AND ENVIRONMENTAL SCIENCES
UNIVERSITY OF SOUTH AFRICA
SUPERVISOR: PROF P M U SCHMITZ
CO-SUPERVISOR: MISS A LOMBARD
2023
Abstract
Abstract
South African project engineers recently pioneered the first agricultural floating solar photovoltaic tech-
nology systems in the Western Cape wine region. This effort prepared our country for an imminent large-
scale diffusion of this exciting new climate solver technology. However, hydro-embedded photovoltaic sys-
tems interact with environmentally sensitive underlying aquatic ecosystems, causing multiple project as-
sessment uncertainties (energy, land, air, water) compared to ground-mounted photovoltaics. The dissimi-
lar behaviour of floatovoltaic technologies delivers a broader and more diversified range of technical advan-
tages, environmental offset benefits, and economic co-benefits, causing analytical modelling imperfections
and tooling mismatches in conventional analytical project assessment techniques. As a universal interna-
tional real-world problem of significance, the literature review identified critical knowledge and methodology
gaps as the primary causes of modelling deficiencies and assessment uncertainties. By following a design-
thinking methodology, the thesis views the sustainability assessment and modelling problem through a geo-
graphical information systems lens, thus seeing an academic research opportunity to fill critical knowledge
gaps through new theory formulation and geographical knowledge creation. To this end, this philosophi-
cal investigation proposes a novel object-oriented systems-thinking and climate modelling methodology to
study the real-world geospatial behaviour of functioning floatovoltaic systems from a dynamical system-
thinking perspective. As an empirical feedback-driven object-process methodology, it inspired the thesis to
create new knowledge by postulating a new multi-disciplinary sustainability theory to holistically characterise
agricultural floatovoltaic projects through ecosystems-based quantitative sustainability profiling criteria. The
study breaks new ground at the frontiers of energy geo-informatics by conceptualising a holistic theoretical
framework designed for the theoretical characterisation of floatovoltaic technology ecosystem operations
in terms of the technical energy, environmental and economic (3E) domain responses. It campaigns for a
fully coupled model in ensemble analysis that advances the state-of-the-art by appropriating the 3E theo-
retical framework as underpinning computer program logic blueprint to synthesise the posited theory in a
digital twin simulation. Driven by real-world geo-sensor data, this geospatial digital twin can mimic the geo-
dynamical behaviour of floatovoltaics through discrete-time computer simulations in real-time and lifetime
digital project enactment exercises. The results show that the theoretical 3E framing enables project due
diligence and environmental impact assessment reporting as it uniquely incorporates balanced scorecard
performance metrics, such as the water-energy-land-food resource impacts, environmental offset benefits
and financial feasibility of floatovoltaics. Embedded in a geoinformatics decision-support platform, the 3E
theory, framework and model enable numerical project decision-supporting through an analytical hierarchy
process. The experimental results obtained with the digital twin model and decision support system show
that the desktop-based parametric floatovoltaic synthesis toolset can uniquely characterise the broad and
diverse spectrum of performance benefits of floatovoltaics in a 3E sustainability profile. The model uniquely
predicts important impact aspects of the technology’s land, air and water preservation qualities, quantifying
these impacts in terms of the water, energy, land and food nexus parameters. The proposed GIS model
can quantitatively predict most FPV technology unknowns, thus solving a contemporary real-world prob-
lem that currently jeopardises floating PV project licensing and approvals. Overall, the posited theoretical
framework, methodology model, and reported results provide an improved understanding of floating PV
renewable energy systems and their real-world behaviour. Amidst a rapidly growing international interest in
floatovoltaic solutions, the research advances fresh philosophical ideas with novel theoretical principles that
may have far-reaching implications for developing electronic, photovoltaic performance models worldwide.
Key terms: Sustainable development, Floating photovoltaic systems, Mathematical modelling, Floating PV habitat mi-
croclimate; Hydroclimatic conditions modelling, Sustainability profiling, Agricultural energy system synthesis, WELF-
nexus, Environmental profiling, Environmental impact assessment, EIA, Balanced scorecard metrics, Sustainable
agriculture, Environmental offset, Water economic surface transformation, Energy transition, Land-water footprint.
iii
Table of contents
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
Abstract .................................................. iii
Table of contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
Definitions and Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Abbreviations and Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
Chapter 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Research Rationale . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.2 Problem Statement and Purpose of Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2.1 Problem identification and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.2.2 Problem statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
1.2.3 Purpose of the study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
1.3 Research Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.3.1 Research design framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
1.3.2 Research design process layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1.3.3 Research hypothesis and research questions . . . . . . . . . . . . . . . . . . . . . 24
1.3.4 Research aim and objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
1.3.5 Concise research methodology and scope . . . . . . . . . . . . . . . . . . . . . . . 28
1.4 Significance of the Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
1.5 Definition of Terms and Research Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . 33
1.6 Thesis Outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Chapter 2. Literature Review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
2.2 Regulatory Context of Sustainable Energy Development . . . . . . . . . . . . . . . . . . . . 39
2.2.1 Global development of sustainable energy . . . . . . . . . . . . . . . . . . . . . . . 39
2.2.2 Development of sustainable energy in South Africa . . . . . . . . . . . . . . . . . . 41
2.2.3 Opportunities for FPV sustainable energy projects in South Africa . . . . . . . . . . 43
2.3 Uncovering Floatovoltaic Technology and its Impact Effects . . . . . . . . . . . . . . . . . . 45
2.3.1 Uncovering floating solar technology . . . . . . . . . . . . . . . . . . . . . . . . . . 45
2.3.2 Salient impact attributes of floatovoltaics reported in literature . . . . . . . . . . . . . 49
2.3.3 Floatovoltaic impact effects supporting sustainability farming . . . . . . . . . . . . . 52
2.4 Developments and Challenges of Floatovoltaic Technology . . . . . . . . . . . . . . . . . . 55
iv
TABLE OF CONTENTS v
2.4.1 Global-scale diffusion potential of floating solar technology . . . . . . . . . . . . . . 55
2.4.2 Diffusion potential for floatovoltaic technology in South Africa . . . . . . . . . . . . . 59
2.4.3 Floating solar PV technology readiness considerations . . . . . . . . . . . . . . . . 63
2.5 State-of-the-Art in Floating PV Sustainability Profiling . . . . . . . . . . . . . . . . . . . . . 64
2.6 Scope for Innovation and Creation of New Knowledge . . . . . . . . . . . . . . . . . . . . . 71
2.7 Summary ............................................. 78
Chapter 3. Philosophical Modelling Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
3.2 Philosophical Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.2.1 Philosophical paradigm and approach . . . . . . . . . . . . . . . . . . . . . . . . . . 81
3.2.2 Framing of research ontology, epistemology, and methodology . . . . . . . . . . . . 83
3.2.3 Philosophical goals for a systemic theoretical sustainability modelling concept . . . . 84
3.3 Methodological Approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
3.3.1 Computer synthesis modelling as a methodological approach . . . . . . . . . . . . . 88
3.3.2 Methodological sequence from initial concept to computer model . . . . . . . . . . . 90
3.4 Methodological Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
3.4.1 Options towards selecting a research framework and methodology . . . . . . . . . . 91
3.4.2 Methodological framework and research procedures . . . . . . . . . . . . . . . . . . 93
3.5 Theoretical Sustainability Assessment Framework for FPV . . . . . . . . . . . . . . . . . . . 95
3.5.1 Observations and identified patterns of error in FPV performance assessments . . . 95
3.5.2 Deriving systemic principles from analytical FPV behaviour data patterns . . . . . . 97
3.5.3 Understanding the systemic dynamics of floatovoltaic operations behaviour . . . . . 99
3.5.4 Formulating a tentative theoretical hypothesis to frame FPV sustainability . . . . . . 102
3.6 Research Model Design and Realisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
3.6.1 Computer modelling of a systems framework for FPV assessments . . . . . . . . . . 109
3.6.2 Model construct of a sustainability assessment framework system . . . . . . . . . . 110
3.6.3 Implementation flowcharts for analytical model and decision model . . . . . . . . . . 114
3.7 Theoretical Substantiation of the Research Methodology . . . . . . . . . . . . . . . . . . . . 116
3.8 Summary ............................................. 119
Chapter 4. Parametric Modelling Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
4.2 Research Model: Systems-level Design and Implementation . . . . . . . . . . . . . . . . . . 121
4.2.1 Systems-based causal analysis in an object-process methodology . . . . . . . . . . 121
4.2.2 Object-oriented system dynamics modelling solution . . . . . . . . . . . . . . . . . . 124
4.2.3 Multiscale system dynamics simulation model for floating PV . . . . . . . . . . . . . 127
4.3 Research Model: Component-level Design and Implementation . . . . . . . . . . . . . . . . 130
4.3.1 Model-based design of floating PV systems components . . . . . . . . . . . . . . . 130
4.3.2 Model-based design parameters, notations and configurations . . . . . . . . . . . . 131
4.3.3 Energy simulation model object design . . . . . . . . . . . . . . . . . . . . . . . . . 133
4.3.4 Environmental simulation model object design . . . . . . . . . . . . . . . . . . . . . 140
4.3.5 Economic simulation model object design . . . . . . . . . . . . . . . . . . . . . . . . 149
4.3.6 Cross-object interactions and stock flow linkages design . . . . . . . . . . . . . . . . 158
4.4 Research Model: Decision-level Layering Design and Implementing . . . . . . . . . . . . . . 163
4.4.1 Integration of decision-support systems layering . . . . . . . . . . . . . . . . . . . . 163
4.4.2 Defining sustainability performance indicators for FPV projects . . . . . . . . . . . . 166
TABLE OF CONTENTS vi
4.4.3 Analytical hierarchically-driven decision-support model . . . . . . . . . . . . . . . . 168
4.5 Integrated Data-collection and Data-processing Instrument . . . . . . . . . . . . . . . . . . 173
4.5.1 Integrated discrete-time Python floatovoltaic synthesis model . . . . . . . . . . . . . 173
4.5.2 Geoinformatics PV-analysis model and decision-support toolset . . . . . . . . . . . 176
4.6 Experimental Design, Data Collection and Evaluation . . . . . . . . . . . . . . . . . . . . . 177
4.7 Ethical considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
4.8 Summary ............................................. 179
Chapter 5. Experimental Results and Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . 180
5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
5.2 Experimental Evaluation Layout, Design and Configuration . . . . . . . . . . . . . . . . . . 180
5.3 Experiment 1: Corroborative Floating PV Model Validation . . . . . . . . . . . . . . . . . . . 182
5.3.1 Goal of experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
5.3.2 Experimental method and procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 183
5.3.3 Experimental case-study results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
5.3.4 Conclusions and critical findings of the experimental case study . . . . . . . . . . . 187
5.4 Experiment 2: FPV Project Performance Assessment . . . . . . . . . . . . . . . . . . . . . 189
5.4.1 Goal of experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
5.4.2 Experimental method and procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 190
5.4.3 Experimental case-study results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
5.4.4 Conclusions and critical findings of the experimental case study . . . . . . . . . . . 203
5.5 Experiment 3: Floating Solar Decision-support Analysis . . . . . . . . . . . . . . . . . . . . 212
5.5.1 Goal of experiment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
5.5.2 Experimental method and procedure . . . . . . . . . . . . . . . . . . . . . . . . . . 213
5.5.3 Experimental case-study results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
5.5.4 Conclusions and critical findings of the experimental case study . . . . . . . . . . . 218
5.6 Summary ............................................. 221
Chapter 6. Summary, Conclusions and Novel Contributions . . . . . . . . . . . . . . . . . . . 223
6.1 Topical Research Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
6.2 Research Summary and Synthesis of Critical Findings . . . . . . . . . . . . . . . . . . . . . 226
6.3 Fundamental Research Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
6.3.1 Revisiting the Research Aim and Objectives . . . . . . . . . . . . . . . . . . . . . . 231
6.3.2 Answering the Research Questions . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
6.3.3 Reflecting on the Research Hypothesis . . . . . . . . . . . . . . . . . . . . . . . . . 237
6.4 Strategic Research Conclusions and Recommendations . . . . . . . . . . . . . . . . . . . . 238
6.4.1 Creation of new theory and knowledge to support floatovoltaic sustainability policy . 238
6.4.2 Floating PV sustainability theory as shared vision for PV energy sustainability . . . . 239
6.4.3 Significance of sustainability theory in sustainable agricultural development . . . . . 240
6.4.4 The role of Geography and GIS in new sustainability framing for FPV . . . . . . . . . 242
6.4.5 Analytical framework for floating PV sustainability assessments . . . . . . . . . . . . 243
6.4.6 WELF-nexus driven decision framework for FPV sustainability assessments . . . . . 244
6.4.7 Fundamental principles of sustainability assessment methodology for FPV . . . . . . 245
6.4.8 Sustainability framework and methodology as computer model logic . . . . . . . . . 246
6.4.9 Sustainability decision supporting framework as computer model logic . . . . . . . . 247
6.4.10 Environmental and resource/climate-health economics into FPV modelling . . . . . . 248
6.4.11 Geographical modelling of floating PV natural environmental micro-habitat . . . . . . 250
TABLE OF CONTENTS vii
6.4.12 Geo-informatics-based data-science technique for predictive FPV assessments . . . 252
6.4.13 Comparison of FPV model with existing PV performance models and techniques . . 253
6.4.14 Effective EIA assessments in environmentally-based FPV sustainability . . . . . . . 255
6.4.15 Closing the circle between FPV-model outputs and EIA reporting . . . . . . . . . . . 256
6.4.16 Defending FPV sustainability in classical sustainability theory context . . . . . . . . . 258
6.4.17 Reflecting on role of 3E sustainability vs classical sustainability theory . . . . . . . . 259
6.4.18 Unexpected contribution in support of ground-mounted photovoltaics . . . . . . . . . 261
6.4.19 International relevance and application potential for the research . . . . . . . . . . . 263
6.5 Summary of Novel Contributions of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . 264
6.5.1 Contextualisation of novel contributions of the study . . . . . . . . . . . . . . . . . . 264
6.5.2 Summary of novel contributions of the study . . . . . . . . . . . . . . . . . . . . . . 265
6.6 Academic and Practical Utility of the Research . . . . . . . . . . . . . . . . . . . . . . . . . 269
6.7 Study Limitations and Directions for Future Research . . . . . . . . . . . . . . . . . . . . . 273
6.8 Summary ............................................. 278
APPENDICES ............................................... 280
Appendix A. Mind Map of the FPV Sustainability Narrative . . . . . . . . . . . . . . . . . . . . 281
Appendix B. Environmental-related legislation in South Africa . . . . . . . . . . . . . . . . . . 282
Appendix C. State-of-the-art Tools: PV Performance Analysis . . . . . . . . . . . . . . . . . . . 283
Appendix D. Research Methodology Development Chart . . . . . . . . . . . . . . . . . . . . . 284
Appendix E. Conceptual FPV Sustainability Philosophy/Theory . . . . . . . . . . . . . . . . . 285
Appendix F. Floating PV Model Development Flowchart . . . . . . . . . . . . . . . . . . . . . . 286
Appendix G. Systems-thinking Hierarchical Pyramid . . . . . . . . . . . . . . . . . . . . . . . . 287
Appendix H. Causal-diagram for Floating PV Sustainability . . . . . . . . . . . . . . . . . . . . 288
Appendix I. Floating PV Model Simulation Synthesis Layers . . . . . . . . . . . . . . . . . . . 289
Appendix J. Floating PV System-of-systems Conception . . . . . . . . . . . . . . . . . . . . . 290
Appendix K. Iterative 3E System Dynamics Processing . . . . . . . . . . . . . . . . . . . . . . 291
Appendix L. Conditioning of PVlib Model for Floatovoltaics . . . . . . . . . . . . . . . . . . . . 292
Appendix M. FPV-GIS Model Layers and Components . . . . . . . . . . . . . . . . . . . . . . . 293
Appendix N. FPV-GIS Model Parameter Notations . . . . . . . . . . . . . . . . . . . . . . . . . . 294
Appendix O. Simulation Experiment Configuration . . . . . . . . . . . . . . . . . . . . . . . . . 297
Appendix P. Geosensor Meteorological Data Input Samples . . . . . . . . . . . . . . . . . . . 300
Appendix Q. FPV-GIS Python Model GUI Interface Screens . . . . . . . . . . . . . . . . . . . . 301
TABLE OF CONTENTS viii
Appendix R. Sampled Real-time Simulation Results Extract . . . . . . . . . . . . . . . . . . . . 303
Appendix S. Sampled Life-time Simulation Results Extract . . . . . . . . . . . . . . . . . . . . 305
Appendix T. Sampled Simulated AHP Results Extract . . . . . . . . . . . . . . . . . . . . . . . 309
Appendix U. EGIS Renewable Energy Application Platform . . . . . . . . . . . . . . . . . . . . 312
Appendix V. Proposed Model Additions for Future Research . . . . . . . . . . . . . . . . . . . 316
Appendix W. Potential for Floating PV Hydroelectrics in Africa . . . . . . . . . . . . . . . . . . 317
References................................................. 318
Definitions and Terminology
The following glossary of definitions and terminology apply to this thesis:
△3E Sustainability Theory means a holistic systems theory for sustainability analysis proposed by this thesis to
characterise the operational behaviour of floatovoltaic power plant installations in terms of the mutual Energy,
Environmental and Economic responses. The theory defines an integrated PV project sustainability criteria as un-
derpinning computer logic and systematic procedure for floatovoltaic performance projections in a newly proposed
predictive methodology data science technique.
△3E or EEE or TEE Framework means an integrated Energy, Environmental and Economic framework for PV
project analysis and assessment proposed by this thesis. Synthesising the 3E theory, the framework opera-
tionalises the 3E sustainability theory to realise sustainability prognosis in terms of 3E sustainability indicator pillar
priorities. The three E sustainability indicator categories comprise an array of around 50 performance metrics to
dynamically profile the sustainability prospects of a floating PV technology installation. The 3E framework also
serves as a PV energy project risk rating (similar to the ESG risk rating assessment that expresses project risk in
terms of Environmental, Social and Governance factors).
△Analytical framework means a framework structure to organise and schedule the analysis of an ecosystem
through the fusion of 3E data in an intelligent system dynamics environment. In an informatics context, the frame-
work concept serves as a means to administrate and organise the analytical process and analytical data streams
into a set of inter-connected 3E categories or discipline domain elements as software objects jointly involved in the
simulated response of a system.
△Analytical reference framework is defined as an integrated theoretical framework designed to structure the re-
searcher’s thinking towards technology performance and impact analysis of floating photovoltaic system prototypes.
It systematically implements the proposed analytical methodology to facilitate the definition of the thesis’s holistic
systems thinking philosophy in a systemic, logical thinking structure.
△CAS Model means a Computer Analytical Simulation Model established by a discrete-time system dynamics simu-
lation model to synthesise the operations of a floating PV system through mathematical algorithms based on actual
real-world geospatial inputs.
△ce-WELF indicators refer to a set of six sustainability indicators comprising climate, economic, water, energy, land,
and food (WELF) metrics defined to measure and portray floating PV sustainability in terms of a climate-economic
WELF index sustainability criteria.
△Computer Model means an established numerical implementation of the mathematical (floating PV) system in the
form of digital discretisation in discrete-time mathematical solution algorithms for which can construct a verification
experiment.
△Conceptual Model defines a collection of systemic assumptions, elements, relationships, algorithms and data that
describes the reality of interest to construct the mathematical computer model and validation experiment.
△Decision support framework defines a framework to drive decision support systems in a class of computerised
information systems to support numerical, computer-assisted or user-assisted support in decision-making activ-
ities towards ensuring transparent (sustainability) decisions based on the impacts, performances and economic
observations of project outcomes.
ix
DEFINITIONS AND TERMINOLOGY x
△Decision Platform refers to a proposed conceptual planning or operational assessment solution to support the
enterprise decision evaluation. The decision platform is designed per an engineering model-based Design Thinking
methodology (logic thinking) and simulation approach to finding a desirable and realistic solution for agricultural
applications to solve complex floating solar energy system problems in an action-orientated solution.
△Digital twin model means a virtual representation model driven by geospatial sensor data to serve as the real-time
digital counterpart of a physical system. In the floatovoltaics context, the geospatial digital twin defines a computer
simulation model driven by actual real-world geodata.
△Ecosystem (specifically a Floatovoltaic Ecosystem) means the overall floating solar business enterprise ecosys-
tem, including its technical energy, financial economic and environmental impact behavioural parameters and prop-
erties as defined in this thesis.
△Economic Surface Transformation (EST) is a new performance evaluation ratio and term defined by this thesis
to describe the economic transformation of a geo-fenced area through the installation of photovoltaic technology.
△Energy efficiency refers to the effective use of energy to produce a given output (in a production environment) or
service (from a consumer point of view), i.e. a more energy-efficient technology has the same service or production
with less energy input.
△Environmental offsets means environmental impacts that offer positive mitigation measures to compensate for
negative environmental actions or impact effects (implementation of the mitigation sequence and achievement of
”net environmental gain" and ”like for like" practice).
△Floatovoltaic system ontology defines the FPV ecosystem’s dynamical domain properties, and relationships
among domain object elements (driven by a systemic framework defining the floating PV ecosystem "language”
and "vocabulary").
△Floatovoltaic system topology defines the FPV ecosystem’s topological structural anatomy comprising knowl-
edge domain objects or object elements according to the 3E framework.
△FPV-GIS model and toolset means the proposed Floating PhotoVoltaic (FPV) GeoInformatic ecoSystem (GIS)
toolset conceptualised, developed and implemented by this thesis to characterise floatovoltaics technology and
to quantify and profile the performance qualities and impacts effects of a future planned floatovoltaic installation
project in terms of the 3E framework.
△Floatovoltaitry is a new term defined by this thesis for the floatovoltaic energy system application field, including
practices, processes, actions, activities and scientific evaluations associated with the planning, design and con-
struction of floatovoltaic energy systems. It establishes and inaugurates one of the new definitional terms created
by the present philosophical research investigation to linguistically support the semantics related to floatovoltaic
technology in the context of geo-information systems.
△Floatoplanometrics is a new term defined by this thesis for the floatovoltaic application field as the definition of
informetrics, indicator metrics or scientific evidence associated with floatovoltaic energy system application field,
including informetrics associated with planning, analysis metrics and decision measures around photovoltaic tech-
nology energy systems. It establishes and inaugurates one of the new definitional terms created by the present
philosophical research investigation to linguistically support the semantics related to floatovoltaic technology in the
context of geo-information systems.
△Floatoplanometry is a new term defined by this thesis for the floatovoltaic application field as a means to define
the activities and measures associated with the planning and analysis (thought processes) related to floatovoltaic
energy system application field. It establishes and inaugurates one of the new definitional terms created by the
present philosophical research investigation to linguistically support the semantics related to floatovoltaic technol-
ogy in the context of geo-information systems.
△Geosgraphical means relating to the features of the surface of the earth or results based on or derived from the
physical features of an area on the earth’s surface. In the floatovoltaic assessment context, the term geographical
refers to the broader geographic habitat of the floatovoltaic system in terms of climate, microclimate and hydrocli-
mate attributions through the engagement of actual real-world geodata such as meteorological data and climatic
data associated with the area of FPV installation.
DEFINITIONS AND TERMINOLOGY xi
△Geospatial digital twin means a virtual representation model driven by geospatial sensor data to serve as the
real-time digital counterpart of a physical system. In the floatovoltaics context, the geospatial digital twin defines a
computer simulation model driven by actual real-world geodata.
△Geographical model-based design relates to the practice of leveraging simulation to understand the behaviour
of a physical geo-system. It thus refers to the stepwise design of computer models from theoretical frameworks
that can engage actual real-world geodata such as meteorological data and climatic data to predict aspects such
as spatial relationships, interaction with or across space, and other issues of geography.
△Impact effects (floatovoltaics) in this thesis includes a subset of the SA NEMA Act to define relevant impacts of
a proposed floating solar project on the atmosphere, land or water quantity, including relevant inter-relationships
between its physical conditions.
△ISA framework for floatovoltaics defines a newly conceptualised 3E tridentate as an integrated performance
sustainability assessment framework and criteria for floatovoltaic sustainability characterisation and profiling. This
Integrated Sustainability Assessment (ISA) Framework implies heterogeneous co-simulation of the operational en-
ergy, environmental and economic system responses in an integrated system dynamics co-simulation methodology.
△Optimal economic dispatch describes an experimental condition whereby the supply side power generation is in
balance with the demand side consumption of the power, meaning all the energy generated by the power plant is
consumed, which in turn implies maximum potential environmental impact for a proposed FPV system and location.
△Profile mapping display means a device for the articulation of (cumulative framework-driven) analytical results is a
radar-type display map, enabling integrated assessment results to be articulated and presented visually in support
of sense-making and understanding.
△Sustainability (conventional/classical) means the avoidance of the depletion of natural resources by maintaining
an ecological balance in the broader geographical context of the social, economic and ecological domains. It in-
volves interpretation and context-specific understanding of the impact of humans on planet Earth through a balance
of system performance/project quality goals across the social, economic and ecological (SEE) dimensions.
△Sustainability (3E) in this thesis defines three lenses of resource-use /performance sustainability, characterised
by the disciplinary fields of energy, economic and environmental systems. The sustainability framework acts in an
overlaying/interactive manner to respect dynamic system integrity in balancing 3E system performance/resource-
use quality goals across the technical energy, environmental and economic (3E, EEE or TEE) dimensions.
△Sustainability assessment is a complex appraisal method to support the notion of collaborative, sustainable
development quantitatively. While sustainability assessment is defined by criteria and processes, the appraisal
of sustainable development projects centres around the three critical pillars of sustainability, namely technical,
economic, and environmental sustainability, in an eco-systemic valuation context.
△Sustainable development is defined as an organising principle or development principle in energy development
practice towards economic development planned without the depletion of natural resources. It is a development
principle that plays an integral role in the energy development practice, especially in strategic environmental as-
sessment, and environmental impact assessment as well as in making development assistance work at the level of
project planning, sustainability policies, and approval programmes.
△Sustainability Impact Assessment Framework is defined by this thesis as an integrated contextual sustainability
valuation framework, expressed as a geo-sensitive approach for exploring the combinational effects of the tech-
nical, economic, and environmental aspects of floating solar development plans. It is a criterion for quantitatively
appraising planning scenarios for floatovoltaic technology in a heuristic analytical decision space.
△Water Economic Surface Transformation (w-EST) is a new performance evaluation ratio and term specific to
floatovoltaics technology defined by this thesis as a quantitative econometric to measure the extent to which every
square metre of formerly unused water surface area is transformed into an income production space or a revenue-
generating space owing to floatovoltaics technology.
△WELF nexus framework means the interactive nexus across the water, energy, land and food (WELF) resource
system as a measure of the potential for improved resource use efficiency and resource interaction efficiency from
an ecosystem perspective.
Abbreviations and Nomenclature
Abbreviations and Acronyms
3E Energy, Economic and Environmental
3P People, Planet and Profit
4IR Fourth Industrial Revolution
AHP Analytical Hierarchy Process
AI Artificial Intelligence
AIP Artificial Intelligence Programming
AIR Application Information Requirements
API Application Programming Interface
APP Mobile Phone Application
BOS Balance of System
CAD Computer Aided Design
CAM Customer Adoption Model
CAR Computer-Aided Reasoning
CART Classification & Regression Tree
CBA Cost-Benefit Analysis
CBR Case-Based Reasoning
CC Carbon Credits
CDM Clean Development Mechanism
CE Computational Estimation
CEF Carbon Emission Factor
CER Certified Emission Reduction
CpI Computational Intelligence
CMM Computational Mathematics Modelling
CMR Climate Model Resolution
CMS Computer Modelling and Simulation
CPSs Cyber-Physical Systems
CPI Consumer Price Index
CSA Climate Smart Agriculture
DDA Due Diligence Analysis
DER Distributed Energy Resources
DEST Digital Ecosystems Technologies
DLL Dynamic Link Library
DP Dynamic Programming
DSS Decision Support System
DTM Digital Terrain Model
DTS Design Thinking System
E3SM Energy Exascale Earth System Model
EA Environmental Approvals
EAP Environmental Assessment Practitioner
EAR Environmental Application Review
EDD Economic Due-Diligence
EGIS Environmental GIS platform
EIM Environmental Impact Model
EEE Energy Environmental Economic
EFS Environmental Feasibility Studies
EIA Environmental Impact Assessment
EIAaaS EIA as a Service
EIAR EIA Report
EIP Environmental Implementation Plan
ELF Energy Land Food nexus
EMF Environmental Management Framework
EMP Environmental Management Plan
ERM Electronic Records Management
ESI Enterprise System Integration
ESR Environmental Scoping Report
ESS Environmental Scoping Study
EWF Energy Water Food - nexus
EWL Energy Water Land - nexus
FPV Floating PhotoVoltaic System
FPVS Floating photovoltaic System
FPVT Floating Photovoltaic Technology
FSPV Floating Solar Photovoltaic system
FPVT Floating Photovoltaic Technology
GAMS General Algebraic Modelling System
GHG Green-House Gas
GIS Geographical Information System
GIT Geographic Information Technology
GPS Global Positioning System
GPV Ground-mounted PhotoVoltaics
GUI Graphic User Interface
HMI Human Machine Interface
ICT Information Communication Technologies
xii
ABBREVIATIONS AND NOMENCLATURE xiii
IDP Integrated Development Plan
IEM Integrated Environmental Management
IES Integrated Energy System
IDE Integrated Development Environment
ISA Integrated Sustainability Assessment
LPV Land-based Photovoltaic installations
MBD Model-Based Design
MCA Multi-Criteria Analysis
MCDA Multi-Criteria Decision Analysis
ML(A) Machine Learning (Algorithm)
MMF Multi-scale Modelling Framework
MPPT Maximum Power Point Tracker
NWP Numerical Weather Prediction
PDA Personal Digital Assistant
PDF Portable Document Format
PFD Process Flow Diagram
PSS Power System Simulation
QGIS Quantum GIS platform
RE Renewable Energy
REC Renewable Energy Certificates
RES Renewable Energy Systems
REDZs Renewable Energy Development Zones
RET Renewable Energy Technologies
RTDS Real-Time Digital Simulator
SDG Sustainable Development Goal
SEA Strategic Environmental Assessment
SEE Social Ecological Economic
SES Smart Energy System
SET Sustainable Energy Technologies
SME Subject-Matter Expert
SOA Service-Oriented Architecture
SOAn Solar Operational Analytics
SOP Standard Operating Procedure
SPA Smart Process Automation
SPI Sustainability Performance Indicators
SREC Solar Renewable Energy Credits
STC Standard Test Condition
TEE Technical Environmental Economic
TMY Typical Meteorological Year (sensor data)
ToD Time-of-Delivery
ToU Time-of-Use (tariff)
TRNSYS TRaNsient SYstem Simulation
VIS Virtual Intelligent Sensor
VSN Virtual Sensor Network
WEF Water Energy Food (nexus)
WELF Water Energy Land Food (nexus)
WPVS Water-based Photovoltaic Systems
ZNE Zero Net Energy
Substance Element Abbreviations
aLP agricultural land preserved
aPE airborne particle emissions
CH4methane
CO2carbon dioxide
CO2e carbon dioxide equivalent
COxcarbon oxides
H2O water or dihydrogen monoxide
HFC hydrofluorocarbons
NH3ammonia
NOxnitrogen oxides
O3ozone
Pesolar power
PM Particulate Matter
POPs Persistent Organic Pollutants
SOxsulphur oxides
Mathematical Symbols and Units of Measure
A Area (m2)
J Joule
kg kilogram
kl kilolitre
kV kilo-Volt
kW kilo-Watt (one thousand Watt)
kWh kilowatt hour
kWe kilo-Watt (one thousand Watt electrical)
kWh kilo-Watt hour (one thousand Watt per hour)
kWp kilo-Watt peak
kWt kilo-Watt thermal (one thousand Watt thermal)
kgCO2ekilogram Carbon Dioxide equivalent
l litres
MW megawatt
MWh megawatt hour
GW gigawatt (one thousand megawatts)
W watt
Wh watt-hour
Meteorology & Radiometry Acronyms & Units
GHI Global Horizontal Irradiance [W/m2]
DNI Direct Normal Irradiance [W/m2]
DHI Diffuse Horizontal Irradiance [W/m2]
GTI Global Tilted Irradiance [W/m2]
RH Relative humidity [%]
UV Ultraviolet irradiance [W/m2]
WD Wind direction [deg]
WS Wind speed [m/s]
1. Introduction
Research Scope
Under the sustainability science discourse (Lee et al., 2022; Maka, 2022), holistic smart-energy technolo-
gies supporting water conservation, resource preservation and the protection of land, soil and vegetation
are receiving increased attention (Hernandez et al., 2022; Vema et al., 2022). Climate-smart trends in
energy systems, security and sustainability have led to recent advances in solar energy towards intelli-
gent, efficient and sustainable methods in sustainable development (Novas et al., 2021; Pouran et al.,
2021). While sunlight is the fuel for all solar energy technologies (IEA, 2021), solar photovoltaic energy
is considered an environmentally-friendly technology generated close to the point of energy use. Recent
environmental engineering enterprises have further led to the invention of floating solar technology (floato-
voltaics) for generating renewable energy that inherently improves the ability of natural systems to preserve
the natural ecosystem (Allen and Prinsloo, 2018; Essak and Ghosh, 2022; Wood Mackenzie, 2019). The
technology contributes to the land-use intensity discussion, a narrative that presents a significant drawback
in most modern electrical power production schemes (Havrysh et al., 2022; Oudes, 2022), especially be-
cause new brownfield energy developments often involve the sacrifice of biomass and crop production in
food-producing regions to deliver marginal residual life cycle climate change benefits (Lovering et al., 2021).
Considerations around improving the understanding of sustainability in landscape-inclusive energy trans-
formations sparked an interest in exploring the catalytic benefits of farm land-surface reclamation and irri-
gation water preservation of floatovoltaics. The technology is considered a climate-solving enabler of sus-
tainable growth in agricultural food production and a frontrunner in cultivation and resource conservation
energy technology of late (Allen and Prinsloo, 2018). However, many knowledge gaps exist around floa-
tovoltaic project sustainability principles (Gadzanku et al., 2021b), especially concerning the technology’s
performance- and resource-use-based sustainability (energy, land, air, water). With sustainability character-
isation of agriculturally-integrated floatovoltaic technologies being the main topic of this thesis, sustainability
performance modelling is required to go beyond the conventional measures of integrated impact assess-
ment of climate change risks (Hottenroth et al., 2022; UN, 2022). The core sustainable energy delivery
theme of the research is highly relevant and meaningful given the context of an ongoing need for sus-
tainable energy production, the rapid emergence and explosive growth of recently discovered floatovoltaic
technology worldwide, and the international need to improve assessment practice in project commissioning
for newly conceived sustainable energy generation technology innovations.
The rapid emergence and worldwide growth of floating photovoltaic installations in recent years can in
part be ascribed to the excellent resource preservation properties (water, energy, land) attributed to this
newly discovered clean energy generation technology (World Bank, 2019b). With floatovoltaic energy sup-
ply installations sited directly on ponds and dams in open-water farming irrigation systems, this technology
interfaces more intensely with the natural environmental system to deliver a broader and more diversified
range of value-laden benefits and co-benefits. It further offers superior sustainability qualities, significantly
improving photovoltaic projects’ sustainability status and conservation qualities. These sustainability bene-
1
fits and qualities include various resource-use efficiencies, impact effects, environmental offsets and water-
energy-land-food nexus positives (Banik and Sengupta, 2021; World Bank, 2019b), as particularised in
the literature study of this thesis. In addition to the documented conservation and resource benefits, each
comes with an associated financial reward thanks to the global climate change drive and its climate-health
compensations and incentives. The extended range of technology and sustainability attributions thus offer
highly beneficial prospects to the agricultural, hydro-power and wastewater-processing industries. There
is a need to develop integrated contextually-sensitive computer models to cater to the different application
contexts that can support these applications. These would be necessary to properly characterise the sus-
tainability status of such projects in terms of the broad-spectrum behaviour and geo-dynamic responses
of floating solar technology (Armstrong et al., 2020; Kumar et al., 2022). At the same time, due to mod-
elling uncertainties, the robust validation of the co-benefits and suggested impacts of the technology on the
agricultural nexus of the local energy-water-food (EWF) system is further lacking (Gadzanku et al., 2021b;
Prinsloo and Lombard, 2015b; Prinsloo, 2020). In studying the sustainability efficiency, effectiveness and
reliability of floating photovoltaic power systems, this thesis argues that the science of holistic-systems sus-
tainability would be a crucial requirement to meet the above needs in developing cutting-edge models to
support the integrated sustainable development discourse.
Under the thesis’s proposed performance- and resource-based sustainability science discourse, the
newly invented climate-solving floatovoltaic energy technology offers a diverse and broad spectrum of en-
vironmental offsets, including natural resource preservation and sustainability attributes that can practically
support environmental sustainability certifications of the technology in the agricultural sector (Allen and
Prinsloo, 2018). However, with floatovoltaic technologies in agricultural applications directly interfacing with
water as a sensitive and critical natural resource in food security worldwide, the lingering threat of inter-
nationally anticipated licensing regulations for floatovoltaic technology installations creates a reason for
concern (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). The legislative requirements for the in-
ternational adoption and environmental licensing of floating solar installations delivering renewable energy
technology in the agricultural sector have already created an urgent need for predictive performance eval-
uation and the characterisation of future floating photovoltaic projects focusing on scientific sustainability
(Gadzanku et al., 2021b; Hernandez et al., 2019).
At the same time, international regulators are interested in gauging the rural agricultural (floatovoltaic)
energy system sustainability portfolios in terms of the support it provides to natural WELF resources and
preservation of global climate change initiatives (United Nations, 2019), as well as the global sustainable
development goals (SDGs) of the United Nations (UN) (UN, 2020a; World Bank, 2017). In considering all
of these strategic goals on an operational level, however, the conventional photovoltaic performance mod-
elling framework paradigm both fails to quantify the broad and diversified spectrum of floatovoltaic system
performances, while failing to provide adequate answers to help solve pressing regulatory requirement chal-
lenges in an integrated manner (Hernandez et al., 2020; Prinsloo, 2019, 2020). The industry is looking to-
wards geographers to solve the (human-environmental-ecological) sustainability assessment problem from
an energy, environmental and economic geography perspective since such sustainability is a geographically
sensitive concept (Hansen and Coenen, 2015). Towards an integrated modelling approach, the information
technology field of Geography has already developed the software tools and cloud computing infrastructure
to study sustainable energy transitions from an energy data systems science perspective (Fu, 2020; Truffer
et al., 2015). While new theoretical strategies, frameworks, models and toolsets are needed to improve
the theoretical characterisation of floatovoltaic technologies (Armstrong et al., 2020; Kumar et al., 2022;
Prinsloo, 2020), this thesis contemplates a more holistic evaluation strategy for sustainability assessment
of agri-floatovoltaic systems. The sustainability assessment goal is to ensure improved alignment with the
multidisciplinary goals of agricultural production sustainability certifications in an FPV project performance
framework and model that can help turn performance predictions into actionable insights.
2
Within this technologically-oriented sustainable energy development context, the thesis is concerned
with a geographical investigation into the conceptualisation and realisation of a geo-informatics approach
to support design stage sustainability assessments of floatovoltaic technology in South African agricultural
applications. With the quantitative sustainability profiling problem seen from a unique analytical and envi-
ronmental data science perspective, the thesis advances the idea that the root of the identified real-world
geographical problem lies at a three-pronged assessment fault line. This study argues that the challenge
relates, in the first instance, to the consensus of scientists that both conventional PV simulation and PV
performance assessment models cannot cope with the anomalous behaviour of floatovoltaic-type systems,
in the second instance, the concerns of impact practitioners that conventional PV simulation and perfor-
mance assessment models cannot perform integrated (environmental) impact assessments and, in the third
instance, the concerns of sustainable developers that classical sustainability policy for energy project as-
sessment (based on people-planet-profit criteria) does not cover all the bases when assessing the complex
spectrum of sustainability attributions delivered by floatovoltaic-type energy systems. Through scientific
inferences from data observations in a systematic literature study, the thesis identifies critical knowledge
and methodological gaps that stand in the way of the diffusion and assimilation of the technology in terms
of, firstly, the characterisation of the technology’s impact effects, secondly, the requirements for proper fi-
nancial due-diligence analysis and, thirdly, the regulatory obligations for approving and permitting the field
deployment of FPV energy technology systems.
Within this context, the thesis recognises the opportunity to create new knowledge to fill the identified
gaps. It sees a hypothetical chance to conceptualise a hybrid sustainability policy theory that can enhance
the concurrent assessment of the behaviour, performance, impacts and sustainability of floatovoltaic instal-
lations in a single integrated algorithm. The thesis identifies theory-building prospects in creating a new
holistically integrated theoretical policy to infuse sustainability and climate change principles into opera-
tional PV performance modelling through a hybrid sustainability criteria framework model as the basis for a
multi-hierarchy dynamical systems algorithm. Such a proposed enabling sustainability theory and improved
framework could offer catalytic prospects in underpinning computer logic in the unique theoretical synthesis
of the operational behaviour of floatovoltaic systems as the basis for the contextual characterisation of both
the sustainability and performance qualities of floatovoltaic installations. Therefore, the thesis poses an
academic argument for crafting a new hybridised sustainability-performance-behaviour policy to establish a
new data science-driven technique to simultaneously model and synthesise the behavioural-, performance-
, impact- and sustainability- responses of floatovoltaic energy systems. Such a proposed methodology
would advance renewable energy valuations at the frontiers of science in the cross-disciplinary fields of
sustainable development, geographical data science, environmental science, and sustainability informat-
ics. Furthermore, it would advance the current state-of-the-art design-stage analytical and decision-support
modelling for floating PV energy systems through numerically automated impact assessment based on
geoinformatics platform processing capabilities.
With the imperative goal of geographical theory building, the thesis formulates an inductive method-
ological framework to spearhead the research towards philosophical theory-building under the sustainable
development discourse. Through this methodology, the study theorises about the role and impact of the nat-
ural resource interactions and aquatic reservoir habitat-induced hydroclimatic variability on the distinct spa-
tiotemporal behavioural and sustainability responses observed in field-installed floating photovoltaic power
generating systems. The thesis hypothesises that these undocumented hydro-climatic-type environmental
feedback interactions and climate-health economics are among the key drivers of the distinct variance and
diversity in the technical behaviour, performance dynamics and sustainability attributions of floatovoltaic
complexes. A hybrid theoretical solution for this three-tiered modelling complexity problem is significant
because theoretical imperfections with current technical photovoltaic behaviour and performance models
are causing deeply problematic challenges in the commissioning of new floatovoltaic energy project plans
3
worldwide (Cohen and Hogan, 2018; Gadzanku et al., 2021a; Hernandez et al., 2019). The thesis proposes
a holistically integrated theoretical framework remedy to address customary tooling deficiencies since the
existing analytical theories, framework paradigms and computer toolsets fail to offer inclusive feasible an-
swers to the real-world analytical and modelling problems associated with floatovoltaics system behaviour,
performance and sustainability (Armstrong et al., 2020; Gadzanku et al., 2021b; Kumar et al., 2022). Aiming
at developing a new theoretical framework and modelling synthesis, it engages in a geographical operations
research approach and methodology to analyse and theorise about real-world system operations of floato-
voltaics from data observations. This methodology helped to conceptualise a new theoretical sustainability
policy as a foundation for formulating a new theoretical framework and modelling synthesis as computer
logic in a next-level state-of-the-art geographical toolset for integrated sustainability assessments of agri-
cultural floatovoltaic power station installation projects.
Towards challenging the current state-of-art in PV sustainability profiling, there are two main strands
involved in the methodology of this research, namely understanding the contextual nature of sustainabil-
ity in floatovoltaic systems (Chapters 2 and 3), and finding ways of incorporating that understanding into
sustainability investigations in support of the sustainable development discourse (Chapters 4 and 5). In
this inductive research study, the research questions warrant the philosophical theory-building and design
research around characterising the spatio-temporal geodynamics of floatovoltaic technology sustainability.
It inspired the formulation of three research questions to solve real-world problems associated with the the-
oretical assessment and contextually-sensitive geographical characterisation of the distinct broad spectrum
of performance qualities and impact benefit offerings unique to floatovoltaic technology. With the research
questions advancing the state-of-the-art to whole-systems level, the thesis extends the diagrammatic view
of ecosystem sustainability, wherein floatovoltaics operations ecosystem theory regards the geodynamical
floatovoltaic system operations as an autonomous climate-smart energy production system nestled within a
productive agricultural habitat and impacting across its water-energy-land-food resource sectors. With agri-
cultural energy generation encompassed within a broader agricultural production ecosystem (Zarei et al.,
2021), the thesis posits a more holistic analytical framework for floatovoltaics built around the intelligence
and logic in modelling a novel floatovoltaic ecosystem. The thesis makes the academic argument that a pro-
posed analytical framework should uniquely incorporate the dynamics behind environmental economics with
an understanding of the environment as a principal component in a theoretically-based modelling framework
for the modelling of a floatovoltaic agri-ecosystem. Giving a unique geographical operations research per-
spective creates the opportunity for the composition of a holistic sustainability policy strategy for agri-energy
production wherein the theoretically established sustainability criteria define part of a cooperative network
or operational energy agency structure that mirrors that of a sustainable agri-ecosystem (water-energy-
land-food sustainability). From a landscape-inclusive energy transition perspective (Hernandez et al., 2022;
Lovering et al., 2021), cross-disciplinary agricultural energy ecosystem sustainability modelling can, as
such, leverage concrete agglomeration benefits offered by the techno-environmental-economic domain
ecosystem ensemble within the aquatic theatre of operation of the FPV system. Modelling sustainability
across the proposed three spheres of sustainability, namely the energy-environmental-economic spheres
(3E domains), constitutes an integrated cross-disciplinary ecosystem wherein agency theory or network
theory can improve the modelling of inter-domain connections. The thinking behind such holistic-systems
orientation inspired the conceptualisation of a novel theoretical project assessment framework and com-
puter model expression distinctively built around fresh perspectives on floatovoltaic modelling logic and
framework for technology ecosystem intelligence.
The study advances the state-of-the-art with the postulation of a foundational theoretical principal, a new
analytical framework theory is composed to explain and account for the distinct broad-spectrum of quanti-
tative sustainability attributions of floating photovoltaic technologies. With the drafting the newly proposed
theoretical principal for geographic-theoretic analysis of floatovoltaic sustainability, the research ensures
4
an improved understanding of agricultural floatovoltaic technology. Formulated around the key drivers of
floatovoltaic sustainability, the theoretical and framework contribution can largely explain the distinct broad-
spectrum variance and diversity in the behaviour and sustainability attributions observed at in-field installed
floating photovoltaic projects. Through the contextualisation of the sustainability theory, the thesis aims to
make novel contributions by identifying and bridging critical knowledge gaps around analytical strategies
and structures to help mitigate multi-disciplinary analytical challenges around the theoretical characterisa-
tion of floatovoltaic project sustainability. As such, the posited theoretical framework gives a significant
edge over conventional toolsets. It helps create new knowledge to cross critical knowledge gaps and
methodological barriers experienced in current deficiencies with computer modelling and tooling for floa-
tovoltaics. The thesis’s theoretical and modelling contributions further inspired the design and formulation
of a comprehensive set of novel sustainability metrics and indicators to assess floatovoltaic projects. Defin-
ing new indicators became imperative as standardised metrics for ground-mounted photovoltaic projects do
not adequately account for floatovoltaic technology’s extended range of distinct and diversified spectrum
of resource-use-efficiencies and impact-effect-positives. The proposed performance metrics leverage the
relevant technical, energy, economic and environmental aspects in indicators-based sustainability assess-
ments and decision-support in floatovoltaic design-stage evaluations. With the proposal of a novel theoreti-
cal framework concept and methodological solution creating a novel geographical data science technique,
it helps, through computer synthesis, to bridge critical operational and systemic knowledge gaps around the
empirical quantification of the complex and diversified set of sustainability attributes of floatovoltaic tech-
nology. In this context, the philosophical research effort and simulation method as technique will help to
advance the state-of-the-art philosophy in floatovoltaic technology characterisation by adding geographical
dimensions to an integrated assessment methodology based on a holistic geo-scientific framework and the
development of a new geomatical synthesis model.
While supporting the global green energy transformation narrative through environmental geo-engineering
endeavours, this thesis’s theoretical and modelling postulations have international relevance in smart energy
systems sustainability research. From an international perspective, this geographical investigation provides
insights into and augments the utility value of a proposed novel theoretical framework and geographical in-
formation tool set capable of providing an early, broad and informative perspective on the expected impact
and sustainability outcomes of floating photovoltaic projects in agricultural applications. While the research
investigation provides a new understanding of the behaviour around the assimilation of floatovoltaic tech-
nology, it also offers new geographical insights to help guide researchers to potential research areas that
would, in the future, support the sustainable development discourse. As regards the downstream academic
and practical utility of the research investigation, its outcomes and results offer significant value proposi-
tions in terms of support for sustainable development and local energy transformation, including the current
context of new environmental authorisations around reporting regulations, obligations around local environ-
mental offsets and, as part of South Africa’s National Climate Change Adaptation Strategy (DFFE, 2019a,b;
RSA, 2019a, 2020), requirements around carbon tax legislation. Moreover, the creation of new knowledge,
facilitated through the philosophical approach and the theoretical research solution postulated by this thesis,
makes meaningful and relevant contributions to the body of international knowledge on sustainable develop-
ment in floatovoltaic systems (Cuce et al., 2022; Prinsloo, 2020). The thesis offers valuable inputs in helping
to bridge critical knowledge gaps that constitute a range of multi-disciplinary technology unknowns in the
scientific-analytical quantification of sustainability attributes. This capability is significant given that these
unknowns currently creates some of the main barriers to regulatory-type approvals and sustainability-type
certifications for floatovoltaic project certification, licensing and deployment commissioning worldwide (Her-
nandez et al., 2019; Spencer and Barnes, 2020). With this sustainability policy framework, the thesis makes
novel multi-disciplinary contributions to the fields of the Geographical Information Sciences, Environmental
Geography and Data Sciences, and in manuscripts submitted for peer-reviewed publication in the Interna-
5
tional Journal of Sustainable Energy Technologies and Assessments (Prinsloo et al., 2021, 2022b). With
the publications providing further validation that the research is novel and innovative, this peer-reviewed re-
search is also beginning to attract international attention in terms of bridging critical knowledge gaps toward
predicting the feasibility, efficiency and sustainability of renewable energy applications in agriculture and of
advancing new simulation tools specific to floating photovoltaic systems (Kumar et al., 2022).
Overall, the thesis makes original academic contributions to the frontiers of sustainable development
by creating new knowledge constructs that fill critical knowledge and methodology gaps through geo-
informatical theory building. With geographical system science spearheading this approach, the research
uniquely addresses the analytical sustainability problem as an energy systems modelling problem caused
by a sustainability policy modelling deficiency that emanates from an inadequate sustainability policy. As
such, it interprets the energy system sustainability problem from a technology-driven behavioural geog-
raphy perspective, in terms of which it engages in a geo-environmental operations research approach to
help explain the real-world sustainability behaviour of floatovoltaics in a more holistic ecosystem context.
Note that in this context, the thesis advances the novel integrated Technical/Energy, Environmental and
Economic sustainability conceptualisation for the sustainability of floatovoltaics. This conceptualisation is
acronymically referenced throughout the thesis by either the EEE, 3E, E3, or TEE as acronyms or terms
that define the same-identical Technical/Energy, Environmental and Economical sustainability integration
conceptualisation for floating photovoltaics posited by the thesis. In this context, further note that the the-
sis’s contributions provide a foundational theoretical and framework principal for an integrated geodynamical
modelling methodology that simultaneously models the concurrent operational behaviour, performance re-
sponses, impact effects, and sustainability characteristics of floating photovoltaic projects under the same
hybrid theoretical sustainability umbrella theory and framework. As such, the harmoniously related ter-
minology, including theoretical -sustainability characterisation, -performance characterisation, -behaviour
characterisation, -impact characterisation, -modelling characterisation, -geodynamical characterisation, and
-response characterisation, as well as related terms, are used interchangeably throughout the thesis to suit
the discussion context or the application of the posited theoretical framework.
Following the above context, the first chapter of this thesis introduces the geographical research topic
and the investigative research process directed towards resolving an identified real-world geographical prob-
lem around floating photovoltaic project assessments. Accordingly, Section 1.1 introduces the study topic
and research rationale around the defined task while providing a conceptual synopsis of the research project
as a logical basis for the study topic and project. Next, Section 1.2 presents the research problem statement
and the purpose of the study towards an active research postulate. Then, Section 1.3 details the research
design process and presents a graphical layout of the research process in line with the research questions,
research aim, and research objectives. Next, Section 1.4 explores the importance of the research, while
Section 1.5 defines the study terminology on which the research project is based and lists the set of ax-
iomatic assumptions. Finally, Section 1.6 outlines the thesis chapters through a condensed summary of
each chapter and the results of this thesis.
1.1. Research Rationale
The International Renewable Energy Agency (IRENA) claims that the energy sector contributes nearly two-
thirds of global greenhouse gas (GHG) emissions. In contrast, the transformation of the worldwide energy
system focuses on the adoption of renewable energy technologies (RETs) and improved energy-efficiency
strategies that could result in at least a 90% reduction in energy-related carbon dioxide (CO2) emissions
(IRENA, 2019). To this end, floating solar photovoltaics is an up-and-coming and cost-competitive tech-
nology for sustainable power supply and green energy solutions to support sustainable farming initiatives
6
locally. However, few studies are still available to assess the sustainability and potential broad-spectrum
impact effects of the interfacing of this photovoltaic phenotype with its environment (Armstrong et al., 2020;
Gadzanku et al., 2021b; Kumar et al., 2021; World Bank, 2019b). Planning for implementing such newly
invented sustainable energy systems makes understanding the fundamentals of sustainability crucial, espe-
cially within the context of the energy business operations and markets of the modern agricultural electrical
power industry. This view is becoming more valuable for certain aspects of environmental science and sus-
tainability because energy is a versatile and pervasive commodity and one of the most important resources
for achieving sustainable economic development and improving the quality of life in our modern society (UN,
2020b; UNESCO, 2020). Energy generation is a scalable function that can power anything from handheld
devices to farming facilities. Electricity can be produced by centralised sources or through decentralised
energy resources to achieve the dynamic supply/demand balance required at the delivery site. While con-
ventional solar photovoltaic (PV) installations often come with the burden of valuable land requirements (a
premium commodity in food production), the installation classification in Figure 1.1 shows the development
of various solar PV installation types to overcome land limitations (Dellosa and Palconit, 2022). This system
may even include different styles of hybrid-type floating solar photovoltaic configurations in water bodies
(Solomin et al., 2021).
Figure 1.1: Floating PV technology as a unique solar photovoltaic energy installation option and technology type, with
equal priority and significance in left-to-right order (source: Sahu et al. (2016), page 2).
In the agricultural context, the climate economy raises concerns about the carbon footprint offset impli-
cations of sacrificing fertile agricultural land, while the water economy requires a reduction in operational
water withdrawals for electricity-generating technologies. This modern trend has inspired clean energy
technology developments such as floating photovoltaic technology (FPV or FPVT) and its water-submerged
photovoltaic (SPV) variant, generally classified as water-based photovoltaics (WPV). As a subfield of pho-
tovoltaics, floatovoltaitry aims to deliver a reliable and sufficient supply to meet the power consumption de-
mands on a minute-by-minute basis, operating at the highest level of optimised technical efficiency and with
the least adverse environmental impact effects, while generating clean energy at the lowest possible cost
and price. Moreover, floating-type solar PV installations offer numerous advantages compared to overland
solar systems, such as land-based photovoltaics (LPV) or ground-mounted photovoltaics (GPV or GMPV)
(World Bank, 2019b). As such, FPV offers a range of superior performance and impact benefits, such as
improved power generation efficiency, cooler operating temperatures, natural resource preservation, fewer
obstacles to block sunlight and the conservation of valuable agricultural land for food production, which is
often challenging to quantify (Gadzanku et al., 2021b; Hernandez et al., 2019; Spencer and Barnes, 2020).
As the global threat of climate change continues to grow worldwide, limited energy infrastructure and
fossil-fuel-driven power generation facilities are struggling to keep up with the increased demand for en-
ergy and the associated requirements of sustainability farming (FAO, 2021; OECD, 2001). This context
emphasises the role of agriculture in ensuring natural resource preservation, bio-security and food security
in developing countries. Furthermore, it highlights the importance of environmental and natural resources
in energy production to help alleviate the problem of agricultural land security and sustainable food produc-
tion by balancing the intricate and dynamic interlinkages among the agricultural water-energy-food (WEF)
nexus and even the agricultural-type water-energy-land-food (WELF) nexus (Gamarra and Ronk, 2019;
7
Pawlak and Kołodziejczak, 2020). Faced with the risk of a limited supply of natural resources, governments
and concerned authorities are steering away from over-reliance on fossil fuels and are consequently pri-
oritising the use of distributed renewable energy resources (DOE, 2018; World Bank, 2013). In promoting
sustainable rural agricultural development, the goal is to ensure that future agricultural energy systems are
environmentally clean and leave a limited carbon and environmental footprint on the earth’s natural systems.
While scientific concerns about the impact effects of environmental change and energy development
have grown markedly (World Bank, 2019b), scientists are making increasingly significant contributions to
new technology developments (Allen and Prinsloo, 2018; SPG Solar, 2010). This contribution includes
understanding the impact effects of such technologies on environmental change (Armstrong et al., 2020;
Cagle et al., 2020; Prinsloo, 2019), especially in the context of the Glasgow COP26 climate conference,
when it announced that South Africa would receive billions of dollars to terminate the country’s reliance on
coal (UN, 2021). In general, such applied research efforts strongly focus on technology ecosystem dynam-
ics and biodiversity impacts by human-induced and energy-system-induced climate change and earth/water
surface processes (National Research Council, 1997). Geography’s contributions to science and society are
thus helping to solve relevant real-world issues emanating from the transformation of rural landscapes and
agricultural land use, mainly where geographical perspectives and techniques contribute to understanding
critical issues in environmental science. The National Science Foundation (NSF) in the USA considers
geographical approaches a scientific means to further strengthen scientific contributions toward mitigating
climate change and resolving critical water-conscious energy-development challenges (NSF, 2011).
From a local development perspective of sustainable energy technology, our country is rapidly transition-
ing toward a more sustainable energy future, where the development of new decarbonisation technologies
such as photovoltaics (PV) expects to facilitate a green energy transformation in various applications and
industrial sectors (NDP, 2019; NERSA, 2021). Regarding agriculturally-based electrical power systems,
the recent invention of floating solar power generation technology has created substantial interest since its
discovery in the agricultural sector only a few years ago (Allen and Prinsloo, 2018). Even in South Africa,
space-saving floating solar PV technology is busy attracting the attention of wine farmers in the Cape
Winelands region because of the unsustainable methods of uprooting and clearing historical vineyards to
set up solar energy units on vineyard land. Worldwide, the rapid diffusion of this exciting new floating solar
invention has created a localised need for new data science techniques and modern digital information tech-
nology tools to help solve real-world geographical problems concerning sustainability analysis appropriate
to the novel impact effects of floatovoltaic technology (Hernandez et al., 2020; Prinsloo, 2019).
The Boplaas Estate in the Franschhoek wine valley recently installed the first on-farm application of a
floating solar photovoltaic system in South Africa. Figure 1.2 shows the floating solar facade for this water-
based floating PV system, classified as a decentralised energy system integrated with a ground-mounted
solar photovoltaic energy facility (refer to Figure 1.1).
Figure 1.2: Versus its ground-based counterpart, floatovoltaics installed on cooler water space to increase power
production and save fertile farmland (source: New Southern Energy (2021), page 2).
8
Remarkably, this cutting-edge floatovoltaic technology prototype system has reduced the Boplaas wine
farm’s carbon footprint by 50% during the first year of installation (Pritchard et al., 2019) to impact the local
environmental conservation efforts positively. In addition, the technology saves fertile agricultural land that
will be used more profitably as a co-benefit for fruit and wine production (Prinsloo, 2017b, 2019). Further
to the South African agricultural sector, floatovoltaic technologies are also drawing increasing worldwide
attention as a novel type of agri-photovoltaic or agro-renewable clean energy class (Allen and Prinsloo,
2018; Pringle et al., 2017; Prinsloo et al., 2021).
An underlying real-world problem addressed by this study is that agricultural floatovoltaic systems (il-
lustrated in Figure 1.2) offer a complex and diversified range of performance benefits and impact effects,
primarily because of the suspension of the FPV system in more excellent and environmentally-sensitive
aquatic ecosystems. The water reservoir is causing localised hydroclimatic interferences on the operation
of the energy system through causal-type interactions with the aquatic environment. Standardised metrics
for ground-mounted analytical photovoltaic energy system sustainability assessments do not adequately ac-
count for these side effects (Cagle et al., 2020; Gadzanku et al., 2021b; Spencer and Barnes, 2020). It is im-
portant to quantify these effects because the underlying aquatic environment is known to support increased
power generation benefits in floatovoltaic technology. The resulting and associated financial positives and
the environmental co-benefits also positively impact the feasibility of a planned floating solar project (Kumar
et al., 2021; World Bank, 2019b). While the complexity and diversity of environmental impacts for floating
PV systems are particular areas of uncertainty (Gadzanku et al., 2021b; Prinsloo, 2020), such information
variability and critical knowledge gaps in the floating photovoltaic research field offer real-world problems
(Gadzanku et al., 2021b; Ranjbaran et al., 2019). From a regulatory perspective, the critical knowledge
gaps and challenges around the scientific-analytical quantification of such "technology unknowns” are the
main barriers to floatovoltaic licensing and deployment worldwide (Haohui et al., 2020; Gadzanku et al.,
2021a; Spencer and Barnes, 2020).
The energy efficiency, the financial risk benefits and the uncertainties around its environmental impact
effects and environmental economics viability threaten the deployment of floatovoltaic systems and regu-
latory approvals of such projects on at least two fronts, namely the environmental front (Hernandez et al.,
2019; Spencer and Barnes, 2020) and the financial front (Cohen and Hogan, 2018; RSA, 2019a). These
risks manifest because floatovoltaic project incentives, environmental taxes and project licensing authori-
sations are all driven globally by evidence-based processes. As such, the associated regulators demand
compelling scientific evidence before granting project incentives and licences, approving the systems in-
stallations and granting permission-to-operate certificates (DEA, 2014; RSA, 2019a). The interdependent
techno-enviro-economic risk profile, which can virtually not be quantified or profiled by any recent-day geo-
graphical tool, compromises project licensing (Prinsloo, 2020). Therefore, most geographical models fail to
provide an adequate integrated and contextual foundation for solving such pressing regulatory requirement
challenges in the sustainability landscape of floatovoltaic technology (Hernandez et al., 2020; Prinsloo,
2019, 2020). Such an analytical tooling deficiency prevails mainly because, when compared to ground-
mounted PV systems (GPV), the hydro-embedded floating PV system interfaces directly with an underlying
aquatic ecosystem to provide a potential range of superior technical benefits, natural capital benefits and en-
vironmental co-benefits (Randle-Boggis et al., 2020; Spencer and Barnes, 2020; World Bank, 2019b). With
an increasing interest in floating solar PV systems worldwide, robust validation of the suggested co-benefits
and impacts of floating PV on the nexus of water, energy, land and food systems is still lacking (Gadzanku
et al., 2021a). These knowledge gaps and related uncertainties constrain fully-functional scientific assess-
ments as a mandatory requirement for evidence-based regulatory project permissions prescribed by current
environmental legislation.
The thesis proposes a new analytics framework to serve as program logic for a predictive sustainabil-
ity analysis model designed around the digital system dynamics modelling of the floating solar technology
9
operational ecosystem. The framework also serves as a governing mechanism in a computer simulation
methodology, capable of performing analytical inferences in evaluation procedures in experimental systems.
The methodology and simulation model further drive a multiscale decision-support profile mapping display
supporting project decision-making around the water-energy-land-food nexus resources. In the context of
floatovoltaic technology proliferation within the South African landscape, the study is framed in the broad
paradigm of sustainable development and scientifically directed at the intersection of the environmental-,
energy- and data-science fields. As such, research on floating PV technology, as in Figure 1.2, aims to con-
tribute to the sustainable development discourse toward the future development of analytical assessment
tools to support the regularisation of impact assessment approvals for local floating solar projects. This
research objective falls under the scope and purpose of the study, which aims to estimate the impact per-
formance of floatovoltaic systems installed on open-air water reservoirs around agricultural areas in South
Africa. The study especially considers those areas where agronomic cultivation intensity is high and where
there are dams or reservoirs with water for floating solar panels. Furthermore, it contemplates that apply-
ing the assessment technology proposed in this thesis would be of more interest to geographical areas in
water-rich regions where water bodies would be more available than land on which to position floating pho-
tovoltaic units. These areas, for example, include the fruit and wine-growing regions of the Western Cape
and other similar agricultural production areas within the South African agricultural landscape (refer to the
context of Research Assumption 3 described later in Section 1.5).
That concludes the introduction to the study topic and study rationale. The following section describes
the overall research process and the steps contributing to solving those mentioned real-world geographical
challenges. It narrates and details the problem statement and the purpose of the study towards addressing
the real-world problem and challenges around geographical decision support for planning a project evalua-
tion model in the context of photovoltaic planning.
1.2. Problem Statement and Purpose of Study
The solution development process begins with identifying a real-world environmental challenge or research
problem and whether modelling results could inform improved decision-related solutions to the identified
problem (EPA, 2022). This section, therefore, discusses the problem statement and purpose of the study
within the context of the research rationale of Section 1.1. It provides a logical basis for the topic of this geo-
graphical study and presents a discourse on an assessment of floating solar technology and the theoretical
frameworks related to the identified knowledge gap. Finally, it provides valid arguments supporting the con-
ceptual rationale and importance of the investigation, especially regarding its contribution to the discourse
on sustainable development. In so doing, it helps resolve a practical, real-world geographical problem for
future planned FPV installations locally and abroad.
Section 1.2.1 presents the problem statement in the context of the development planning domain of a
floatovoltaic energy system. Particular attention is directed towards the challenges around the complexity
of the geospatial system in respect of environmental scoping, techno-economic analysis and contextualised
decision support for floating solar system planning. This discussion lays the foundation for defining the prob-
lem statement formulated in Section 1.2.2. Section 1.2.3 summarises critical factors from the background
information and supporting arguments that led the researcher to the meaning behind this study.
1.2.1. Problem identification and analysis
The literature review in Chapter 2 brainstorms the research problem of FPV energy technology sustainabil-
ity as part of project planning challenges from a multi-level sustainable development perspective. Against
10
the background of agricultural-type floatovoltaic energy system deployment, this section highlights the busi-
ness information systems analysis that deconstructed the real-world problem of floatovoltaic sustainability
assessment into organisational-level, process-level and activity-level challenges.
In shortly summarising the literature study of Chapter 2, the problem identification analysis runs in
tandem with the scholarly review. It starts with a brief discussion on the performance modelling prob-
lems and challenges in Section 1.2.1.1 before considering the challenges associated with floating PV sus-
tainability in a planning and assessment context into a three-tiered hierarchy. This hierarchy considers
the statutory/organisational-level challenges, process-level challenges and activity-level challenges, as de-
picted in Figure 1.3, before considering the purpose of the study towards a new theory, methodology or
model for floating PV analysis in the rest of the thesis.
Regulatory process-level
problems and challenges
Organisational-level prob-
lems and challenges
Performance modelling
problems and challenges
Problem identification
Activity-level problems and challenges
Problem statement
Purpose of the study
Figure 1.3: Problem identification, breaking down the research problem around floating PV planning and assessment
challenges to define the problem statement and purpose of the study (source: author).
Within the context of Figure 1.3, Section 1.2.1.1 narrates performance modelling problems and chal-
lenges, while Section 1.2.1.2 informs of the strategic regulatory-level challenges and problems. Process-
level challenges and issues associated with floating PV sustainability valuations in project planning exer-
cises, as detailed in Section 1.2.1.3, while Section 1.2.1.4 includes an extensive overview of the activity-
level challenges and problems associated with the design-stage analysis and assessment of sustainability
in floating PV project plans.
1.2.1.1. Performance modelling problems and challenges
By brainstorming the research problem from a performance modelling perspective, this section provides
insight into the root-cause analytical challenges around floating PV sustainability assessments that emerged
within the activity-level tier of the three-tier problem hierarchy discussed in Chapter 2. As such, it offers
a hands-on description of the practical, real-world observations in the actual functioning of floating PV
installation projects documented by scientists.
In this context, agricultural floatovoltaics represents an alternative energy technology solution defined
by a novel type of solar PV technology installation type (Figure 1.1) in which solar PV panels are sited
directly on open-water surfaces of irrigation dams and reservoirs of farming establishments (Figure 1.2)
(Allen and Prinsloo, 2018; Esmaeili Shayan and Hojati, 2021). While various innovative over-water float-
ing PV technology configurations offer a conservation-friendly solution to ensure energy independence in
a more efficient low-carbon energy generation technology (Gorjian et al., 2021; Kjeldstad et al., 2022), it
presents valuable opportunities to deliver decentralised power generation to remote and rural farming ar-
eas that can change the energy landscape (NREL, 2019a; World Bank, 2019b). Further offering improved
11
decarbonisation potential with efficient power generation capacity (Luderer et al., 2019; Moghadam et al.,
2021), FPV technology offers valuable decarbonisation opportunities around urban utilities and peri-urban
facilities such as hydro power, water processing stations, wastewater treatment plants and desalination
reservoirs (da Costa and da Silva, 2021; Fitch, 2020; TNO, 2021). At the same time, floatovoltaic technol-
ogy can transform several square metres of previously unused water surface area into an income-producing,
revenue-generating space thanks to the income generated from floating solar energy farming. These as-
pects make environmental sustainability an indispensable part of a project due diligence study.
New energy developers worldwide are further interested in the sustainability economics and resource
economics of the technology, as farmers and investors are showing interest in investing in climate-smart
floating PV power plants as yield efficient, water-preservation, land-saving, and carbon emission reduction-
type energy generation technology (IFC, 2020; Meyerhof and Prinsloo, 2019; World Bank, 2019b). Offer-
ing a range of value propositions as an alternative to over-land PV, especially where water reservoirs are
available and land is scarce and expensive (Allen and Prinsloo, 2018; Pritchard et al., 2019), floatovoltaic
technology helps to establish the so-called concept of sustainable solar farming more firmly (Gorjian et al.,
2022a; Kuschke and Cassim, 2019). Furthermore, in terms of the international Conservation Champions
Initiative in support of local wineries in SA (WWF, 2021), FPV technology’s broad spectrum of land-surface
saving, water preservation, and carbon emission reduction properties combined with its positive environ-
mental offset and water-energy-land-food (WELF) footprint attributes make floatovoltaics one of the leading
enabling technologies of sustainable farming in the wine-producing regions of South Africa.
From an economic and environmental impact assessment perspective, regulatory requirements hamper
floating PV deployments despite the remarkable sustainability attributions due to the lack of geographical
tools to scientifically quantify the diverse spectrum of interrelated sustainability attributes of floatovoltaic
systems. These are deeply problematic challenges causing particular concerns in the industry, as de-
lays with project deployment licensing and investment adjudications hamper FPV deployment Barbuscia
(2018); ELaw (2022); Gadzanku et al. (2021b). Furthermore, assessing floatovoltaic technology sustain-
ability impacts emerging from energy, environmental and economic operations perspectives face substantial
challenges regarding the lack of tool availability. At the same time, the availability of data on the field perfor-
mance of floatovoltaic technology development projects is also deficient due to the novelty of the technology
and the immaturity of the techniques and processes to predict system performances and impact effects (Ku-
mar et al., 2022; NREL, 2019a).
With the thesis campaigning for a geographical solution to solve this real-world geographical prob-
lem, the academic merits of the research problem around the scientific-theoretic characterisation of the
floatovoltaic sustainability complex are interpreted from a practical geo-environmental operations research
perspective. Within the context of the renewable energy development decision space, there exists a need
for improved alignment between climate change and sustainable development goals. Scientists have de-
veloped conventional technical performance modelling frameworks and toolsets for assessing overland PV
performances, but these models focus mainly on standardised technical procedures and performance met-
rics for land-based or ground-mounted photovoltaic installations (Sandia, 2020). As such, existing technical
framework principles behind these models do not adequately account for the distinctly diverse and broader
spectrum of floatovoltaic technology deliverables (Cagle et al., 2022; Hernandez et al., 2020).
For example, conventional technical performance models mainly focus on technical and techno-economic
performances; as such do not fully account for the extended range of environmentally-related impact-effect-
positives, resource-use-efficiencies, resource economics, and climate-economic benefits offered by water-
surface mounted floating photovoltaic installations (Prinsloo, 2020). Land-surface-saving floatovoltaics offer
valuable synergistic benefits in using under-utilised open water spaces to generate sustainable energy
and additional revenue from solar power generation while concomitantly providing land-surface-reservation
benefits. This high-value agricultural land surfaces reserve fertile agricultural land for farming applications
12
(Cagle et al., 2020; Spencer and Barnes, 2020).
Overall, field experiments by scientific scholars indicate that floating PV technology offers a distinctively
different nuance of impact effects while delivering a broader spectrum of performance and co-benefits. This
includes an array of technical performance benefits (Altegoer et al., 2021; Kjeldstad et al., 2022), environ-
mental offset effects (Allen and Prinsloo, 2018; Prinsloo, 2019), economic impact positives (Dizier, 2018;
Micheli, 2021; Makhija et al., 2021). Prinsloo (2015) identified and recorded that many of these benefits
and co-benefits of floatovoltaics show remarkable concurrence WELF resource preservation benefits, an
aspect other researchers have since also recorded as a quantitative research challenge (Armstrong et al.,
2020; Gadzanku et al., 2021b,a; Prinsloo, 2019). Conventional technically-oriented analytical PV perfor-
mance assessment tradecraft (frameworks, models or indicator metrics) do not adequately account for this
kaleidoscope of performance and impact effects (Armstrong et al., 2020; Gadzanku et al., 2021b). Although
incremental modelling improvements find it difficult to characterise the full spectrum of sustainability impera-
tives of floatovoltaic installations, recent research efforts into specific FPV applications selectively developed
certain model elements to capture individual performance/impact effects in technical and techno-economic
floating PV performance models on an ad hoc requirements basis (Eyring and Kittner, 2022; Majumder
et al., 2021; Refaai et al., 2022; Tina et al., 2021a).
The mapping of the peculiar performance discrepancies in the mind map of Figure A.1 leads the thesis
to the conclusion that there exist structural imbalances in the current modelling approaches of GPV systems
that cannot be addressed by simply haphazardly adding model parts (refer Sections 2.6 and 3.5). Despite
constraints with model framework flexibility, there are deep-seated tool model vulnerabilities in conventional
frameworks wherein insufficient consideration for environmental impacts causes serious drawbacks. This
deficiency means that, under the flexibility constraints, ad hoc model improvements to the conventional
analytical framework (Figure 3.9) would be unable to alleviate the underpinning theoretical problem. Instead,
it should theoretically address systemic imperatives at the meta-level to fully accommodate the diverse
modelling of the FPV sustainability ecosystem.
The assessment of the technology unknowns is a problem that needs addressing at the theoretical level,
ideally by posing an improved theory that can help to overcome the fundamental issue around the quantifi-
cation of what became known among scientists as the "technology unknowns” of floatovoltaic technologies
(Banik and Sengupta, 2021; Gadzanku et al., 2021b). The following three sections narrate the implicated
problems and challenges caused by the performance modelling deficiencies discussed above, as well as
the inadequacy of performance modelling to deal with the assessment of sustainability required to support
sustainable development in terms of strategic, tactical and activity level objectives.
1.2.1.2. Strategic regulatory level challenges and problems
South Africa is considered one of the world’s most carbon-intensive economies. At the same time, the
country is fortunate to have a sound legal system and comprehensive environmental legislation aimed at
achieving sustainable development, effective environmental management, and conserving the country’s rich
pool of natural resources (IFC, 2021). South African environmental law and policy, in particular, reinforces
the National Development Plan (NDP) and guides the implementation of the Sustainable Development
Goals (SDGs), focusing on virtually all of the UN SDGs, but with particular emphasis on eliminating poverty
(SDG 1), providing energy security (SDG 7), and reducing inequalities (SDG 10) (System Dynamics Society,
2021).
In support of South Africa’s legislative and policy frameworks, the SA Department of Forestry, Fisheries
and Environmental Affairs (DFFE) mandate working towards a fully-integrated and adequate spatial data
infrastructure. To provide a fully-fledged online service to environmental practitioners and project develop-
ers, the DEA has established the South African Environment Geographic Information System (EGIS) on
13
the ESRI platform (EGIS, 2022). The focus of the EGIS platform is on the monitoring and management of
the environment; it prioritises the digitalisation of environmental data on the electronic spectrum and the
processing of impact assessments in digital formats. However, with the DEA’s aim to manage integrated
environmental management processes on a web-based platform, the EGIS online processing requirements
create an immediate problem for impact practitioners. As a result, the government has created a pressing
need to develop new geographical assessment tools capable of digitalising environmental impact assess-
ments (DFFE, 2018). As proposed in this research project, the thesis aims to overcome this problem by
digitalising project assessments in preparation for the future online filing of new applications for approvals,
endorsements, and licensing regarding planned renewable energy projects.
Further to the context of the geographical, legislative and technological requirements is considered to
be of utmost importance in the current research, especially given the urgency of the problem of meeting
the requirements of South Africa’s newly-promulgated Carbon Tax Act (Act No. 15 of 2019) (RSA, 2019a).
As this Act is to be fully implemented by 2022, the stipulations of the Act significantly sharpen the focus
on planned project credentials and environmental reporting requirements. This requirement intensifies the
need for scientifically-based environmental approvals subject to sound scientific evidence (predictions or
measurements). The Carbon Tax Act further dovetails with the Environmental Impact Assessment (EIA) re-
quirement to authorise South African solar farming projects during the conceptualisation and design phases
of a project. Here, the DFFE (2015) also requires project developers to submit scientifically-founded a-priori
analytical measures on environmental offset performance predictions during the project design and planning
phases as part of the new Environmental Offset Policy.
Therefore, from a strategic regulatory level challenge perspective, authorities demand compelling scien-
tific evidence as a prerequisite for reporting on new directives such as environmental offset, climate change
and carbon taxation. However, in the case of floatovoltaic technologies, impact practitioners do not have the
proper scientific tools to affect such analysis or to submit scientifically founded a-priori predictions required
for new energy sustainable development reporting.
1.2.1.3. Tactical process level challenges and problems
What further intensified the need to prioritise the research is the urgency and relevance of the study to-
wards reinforcing sustainable farming practices in the local agricultural industry (Meyerhof and Prinsloo,
2019; Pritchard et al., 2019). Since the environmental impacts of water-borne floating solar systems differ
substantially from the impact effects of ground-mounted photovoltaic systems, solar project developers in
South Africa are facing challenges with the regularisation of analytical methodologies, computer forecasting
models and assessment methodologies to streamline floatovoltaic regulatory approvals for the authorities
granting permits for local projects (New Southern Energy, 2021).
In the category of novel solar farming solutions, floatovoltaic technology as a water-based photovoltaic
system (WPVS or FPV) is a novel upcoming approach toward achieving more sustainable energy devel-
opment targets as part of the larger sustainability farming narrative for agriculture in South Africa (Gaz-
dowicz, 2019; Pritchard et al., 2019). Floatovoltaic technology positions itself to support the conservation
goals of the WWF (2021) "Conservation Champion” initiative for South African wineries. For instance, it
has spotlighted protecting conservation-worthy agricultural farmland, reducing water usage and implement-
ing energy-efficient solutions. In this manner, the development of agricultural floating solar technology
has brought new meaning to sustainable farming in the local agricultural industry, as the technology of-
fers an extended range of benefits in terms of natural resource conservation. However, project planning
is becoming increasingly problematical since this technology is starting to draw attention away from the
agricultural industry in such a context. Moreover, the technology’s interdependent, multi-dimensional and
trans-disciplinary attributes intensify the FPV project-planning problem. Of particular concern are FPV tech-
14
nology’s exploits in surface-drinking-water beneficiation and its potential land-saving endeavours whereby
valuable agricultural land needed for agriculture is spared under local conservation legislation, licensing and
levies.
These drivers of the new tendencies and developments in FPV are formalising an urgent requirement
for the credible scientific quantification of floatovoltaic technology benefits in at least three areas, namely
agricultural land-saving, water surface transformation, and irrigation water preservation, the last-mentioned
three requiring quality improvements (Cagle et al., 2020; de Lima et al., 2021a; Mathijssen et al., 2020).
In these three areas, floatovoltaic systems deliver a predictable range of environmental offset co-benefits,
together with quantifiable sustainability impact positives (Prinsloo, 2019, 2020) in support of the modern
mitigation of climate change and opportunities for climate adaptations (FAO, 2021; World Bank, 2019a).
Such sustainability benefits have become the most important in the agricultural economy. Hence, they
are valuable to modern-day agricultural farming narratives and modern approaches such as sustainability,
biodynamic, organic, regenerative, and alternative agricultural paradigms.
The above discourse around energy sustainability emphasises the need for dedicated geographical tools
suitable for computer-simulated analyses and the gauging of floatovoltaic impact performances as part of
environmental risk analysis during the project’s design stage. The goal to accurately predict the added
environmental and sustainability benefits of floatovoltaic technology clearly emphasises that present-day
geographical tools are failing to provide an adequate contextual foundation for solving pressing present-
day challenges in the sustainability landscape of floatovoltaic technology (Hernandez et al., 2019; Prinsloo,
2019). Existing tools fall short mainly in capturing information on the broad spectrum of environmental
benefits and co-benefits and the ability to account for financial and economic aspects such as carbon
taxation and incentive programmes (Cagle et al., 2020; Prinsloo, 2019). This aspect is even more of the
case in the South African agricultural context, where little field data is currently available from field tests
around floating solar impact performances, simply because the first local agricultural floatovoltaic system
has only recently been installed (New Southern Energy, 2021).
With the regularisation of environmental scoping and environmental impact assessments for prospective
geographical FPV installation sites in mind, the study moves towards generating new geographical knowl-
edge and re-contextualising knowledge to fill the existing knowledge gaps required to support the prolifera-
tion of floating solar ecosystems. It advances state-of-the-art technologies and creates positive downstream
research opportunities to gain geographical insight into the impact of FPV, and thus help shape the technol-
ogy planning and environmental authorisations around the FSPV discourse. In the South African landscape,
the digital synthesis model integrates systems-wide photovoltaic, environmental, and financial-performance
properties into an enterprise-wide floatovoltaic ecosystem. The goal is to establish a fully digital experi-
mental geoinformatics tool to predict and empirically study the geodynamic performance characteristics,
technical potential and environmental sustainability properties of an FPV system in local agricultural appli-
cations.
The chief problem is that environmental project authorisations have prerequisite integrated assessment
requirements on a tactical level. Being part of project performance and impact effects for new energy
development projects, impact practitioners do not have the proper modelling tools to accurately forecast the
performance features or impact effects of floatovoltaic technologies.
1.2.1.4. Activity level challenges and problems
The main problem with floatovoltaic systems is that they are suspended in environmentally sensitive aquatic
ecosystems, leaving contrasting technology impact differences between over-water floating PVs and con-
ventional over-land PVs. Accomplished by implication means that standardised metrics for ground-mounted
PV projects do not adequately account for the broad spectrum of impacts on aquatic habitats, agronomic
15
impacts, resource-use efficiencies and impact-effect positives as characteristic features of floating solar
technology. These critical knowledge gaps are barriers to the planning and deployment of climate-solving
floating PVs as a renewable energy resource.
As the adoption rate for floating solar systems is destined to increase in South Africa (Meyerhof and
Prinsloo, 2019; World Bank, 2019b), a corresponding set of planning requirements for floatovoltaic tech-
nology projects needs to grow in the South African and international agricultural landscapes (DFFE, 2015;
Norton Rose Fulbright, 2021; Prinsloo, 2019). An international study supports this viewpoint in reporting on
the floatovoltaic landscape survey scenarios (Cohen and Hogan, 2018; Wood Mackenzie, 2019), drawing
attention to the need to formulate dedicated new floating solar project plans and sustainability assessment
frameworks to accommodate the increases in this respect in the future. While official EIA processes are
evidence-based in practice, impact practitioners are trying to use traditional GPV analytical tools. However,
they are proving inadequate because they do not meet the comprehensive legal obligations. Moreover, the
GPV tools generally fall short in supporting the compilation of project scoping and the review reports, which
are supposed to document the full spectrum of the anticipated impact effects of a prospective project (RSA,
1998). This deficiency burdens the mandatory requirement for impact practitioners to scientifically predict
the expected impact effects of a new energy project throughout the project’s lifetime. Should make such
predictions during the development steps and the project’s conceptualisation phases.
Defined according to the US Energy Department, developing a new energy project is an iterative process
that plays out in five phases (USDoE, 2020). This five-step process is depicted in the framework presented
in Figure 1.4. It covers all aspects needed to inform the entire project development process towards com-
piling a comprehensive energy project plan before the installation approval. For a detailed definition of each
of the five development phases or steps in the project’s implementation process, as indicated in Figure 1.4,
the reader may refer to the definition of the Energy Project Development Process Framework as defined by
USDoE (2020).
Figure 1.4: Floatovoltaic energy project implementation steps to highlight the importance of analytical design predic-
tions in Steps 1 to 3 of the project planning process (source: USDoE (2020), page 3).
Regarding this thesis’s floatovoltaic project planning focus, the study’s problem statement mainly relates
to data processing and data analysis during the project planning phases (first three steps) in the project’s
lifecycle. This focus includes the design phases, namely the analytical-potential, mitigation-strategy and
refinement-decision phases at the commencement of the project in Figure 1.4. In this universal planning
framework, floatovoltaic project planning starts in Step 1 with the strategic analysis of the project’s techni-
cal viability, environmental accountability and financial feasibility. This step includes site selection, project
16
scaling, resource analysis, impact analysis and income analysis to ensure that the project provides an at-
tractive investment opportunity. Then, Step 2 focuses on analysing the regulatory approval considerations
towards obtaining the project permits, authorisations and legal licenses required to install and operate a re-
newable energy facility. Finally, moving to Step 3 of the development framework of the floatovoltaic project,
Figure 1.4, the associated activities include refining project design options through a multi-criteria decision-
support analysis to ensure meeting the project objectives with optimal investment returns in terms of the
sustainability criteria of the project (CIF, 2018). As such, each step or phase in project planning categorises
the key activities and decision points in the development framework of the energy project. At the same
time, it highlights the required planning decisions around FPV sustainability, in terms of expertise, project
memory and experience, towards viable project outcomes in terms of secure project investments and the
sustainable design-option goals set for the envisaged FPV energy project (Springer, 2013; USDoE, 2020).
In terms of sustainability performance assessment, impact profiling and climate risk analysis during
the conceptual design and analysis phase of the solar floatovoltaic technology project is crucial. As pre-
sented in the pipeline in Figure 1.4, geographical planning models are required in the local context to assist
decision-makers to optimise their decisions around sustainable floating solar operations and investment
planning (Allen and Prinsloo, 2018; Meyerhof and Prinsloo, 2019). Equally important, given the relevant
regulatory regimes of the day, the project-planning decisions around project validation regimes are simi-
larly crucial (Cohen and Hogan, 2018). In this context, Section 1.1 accentuates the disjunctures in terms
of the sustainability criteria and the analytical frameworks between on-ground or on-land photovoltaic sys-
tems and floating or on-water photovoltaic systems (Cox, 2019a; Rosa-Clot et al., 2017). As opposed to
the case with traditional GPVs, the disjunctures and imperfections with conventional tooling in assessing
FPV sustainability performance and its impacts are causing serious sustainability assessment challenges.
These modelling imperfections are proving to be among some of the greatest project-planning hindrances
and regulatory obstacles to investors in the projects, environmentalists and impact practitioners worldwide
(Cuce et al., 2022; Hernandez et al., 2019; Prinsloo, 2019; Spencer and Barnes, 2020).
Brainstorming the challenges of floatovoltaic sustainability assessments has led to the compilation in
the thesis of the real-world floatovoltaic planning challenges in the mind map illustration of Figure A.1
(Appendix A). This mind map illustrated the complexity and diversity of the planning and sustainability
dimensions of a floating PV project in a climate-smart agricultural context. Following the guidelines on
the development, evaluation and application of environmentally-related models (EPA, 2009), the mind map
analyses and highlights the integrated roots of the floatovoltaic ecosystem’s web of interrelated sustainability
assessment aspects and complexities. Furthermore, as a technical framework to guide the development
of the model, the mind map further emphasises the demand, in terms of the purpose of the study, for a
philosophical systems-thinking approach to sustainability. While helping to identify the fundamental cause
and risk determinants in FPV project assessments, it portrays the decision/planning dimensions involved in
sustainable planning operations for floatovoltaic systems in a complex systems model (Prinsloo, 2020).
In addressing the planning challenges listed in the mind map of Figure A.1, the thesis aims to over-
come real-world problems associated with the modelling and analyses of the sustainability of floatovoltaic
systems. The goal is to effect optimal sustainability decision-making for floatovoltaic projects in the lo-
cal geographical planning environment. Furthermore, with the root cause of many real-world floatovoltaic
planning problems revolving around quantifying systems performances through sustainability indicators, the
goal is to support geo-sensitive sustainability decisions needed to help decision-makers overcome issues
associated with the planning factors inherent in floatovoltaic projects (and its sustainability assessments).
In this context, the mind map of Figure A.1 highlights complex interdependencies among the different
resource domains of the floating solar ecosystem. These cross-linked cascading effects influence the plan-
ning decisions around the floatovoltaic project, further complicating the already confusing decision-making
environment behind it. Accordingly, decision-makers require more profound knowledge and business intel-
17
ligence insights into the analytical aspects of techno-economic, environmental and techno-ecologic diag-
nostics. Relevant indicators are thus needed to enable decision-makers to analyse their options within a
multi-dimensional decision space.
Therefore, project planning requirements demand scientifically derived information on project sustain-
ability at the activity level, but the current models cannot accurately predict these attributions. Furthermore,
conventional PV performance assessment models cannot correctly deal with the complexities of floato-
voltaic operational analyses and are unable to provide comprehensive answers to the broader requirements
of sustainability analyses. Therefore, there is a need to conceptualise a new theoretical framework for FPV
energy technology that can deal with the analytical appraisal for floatovoltaic projects. The following section
defines the problem statement in preparation for the definitional purpose of the study.
1.2.2. Problem statement
The floating PV sustainability challenges depicted in the mind map of Figure A.1 show how floating PV
technology in agricultural settings has exposed critical theoretical knowledge gaps and methodological gaps
around the broad spectrum characterisation of the distinct and diverse set of sustainability properties of
floating PV energy systems.
The above requirements define apparent knowledge gaps and methodology gaps in the reporting re-
quirements concerning the sustainability features of the technology. Critical knowledge and methodological
gaps worldwide, cause deeply complex challenges, especially in failing to meet stricter regulatory require-
ments for the authorisation and licensing of such PV system installations amidst global climate change
imperatives (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). In this context, the problem statement
needs to state the root cause of the problem as the foundation for defining the study’s purpose.
The discussion has thus far put the spotlight on how floatovoltaic planning decisions towards improve-
ment in project sustainability are intrinsically complex because, in principle, they embrace interdependence
in an interdisciplinary arena. In the design-stage analytical context shown in the mind map of Figure A.1,
challenges arise from a wide range of inter-related knowledge fields covering multi-objective decision-
making around: (a) contrasting technical performances between FPV and GPV; (b) the complexity, frag-
mentation and (in)accessibility of information pertaining to the floatovoltaic habitat/ecosystem; (c) the in-
teractive dependencies between the decision-domain fields of the information ecosystem, and its variables
and indicators associated with the receiving environment; (d) contextual financial policy formulations as-
sociated with the application region; (e) prevailing legislation, legal parameters and programme variables
for that particular area of jurisdiction; (f) conceptual limitations in terms of the ability of decision-makers
to grasp the full impact of their systemic parameter choices and decision options; (g) limited time-horizon
data collection and analysis causing project decision-making with short-term static performance projections
instead of lifetime projections; and (h) the multiple objective choices that must be considered by project
developers and impact practitioners in terms of planning decisions around a range of project performances
within varying legislative government requirements.
The mind map of Figure A.1 shows that most of the above-mentioned real-world challenges or difficulties
in FPV project planning are, in fact, interlinked planning-decision complexities in the decision-making mech-
anism of the floatovoltaic ecosystem. Furthermore, they highlight the knowledge gap that exists around the
sustainability assessments and reporting requirements for floatovoltaic technology installations, which, to
a certain degree, represents a hindrance to the prospects for future growth in the diffusion of floatovoltaic
technology in South Africa’s agricultural industry. The dichotomy between FPV and GPV performances
emphasises this knowledge gap emphasises and the need for an investigative research study to formulate
a scientific research problem statement, which is defined as follows:
18
Problem Statement:Standardised metrics and models for ground-mounted PV projects do not ade-
quately account for floating PV technology’s complex range of distinct and diversified range of resource-use-
efficiencies and impact-effect-positives, mainly because existing analytical theories, framework paradigms
and computer toolsets fail to offer feasible and inclusive answers to the real-world analytical and modelling
problems associated with floatovoltaic system sustainability. On a strategic level, existing models need to
be extended as they cannot meet the broader requirements of sustainability assessments in a South African
agricultural context.
Addressing this problem is significant, especially with quantifying the technology unknowns of floato-
voltaics being a significant project assessment requirement during the project planning analysis or decision
advisement phases (World Bank, 2019b; IFC, 2020). The problem statement above underscores the asso-
ciated knowledge gaps in sustainability policy modelling as it calls for pioneering research towards creating
new knowledge that needs to be developed around scientifically-based quantitative environmental impact
assessments. Regarding sustainable development analyses and reporting requirements towards environ-
mental project authorisations, these gaps are causing deeply problematic challenges it causes worldwide in
meeting stricter regulatory requirements for the authorisation and licensing of such PV system installations
amidst global climate change imperatives (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). Given
the technical and analytical decision-support challenges associated with floatovoltaic planning around the
anticipated diffusion of floating solar technology in South Africa’s future, the main requirement is to deal with
geographical narratives to contextualise sensitivities and the complex interdisciplinary characteristics of the
dynamic behaviour of the floatovoltaic ecosystem in terms of information system dynamics, collaborative
responses and performance-impact analytics.
While this section defines the real-world geographical problem within the discourse of sustainable pho-
tovoltaic energy strategies and the context of multi-criteria decision analysis, the following section translates
this knowledge gap into a formulation of the purpose of the study as a proposed attempt to fill in the knowl-
edge and methodology gaps in the research, as identified above.
1.2.3. Purpose of the study
Whilst the need for new green energy infrastructure is clear, this section addresses the practical geograph-
ical challenges in the real-world problem of the sustainable development of due diligence in agricultural-
type floating solar technology from a study-purpose perspective. The study’s purpose highlights that this
research investigation intends to theoretically address the sustainability assessment challenges for floato-
voltaics technology. It considers options for the conceptualisation of an underpinning theoretical solution
that can simultaneously serve to address strategic, tactical and activity layer project planning challenges
currently experienced with the deployment of floatovoltaic technology.
The research problem statement in the previous section underscores the hierarchy of challenges to ad-
dress as part of the study’s purpose. It explicitly highlights that when floating PV technologies arrived on
the scene, they provided a variation of the photovoltaic theme that exposed critical theoretical knowledge
and methodological gaps towards the more complex theoretical characterisation of the technology (Arm-
strong et al., 2020; Cagle et al., 2020). With the more complex and diversified theoretical characterisation
requirements, especially when considered in a climate-smart agricultural context, these knowledge gaps
run across environmental, information, economic, and engineering sciences. It not only leaves scientists
confronted with a fundamental problem around the quantification of the above-mentioned "technology un-
knowns” of floatovoltaic technologies (Banik and Sengupta, 2021; Gadzanku et al., 2021b), but also with
a theoretical challenge to overcome these technology uncertainties through computer models, as these
are currently the main deployment barriers and regulatory approval barriers of floatovoltaic technologies
19
(Gadzanku et al., 2021a; Gorjian et al., 2022c). While the root cause of the research problem concerns the
theoretical characterisation of floatovoltaic technology in the context of a broader sustainability assessment
framework, it is imperative to resolve the sustainability assessment problem on a philosophic-theoretic level
as well as on a computer simulation logic level.
While floatovoltaic installation sustainability presents a next-level geographical problem that is poorly
understood at present, it underscores the need for relevant academic research around the contextualised
integrated sustainability assessments for agricultural-type floatovoltaic technologies, with specific reference
to the conceptualisation of a contextualised theoretical assessment and modelling framework. A new frame-
work logic further requires an underlying basis for developing a geospatial digital twin model or digital geo-
matics tooling intervention capable of replicating the operations of real-world floating solar PV systems. In
this context, the study’s purpose defines the solution specification from the problem-identification stage to
determine the right questions relevant to the decisions and to establish relevant modelling objectives. From
a requirements perspective, the first element of the requirement centres around impact assessment for floa-
tovoltaic installations. In this case, there is an immediate need to focus on an impact assessment solution to
determine the sustainability status of a project. Furthermore, on account of the growing status of FPV tech-
nology in South Africa, such a proposed solution should purposefully resolve geographical issues around
floating photovoltaics, especially in agricultural applications and complex sustainable food production en-
vironments. As a contextually sensitive priority, this need arises as a promise of a new FPV technology
offering a sustainable electric power generator for the local agricultural sector and enabling technologies to
support the progressive concepts of sustainable farming and regenerative agriculture under the 4IR drive.
In terms of functional behaviour analysis, studying and documenting the sustainability attributes of float-
ing PV technology requires a proper understanding of the operational aspects of the floating solar PV enter-
prise in agricultural production settings. This gap in understanding refers to the dynamics of the technical,
environmental and economic aspects operating within the floating PV ecosystem context. In this regard,
numerous scientific publications refer to the real-world techno-hydrological and spatial attributes of floato-
voltaic technologies, primarily investigations spanning an array of different contextual and climatic regimes
(see comprehensive summary later in Section 2.3.2). Thanks to the sustainability attributes of floatovoltaic
technology in terms of carbon-free energy delivery, high-efficiency rates, sustainable land conservation,
scarce resource preservation qualities, and water-saving benefits, the technology is bound to gain popular-
ity and trend in the agri-PV market for both natural and man-made water bodies in South Africa and on the
African continent (Gonzalez Sanchez et al., 2021; World Bank, 2019b).
With a growing local interest in the deployment of floatovoltaic installations to support the local agricul-
tural energy transition, the project planning research around floatovoltaic technology in this thesis is mainly
concerned with mitigating geographical assessment problems and challenges during the planning steps
of the project. In this respect, as mentioned earlier, situational analysis is a crucial goal of the analytical
design phase of the floatovoltaic planning process, with specific reference made to Steps 1 and 2 of the
floatovoltaic project development framework of Figure 1.4. Expanding on the execution activities during
these project-planning steps, as illustrated in Figure 1.5, underscores the need for ongoing research into
the layered sets of environmental impact effects of floatovoltaic system installations ahead of the approval
of a project for construction.
In striving to implement advanced technologies during these analytical planning steps and before the
installation approval, simulation tools and techniques are vital to characterising the performance qualities
of floating PV system installations through simulation-based project emulation enactment. Furthermore,
simulation modelling is further pivotal in the analyses of sustainability and business scenarios, project man-
agement trajectory planning and to help ensure that projects deliver on their planned sustainability and
performance outcomes (NREL, 2021; Suslov and Katalevsky, 2019). With the design analytics and regula-
tory approval considerations being the critical planning steps before the installation approval at the onset of
20
an energy pipeline project (Figure 1.5), the analytics mission of the thesis is to seek geospatial solutions to
help overcome complex planning-decision challenges.
Figure 1.5: Project planning steps and activities for new energy projects (source: USDoE (2020), pages 4 & 25)).
Fortunately, the digital 4IR Agriculture 4.0 paradigm contributes to the development of flexible geospatially-
driven "digital twin” modelling tools for the virtual commissioning of projects during the early stages of the
design and planning process (Clifton, 2022; ESRI, 2022; Maplesoft, 2022b). However, when it comes to
digital twinning design processes and practices, an accurate digital twin abstraction that replicates a real-
world floatovoltaic system needs to be developed from the ground up and based on the final specifications
of the real-life system (Jones et al., 2020). From a digital twin perspective, this thesis generally postulates
that should give the natural environmental and natural environmental aspects have greater prominence in
the designing and modelling of a digital twin floating PV system archetype. This requirement is imperative
since the microhabitat and the prevailing microclimate enveloping the floating PV system are driving many
of the unique broad-spectrum performance and impact effects of FPV technology. It makes environmental
understanding a crucial aspect of the digital twin characterisation of floating PV technology.
From the perspective of emulating the geodynamical responses of floatovoltaics technology in agricul-
tural environments, developing an FPV digital twin model should aim at replicating a floating photovoltaic
power plant as a virtual energy manufacturing enterprise. Generally, the digital twin should be able to
predict performances digitally and mitigate undesirable behaviour in a complex system context (Grieves,
2014; Grieves and Vickers, 2017). As a designated virtual power plant (operating in digital cyberspace),
characterising the geospatial digital twin model for a floating PV system should be designed to contain prin-
cipal operational objects and its network coupling interactions, defined in terms of technical, geometric and
semantic information. As a geospatial digital twin variant, the floating PV model should contain real-time
meteorological and climatic sensor data, and its simulation workflow should ideally integrate a variety of
analyses to ensure evaluation of the best design, best planning and optimal design intervention decisions
(Stoter et al., 2021). Given the interdependency complexities perceived to transpire within the complex envi-
ronmentally driven floatovoltaic ecosystem, the thesis is to define a new simulation modelling framework to
challenge the current prior-art assumptions around land-based PV modelling. The floatovoltaic simulation
model design process inspires the study to revert to a more integrated systems-design approach towards
developing a high-fidelity dynamic FPV model as a geospatial digital twin model (Goossens, 2021; Prinsloo,
2020).
In this regard, the study sets its sights on designing and developing a high-fidelity dynamic digital twin
simulation model to characterise the operations of floatovoltaic systems in digital terms. The problem state-
ment has already highlighted a clear research opportunity to conceptualise and create new knowledge
frameworks to fill in unique geographical knowledge gaps around determining the sustainability status of a
21
project in quantitative terms. In this geographical context, the thesis highlights the research’s requirement
for new knowledge as it defines the primary study purpose as follows:
Study Purpose:To challenge state-of-the-art in the geographical science field by creating new knowledge
to fill in the knowledge and research gaps around design stage floatovoltaic performance and sustainability
status assessments. To formulate research questions around the theoretical conceptualisation of an inno-
vative scientific roots-driven sustainability assessment framework that can serve as underpinning analytical
methodology. To drive decision-supporting capabilities in a computer-assisted geo-aware geoinformatics
information system for floating PV project valuations. To ensure improved synergy between sustainability
valuations for farming-based PV energy systems and their agricultural habitats. Architecting and designing
an enhanced sustainability assessment policy framework for agricultural-type floating PV energy systems.
To ensure enhanced compatibility between a proposed farming FPV energy system sustainability framework
and that of agricultural sustainability valuation frameworks.
In terms of functional behavioural analysis, studying and documenting the attributes of floating PV tech-
nology requires a proper understanding of the operational aspects of the floating solar PV enterprise, in-
cluding the dynamics of the technical, environmental and economic aspects operating within the floating PV
ecosystem.
In terms of study purpose, this brings three main research opportunities for developing and implement-
ing a sustainability reference framework for assessing floating solar PV system performances and impact
effects. The first opportunity is to design a reference sustainability criterion for a floating solar project
amidst the already high complexity of the sustainability concept (Armstrong et al., 2020; Moldavska and
Welo, 2015; Prinsloo, 2020). The second opportunity is to use the reference framework in a computer-logic
architecture to define a computer synthesis model to study thus the properties and dynamism of a floating
solar PV ecosystem and enterprise (Prinsloo, 2020). Finally, the third opportunity is to employ the frame-
work construct and model expression in project decision-making support, especially around exploring the
role and value of floating photovoltaics in terms of the WELF nexus-resource interactions (Gadzanku et al.,
2021b). In this context, the study can apply complexity theory to quantify and explain the sustainability traits
in a systemic assessment process that may identify interlinked criteria and cross-cutting benefits among
operational functions within the floating photovoltaic ecosystem.
In the context of these opportunities, this thesis aims to postulate an improved sustainability criterion
as a theory and theoretical framework for floatovoltaic sustainability characterisation as a logical framework
for developing new digital computerised simulation-based evaluation models capable of predicting the ben-
efits. This theory intends to overcome the critical knowledge gaps in determining the role of floating solar
photovoltaic technology in the water-energy-land-food nexus (Prinsloo, 2020). In this WELF nexus-driven
sustainable development context, a newly anticipated geospatial solution needs to fill in the knowledge
gaps around the analytically-integrated impact assessments around floatovoltaic performance based on
programmable intelligence on an adaptive geospatial platform. The research design process in the rest
of this chapter details the research questions and how the research aim and objectives can be defined to
physically establish the conceptualised framework as a practice-oriented computer simulation model and
governing methodology.
The discussion concludes by introducing the problem statement and study purpose in terms of a re-
quirements analysis and functional behaviour analysis proposal. The following section informs the research
design, including the research processes’ formulation and the research questions’ definition. It also formu-
lates the research aim and objectives with a concise methodological scope.
22
1.3. Research Design
This section presents the research design process and serves as a scientific framework to systematically
analyse scientific data and research information in the application field of floating solar electrification. In
addition, the research process serves as a means to perform diagnostic analyses to appraise critically
and make sound judgements on and creatively deal with complex issues around the installation options
and schemes of next-generation floating solar renewable energy systems. Section 1.3.1 introduces the re-
search design framework, while a graphical layout of the research design process features in Section 1.3.2.
It presents the research process to help solve real-world challenges through floating solar planning and
formulating the leading research questions in Section 1.3.3. The definition of the research aim and the main
research objectives are provided in Section 1.3.4, while an overview of the methodological scope of the
research is presented in Section 1.3.5.
1.3.1. Research design framework
While the study purpose defined the research focus, the research design explores a computer modelling
methodology (Law, 2014) to investigate, analyse and predict the performances and impacts of floating PV
systems in the local agricultural landscape. In the context of Figure 1.6, the research design logic is con-
cerned with defining research questions and turning research questions into research aims and objectives
towards defining an appropriate research methodology and mathematical simulation method (Schlesinger,
1979). The research design process defines the research process and must decide on the elements for ex-
perimental research methods, the research strategy and the selection of a sample (McCartan and Robson,
2016). Strategically, the research design framework of Figure 1.6a sets the strategic goal as the research
foundation (problem, purpose, theoretical underpinning, research questions), leading to the selection of the
research method and methodology (data sampling, collection and analysis).
(a) (b)
Figure 1.6: Scientific procedure diagrams for the (a) research design framework (source: McCartan and Robson
(2016), page 3), and (b) potential avenues towards studying a system (source: Law (2014), page 4).
While Chapter 3 provides a more detailed motivation and insight into the thesis’s philosophical paradigm
and approach, it suffices to state that the study follows a systems-thinking philosophy (Mingers, 2014; Midg-
ley et al., 2014). It guides a quantitative research methodology through the engagement of computer simu-
lations in the data-driven modelling method of Figure 1.6b as a means to study floating PV systems and to
obtain empirical evidence (Law, 2014; NSF, 2006). In this context, the research design framework of Mc-
Cartan and Robson (2016), depicted in Figure 1.6a, provides a clear route map of the various components
and steps required when carrying out applied research in an analytical, quantitative research paradigm. The
23
framework also guides the research purpose and theoretical backgrounds to jointly feed into the research
questions, while the research questions drive the choice of research methods and the data-sampling strat-
egy and process. In this context, Chapter 3 motivates the value of the systems-thinking philosophy within the
context of the philosophical theory of positivism (Comte, 2019), and details the philosophical assumptions
around the research ontology, epistemology and theoretical perspective towards the chosen quantitative
research methodology and computer simulation modelling method.
With the research design framework of Figure 1.6 defined, the following section navigates the research
design process towards the overall strategic layout formulated to carry out the research by progressing
through the chronological research sequence.
1.3.2. Research design process layout
This section summarises and presents the research design layout according to the strategic research design
framework of Figure 1.6. While this research design process governs and navigates the research process,
it provides guidance on the methods and techniques selected to logically combine the various components
of the research project.
Addressing both conceptual and practical issues, the overall layout of the research design provides a
framework for planning the current research project and for probing the research questions. As such, the
research design layout of Figure 1.7 follows the research design logic of Figure 1.6 to specify the logical
structure of the research study, while providing an execution plan to guide the research towards defining and
answering the research questions (Sim and Wright, 2000). In terms of the workflow process of the research
design flow of Figure 1.7, the arrows denote the direction of the research from the problem statement
towards the purpose of the study, the formulation of the research questions, the aim and objectives, and the
methodological scope of the research project.
The research design process layout, shown in Figure 1.7, covers the overall strategy used to carry out
research and to progress through the chronological research sequence. It also defines a succinct and logical
plan to determine the research questions and answer the established research questions through collecting,
interpreting, analysing, and discussing experimental data. The research design layout also serves as a
framework for planning the rest of the investigation into conceptual geographical models for floating solar
PV sustainability assessments towards defining and answering the main research questions. Furthermore,
as the research design layout further governs the arrangement of the thesis chapters, it also provides a
visual guide to formulating the research design. In this context, Figure 1.7 guides the research through a
logical pattern of research activities and the flow of the research process from the research problem to the
research methodology.
With the research design process formulated, the following section defines the research questions,
which allow for the strategic navigation of the research process within the context of the study purpose.
1.3.3. Research hypothesis and research questions
While questioning the effectiveness of existing methods and paradigms to solve the research problem,
the literature review motivated this thesis to investigate systemic-type cause-effect analyses in a floating
solar ecosystem environment. To this end, the study formulates research questions aimed at energising an
investigative academic research inquiry into the implementability of a holistic and agricultural ecosystem-
orientated sustainability integrity valuation methodology for floatovoltaics.
In preparation for formulating the research questions, the literature study in Section 2 reviewed the
history of existing scholarly works of similar research to relate real-world observations to scientific theory
(INNS, 2021). This exercise set its target on finding knowledge gaps in systemic relational propositions.
24
Real-World Challenges /
Planning Requirements
Research Design Layout
Literature
Review
Problem
Statement
Research
Questions
Research
Aim & Objectives
Research
Methodology
Technical
engineering
complexities
Sustainable
development
goal impacts
Environmental
offsets & EIA
approvals
Investment
plan/economic
frameworks
Carbon Tax
legislative
requirements
Fragmented
information
Narrow
understanding
Interactive
dependencies
Decision
complexity
Decision
time-horizon
Multiple
objectives
Problem Statement/Knowledge Gap: Need for theory building around a sustain-
ability assessment framework and digital computer model to aid environmental sus-
tainability analysis in floatovoltaic project planning decision support in SA agriculture
Literature Review: Based on real-world challenges and problem statement - com-
prehensive literature study and critical review of existing research on energy devel-
opment requirements and floating-solar-planning assessments and decision support
Research Aim: Geo-sensitive information system for floating PV system sustainability
profiling based on a sustainability assessment framework in a digital twin desktop model
towards WELF nexus-integrated geoinformatics decision support for SA agriculture
Objective 1: Define an FPV
assessment framework, to
integrate energy, enviro &
economic domain elements
Objective 2: Formulate FPV
system dynamics simulation
model, logic and coding, for
FPV analysis-by-synthesis
Objective 3: Calculate
techno-enviro-economic FPV
assessment metrics, display
as WELF decision indicators
Methodology: Quantitative theoretical modelling and simulation methodology, with
algebraic system dynamics modelling instrument model, to investigate geoinformatics-
based floatovoltaic planning analyses and decision support in the SA agricultural context
Question 1: How can a
systems-thinking framework
be formulated as technical,
economic and environmental
tridentate, to theoretically
characterise FPV responses?
Question 2: In which way
can a simulation model be
established from a triden-
tate framework logic in FPV
project assessment and
decision-support profiling?
Question 3: To what extent
can a tridentate framework
support WELF nexus
driven decision-support in
visual/empirical FPV project
sustainability assessment?
Figure 1.7: Flow diagram of the research design process, the arrows denoting the direction of research from the
definition of the real-world challenge, the literature study, problem statement, research questions, research aim and
objectives towards the research methodology (source: author).
25
These systemic relational propositions can help to conceptualise an improved set of principles to sup-
port the environmental scoping, prediction and due-diligence reporting stages in digital sustainability as-
sessments of floatovoltaic projects in agricultural landscapes. It leads to the compilation of the problem
statement, which identified critical knowledge gaps that causes standardised metrics for ground-mounted
PV projects to properly account for the complex range of distinct and diversified resource-use-efficiencies
and impact-effect-positives of floatovoltaics. The study purpose, therefore, sets the goal to challenge state-
of-the-art in the geographical science field by creating new knowledge to fill in critical knowledge and
methodological gaps around design stage floatovoltaic performance and sustainability status assessments.
With the literature review playing an instrumental role in formulating the problem statement and study pur-
pose, Chapter 3 provides an extensive layout of the philosophical thinking behind formulating a tentative
research hypothesis around systemic relational propositions to help solve the real-world problem (refer to
Section 3.5.4). This process formulates a tentative hypothesis to give an idea and an orientation of the
research (Jashari, 2017), in terms of which the thesis drafts the tentative research hypothesis as follows:
Research Hypothesis:The sustainability attributes and natural resource impacts of agricultural-type floato-
voltaics can be derived from the tridentate responses of the technical-energy, economic and environmental
domain operations and its collective ecosystem-level interactions.
This tentative philosophical research hypothesis serves as a philosophical compass to guide academic
research toward new geographical knowledge creation while compiling the hypothetical attitude of this sci-
entific work (Chen and Cui, 2020). Strategically this hypothesis features the dynamic phenomenon and
systemic behaviour of floatovoltaic system behaviour that this thesis intends to explore while highlighting
the interest in the factors that cause the complexity and diversity of floatovoltaic sustainability. The state-
ment advances a testable proposition about the possible outcome of this scientific research study in line
with the purpose of the study (McLeod, 2018). It also acknowledges the dilemmas and ambiguities about
the complexity of the real-world research problem of floatovoltaic project sustainability. Furthermore, di-
rectly stating the predicted relationship functions between system actor variables enables the study to ask
progressive research questions on how and to what extent the three different operational factors and their
functionality affect the floating PV sustainability phenomenon in the agricultural ecosystem context.
Towards substantiating the ecosystem principles informed by the research hypothesis through experi-
mental data, the research questions are progressively focused on the potential conceptualisation and ap-
propriation of a more holistic and WELF-resource-friendly sustainability evaluation framework better suited
for floatovoltaic technology sustainability assessments in agricultural ecosystem contexts. In aiming to com-
pose a new theoretical basis and program logic in a whole-system analytical computer simulation modelling
methodology, the research thinking philosophy is, as such, directed at a quantitative computer modelling
and synthesis methodology towards achieving geo-aware analytical decision support. Towards the sys-
temic sustainability precepts of feasibility, viability and accountability in a practical project sustainability
due-diligence assessment and analysis technique, the thesis formulates three research questions that read
as follows:
1. Research Question 1: How can systems-thinking principles be used to conceptualise and formulate
an integrated tridentate framework to theoretically characterise the combined technical, economic and
environmental responses of floating photovoltaic technology in support of the sustainable development
discourse?
2. Research Question 2: In which way can a geo-sensitive analytical computer simulation model be es-
tablished in terms of an integrated theoretical framework logic to characterise the operational dynamics
26
of floatovoltaics in a tridentate systems-thinking model for empirical project assessment and decision-
support in aid of sustainable development narratives?
3. Research Question 3: To what extent can an integrated tridentate assessment framework and com-
puter model expression support the sustainability profiling of floating photovoltaic projects in terms of
water-energy-land-food nexus qualities as a means to drive visual/empirical project decision-making in
a sustainable development context?
Taking guidance and cues from the general theoretical research design and research formulation con-
cepts proposed by researchers such as Marshall and Rossman (2011), the research questions above have
the goal of inspiring novel theory-driven academic research aimed at exploring new ideas around existing
uncertainties in floating PV project sustainability assessments as an area of concern. It further points to
the need for deliberate research into the performance modelling of floating PV ecosystems based on the
systems-thinking ontology and computer synthesis methodology. The research questions make it possible
to more effectively decide on the goal and route toward achieving the practical outcome of the research and
also guide the research study’s relevant research aim and objectives to be formulated within the setting of
the research questions.
The following section presents the research aim, giving the research activity the desired thematic and
theoretical direction. It also defines the research objectives to provide concrete steps to observe and ad-
dress current conflicts and knowledge gaps through a proposed decision-analysis framework, theory, con-
cept or model.
1.3.4. Research aim and objectives
This section formulates this research study’s relevant aim and objectives within the framework of the three
research questions stated in the previous section.
The research aims to define a scientific sustainability assessment framework to investigate the applica-
tion of quantitative analytical methodologies in computer-based decision-supporting applications. Further-
more, it seeks to establish an information (eco)system in which to seat a unified sustainability theory and
assessment framework, integrated as program logic, in a digital twin computer model, with the aspiration to
accommodate sustainability assessments in an environment in which the technical and operational floating
solar technology options are varied in terms of a range of technical and business permutations. Against the
backdrop of the contextual research questions for the research project, the quest is to define a research
aim that would ultimately help to facilitate regulatory approvals in sustainable floatovoltaic energy project
developments in a local agricultural landscape context. In this context, the study formulates the research
aim as the following:
Research Aim: Conceptualise a geo-sensitive information system for floating PV system sustainability
profiling in terms of theoretical performance and impact assessments based on a novel analytical framework
and computer model expression in a real-time computer-modelling and simulation solution, embedded as a
heuristic-type geoinformatics decision-support system suitable for South African agricultural applications.
The research aim directs the development of a sustainability theory and computer simulation model,
intending to develop a conceptual model that reflects the underlying science of the modelled processes,
as well as a mathematical representation of that science whereby the model encodes those mathematical
expressions in a computer program (refer to (EPA, 2022)). As such, a floatovoltaic digital twin or FPV-
geomodel should aid EIA and WELF-resource-oriented project planning in the theoretical appraisal of future
planned sustainable floating PV energy projects earmarked for off-grid and grid-substitution applications in
the South African agricultural context.
27
In line with the associated research questions and the research aim formulated above, the progressively
focused research objectives conceptualise and define the sustainability theory and computer models for
finding heuristic decision-planning support for establishing FPV-geomodel ecosystem components. They
aim to develop proposed whole-system sustainability criteria and valuation methodology underpinned by
a conceptually integrated sustainability assessment framework for floatovoltaic technology. The research
objectives progressively focus on implementing and applying a holistic computer modelling and simulation
methodology capable of realising a cause-effect analysis in a whole-systems-based operational floating
PV ecosystem. Three research objectives, closely aligned with the research questions and research aim,
define the goals that the research sets out to achieve as the following:
1. Objective 1: Conceptualise and evaluate an analytical sustainability reference framework design that
includes functional energy, environmental and economic domain operations as systemic integrants. Its
purpose should be to serve as computer program logic towards an integrated FPV-geomodel, also per-
forming as an analytical ecosystem, for theoretical floating PV project assessments within the context of
the South African landscape.
2. Objective 2: Use the sustainability reference framework as computer logic to conceptualise a dynamic
systems-thinking modelling and simulation methodology. On this basis, design an analytical model
(FPV-geomodel) that defines and dynamically integrates empirical real-time simulation models for the
functional energy, environmental and economical operational integrants towards theoretically predicting
the pre-installation performance and sustainability impact profiles of future planned floating PV projects
for SA agricultural landscapes through analysis-by-synthesis experimentation.
3. Objective 3: Devise and design an integrated analytical and decision-supporting profile mapping display
for the integrated FPV-geomodel to serve as a visual geo-informatics decision-supporting interface for
floatovoltaic project planning assessments. It should use balanced, evidence-based indicator portfolio
mapping suitable for heuristic floatovoltaic planning support in grid-substitution applications for the South
African agricultural context.
The research investigation then reverts to an experimental case-study scenario in an evaluation exer-
cise to determine the level at which the research aim and objectives can perform. This empirical evaluation
will critically examine the research results in search of verifiable evidence that the research has already at-
tained the specific accomplishments defined by its purpose, aim and objectives. The thesis would thus ap-
praise the research objective through operational system simulations within local environmental conditions
for assessing floatovoltaic project sustainability. These simulations would emulate floatovoltaic functional
systems-planning mechanics to determine whether the proposed solution suits farming landscapes in the
South African region.
While the aim and objectives above support the research questions, the following section informs of the
methodological scope of the study, with the scientific research methodology to guide this philosophically-
based research investigation.
1.3.5. Concise research methodology and scope
Given the context of the research aim and objectives described in the previous section, this section provides
an abstract view of the methodological approach and scope of the study, which are to be dealt with exten-
sively in Chapters 3 and 4 of the thesis. The methodology takes cues from the systematic review of current
research described in Chapter 2, specifically the strategic direction depicted by the flowchart in Figure 2.1
as a guide to select those papers and manuscripts that deals explicitly with floatovoltaic project analyses,
risk prognosis and sustainability assessments of future planned floatovoltaic projects.
28
Based on a systematic review of current research literature of similar scientific works in Chapter 2, the
thesis provides an extensive description of the philosophical research methodology in Chapter 3 for the
development of an analytical sustainability reference framework (Research Objective 1). With the impera-
tive goal of geographical theory building guiding the drafting of a research hypothesis, this investigation’s
methodology starts with engaging in an inductive research strategy and philosophical theory-building ap-
proach to analyse floatovoltaic system operations from a geographical operations research perspective.
This methodology guides the comprehensive coverage of the relevant scholarly review in the literature study
to infer conclusions from scientific field data as a geographical operations research approach. It regards the
floating PV system as a 4IR-type smart energy production system for which sustainability attributes must
align with that of climate-smart agricultural goals (Scherr et al., 2012; WWF, 2021). With systems thinking
philosophy spearheading the quest for new insight and understanding into the broad spectrum of floato-
voltaic system behaviour phenomena, a geographical system science methodology enabled the thesis to
provide a trans-disciplinary geo-operations perspective on the conceptualisation of a holistic sustainability
valuation strategy for floatovoltaic technology systems. This perspective creates the opportunity to advance
the research hypothesis of the thesis as a diagrammatic view of an operations ecosystem, wherein the
floatovoltaic production system operates within a geodynamical agricultural habitat as part of a cooperative
network or operational agency structure. Since this holistic, systemic intervention provides an improved
understanding of the dynamical system behaviour responses, it serves as a theoretical basis for the more
efficient and effective empirical assessment of the sustainability integrity and coherence qualities of floa-
tovoltaic. In this way, the thesis establishes the foundations for creating new knowledge to support the
improved theoretical characterisation and understanding of sustainability in FPV power generation.
While the philosophical research methodology in Chapter 3 provides a unique geographical, behavioural
science perspective on floatovoltaic system behaviour and sustainability modelling methodology, Chapter 4
works toward the computer modelling of the posited sustainability reference framework as computer pro-
gram logic in a computer simulation modelling and decision support methodology (Research Objectives 2
and 3). This parametric modelling methodology works toward synthesising the posited theoretical frame-
work for deciphering the distinct sustainability signals of floatovoltaic technologies in a computer synthesis
and decision support model. This methodology comprises three parts: (a) the systems-level design and
implementation; (b) component-level design and implementation; and (c) the decision-level layering design
and implementation of the desktop-based experimental research model. With the main research goal set
on answering the research questions within the context of the research aim and objectives described in the
previous section, the broad scope of the research methodology is captured in the research methodology
formulated below:
Concise Research Methodology: Actualise a quantitative theoretical computer modelling and sim-
ulation methodology for floatovoltaics by conceptualising an enhanced sustainability criterion as a com-
puter program logical for a mathematical simulation and algebraic decision-support model (Floating PV-
geomodel), and use the model in pre-qualifying project-assessment experiments to study the sustainability
attributions of a typical floatovoltaic system operating in the South African agricultural landscape.
Towards developing new geographical methods and tool models, the NSF (2011) promotes computer
modelling and simulation as an advanced scientific research methodology in systems geography. The
focus is specifically on research into systems-design thinking supported by data-science and computer-
simulation methodologies in the spatial sciences to address problems, advance new theories and help
understand society’s challenges from an energy geography perspective (Calvert, 2016; Solomon et al.,
2003). In terms of considering methodological options in the geographical science engagement through
applied geographical theory (NSF, 2011; Strahler, 1980), an abstract representation of the positioning of
29
computer simulations as a recognised investigative geographical method and tools are depicted in the
geography-centric diagram of Figure 1.8a. Physical geography activities, pictured on the left side of the
diagram, highlight the importance and scientific positioning of the simulation tool driven by, amongst others,
data science, the geographical sciences, spatial sciences and other cognate disciplines.
Mathematics &
Statistics
Domain
expertise
Computer
science
Data
Science
data
analysis machine
learning
software
development
(a) (b)
Figure 1.8: Methodological options in the geographical science engagement, with the focus on (a) the role of mathe-
matical and computer science models (source: NSF (2011), page 10), and (b) knowledge integration inherent to the
geographical data science philosophy (after: Batty (2011)).
The data science philosophy and model-driven methodologies, as depicted in Figure 1.8b, illustrate how
the data-science-driven computer synthesis relies strongly on geo-intelligence and knowledge integration
from the domain-specific science, computer science and mathematical-statistical science perspectives. This
geographic data science and geo-computation context enables the use of spatial data science and data
engineering principles in geo-system modelling.
The geographical modelling conceptualisations in Figure 1.8 can potentially play an integral part in
modelling energy system sustainability while impacting the production of new knowledge. Towards estab-
lishing an opportunity for new knowledge creation towards constructing an improved sustainability theory,
Figure 1.8 presents the opportunity to develop new knowledge-based frameworks and to operationalise
such frameworks in sustainability assessments. The development of new geographical methods and tools
(Figure 1.8a), it can serve as an underpinning basis for the development of new geographical data science
techniques (Figure 1.8b) in geospatial digital twin models to generate the relevant geospatial-data in support
of assessing and studying the reality of PV energy system sustainability in the virtual dimensions.
In the context of floatovoltaic systems modelling, Figure 1.8 can guide the development of an experi-
mental computer simulation synthesis model that can play an instrumental analytical role in explaining future
systems behaviour or in examining how a floating PV system will perform, or to predict the future sustain-
ability impacts of a system installation. The process of computer simulations in digital twin models can also
support human decision-making (ESRI, 2022; Goossens, 2021), especially with floatovoltaic ecosystems
being so complicated that they are not easily understood. The assignment can also use the simulation
process in an educational role to help elucidate new knowledge or information by clarifying the effects or
impacts the systems planner may not have foreseen. From a sustainable development modelling perspec-
tive, geographical system science theory is key to sustainable development support (Fu, 2020; Mehren and
30
Rempfler, 2022), while new geospatial data science techniques (Karimi and Karimi, 2020; Singleton and
Arribas-Bel, 2021) in geospatial digital twin models (Clifton, 2022; Stoter et al., 2021) to provide a digi-
tal geoinformatics platform for analytical analyses and decision-support in floating PV project sustainable
developments.
By providing an overview of the methodological scope, the thesis thus concludes the discussion on the
research design and process and the definition of the research questions in conjunction with the research
aim and objectives. The following section looks at the significance of the research from an international and
a South African geographical perspective.
1.4. Significance of the Research
Given the context of the global energy transformation under the umbrella of South Africa’s National Climate
Change Adaptation Strategy, this section highlights how this thesis’s proposed research topic and approach
are meaningful and relevant.
Under this strategy, the overarching research significance derives from the rapid international emer-
gence and growth of new floatovoltaic climate-solving technology worldwide (World Bank, 2019b). The
research is relevant because the thesis supports the ongoing need for new information systems to support
sustainable energy production through economically-viable and environmentally-responsible solutions. The
following significant aspects of the research deserve special consideration:
1. Computer-assisted EIA processes: The research in this thesis is of substantial significance in its primary impact
focus on developing 4IR technology solutions to reinforce complex computer-assisted EIA process requirements
in a virtual platform environment (4IRSA, 2020). As such, the 4IR approach emphasises the significance of the
spectrum of EIA determinations. The complexity of EIA increases dramatically with agricultural floatovoltaics, owing
to the complex water-energy-land interactions uniquely inherent to floating solar systems. Furthermore, with the
EIA’s role being for the protection and improvement of the environmental quality of life on earth, the thesis promotes
the value of computer-aided EIA procedures and its more significant intention of simplifying such processes to
identify and evaluate the effects of energy project on the receiving environment (DFFE, 2021a).
2. EIA processing demand requirements: The demand for EIA processing has shifted to meet requirements asso-
ciated with disasters and epidemics which force people to work remotely. This ability has been facilitated by the
Covid-19 lockdowns, where the era beyond the global outbreak of the COVID-19 pandemic brought a whole new
approach to conducting EIA on a remote online basis. The thesis reinforces augmented assistance services to
support decentralised digital desktop EIA processes to meet such requirements. The thesis outcomes thus help
to address new digital worker initiatives enforced on practitioners by COVID regulations (decentralise processing,
boost digital capabilities, conduct pervasive computing, digitise collaboration and operations, first engage with dig-
ital customers) (Acker et al., 2020; Madurai Elavarasan et al., 2021). In this respect, the research meets the new
demands to ramp up decentralised EIA advisories and digital technology impact advertisements through mobile
and software-enabled intermediate digital expertise in mobile applications to help with the online decentralisation
of agricultural project appraisals (EGIS, 2022).
3. Prioritising sustainable agriculture through renewable energy: The research is further of particular signifi-
cance in that it supports agricultural sector goals at a point in time where floating solar systems are considered
the new frontier of sustainable farming through renewable energy (Norton Rose Fulbright, 2021). As a new source
of distributed solar power generation, it reinforces the South African national 2030 renewable energy-mix contri-
bution goals defined in the national IRP (DOE, 2018; RSA, 2019b). Such means of distributing solar renewable
energy can enable significantly better reliability, and reduce the levelised cost of electricity (LCoE) tariffs through
smart energy systems based on decentralised solar power systems in most sub-Saharan African countries (Lee
and Callaway, 2018; Schelly et al., 2021). The research can thus help reduce rural electricity costs and the risks of
exposure to unreliable grid power, load shedding and diesel generators, while positively contributing to reducing the
greenhouse effect in South Africa. Most parts of South Africa exhibit the potential for excellent solar energy gener-
ation (van Niekerk, 2010). The national government is widely applauding remarkable successes attained through
31
solar energy’s contribution as instrumental in rural development and spatial planning for sustainable development
(DMRE, 2019; Global Legal Insights, 2021).
4. Supporting the implementation of the new carbon tax legislation: What gives further substance to the signif-
icance of the research in this thesis is the legislative implementation of South Africa’s new carbon tax or eco-tax
obligations. This obligation requires proper financial performance modelling and carbon tax factoring into solar
energy market development projects (RSA, 2019a). Furthermore, given the National Climate Change Adaption
Strategy for South Africa (DFFE, 2019a) in the context of the latest South African carbon tax legislation, the thesis
research is considered most significant and highly valuable. In this sense, the study offers the delivery of an exten-
sive set of sustainability metrics required for participation in the Kyoto Protocol’s Clean Development Mechanism
(CDM) (IRENA, 2019) and in emissions trading schemes for carbon reduction and projects for removing emissions.
5. Promoting green energy empowerment through access to 4IR information: Equal and equitable access to
information and control over sustainable energy services is a fundamental right in respect of sustainable rural de-
velopment (Energia, 2020). It is also imperative to tackle the digital divide toward meeting the United Nations (UN)
Sustainable Development Goals (SDG) and Sustainable Energy to All (SE4ALL) objectives in driving faster action
towards the achievement of SDG 7 (SE4ALL, 2017; UN, 2020a). On the road to ensuring universal energy access,
floating solar can be seen as an engine for economic growth. Furthermore, it can help agricultural, rural and re-
mote communities overcome barriers preventing them from entering the small water footprint of the photovoltaic
energy transition. In this context, this geographical research is significant because it conceptualises geographical
technologies driven by the 4IR to help create energy prosperity and possibilities for reshaping the future of solar
by making floating solar energy more accessible to all (SE4ALL, 2017). The thesis research thus supports envi-
ronmental awareness amongst a new breed of agripreneurs and energypreneurs, especially in an era where our
government has identified science, technology and innovation as primary drivers of local job creation, economic
growth and socio-economic reform (DSI, 2021; Lombard et al., 2021). Both the analytical and decision-support
around energy system capabilities described in this thesis could, as such, aid in identifying collaborative opportu-
nities in water-based floating solar energy businesses in support of the microclimate and job-creation goals in a
local community environment (Khan, 2021). It thus supports the advancement of rural-centred entrepreneurship
and social enterprises, potentially within the corporate social responsibility narrative, to bring off-grid electricity to
underserved agricultural communities across South Africa.
6. Informatics for the prospective regularisation of new FPV technology: The research is of significance owing
to the novelty of the recent invention of state-of-the-art floating solar systems as a new means of harvesting solar
radiation energy from the sun. This new frontier in technology discovery creates astonishing sustainable farming
opportunities in the South African agricultural sector, as the technology offers sustainability benefit effects ready to
be translated into pragmatic environmental plans (Hernandez et al., 2019; Pritchard et al., 2019). While many land-
mounted solar photovoltaic energy systems in South Africa already generate clean electricity from the non-polluting
sunlight resource, floatovoltaic as a water-conscious technology offers a fresh opportunity for decarbonising farming
operations through solar farming and the solar augmentation of on-farm electrical power generation. This thesis
provides a valuable geographical contribution to the application field. On the other hand, very few scientific studies
in South Africa have developed quantitative analytical approaches to predict the broad spectrum of environmental
offsets and impact benefits that this new floatovoltaic climate-solving type of technology has to offer.
7. Supporting local WWF conservation wine-farming initiatives: The research is of particular significance to the
winemaking industry in South Africa, where sustainable farming practices and environmental risk management
initiatives prioritise the protection of agricultural farmland, reducing water usage and implementing energy-efficient
solutions. Prominent environmental wine leaders in the South African wine industry are collaborating with the inter-
national World Wildlife Fund (WWF) on these goals for wine-makers to become certified as conservation champions
(WWF, 2021). This opportunity for prestigious accreditation for the local agricultural sector emphasises the need
to prepare the scientific world for obligatory sustainable development requirements around project authorisations
for a newly emerging wave of interest in FPV solar technologies to support WWF sustainable winery certifications.
8. Satisfying the urgent need for new EIA tools and geographic models: The tasks of impact assessment are
generally found to be complicated, time-consuming and even computationally intensive worldwide. Even if im-
pact assessment computer models were available, such analytical models do not always provide an integrated
solution that covers all three spheres of agricultural sustainability, namely energy, economic and environmental
analysis. This aspect is even more evident in the case of agricultural FPV energy installations, wherefore present-
32
day geographical tooling models fail to provide an adequate contextual foundation for solving pressing present-day
challenges in the floatovoltaic technology sustainability landscape (Hernandez et al., 2020; Prinsloo, 2019). This
non-fulfilment is primarily due to the extended range of superior technical benefits and environmental co-benefits.
The study addresses an identified dilemma concerning environmental impact analyses for floatovoltaic technology,
as well as the requirement for improvements in environmental practice in developing a computer-aided EIA solution.
The study thus offers scope for innovation and an opportunity to exploit the extended environmental offsets pro-
vided by the innovational trends in solar renewable energy around state-of-the-art floating solar technology (NREL,
2019a; World Bank, 2019b).
While this section highlights the significance of the current research, the following section introduces the
overarching assumptions made in this study.
1.5. Definition of Terms and Research Assumptions
In developing new paradigms or theoretical frameworks to make logical research deductions, the researcher
is compelled to define the terminology and identify assumptions that present the logical real-world scenarios
around the floatovoltaic sustainability discourse and the study’s research framework.
The list of the terminology and the definition of each of the terms used throughout the thesis are docu-
mented at this point. The terminology and definitions give the reader an understanding of the concepts and
terms of reference used throughout the thesis, and familiarise the reader with the contextual information
about how the terminology is meant to express the geographical concepts in this study. In terms of novel
semantic terms for studying floatovoltaic technology in the context of geo-information systems, the research
investigation defines three new definitional terms, namely (a) floatovoltaitry, being the planning, design and
construction of floatovoltaic energy systems; (b) floatoplanometry, being the activities and measures associ-
ated with the processes of planning and analysis of floatovoltaic energy systems; and (c) floatoplanometrics,
being the informetrics and indicator metrics associated with the planning for and the analytical decisions in
respect of floating PV energy systems.
Given the above definitions, the thesis highlights the axiomatic research assumptions of the thesis,
with specific reference being made to the conceptualisation of resilient heuristic geo-informatical analytics
and decision support. In the broadly-defined geographical context of techno-economic, techno-enviro, and
enviro-economic profiling, the research is focused on heuristic decision-making analyses, in which the study
endorses the set of assumptions given below:
1. Assumption 1, Life-cycle consistency: In the geographical reasoning context of techno-economic and
enviro-economic decision analyses, the current research accepts that government legislation and policies
(interest rates, environmental taxes, electricity costs, etc.) are flexible and dynamic enough to impact on
future aspects of an incentive-based, problem-solving environment around government and industry. The
same assumption holds for variations in predicted solar radiation, as climate change and natural earth
processes may impact multi-year weather and solar data for typical meteorological year data (TMY data).
As such, the thesis does not include any inter-annual climatic variability, year-by-year weather changes or
longer-term cycles. For those interested in numerical weather predictions, these effects can be modelled
as future research, as they are typically stochastic. According to standard industry practice, and in the
best interests of geographical reasoning and heuristical decision support in a dynamic real-world environ-
ment, the current research assumes that the relevant regulations, financial taxes, subsidy formulations,
investment incentive values, interest rates, electricity tariffs, TMY climatic/weather data projections, and
so forth, would remain relatively stable or consistent throughout the lifetime of the project (with certain
escalations where indicated). In terms of techno-economic and enviro-economic systems profiling, these
enable the study to perform explicit impact estimates and predictions based on the present-day policy
33
and regulatory assumptions in the current South African regulatory context. Similarly, typical meteoro-
logical year-ahead weather conditions for full-year TMY data would be assumed to be repeated over the
project’s lifetime. This assumption enables the study to perform explicit impact estimates and predictions
based on the present-day policy and regulatory assumptions in the present-day South African context.
2. Assumption 2, Energy supply versus demand anomalies: In terms of techno-economic and environ-
mental profiling, the research is focused on heuristic decision-making analyses in terms of photovoltaic
self-consumption in off-grid or grid-substitution applications, where the dispatch of optimal energy does
not form part of the study focus. In the context of this type of geographical reasoning, the assumption
is made that floating solar farms consume all of the electrical power generated by the planned floating
solar PV system on any hypothetically-proposed site. The variations in the consumption profile of the re-
spective sites, the engineering technicalities around intelligent demand/response reactions, the storage
of battery energy and the schemes around the intelligent management of energy (Mazzola et al., 2017)
are issues beyond the scope of this study. These factors have not been factored into the research’s ge-
ographic floatovoltaic capacity-based model developed by this research. This assumption further meets
the definition formulated by this study for agricultural grid-substitution applications (prosumer-based on-
site generation, on-site consumption, 100% self-consumption), typical of independent or isolated power
systems in rural farming areas. These are realistic assumptions for the South African context, where
the guidelines and legalities around grid-energy wheeling and feed-in tariffs with independent power pro-
ducers have not yet been firmly established in government and Eskom circles. While future research
may consider an analytical method for demand response systems or load determination on floating solar
farms (Ikhennicheu et al., 2021), Assumption 2 is logical in terms of the understanding that the com-
puted impact figures will reflect the maximum potential environmental impact for a proposed FPV system
and location site. An optimal economic-scheduling assumption is rational as the typical daily electricity
consumption profile (solar-powered irrigation, solar-powered cooling, cold storage, etc.) on farming and
fruit-processing facilities shows that the bulk of the energy is consumed during the day when an FPV
system would generate all of its electricity.
3. Assumption 3, Floatovoltaic applicability: While valuable agricultural land surfaces can be saved by
using floating solar installations on local man-made freshwater reservoirs, the scope of this research is
focused on floating solar devices for suitable areas where the intensity of cultivation is much higher and
where eligible dams with water can be used for floating solar panels (Spencer et al., 2019). This includes
remote agricultural areas where electricity supply is unreliable or electricity prices are high. Therefore,
the study assumes that the application of floating panels is better suited to the water-rich areas of South
Africa, such as the wine-growing regions of the Boland and the Western Cape, where water bodies
on which to place photovoltaic units are more freely available than land. Apart from water-rich areas
such as the fruit and wine-growing regions of the Western Cape and other similar regions in South Africa
(Prinsloo, 2019), the study also contemplates that freshwater floating solar solutions may also be suitable
for on-shore lagoons along the coast. Future models on the visionary horizon may include options for
offshore floating solar projects, especially offshore marine-floating solar-energy installations in coastal
areas that offer enclosed bodies of seawater calm enough to allow for marine energy through floating
or submerged solar panels. Although the issue of visual impact may start playing a role, floatovoltaic
systems can be integrated into the farming landscape on an aesthetic and practical aqua-farming level
(Weigl, 2021). With Assumption 3 as the backdrop, it should be acknowledged that large parts of South
Africa have a shortage of water but no shortage of agricultural land, particularly in areas of high solar-
energy potential. Hence, land-based photovoltaic installations could be considered as an alternative
EIA option in water-stressed areas where farmland has very poor grazing potential and where there
is far too much of a water-shortage issue to specifically create suitable water ponds on which to float
34
photovoltaic units. The experimental section of this thesis endorses such comparisons from a project-
planning perspective.
4. Assumption 4, Ecological and social impact modelling elements: The problem statement and re-
search aim to focus on the effects of multipronged interactions between the knowledge domain elements
of the primary energy, environmental and economic impacts of the floatovoltaic technology ecosystem.
The social and ecological aspects form an integral part of classical sustainability conceptualisation, de-
fined in terms of the typical three "P” relationship pillars, namely People-Profit-Planet (3P) or Social-
Economic-Ecological (SEE) sustainability triangle (Gbejewoh et al., 2021; Sala et al., 2015). As such,
many researchers argue that floatovoltaic sustainability models are not entirely complete without quan-
tifying and delivering ecological or social impact results. While the standard definition underscores the
classical thinking behind the relationships in the 3P/SEE sustainability definition, the system dynamics
model proposed by this thesis provides additional subsystemic models and interactions that dovetail with
the standard sustainability definition. Although much ongoing research is venturing into the ecological-
impact space (Cagle et al., 2020; Hernandez et al., 2020), ecological and biological complexities and
social corporate social responsibility and social acceptability (Bax et al., 2022; Schelly et al., 2021), these
aspects are beyond the scope of this thesis. The study opted to include social and ecological aspects as
a potential direction for future research in this geographical study.
5. Assumption 5, Electrical systems modelling and site morphology: Technically, a typical water-based
floatovoltaic system includes a floating pontoon structure with securely attached photovoltaic panels
to generate electrical power within a specific geospatial and power-conditioning environment. The im-
pacts of site-specific electrical engineering technicalities and variations in site morphology are beyond
the scope of generic floating PV-performance assessments in this Geographical Information Systems
study. This means that the thesis focuses on direct current (DC) parameters, where electrical power
references are concerned, and does not consider PV systems peripherals such as energy-conditioning
and energy-carrier equipment in a smart grid or smart microgrid environment (i.e. ignoring BoS balance
of system components such as electronic power conditioning, solar batteries, AC/DC inverters, power
electronics, power switching, cabling/wiring, battery-energy storage, transformers, connecting boxes, etc.
(Baum-Gartner, 2017; Oliveira-Pinto and Stokkermans, 2020)). In the context of comparative heuristic
decision-supporting analysis in a geographical context, the present study assumes universal industrial
standards for the local context and benchmark costs for small-scale power generation (Kuschke and Cas-
sim, 2019). When computing the system’s LCoE power and maintenance costs over the lifetime of the
FPV economic model, the study, therefore, engages unitised capital watt-peak, cost-rating guidelines for
both the solar panels and floater components (Pritchard et al., 2019; Sustainable Energy Africa, 2017).
6. Assumption 6, Standalone floatovoltaic systems: While there is consensus that decentralised energy
projects, such as floatovoltaic projects, are a necessary means to improve universal access to electricity
supplies, most of the current energy-planning tool models do not include metrics for grid extension or grid-
substitution reliability, nor indicators to quantify the impact of reliability cost factors (i.e. operational losses
on the account of load shedding, power outages, energy wheeling, etc.) as part of the cost assessments
for decentralised solar energy systems (Lee and Callaway, 2018). As the present geographical study
is more interested in the use of floatovoltaics in behind-the-meter grid substitution type applications, the
inclusion of such complex and variable reliability cost factors would generally impact positively on the
decision prospects towards the installation of floatovoltaic systems. However, the randomness of these
costs and events may vary from region to region, thus signifying that complicated statistics may need
to be developed in future to account for these factors. The present geographical study would not at
this stage include any cost factors or indicators to quantify the impact of reliability cost factors in the
geoinformatics model.
35
The identified study assumptions may appear to restrict the research. However, these axiomatic-type
assumptions are consciously made to focus the study on the philosophical and geographical research
aspects rather than venture too intensely into the problem fields of control engineering, computer science
and agricultural physics. At the same time, the assumptions may present opportunities for extending the
focus of future research regarding the utility of the research results. These will be explained later in the
conclusions of the thesis.
This section concludes the discussion of the research assumptions. The following section presents an
outline of the thesis chapters, with a short précis on the content of each chapter.
1.6. Thesis Outline
In terms of the organisational framework of the thesis, this PhD thesis is subdivided into six chapters. The
chapter divisions are set out as follows: Chapter 1 embodies the introduction to the thesis with a statement
of the research topic, an outline of the scope of the research, a definition of the problem statement, and
a definition of the study purpose towards providing a statement of the final hypothesis. It further poses
the research questions, presents the study aim and objectives, explains the methodological approach, and
defends the research’s significance while detailing the thesis definitions and study assumptions. Chapter 2
presents a literature review, including information on geographic information systems, environmental im-
pact assessment studies and sustainability assessment considerations, with specific reference being made
to the knowledge gaps around floating solar technology systems. Chapter 3 introduces the methodological
approach by presenting the study’s philosophical approach and research method. It presents conceptual
ideas that display how the subsystem simulation objects are taxonomically integrated into a 4IR-type system
dynamics model and motivates a computer modelling and simulation methodology to study the performance
profiles and impact effects of floating PV technology systems. Chapter 4 details the implementation of the
research methodology and floatovoltaic system simulation model as a fully functional research instrument.
While this chapter provides detailed mathematical descriptions of the systemic energy, environmental and
economic subsystem models, it also presents the underlying causal interactions that constitute a concep-
tualised geoinformatics model. It further provides guidelines for the methodological procedures, explaining
how to conduct the experimental evaluation of this research project. Chapter 5 details the experimental eval-
uation findings, as determined through a scenario-based experimental case study. Based on case-based
scenario narratives, the geoinformatics desktop model is used to characterise and map the sustainability
profile for a floating solar system in a hypothetical grid-substitution application within the local agricultural
landscape. The thesis conclusions are summarised in Chapter 6. This chapter further answers the research
questions, provides strategic findings and conclusions with recommendations, summarises the novel contri-
butions of the thesis, and discusses options around the practical and academic utility of the research while
suggesting directions for future research.
This introductory chapter concludes the discussion on the investigative research approach aimed at help-
ing to solve a contemporary real-world problem around the planning of contextualised floating solar energy
systems through scientifically-driven decision-support mechanisms. The next chapter details the literature
review on the synthesis of similar studies and reviews the definitional and territorial difficulties and contro-
versies around planning decisions in the analysis domain of the floating solar energy scenario.
36
2. Literature Review
2.1. Introduction
This chapter expands on the study topic by exploring the scientific literature around this philosophical in-
vestigation into floating PV project assessments. It offers a comprehensive and critical literature review
that extensively deals with the merits of the research and elaborates on the real-world challenges asso-
ciated with the study aim against the backdrop of previous research conducted in the field of study. The
review constantly focuses on real-world research problems arising from the body of knowledge developed
in previous research. As such, the scientific review highlights existing research, best-practice methods and
modelling frameworks toward floating solar analysis and decision support from a systems planning and
investment viewpoint. Furthermore, the assessment underscores the scope for innovation and the opportu-
nities for novel geographical research toward solving real-world geographical problems experienced within
the planning operations domain around the operational floatovoltaic ecosystem.
Exploring scientific literature around floatovoltaic system sustainability performance modelling and as-
sessments engages in the analytical systems approach to scientific business information devised by Rumm-
ler (2007) to break down the sustainable development research problem and questions into a three-tier prob-
lem hierarchy. In the process depicted by Figure 2.1, the literature review assessment hierarchy considers
the statutory/organisational-level requirements, tactical process-level goals and activity-level challenges be-
fore resuming a discussion on the current state-of-the-art scenario toward the conceptualisation of a new
theory, methodology or model for floating PV analysis. The research methodology of the thesis ultimately
derives a new FPV assessment and modelling solution from the systematic review of current research and
the strategic direction depicted by the flowchart in Figure 2.1. This flowchart thus serves as a guide to
selecting those papers and manuscripts in the literature that deals explicitly with the pre-installation as-
sessment and analyses of future planned floatovoltaic projects in the context of the world’s larger energy
economy and water economy infrastructure support projects.
Process-level goals & challenges
(Section 2.3)
Statutory-level requirements
(Section 2.2)
Literature review
Activity-level challenges
(Section 2.4)
Review on state-of-the-art in floating PV characterisation
(Section 2.5)
Scope for new theory, framework, methodology or model
(Section 2.6)
Figure 2.1: The literature review uses a hierarchical analytical systems approach to break down the research problem
and questions around the challenges of floating PV (source: author).
37
In terms of Figure 2.1, the review surveys the research opportunity from a strategic/statutory problemat-
ical perspective (in terms of sustainable development goals, legislative and policy frameworks, and global-
context aims); moves to consider the obstacles (in terms of options for the deployment of the technology,
the goals to attain sustainability in agriculture, and impact assessment governance) to the planned process;
and examines the shortcomings on the operational level (tactical implementation in terms of best-practice
options, tools, techniques, models) to enable progress towards the conceptualisation (a practical design,
model, framework, theory concept) of a design solution. Such a tiered approach helps to elaborate on the
central theme of the philosophical research study. It further produces a logical survey of the definitional and
territorial difficulties and controversies in the analysis domain of the floating solar scenario.
The primary aim of the scholarly review on sustainable development is to contextualise floating PV
sustainability assessments, and to shed new light on the theoretical underpinnings of the empirical char-
acterisation of floatovoltaic technologies, both core real-world geographical problems presently understood
poorly. The review investigates opportunities for an improved theoretical basis for solving current geospa-
tial problems with PV performance modelling, especially finding an improved theoretical solution for the
empirical quantification of the so-called "technology unknowns” floatovoltaic technologies as the primary
regulatory deployment barriers of the technology (Armstrong et al., 2020; Gadzanku et al., 2021b). While
the review addresses the research problem in a three-tier problem hierarchy, it intends to show that existing
critical knowledge and methodology gaps around the fundamentals of scientific sustainability assessments
are cutting right across the three-tier problem hierarchy. It percolates from the assessment activity level,
through the tactical process level challenges, to the regulatory legal compliance requirements for new en-
ergy development project authorisations.
Although floatovoltaics characterisation primarily resorts under the activity-level tier of the three-tier
problem hierarchy, the review shows that it is a fundamental underlying problem that impacts the conse-
quential regulatory-, carbon taxation-, investment due diligence-, and sustainable development-compliance
requirements. It means that a proposed theoretical intervention needs to consider the implications of the
activity-level challenges to address strategic/statutory and tactical/process-level challenges. Therefore, an
extensive overview of the derivative difficulties of the research problem is presented in a three-tier hierarchy
to understand the root cause of the problem better. The review concludes with an objective analysis of
the State-of-the-Art in Floating PV Sustainability Profiling in Section 2.5, followed by a critical debate about
opportunities for a new theoretical principle in the discussion on the Scope for Innovation and Creation of
New Knowledge in Section 2.6.
On a strategic level, the literature review prepares for the proposal of a conceptual opportunity-driven
modelling framework solution as a new theoretical basis for floating PV sustainability assessments. To this
end, Section 2.2 introduces the concept of sustainable development from a global perspective and presents
an overview of the regulatory regimes and framework policies to be considered when addressing sustain-
able developments in South Africa. Section 2.3 details the uncovering of the floatovoltaic technology, its
agricultural impact benefits and co-benefits, and the substantial range in environmental impact variations
between floating solar and land-based solar power installations. Section 2.4 explores the phenomenal ex-
ponential growth rates in floatovoltaic installations around the globe. It also considers the diffusion potential
of floatovoltaic technology for South African agricultural applications. Section 2.5 surveys the research
literature on current best-practice methods in energy sustainability science in terms of the existing method-
ologies for appraising sustainability performance in planned floatovoltaic project deployments. Section 2.6
concludes by summarising the data observations by narrating the scope for innovational schemes toward a
plausible solution for resolving prediction challenges in technical performance around floatovoltaic project
designs in support of the diffusion of the technology in the South African floating solar PV landscape.
38
2.2. Regulatory Context of Sustainable Energy Development
A sustainable development focus initiates the current research project toward assessing and profiling novel
floating solar sustainability. Therefore, this inauguration section provides a logical basis for the geographical
study by introducing the sustainability concept within the broader global and local energy development con-
texts. Furthermore, it furnishes details of the international regulatory requirements that shape the current
research project and its theoretical framework details. Section 2.2.1 introduces the concept of sustainable
development from a global perspective. Section 2.2.2 presents an overview of the regulatory frameworks to
address sustainable development challenges in South Africa, while Section 2.2.3 focuses on local opportu-
nities for floating solar energy projects from a regulatory perspective.
2.2.1. Global development of sustainable energy
This section provides insights into the sustainability aspects of global energy transformation amidst the
burgeoning demand for energy globally. It underlines the global push for green economy technologies and
renewable resources to meet the transition goals for sustainable energy. It sketches the global context for
current sustainable development challenges in the South African context. Finally, it highlights the worldwide
need for developing appropriate geospatial tools for scientific sustainability modelling and assessments in
new energy project developments that account for the impacts of these planned projects worldwide.
While global climate change impacts water and energy resources, scientists must address two signif-
icant global imperatives climate change has created worldwide. Firstly, natural or anthropogenic climate
change events amplify the need to reduce energy-related carbon dioxide emissions to curb rising temper-
atures globally. Secondly, they alert scientists to the need to consider the cascading effects as part of
the dynamics of green transition, such as improving the conservation of water resources to help combat
worsening drought conditions (Cohen and Hogan, 2018). Compounding the sustainability problem are the
increased water temperatures for open-air water reservoirs, especially since there is the additional problem
of the increased heat caused by climate change (UNESCO, 2020). These are causing increased evapora-
tion levels that threaten clean drinking and irrigation water availability. At the same time, it will further help
to block sunlight from entering the water storage to limit intensified algal growth and the production of toxic
algal blooms (Heisler et al., 2008), while restricting the chemical formation of carcinogenic bromate from
freshwater bromide (Naushad et al., 2019).
With the world’s energy project developments moving to support smart-city, smart-village and smart-
farming concepts, energy is the dominant contributor to climate change as it accounts for around 60% of
the total global greenhouse gas (GHG) emissions (UN, 2020a). Furthermore, the dependence on coal-
fired power stations in many parts of the world further exacerbates the climate change problem, as these
plants use the already diminishing water supply as a coolant, thus adding to potential disturbances in the
water balance in countries such as South Africa (De Villiers and de Wit, 2010). The current strain on
global water supplies emanating from the continued development of coal-fired power stations is considered
one of the most harmful worldwide carbonisation impact risks for rising temperatures globally (Cohen and
Hogan, 2018; De Villiers and de Wit, 2010). Therefore, current environmental engineering efforts include
hydraulic engineering projects, such as the hydro-technical construction of modern water storage reservoirs
and hydro-technical installations of hydro-electric power plants at water collection and processing facilities,
as part of the application of scientific and engineering principles to improve and maintain the environment
(World Bank, 2019b).
Since energy stands central to virtually every significant opportunity and challenge the world population
faces in the current millennium, achieving energy sustainability targets would help meet all other global sus-
tainable development goals (SDGs) (ESMAP, 2021; UN, 2020a). In managing and monitoring the impacts
39
of global energy developments as a worldwide decarbonisation goal, the UN has defined a total of 17 SDGs
designed to transform sustainable development worldwide in a carbon-neutralising direction economically
(UN, 2020a). Figure 2.2 depicts the energy development goal SDG 7 as an anchor goal, as it stands
central to social development, while directly linking all of the 17 other SDGs (Madurai Elavarasan et al.,
2021; World Bank, 2017) in an Integrated Sustainable Development Goal (iSDG) model (System Dynamics
Society, 2021).
Figure 2.2: Energy in terms of Goal 7 of the SDGs shows how energy project developments take on the central stage
role in the context of sustainable development (source: World Bank (2017), page 15).
As one of the main challenges of resilient and sustainable infrastructural development in emerging
economies, the SDG7 goal is to strive toward "access to affordable, reliable, sustainable and modern energy
for all” and drive faster energy development actions under the SE4ALL programme (SE4ALL, 2017; World
Bank, 2017). Furthermore, while the SDGs shown in Figure 2.2 are all intertwined, Vairavamoorthy (2019)
highlights the fact and discusses the option that water could also be considered an essential resource in
the central part of the SDG set (similar to the central role of energy in Figure 2.2).
The agricultural sector in Africa is taking necessary steps towards greening new energy investments,
especially in off-grid and rural energy access projects (Medinilla et al., 2022). Concerning energy in the rural
agricultural sector, where the rural economy is heavily dependent on fossil fuels, scientists estimate that the
world’s agricultural activities contribute around 19 to 29% of the world’s total annual IPCC audited GHG
emissions (CGIAR, 2020). Furthermore, despite the significant urban/rural divide in the gross domestic
product (GDP), agricultural food supply chains are estimated to consume around 30% of the total global
energy production (CGIAR, 2020; FAO, 2021). With energy being a recognised vital driver supporting
the UN SDG7 (World Bank, 2017), this is a critical component in virtually all agricultural food production
and processing chain stages. Serving as a testimony to Figure 2.2, energy access in agriculture essentially
drives the supply chain from agricultural crop production to harvesting and post-harvesting activities, cooling
and storage, and through to value-added processing, transport and distribution (FAO, 2021).
To help overcome the sustainable energy challenge, the climate-protection efforts of sustainable agri-
culture drive three pillars of sustainable development as part of the UN Agenda 21, namely adaptation to
40
and flexibility in terms of climate change, the sustainable increase of agricultural productivity and the mit-
igation or elimination of GHG emissions (FAO, 2021; Mavroidis and Neven, 2019). Towards solar energy
and the fuelling of a zero-emissions future, the focus is on measuring the technical efficiency, feasibility and
sustainability of renewable energy applications in agriculture (FAO, 1995, 2021).
The discussion presented in this section highlights the importance of sustainable projects around water-
conscious energy development to support the SDG goals in the rural and agricultural areas of developing
countries worldwide. The following section focuses on the South African sustainable development context
for new energy projects, as it highlights the legislative and policy framework contexts for predicting the
energy impacts of new energy project development.
2.2.2. Development of sustainable energy in South Africa
This section presents a general overview of sustainable development challenges and the associated leg-
islative context of South Africa’s climate change action plan. Thus, it highlights the regulatory frameworks
to deal with the immense pressure on our country’s natural resources from the perspective of sustainable
energy development. It presents the current legal and regulatory contexts to underscore the compliance
requirements and associated policy obligations around the drive for sustainable development.
South Africa’s green economy approach aims to implement programmes and technology solutions that
support practical and implementable action plans toward the clean energy "grid of the future”. The goal ex-
plicitly states advancement "towards a low carbon footprint and a resource-efficient pro-employment growth
path, by building the best processes, initiatives and indigenous knowledge frameworks to support key eco-
nomic sectors” (DFFE, 2020). Currently, electricity generation is primarily undertaken by the state-owned
Eskom power utility company, with smaller contributions from Independent Power Producers (IPP), espe-
cially in wind and solar generators. Regarding critical infrastructures and the future of energy generation
technologies in the country’s energy-mix compilation, the South African power pool comprises gas, hy-
droelectric power, coal, nuclear power and renewables (Global Legal Insights, 2021). Regular reviews on
the current energy mix compilation and the development of new energy generation technologies are likely
to impact the future direction of energy policy. At the same time, In the current coal-dominated scenario,
economic threats associated with South Africa’s coal-fuel cycle externalities, linked with the generation of
water-thirsty, coal-based energy, keep rising (De Villiers and de Wit, 2010; World Bank, 2013).
While South Africa is considered one of the most carbon-intensive economies in the world, it is fortunate
to have a sound legal system and climate litigation strategies backed by extensive environmental legislation,
judicial procedures and legislation, as portrayed in Figures 2.3 and B.1. These government frameworks are
aimed at achieving sustainable development, effective environmental management and the conservation of
a rich pool of the country’s natural resources (IFC, 2021; Semelane, 2021).
In the context of Figure 2.3, photovoltaic technologies offer one of the largest levers that the SA soci-
ety has in the fight against climate change, especially since it further offers the capacity to help produce
carbon-neutral synthetic fuels or electro fuels (e-fuels) from recycled CO2(Ababneh and Hameed, 2022).
However, within the various mandates of the legislative and regulatory frameworks, a foundational policy
landscape informs the local energy sector transition of SA. In terms of the local energy policies summarised
in Figure 2.3, energy service companies (ESCO’s), new energy project developers and project owners carry
heavy responsibilities and liabilities to guarantee high environmental quality standards throughout all the life
stages of the newly planned energy projects.
In Figure 2.3, the National Environmental Management Act (NEMA, Act no. 107 of 1998) (RSA, 1998),
read jointly with the Environmental Management Air Quality Act (Act no. 39 of 2004) (RSA, 2004), estab-
lishes a statutory framework that allows for the regulatory enforcement of Section 24 of the Constitution
of the Republic of South Africa. While it recognises the necessity of economic development, the frame-
41
Figure 2.3: Summary of environmental-related legislation and historical policy landscape in South Africa informing
the local energy sector transition (source: Semelane (2021), page 4).
work promotes cooperative governance to ensure upholding the rights of citizens (King, 2016). While SA’s
Climate Change Strategy (CCS) is set to augment SA’s Sustainable Development Discourse, her EIA frame-
work (together constituting environmental frameworks and policies) ensures that prospective sustainability
assessments are an evidence-based EIA practice. The idea is that occupational practices’ sustainability
assessments should provide reliable scientific evidence.
That sharpens the focus on the reporting requirements, and the fact that regulatory project approvals
and environmental air-quality impact estimates must ensure reliability through bona fide scientific evidence
(predictions or measurements). Sound scientific evidence around sustainability assessments during the
planning phase of a new energy project (refer to Figure 1.4) is a crucial requirement when seeking new
pathways to cleaner production. Progress-tracking, as such, forms part of global stocktaking through the
digital electronic media (ESMAP, 2021). These scientific reporting requirements present unique oppor-
tunities to develop innovative 4IR data tools built upon simulation-type environmental impact models and
geographical systems toward computer-assisted assessment technologies related to impact analyses.
Considering our government’s international obligations and local objectives in Figure 2.3, South Africa’s
National Development Plan (NDP) sets the goal of investing in a strong network of economic infrastructures.
It aims to develop an integrated energy plan to support the NDP’s economic and social development goals
within the context of an appropriate energy mix to meet the electricity needs of South Africa over a 20-year
horizon to 2030 (NDP, 2019). Accordingly, South Africa recently gazetted its Integrated Resource Plan
(IRP), which defines projected energy availability factors in the future energy mix to guide the transition
toward reduced carbon-emitting technologies, which are crucial to the local economy (RSA, 2019b). The
IRP primarily promotes infrastructural developments in renewable electricity by investor-owned utilities that
are renewable-driven, based on a least-cost power supply and a demand-balancing principle (DPE, 2019).
These initiatives aim to battle against climate change through the upfront definition of a diverse energy mix
that respects the global COP21 plight for keeping the global temperature increase below the 1.5 to two
percent (2%) threshold (COP21, 2016). SA’s IRP contributions can support this commissioned goal before
2030. In reducing emissions, this determination is expected to contribute towards a 28% CO2-equivalent
atmospheric pollutant reduction plan between 2019 and 2050 (Global Legal Insights, 2021). Supporting
South Africa’s climate change strategy (DFFE, 2019a), the NDP Vision 2030 document provides a succinct
summary of the indicators to be considered in mitigating environmental problems in South Africa from an
economic development perspective (NDP, 2019).
While SA’s Climate Change Strategy (CCS) augments SA’s Sustainable Development Discourse, the
EIA framework of the NEMA Act (DEA, 2014) supports the Treasury’s new Carbon Tax Act (RSA, 2019a),
are both designed to dovetail in their emission goals through the planning phases of sustainable devel-
42
opment projects. As such, the analytical reporting towards IPP and EIA project approvals (DFFE, 2015;
DMRE, 2019; DEA, 2014) ultimately operates in conjunction with SA Treasury’s Carbon Tax Act, recently
promulgated as the Carbon Tax Act No. 15 of 2019. In addition, the SA government introduced a pre-cursor
policy on the requirements for Environmental Offset Policy reporting (DFFE, 2015). The implementation
of this policy aims to strengthen the Carbon Tax Act (RSA, 2019a; SAIT, 2020) towards promoting a just
transition from coal to renewable energy to limit GHG emissions under the RSA (2019b).
Given the nature and gravitas of the local environmental problem in terms of the rapid population growth,
urbanisation and the slow economic growth rate, new eco-friendly energy developments are needed to con-
tribute to the unprecedented demands made in capitalising on our country’s WELF (water, energy, land
and food) resources (King, 2016). In the context of global warming and climate change concerns under
UN Agenda 2030 (UN Chronicle, 2015), studies on the inter-dependence between these resources (Law-
ford, 2019; Ringler et al., 2013) have assisted the DFFE (2019a) in its implementation of the South African
National Climate Change Adaptation Strategy. This strategy presents yet further 4IR support opportuni-
ties, where data scientists can harness data intelligence in geoinformatics systems to help evaluate and fix
critical environmental impact challenges.
This section presents a general overview of sustainable development challenges and the associated
legislative context of SA’s climate change action plan. The following section now focuses on local opportu-
nities for floating solar energy projects from a regulatory perspective.
2.2.3. Opportunities for FPV sustainable energy projects in South Africa
This section highlights local opportunities for floating solar energy projects in South Africa from a regulatory
perspective. It presents the latest changes to the legal and regulatory frameworks to underscore the oppor-
tunities that floating solar energy presents on a small scale for the self-generation of embedded electricity
under the local drive for sustainable energy development.
Political and economic initiatives in South Africa intend to increase energy production by driving the de-
mand for solar energy while reducing carbon emissions and employing water-saving technologies. With the
local climate conducive to sun-driven solar energy expansion, the opportunities to deploy new floating solar
installations are rising. Such opportunities are set to increase as the country recently earmarked around
6000 MW of new solar photovoltaic power capacity from distributed generation by IPPs (RSA, 2019b).
This goal has the societal interest at heart as it supports the notion of energy localisation through dis-
tributed generation, particularly in the form of remotely dispersed solar energy generation. The policy has
the added benefit of producing electrical energy through small-scale technologies closer to the power end-
users (Global Legal Insights, 2021). Future energy development expansions, such as floating solar projects,
can thus benefit from renewable portfolio standards and developments to instil investor confidence in a con-
stant pipeline of distributed energy through national initiatives such as those shown in Figure 2.4. Such
development actions would include locally distributed renewable energy projects, improvements in energy
efficiency, enabling smart grids, energy storage and a digital backbone supporting the electricity network
grid to integrate capacity and ensure the efficient delivery of an economical, sustainable and secure elec-
tricity supply (RSA, 2019b).
New opportunities, such as shown on the map of Figure 2.4, to strengthen the argument for local floa-
tovoltaic deployment toward achieving the country’s IRP targets have emerged from the South African gov-
ernment’s countrywide renewable-energy priority areas and corridors. The designated geographical areas
aim to enable higher levels of renewable power exploitation through wind turbines and solar energy projects
in strategic regions of the country (DFFE, 2022; McEwan, 2017). Known as the Renewable Energy De-
velopment Zones (REDZs) and corridors, certain geographical areas and corridors portrayed in Figure 2.4
43
Figure 2.4: Supporting the mitigation of climate change and reductions in local emissions, SA’s grid corridors (DFFE,
2022) already spurring new energy spatial developments (source: Semelane (2021), page 10).
have been selected to perform Strategic Environmental Assessments (SEAs). The SEAs serve to fast-track
new local wind and solar energy development projects in the nominated regions (EGIS, 2022).
In the identified RE zones and corridors depicted in Figure 2.4, the SEA mechanism aims to unlock
solar energy developments in the selected geographical areas (Semelane, 2021). The identified zones
and corridors in Figure 2.4 further ensure that neighbouring areas receive improved economic stimuli in
the energy context and help reduce the virtually complete reliance of regions on agricultural and mining
economies (Gorjian et al., 2022b). In addition, the REDZs-targeted solar energy areas are ideal locations for
promoting regional auctions to ensure private offset agreements for energy purchases from IPPs. In spatial
deployment terms, Figure 2.4b already confirms substantial geographic shifts in deploying new renewable
energy plants and creating new energy jobs thanks to these meta-level renewable energy infrastructure
plans to support SA’s energy, water, ocean and marine economy (Semelane, 2021).
Beyond the geographical zoning, as depicted in Figure 2.4, a more ubiquitous legislative and financial
opportunity strengthens the argument for local floatovoltaic deployment, namely the recently promulgated
South African Carbon Tax Act (RSA, 2019a). This act aims at balancing climate action and economic de-
velopment with the country setting ambitious goals for emission reduction driven through the taxation of
carbon emissions. In terms of newly envisaged floatovoltaic energy deployments in the South African agri-
cultural landscape, this new environmental taxation legislation offers tremendously positive opportunities for
local farmers to benefit from the economics of pollution and emission mitigation through the carbon-offset
credit and trading exchange provisions promulgated in the new act (RSA, 2019a). To meet the analytical
requirements prescribed by carbon taxation, the SA government introduced a further analytical reporting
obligation, namely the Environmental Offset Policy (DFFE, 2015). The implementation of this policy is meant
to strengthen the Carbon Tax Act towards promoting a just transition from coal to renewable energy to limit
GHG emissions under the RSA (2019b) objectives.
Amidst all the attractive opportunities for floatovoltaic energy projects countrywide, the set of Environ-
mental Reporting Regulations (ERR) is an integral part of the country’s carbon reporting under the evolving
post-2020 mitigation regime (DFFE, 2015; DEA, 2014). As the new carbon tax and environmental-offset
44
framework amendments are being instituted, the newly proposed decentralised solar floatovoltaic technol-
ogy can help to shape the agricultural sector’s decarbonisation campaign for its future power scenario by
shifting the producer-consumer energy dynamics. As such, the new legislation mainly puts the spotlight on
the policy obligations and associated regulatory compliance requirements in preparation for geographical
tool development to support floatovoltaic project deployments within the local agricultural sector.
Crucial to compliance with the Carbon Tax Act and the NEMA Act, the policy-reporting guidelines for
environmental offsets oblige the project developers to submit scientifically-founded a priori analytical mea-
sures to determine tax rewards and to hone their awareness of environmental offsets in projects at an early
phase in the process. These guidelines call for the appropriate design phases in terms of the decisions
made on the analytical-potential, mitigation-strategy and refinement fronts, especially during the project-
design and project-planning phases (refer to Figure 1.4)of the planning conceptualisation of a new energy
project. At the same time, scientific proof of a project’s environmental and sustainability integrity needs to
consider the implementation of a clean development mechanism, sourced in the experience of the Kyoto
Protocol flexibility mechanisms for international emissions trading (Allen et al., 2021).
Thus far, the discussion has dealt with the global and South African regulatory contexts to help overcome
sustainable development challenges associated with green energy transformation through floatovoltaic tech-
nology. The following section introduces the floatovoltaic technology phenomenon, and the remarkable
technical attributes and sustainability impact features it can offer in helping to reach sustainable develop-
ment goals in the local agricultural sector.
2.3. Uncovering Floatovoltaic Technology and its Impact Effects
Having studied the contextual regulatory requirements and opportunities for floatovoltaic technologies, the
discussion in this section highlights the need to advance the state-of-the-art analytical and technological
developments in geography by conceiving and creating a scientifically-based digital twin prediction model
to estimate the sustainability impacts of an exciting new technology known as floatovoltaics. It also highlights
some critical environmental impact attributes of the technology to enable the reader to understand better
the geospatial complexity spectrum and multidimensionality of the real-world geographical problem within
the sustainable development discourse on the spatial aspects of floatovoltaic technology.
Section 2.3.1 uncovers the suitability of floatovoltaic technology for sustainable energy development
projects in line with local agricultural needs and the international SDGs. Section 2.3.2 offers an extensive
literature survey to document and reference the technology’s range of features, environmental benefits and
co-benefits to highlight the substantial variations in environmental impacts between floating solar and land-
based solar. Finally, Section 2.3.3 discusses the implications of the impact effects of floatovoltaic technology
on sustainable agriculture, emphasising opportunities for advancing practices through the development of
floatovoltaic technology projects.
2.3.1. Uncovering floating solar technology
While the old climate vulnerability problems and their new social threats are asking for innovative tech-
nology and regulatory solutions (Cohen and Hogan, 2018), one relatively new renewable energy strategy
to address land-use, water preservation, and green energy supply challenges simultaneously is installing
eco-friendly solar floatovoltaic energy systems on water surfaces.
Floatovoltaic technology relates to installing photovoltaic systems upon pontoon-based floating struc-
tures to actualise solar energy farming on open water bodies (irrigation reservoirs, irrigation ponds, hydro-
electric dams) (Friel et al., 2019; Singh et al., 2022; TNO, 2021). As shown in Figure 2.5, the technology
45
offers a unique individual flavour of photovoltaics because the installation of floating solar panels over multi-
purpose reservoirs, canals, rivers or above other open-air waterways yields excellent water, air quality, land,
financial and climatic payoffs (Liber et al., 2019; McKuin et al., 2021; SERIS, 2021). In farming, land or
building-integrated photovoltaics are often placed on parcels of land and farm-building rooftops as primary
or auxiliary power supply systems.
Figure 2.5: Typical components of a floating solar technology testbed (source: World Bank (2019b), page 2).
Floating photovoltaic technology offers local energy in spatial proximity to energy prosumers and their
local energy resources (Hoffacker and Hernandez, 2020). The concept of floatovoltaic technology further
enables the siting of photovoltaic devices on open water reservoirs or other often under-utilised water body
surfaces. By way of example, Figure 2.5 shows an illustration of a typical standalone floating solar con-
struction to establish a green energy solar farming architecture on a freshwater or ocean surface as part of
a distributed energy resource (DER) integration (World Bank, 2019b).
Accordingly, floatovoltaic technology essentially exploits a niche mutation in the solar photovoltaic de-
velopment spectrum in which solar systems enter the water domain, thus transforming this hydrological
domain into a notion of a usable island space in the form of a floatovoltaic island sited directly on an open
water space (NREL, 2019a; World Bank, 2019a). The technology is considered a revolutionary cleantech
innovation from an environmental impact and sustainable development perspective. The floating solar in-
vention delivers an extensive range of renewable energy and environmental conservation benefits (Abid
et al., 2019; World Bank, 2019b), creates a range of co-benefits owing to the symbiotic relationship be-
tween the floating solar panel habitat and the underlying water habitat (Abid et al., 2019; Hernandez et al.,
2014, 2020).
While helping to transform unused bodies of water into highly efficient green power plants, floating solar
systems installed over artificial or marine bodies of water, as in Figure 2.5, offer a range of downstream
advantages and environmental co-benefits (Grand View Research, 2020; World Bank, 2019b). To introduce
but a few: owing to the cooling effects of the water environment, increased photovoltaic power yields;
outstanding land-sparing benefits; water preservation and quality improvement benefits; minimised water
evaporation by covering the water surface; comparatively easy installation and maintenance; robust and
stable floating PV substructures; the ability to cover large potential areas at a comparatively low cost; and a
potential reduction in the building construction and installation costs of floatovoltaic systems when compared
to similar sized ground-mounted solar PV systems (Allen and Prinsloo, 2018; Wood Mackenzie, 2019).
In these contexts, floatovoltaic technology offers a decentralised clean energy solution that can mobilise
farmers to leapfrog traditional onsite fossil-based energy supply approaches. As such, floating solar solu-
tions can also help overcome the problems often experienced in rural areas where the power transmission
in conventional centralised electricity supply chains directed from coal-based or gas-fired power stations is
unreliable. Similar co-benefits can be gained by non-tracking or rotating floatovoltaic facilities in hybrid wind
46
power configurations (Solomin et al., 2021). Irrespective of the trophic status of the lake, watershed, reser-
voir or water governance areas, floating PVs (in ordinary PV or concentrated PV CPV format) can also be
applied to freshwater reservoirs, land reclamation areas, retention ponds, polder reservoirs, mine pit lakes,
quarry lakes, sand extraction ponds, irrigation reservoirs, hydroelectric dams, reservoirs, water-treatment
plants, aquaculture ponds, aquaponics fishponds, flood detention ponds or floodplains (Benson et al., 2015;
Baradei and Sadeq, 2020; World Bank, 2019b).
In consideration of the notion of a floating solar island, the physical footprint of the modular floating
architecture, which houses the embedded solar PV technology, harnesses the economics behind green
energy savings, together with the savings on agricultural land-price costs in using unused water surface
spaces previously unused or even unsuitable for the prevailing agricultural activities (Saunders, 2020; SPG
Solar, 2010). Because of its energy efficiency, in conjunction with its impact on land use and the premium
value of land (Dizier, 2018; Saunders, 2020), floating solutions such as PV technology are considered a
desirable option for water-abundant areas. Here, the technology can harness renewable energy on under-
utilised water surfaces in areas of complex terrain or where the availability of productive farmland is limited.
In this context, artificial agricultural water surfaces are ideal spaces for establishing floating solar farms as
such waters are often calmer than marine habitats. Furthermore, freshwater reservoirs allow for the cooling
of the floating panels through water sprinkler systems, thus further increasing the electricity performance
values, revenue income and environmental-offset impacts of floatovoltaic systems (Dizier, 2018; Cohen and
Hogan, 2018). Human-made water bodies can ideally be viewed as the aquatic equivalent of brownfields
suitable for floatable photovoltaic green energy developments (NREL, 2019a; Spencer et al., 2019).
From a geographical land-use perspective, the agricultural land and environmental conservation co-
benefits of a floatovoltaic facility are bolstered by the characteristic ease of installing the hardware for the
technology architecture. These aspects are illustrated in the typical low-intensity floating solar system
construction site in Figure 2.6. It demonstrates the floatovoltaic mounting structure as a modular concept
design (often supplied as a ready-to-ship assembly kit), built upon a row-by-row floating block-by-block
connect basis on a working platform at the water’s edge (DNV, 2021; Friel et al., 2019). Such a modular
pre-assembly deployment approach ensures additional land conservation benefits owing to the minimum
disturbance of the soil through a reduced need for site preparation activities (such as land surveying, land
excavation, earthworks and concrete works) to advance the idea of championing the potential practice of
the 100% carbon-free construction of solar parks (Weigl, 2021).
The floatovoltaic island construction in Figure 2.6 is deployed from a launching ramp at the water’s edge
during the installation of either a fixed anchored or a flexible surface-mounted floatovoltaic architecture (Ciel
et Terre, 2022; ISIGenere, 2020; Seaflex, 2020; Krug, 2021). As illustrated in Figure 2.6, the floating so-
lar racking system fixed to a buoyant structure reflects an opportunity and scope for innovation towards
the development of potentially new floatation configurations to simplify the on-site and on-farm installations
(Allen and Prinsloo, 2018; Deriu, 2021). In terms of the geometric dimensioning of an FPV system, a pro-
fessional geomorphological feature assessment and bathymetric analysis could help design or procure an
appropriate floatation device suitable for the hydrogeomorphic variability (Ciel et Terre, 2022; Wagner et al.,
2009) or bathymetric profile characteristics of the farm’s irrigation pond (Soulemane Hayibo et al., 2022;
SunGrow, 2020). While it is often argued that the price to install floating solar is around 10-15% higher, the
gap between the cost of floating solar and land-based solar is narrowing and is close to reaching price parity
(Haggerty, 2020). To reduce costs, technically-proficient farmers may alternatively develop their floatation
constructions to establish a customised floating solar photovoltaic farm in static or rotating solar-tracking
configurations (Allen and Prinsloo, 2018; Choi and Lee, 2014; Laing et al., 2011). While the jury is still out
on the comparative benefits and technical challenges concerning floatovoltaic maintenance in terms of oc-
cupational health and safety (OHS) functionality, contextual risk factors in the floating solar platform design
are being addressed through ergonomics design interventions to help achieve the UN SDGs (Sen et al.,
47
Figure 2.6: Floatovoltaic construction, showing the ease of installation through deployment from the water edge and
the limited need for concrete and earthworks (source: NREL (2019b), page 4).
2021). Towards developing climate-resilient decentralised floating solar installation solutions, geographers
and project developers should be aware of and knowledgeable about the potential physical and environ-
mental risks associated with floatovoltaic technology (Deriu, 2021; Judek, 2020), especially since designs
parameters for floatovoltaic systems and floaters are often not sufficiently robust for severe environmental
conditions (Choi et al., 2022; Kaymak and Sahin, 2021). Furthermore, the technology presents particular
geo-sensitive technical risk challenges and hydrodynamic loads on floating PVs in extreme weather con-
ditions, especially those that cause high winds and water waves (as in the case of hurricane crosswinds
or heavy snowfalls), requiring proper hydroelastic modelling and design-around strategies to counter these
geographically-linked risk factors (Dallery, 2020; Karatas and Yilmaz, 2021; World Bank, 2019b).
The range of technical and environmental benefits, with the added co-benefits of floating solar technol-
ogy, has sparked international interest in the advancement of the technology and its performance impacts,
and ways of dealing with uncertainties around installation impacts on various water bodies (Gadzanku
et al., 2021b; Gupta, 2021; Santos-Borja, 2021). Credible scientific research expeditions such as those by
El Hammoumi et al. (2021) and SERIS (2021) have set up experimental floatovoltaic testbeds to help fill
in the knowledge gaps around floating solar technology. To help quantify FPV technology unknowns, the
SERIS group constructed an open-air laboratory in Singapore, with an experimental FPV testbed for leading
floating solar manufacturers to use as a testing sandbox or living laboratory to conduct side-by-side instal-
lations for their solutions (López et al., 2022). As such, scientists can perform comprehensive research on
FPV systems and compare real-time performance monitoring analyses on the performance and impacts of
physical systems in large or small-scale remote off-grid and grid-tied applications. Moreover, global interest
in the ecological modelling and associated impacts of floating PV installations on the biosphere has further
inspired ongoing research into the impact effects of floatovoltaic systems on wildlife habitat, bird species,
algal growth, and aquatic fauna and flora in terms of microbiological, ecological and chemical parameters
(Armstrong et al., 2020; Cagle et al., 2020; Mathijssen et al., 2020).
Such research efforts set the sites on improved technology transfer efforts to advance experiential
knowledge and promote innovation excellence in floatovoltaic technology and the contextualities around
floating solar plant operations. This effort can support the worldwide trend to obtain deeper insights, gain
new knowledge, develop best-practice methods, and help formulate technical and sustainability standards
as enabling factors for this newly conceived floating PV technology invention (SERIS, 2021; Wood Macken-
zie, 2019). In the context of the present thesis, such research further calls for comprehensive geospatial
48
analysis to support the future understanding of floating solar technology impacts. Furthermore, such re-
search calls for 4IR-based information systems to reinforce scientific-analytical and operational perspec-
tives on the operational challenges, requirements, benefits and environmental impact co-benefits of floating
solar on water. This thesis will present tools for a better understanding of the real-world problem through a
proposed geographical solution and theoretical modelling conceptualisation.
Section 2.3.1 introduced the floatovoltaic technology and its benefits, while the following section docu-
ments the results of the author’s extensive literature survey around the scientifically-confirmed performance
and impact attributes of this new floating solar technology.
2.3.2. Salient impact attributes of floatovoltaics reported in literature
While the research aims to realise a digital twin model of a floatovoltaic installation as part of a computer
modelling and simulation methodology for floatovoltaic project decision support, it is essential to consider
the salient features of the real-world prototypes of the digital twin counterpart. The digital floatovoltaic
ecosystem model must consider the range of scientifically-confirmed impact attributes prevalent in real-
world floating solar installations. For this purpose, this section documents a detailed survey conducted
for this thesis to decipher the broad spectrum of impact effects, the salient environmental features and
sustainability characteristics attributed to floatovoltaic technologies.
While the geographic modeller must learn to understand much about the system to be simulated, the
graphical representation by Armstrong et al. (2020) in Figure 2.7a presents comprehensive insights into
the rich spectrum of floating solar system impact effects on the environment and the neighbouring fauna
and flora (impacts on the earth’s biosphere or ecosphere). As a function of the biogeographic region, the
theoretically-derived hierarchical-effects framework of Figure 2.7a was developed from a hierarchical spatial
reasoning perspective to emphasise the complexity and diversity of floatovoltaic interactions.
(a) (b)
Figure 2.7: Conceptual depiction of (a) a layered three-dimensional impact-effects hierarchy in a floatovoltaic ecosys-
tem (source: Armstrong et al. (2020), page 6) to be modelled in (b) an integrated environmental impact assessment
process (source: DEA (2014), page 12).
Together with scientific evidence from a systematic review and expert stakeholder viewpoints as a
benchmark, the framework highlights the multidimensional and layered nature of the impact-effects hier-
archy of the floatovoltaic ecosystem, while illustrating the critical variables affected in each impact region
49
(namely, the air, the water surface and the underwater aquatic region) (Armstrong et al., 2020; Exley et al.,
2021b). As such, the illustration in Figure 2.7a succeeds in highlighting the complex nature of the inter-
related and layered nature of the impact qualities of the water surface, the land surface and the air-water
interface of a floatovoltaic system. An ultimate analytics solution for floating PV would aim to support the
EIA processes in the context of Figure 2.7b, both in terms of the impact of the floatovoltaic solar energy
development on the environment, and particularly on the biosphere in the agro-industry, directed towards
environmental legitimacy, environmental accountability and environmental proactivity in support of sustain-
able development (FAO, 1995, 2021; IISD, 2016).
However, the extensive literature survey conducted by the author concluded that, while many of the
identified impact effects hidden in the layered impact-effects hierarchy of Figure 2.7a have been verified
in scientific literature, others are still being uncovered daily by experimental research or in real-world floa-
tovoltaic facilities. Furthermore, scientometric studies and bibliographic analysis of floatovoltaic literature
spotlight the analytical toolset modelling challenges associated with characterising floatovoltaic behaviour
(Gadzanku, 2020; Kumar et al., 2022).
Aiming to conceptualise a high-precision FPV analytical modelling framework and a geographical engi-
neering solution for FPV (the goal set by the research aim), the survey contributes to the study objectives
in relieving technological uncertainties and demystifying the contextual controversies around the planning
narratives of floating systems. As such, the survey seeks to list the most comprehensive spectrum of di-
verse impact effects of a floatovoltaic system towards uncovering many interdependencies and interlinkages
between its composite set of observable characteristics or traits, hidden among or camouflaged by one an-
other. Such inter-relational patterns often result from the interaction of the components of this technology
habitat with the natural environment and further manifest themselves in dynamic patterns of interactive be-
haviour. Detailed analyses in this respect are crucial to successfully modelling the homeostatic floatovoltaic
ecosystem, designed to provide accurate projections according to the composed modelling framework out-
lined in the mind map of Figure A.1. Therefore, keeping the core EIA phases of Figure 2.7b in mind, the
survey lists the experimentally-confirmed set of layered sustainability features reported for floating solar
technology in the literature. These features serve to amplify the shortcomings of the current analytical as-
sessment tools in predicting a broad spectrum of layered interactive assessment attribute measures in the
context of floatovoltaic impacts.
The detailed technical briefing below summarises the benefits and advantages of floating solar installa-
tions instead of the attributes of above-the-ground surface-mounted photovoltaics, referencing a wide range
of scientific publications as evidence of the pivotal upside performance and impact benefits inherent to the
floatovoltaic DNA. The dominant upside benefits, determined through scientific research, include the fol-
lowing: (a) avoidance of the land acquisition and site-preparation issues associated with traditional solar
installations (Hoffacker et al., 2017; Sharp Corporation, 2008; McKuin et al., 2021; Trapani and Santafe,
2015); (b) the application of more efficient, cooler photovoltaic systems as a result of evaporative cooling,
and in conjunction with the aquatic heat sink, thus assisting heat exchange between the solar panels and
the underlying water body (Choi, 2014; Karatas and Yilmaz, 2021; NREL, 2019a; Spencer et al., 2019);
(c) improved efficiency in the optical-electrical conversion, thus delivering a higher energy yield (a 5-15%
improvement) owing to the cooling effect of water that keeps the photovoltaic panels at lower temperatures
(Choi, 2014; Goswami et al., 2019; Karatas and Yilmaz, 2021; Kumar et al., 2017; Moharram et al., 2013;
Rahnama and Aghbashlo, 2019); (d) an improved energy yield as a result of the reduced impact of dust
collection, wildfire ash and ground soiling on the panel module surfaces (Bonkaney et al., 2017; Jordan
et al., 2018; McKuin et al., 2021); (e) improved mitigation of the carbon footprint and of the climatic im-
pacts owing to an elevated environmental offset profile and improved environmental impact characteristics
(Gadzanku et al., 2021b; Gunjan, 2019; NREL, 2019a; Tawalbeh et al., 2021); (f) cleaner panels on account
of the availability of spray water to wash, clean and cool the panels (Choi, 2014; Majumder et al., 2021),
50
thus reducing panel degradation and the negative optical consequences of dirt, soiling and the snail-trail
discolouration of PV modules (Rosa-Clot and Tina, 2017; SPG Solar, 2010); (g) water evaporation control
and water preservation benefits owing to the limited evaporation of water for the area under the canopy of
the power plant (a 70-80% reduction in evaporation, with sunlight attenuation), panel shadowing and the
screening of the wind effect, thus inhibiting water evaporation (similar to a water curtain or shade balls)
(Abd-Elhamid et al., 2021; Ciel et Terre, 2022; Gazdowicz, 2019; Gupta, 2021; Scavo et al., 2020; Taboada
et al., 2017); (h) improvements in the quality of irrigation water and in plant health owing to reductions in
the growth of toxic algae (panel shadowing restrains poisonous cyanobacterial algae blooms, while without
it, plants can be poisoned with contaminated water) (Abdelal, 2021; Boersma et al., 2019b; de Lima et al.,
2021a; Mathijssen et al., 2020; Trapani and Santafe, 2015); (i) conserving water by directly collecting rain-
water via an effective water/rain redistribution system, collecting dew, and also storing water as a potential
future generator of atmospheric water (Pritchard et al., 2019; Weigl, 2021); (j) positive ecological effects
with the floating frame in creating a natural habitat for the biosphere (birdlife, fish) (Allen and Prinsloo, 2018;
Chateau et al., 2019; SPG Solar, 2010); (k) reducing the impact on the ecosystem and a reduced ecological
footprint as a result of diminished interference in the biocapacity of the farm and the biologically productive
land (concrete civil works, soil compaction, etc.) (DNV, 2021; ISIGenere, 2020; Weigl, 2021); (l) limiting
the use of land space which tends to reduce land acquisition costs and the loss of fertile agricultural land
for crop production (Gazdowicz, 2019; Spencer et al., 2019; Haohui et al., 2020); (m) positive contributions
to all four of the WELF-nexus pillars and interactions (Campana et al., 2019; Gamarra and Ronk, 2019;
Prinsloo, 2017b, 2019, 2020); (n) exploiting the distributed energy resource as a self-generating energy
supply facility and an on-site power source during agro-business hours (World Bank, 2019b); (o) easy-to-
install solar tracking on a floatovoltaic buoyant structure (greater scope for hydrodynamic tracking); while
the sliding of the rotating solar panels on the water allows for the constant alignment of panels with the
sun (Choi and Lee, 2014; Schubert et al., 2017); (p) quick and robust construction and installation methods
for floating solar infrastructures, since, owing to a high degree of systems modularity, this involves an ele-
mentary deployment process that is easily installed (Pritchard et al., 2019; Sharp Corporation, 2008; TNO,
2021; World Bank, 2019b); (q) eliminating the need for major site preparation, thus excluding the need for
soil compaction/levelling and foundation laying (Pritchard et al., 2019; World Bank, 2019b); (r) ecologically-
friendly agriculture owing to the elimination of contaminating herbicides or chemical weed killers to control
the vegetation growing under ground-mounted systems (Hernandez et al., 2019; World Bank, 2019b); (s)
and less shading of panels by nearby objects, thus offering a reduced risk of sunlight blocking of panels
by surrounding buildings or structures (Gunjan, 2019; Martins, 2019); (t) suitability for remote off-grid ar-
eas where on-site generators are needed, especially where the grid power infrastructure is sparse (Suh
and Brownson, 2016; Vides-Prado et al., 2018); (u) suitability as a means to mitigate droughts on ac-
count of the water-preservation and drought-adaptation benefits to build resilience in remote continental
drought-response strategies (Gupta, 2021; Gonzalez Sanchez et al., 2021); (v) ease of accessing available
finances for supporting floating photovoltaic systems and services in compliance with the UN’s SDGs (IFC,
2015, 2020); (w) floating solar PVs complement hydroelectric power with daytime peak electricity demand
(Gupta, 2021; Gonzalez Sanchez et al., 2021); (x) economic benefits of floating solar arrays in the electrifi-
cation of remote rural areas (Makhija et al., 2021); as well as (y) increased revenue streams from incentive
and transactive motivations, eco-taxes and energy-trading revenues as a result of higher energy yields and
reduced carbon-footprint impacts (Barbuscia, 2018; Van de Ven, 2016; World Bank, 2019b).
Given the broad spectrum of floatovoltaic impacts described in this section, the EIA process, as pre-
sented in Figure 2.7b, has a strict judicial obligation regarding project approvals through environmental
scrutiny. These EIA requirements are mandatory and geared toward legal certification to the effect that a
planned photovoltaic installation would be deemed suitable for agricultural purposes and that the installation
site would be both environmentally and technically feasible (RSA, 1998). The FPV EIA regularisation under-
51
scores the requirement challenges associated with creating and re-contextualising geographical framework
knowledge in support of accurate floatovoltaic EIA assessments, leaving little room for prediction inaccura-
cies. Additionally, it underscores a need to conceptualise and develop new locationally-aware analytical tool
models to predict floatovoltaic performance characteristics with associated inter-dependent impact-effect
profiles as sustainability attributes of floatovoltaics (Prinsloo, 2019, 2020).
This section narrates the scientifically recorded and documented benefits and co-benefits of floating
solar technology installations. From a geographical impact analysis and sustainability appraisal perspec-
tive, little work has been done to predict or capture the remarkable environmental protection benefits of
floatovoltaic technology in regulatory-type environmental plans from a geographical modelling perspective
(Cagle et al., 2020; Prinsloo, 2019). In the age of digital farming and climate-smart agriculture, a practical
4IR-type digital analytical technology tool would help reinforce sustainability and precision farming princi-
ples in support of the sustainable farming narrative (Binder et al., 2010; FAO, 2021). This thesis addresses
analytical sustainability modelling issues regarding agriculture and sustainable energy development from
an international geographical needs perspective, specifically from a local SA perspective (Prinsloo, 2020).
2.3.3. Floatovoltaic impact effects supporting sustainability farming
In this section, the thesis discusses the cost-benefit implications that floatovoltaic technology brings in
support of the sustainable farming paradigm. It underlines opportunities to advance towards sustainability
in farming and sustainable farming practices through the development of floatovoltaic technology projects
to establish floating solar farms.
In addition to FPV technology’s increased green power generation and land conservation co-benefits,
floating PV solutions can enable farmers to transform their prevailing water surfaces into multi-functional
landscapes that generate revenue from renewable energy while delivering sustainability co-benefits such
as land and water conservation (Cohen and Hogan, 2018; Gorjian et al., 2020). In addition to environ-
mental restoration, micro-grid-based floatovoltaic systems in solar farming can help wine farmers gain by
increasing their real sustainability income prospects in wine farming as they advance towards the integrated
production of wines, organic wines and biodynamic wine certifications (Wines, 2019). As a grid-substitution
solution for renewable energy (Prinsloo, 2019), floatovoltaic technology reinforces those agricultural sus-
tainability practices required to reduce a farm’s ecological and carbon emission footprints through energy
management based on floating solar energy concepts (Brent et al., 2016; Exley et al., 2021a; de Lima et al.,
2021a). As part of the new frontier of renewable energy, floating photovoltaic farms combine well with wind
energy, wind turbines and wind power generation technology (even floating wind turbines) (Solomin et al.,
2021). Meeting a farm’s sustainability goals through green FPV technologies advances its green econ-
omy transition to optimise its carbon offsets while increasing its natural capital footprint (Van de Ven, 2016;
Wackernagel et al., 1999), especially in self-energy production by floating photovoltaic covers for irrigation
reservoirs (Santafé et al., 2014).
Serving as a testament to the drive for conservation agriculture, the invention of floating solar technology
has its roots in the agricultural sector, where the Californian winery, Far Niente, established the world’s first
floating photovoltaic farm in the Napa Valley wine region of the United States (US) (Allen and Prinsloo,
2018; Sharp Corporation, 2008). The invention of this pioneering floating solar station surfaced when the
wine farmer set the goal to become one of the first net-zero electricity-using wine producers. The winery
subsequently prioritised the conservation of its historical grapevines over sacrificing more vineyard land
to extend its green energy development initiatives (Pentland, 2011; SPG Solar, 2010; Wines, 2019). With
Far Niente’s conception of its world-first pioneering floatovoltaic system, the revenue income from green
electricity generation by floating PV systems has helped to maximise the winery’s return on investments -
both in terms of financial currency capital and sustainability in environmental investment or natural capital
52
(Allen and Prinsloo, 2018; Randle-Boggis et al., 2020). Furthermore, solar farming on a previously non-
revenue-generating agricultural water surface area immediately maximised the under-utilised water surface
for profitable energy generation. Profitability improved due to the water-embedded floatovoltaic system,
which delivered a more significant energy output than its ground-mounted photovoltaic counterpart on the
water’s edge (Allen and Prinsloo, 2018; SPG Solar, 2010).
The floatovoltaic invention thus proved its worth as an on-site source of distributed renewable energy
capable of delivering clean energy on the doorstep of the winery, while further supporting the winery’s
certification as a sustainable green energy project and its associated green marketing objectives (Allen
and Prinsloo, 2018; Hira and Swartz, 2014; Wines, 2019). From a sustainability perspective, the operation
of floating solar panels benefits directly from climatic conditions that are equally favourable and ideal for
agricultural production such as favourable air temperature and humidity conditions. In addition, the shading
protection provides a multipurpose panelled canopy on the floating platform that attenuates on-water solar
radiation to reduce water surface evaporation while creating favourable microclimatic conditions suitable to
downstream agro- and aqua-farming in conjunction with floatovoltaic energy generation (can support fish
and shrimp breeding aquaculture, similar to agrivoltaics and agrovoltaics) (Peacock, 2021). Such synergies
for the dual use of a water area for solar photovoltaic electricity generation and aquaculture create a new
opportunity for aquavoltaics (Pringle et al., 2017).
From an agricultural market needs perspective, and as shown in the illustration of Figure 1.2, floating so-
lar systems on farms are also becoming popular among farmers because these floating PV microgeneration
systems are noise-free, pollution-free, reliable, efficient and durable non-fossil energy systems. Moreover,
the technology around floating power generation inherently possesses certain advantages and co-benefits
that drive the agricultural market for floating power plants. These sustainability and profit-driven advantages
include an immediate local electricity supply in areas with limited grid infrastructure. They are also a source
of distributed energy located in rural areas where self-generated electricity is required; where there is an
on-site supply of green energy, and it is delivered in support of sustainability practices and PV power gener-
ation; where no land space is available for the installation of land-based power plants; and where a secure
power supply is provided in areas prone to earthquakes and floods (Sudhakar, 2019; World Bank, 2019a;
Ziar, 2021). Additionally, as a result of its positive environmental fingerprint, the technology is fast becoming
the preferred choice for agricultural on-site power generation, where it is becoming increasingly difficult to
realise ground-mounted photovoltaic systems because of the premium cost of land and social pressure to
use land productively (Saunders, 2020; Spencer et al., 2019).
Floatovoltaic energy solutions generally open up new opportunities for the local energy transition. They
support revenue generation for clean energy while supporting sustainable farming principles, but without
competing with agricultural activities. By cleverly associating sustainable crop/wine production with floating
PV technology, the Far Niente winery system has successfully delivered on-farm access to a sustainable
energy source. Furthermore, with its implementation, the floatovoltaic project also immediately reduced the
carbon footprint of the wine farm and winery while simultaneously reducing the land footprint of the farm’s
newly extended photovoltaic project, also in turn benefiting from the characteristic temperature sensitivity
of PV performance in lower operating temperatures (Altegoer et al., 2021; Choi et al., 2016). Altogether,
the distinctive support of clean technology for interrelationships among the energy, water and food nexus
elements of a biologically-sensitive water ecosystem helped improve agricultural productivity and increase
food security while ensuring the availability and the conservation/preservation of potable water (Allen and
Prinsloo, 2018). FPV technology can thus help enrich farmers’ portfolios without compromising agricultural
production while potentially enhancing the capacity of on-site photovoltaics towards mitigating climate risks.
In enriching the farming portfolio, the conservation of irrigation water, water preservation and water
quality improvements are placed high on the priority list. One of the leading agricultural market drivers
for floating solar technology in support of these priorities is the potential of floatovoltaic systems for water
53
conservation and preservation. As floatovoltaic panels are typically deployed over-irrigation reservoirs and
dams on farms, the resultant solar panel shading over the aquatic habitat inherently ensures the co-benefit
of shading and cooling the water surface to reduce the rate of evaporation (Abdelal, 2021; Scavo et al., 2020;
World Bank, 2019b). Such preservation of irrigation water can be attributed to the avoided evaporation of
irrigation water due to the solar-shading and wind-screening effects of the solar panels as a function of site-
specific solar radiation and aerodynamic heat-exchange effects. As such, floating PV technology also offers
a tool to control evaporation and manage reservoir water quality. The water preservation impact effects of
floatovoltaics come as a contextually-sensitive two-way interactive process (Exley et al., 2021a). Thus, the
warmer the climate of a region, the greater the amount of surface water saved from evaporation losses, and
the more valuable the increased amount of cooler energy generated from the floatovoltaic system (de Lima
et al., 2021b).
Land-use calculations for land-integrated photovoltaics often reveal the actual economic impact of an
energy development project in terms of crop production forfeited. In the case of floatovoltaic systems,
no crop production is forfeited, while the farm can further benefit from power generation on its irrigation
hydrosphere (Spencer et al., 2019). Thus, a calculation to determine the land-use footprint of solar energy
systems, in conjunction with a real estate property valuation in land suitability, should play a decisive role
in planning for water-integrated floatovoltaic systems. Furthermore, stricter environmental practices in the
administration of solar farming developments are necessary as the result of solar energy installations on the
land impacts directly on the land cover, which may be subject to restraints on the repurposing or alienation
of land in protected areas (Hernandez et al., 2016). The landowner’s ability to dispose of fertile farmland
in favour of photovoltaic developments, and the flexibility around the possible risks of land-use change,
are being closely monitored by landscape planners since clear directives are being enforced by ecosystem
service experts who aim to minimise the fragmentation and alienation of valuable land in the agricultural
sector (Hoffacker et al., 2017).
Since agricultural sustainability describes a multidisciplinary goal in a whole-systems approach (impact-
ing on energy development, environmental protection, economic feasibility and food security) (Luca et al.,
2015), the thesis’s research questions contemplate a more holistic agri-friendly sustainability valuation strat-
egy for floatovoltaic technologies that better aligns with agricultural production sustainability. Furthermore,
in considering the integrated model to assess environmentally sustainable agriculture given by Figure 2.8
(Alkemade et al., 2012), the thesis framework development effort can be guided by the requirements for
sustainable farming certifications.
Certifications for sustainable agricultural practices and stewardship over the natural habitat in terms of
the principles depicted by Figure 2.8 constitute an integral part of sustainable fruit production and wine grow-
ing. As such, floatovoltaic technology generally complements the sustainability farming ideology since the
technology actively supports progressive environmental conservation principles and the associated qualifi-
cation criteria support the adoption of FPV technologies (Allen and Prinsloo, 2018).
Since protected harvest-type certifications continue to gain traction with consumers internationally (IFC,
2020), sustainability certifications are essential in an era where energy project investors prefer conservation-
oriented projects and production methods. At the same time, Furthermore, in terms of the environmental
impact protocols of the fruit and wine production sector, sustainability-conscious certifications (eco-label
certifications, protected harvests, food labelling and health-grown labels) involve inspections and audits on,
amongst others, aspects such as the reliability of resources for electrical power generation (Moscovici and
Reed, 2018; Sonke, 2008). While conservation-sensitive floatovoltaic technology generally falls under the
solar farming sustainability criteria, its natural resource conservation attributes may inspire higher rankings
on the sustainability index under SCS and ANSI standards because it enhances the main qualifying drivers,
such as the carbon cost of wine determined as part of the vineyard’s carbon budget or C-cycle (Decock,
2020; Scandellari et al., 2016). While conservation-sensitive floatovoltaic technology generally falls under
54
(a) (b)
Figure 2.8: Agricultural sustainability frameworks in terms of (a) resources-based WEL-nexus pillars (after: McCarl
et al. (2017)), and (b) OECD’s integrated model to assess environmentally sustainable agriculture (source: Alkemade
et al. (2012), page 2).
the solar farming sustainability criteria, its natural resource conservation attributes may inspire higher rank-
ings on the sustainability index (under SCS and ANSI standards). This improvement is because it enhances
the main qualifying drivers (e.g. the carbon cost of wine determined as part of the vineyard’s carbon budget
or C-cycle) (Decock, 2020; Scandellari et al., 2016).
Given the range of environmental benefits and co-benefits of floating photovoltaic systems as clean,
green, renewable energy sources, the technology is set to emerge as a dynamic and pragmatic energy
source and a self-generating solar farming solution, positively inclined to support the sustainability farming
discourse. Having considered the various impact opportunities that floatovoltaic technology can offer over
land-based photovoltaic projects to help sustainable farming practices, the following section discusses some
of the international options for the technology. It also highlights the practical challenges caused by the lack
of modelling tools to prepare due diligence and impact assessments for new project development approvals.
To further emphasise the practical diffusion-support requirements for a real-world technology, the follow-
ing section describes the diffusion trends of international technology and the impressive worldwide growth
and expansion that floatovoltaic technology projects have made over recent years.
2.4. Developments and Challenges of Floatovoltaic Technology
This section surveys global developments in floatovoltaic technology, as well as the practical and scientific
challenges in the technology’s sustainable development discourse. It focuses on the significance and ur-
gency of the research and the functional requirements to help achieve the research aim and purpose of the
study within the topological field of global sustainable development. Section 2.4.1 explains the phenomenal
growth rates in applying floating solar technology around the globe, while Section 2.4.2 highlights the po-
tential for the development and diffusion of new floating solar energy solutions in the South African region.
The regulatory readiness considerations for Floating solar PV technology are discussed in Section 2.4.3.
2.4.1. Global-scale diffusion potential of floating solar technology
This section focuses on the explosive growth of floating solar-powered generator technology worldwide and
the role that its diffusion can play in supporting the global and local energy transition drive in regions where
arid climatic conditions leave some regions experiencing acute freshwater shortages. It also deals with
the challenges experienced internationally with project licensing and environmental approvals. The chief
55
problem is the limited ability of analytical prediction models to quantitatively assess the broad spectrum of
sustainability and feasibility parameters for floatovoltaic technology.
From a global technology diffusion perspective, state-of-the-art climate-solving floating solar technology
offers significant opportunities for generating an additional supply of energy to aid the global expansion
towards fuelling and penetrating a rural distributed renewable energy market. This trend has been motivated
by the UN SDGs (World Bank, 2019b; TNO, 2021). Analyses and projections in a recent study by the World
Bank (2019b) have reported explosive growth rates in the worldwide installation of floatovoltaic systems
over the last five years. The trend is depicted graphically in the global growth curve for the technology in
Figure 2.9, which portrays the aggregate installed capacity.
Figure 2.9: Cumulative accrual of green energy from floatovoltaics worldwide, emphasising the exponential escalation
in the scope for technology diffusion (source: World Bank (2019b), page 4).
Moreover, the International Renewable Energy Agency (IRENA) emphasises the urgent need for the
diffusion of renewable photovoltaic technology, especially when global renewable electricity production is
needed to deploy new technologies as private or district energy systems. The deployment is required to
grow eight times faster as a means of helping to control average global temperatures and global heating
levels (Cazzaniga and Rosa-Clot, 2021; The Guardian, 2021). Based on the World Bank (2019b) projections
for technology diffusion, researchers expect the global diffusion rate to grow by an average of 22% on a
year-to-year basis from 2019 through to the year 2024 (Cox, 2019a; Cuce et al., 2022).
This trend means that floatovoltaic installations could satisfy about two per cent (2%) of the global
annual solar demand by the year 2022, as they had already delivered just under one per cent (1%) of the
requirements of solar’s international energy projects by the end of 2019 (Cox, 2019b; NREL, 2018a).
One of the primary reasons the world is adopting floatovoltaic technology systems so rapidly is linked
to how the technology foregoes the need to acquire large tracts of agricultural land. At the same time, it
also delivers increased energy outputs, together with a range of environmental side benefits, such as the
preservation of water and improvements to the quality of the water (NREL, 2019a; World Bank, 2019b).
The floatovoltaic growth trends and future estimates detailed above highlight how floating solar energy
technology systems are paving the way for the water-rich regions of the world to significantly scale up the
use of solar renewable energy, particularly in the faster-growing areas of the world (Weigl, 2021; Wood
Mackenzie, 2019).
Several studies are already zooming into the future potential for the continental-wide diffusion of floating
solar technology (Dunham, 2021). A study by (Spencer et al., 2019) used a GIS system with high-resolution
mapping of water reservoirs and dams across the USA to map and predict the spatial distribution and the
anticipated generation capacity through the diffusion of future floating solar systems. This study, commis-
sioned by NREL, offers valuable lessons in helping to determine analytical modelling needs to prepare the
USA for a large-scale roll-out of floating PV systems across the continent (NREL, 2019a). The NREL team,
56
led by Spencer et al. (2019), has zoomed in on the technical potential of floatovoltaic energy systems for
man-made water bodies, arguing that managed bodies of water may bring fewer environmental concerns
than installations on natural water bodies. The geospatial outcomes of the research, as depicted in Fig-
ure 2.10, show the potential spread of suitable water bodies identified by the GIS filtering of geographical
data from the USA Geological Survey Hydrography Dataset, together with that from the National Inventory
of Dams in the USA, thus highlighting the surface area map for water bodies available for FPVs.
Figure 2.10: Technical potential for FPVs on water bodies in the USA (source: Spencer et al. (2019), page 1686)
against a backdrop of price declines driving rapid PV deployment in the USA (after: SEIA (2020)).
According to the available water area in the mapping context for FPVs in Figure 2.10, the Spencer
et al. (2019) study predicts that floating solar systems covering just 27% of the surface area of the identified
suitable water bodies in the USA could produce as much as 10% of the current total US electricity generation
capacity. The floatovoltaic resource potential in Figure 2.10 should be viewed against the backdrop of solar
industry research data, where solar market insights confirm that the solar industry is already booming as
the solar share of new capacity is rapidly growing in the USA as a result of PV price declines (SEIA, 2020).
Together with the geospatial results of Figure 2.10, Spencer et al. (2019) also highlights the broad range of
impact factors that could be quantified as co-benefits when integrating FPVs with human-managed water
bodies. In a related study on the technical potential for FPVs in Europe, accommodating floating PVs on
just 10% of European freshwater reservoirs alone would generate around 200 GW of power (Weigl, 2021).
In a similar study for the African continent, Gonzalez Sanchez et al. (2021) performed an analysis to
assess the technical potential of floatovoltaic technology by profiling reservoir areas at hydropower facilities
throughout Africa. The study evaluated the floating solar photovoltaic potential in the current hydropower
reservoirs in Africa by pairing floating PVs with hydroelectric dams. The results for linking FPV with hy-
dropower facilities on the African continent (refer to Figure W.1) show that FPV panels covering just one
per cent (1%) of hydro-carrying water reservoirs in Africa could double the continent’s hydropower capacity
in terms of an increase in electricity generation. In computing the cumulative surface area of feasible water
bodies for FPV systems and the average state land values, both the USA and African studies on floating
PVs call attention to the extent to which energy technology may be able to help land-constrained areas to
better deal with the demands of competing for fertile, productive agricultural land.
57
The World Bank (2019b) projections for FPV energy development projects in Figure 2.11 anticipate
an accelerated growth period for floatovoltaic technology globally. In this context, the exponential growth
rates for floating solar energy in Figure 2.9 further anticipate a promising emerging floatovoltaic industry
already gaining international momentum. In celebration of the achievements in the growth of technical
capacity in respect of new floating solar power projects worldwide, the World Bank (2019b) has published
the anticipated global development map, as featured in Figure 2.11, for the expected future diffusion of
floatovoltaic technology. The projected diffusion trajectory or geospatial roadmap in Figure 2.11 draws
attention to the dramatic proliferation potential for new floatovoltaic projects internationally, testifying to
an expected continuation in the exponential growth in a cumulative international capacity, as depicted in
Figure 2.9, through the future deployment and installation of new floatovoltaic systems.
Figure 2.11: WorldBank analysis of the technical potential of floatovoltaics worldwide, including South Africa
(blue=installed intensity; red=projected capacity) (source: World Bank (2019b), page 57).
Interestingly, the worldwide assessments in Figures 2.9 and 2.11 point toward valuable opportunities for
floatovoltaic project financing both in Global North and Global South countries (Abid et al., 2021; Rangaraju
et al., 2021; Sathe, 2017), with China being the leading research agent for FPVs amongst the BRICS coun-
tries (World Bank, 2019b; Xia et al., 2022). The financing of modern-day renewable energy projects even
includes options for the long-term leasing of unused water bodies and water reservoirs, making solar leas-
ing or solar rentals for turnkey floating solar solutions a viable commercial option for floatovoltaic technology
(Micheli et al., 2022; World Bank, 2019b).
As floating solar is busy emerging as a leader in next-generation distributed energy innovation, floa-
tovoltaics is poised to establish the third pillar in solar photovoltaic sector development among renewable
energy advocates. This opportunity exists because the technology shows tremendous market potential to
become the world’s third-largest sector for photovoltaic energy project deployment after ground-mounted
and rooftop solar PV energy systems (Acharya and Devraj, 2019; Haohui et al., 2020).
In this context, the significance of new financing regimes, renewable energy certificates, and research
to support the floatovoltaic technology pathway is exceptionally high in areas where land availability or
land costs are putting constraints on the growth of renewable projects (i.e. water-wealthy regions, typically
characterised by high population densities or high WELF plant-growth densities) (World Bank, 2019b).
Moreover, even government policies and initiatives are starting to support floating solar technology, with a
proviso that experts must scientifically quantify the impact benefits in EIA studies and environmental offset
analyses in the conceptual phases of project plans (DFFE, 2015; Grand View Research, 2020). This new
58
energy technology is thus poised to continue to snowball worldwide, despite various deployment challenges
and knowledge gaps, some of which are addressed in this thesis.
To help reinforce the worldwide organic growth trend and to help meet floating solar project appraisal re-
quirements, the scientific community has identified a rising international need for integrated floatovoltaic an-
alytical planning tools (Armstrong et al., 2020; NREL, 2019a; TNO, 2021). With this goal in mind, Boersma
et al. (2019a) and the World Bank (2019b) have emphasised the distinctive potential for microscale floato-
voltaic systems suitable for agricultural-type man-made water bodies across the world. However, with the
knowledge gaps and technological unknowns being a barrier to entry into the floatovoltaic arena in many
regions, some work remains to be done on the performance prediction aspects of the technology (Cuce
et al., 2022; Spencer et al., 2019). This requirement underlines the need to develop tools to quantify the
impact effects of prospective floatovoltaic installations, especially in support of environmental project plans
per government legislation and regulatory requirements (Grand View Research, 2020; Hernandez et al.,
2019; Prinsloo, 2019). This requirement gives impetus to the research opportunity to create new knowl-
edge through technical and geographical innovations in order to reinforce the knowledge base required
by the fast-spreading adoption of floatovoltaic technology inventions in the rural agricultural regions of the
world (Hernandez et al., 2019; Gonzalez Sanchez et al., 2021; Sylvia, 2019).
Those knowledge gaps around FPV impact analysis raise particular concerns worldwide, including the
general shortage of local contextual information, technical knowledge around the model framework, prac-
tical experience, and the lack of geographical computer tools to perform viability analyses of floatovoltaic
projects within a broader techno-economic, techno-enviro scoping framework (Cagle et al., 2020; Hernan-
dez et al., 2019). From a planning perspective, future envisaged geographical tools/models are needed to
help appraise the future impacts of floatovoltaic technology systems in various countries. Such geospatial
tools should be sensitive to the application context and provide decision-making skills that would reaffirm
and account for the broad spectrum of interdisciplinary impact benefits emanating from local floatovoltaic
systems (Prinsloo, 2019; Spencer et al., 2019). This geographical knowledge gap significantly hampers
or delays EIA analyses regarding the legislative planning and reporting requirements around a country’s
adaptations to the latest climate change conditions and carbon taxation strategies (Cohen and Hogan,
2018; DFFE, 2019a). With a young lifetime of floatovoltaic systems in some countries, there are still many
challenges around the lack of data, tool availability and the absence of a long-standing track record with the
roll-out and diffusion of floating solar systems in the local context of each country.
In summary, this section provided insights into global energy transformation aspects that drive or hin-
der floatovoltaic renewable energy developments and into the appropriate impact attributes of this newly-
emerging energy technology worldwide. The following section gives a broad overview of the local emer-
gence in South Africa of the latest floating solar system designs and underscores the contextual challenges
and opportunities for this technology in the local sustainability farming landscape.
2.4.2. Diffusion potential for floatovoltaic technology in South Africa
As floating solar renewable energy technologies continue to make headway worldwide, the discussion fo-
cuses on the diffusion potential for floatovoltaic technology in the local agricultural landscape. It prepares
the reader for a discussion on the anticipated environmental impact and economic due-diligence analytical
challenges associated with the implementation and diffusion of floating solar technology in the SA land-
scape context.
State-owned supplier Eskom’s backing for net-metering and the self-generation of power through small-
scale embedded generation helps make microgrid-scale floating solar technologies one of the new frontiers
of agro-renewable technologies. Such self-generation in an energy cloud management environment offers
promising growth trends anticipated for the sustainable agriculture sector, especially with Eskom’s time of
59
use tariff structure creating opportunities for energy arbitrage (Prinsloo, 2019; Pritchard et al., 2019). The
local agricultural industry is often severely exposed to the risks of power blackouts, adaptive load shedding,
load curtailment, unreliable grid-power supplies and the vulnerabilities of using diesel generators in rural
farming areas. To counteract load-shedding blackouts, restrictive power curbs, and electricity rationing,
the outstanding potential of floating PVs for solar energy generation comes to the fore (Allen and Prinsloo,
2018; Pritchard et al., 2019). As an international trending solution, on-demand floating PV renewable energy
enables on-site water-embedded power generation in water-rich areas to help mitigate the risks of power
losses and fossil fuel emissions (World Bank, 2019b; New Southern Energy, 2021). Floatovoltaic technology
combines well with land-mounted solar photovoltaic energy generation systems in South Africa (Pritchard
et al., 2019). With the proper licensing for embedded generation through floatovoltaic strategies, farmers
can generate their clean electricity from the non-polluting sunlight resource, with promising success rates
in reducing greenhouse effects and as part of sustainable farming practices (Prinsloo, 2019).
Besides power supply risks, local consumers and farmers face ever-increasing electricity tariff increases.
Such dramatic price increases for electricity from the national energy utility supplier, Eskom, are reflected in
the national electricity tariff increase rates shown in Figure 2.12. In this graph, Moolman (2019) has plotted
the local electricity tariff rates against the comparative inflation or consumer price index (CPI) from 1998
to 2019, with projections for further increases from 2020 up to 2022. These grid-pricing trends highlight
that South African electricity tariffs increased by 446% between 2007 and 2019, such increases being more
than four-fold over 12 years, with an inflation figure of 98% over a similar period.
Figure 2.12: South African electricity tariffs plotted against CPI to highlight the dramatic electricity price increases in
South Africa over the last few years (source: Moolman (2019), page 2).
Based on the currently approved increases over the 2020 to 2021 period, the total increase in local
electricity costs will have increased by 520% from 2007 to 2021. The study concludes that at some point in
this run-away micro-economic cycle, the customer’s willingness-to-pay (WTP) threshold of energy tariff/cost
increases will be breached (WTP being a proxy to consumer cost-benefit analysis) (Moolman, 2019). Farm-
ers and other willing and able energy prosumers will finally recognise the window of opportunity for installing
smart renewable energy systems. While solar energy can increase energy self-sufficiency, reduce pollution
and help to reduce a farm’s electricity and heating bills (Gorjian et al., 2022c), the farming sector is already
actively contemplating solar farming (solar photovoltaics or floatovoltaics) to meet local energy needs, or to
sell electricity back into the grid at a feed-in tariff premium (Moolman, 2019; Pritchard et al., 2019).
At the same time, international market forces are putting tremendous pressure on the South African
agricultural sector, especially wine producers, to advance toward sustainable farming and environmentally-
friendly food and wine production practices (European Commission, 2008; Sonke, 2008). Towards sustain-
ability with the integrated production of wine, organisations such as Sustainable Wine South Africa (SWSA)
are helping to inspire sustainable wine growing, biodynamic farming practices and social responsibility
60
through wine-of-origin seal labels. Environmental awareness inspires food and wine producers to recon-
sider the environmental implications of their energy usage and power generation resources. In aiming at
sustainable farming certifications (Kuschke and Cassim, 2019; Moscovici and Reed, 2018), considerations
towards improving farming business sustainability and resilience through renewable energy diversification
are taking top priority. Solar renewable resources are particularly appealing as an option to support the
sustainable farming discourse as the solar topography of South Africa offers promising opportunities for the
decentralised self-generation of power (van Niekerk, 2010).
With the main focus on sustainable food and wine production, the South African Government and
the WWF support wine-farming landowners by committing to ecological progression through biodiversity-
friendly carbon farming and regenerative farming practices (Kuschke and Cassim, 2019; WWF, 2021). Ac-
cordingly, this sector especially values efforts to improve water and energy efficiencies and conserve natural
farming areas. Such conservation efforts spotlight sustainable farming practices that protect conservation-
worthy agricultural farmland, reduce water usage and implement energy-efficient solutions (WWF, 2021).
These WWF accreditation requirements are inherent sustainability co-benefits of floatovoltaic energy sys-
tem deployments as distributed power generation solutions suited for the South African agricultural sector
(Prinsloo, 2019; World Bank, 2019b).
To meet international sustainable farming goals, South African farmers and wine producers can learn
valuable lessons from the Far Niente floatovoltaic energy system in the USA. Depicted in the aerial image
view of Figure 2.13a, this pictorial representation of the Far Niente Winery’s floating solar system shows the
world’s first hybrid land/water floatovoltaic system with its co-located ground-mounted photovoltaic system.
Figure 2.13a illustrate the concept of a floatovoltaics installation on a water body near an agricultural field,
co-located with a neighbouring land-based GPV installation (Allen and Prinsloo, 2018).
(a) (b)
Figure 2.13: First agricultural floating solar systems (a) in the USA at the Far Niente Winery in the Napa Valley
(source: Gamarra and Ronk (2019), page 39), and (b) in South Africa at the Boplaas Farm Estate in the Franschhoek
Valley (source: Caboz (2019), page 2).
Far Niente Winery has distinguished itself as the leading inventor of new state-of-the-art floatovoltaic
systems, with an innovative floatable racking system design implemented in the agricultural sector on a
global scale (Sharp Corporation, 2008; Wines, 2019). As an early pioneer, which spearheaded the in-
vention and subsequent adoption of floatovoltaic technology, this winery has played a commanding role in
the leading industry around sustainable agriculture and organic farming by applying floatovoltaic technol-
ogy in the eco-conscious viticultural industry (Nowak and Washburn, 2002; Prinsloo and Lombard, 2015b;
Prinsloo, 2019).
61
In taking a leap toward sustainability farming, Western Cape farmers of Boplaas (anno 1743) recently
followed the example set by the Far Niente Winery by inaugurating the first floatovoltaic power plant on the
African continent (see Figure 2.13b (Caboz, 2019). Forming part of the famed Franschhoek wine valley, the
landowners of Boplaas (Marlenique Estate) wanted to improve the energy resilience of their farming busi-
ness and decided on floatovoltaic renewable energy diversification as a means to tackle their intermittent
grid and load-shedding problems. Having followed the tedious environmental impact approval processes
(Pritchard et al., 2019), the result, as depicted in Figure 2.13b, shows the facade of South Africa’s leading
floatovoltaic demonstration plant (installed around March 2019). With this local flagship floating renewable
energy project, the historical Boplaas farm (the oldest family farming business in SA) successfully designed
and implemented the local winery industry’s first floating solar power plant to reduce the farm’s carbon
footprint by 50% (Pritchard et al., 2019). As an experimental pilot plant system premiering floating solar in
South Africa, the newly installed system established SA’s first operating FPV land/water hybrid photovoltaic
variant as a solar farming application installed by expert solar installers (New Southern Energy, 2021).
Since being commissioned in March 2019, the floating green energy solution of Boplaas featured in
Figure 2.13b has significantly increased the farm’s energy independence and energy security while fore-
going the need to uproot historical vineyards or to use up valuable orchard/vineyard land space to install
more photovoltaics (Pritchard et al., 2019). The economies of scale and the cost advantages obtained as
a result of the solar farm’s scale of operation are currently being discounted in microeconomic terms to
verify the on-site monetary savings in preventing production losses owing to fewer power disruptions and
intermittent power cuts during the peak periods of fruit/food production and wine-processing. Scientists will
further perform an accompanying technical cost-benefit analysis in a range of long-term analytical exper-
iments. Proprietary in situ measurements are currently collected to scientifically confirm the direct gains
inherent to the Boplaas floatovoltaic system variant (Pritchard et al., 2019). This scientific scrutinisation of
the Boplaas floatovoltaic system will include evaluations of the observed increased energy yields and the
impact benefits emanating from the water evaporation regimen of the system for that location. However,
proprietary experimental data on these effects may not be released or published. Nevertheless, preliminary
observations by Pritchard et al. (2019) have accorded value to the purpose of the system in terms of its
mitigation of operational risks to the farm around intermittent agricultural power outages arising from grid-
vehicle disturbances, especially as the system has already counteracted the effects of unannounced power
blackouts owing to weather disasters or interferences in the transmission and distribution of power.
The technology is already demonstrating its value proposition in terms of an opportunity to gain lever-
age on Boplaas’s floatovoltaic technology installation to benefit the rest of the South African agricultural
sector. It has brought about an improved balance between nature, power generation and sustainable farm-
ing (Pritchard et al., 2019; Weigl, 2021). As an exemplary environmentally-friendly demonstration site, it
may motivate the local wine industry to engage in the technology to move faster towards the forefront of
gaining international recognition as a leader in the global wine production sector (Kuschke and Cassim,
2019; Nedcor Foundation, 2004). The Boplaas floatovoltaic technology installation may further demon-
strate its ability to remedy sustainability problems, inspiring leaders in environmental care and innovation
who want to cooperate closely in farming and conservation matters (WWF, 2021). Such inspiration helps
to improve the local credentials of floating PV systems, supporting the technology’s proliferation to help
catalyse behaviour changes towards more sustainable farming practices within the wine-growing regions of
South Africa.
While floatovoltaic technology is a novel energy technology configuration earmarked to continue accel-
erating amidst the fight against the global climate crisis (Weigl, 2021; World Bank, 2019b), project licensing
requirements for the technology means that new analytical tools are needed to qualify sustainable invest-
ments in renewable energy technologies. The lack of appropriate research tools to profile floating PV
technology sustainability creates opportunities to overcome the obstacles to project planning consent for
62
agricultural floatovoltaics. Such research can also help improve the regulatory and market readiness of
the technology and stimulate transformative investments in the technology in the South African agricultural
landscape.
2.4.3. Floating solar PV technology readiness considerations
Regarding the diffusion of floatovoltaic technology in South Africa, the awareness created by Boplaas in
the floating PV technology domain has already borne fruit. For example, Alzanne Fruit and Vegetable Farm
in Vredendal recently installed an FPV system on a local water weir on the farm to support their zero-
carbon footprint sustainable farming goal (Laubscher, 2022). Furthermore, South Africa’s Water Research
Commission (WRC) recently also joined hands with the University of Cape Town (UCT) to establish a new
floating PV pilot plant at the City of Cape Town’s Kraaifontein Wastewater Treatment Works and Sewage
Farm (Smith, 2021). Since this project aims to support the smart city concept, this WRC floater installation
includes a hybrid floating solar PV array with a ground-mounted PV system to serve as a research pilot. The
goal is to determine the evaporation savings and relative energy generation performance of floating solar
PV technology in an urban environment. The development of the Boplaas Fruit Farm and the Kraaifontein
Sewage Farm floating PV pilot plant is already significant, as these demonstration prototypes have heralded
the geographical presence of floating solar technology in the South African landscape.
Generally, the international country-specific Technology Readiness Level (TRL), Technology Regulatory
Readiness Level (RRL), Investment Readiness Level (IRL), and the Technology Marketing Readiness Level
(MRL) scale reflect technology readiness for diffusion in a particular region, market or application. These
scales, portrayed in Figure 2.14, were initially developed by NASA to reflect the increase in the value
of the technology during the valorisation process and to serve as a Return-on-Innovation indicator (ROI)
(Goldense, 2017). Regarding the TRL criteria, the Boplaas floatovoltaic technology prototype installation
rates relatively high on the TRL scale, possibly between six and seven. However, due to the lack of analytical
tools to measure or predict the sustainability profile of floatovoltaic technology in South Africa, it delivers an
unfavourable rating on the Technology Regulatory Readiness Level rating scale.
Figure 2.14: Floatovoltaics technology in SA in the demonstration phase rating quite favourably on the TRL rating
scale. However, the lack of analytical tools to measure or predict the sustainability profile of the technology in SA
cause an unfavourable rating on the RRL and MRL scales (source: Kobos et al. (2018), page 29).
The excellent environmental profile of floating PV technology, combined with a supposedly high TRL
rating for floatovoltaic technology in SA, can create new leverage points for the local sustainability transfor-
63
mation, thus offering a unique value proposition for sustainable farming accreditation. Furthermore, judging
from the awareness of the local technology domain created by the Boplaas floatovoltaic system in terms
of its positive environmental attributes, this pioneering installation in the Cape Winelands may catalyse the
increased proliferation of local floating solar installations and investments. As a result, the expected diffuse
rate of floating solar technology in South Africa may rise significantly over the next five to nine years (World
Bank, 2019b). This positive expectation is eminent from discussions with local and international engineering
experts, Meyerhof and Prinsloo (2019), and Pritchard et al. (2019), a view further supported by international
floatovoltaic landscape survey studies (Cox, 2019b; Wood Mackenzie, 2019).
However, a narrative of increasing the number of FPV project plans would trigger large numbers of
listed EIA activities, and critical knowledge gaps still exist around the scientific sustainability assessments
for floating solar projects in EIA studies. These knowledge gaps’ impact effects perpetuate due to a lack of
empirical data on the performance of floatovoltaic systems in the local context. Understandably, given the
short lifespan of the ground-breaking agricultural-scale prototype in the country, performance profiling and
project bankability evaluations for the South African context would remain a barrier to floatovoltaic deploy-
ment (Pritchard et al., 2019). From a geographical, computer science and information systems engineering
perspective, the solution should come from data science studies and computer simulation models to speed
up this 4IR technology transfer cycle. Given the short lifetime of the first physical, real-world local floating
solar system installation in South Africa, newly proposed geo-aware digital models are crucial to support
learning toward an anticipated increase in local floating solar technology installations (Prinsloo, 2020), de-
spite data availability limitations.
From a local South African context perspective, virtually no other geographical studies have quanti-
fied the environmental and economic co-benefits of FPV technology in South African conditions (Prinsloo,
2020). Furthermore, there are limited capacity and research funds dedicated to geoinformatics tools to
support the EIA processes for floating solar technology, apart from the related studies that led to the intro-
duction of this research project (Prinsloo, 2017b, 2019). In contrast, this study focuses on developing a new
philosophical framework as a computer simulation logic to improve FPV technology’s regulatory readiness.
The study anticipates that the theoretical research of this thesis may be of broader significance in terms of
its implications for policy and utility in impact assessment studies.
This section spotlighted the local and international introduction and anticipated diffusion of floating so-
lar technology. The following section provides an overview of the state-of-the-art position around existing
analytical best-practice tools to analyse and assess the energy and environmental performances of floato-
voltaic and photovoltaic systems to help diminish the complexity of the appraisal challenges of real-world
geographical projects.
2.5. State-of-the-Art in Floating PV Sustainability Profiling
This section presents a condensed precis on the underpinning best-practice methods in floatovoltaic perfor-
mance profiling from a methodological framework and a modelling tool perspective. Providing a systematic
review of photovoltaics modelling, it details the current state-of-the-art in sustainability assessment theory
and practice. While the discussion considers techniques developed by other scholars to assess floatovoltaic
systems attributes in similar studies, it deals with past and present research into assessments of floating
solar photovoltaic solar systems in various industrial or agricultural applications.
The pre-construction evaluation and justification of experimental photovoltaic systems should ideally be
based on mathematical models and computer simulation programs as desirable real-world alternatives. In
this respect, the IEA Photovoltaic Power Systems Programme (PVPS) aims to lower the barriers and costs
of PV grid integration in the transition to sustainable energy systems while lowering PV systems planning
64
and investment costs through enhancing the performance forecasting quality of forecasts and resource as-
sessments (IEA, 2017). The broad spectrum of environmental implications, diverse impact profiles, and
impact-effect variations around the floatovoltaic habitat are causing practical analytical assessment issues
and challenges, stressing the need to analytically model key aspects at the root of these impact disjunc-
tures (Cagle et al., 2020; Gadzanku et al., 2021b; Prinsloo, 2019). The nature of the hierarchically-layered
techno-enviro economic benefits serves as an incentive for further research to close the gaps in the current
understanding of floatovoltaic technology impacts (Haohui et al., 2020; Gensol Engineering, 2018; Gorjian
et al., 2021; Tawalbeh et al., 2021).
Historically, accurate PV energy system modelling efforts benefit from the Sandia PV Performance Mod-
elling Collaborative (PVPMC) initiative, where stakeholders regularly meet, intending to advance the "state
of the art” in PV performance-prediction modelling and practice (IEA, 2017; Sandia, 2022a). Under this ini-
tiative, the current real-time synchronous analysis uses photovoltaic array performance models and meth-
ods for the pre-installation prediction of conventional overland PV system performances. These have thrived
for several decades, thus significantly contributing to understanding the technical and techno-economic per-
formances of ground-mounted PV systems (Sandia, 2020). Apart from combating climate change, such dis-
crete time-analytical models and methodologies in open-source modelling libraries are generally applicable
to the sizing and design of land-based PV systems, the optimisation of systems performance, estimat-
ing field installation performances, predicting the economic viability of land-based PV systems, translating
predicted-performance measurements into standard reporting conditions, and the real-time comparison of
model measured versus expected systems performance in ground-mounted PV systems (King et al., 2004;
IEA, 2017).
With collaborative R&D being essential to a resilient, clean, renewable energy future, Sandia and NREL
engineers have succeeded in applying the industrial standard models (i.e. PVsyst, PVlib, PVwatts) to de-
velop an array of extended photovoltaic computer simulation tools for ground-mounted PC systems, as
depicted in the Spatio-temporal Power Systems Model Map of Figure C.1 (NREL, 2018b). However, many
of the universally used land-based photovoltaic performance-modelling tools in this map (REopt, Homer,
ReEDS) have been custom developed around ground-mounted PV systems (NREL, 2020, 2015; Aznar
et al., 2019). Therefore, they do not yet fully meet the environmental conditioning requirements and analyti-
cal flexibility required to accurately model floating PV performance in multi-scale sustainability assessments.
The model simulation tooling developments mapped in Figure C.1 are universally used land-based photo-
voltaic performance-modelling tools, making them valuable industrial standard GPV-specific models with a
ground-mounted PV analysis focus to serve a technical or techno-economic analytical purpose.
As such, the problem in the model map of Figure C.1 is that most of the PV engineering-performance
models in the map do not as yet include any conditioning options to accommodate floatovoltaic modelling
variations. Although there have been some planning efforts in working towards the goal of modelling FPV
from an engineering-design-support perspective (Aznar et al., 2019; NREL, 2019b; Spencer et al., 2019),
most engineering models would not necessarily include or aim at broad-spectrum environmental impact
modelling. However, NREL’s endeavours over the years have led to the creation of the most valuable
mathematical models and software platforms to support the original ground-mounted photovoltaic system
modelling goal, among those depicted in Figure C.1, some of which can be conditioned toward floatovoltaic
energy system analyses. Such models and software tools allow for geo-modelling, which in this instance
is built upon relevant technical content and business intelligence established in the decision sciences, sci-
entific data management and high-performance computing. The models further incorporate computational
science, applied mathematics and informatics principles in data processing and the visualisation of scientific
detail in decision-support systems (Lufkin, 2015; NREL, 2015, 2019a). It is anticipated that NREL will de-
velop future engineering PV performance modelling tools to accommodate floatovoltaic modelling software
tools in the modelling map of Figure C.1. Although future floating solar tools are likely to be developed more
65
from a techno-economic engineering perspective (Gadzanku, 2020), the literature review (Section 2.3.2)
showed that floatovoltaic technology has taught that environmental modelling should form an integral part
of futuristic analysis methodologies for floatovoltaic performance evaluations (Cagle et al., 2020; Luderer
et al., 2019; Prinsloo, 2019).
In the context of real-time floating PV modelling improvements, the range of valuable land-based PV
modelling software platforms developed by the Computational Science Programme of the National Renew-
able Energy Laboratory (NREL) provides crucial computer support to assist solar energy research suitable
for generic applications. Despite most PV modelling solutions being strongly biased towards land-based
PV systems in linear-type unidirectional processing frameworks, these engineering efforts provide a valu-
able basis for modelling the power generation aspects of ground-mounted photovoltaic systems on software
tool platforms such as the industrial standard models PVsyst and PVlib (IEA, 2017; Sandia, 2020; PVsyst,
2022). The PVlib Toolbox, for example, provides a set of documented functions to theoretically model
ground-mounted photovoltaic energy systems through computer simulations (Sandia, 2020). A range of
similar PV laboratories, offering solar PV-designed software tools, solar resource assessment models,
prospecting tools for energy yield assessments, and long-term PV electricity production models, include
algorithms such as PVsyst, PVcase, Aurora solar, BlueSol, PVsol, PV CAD, Solarius PV, TrnSys, Energy-
Plus, EnergyPlan, PV Planner, EasySolar, PV Sketch, SolarEdge site designer, INSEL, PV Sim, PVpmc,
PV-Watts, PVform, Hysim, F-chart, RETscreen, PV Complete, Solargis iMaps, pvPlanner, Archelios or Poly-
sun (Holmgren et al., 2017, 2018; Klise and Stein, 2009; Mayer and Gróf, 2021; Santiago Silvestre, 2012).
The topographical solar PV design tool Helioscope (HelioScope, 2021) and the geographical information’s
System Advisory Model (SAM) (Gilman, 2018), have already laid solid foundations for the techno-economic
characterisation of photovoltaic system performances, these models are not as such yet directly suitable to
floatovoltaics.
NREL’s vanguard SAM power systems advisor model specifically offers a software platform that en-
ables detailed technical performance and financial analysis for renewable energy systems (Freeman, 2022;
NREL, 2022). As a parametric desktop application, shown in Figure 2.15, it is one of the formal practical
(virtual) solar modelling tools mapped in Figure C.1 which is especially relevant to the research in this thesis.
However, the SAM virtual power plant modelling solution caters mainly for contextually regulatory conditions
and funding mechanisms pertinent to the US landscape and does not give much attention to environmental
modelling required for EIA reporting.
Figure 2.15: Model parameters and processing steps of NREL’s techno-economic system’s advisory model SAM for
land-based photovoltaic systems (after: NREL (2015)).
While SAM offers a software routine and components for modelling the energy and economic per-
66
formances of solar energy systems in Figure 2.15, its capabilities focus mainly on the techno-economic
domain. The modelling of floating PV performances in SAM is planned as a potential future development
(Freeman, 2022). The present SAM computer simulation model, together with the SAM modelling steps
in Figure 2.15, offers the technical capacity to calculate the hourly energy output for a renewable energy
system over an entire year of photovoltaic and solar plus storage solutions (Eurek et al., 2013; Gilman,
2018). Although some attempts to adapt SAM for comparative floating PV and land-based PV analyses
exist (Makhija et al., 2021), the SAM model currently supports only cost-benefit research for the traditional
grid-connected land-based photovoltaic projects (Gilman, 2015). While the SAM model offers interesting
features suitable for analytics in the economic context, it also focuses more on the USA’s economic con-
text. Moreover, it provides a less focused view of environmental impacts, making it less suitable for the
environmental sustainability modelling of floatovoltaic installations for local SA conditions. Apart from com-
puting the carbon footprint of a PV installation, neglecting the symbiotic microscale environmental feedback
in water-based PV system characterisation compromises the performance-prediction accuracy in linear or
open-loop floating systems assessments. From an FPV modelling perspective, this aspect is considered a
universal limitation of virtually all of the state-of-the-art solutions described in the rest of this section.
Engineers have developed similar unidirectional techno-economic models to model the energy and eco-
nomic assessment of floating photovoltaics in reservoirs to determine their cost competitiveness and the
role of temperature (Dizier, 2018; Micheli, 2021). Dizier (2018) specifically modelled the benefits of ac-
tive cooling interventions on floatovoltaic technology as part of the goal of the Ciel et Terre R&D team
to improve FPV systems efficiencies. In this application, the researchers developed an open-loop Matlab
computer model for studying the techno-economic yield analysis of future planned floatovoltaic systems
spray-cooled with water (Dizier, 2018). The FPV techno-economic model, depicted in Figure 2.16, is se-
quentially linear by nature, in the sense that the model considers impact analysis linearly, which differs from
the interconnected causal system dynamics approach proposed by this thesis.
Figure 2.16: Matlab model and procedures for techno-economic impact assessments in active floatovoltaic cooling-
technology propositions (source: Dizier (2018), page 4).
The Dizier (2018) study is unique in a strategic sense in that it models a proposed technically superior
modification of a floating solar system that adds active cooling as a design option. The assessment model
design in Figure 2.16 digitally tests the performance merits of a floatovoltaic technology improvement option,
predicting the impact gain when using on-site water to cool down the surface floating panels by regularly
spraying irrigation water onto them (Dizier, 2018). The dedicated Matlab model has an engineering focus,
67
designed around the virtual testing of the techno-economic performance improvements when implementing
an active water cooling system on a floating power plant.
Also, considering floating PV systems analyses, similar type of linear or unidirectional techno-economic
model has been developed by Goswami et al. (2019); Semeskandeh et al. (2022). These studies focused
on evaluating the performance of floatovoltaic systems from a sustainable energy progress angle, adding
economic and carbon footprint analysis as post-processing analysis. As such, these studies perform a
techno-economic feasibility analysis for a floating solar project to determine the factors that drive and gov-
ern the benefits of the technology. The experimental evaluations of Goswami et al. (2019) developed a
pilot plant model to test the (environmental) carbon footprint suitability and economic viability of floatovoltaic
technology towards maintaining an ecological balance. By adjusting the temperature of the floating PV
modules, the model predicted that the floating photovoltaic power plant could offer yield increases by as
much as 10.2% when compared to the land-based systems. The water-based power plant’s increased gen-
eration capacity can deliver such gains over the project’s entire life cycle. The floating solar system reduced
the levelised cost of electricity by around 36%, which is remarkable when considering that land-based sys-
tems deliver energy at the cost of US$ 0.026 kWh. While such techno-economic-focused approaches are
valuable, it falls short in predicting and documenting the full extent of the broad-spectrum environmental
benefits/co-benefits required by EIA reporting for floatovoltaic projects.
The technical, economic, and environmental feasibility of floating solar power generation has been
studied by Bui (2019) for supplementing hydropower capabilities in Vietnam, by Testori (2021) for energy
development in Dutch cities, and by Makhija et al. (2021) for the electrification of remote areas in India.
In. terms of energy modelling, Bui (2019) used the European Commission’s Institute for Energy and Trans-
port (IET) Photovoltaic Geographical Information System (PVGIS version 5), Testori (2021) used SketchUp
Pro with PVsyst 7.2, while Makhija et al. (2021) used the conventional SAM model. These models do not
actually model floatovoltaic technology as such but instead use conventional photovoltaic tools to deter-
mine energy outputs based on the overall scale of a photovoltaic project on water surface coverage. From
the estimated energy production figures, these studies progress toward the prediction of the selected en-
vironmental factors and economic costs in terms of financial variables. The environmental models of Bui
(2019); Testori (2021) estimates the water evaporation benefits of the floating cover (by way of Penman
formulas). The Bui (2019); Makhija et al. (2021); Testori (2021) approaches all follow linear unidirectional
modelling guidelines in which elementary technical, economic and environmental assessments are carried
out individually and separately from one another. However, these FPV models fall far short of meeting all
the feedback-type environmental and sustainability impact assessment requirements and are thus unable
to model the microclimate and hydroclimate feedback effects observed in real-world FPV systems.
In South Korea, prospective mainstream investments in the generation of floating photovoltaic energy
and its feasibility features as a renewable energy alternative are attracting increasing interest from Korea’s
research fraternity (Kim et al., 2016b; Suh et al., 2020). In terms of the performance modelling of the
planned floating solar systems, some progress are made in research towards developing planning-support
platforms (Choi, 2014; Kim et al., 2016b). One Korean engineering study, for example, has quantified the
environmental impacts of medium-sized plants and has shown that returns on investment for such systems
could be as much as 10% higher for floating solar systems as opposed to those for land-based systems
(Choi, 2014). The study highlights the importance of performance modelling in analysing and investing in
floating solar technology and its diffusion processes (CIF, 2018). However, it still leaves significant gaps
in its environmental impact predictions for newly planned floatovoltaic systems. Such economic research
into floatovoltaic experiments offers valuable insights to assist with performance modelling of floating solar
performance as opposed to performance modelling of ordinary land-based solar power systems.
In considering decision-support research towards the promotion of floating solar technology, new mod-
els have been defined to evaluate the optical conversion rates for floating solar irradiation for floating solar
68
power generation schemes installed on water storage facilities (Martins, 2019; Perez et al., 2018). In this
context, Marstein et al. (2020) developed a 3D computational fluid mechanics framework and model to pre-
dict the cooling effects of floatovoltaic systems. Lindholm et al. (2021) continued this work towards devel-
oping a numerical model to compute the module efficiency by determining the operational cell temperature
for a given FPV technology with the efficiency of heat loss to the environment. This effort follows earlier
research in Italy, where Tina et al. (2018) worked on developing software tools to analyse the geographical
and technical potential of FPV systems, including the performance capabilities of different floating photo-
voltaic designs. Various geographical indicators were integrated and combined to determine a technology
suitability factor, subsequently used to evaluate usable land or water space in both land- and floating- solar
structures. The analysis results prove that the available water surfaces can accommodate floating solar
structures. At the same time, engineers can develop improved modelling techniques to help predict the
potential of the energy outputs from floating solar plants and their water preservation benefits, as opposed
to those from land-based solar plants (Cazzaniga et al., 2018; Scavo et al., 2020; Tina et al., 2016).
Further, Rosa-Clot and Tina (2017) FPV carried out experiments with modelling designs and case stud-
ies to predict the performance of floating PV, submerged photovoltaics, and water immersion on PV panels.
The research facilitated scientific investigations into computer programmes to potentially model the inte-
gration of water-based solar power plants into electricity generation and the impact of these systems on
the environment. The study employs complementary simulation techniques in decision-support analyses to
enhance the learning experience and offers valuable theoretical and practical explanations to enhance en-
gineering design options for photovoltaic energy systems. Rosa-Clot and Tina (2017), together with Scavo
et al. (2020), further engaged their computer models to investigate the extent to which water-based floating
solar systems can increase cost-effectiveness while creating positive synergies with smart environmental
and economic investment decisions. The research offers significant opportunities to apply engineering-
modelling approaches to the field of geography to facilitate selected environmental impact assessment
aspects for water-borne floating photovoltaics or water-submerged solar energy technologies.
More recently, Kumar and Majid (2021) studied the potential for floatovoltaic power plants to ensure a
rapid transition to a green energy and sustainable future in India. The study was extended to unveil the
opportunities and challenges with the development of floating photovoltaic systems (Manoj Kumar et al.,
2021; Kumar et al., 2021). A subsequent comparative analysis of field-measured and simulated energy
performance led to the realisation of the need to develop a new simulation tool specific to floating solar pho-
tovoltaic systems (Kumar et al., 2022). Assessment of the potential of different floating solar technologies
is further complicated by different FPV energy conversion configurations (Oliveira-Pinto and Stokkermans,
2020). Refaai et al. (2022) presented the implementation of a floating PV model to analyse the system’s
power generation. While these research sources provide valuable information consulted in this thesis, most
of these models focus on the technical performances (energy output) of floatovoltaics and do not deliver a
sufficiently broad spectrum of outputs required for mandatory regulatory project approvals.
From a South African contextual perspective, a local environmental research enterprise has made
progress in preparing a generic carbon footprint assessment tool to help the local agricultural industry
prepare for the introduction of South Africa’s new carbon tax legislation (RSA, 2019a). This geographical
tool named the Carbon Footprinting Tool (CFT) was developed by the professional bodies of the local fruit
and wine industries as a means for farmers to perform onsite carbon-impact assessments in respect of their
operations (Blue North, 2017; Blignaut, 2017). This carbon management tool has been brought into exis-
tence under the Confronting Climate Change Initiative, a broader paradigm initiated to help farmers smooth
the transition to the Carbon Tax Act (RSA, 2019a). They are based on international carbon accounting
and reporting tooling protocols (FIVS, 2008), the CFT-tool aims to provide environmental information in the
local context to support realistic onsite carbon auditing. However, the CFT tool lacks the specific proper-
ties for dealing with sustainability impacts across the full spectrum of impacts required for analysing local
69
floating solar systems. The tool predicts only the impact’s carbon dioxide (CO2) dimension and cannot
deal with PV energy yield analysis or other environmental assessment issues. Furthermore, as far as can
be determined, no provision has yet been made for technology-specific analyses of FPV systems or of
the techno-economic analytical properties required in broad-spectrum sustainability analyses for assessing
floatovoltaic environmental impacts.
Continuing in the South African context, Stellenbosch University researchers (Gurupira and Rix, 2016)
performed experimental comparisons between two of the most popular industrial-standard models, namely
the PVlib (Sandia, 2020) and PVsyst (PVsyst, 2022) platforms for ground-mounted photovoltaic systems.
The research found that the PVlib model application as a software development kit (SDK) generally of-
fers greater flexibility, while the model’s performance credibility proved its suitability for Southern African
conditions. While the PVlibrary algorithm does not as yet make provision for floating solar system mod-
elling, this thesis can provisionally condition the PVlib inputs for floatovoltaic applications with simplified
yield workarounds. This work started in the GIS-based performance-profiling tool for floating solar devel-
oped by the author of this thesis in earlier floating solar research (Prinsloo, 2017b, 2019), and potentially in
Profloating’s online Flotar calculator for floatovoltaic performance analysis on demand (Profloating, 2021).
In the South African landscape context, the need arose to fulfil the requirements of EIA assessments for
upcoming floating PV technology installations in the local agricultural sector. Starting early geographical re-
search on the technical performance and environmental credentials of floatovoltaic energy systems in South
Africa, Prinsloo and Lombard (2015a) identified a geographical research gap in local environmental impact
assessment narratives around energy development planning for local context floatovoltaic system installa-
tions. To support EIA integrations into floating PV modelling, Prinsloo (2017b) originally defined a TrnSys
model and later developed a PVlib-driven model to drive a basic geospatial platform. This project and ini-
tiative aimed at pioneering PV systems modelling to help guide energy developers and impact practitioners
through the complex maze of environmental impact assessment procedures for agricultural-type projects for
floating solar energy infrastructures based on floating and hybrid land/water photovoltaic systems. Prinsloo
(2019) continued to improve this floating PV assessment model when he conceptualised and developed the
EIAcloudGIS tool for assessing specific floatovoltaic environmental impact effects. The schematic diagram
of this integrated model in Figure 2.17 shows how this real-time synchronous FPV modelling methodology
duly focuses on integrating the "natural environmental system” into the floating PV ecosystem model.
GIS Host Layer z: WELF Nexus Objects
GIS Host Layer y: EIA Objects
SolarM odel
Energy
Simulation
Model
Environmental
Simulation
Model
kWs(t)
kWe(t)
Pe(kWh)
H2O (kg)
SOx(kg)
NOx(kg)
CO2(kg)
Solar data
Weather data
Climatic data
Figure 2.17: Schematic diagram of energy and environmental simulation model objects embedded in a GIS tool to
appraise floatovoltaic performance profiles (source: Prinsloo (2019), page 36).
The model, depicted in Figure 2.17, outlines the energy and environmental simulation model objects
embedded in a GIS tool to appraise floatovoltaic performance profiles in a balanced electronic EIA score-
card and reporting approach (Prinsloo, 2019). As an upgraded GIS version of his conceptualised floa-
70
tovoltaic computer simulation model, the geo-aware computer model can mimic some of the energy and
environmental-offset impact performances of a real-world floatovoltaic system. This Google-maps-enabled
web-based analytical floating PV design tool includes features aimed at recreating the real-time perfor-
mances of a floating solar PV system, reshaping the floating PV project-proposal generation process
through the definition of balanced electronic scorecard metrics. The execution of this linear model allows for
the preliminary characterisation of the performance of a floating solar system using a statistical analytical
approach to help practitioners navigate the EIA process. On a strategic level, the research established
the groundwork for computational modelling and simulation techniques for determining some grid-replacing
environmental offsets gained from floatovoltaic energy systems. Prinsloo’s (2019) GIS-based computer
modelling tool, transliterated in digital software form (Figure 2.17), ventures towards the idea of using com-
ponents and software objects in pursuit of a geographically-motivated descriptive virtual floatovoltaic power
plant model. On this geospatial platform, the contextualisation of the floating solar impacts for local environ-
mental conditions embedded in the GIS host layers of Figure 2.17. The novel software objects embedded
in the GIS layers in the model of Figure 2.17 allowed the research to simplify the critical aspects of the
procedures for local environmental impact analyses for newly envisioned floatovoltaic systems installations
in SA’s farming arena (Prinsloo, 2017b, 2019).
In conclusion, this section considered the current best-practice and state-of-the-art tools, methods and
approaches in terms of current geographical engineering performance modelling, sustainability analysis
and assessment frameworks in the floating solar scenario analysis domain. The main conclusion inferred
from the research is that current modelling efforts focus more on "performance modelling” than on the
more complex "sustainability modelling”. Furthermore, the review highlights how floating PV behaviour
analysis, performance assessments and sustainability valuations are three individually different modelling
concepts with different objectives and scales, with sustainability appearing to be at the pinnacle of the
hierarchy. In this context, the review highlighted the definitional and territorial difficulties in conjunction with
the controversies with theoretical modelling approaches toward project-related sustainability appraisals of
floating PV energy technology. The review finds that current best-practice efforts still struggle to fill all the
knowledge gaps around floating PV behaviour and performance modelling, which in turn causes a range
of "technology impact unknowns” and "technology impact unknowns”. This baseline void in current models
leaves real-world problems in EIA, as it constitutes barriers to entry into the floatovoltaic arena for many
regions (Spencer et al., 2019; Gadzanku et al., 2021a). At the same time, the review acknowledges that we
can still learn valuable lessons towards improving pragmatic model features t from international research
teams and scholars who are currently experimenting with real-world floating solar projects and case studies
(NREL, 2019a; SERIS, 2021; Verbruggen, 2018; World Bank, 2019b).
In proposing a new analytical solution for floating PV project assessments, the following section high-
lights the scope for innovation in the current thesis. It also provides guidelines for a solution to fill the
knowledge gaps around the anticipated challenges associated with integrated floating solar technology as-
sessments to support the diffusion of floatovoltaic technology worldwide.
2.6. Scope for Innovation and Creation of New Knowledge
While the literature review articulates the need for improvements to floatovoltaic technology assessment
modelling, this section summarises the data observations towards defining the scope for innovation and
the creation of new knowledge in the field of geographical information sciences to support floating PV
characterisation. It concludes the literature review by emphasising the challenges and opportunity areas
where current model deficiencies in state-of-the-art approaches leave knowledge gaps in the field of sus-
tainability analysis. The discussion further highlights opportunities for innovation towards conceptualising a
71
research hypothesis for a theoretical framework that could ensure improved theoretical characterisation of
agricultural-type floatovoltaic technologies.
Overall, the literature study introduces floatovoltaic technology as an exciting new eco-friendly cleantech
energy solution in support of sustainable agriculture (Allen and Prinsloo, 2018; Pritchard et al., 2019). In this
context, the review highlights that floatovoltaic technology is environmentally wise and geared to help ad-
dress the three-fold real-world challenge of sustainable land development, generating additional carbon-free
energy while conserving scarce water supplies (Cohen and Hogan, 2018; World Bank, 2019b). However,
the review also highlighted the fundamental disjunctures between the diverse sustainability performances
and impacts GPV and FPV systems are causing serious sustainability assessment challenges, although
sustainability assessment is required to go beyond the conventional climate change measures. The unique
combination of benefits and co-benefits delivered by FPVs, therefore, emphasises the critical knowledge
gaps around FPV performance and impact assessments, as well as the need for a new generation of tools
for enabling synergetic floatovoltaic ecosystem impact analyses as part of floating solar technical potential
assessments (Hernandez et al., 2019; Gadzanku, 2020; Gadzanku et al., 2021b). Strategically, the review
highlights explicitly how the identified knowledge and methodological gaps are emergent from a broad ba-
sis, covering the fields of environmental-, information-, economic-, and engineering- sciences (Armstrong
et al., 2020; Cagle et al., 2020). At the same time, the literature review underscores how the strategic-
, tactical- and activity-level requirements define apparent knowledge gaps and methodology gaps in the
reporting requirements concerning the sustainability features of the technology. It thus underscores the
scope for innovation to support this sustainable energy technology diffusion in the South African floating
solar technology landscape.
In considering potential solutions to this real-world problem from a sustainable development perspective,
the thesis identifies a strategic opportunity to address current real-world challenges through a newly envis-
aged conceptual development framework for FPV system sustainability (Prinsloo, 2020). Such a philosophy
is to draft a novel hypothesis that should be oriented towards formulating a new framework in the form of
a sustainability theory purpose-built for agricultural-type floatovoltaic technologies. From a philosophic-
geographic perspective, the thesis observed a fundamental problem with current PV performance models,
neglecting the importance of the PV system’s interactions with the natural environmental system. It mainly
concerns existing modelling frameworks ignoring considerations around the symbiotic microscale environ-
mental feedback in water-based PV system characterisation compromises performance-prediction accuracy
in almost all linear or open-loop floating system assessments. This environmental void aspect is an inherent
problem with current PV performance models insofar as the limitations of its perspective on the environmen-
tal impact aspects of PV technologies. The thesis considers this one of the universal limitations of virtually
all state-of-the-art PV performance modelling solutions (Void to Opportunity, as described in Section 2.5).
This geographical problem is fundamental by nature and needs to be addressed at the theoretical level.
New academic theories, strategies, frameworks, models and toolsets are required to improve floatovoltaic
technology sustainability characterisation (Armstrong et al., 2020; Kumar et al., 2022; Prinsloo, 2020).
This thesis strategically argues that a newly envisaged theoretical framework solution should be grown
from the roots of agricultural sustainability thinking, as it will then be inclined to have a much more sig-
nificant impact in terms of capturing a broader spectrum of performances when compared to that of the
current state-of-art PV performance models (Prinsloo, 2020). Since agricultural sustainability describes a
multidisciplinary goal that impacts energy development, environmental protection, economic feasibility and
food security (Luca et al., 2015; Alkemade et al., 2012), the thesis, therefore, contemplates a more holis-
tic agri-friendly sustainability valuation strategy in a sustainability framework for floatovoltaic technologies
that better aligns with OECD’s agricultural production sustainability depicted in Figure 2.8. Such a newly
proposed framework would be able to address legislative requirements for the international adoption and
environmental licensing of floating solar renewable energy technology installations (especially in the agricul-
72
tural sector). In this sector, there exists a dire need for an environmentally-oriented performance evaluation
framework for the scientific sustainability characterisation of future planned floatovoltaic projects (Gadzanku
et al., 2021b; Hernandez et al., 2019). Because sustainability assessment is by definition required to go
beyond the conventional measures of climate change and environmental impact assessments (Hottenroth
et al., 2022; UN, 2022), envisaged sustainability performance profiling capabilities for PV need to go be-
yond the environmental risk paradigm and venture into the integration of environmental economics into the
sustainability equation (Prinsloo, 2020).
Furthermore, since floatovoltaic technologies in agricultural applications interface directly with water
as a sensitive and critical resource in food security worldwide, the lingering threat of new international
licensing regulations specific to future floatovoltaic technology installations creates a reason for improved
regulatory compliance considerations (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). Moreover,
while international regulators are interested in gauging the rural agricultural (floatovoltaic) energy system
sustainability portfolios in terms of the support it provides to natural WELF resources and preservation of
global climate change initiatives (United Nations, 2019), the framework should ideally be able to address
these issues based on the principles graphically depicted by Figure 2.8. At the same time, a proposed
framework should be able to project the role of technology in driving a sustainable future through the linking
of the SDGs shown in Figure 2.2 (UN, 2020a; World Bank, 2017). In considering all of these strategic
goals on an operational level, the conventional photovoltaic performance modelling framework paradigm
both fails to quantify the broad and diverse spectrum of floatovoltaic system performances while failing
to provide adequate answers to help solve pressing regulatory requirement challenges in an integrated
manner (Hernandez et al., 2020; Prinsloo, 2019, 2020). This failure creates the opportunity for the creation
of new knowledge through the composition of a new theoretical framework and tooling solution from an
agricultural perspective, wherein the layered techno-enviro-economic domain responses could address the
current real-world dilemmas with floatovoltaic impact assessments.
Observing the perceived theoretical framework problem from an EIA perspective, the lack thereof results
in regulatory project delays principally because the lack of EIA assessment protocols for FPV causes an
unfavourable rating for the technology on the technology readiness on the TRR (Regulatory Readiness) and
TMR (Market Readiness) scales portrayed in Figure 2.14. Therefore, while floatovoltaic technologies in SA
are in the demonstration phase and it technically rates quite favourably on the TRL rating scale shown in
Figure 2.14, the lack of a suitable sustainability assessment framework to drive analytical tools that measure
or predict the sustainability profile of the technology in SA are causing real-world technology assimilation
challenges. The goal is, therefore, to compose a new geographical framework and tooling solution from an
agricultural perspective wherein the layered techno-enviro-economic domain responses could address the
current real-world dilemmas with floatovoltaic impact assessments.
In the context of a multiscale regulatory sustainability definition and EIA criteria in Figure 2.7, two signifi-
cant limitations are causing analytical imperfections in most floating PV analysis models. They are depicted
in Figure 2.18 as the unidirectional linear succession of the process employed by most floating PV char-
acterisation models and the low priority given to environmental considerations as part of the floating PV
modelling regime. The first problem, which emanates from an open loop-type linear representation of a
unidirectional analytical process, as depicted in Figure 2.18, often fails to fully account for real-time floating
PV performance attributes caused by systemic causal interactions with the natural environmental ecosys-
tem. According to the idea of causation, the environmental modelling void is particularly devastating as it
largely seems to ignore the microscale and mesoscale climatic feedback effects, such as the hydroclimatic
variations inherently observed in floatovoltaic installations. Putting the environmental subsystem at the end
of the analytical cycle (refer to the environmental model in Figure 2.18), open-loop processing leaves sig-
nificant knowledge gaps around modelling the recursive feedback that seems to characterise floatovoltaic
performances. This effect happens because the real-time microscale environmental heat exchange feed-
73
back between the floating PV panels and the underlying aquatic habitat significantly impacts the real-time
performance of the floating PV system (Prinsloo, 2019, 2020; Prinsloo et al., 2021).
microscopic linear techno-economic perspective => monocausal linkages between energy and economic objects
Environment
Model
individual energy and economic domain object components linked linearly in series => not conventionally connected to an environmental impact model
Figure 2.18: Unidirectional analysis framework driving typical PV system assessment models (after: Baring-Gould
(2014)), failing to fully account for real-time FPV systemic causal interactions between energy, economic and environ-
mental operational sub-models (source: author).
While the open-loop type unidirectional processing framework depicted in the typical process repre-
sented in Figure 2.18 is used by most of the PV performance modelling tools of this era, such a modelling
framework compromises performance and impacts upon prediction accuracy in floating system assess-
ments. For example, a typical techno-economic project evaluation model such as SAM, in Figure 2.18,
calculates the hourly energy output of the PV system. Energy outputs are used afterwards to calculate the
energy cost for the PV energy project over the project’s lifetime. Being characterised by a low level of exe-
cution connectivity in the open-loop techno-economic sequence (often no real-time feedback between the
technical, economic and environmental ecosystem elements during the simulation run), the linear frame-
work may miss important performance and impact aspects required by carbon taxation and environmental
impact reporting. Sequential processing is a universal limitation in most state-of-the-art solutions described
in the literature review, open-loop analytical processing that does not integrate microclimate environmental
system effects. It also explains why standardised metrics for ground-mounted photovoltaic project models
such as Figure 2.18 do not adequately account for the extended range of resource-use efficiencies and
impact-effect positives offered by floating PV systems.
From an operations-theoretic research perspective, new methodological developments triggered the
goal of productivity assessment in manufacturing for effectiveness, in which organisations pursue appropri-
ate holistic goals for adoption in practice and to improve the usability of assessment reports. This inspired
research into developing new data-driven assessment methodologies to improve production productivity by
optimising the energy, environmental, and economic performances at the enterprise manufacturing level
(Abdullah and Nishimura, 2021; Cartelle et al., 2015). This multi-disciplinary perspective on cleaner pro-
duction inspired Kumar and Mani (2021, 2022) to theorise about the methodological divide and distinc-
tions between "efficiency vs effectiveness” in production assessment perspectives (reductionist vs systemic
philosophies), as presented as Figure 2.19.
Figure 2.19: Contextualisation of the concepts of efficiency vs effectiveness in assessments, presenting a reductionist
vs systemic perspective on assessment for manufacturing enterprises (source: Kumar and Mani (2022), page 4).
74
The argument of Figure 2.19 is that the modern (geographical) operations research should focus on
broadening production sustainability assessment perspective from process to enterprise-level, from a frag-
mented to a holistic point of view, from single to multiple cascaded parameter assessment, and from local
project business perspectives on global sustainability and circularity. It was asserted that the shift toward the
holistic production sustainability assessment towards the so-called "effective assessment practices” through
interdisciplinary frameworks for effective sustainability assessments.
Such assessments should be driven by factors such as the awareness of resource scarcity, the global
environmental burden, and modern-day challenges in human health. Consideration for these factors in man-
ufacturing sustainability (in the context of Figure 2.19) serves as motivating concerns for the inspiration of
new research in theory (academia) and practice (industry) to develop systemic-type holistic (effectiveness)
sustainability assessment methodologies that focus less on silo-type reductionist appraisal methods, and
more on the holistic-type systemic productivity appraisal methods (reductionist vs systemic). From a geo-
graphical operations research perspective, this thesis perceived these recommendations as having potential
for new geographical theory-building research in the context of energy production system installations.
While this thesis views a floating PV system through a 4IR lens, this research views the PV energy
system as a smart energy manufacturing enterprise wherein sustainability productivity should be evaluated
beyond the technical or techno-economic parameters as illustrated in Figure 2.18. The effective causes of
the imperfections in conventional analytical processing frameworks used in current PV performance models,
such as those deficiencies depicted in Figure 2.18, can be theoretically explained at the hand of the prin-
ciples portrayed in Figure 2.19. To overcome current modelling limitations, that currently hinder evidence-
based scientific assessments in regulatory project permissions mandated by law, the broader perspec-
tive on holistic production efficiency and the economy of production performance helps to expose serious
knowledge gaps around systemic perspectives on PV performance assessments. From an environmentally-
oriented EIA perspective, the layered hierarchical impact-effects hierarchy of Figure 2.7a serves as testi-
mony to this hypothesis, and at the same time, serves as strong motivation to base future sustainability
assessments in floating PV analysis on holistic or systemic principles (contextualised in the right-hand of
Figure 2.19). To help overcome the analytical limitations in EIA assessments of the broader spectrum
of impact effects of floating PV installations, as illustrated in Figure 2.7b, Prinsloo (2020) proposes an im-
provement in the EIA assessment process and modelling framework for FPV (Prinsloo, 2020). This proposal
inspired the research of this thesis and includes as a basis the extension of the model of Figure 2.17 to-
wards an improved architecture that would consist of both environmental and economic impact modelling as
part of floating PV systems modelling. Since the floating PV assessment model developed by the author in
Figure 2.17 uniquely makes provision for the modelling of the environmental aspects associated with float-
ing PV ecosystem analysis, Prinsloo (2020) further incorporated environmental economics and financial
viability measures into an analytical FPV sustainability assessment model depicted in Figure 2.20.
To investigate the opportunity of integrating environmental EIA, as well as financial sustainability and
environmental economics, into a due-diligence analysis of a technical floating PV system, Prinsloo (2020)
compiled his research proposal to also include fundamental improvements to the geomatic FPV model of
Figure 2.17 by incorporating financial due-diligence into the techno-environmental modelling analysis. This
integrated approach to the characterisation of a floatovoltaic system, through a digital twin virtual proto-
type model depicted in the multivariate data model of Figure 2.20, offers a conceptually-integrated techno-
enviro-economic (TEE) analytical system improvements based on the technical energy, environmental and
economic (3E) dimensions. Furthermore, as delineated in the diagram of Figure 2.20, this model provides
an extended floating PV system focused on the functional systemic role of the environmental system. Con-
tinuing the sequel depicted in Figure 2.17, the integrated Figure 2.20 puts the spotlight on a proposed
investigational-type platform for scientific experimentation, with a broader spectrum of metrics built into an
analysis framework and concept of a planning scenario model of a floating solar system.
75
EIA Offset & WELF Reporting Layer
SolarModel
Energy
Simulation
Model
Economic
Analysis
Model
Environmental
Simulation
Model
kWs(t)
fy(?)
fx(?)
fz(?)
kWs(t)
CO2(kg)
H2O (kg)
SOx(kg)
NOx(kg)
aPE (kg)
Solar data
Weather data
Climatic data
Cloud modulation
Capex costs (R)
LCoE electricity (R)
E-tax credit (R)
Revenue (R)
Figure 2.20: Block diagram of a more integrated design of an FPV systems model, showing the assembly and linear
process workflow for a multi-core system, depicting the Energy-, Environmental- and Economic- or EEE-subsystem
simulation models (after: Prinsloo (2020)).
However, from an operational research perspective, one problem remains: Would the floating PV system
model proposed in Figure 2.20 still operate if based on a unidirectional linear open-loop processing frame-
work? Because there is a missing conceptual link between the microscopic and most of the macroscopic
effects (refer to functions f(?) in Figure 2.20), a universal construction of the principle of shared interests
and systems sensitivity is required (Habermas, 1979). Furthermore, for the proposed model of Figure 2.20
to account for those cascaded feedback effects caused by systemic causal interactions between energy
and economic and even the environmental ecosystem subsystems in the floating PV ecosystem, the float-
ing PV performance-profiling model needs to be extended towards dynamical systems modelling to meet
the accuracy and flexibility requirements of EIA-integrated floating PV project assessments.
Therefore, from a systems research and behavioural science perspective, more of a systems-thinking
research effort is needed to condition the integrated techno-enviro-economic (3E) model for floatovoltaic
systems properties, specifically to include country-specific economic and environmental impact model pa-
rameters suitable to floatovoltaic energy systems within the South African agricultural landscape (Prinsloo,
2019, 2020; Prinsloo et al., 2021). Towards opportunities and scope for innovation in applying synergetic
systems thinking, this thesis offers novel research to emulate the geodynamical responses of floatovoltaics
technology in a computer simulation model through a multi-layer modelling framework for techno-economic-
environmental assessments (EEE). As such, the thesis advances a three-tiered floating PV business sus-
tainability model (3E model) as a basic modelling schema proposed by Figure 2.20.
Scope for new innovation
Note the value proposition around integrating the technical Energy-Environmental-Economic or E-E-E interdependency as
a dynamic system. Such dynamic interactions form part of the more extensive set of "technology unknowns” caused by
critical knowledge gaps in the field and identified as critical barriers to floatovoltaic project deployments and project ap-
provals. Such complex correlative associations play a crucial role in motivating the thesis’s conceptualisation of a broader
inter-dependent project assessment framework, namely the holistic sustainability E-E-E framework, as a means to capture
most impact interactions within all three types of the "E”nergy (or "T”echnical), "E”nvironmental and "E”conomic elements
within the floatovoltaic ecosystem. In terms of the postulates of this thesis, a holistic EEE or TEE or 3E or E3-modelling
framework can establish the analytical mainstay and modelling hallmark for broad-spectrum floatovoltaic ecosystem mod-
elling since causal system dynamics are the drivers of accurate predictions of FPV project impact assessments. The
thesis defines the term 3E to define the holistic sustainability framework in terms of the energy-environmental-economic
dimensions. It is hypothesised that the theoretical 3E framework may help to uncover and identify many more of the
systems-based interdependent linkages of the real-world FPVs towards quantifying many of the "technology impact un-
knowns” in a fully-integrated analysis-by-synthesis model for the assessment of the sustainability of floating PV technology.
76
In terms of the novelty of the research of this thesis, most state-of-the-art models use the underpinning
open-loop linear framework of Figure 2.18 to perform floating PV assessments with an additional panel
temperature model. However, engaging a silo-based approach to solving the analytical problem by linearly
or serially linking technical energy performance modelling with economic- and environmental- performance
modelling (E+2E) may miss critical causal feedback linkages in the holistic floating PV operational ecosys-
tem (EIA, climate economy, etc). Although the approach of linearly adding more components to the analyt-
ical system may provide some improvement, unidirectional linear processing still does not account for the
cascaded hierarchy of causal interactions observed in real-world floating PV ecosystems (Prinsloo, 2020).
Here, the literature review can highlight the triple-element linear analysis approach to biorefinery-specific
applications proposed by Lindorfer et al. (2019). This approach uses an Excel spreadsheet to assess envi-
ronmental and financial impacts in a biotechnological MIS environment. However, like most other open-loop
assessment frameworks, this Excel-based stakeholder database approach evaluates existing biorefinery
plants by considering each TEE element silo independently in a linear sequence. Sahoo et al. (2018) per-
formed an energy, exergy, economic analysis (4E) analysis, but this is to optimise a biomass system in
a hybrid solar polygeneration scheme. Another open-loop type linear TEE or 3E framework, termed the
Building for Environmental and Economic Sustainability (BEES) framework, was developed by the National
Institute of Standards and Technology (NIST). It integrates environmental and economic performance met-
rics to support investment choices in a holistic-building energy system assessment database called Building
Industry Reporting and Design for Sustainability (BIRDS) (Kneifel et al., 2017). Similar to the Lindorfer et al.
(2019) model, the Kneifel et al. (2017) model focuses on lifetime performance and costing assessments
rather than on real-time modelling as proposed in this thesis.
Regarding technology-specific floating PV performance models, the Dizier (2018) model is limited to
a conventional open-loop two-element techno-economic (TE) model that operates in a linear-processing
fashion. This model has a different application purpose, as it applies the TE methodology to evaluate active
cooling interventions, which is significantly different from the holistic systems-based cybernetics approach
proposed in this thesis. The OECD also proposed a linear informatics assessment approach (Alkemade
et al., 2012). Still, the analysis is constrained by the fact that TEE processing again takes place linearly
in an open-loop fashion, similar to the model of Omrani and Omrani (2022). Bui (2019); Makhija et al.
(2021); Testori (2021) used conventional PV tools in linear, unidirectional modelling approaches to assess
the Techno-Enviro-Economical (TEE) aspects of floating photovoltaics systems in hydropower, city and rural
electrification applications in Vietnam, India and the Netherlands respectively. These studies follow current
best practice techniques discussed in Section 2.5, meaning they fall short of meeting feedback-type environ-
mental and sustainability impact assessment requirements expected in evidence-based regulatory project
licensing and approvals. Toba and Seck (2016), on the other hand, looked at enhancing energy policy deci-
sions through a generic electricity system modeller, linearly processing the technical, social, economic and
environmental elements (S-TEE) in a supporting model for country-specific energy-system policy valuation.
This model represents a simplified macro-energy utility system model to acquire, generate and distribute
electricity in a country context, and specifies a set of sequential relationships between technical, social, en-
vironmental, and economic aspects of an energy system as a means to account for policy definition effects
and implications. Despite the dissimilarity in applying the 3E-element analysis in the existing literature, the
inclusion of the above research findings in this thesis review highlights the practical benefits of potentially
integrating environmental impact assessment with energy and economic system performance analyses.
With the review highlighting the novelty of the research of this thesis, the review puts the spotlight
on new geoscientific model development based on the identified bi-directional feedback system solution
towards delivering an integrated framework based on technical energy, environmental and economic geog-
raphy. It further offers new opportunities for geomatics processing and system dynamics modelling towards
floatovoltaic performance and impact effects profiling based on a newly proposed theoretical framework
77
and geographic information processing technique. In this regard, the thesis’s extensive review of scientific
literature could not at this stage find any meaningful competing research attempts towards (a) the feedback-
integrated 3E-modelling of floatovoltaic analytics and decision support; (b) modelling systems suited to the
South African context; (c) system dynamics thinking or cybernetics-type systems-based analytical simula-
tion models for floatovoltaic technology; (d) nor any dynamic systems-integrated energy-enviro-economic
real-time computer synthesis models and data-processing approaches to compete with the unidirectional
systemic integration framework for floating PV characterisation anticipated by this thesis.
Therefore, as far as the literature study could determine, the proposed holistic networked information
systems approach to the geomatical modelling of the combined interactions between the floatovoltaic tech-
nology ecosystem domains (refer to Figure 3.14 in the next chapter) seems to be novel. In addition to the
planning-operations framework for the floatovoltaic ecosystem, the fact that the processing methodology for
floatovoltaic technology ecosystems should be based on system dynamics modelling in a hybrid system’s
network adds a dimension of novelty. Furthermore, filling in the research gaps around floatovoltaic project
appraisals through the systems-integrated scientific framework can offer a unique opportunity for dynamic
cause-effect analyses. Such consistent cascading interactions in a 3E trifecta domain have not yet been
compositely model-linked in real-time simulative broad-spectrum planning analyses or decision-support ap-
plications for floatovoltaic technology development projects, thus adding a new dimension to the novelty of
the research.
Supporting the perforation of floating photovoltaics in South Africa, technology adoption calls for the
re-contextualisation of floating solar decision-planning analysis towards an integrated techno-economic-
environmental assessment of the floating PV system. The thesis aims to combine the operational energy,
environmental, and economic (3E) processing elements in a new systems-thinking paradigm and analytical
reference framework. The aim is to formulate a newly proposed technology-specific analytical framework
(refer to Figures 3.13 and 3.14 in the next chapter) to operate as a triple-helix conceptualisation in a logical
computerised architectural diagram (refer to Figure 3.16 in the next chapter). With such an integrated
framework, program logic and computer architecture, the thesis foresees that it can fill critical knowledge
gaps through a systems-integrated holistic geographical modelling innovation for floating PV technology.
That, in a nutshell, describes the aim and objectives of the research in the present thesis as a contribution
to the sustainable development discourse in South Africa.
With the current challenges and the underpinning of the best practice methods detailed in terms of
the present state-of-the-art floatovoltaic assessments, the next chapter describes the proposed research
methodology and theoretical framework to help overcome these real-world analytical challenges.
2.7. Summary
This chapter offers an extensive literature review on the study topic and theoretical framework of the study
within the scope of the research and the study rationale around the topological field of floatovoltaic tech-
nology modelling. It engages the science and practice of knowledge management in an organised train
of thought to expand the scope for innovative opportunities for filling knowledge gaps around floating pho-
tovoltaic technology analytics and decision support. The review underscores the need to conceptualise a
new contextually-sensitive sustainability assessment and modelling framework for floating solar technology.
This conclusion includes the need to apply the proposed framework in developing new geoinformatics tools
and GIS models to assist farmers and environmentalists in studying the long-term performance profiles,
environmental impacts and economic feasibility of floatovoltaic energy systems.
By breaking the problem down into a three-tier hierarchy, considering statutory-level requirements,
process-level goals, and activity-level challenges, the review kicks off with a statutory-level analysis to
78
introduce the concept of sustainable floatovoltaic project development from a global perspective, before
addressing the regulatory frameworks toward sustainable development in South Africa from a local per-
spective. The process-level analysis then details the phenomenal growth rates in applying floating solar
technology around the globe and highlights the exceptional potential for the development and diffusion of
floatovoltaic technologies in the South African region. This analysis uncovers a range of sustainability fea-
tures of floating solar systems referenced in the literature and highlights the severity of the analytical and
modelling knowledge gaps around the technology’s impact effects. In this context, the activity-level analysis
overviews the underpinning best-practice approaches by reviewing existing analytical technology assess-
ment tools developed by scientists to measure or predict the impact performance of floatovoltaic systems in
sustainability metrics and econometrics.
Bringing the literature study to a conclusion, the review narrates the scope for innovation towards a pro-
posed solution for resolving strategic challenges around floating solar project design evaluation. It proposes
a conceptual solution for an opportunity-driven simulation model framework supporting the local develop-
ment and diffusion of floating solar technology in the South African landscape. Towards implementation, the
following chapter details the overall research methodology and modelling process that contribute to solving
those mentioned real-world geographical challenges.
79
3. Philosophical Modelling Methodology
3.1. Introduction
This chapter deals with the research design process and the methodological framing toward the realisation
of the aim through Research Objective 1. The discussion focuses on philosophical theory-building and
provides an intelligent narrative around the goals of this research objective. It gives theoretical substanti-
ation to the systems thinking approach and methodological processes behind establishing a sustainability
reference framework for assessing the sustainability qualities of future planned floating PV installations.
The proposed research methodology lays the theoretical foundations to facilitate the creation of new knowl-
edge within the context of the floating solar renewable energy systems discourse. While the literature study
allowed for the interpretation of the requirements for new design stage geographical tool development for
floatovoltaic project assessments, this chapter undertakes novel research in terms of systemically analytical
considerations and system dynamics design principles for the simulation-based evaluation of floating solar
energy systems. It deals with developing a proposed theoretical framework and model toward a method of
execution whereby quantitative data can be collected and analysed.
Towards helping to answer the first research question, the theoretical assessment framework develop-
ment efforts of this chapter aim at the implementation of the research Objective 1, namely how to define an
integrated systems-thinking project analysis and sustainability-oriented decision framework concept to re-
alise a geo-sensitive approach capable of theoretically exploring the combinational effects of the technical,
economic and environmental narratives in floating PV development plans for the South African agricultural
context. As such, the flowchart in Appendix D details the research framework and model development
steps, whereby this chapter navigates the reader through the philosophical and methodological approaches
towards establishing a quantitative research methodology. Section 3.2 presents a brief overview of the
philosophical systems thinking approach behind the methodological approach. Section 3.3 motivates the
application of computer modelling as a practical research methodology and guides the reader through the
steps of translating a conceptual systems model into an experimental computer model. The methodological
framework in Section 3.4 presents the options for selecting and defining an appropriate research framework
and methodology to conduct the research investigation in a step-by-step procedure. The process of devel-
oping a theoretical assessment framework in Section 3.5 follows this methodological research procedure to
thereby formulate a tentative hypothesis as a conceptual, analytical framework model to characterise the
theoretical performance of floatovoltaic ecosystems and profile them in sustainable terms. Section 3.6 de-
tails the logical and physical model design processes in a system dynamics thinking methodology towards
the realisation of a computer synthesis model and geographic toolset for floating PV system assessments.
Before Section 3.8 concludes the chapter, Section 3.7 provides the theoretical substantiation for the re-
search investigation to motivate and detail the theoretical underpinnings of the proposed systems-based
analytical framework of a modelling approach and computer synthesis solution toward the scientific realisa-
tion of the research aim and objectives.
80
3.2. Philosophical Approach
This section reveals the deeper conceptual and theoretical foundations for defining, realising and validat-
ing the hypothetical research purpose in a scientifically-founded and experimentally modelled valuation
framework paradigm for answering the research questions. It advances the research process towards
applying systems thinking in exploring underlying systems principles to improve the sustainability mod-
elling of floating PV systems. It navigates the reader through the philosophical approach as it presents
a narrative overview of the underlying theoretical concepts in establishing a digital floatovoltaic analytical
technique and a computer synthesis model as foundational parts of the proposed geoinformatics analyti-
cal decision-support system. Section 3.2.1 guides the reader through the steps of the research paradigm
and the paradigmatic assumptions around systems thinking in theory-driven development. Section 3.2.2
frames the philosophical research paradigm in terms of the research ontology, epistemology, theoretical
framework, methodology, and modelling approach options within the context of floating PV (eco)systems.
Finally, Section 3.2.3 presents a unique philosophical perspective to develop a systemically integrated sus-
tainability assessment solution to encompass the analytical assessment of floatovoltaic system operations,
performances and impact effects in one sustainability criteria.
3.2.1. Philosophical paradigm and approach
This section delineates the philosophical paradigm and methodological approach toward establishing a
quantitative research methodology to evaluate the research aim and objectives and to answer the research
questions. It tables the key philosophical assumptions and paradigmatic features of the systems thinking
research paradigm in the context of the positivist and systems thinking research philosophies.
In terms of systems research and behavioural science, the character of analytical research philosophy
has a distinctive role in the philosophical ecosystem and workflow simulation chronology of Figure 3.1. In
terms of the research requirement defined by the study purpose (Section 1.2.3), the research paradigm
focuses on the virtual issue of analysis and modelling of floatovoltaic sustainability, with specific reference
to the development of a new theoretical framing and systemic synthesis tools in an enhanced academic
philosophy that can address this real-world topical issue.
Figure 3.1: Progressing through the research paradigm steps towards the research methodology and method of data
processing, data collection and data analysis (source: Patel (2015), page 2).
In terms of the philosophical research process depicted by Figure 3.1 the philosophical research paradigm
serves as the overarching system of inquiry as it defines how to acquire knowledge about the reality of
interest, in this case, floatovoltaic system sustainability. Regarding organisation theory, the paradigm com-
prises branches of philosophical studies: namely ontology, epistemology and methodology. The ontological
and epistemological dimensions influence the research methodology by clearly understanding the chosen
methodological approach and design.
81
The research methodology drives the research paradigm, which offers a method, model and pattern for
conducting research and describing how to systematically solve the research problem from a geographical
resource engagement perspective. The methodological research paradigm generally includes the various
logical steps adopted to study the research problem. While the thesis has already dealt with the step of
investigating the ontology (What is reality?), this chapter deals with the steps that are taken to formulate
the epistemology (how to know reality), the theoretical framework (how to study/model reality), the chosen
research methodology (How to uncover reality?) and the research method (data collection method and
necessary tools or techniques to use to acquire knowledge about reality?) (Patel, 2015).
In this context, the graphic depiction of the chronological scientific research process in Figure 3.1 visu-
alises the chronological progress through scientific engagement and the research paradigm towards choos-
ing the appropriate research methodology and methods for the data processing, data collection and data
analysis processes. While the scientific philosophy links the research ontology, epistemology and research
methodology through the chronological research sequence of Figure 3.1, the philosophical paradigm sup-
ports the research design framework of Figure 1.6 in that it translates the strategic goal of the research, or
the research foundation (problem, purpose, theoretical underpinning, research questions), into the selec-
tion of the research method and methodology (data collection, data sampling and analysis) (Mingers, 2014;
McCartan and Robson, 2016).
The research terminology and the procedural steps of the workflow simulation chronology of Figure 3.1
bring the philosophical investigation to the definition of the philosophical assumptions, attributes and bound-
aries of the paradigm in Table 3.1. With the philosophical argumentation and assumptions presented in
Table 3.1, the study can implement the research aim and objectives according to a recognised data-science
methodology and modelling method.
Table 3.1: Philosophical paradigm and assumptions towards investigating FPV systems, as defined by this thesis
(source: author).
Paradigm Ontology Epistemology Theoretical per-
spective
Methodology Method
Positivism Single reality,
objective and
singular, inde-
pendent, apart
from researcher
Reality can be under-
stood or constructed,
replicable findings, re-
ality can be explained
relatively accurately
Positivism,
Post-positivism
Quantitative
research, ex-
perimental
research
Empirical research,
computer modelling,
statistical analysis,
quasi-measurement
and scaling
Systems
Thinking
Combining
concepts, rela-
tionships and
governance
rules to admin-
ister operations
and linkages
Systems epistemol-
ogy, seeking to derive
knowledge from a
strategic vantage
point, making sense
of causality from
various perspectives
Object-oriented
principles,
Dynamic sys-
tems think-
ing, Deduc-
tive/Inductive
reasoning
Quantitative,
empirical, real-
ity simulated,
quantitatively
measured,
scientific fore-
casting
Parametric mod-
elling, data sci-
ence models,
computer synthe-
sis, quantitative
sampling, analysis-
by-synthesis
With the sequence and boundaries of philosophical argumentation defined in Table 3.1, the systems
thinking research paradigm followed in this thesis offers an opportunity for a broad-based methodology
for systemic intervention through engagement in a design-based research approach aimed at value chain
analysis. Following the introduction of the philosophical approach in terms of the research perspectives
of the methodology, the following section frames the research ontology, epistemology and methodology in
terms of the combined positivist and systems thinking philosophical research approach.
82
3.2.2. Framing of research ontology, epistemology, and methodology
This section contextualises the research design process of Figure 1.6 by framing the philosophical inves-
tigation into floating PV systems in terms of research ontology, epistemology and methodology toward the
combined positivist and systems thinking philosophical research paradigm defined in the previous section.
From an ontological perspective (What is reality? What is the nature of reality?), the reality of interest
can be defined as a particular aspect of the world (a system, subsystem or component problem) to be mea-
sured and simulated (Thacker et al., 2004). From this perspective, the study aims to investigate the reality
of sustainability in future planned floating PV systems. The goal is to quantify and understand the theo-
retical pre-installation performances and impact effects of floating PV systems in agricultural applications.
This focus sets aim at predicting the broad spectrum of sustainability attributes or geodynamical responses
inherent to floatovoltaic technologies. While a multidisciplinary goal defines the criteria for agricultural sus-
tainability, its impacts range from energy development, environmental protection, economic feasibility and
food security. In this context, new sustainability definitions and governance frameworks for floating PV (and
land-based PV) must be developed and implemented around water preservation, resource conservation
and soil/vegetation protection. Towards ensuring a more favourable Technology Regulatory Readiness rat-
ing (refer Figure 2.14) for agricultural floatovoltaics, the sustainability status of agricultural energy projects
should be framed holistically as a theory to provide policymakers with effective alternatives toward defining
a more resilient future sustainable world.
From an epistemological perspective (How to know reality? How to find reality?) and the theoretical
framework perspective (How to study/model reality?), the aim is set to design an information system to
support sustainable project planning around the reality of floating PV system sustainability. With specific
reference to theoretical sustainability performance profiling, as illustrated by the reductionist vs systemic
philosophies presented in Figure 2.19, the literature review narrates the scientific foundations of the vari-
ous theoretical and simulation methods that are available for study and for knowing the real-world reality
around floating PV sustainability through computer model analysis. However, in the context of EIA-inspired
floating PV assessments, the literature review in Chapter 2 shows that there are fundamental analytical chal-
lenges with the existing frameworks and methods available to theoretically study and model the reality of
floating PV sustainability attributes. Therefore, on the philosophical level, floating PV characterisation in ge-
ographic information systems calls for a fresh system thinking research perspective towards improvements
in the epistemological and theoretical frameworks to fill the knowledge gaps toward improved sustain-
ability assessments in floating PV system profiling through the modelling and characterisation of theoretical
systems. In the context of the 4IR paradigm, the Systems Thinking 4.0 paradigm advances research in
universal systems thinking and system dynamics theory with cognition-informed solutions toward solving
technical, managerial, environmental and societal problems (System Dynamics Society, 2021). With this
value chain analysis goal in mind, Figure 3.2, the diagram representing knowledge, summarises the basic
concepts, principles and patterns of the systems thinking paradigm, thus serving as a means to guide the
systems thinking framework behind floating PV systems modelling and synthesis improvements.
While Figure 3.2 visually summarises how scientists should present the knowledge about the system, it
also serves as a guide to the critical components of the systems thinking paradigm in terms of recognising
the elements, relationships and perspectives of the systemic resolution. For the development, in terms of
science, of a systems thinking model, and to scientifically advance the characterisation of energy systems
and to assist environmental impact practitioners to participate knowledgeably in science-related impact
assessments and policy decisions, the commonly used reductionist approaches in sustainability conceptu-
alisation and assessments must be complemented with a more holistic approach, based on the systems
thinking paradigm. In this context, the systems thinking hierarchical modelling pyramid to implement Fig-
ure 3.2 is depicted in Appendix G.
83
Figure 3.2: Abstract representation of the fundamental concepts of systems thinking within the systems science
paradigm of interpreting real-world evidence to identify systemic principles and patterns during data analysis (after:
Lawson and Martin (2008)).
It characterises the breakdown of systems thinking model development actions in the respective phases
of systems development in terms of the analysis of the systemic components, the mathematical synthesis
of them, and the implementation of new systems models (Orgill et al., 2019).
From a research methodology perspective the study aims to develop a geographic information sys-
tem to realise a synthetic reality-oriented assessment of the prospective sustainability of floating solar-PV
deployment. The proposed solution aims at an interdisciplinary theoretical conceptualisation, whereby a
systemic sustainability resolve can address many environmental concerns around integrated sustainabil-
ity assessments of floating PV systems through developing a geographical information system (Strahler,
1980). The systems thinking philosophy supports design thinking as it offers a line of thought in the op-
erations assessment field, which stresses the interactive nature and interdependence of both the external
and internal factors in the ecosystem. Systems thinking can support the creation of causal loop diagrams
in the life cycle and instantaneous sustainability assessment strategies. Such diagrams help identify lever-
age points where whole-system interventions in a complex systemic environment will be more efficient and
effective (Sandia, 2016). Diagramming causal loop analysis can support operational PV and renewable
system design choices (Gonzalez et al., 2016; Halog et al., 2017; Sandia, 2016).
Since a proposed theory and sustainability framework ultimately serves as a blueprint for developing
an integrated computer program logic and dynamic software architecture for floating PV ecosystem char-
acterisation, the following section defines the philosophical goals for establishing a systemic sustainability
modelling concept.
3.2.3. Philosophical goals for a systemic theoretical sustainability modelling concept
Thanks to the arrival of floating PV technology, scientists worldwide are invited to re-image the concepts
of performance analysis, impact assessments and sustainability profiling, given the technology’s broad
spectrum of attributions.
The notion of sustainability in the context of floatovoltaic technologies is a reality that is currently a
poorly understood philosophical concept. The prevailing academic question then relates to the research’s
requirement; What should the philosophical approach and goals be for a systemically-based theoretical
sustainability modelling concept for agricultural floatovoltaics? The idea of this question is to engage the
thinking of Laszlo (1972), who describes systems thinking synthesis as a methodology that employs ecosys-
84
tem perspectives to model the nature of reality (FPV sustainability) and to use this systemic approach to
help solve critical real-world problems by predicting future behaviour (FPV sustainability synthesis). While
criteria and processes define sustainability assessment, Bond et al. (2012) defines the assessment of sus-
tainability as "any process or criteria that direct project decision-making towards resource sustainability”.
This sustainability view calls for modern sustainability indicators that are more integrated, dynamic and par-
ticipatory (Carmichael et al., 2017). To this end, the literature review highlights that floating PV behaviour
analysis, performance assessments and sustainability valuations are three individually different modelling
concepts with varying objectives and scales. While sustainability criteria and valuations are at the pinnacle
of modelling, all three modelling concepts in a floatovoltaic technology context are poorly understood. How-
ever, from a philosophical sustainable farming energy modelling point of view, an opportunity is seen in the
fact that they cognitively and functionally appear hierarchically interlinked in the modelling conceptions of
floatovoltaic sustainability.
Sustainability assessment is a complex systemic concept compared to performance analysis and impact
assessment (Al-Subhi et al., 2020), and it requires an integrated multidimensional assessment regime for
sustainability (Buchmayr et al., 2021; Naegler et al., 2021). Sustainability criteria are thus needed to go
beyond the traditional techno-economic and climate change measures, urging scientists to think beyond
the power output and hydrocarbon emission risk paradigm in energy systems development (Hottenroth
et al., 2022; Naegler et al., 2021). At the same time, it requires an architectural concept to automate
the computational scientific workflows in sustainability assessments (Liu et al., 2019; Morrison-Saunders
and Pope, 2013) through computer aided process engineering methods. Towards more intelligent process
automation for sustainability assessments, the thesis posits that the real-world geographical problem. It
posits that the characterisation of FPV sustainability, is caused by a three-pronged analytical fault line in that
(a) conventional PV simulation and PV performance assessment models are unable to cope with anomalous
behaviour of floatovoltaic-type systems; (b) conventional PV simulation/performance models do not include
sufficient environmental understanding to ensure integrated (environmental) impact assessments, and; (c)
classical sustainability policy for energy project assessment (based on people-planet-profit criteria) does
not cover all the bases when assessing the complex spectrum of sustainability attributions delivered by
floatovoltaic energy systems.
While scientists agree that conventional PV behaviour and performance assessment models cannot
cope with the distinctly diversified and complex behaviour of floatovoltaics, what further complicates the
appraisal challenge for this thesis is the fact that sustainability assessment needs to go beyond the environ-
mental risk measures of the climate change paradigm (refer to (Al-Subhi et al., 2020; Hottenroth et al., 2022;
Naegler et al., 2021)). Moreover, in the case of appraising agricultural floatovoltaics, one must take cogni-
sance of the requirement for alignment with agrarian sustainability goals that govern the conditions for new
energy installations on farms as part of the green energy production movement (Cagle et al., 2020; WWF,
2021). These prerequisites instil a binding obligation on new theory creation. A new theoretical construct
must overcome an inherent problem of conventional PV project assessments by conceptualising a sus-
tainability criterion framing that matches sustainable agricultural production goals. The thesis is, therefore,
faced with cultivating new ideas towards defining a fit-for-purpose theoretically-based sustainability framing
for agricultural floating PV technology. From systems thinking perspective, the literature review forms the
opinion that critical geographical systems thinking and behavioural science techniques may offer exciting
analytical prospects in the dynamical behaviour of floatovoltaic technology and help find a new theoretical
underpinning for multi-domain intellectual analysis modelling and simulation concept capable of emulating
real-world agricultural floatovoltaic enterprise ecosystems in a geospatial digital twinning solution.
This geographical investigation is, as such, strategically confronted with the academic question of philo-
sophically resolving a real-world problem through a theoretical-level systemic modelling solution. A pow-
erful philosophy for developing a more holistic FPV modelling conceptualisation would be to advance the
85
academic-theoretic proposition to establish a new theory and framework to concurrently synthesise FPV
operations, performances, impact effects and sustainability in the same model. At the same time, such a
theory and framework for assessing agricultural-type (floating) PV energy systems must ensure coherence
with agricultural production-type sustainability assessment frameworks (in a WELF nexus environment). To-
wards a newly proposed functional sustainability criterion for such a comprehensively integrated model, the
thesis identifies theory-building prospects in creating a new integrated theoretical policy that can uniquely
infuse sustainability principles into the theoretical characterisation of the real-time simulation of the oper-
ational behaviour of an FPV system. Hybrid sustainability criteria can thus underpin a PV sustainability
modelling as a framework for a predictive multi-hierarchy dynamical systems algorithm. On the philosophi-
cal level, the conceptual illustration in Figure 3.3 depicts the conceptualisation of the integration of behaviour
synthesis, performance evaluation, and impact assessment all in a single consolidated sustainability mod-
elling synthesis as integrated theoretical modelling abstraction posed by this thesis.
Operational
behaviour
Performance
features
Impact
effects
Sustainability
attributions
metrics metrics
metrics
Operations
modelling
Performance
modelling
Impact
modelling
Sustainability
modelling
responses responses
responses
(a) (b)
Figure 3.3: Prospective opportunity for the formulation of a new hybrid theoretical sustainability formulation that (a)
integrates operational behaviour, performance assessment, and sustainability attributions, in hybrid sustainability cri-
teria towards a hybrid modelling framework that simultaneously emulates (b) the operational behaviour, performance
responses and sustainability signals in a single integrated algorithm (source: author).
The philosophical idea depicted by Figure 3.3 is powerful since scientists have not yet completely solved
the problem of synthesising the operational behaviour of floatovoltaics and are still struggling with concep-
tualisations around the impact assessment of floatovoltaic technologies. The philosophical argument boils
down to the proposal that an integrated design for a real-time discrete event simulation model can ensure
an improved sustainability assessment through a computer synthesis, which can ultimately help to improve
the quality of floating PV planning support. Furthermore, such a consolidated sustainability synthesis fram-
ing, conceptually portrayed in the abstraction of Figure 3.3, may inherently advance the state-of-the-art as
it would be able to cope with the complex cascading nature of the "technology unknowns” of floating PV
performance modelling and the correlative conjunctions observed in field-installed floatovoltaic systems.
From a methodological perspective, Figure 3.4 enables this study to recognise the hypothetical chance
to conceptualise a hybrid sustainability policy theory that can enhance the concurrent assessment of the
behaviour, performance and sustainability of floatovoltaic installations in a single integrated algorithm. With
the systems thinking paradigm underpinning the concept depicted by Figure 3.3, the thesis poses the ar-
gument that a strategic focus on solving the sustainability modelling problem could inherently encompass
solutions for operational behaviour synthesis, performance analysis, impact assessment and sustainability
86
synthesis. Towards developing an evaluation methodology and introducing the relevant criteria in informa-
tion systems development, Richardson and Pugh (1981) describes the processing steps towards defining a
system dynamics modelling process in terms of Figure 3.4.
Systems Thinking
System
conceptualisation
Identify Problem
Ask Questions
Conceptual framework
hypothesis postulate
Logical model
architecture formulation Physical simulation
model formulation
Simulation model
behaviour analysis
Model application
/experimentation
Figure 3.4: Research methodology process for developing a novel system dynamics thinking concept into an experi-
mental computer synthesis model (after: Richardson and Pugh (1981)).
From a methodological perspective, Figure 3.4 depicts how any newly conceived systems theory and
framework is eventually required to serve as a blueprint logic in a computer program (Richardson and
Pugh, 1981). The theoretical framework thus serves as the basis for developing an integrated computer
programme logic and dynamic software architecture for floating PV ecosystem characterisation in terms of a
new system dynamics computer modelling and simulation methodology for floating PV systems. In the real-
world context of a predictive floating PV characterisation methodology, reframing a quantitative sustainability
assessment framework is crucial for remodelling floating PV characterisation through a system dynamics
modelling process shown in Figure 3.4.
From information systems research method perspective (tools or techniques chosen to acquire knowl-
edge), the aim is to consider experimental options in computer synthesis and scientific forecasting models
and to quantitatively measure or simulate numeric variables in a multiscale sampling research method. In
this context, experimentation with computer synthesis models can engage controlled designs, select simu-
lation timeframes, and consider sampling rates for data collection and data analysis in geomatics solutions
(Besenicara et al., 2008). From systems thinking and information systems modelling viewpoints, the de-
velopment diagrams depicting the designs in Figures 3.2 and 3.4 can serve as cognitive and procedural
guides for the reframing and remodelling of quantitative sustainability assessment modelling frameworks in
computer models for the assessment of floating PV system installations. The particular theoretical frame-
work can then serve as computer programme logic for studying the reality of sustainability in floating PV
ecosystem operations through a newly conceived computer modelling and simulation methodology.
Against the backdrop of this thesis’s proposed philosophical systems thinking approach in terms of the
ontological, epistemological and methodological research perspectives, this section posed the philosoph-
ical idea of combining behaviour synthesis, performance evaluation, and impact assessment in a consoli-
dated theoretical framework for agricultural sustainability modelling synthesis for floatovoltaics. The follow-
ing section guides the research approach’s direction in deciding on the appropriate methodological research
framework to drive a design-based research approach to achieve floatovoltaic sustainability through theory-
building research.
3.3. Methodological Approach
This section defines the research methodology to define a new quantitative theoretical computer modelling
technique dedicated to floating solar technology. It informs the methodological approach and process to
model the methodological framework proposed by this thesis into a quantitative computer simulation model
template. It supports the philosophical approach whereby an innovative systems modelling solution requires
87
a new theoretical framework to support system dynamics thinking according to the methodology frameworks
of Figures 3.2 and 3.4. In this context, the discussion deals with framing the research methodology and
defining it in terms of the step-by-step procedures toward conceptualising and realising an improved theory-
driven method for floating PV sustainability assessments. Section 3.3.1 motivates the computer simulation
as a methodological approach towards computer synthesis modelling as a conceptual modelling approach
to implement a theory-driven assessment framework for floating PV assessments. Section 3.3.2 delineates
the methodological sequence to move the research from a conceptual model to a computer model capable
of serving as a system of inquiry into the reality around floating PV sustainability.
3.3.1. Computer synthesis modelling as a methodological approach
While state-of-the-art analytical photovoltaic project assessments have been established firmly in the quan-
titative research paradigm, this section motivates the rationale behind computer synthesis as an overarching
research methodology by applying computer modelling and simulation methodologies.
In terms of the justification of computer modelling and simulation as a recognised research methodol-
ogy, the NSF categorises the use of digital simulation methods as the third methodology pillar in scientific
research (NSF, 2006; Vos and Shults, 2015). As such, the development of computer modelling and simu-
lation is appraised as a reliable scientific research instrument that stands on an equal footing alongside the
more traditionally recognised theoretical and experimental research methodologies (Canham et al., 2003;
Vos and Shults, 2015). Computer synthesis, as such, stands on its own legs as a quantitative research
method to represent a formal objective research approach for which the study must collect quantitative data
for information processing (Burns and Grove, 2005; Donatelli, 2008). The computer simulation method of-
fers a valuable modern-day research tool that extends the scientific research method by employing a digital
computer model representation of a real-time system (Estefan, 2008; Tina et al., 2017). As a recognised
quantitative data collection technique, computer synthesis offers a beneficial research instrument, present-
ing functional technology options in environmental assessment research of the diagnostic modelling type
(Frenkiel and Goodall, 1978). Computer simulation is, as such, recognised as a valuable scientific method
for collecting and processing environmental research data as part of investigations into the technology
around climate change (Fischlin, 1990).
In the context of this thesis, the main aim of the methodological approach is to establish a model or plat-
form to simulate the operation of a future planned floating PV panel system by conducting a pre-installation
analysis of its modelling and theoretical performance (Prinsloo, 2020). In this context, the thesis defines
a digital simulation model to mimic the virtual operation of any integrated floating PV system and incorpo-
rate the energy, environmental and economic components through discrete time-interval iterations to study
the productive capacity, financial implications, operational aspects, and environmental impacts of any pre-
defined floating PV micro-utility power-plant. Numerically-based experimental studies use computer mod-
elling and simulation methodology to examine the relationships among the various systemic components
and variables. This approach can assist in statistically analysing and determining the cause-and-effect
interactions between one or more transdisciplinary linkages or interdisciplinary variables (Kothari, 2004).
Furthermore, a systemic analysis of causal connections in an ecosystem context can help deal with predic-
tive problems in the theoretical assessment of project performances in the pre-installation phase (Khaiter
and Erechtchoukova, 2007).
As in the case of the real-time analysis of floating solar PV power plant projects, and almost every other
area of ecosystem science, computer models can provide a logical structure of the underlying system as a
means to guide and inform empirical observations of ecosystem processes as part of investigations around
the impacts of climate change technology (Fischlin, 1990). Most valuable as a scientific method for col-
lecting environmental research data are the tridimensional roles of digital computer models, as highlighted
88
by Canham et al. (2003) in the distinct ecosystem science domain roles in Figure 3.5a. In the situational
analysis of floatovoltaic technology, Figure 3.5a puts the spotlight on the model simulation roles of obser-
vation and experimentation, synthesis and integration, as well as prediction and forecasting. In the floating
solar planning activities of the present study, all three roles of mathematical modelling are engaged in a
model-driven simulation methodology.
(a) (b)
Figure 3.5: Graphical representations of the computer modelling and simulation research methodology, showing (a)
the three roles of models in a model-driven simulation methodology (source: Canham et al. (2003), page 2), and (b)
the process development phases towards experimentation with a computer model representation (source: Magliocca
et al. (2015), page 3).
With the emphasis on the workflow simulation towards the virtualisation of floatovoltaic systems in the
context of Figure 3.5b, the modelling and validation of the boxes of the simulation model design process
of the floatovoltaic ecosystem in Figure 3.5 indicate the critical stages in the development process of the
software model. Towards establishing an abstract representation of a production-environmental-economic
model in the design of a floating solar ecosystem, the solid lines in Figure 3.5 indicate modelling activities,
while the dashed lines between the process steps indicate the evaluation of the respective modelling activ-
ities towards the establishment of a digital twin model (or mode) of a synthetic floating solar PV ecosystem.
From sustainability analytics and the decision-making point of view, the proposed analysis-by-synthesis
simulation method for floatovoltaic project valuations lays the foundation underpinning the implementation
of Research Objectives 1, 2 and 3. By following the flowchart steps in the development sequence of the
simulation model in Figure 3.5b, the computer modelling and simulation methodology can facilitate the
formulation of research procedures and techniques (building blocks and interlinkages) in support of the
implementation of domain-specific solution objects or models towards an integrated dynamic floating PV
ecosystem model.
This overview motivates the methodology around computer modelling to answer the research questions
and validate the research aim and objectives in synthesising real-world photovoltaic ecosystem projects. To
this end, this section introduced and justified computer simulation as a methodological research approach.
The following section details the theoretical formulation, systems thinking, and systems theoretical design of
the proposed computer model for experimental data collection and analysis through a quantitative computer
simulation model.
89
3.3.2. Methodological sequence from initial concept to computer model
This section presents the methodological sequence from initial conceptualisation to computer modelling. It
moves the research from the conceptualised model to the computer model as a system of inquiry into the
reality of floating PV sustainability. In this context, it considers the methodological sequence of the research
design methodology to study the floating PV systems in terms of the research design process, as depicted
in Figure 1.6.
Applying the procedural steps of Figure 3.5b to the context of decision-practice-oriented scientific stud-
ies around geographical systems (i.e. EIA assessment studies), the schematic framework in Figure 3.6
outlines the essential modelling steps to perform a complex system thinking research project through in-
ductive research methods. With these steps, a system-thinking-oriented study can address the physical
modelling challenges emerging from the following aspects, namely (a) observing data towards identifying
and evaluating the impact factors associated with ecosystems; (b) investigating the sustainability impacts
of tempo-spatial patterns in terms of entities and relationships; (c) exploring opportunities to hypothesise
around explanations for the patterns in the data or techniques through systemic diagrams showing entity
relationships; and (d) exploring opportunities for project decision support towards effective goal-oriented
planning-intervention towards sustainability (Richardson and Pugh, 1981; Xia et al., 2017). In this context,
the conceptual modelling elements, encapsulated by Figure 3.6, offer the means to follow an inductive rea-
soning process toward building the theoretical and computational prototypes of the ecosystem in a way
commensurate to representing a real-world system as a means to overcome analytical problems.
Figure 3.6: Essential steps to perform complex systems thinking research, with directional arrows showing functional
modelling interrelationships (source: Xia et al. (2017), page 4).
In developing systems thinking principles to support practice-oriented analytical decision-support solu-
tions, Figure 3.6 underscores the core of the problem-driven conceptual modelling process towards defining
and refining the systemic components with the systemic interactional relationships and impact factors to
model an abstract representation of the real world. The EPA (2022) defines the four stages of the model’s
life-cycle in Figure 3.6 as the steps in the process towards taking the conceptual understanding of an en-
vironmental process to a fully analytical model in terms of problem identification, framework development,
model development, and model application in practice. In explaining Figure 3.6, the problem-driven con-
ceptual modelling step logically translates a real-world geographical solution into a conceptual model in the
90
theoretical or computational domain. The data-oriented real-world basis is devoted to defining the computer
programme’s logic and modelling the conceptual model’s architectural embodiment using real-world data or
the statistical analysis of real-world observations to determine the model parameters that define the optimal
analytical method. With the analytical computer model defined in Figure 3.6, the goal-directed analytical
inference aspect concentrates on applying the analytical solutions to address specific real-world problems
through analytical surveillance. Finally, evidence-based practice actions or decision support apply the ana-
lytical results to help find optimal solutions to meet any particular project goals. In this phase, goal-directed
analytical inference engages data from the developed model. It acts as an analytical decision-support tool
to provide problem-solving design methods or configuration solutions to address specific real-world prob-
lems. The overall research process, defined by Figure 3.6, moves the research from a theoretical concept to
implementing a model and validating and improving an envisaged analytical and decision-support solution.
It offers practical research steps to bridge the identified theoretical knowledge and computational analysis
gaps with support from observations made on real-world data.
With the philosophical and methodological approaches introduced, this section details the fundamentals
of the empirical research and development research approach to advance modelling research in dynamic
PV systems from a conceptual model to a computer model for experimental, analytical inference. The
following section focuses on defining the appropriate design-based research process steps required to
apply systems thinking in a theory-driven developmental research framework for floating PV technology
modelling to evaluate the model’s performance.
3.4. Methodological Framework
While the research design framework of Figure 1.6 offers an appropriate strategy or blueprint for deciding
how to approach the research study, this section defines the methodological framework or the steps in the
research framework to implement the research design process of Figure 1.6. In this context, the thesis dis-
cusses choices around selecting the research paradigm as a reasoned approach, which evaluates options
for choosing an appropriate research framework and methodology in which to conduct the research. The
discussion introduces the deductive and inductive research methods as methodological options for devel-
oping a new theory-driven framework for quantitative analytical synthesis. It considers the attributes and
benefits of the deductive and inductive reasoning approach as a paradigm for a systemic inquiry toward an
appropriate intervention measure into a solution. Section 3.4.1 discusses the options available for select-
ing an appropriate research framework and research methodology to work toward a proposed theoretical
assessment framework for floating PV systems in terms of acquiring knowledge about the reality around
floating PV sustainability. It also selects the appropriate research framework and methodology regarding the
attributes and benefits of the deductive and inductive reasoning approaches. Section 3.4.2 subsequently
defines the steps in the research framework methodology to establish a system of research inquiry that
describes the inductive research procedure for acquiring knowledge about the reality around floating PV
sustainability.
3.4.1. Options towards selecting a research framework and methodology
This section presents a narrative overview of the methodological approach that focuses on the research
framework and options available to apply computer modelling and simulation in the quantitative research
methodology. It highlights the goal of the research to choose a methodological framework capable of fa-
cilitating theory building towards defining a new theoretical assessment framework based on the systems
thinking methodology suitable for assessing floating solar technology projects.
91
Regarding applying the systems thinking philosophy aimed at combating analytical and impact assess-
ment deviations observed in the literature review of floating PV performance, this scientific study aims to
use systems thinking to improve the modelling of real-time floating PV technology ecosystems. To this end,
scientific research has the option to explore deductive or inductive scientific research methods in quanti-
tative data modelling, data collection and data analysis. Traditional empirical research has drawn a clear
line between research conducted with traditional empirical goals in mind and research inspired by theoreti-
cal or theoretical framework development goals (Reeves, 2006). Conventional empirical research methods
(including predictive methods), as depicted in Figure 3.7a, offer the opportunity for deductive reasoning,
starting with a general statement (hypothesis) and progressively examining the possibilities for reaching a
specific logical conclusion. This so-called "bottom-up” research direction descriptively studies the reality of
the outputs or impact occurrences of FPV systems through the collection of analytical or observational data.
The literature review of this thesis shows that deductive empirical research results have helped to uncover
anomalies in the tempo-spatial patterns of the systemic impacts of floating PV systems. At the same time,
this literature review has also helped to identify the characteristic variations in the impact effects of floating
PV systems in terms of their systemic attributes. In scientific floating PV assessments, however, deductive
empirical research has also uncovered several critical knowledge gaps (Section 3.2.2). It confirms that new
analytical frameworks and experimental methods are needed to evaluate new sustainability framework hy-
potheses relating to the proposed causes of the observed systemic impact effects within analytical floating
PV system modelling.
Theory
Hypothesis
Methodology
Observation
Confirmation
Theory
Observation
Data pattern
Tentative hypothesis
Methodology
Confirmation
Framework / Theory
(a)
(b)
Figure 3.7: Scientific diagram depicting empirical research process options, namely, (a) the deductive research
framework, and (b) the inductive research framework (after: Reeves (2006)).
As opposed to the traditional empirical research design of the pure evaluation, as depicted in Fig-
ure 3.7a, the design-based inductive research in Figure 3.7b is characterised by innovation, whereby
cutting-edge theoretical knowledge is analysed by utilising quantitative or qualitative methods to allow for
adjustments to the implementation of the theoretical design. Often applied as a design science research
method in the field of the environmental and education sciences (Ashley and Boyd, 2006; Reeves, 2006),
the inductive reasoning research framework, as depicted in Figure 3.7b, is sometimes referred to as design-
based research or new theory-driven developmental research. A so-called "top-down” research direction
reminds us of rounded theory, whereby a research method is concerned with generating a theory. As such,
the inductive research framework approach is differentiated in the need to resolve critical but fundamental
92
analytical characteristics through the development of solutions that require a new theoretical framework as
part of the research design and experiments.
In the context of extending the floating PV investigative research design depicted in Figure 1.6, the
research methodology can accommodate the steps of gathering data or reporting and observing data or
technique outcomes; detecting patterns or regularities or frequently reporting patterns and frequencies in
the data generated by FPV analytical techniques; deriving general principles or developing explanations;
making assumptions to formulate tentative hypotheses on theoretical principles to develop theory; or cre-
ating a model to validate the proposed theoretical construct of the assessment framework. Terms of the
scientific research framework diagram of Figure 3.7b, Reeves (2006) define theory-building research as an
iterative research process focused on: (a) addressing complex problems in real-world contexts by observ-
ing behavioural pattern discrepancies (often in collaboration with skilled practitioners); (b) defining tentative
hypotheses as solutions; (c) integrating robust analytical framework theory and theoretical design princi-
ples with technological potential (goal-oriented behaviour) to render plausible solutions to solve complex
real-world analytical problems; (d) conducting a rigorous scientific inquiry to evaluate or confirm the impact
of the new innovative framework or design intervention; and (d) conducting reflective scientific inquiries to
refine new framework models toward new planning decisions and design principles.
Considering the methodological options for the research design methodology for studying the floating
PV system dynamics that are based on the research design process of Figure 1.6 and the procedural
methodology steps of Figure 3.4, scientific research has the option to explore deductive or inductive sci-
entific research methods in quantitative data modelling, data collection and data analysis. Inductive and
deductive reasoning differ in terms of their reasoning logic and orientation. Inductive reasoning logic works
from the particular to the general, while deductive research logic works from the general to the particular
in using general rules or laws to test individual cases (Ashley and Boyd, 2006). Since the research frame-
work development technique of Figure 3.7b inherently improves interdisciplinary collaboration, it primarily
supports inductive-type theory development research that focuses on inductive reasoning to help create a
formal understanding and characterise real-world systemic properties observed in the floating PV system
sustainability reality. In the context of the research design framework of Figure 1.6, the inductive research
approach of Figure 3.7b offers an appropriate and well-established paradigm for scientific inquiry where
improvement of the theoretical assessment framework needs to play a central role in the design of the
research.
As this section details the fundamentals of the empirical research and developmental research ap-
proaches in the context of dynamic floating PV systems modelling, it highlights how the inductive design-
based research method or framework may serve as the foundation for developing a new theory-driven
framework for quantitative analytical synthesis and measures for solution interventions. Therefore, the fol-
lowing section defines the appropriate designed-based research process steps required to apply systems
thinking in a theory-driven developmental research framework for floating PV technology modelling.
3.4.2. Methodological framework and research procedures
While the methodological framework serves as a crucial tool to guide the research through the sequence of
steps towards completing the research procedure, this section defines the procedural steps in performing
the designed-based research process. It details the procedural steps towards implementing an empirical
systems thinking modelling methodology as the final part of the strategic research-design process. The
methodological framework defined in this section addresses the basic research components in Figure 3.6,
thus enabling the study to engage systems thinking in a theory-driven developmental research methodology
for floating PV systems.
93
The research problem and literature review established the need to construct an improved systems-
based theoretical framework to quantitatively model the real-world systemic properties and dynamics ob-
served in the reality of the floating PV system sustainability. From systems thinking perspectives, the re-
search methodology calls for generating new systems thinking theories and frameworks for conceptualising
a new analytical floating PV model design. To formulate a new integrated system thinking sustainability
assessment framework, inductive reasoning aims to establish a representative system and associative-type
causal relationships between the systemic properties and their impact occurrences. Furthermore, as a type
of rapid prototyping method (often improved in the form of iterative comparisons with reality), it includes the
use of interactive mathematical or computational synthesis models in the scientific enactment of floating
PV projects as a means to study and evaluate the effects of dynamic interrelationships among the range
of systemic components and impact factors. In this context, the inductive research design method involves
conceptualising a new theoretical framework as a logical blueprint for developing computer-assisted math-
ematical, computational or physics models to support analogical inferences in virtual dimensions. Such
models can be used jointly with case-driven scenario-based simulations as a predictive tool for theoretically
characterising the properties and dynamics of floating PV performance and the impact effects.
While the nature of a theory-driven research framework inherently leans toward the inductive reasoning
approach of Figure 3.7b, it calls for a practical definition of the specific methodological steps defined by
Figure 3.6 as a means to determine the exact steps required to move the research from the conceptual
framework modelling phase to the analytical inference phase. In this thesis, Figure 3.8 strongly correlates
with the procedural steps of Figure 3.4 to offer an appropriate inductive reasoning pathway for methodologi-
cal research to add to the body of knowledge or state-of-the-art subject theory in the field of the geographical
sciences. The design-based research steps of Figure 3.8 allow the systems designer to engage in the con-
ceptualisation of a new sustainability reference framework solution (tentative hypothesis) to help improve
the effectiveness of modelling the floating PV system components and interlinkages through model-based
design and networking parameters.
Study underpinning theory
of FPV sustainability
(Chapters 1, 2 & 3)
Observe data on FPV
sustainability assessments
(Section 3.5.1)
Find patterns/regularities
in FPV data or techniques
(Section 3.5.2)
Tentative hypothesis
to explain/frame patterns
(Section 3.5.4)
Conceptual modelling to
implement framework model
(Section 3.6, Chapter 4)
Experimental evaluation
to test hypothesis
(Chapter 5)
Figure 3.8: Methodological framework, research procedure and actionable steps towards a sustainability assessment
framework and model for assessing floating PV technology (source: author).
The development of an analytical framework focused on research methodology, as depicted in the
chronological research activity steps in Figure 3.8, provides essential milestone steps toward the creation
of new knowledge in support of the formulation of a new theoretical framework and computer model through
the inductive reasoning process. The inductive research process offers an integrative methodology when
an innovative modelling solution requires a new theoretical framework to support systems modelling. The
94
research steps, defined in Figure 3.8, methodologically serve to navigate the research process through
the steps of: (a) studying the underpinning theoretical principles of sustainability in FPV technology; (b)
observing the floating PV sustainability assessment data and techniques; (c) analysing assessment data
to find regularities or error patterns in the data from existing models or techniques; (d) conceptualising a
tentative hypothesis in terms of the framework of the reference system to explain patterns and regularities;
(e) formulating a systems methodology to bridge the knowledge gaps and limitations of existing models
and paradigms; (f) defining a conceptual systems model to implement the framework hypothesis; and (g)
using the framework theory as logic in a computerised systems model to validate the framework concept in
making an experimental evaluation towards confirming the framework hypothesis. As listed in Figure 3.8,
these steps direct the characterisation of the floatovoltaic system as implemented in the rest of this chapter.
This section detailed the fundamentals of the developmental research approach driven by the inductive
approach and in the context of dynamic floating PV systems modelling. It motivates the use of the inductive
research framework approach (Figure 3.8) as a sound paradigm for scientific inquiry while providing detailed
steps towards developing a new quantitative analytical synthesis and measures for solution intervention
for floating solar systems. To this end, the following section details the implementation of the inductive
research approach toward engaging system dynamics thinking into theory-driven developmental research
frameworks.
3.5. Theoretical Sustainability Assessment Framework for FPV
This section deals with framing the research paradigm in the context of conceptualising and realising an im-
proved systems-based theory-driven methodology for floating PV sustainability assessments. Keep in mind
the context wherein Section 3.2.3 presented a unique philosophical perspective to develop a systemically
integrated sustainability assessment solution to encompass the analytical evaluation of floatovoltaic system
operations, performances and impact effects in one sustainability criteria. In this context, the discussion
applies the research methodology in Figure 3.8 toward using systems thinking to explore the underlying
systemic principles and challenges of performance- and impact- assessment to improve floating PV sys-
tem sustainability modelling through the analytical characterisation of the floating PV technology ecosys-
tem. The discussion follows the research steps defined in Figure 3.8 to summarise data observations and
analyse data patterns. It also introduces a tentative hypothesis as a conceptual, theoretical sustainabil-
ity modelling solution for floating PV systems, explicitly referring to a theory-building framework approach.
Section 3.5.1 summarises the critical observations in the current data assessment techniques pertaining to
floating PV performance modelling in terms of the data patterns and challenges observed in floating PV as-
sessments. Section 3.5.2 derives systemic principles from assessing FPV data patterns to help overcome
the key challenges emerging from the observed patterns. Section 3.5.3 considers the systemic dynamics of
floatovoltaic operations behaviour, while Section 3.5.4 considers options for more advanced sustainability
criteria for floating PV modelling.
3.5.1. Observations and identified patterns of error in FPV performance assessments
This section summarises the observations and patterns that emerge from the theory and data, identifying
patterns that offer challenges and opportunities for defining the scope for innovation in the field of floating
PV characterisation. It extends the discussion of Chapter 2 on the general research and data observations
around knowledge gaps associated with floating PV sustainability assessment data and techniques.
Regarding the methodological research steps of Figure 3.8, the first two inductive reasoning milestones
have already been reached in the meta-analysis review of literature in Chapter 2. While Sections 2.2 to 2.5
study the underpinning theoretical principles of sustainability in FPV technology, Section 2.6 of Chapter 2
95
specifically deals with observations around floating PV sustainability assessment data and techniques. The
literature review revealed fundamental analytical problems with existing frameworks and methods available
to theoretically study and model the reality of sustainability in floating PV systems.
The fundamental challenge with characterising floating PV systems is that most present-day state-of-
the-art analytical modelling solutions for PV systems assessments are founded on the reductionist theory.
In the context of floating PV system characterisation, the reductionist theory is depicted by the abstract
subsystem representation of the linear processing framework model of Figure 3.9. While squares or circles
serve as metaphoric subsystems to explain systemic principles in an abstract representation, this figure
illustrates how performance calculations pertaining to technical energy engineering models are serially
fed into a subsequent economic subsystem sequentially or linearly (as also illustrated in Figures 2.15 to
2.18). Known as operational sector-based resource analysis and management approaches (Taguta et al.,
2022), reductionist thinking approaches such as Figure 3.9 fail to acknowledge and capture the interlinkages
between these connected system elements (including the WELF resources).
Technological Economical Environmental
Figure 3.9: Reductionist thinking in floating PV systems analysis, highlighting linear processing in a unidirectional
assessment framework, the absence of causal feedback, and the absence or low-priority role of the environmental
function in photovoltaic system modelling (source: author).
By implication, and according to Figure 3.9, reductionist thinking in silo-based analytical processing of
the floating PV complex cause the analysis of the energy subsystem model running its complete one-year
course before the chronological commencement of the lifetime processing of the economic model as the
next silo for post-processing. Therefore, it is scientifically logical to argue that the conventional techno-
economic PV performance model focuses on synthesising the energy operations to extract the technical
performance metrics and then derive the economic performance metrics in isolation mathematically. For
the benefit of EIA processes, the EIA practitioner may afterwards perform an impact assessment in isolation
(disconnected environmental subsystem in Figure 3.9).
With data processing in most conventional project evaluation models running according to the logical
sequence of Figure 3.9, it should be clear that performance processing in this silo-based approach (re-
ductionist thinking) suffers from a general lack of feedback between subsystem component interaction.
Moreover, a significant limitation of this linear framework approach in Figure 3.9 is that it primarily treats
economic and environmental performances as mere supplementary additions to the performance assess-
ment process. As a result, determining the environmental and economic impact effects of a PV system
in a linearised post-processing data analysis manner creates knowledge gaps that limit the pre-installation
project assessments to a periscopic view on theoretical performance assessment because it ignores the
micro-scale feedback of the natural environmental system on the performance of the energy system. As a
no-feedback system, the predictive calculations presented in Figure 3.9 that pertain to the performance of
the energy engineering model sequentially precede the performance analysis of the economic silo/bubble
(or, for that matter, the environmental silo/bubble), and without feedback.
The recent arrival of technology technologies has taught scientists that the inherent impacts of floating
PV systems (land-reclamation, water-preservation) play a more significant role in the natural environmental
96
system (Armstrong et al., 2020; Exley et al., 2021b). Therefore, this thesis argues that the reductionist
analysis framework, as an open-loop system, creates particular problems in the accurate assessment of
floating PV project qualities because FPV energy systems have much greater interaction with the natural
environmental system on the micro-climate and meso-climate scale compared to conventional GPV sys-
tems. Furthermore, when viewed from an EIA-driven sustainability assessment perspective, the open-loop
framework representation of Figure 3.9 underscores the practical limitations caused by unidirectional linear
techno-economic processing models if the micro-scale climatic effects induced by the natural environmental
subsystem are ignored, or treated as an add-on.
With the above discussion, the thesis reaffirms that the root cause of these analytical problems points to
critical knowledge and methodology gaps in three dimensions. Firstly, the reductionist thinking philosophy
is an overwhelming approach engaged by most current PV systems assessment models; secondly, the
low-priority role of the natural environmental micro-habitat in the modelling characterisation of floating PV
systems; and thirdly, inadequate modelling linkages with the WELF system resources. These aspects put
the spotlight on overcoming the observed knowledge gaps as opportunities for new geographical innovation,
as it provides a means to guide philosophical systems thinking in the next section.
3.5.2. Deriving systemic principles from analytical FPV behaviour data patterns
This section shows why anomalies are experienced with sustainability assessments when treating envi-
ronmental and economic considerations as a modelling add-on or post-processing afterthought (refer to
Figure 3.9).
In conventional PV performance models, the photovoltaic energy production system performance is a
function of the system’s location, solar resources, climatic conditions, technical design, mounting config-
uration, and cell module technology. On a rational level, the PV energy subsystem of Figure 3.9 already
establishes a complex system in terms of technical energy production (refer to energy processing by Sandia
(2020), for example). With the energy system being a system within a system in its own right already (multi-
physics PVlib and PVsyste models), the predictive PV energy model is effectively driven from the outside
by optical solar fuel, primarily represented by the optical solar incidence onto the floatovoltaic panels (refer
to Figure 2.17). This explanation means conventional modelling characterisation of the PV energy system
model already incorporates, to a limited extent, aspects of the natural environmental system to drive it.
In terms of incorporating natural environmental system functions into conventional PV modelling schemes,
the environmental micro-habitat is often reduced to a panel temperature conditioning factor influenced by
the structure and placement of the floating PV-carrying structure (direct and indirect evaporative cooling
under a cross-flow arrangement). Although the natural environmental influence may, on a simplistic engi-
neering level, logically seem to boil down to the estimation of the PV module temperature for water-based
photovoltaic systems, the temperature-dependent power modelling of photovoltaics may deviate under var-
ious (micro-)climatic conditions and FPV systemic interactions.
As such, FPV modelling imperfections or abnormalities occur because environmental and economic
elements are part of an operational (eco)system, and consideration has to be afforded to its mutually bene-
ficial influences in (eco)system context. On a rational level, the limitations posed by the reductionist thinking
approach in Figure 3.9 pose a fundamental problem on the systemic-theoretic level. It causes discrepancies
in FPV assessments, commonly known among scientists as "technology unknowns” of floatovoltaics (Banik
and Sengupta, 2021; Cagle et al., 2020; Gadzanku et al., 2021a). The thesis argues that many of these
technology unknowns in the peculiar performance phenomena of FPV technologies are systemic causes
having their roots in environmental and economic oblivion.
In this context, the study aims to create new knowledge by posing a system-based theoretical construct
that can explain the anomalies in water-mounted PV behaviour (compared to conventional ground-mounted
97
PV). In this respect, the guidelines for systems thinking innovations in the pyramid of Figure G.1 should
serve as a navigation compass. Identifying systemic anchors in floatovoltaic behaviour analysis can start
by taking the reductionist thinking’s deconstructed elements in Figure 3.9 as a point of departure. The next
step is understanding how these three functional systemic components (energy, economics, environmental
= 3E operating elements) function together in real-world operations. The approach inspired the idea of
engaging geographical operations research principles (Kumar and Mani, 2021; Sterman et al., 2015), and
analytical smart systems thinking principles of the 4IR paradigm (Groundstroem and Juhola, 2021; Stanton
et al., 2019), towards the exploration of the real-world evidence though a multi-sector system dynamics
causal diagram (Kundi, 2006; Rodriguez-Ulloa and Paucar-Caceres, 2005). Studying the dynamics of the
3E elements of the technology system in a fully functioning FPV installation in farming-scale operations
can help to identify synergistic interactions between the three system elements. The operational goal is
to identify and explore particular mutual interactive responses in the operational ecosystem that may have
been neglected if viewed in isolation or the systemic functioning interactions that is unique to the operational
FPV ecosystem. The strategic goal is to derive system thinking principles from pattern analysis and to
capture foundational dynamic interactions and principles in a cognitive mind model that can form the basis
for conceptualising a new theoretical construct.
However, these environmental considerations include environmental sensor data centred around ex-
traterrestrial solar irradiation and mesoscale meteorological conditions that impact solar fuel as local solar
energy insolation. This aspect is metaphorically illustrated in terms of the abstract environmental half-
system on the left-hand side of Figure 3.10. The dotted lines in Figure 3.10, however, illustrate how most
conventional PV performance models either ignore or give insufficient attention to the remainder of the
natural environmental system (refer to (Ahn et al., 2021; Armstrong et al., 2020; Prinsloo, 2020)).
Environmental
(Solar)
(Meteorology)
Technological Economical
(Microhabitat)
Environmental
Figure 3.10: Considering the position of the environmental function in a floatovoltaic model automata diagram, rel-
evant systems thinking considerations for positioning environmental systems in the floating PV ecosystem as an
integrated cohesive framing ensemble (source: author).
Having identified the indispensable role of the environmental system and its associated climate-health-
and resource-economics as anchor elements and imperative aspects of floating PV performance modelling,
the question then asked is how to integrate the environmental and economic system influences in a sys-
temic invention that would ensure improved floating PV performance modelling (to provide enhanced or
more realistic performance assessment predictions in the case of FPV)? The methodological framework
of Figure 3.8 can help to overcome this problem through the inductive reasoning process, as it allows for
important ideas to emerge from the data as it searches for patterns and trends in the data observations
to make generalisations about progressive theoretical framework concepts. Towards making inferences
to best explain the observed data patterns, this section learns from the data observations to define data
patterns and derive general principles from the FPV assessment data patterns and trends.
Not incorporating an environmental function in floatovoltaic modelling, as illustrated in the dotted lines of
the expanded illustration in Figure 3.11 emphasises the consequences of ignoring micro-scale environmen-
tal functions and feedback functions in the PV model. Therefore, according to Figure 3.11, representation
of the microscale meteorological conditions and reciprocity-type hydroclimatic impacts in the floating PV
98
system’s microhabitat is critically essential for floatovoltaic modelling. The current floating PV performance
modelling’s missing environmental and economic components are generally represented on the right-hand
side of the processing sequence of Figure 3.10.
Environmental
(Solar)
(Meteorology)
Technological
(Energy production)
(Climatic reform)
(PVlib/PVsyst)
Economical
(energy-economics)
(resource-economics)
(climate-economics) (Aquatic-habitat)
(Hydroclimate)
Environmental
Figure 3.11: Considering the position of the environmental function in floating PV systems modelling automata dia-
gram, relevant system dynamics thinking considerations for linking environmental and environmental economic system
functions into the floating PV ecosystem towards a systemically integrated modelling framework (source: author).
As a structured approach to object-oriented analysis and design type, Figure 3.11 offers a type of ac-
tivity and sequence diagram to help the system designer identify the FPV ecosystem’s systemic objects,
attributes, associations and behaviours (Yadav et al., 2021). As such, Figure 3.11 illustrates the thesis’ ar-
gument that, first and foremost, a numerical reservoir-scale hydroclimatic model is needed to characterise
the FPV microhabitat properly. This factor includes defining the representative environmental function and
feedback of the farming area and the reservoir, the land and the aquatic habitat of the floating PV system
ambience. Secondly, Figure 3.11 illustrates the need to consider natural resource economics and climate
health economics as feedback mechanisms within the same floating PV ecosystem complex. Thirdly, the
dotted lines in Figure 3.11 highlight the oscillating or reverberating effects of the micro-environmental habi-
tat on the floating PV ecosystem. Overall, these influences and feedback effects are directly and indirectly
responsible for the broader and diversified spectrum of characteristic impact effects of floatovoltaic tech-
nologies.
The general representations of Figures 3.10 and 3.11, as such, call for modelling improvements that
make provision for the various broad-spectrum environmental and environmentally-derived technical and
economic feedback effects. Furthermore, adequately defined environmental and economic sub-models
would help to overcome the deficiencies caused reductionist-type silo-based floating PV systems model
(microclimatic impacts, land/water-surface albedo, land-water resource impacts, carbon footprint, carbon
taxation, carbon credits, etc.). As such, proposed models to predict floating PV sustainability under a
broader environmental impact definition require deeper insights than just the operating temperature of pho-
tovoltaic module technologies under varying climatic conditions.
3.5.3. Understanding the systemic dynamics of floatovoltaic operations behaviour
To overcome the limitations of the open-loop modelling framework of Figure 3.10, this thesis argues that a
theoretical performance assessment of sustainable energy technologies needs to emphasise the systemic
dynamics of environmental influences together with that of the financial benefits of climate health in the
holistic sustainability assessment of new energy technologies.
Systems thinking according to the graphical representation of Figure 3.11 inspires the thesis to probe
the response of the systemic interfaces as an evolving environment for response and understanding of
99
the environmental influences and impacts, as well as the climate change economic influences and impact
effects on the sustainability of an FPV system model in Sections 3.5.1&3.5.2. A complex systems modelling
problem, such as depicted by Figure 3.11, requires systems theory as a method of action research to
improve the understanding of the structure and dynamic behaviour of the integrated (eco)system (Jackson,
2010; Sterman, 2003). With its roots in the theory of communicative interaction (Habermas, 1981), dynamic
systems theory gives recognition to the strategy that the structure of the (eco)system with its interlocking
and circular relationships among the components of the FPV system (Figure 3.11) needs to be understood
as these elements determine the behaviour of the entire system (Midgley, 2003; Newman and Newman,
2020).
As a starting point for considering the environmental impact effects of floating PV systems at ecosystem
level, the thesis focuses on the requirement for environmental variables to be considered in floating PV
assessments as an extension of EIA requirements (void to opportunity). This approach brings to the fore the
issue of inter-organisational relations between the energy production ecosystem components in agriculture
contexts, and how it can help to conceptualise a framework ontology for understanding the impact of energy
production on sustainable agriculture (FAO, 2022; Velten et al., 2015).
In this context, the known dynamics of the systemic interactions among the water-energy-land system
components, depicted in Figure 3.12a, already underscore the crucial role of the natural WELF nexus
resource economy as a critical part of the integrated sustainability assessment of floating PV ecosystems
(Gadzanku et al., 2021b).
Technological
Environmental Economical
Energy
Water Land
(a) (b)
Figure 3.12: Abstract view of floating PV system automata diagram in which (a) an absent environmental system
function in conventional techno-economic analysis misses key links in the dynamic role of the natural environment in
the FPV ecosystem, especially in terms of (b) the WEL(F)-nexus resource system interactions (source: author).
The framework paradigms depicted by Figure 3.12 provide a holistic approach to the sustainable devel-
opment of energy, offering a different school of thought that considers agricultural energy production as part
of the broader realm of water, land and environmental ecosystem (refer to (Armstrong et al., 2020; Haohui
et al., 2020; Urbaniec et al., 2017)).
Even when considering the operational floating PV system components of Figure 3.10, the positioning of
the environmental function in floating PV systems modelling requires rationalised systems-design thinking to
establish a new non-conventional modelling framework for systemic assessments according to Figure 3.12a.
In natural geography, nature teaches us that environmental interactions are part of larger natural cycles. In
the same way, the cyclic floating PV ecosystem representation depicted in Figure 3.12b offers a more
organic systemic structure as a natural resource representation of the environmental system’s functions
around the nerve centre of the floatovoltaic ecosystem.
100
From a principal component analysis perspective, rethinking of the ecosystem network structure and
priorities calls for cognitive synthesis of the dynamical operations of a functioning floating PV system. Such
PV project recreation essentially requires causal network reconstruction to ensure contextual intelligence is
maintained in a location-based data framework architecture (Runge, 2018). As such, an integrated system
dynamics assessment concerns their systemic modelling and simulation, and their integrated optimisation
of sustainable technologies. Networking type agency theory from the information and communication tech-
nology field (Chen and Poquet, 2022; Klemm et al., 2012) offers a link between (geographical) rational
theory and organisational theory (Zey, 2001). This theory enabled the thesis to graphically portray crucial
systemic elements, boundary conditions and behaviour patterns to help characterise the operational and
sustainability behaviour of a floating PV system in a cross-domain object-process network (entity relation-
ship causation diagram in Figure H.1). This systemic whole-systems entity-relationship diagram further
facilitates identifying and capturing the fundamental systemic boundaries and components as imperatives
of floatovoltaic identity, behaviour and responses in the simplified entity relationship diagram of Figure 3.13.
Energy
Environment
Economic
Solar
Air
Land
W ater
Figure 3.13: A holistic causal loop diagram of the floating PV ecosystem in a closed-loop framework to illustrate
the key role of the natural environmental system in the behaviour of the FPV system and to highlight the important
ontological causal relationships that are often ignored in open-loop analytical assessment models (source: author).
Through the application of the proposed sustainability portfolio and theoretical principles of Figures 3.12
and 3.13, the research of this thesis can postulate a more organic structure for floating PV model devel-
opment, whereby the topological system abstract of Figure 3.13 resemble a more holistic-type systemic
intervention. In this context, the trans-disciplinary inter-domain 3E relationship diagram in Figure 3.13
emphasises the imperative role of the underlying natural environmental system and cycles as part of the
dynamic networking relationship in digital photovoltaic system modelling.
Since Figure 3.13 identified the importance of an environmental focus and the indispensable role of the
environmental system in a resource-oriented floating PV performance modelling, the quest is to facilitate
systemic integration of the environmental system and its environmental economics parallel according to the
principles of Figures 3.12 and 3.13. While system dynamics causality diagramming offers a graphical tool
that visualises the relationships between the exposure of interest and draws systemic assumptions from
these conclusions (Rodriguez-Ulloa and Paucar-Caceres, 2005), the causal diagram helps to inform the
systemic design and interpretation of observational studies (Staplin et al., 2017).
101
Based on the interpretation of real-world evidence, a causal analysis facilitated the exploration of the
systemic properties of the floatovoltaic ecosystem in terms of population equivalences, systemic associa-
tions, temporal relations, environmental equivalences, and boundary conditions. Inferences from conceptual
data analysis thus helped to capture properties of the underlying systemic processes in the relationship dia-
gram developed by the thesis in Figure H.1 (refer to the causal chart as a type of object-process diagram in
Appendix H (Coletta et al., 2020; Kohen and Dori, 2021)). The floating PV ecosystem model components,
boundaries and inter-relations are identifiable in Figure H.1, offering a sound systems engineering basis
for applying an object-oriented approach. This methodology is a comprehensive modelling approach for
integrating the function, structure and behaviour of floatovoltaic operations in a single unified model based
on an object-process method (refer (Kohen and Dori, 2021)).
While this section aimed at explaining data patterns and deriving general principles from FPV assess-
ment data patterns to learn from the body of observations, the following section considers how the study
can interpret these systemic patterns to come up with general principles that can synthesise a tentative
hypothesis towards a new integrated assessment framework for floating PV systems.
3.5.4. Formulating a tentative theoretical hypothesis to frame FPV sustainability
Towards creating a technology-enhanced assessment environment for floating PV, this section introduces a
newly proposed systems-based assessment framework to provide new design principles for a hypothetical
system with the necessary technological provisions capable of rendering plausible analytical solutions in
the floating PV context.
Aiming to meet Research Objective 1, the study can consider the inference mentioned above that floato-
voltaic energy ecosystem sustainability modelling can leverage concrete ecosystem agglomeration benefits
offered by the techno-environmental-economic ensemble to improve cross-disciplinary ecosystem domain
connection modelling through agency forming. To this end, the thesis postulates that the dynamical sys-
tems view of Figure 3.13 shifts the paradigm toward a broader-spectrum sustainability modelling strategy
toward an improved framework for characterising and assessing a floating PV technology phenotype. As
such, it highlights the need to augment the scoping capabilities of FPV projects towards a more holistically-
integrated approach that incorporates joint energy-environmental-economic considerations into the theoret-
ical quantification of floating PV performance assessments.
The thesis’s proposal is a novel integrated multi-criteria sustainability foundation for floatovoltaics that
uniquely incorporates environmental understanding with environmental economics concepts into the tech-
nical energy performance and impact assessment of the diverse geographical behaviour of floatovoltaic
technology. Within this diversified and yet integrated view of sustainability, according to the causal network
proposed by this thesis in Figure 3.13, this study presents holistic sustainability integrity (HSI) valuation
strategy for floatovoltaic performance modelling. As a tentative proposition suggested as the solution to
a vital part of the research problem, the proposed holistic sustainability integrity valuation strategy estab-
lishes the baseline for a holistic thinking paradigm as a tentative research hypothesis, which is defined by
the thesis in the logical supposition as follows:
Lemma 3.5.1 Tentative Sustainability Criteria Hypothesis:The sustainability attributes and natural re-
source impacts of agricultural-type floatovoltaics can be derived from the tridentate responses of the technical-
energy, economic and environmental domain operations and its collective ecosystem-level interactions.
While the associative relational propositions proposed by this hypothesis are visually depicted in the
mind model diagrams of Figures 3.12 and 3.13, the proposed holistic sustainability hypothesis offers an
objective, logical framing for a proposed integrated sustainability integrity valuation strategy for floating PV
102
technologies. These contextual mind model diagrams further emphasise the indispensable role of the envi-
ronmental system in characterising the diverse set of broad-spectrum sustainability aspects of floating PV
systems. As an abstract view of the operational floating PV energy system, Figures 3.12 and 3.13 further
underscores the crucial role of the environmental, water and land components as part of the WEL(F)-nexus
system in Figure 3.12a. As such, Figure 3.12b shows how an absent environmental function in a conven-
tional analytical framework, as presented here, can miss crucially essential links in the dynamic feedback
role of the natural environmental system in the FPV ecosystem. This viewpoint overcomes the problem
where the interaction between the WELF resources is considered independently of the critical aspect of the
modelling system, without considering the opposing view or modelling the internal relationships among the
subsystems within the concept of the overall system (Afshar et al., 2021; Gadzanku et al., 2021b). From
an environmental modelling perspective, the EPA (2009) defines a conceptual model as a descriptive nar-
rative depicted by a conceptual diagram of a system that shows the complete set of relationships and flows
amongst the system’s components.
From a scientific, operational research perspective, the study needs to translate the (sustainability)
hypothesis into a conceptual framework model to offer a working theory of the essential factors that govern
the behaviour of the process of interest (FPV sustainability) (EPA, 2022). In this context, Figure 3.12 makes
a point of the fact that there is significant scope to engage complex holistic systems thinking as an improved
paradigm to bridge the real-world knowledge gaps through a newly proposed philosophical systems thinking
and modelling approach towards an enhanced understanding of floating PV systems. In this respect, the
thesis formulates the following tentative framework theorem proposition:
Lemma 3.5.2 Integrated Sustainability Assessment Framework (ISA-framework):An integrated sus-
tainability assessment framework can be defined as a combined co-simulation framework that models the
combined effects of technical energy, economic and environmental trilemma as interacting principal compo-
nents. It uses an integrated dynamic geo-sensitive systems approach to serve as a contextual framework to
explore value chain analysis in floating solar developments, and to quantitatively appraise the sustainability
attributes of floatovoltaic technology planning scenarios in a heuristic analytical decision space.
As such, the tentative hypothesis of the thesis rests on the philosophical principles that agricultural-type
floating photovoltaic technology sustainability can be characterised through a theoretical framework that
combines the tridentate technical, economic and environmental responses of floating photovoltaic opera-
tions at the ecosystem level. This hypothesis directly translates from the relationship diagram developed by
the thesis in Figure H.1, wherein the literature review inspired the decryption of the cause-effect interactions
identified in an ecosystem agency analysis approach. As such, Figure H.1 highlights the value proposition in
orienting the baseline hypothesis of the thesis toward the systemic integration of sustainability performance
characterisation towards drafting the system boundaries, identifying the principal components and principal
agents, as well as the recurring dynamic interactions between the ecosystem actors as cause-and-effect
relationships of FPV system sustainability.
As a metaphor for systems thinking, the philosophic abstraction of the system’s ISA-framework can
therefore be graphically represented by concentric circles, where any intersection of the overlapping cir-
cles in a cognitive schematic diagram represents systemic interactions, similar to mathematical set theory.
Rethinking environmental sustainability in the FPV context on the philosophical level, the thesis postulates
that systems thinking and the systemic institutionalisation of the floatovoltaic ecosystem is key to filling the
critical knowledge gaps around the distinctive characterisation of floatovoltaic technologies. This thinking
advances the state-of-the-art toward establishing a conceptual reference framework for a sustainability as-
sessment of an ecosystem as a means of modelling the contextual and geodynamical responses inherent
to floatovoltaic technologies is cognitively depicted in the conceptual abstractions in the schematics diagram
103
of Figure 3.14a. According to the systemic thinking principles illustrated by Figure 3.13, this newly proposed
analytical processing framework is defined around incorporating environmental understanding into the char-
acterisation of floatovoltaic technology behaviour. The postulated framework depicts how the environmental
effects and environmental resource/health economics should be firmly engrained into the characterisation
(DNA model) of the floating PV ecosystem. Furthermore, instead of applying a serially-phased approach
between the three principal components, the application of a system dynamics approach according to Fig-
ure 3.14a address the issue of processing fragmentation. The abstract mental model conceptualisation in
Figure 3.14 allows the thesis to uniquely incorporate environmental functions and their associated climate-
health- and resource-economic functions with the energy system functions as anchor elements in floating
PV operations modelling.
Technological
Economical Environmental
sustainable
feasible regenerative
viable
Technological
EnvironmentalEconomical
sustainable
(a) (b)
Figure 3.14: Posited sustainability impact assessment framework as a balancing trilemma, depicting (a) the integra-
tion of 3E domain expressions into a synergetic smart FV-systems-thinking sustainability framing, to form (b) a novel
sustainability triangle as integrated multi-criterial sustainability system with trade-off scaling opportunities between its
cross-functional systemic elements (source: author).
With the 3E trinity conceptualisation for the theoretical assessment and modelling framework for the
sustainability profiling in the 3E framing abstraction in the cohesive ensemble shown in Figure 3.14, the
thesis, in effect, postulates a generic sustainability theory as a whole-system integrated sustainability cri-
teria and assessment strategy for floatovoltaic technologies. On a philosophic-theoretic level, the universal
framework archetype posited by Figure 3.14 takes a fresh theoretical look at multi-disciplinary environmen-
tal and climate risk analysis through the lens of 3E integrated sustainability analyses, whereby sustainability
quality is distinctively assimilated in terms of the whole-system energy, economic and environmental (3E)
fundamentals as pillars of sustainability. While considering the critical imperatives in the science of the total
environment of operation, the proposed multi-disciplinary framework of Figure 3.14 changes the analyti-
cal perspective from an environmental impact assessment view to a more integrated and systems thinking
sustainability impact assessment viewpoint.
The tentative hypothesis depicted in the theoretical reference framework of Figure 3.15a provides for
the modelling of the three critical floating PV system components and the dynamic cross-functional causal-
ity properties of the system to realise cross-system impact effects over time. As an integrated reference
framework for sustainability assessments, used in studying the reality of sustainability in floating PV ecosys-
tem operations, the systemic sustainability framework of Figure 3.15a inherently incorporates all the WELF
nexus systemic interactions cognitively depicted by the WEL nexus abstraction of Figure 3.15b. Figure 3.15
104
shows the proposed conceptual, integrative framework for real-time system dynamics sustainability mod-
elling for establishing a floating PV virtual powerplant archetype as an analysis-by-synthesis FPV model.
As a pluralistic-type systemic intervention, the abstract framework illustrated in Figure 3.15 highlights the
operational floating PV energy system in terms of: (a) the integrative reference framework showing the
technical, environmental, and economical operations; and (b) the mutually-inclusive water-energy-land(-
food) resource ecosystem as an integrative WEL(F)-nexus systems framework. This integration can model
the functionally operating systems objects of the floating PV system, together with the dynamic interplay
between the TEE and WEL couplings of the floating PV system in an integrative system thinking paradigm.
Technological
Economical Environmental
sustainable
feasible regenerative
viable
Energy
Water Land
sustainable
feasible conserve
viable
(a) (b)
Figure 3.15: Systems thinking in terms of a reference framework for a conceptual sustainability model, with abstract
representations for (a) an operational floating PV ecosystem and interactional mapping, and (b) a floating PV energy-
land-water-resource ecosystem and nexus interactions (source: author).
Furthermore, apart from the natural environmental systems considerations included in the floating PV
system, Figure 3.15 also allows for integrated water-energy-land or WEL nexus assessments as part of the
theoretical floating PV system performance analyses. As a mutually inclusive set of systems, the environ-
mental and energy models of the floating PV operational ecosystem that functionally operate between the
3E system elements and the WEL resource system in Figure 3.15 offer an interdisciplinary view of the sus-
tainability resolve, which can address many of the environmental and resource-nexus concerns. As such,
the integrated assessment framework of Figure 3.15 further aims at objectively operationalising essential
scientific indicators to characterise agricultural floatovoltaic ecosystems along the WELF nexus paradigm.
In this context, Figure 3.15 aims to re-link the resource dependencies among the WEL nexus resources and
to capture the FPV-WEL identity of the intricate dynamic interlinkages. In so doing, it ensures the profiling
of floatovoltaic installation performance profiles along the agricultural WEL resource collaboration context
in a balanced scorecard approach (Prinsloo, 2020). With the inter-dependent geographical domain objects
combined in unity, each functional domain object contributes its respective mutual resource exchange inter-
actions to affect the ecosystem’s mutualistic symbiosis. Illustrated by the modulated intersections among
the unified simulation framework of Figure 3.15, the framework accommodates both the operational and
natural flow of events in a virtual digital business model, a typical floatovoltaic enterprise.
Moreover, based on the modelling considerations depicted in Figure 3.3b, the thesis poses the argu-
ment that the theoretical solution for solving the sustainability modelling problem, through the theoretical
framework conceptualisations in Figure 3.14fig:GeoTriangleYz, can automatically encompass a solution
105
for the operational behaviour synthesis, performance analysis, impact assessment and sustainability syn-
thesis of floatovoltaic systems. Therefore, shifting the floating PV project assessment paradigm towards
broader-spectrum sustainability suitable to floating PV modelling and the stipulated regulations for EIA and
carbon taxation, there is a need to augment floating PV project-scoping capabilities towards integrating
environmental-economic considerations into the theoretic quantification of floating PV performance assess-
ments. To investigate dynamic interaction models within the broader operational floating PV ecosystem,
Figure 3.14 provides a dynamic abstraction of the research’s theoretical intervention. If implemented as a
time-driven framework, the diagrammatic scientific framework of Figure 3.14 can act as a dynamic opera-
tional framework or computer programme logic for ontology modelling in a real-time dynamic systems-based
computer model construct. In this respect, the thesis formulates the following modelling hypothesis as a
tentative modelling theorem proposition:
Lemma 3.5.3 Computer Analytical Simulation Model (CAS-model):An analytical computer simulation
model can establish a contextual geo-sensitive sustainable development assessment and decision-support
ecosystem by logically integrating floatovoltaic operational dynamics into the tridentate co-simulation as-
sessment framework topology of Figure 3.18. In this figure, empirical algebraic functions represent the
systemic components and inter-playing ontological relationships in a computer-synthesised environment
that requires the energy, environmental and economic models to run sequentially at a single simulation
timestep.
The CAS-model modelling hypothesis sets the goal of extending the existing technical floating PV project
prototyping capabilities, in that the modelling project does not require the full development of a new model,
but rather the conditioning or extension of existing and established photovoltaic models (Gadzanku et al.,
2021b; Spencer et al., 2019). This scenario would shorten the model life-cycle as it involves model de-
velopment steps that integrate technical-energy-environmental-economic considerations into a quantified
theoretical performance model of a PV system. This approach allows for the realisation of a more holistic
integrative parametric model that gives increasing prominence to the environmental function and the re-
quirement for reinforcing environmental responses in the responses of the floating PV ecosystem model.
As such, the geographical systems architect or systems designer needs to reverse engineer the floating PV
architecture by mapping its analytical strategy, based on the systems thinking philosophy, to establish an
experimental technique that creates and implements a unified run-time analytical tool. With the framework
of Figure 3.15 serving as a blueprint for a computer programme logic, a new system dynamics framework
and the computer model are to be developed by this thesis to help understand floating PV systems in terms
of systemic principles. A more dynamic systems thinking framework as computer logic for a dynamic sim-
ulation model for FPV could fast-forward the real-time simulations to predict broad-spectrum sustainability
performances ahead of time (on a theoretical basis and ahead of installation). Such a dynamic interactive
framework, model, and methodology may improve the performance profiling and prediction accuracy of ex-
isting theoretical floating PV models, cut down on time spent in complex floating PV project assessment
analyses, and provide a more consistent framework for more comparative project assessment approvals.
According to this syndicated definition of sustainability defined in the framework hypothesis of Fig-
ure 3.14, the thesis can continue to apply the postulated theoretical framework as a means to investigate
the relationships between a floatovoltaic system and its environment. In terms of creating new knowledge,
the thesis postulates that system-wide trilogy real-world interdependencies between techno-economic fea-
sibility, enviro-economic viability, and techno-enviro regeneration intrinsic to a floatovoltaic ecosystem in
Figure 3.14 are critical drivers to the accurate sustainability characterisation of floatovoltaic technology. To-
wards the research design processes of establishing a software model to implement the research method-
ology, Geoscientific Model Development.
106
In this context, the schematic representation of the conceptual modelling synthesis framework for the
sustainability assessment model in Figure 3.16 can model the principal components and interlinkage as
vital elements to help distinguish sustainability assessments from the integrated sustainability assessment
reporting required by EIA processes and protocols. Therefore, towards the operationalisation of the newly
proposed framework archetype, the abstract mental model conceptualisation in Figure 3.14 allows the thesis
to uniquely apply a concurrent/parallel processing type of system dynamics co-simulation approach to the
simulation of a floating PV ecosystem in an integrated 3E project value-chain analysis and assessment as
an interlaced triquetra.
(microscopic perspective = individual functionary domain components)
Energy sub-system Environmental sub-system Economic sub-system
Integrated Floating PV System Analysis
(macroscopic perspective - combined structure, function and goal)
Figure 3.16: Systems thinking applied to analytical floating PV system modelling, integrating functional intelligence
components in a synergetic systems-oriented digital synthesis model (source: author).
According to the proposed co-simulation framework depicted by Figure 3.14a, and the floating PV mod-
elling regime of Figure 3.16, multiple 3E submodels (or EEE model objects) are exchanging data as the
simulation advances through each time-step iteration. From a cross-functional perspective of the technical-
environmental-economic decision-making theory, principal component analysis based on the integrative
energy-environmental-economic (3E) trilemma variables provides a valuable framework for considering the
three main broad objectives required for installing a sustainable energy system. Instead of treating the
economic and environmental responses of floating PV systems as supplementary processing entities or
objects in an open-loop relationship with the technical response, the theoretical framework conceptuali-
sation of Figure 3.14b proposes a novel theoretical framework foundation to process the economic and
environmental responses as closed-loop processes to be tightly integrated and to form part-and-parcel of
each of the individual floating PV project enterprise simulation process steps. The 3E trilemma, depicted
by the theoretical framework conceptualisation of Figure 3.14b, thus implies that all three integrated dimen-
sions need to be considered when assessing or making fundamental trade-off decisions about floating PV
project sustainability in the conceptual framework of Figure 3.14a.
The proposed synergetic floating PV modelling regime of Figure 3.16 aims to help digital twin creators
to establish an operational FPV model. This modelling regime can engage communicative coordination and
a macro-theory perspective according to the Habermas (1981) discourse theory as a means to account
for the microscopic processes that constitute the web of interactions capable of redefining the assessment
framework to account for the combined structure-function and goal of the systems paradigm. Furthermore,
as a transitional approach to improved floating PV characterisation, the proposed synergetic framework
modelling concept of Figure 3.16 provides an interdisciplinary approach to whole energy systems modelling
for floating PV systems, thus serving as a means to provide insight and understanding into systemic op-
erations in the context of multiscale challenges, opportunities, and solutions at different levels and scales,
and on different time steps. As such, the newly proposed and defined floating solar power plant framework
and model can help to uncover and identify many more of the system-based interdependent linkages of the
107
real-world FPV (hidden FPV project ecosystem treasures), and quantify many of these "technology impact
unknowns” in a fully integrated analysis-by-synthesis model for floatovoltaic technology.
The proposed conceptual geographical framework of Figure 3.18 and the proposed computer framework
model of Figure 3.16 serve to reveal the basic systemic and theoretical foundations in a new methodolog-
ical approach toward realising and validating the hypothetical research purpose in a scientifically-founded
paradigm geared to experimental modelling evaluations. In this respect, the thesis formulates the following
tentative decision-support hypothesis as a theorem proposition for decision-space modelling:
Lemma 3.5.4 Analytical Hierarchy Process decision-model (AHP-model):Indicator metrics derived
from the analytical computer simulation model of Figure 3.16 can drive a practice-based decision-support
system as an indicator and decision-support profile mapping display to serve as a visual decision-supporting
interface for floatovoltaic project planning assessments in a heuristic analytical decision space.
As a means to manage the trade-off between water-energy-land resources in small-scale agricultural
energy settings through a dynamic digital system, the modelling methodology needs to augment the floating
FV ecosystem panorama by adding local ecosystem contexts and functions to take project-planning deci-
sions around the sustainability of a proposed floating PV installation. While predicting and increasing the
sustainability features of a project can be embedded in a systems thinking approach, the model for systemic
change on project sustainability assessment analysis needs to be integrated into the fabric of an analytical
and decision-support model towards improving the existing real-time photovoltaic modelling endeavours.
This section addresses barriers to floating PV project deployments by defining an analytical framework,
computer model and decision-support display by emphasising efforts and concepts designed to enhance
sustainability assessments around the environmental function according to Research Objective 1. Framed
around the idea that an increase in the prediction accuracy of floating PV systems is likely, the following
section discusses the research model design and realisation in preparation for implementing Research
Objectives 2 and 3 in the next chapter.
3.6. Research Model Design and Realisation
Finalising the research model design and realisation, this section integrates the proposed systems-based
floatovoltaic computer synthesis approach toward digital scientific computer modelling and the simulation
technique as a modern-day research methodology and instrument. It focuses on the computer modelling
strategy and flowcharts to advance the project from engaging the framework product of Research Ob-
jective 1 towards implementing a computer model and decision-support interface according to Research
Objectives 2 and 3 in the next chapter. Thanks to the conceptualisation of a new theoretical framework
and modelling approach to floatovoltaic performance profiling and impact characterisation in the previous
section, this section aims to establish the foundations for translating the proposed analytical modelling
framework of Figure 3.14 into a ground-breaking modelling expression or synthesis model for floatovoltaic
ecosystems.
The discussion thus guides the reader through the final step of the research methodology framework
of Figure 3.8 towards realising and validating the hypothetical research purpose in a scientifically-founded
experimental modelling solution and evaluation instrument. The overarching steps toward the computer
modelling of the new framework for FPV assessments are narrated in Section 3.6.1. The sustainability
assessment framework system and the logical representation of the sustainability computer model construct
are detailed in Section 3.6.2. Section 3.6.3 lays the foundation for implementing the research model in the
next chapter, as it describes the integrated research technique and flowchart diagrams for implementing the
floating solar photovoltaic systems models as geoinformation decision-support layers.
108
3.6.1. Computer modelling of a systems framework for FPV assessments
The research focuses on finding a solution for real-time and life-cycle analysis towards theoretical sus-
tainability performance characterisation through predictive digital systems modelling and analytical digital
energy systems planning. Based on the consideration of Figure 3.3), this section focuses on the realisation
of a regulatory assessment framework in the technology-enhanced assessment environment for floating
PV systems in that it describes the process of modelling the conceptual framework as a system dynamics
computer model.
In the context of interactive modelling for information systems design, Figure 1.6b differentiates between
the avenues of exploring a system, whereby we can study the energy system through experimentation with
the actual real-world system or with an experimental synthesis model of the energy system. While the sys-
tems thinking methodology, together with its models, fully appreciates the complexities of structures and of
the underlying mental model which drive systems behaviour (Dori, 2011; Monat and Gannon, 2015; Strahler,
1980), it is ideally suited to model the proposed integrative floatovoltaic framework or mental model of Fig-
ure 3.14 in a computer model expression orchestrated by scientific theoretical methodology. Consistent with
the appropriate research methodology and theoretical and philosophical perspectives underlying the study,
discrete event simulation modelling serves to aid the research methodology of an environmentally-centred
knowledge economy. It can offer an effective dynamic analytical and decision-support tool to represent the
real world quantitatively and to simulate systems dynamics on an event-by-event basis (Law, 2014).
Revisiting the four essential steps to define a computer model expressed in terms of complex systems
thinking research according to Figure 3.6, the (yellow) practice-based decision-support element in Fig-
ure 3.6 is omitted for the moment to reduce the development process to the three core essential modelling
steps that focus on the progress of the model. As such, Figure 3.17a depicts the reality of interest, repre-
sents the physical floating PV system for which data are being obtained, and the conceptual model of reality
already defined by the ISA-framework model depicted in Figure 3.14.
Technological
impacts
Economical
impacts
Environmental
impacts
Integrated
Analytical Framework
for floatovoltaic
Sustainability Profiling
techno-econo
responses techno-enviro
responses
econo-enviro
responses
(a) (b)
Figure 3.17: Model implementation process as (a) studying the reality of interest, defining a conceptual systems
model as computer logic, implementing a computerised model (source: Schlesinger (1979), page 4), and as (b)
founded on the conceptual model framework for sustainability (source: author).
Note that from a computer programming perspective, the structured floatovoltaic sustainability assess-
ment framework of Figure 3.14 can be translated into a more practical systems-programme-friendly form in
Figure 3.18b. In this translation, the conceptual computer modelling framework depicted by Figure 3.17b es-
109
tablishes the conceptual model as a programme-logic abstraction towards creating a computerised physical
model (CAS-model) for the collection and processing of experimental data.
According to the process of Figure 3.17a, the conceptual model of Figure 3.17b acts as software pro-
gramme logic for the implementation of the software of the computerised model as a computer programme
code. Through a geo-engineering model-based design process, the computerised model in Figure 3.17a
is to be established as a parametric mathematical model, driven by constitutive mathematical equations,
together with modelling data required to describe the "reality of interest” in terms of mathematically relevant
physics and algorithms (Thacker et al., 2004). From an analytical science perspective, the system-translated
conceptual ISA-framework model of Figure 3.17b inherently represents a simplification of the floatovoltaic
ecosystem. As such, it means the geographic researcher’s perception of the reality of a floating PV energy
ecosystem sustainability. In terms of the philosophical goals of the research toward conceptual modelling
and conceptual validation, the computerised model (CAS-model) should be designed to digitally establish
the formalisation of a virtual floatovoltaic power plant in terms of the ecosystem’s elements, processes,
components and functional interrelationships between the elements and process objects.
Strategically, this section shows how the synthesis of knowledge pertaining to reality and practice to de-
velop environmental modelling software to build a suitable process-based floating PV sustainability-profiling
capacity. The following section provides deeper theoretical foundations for defining the sustainability as-
sessment framework as a construct of a computer systems model.
3.6.2. Model construct of a sustainability assessment framework system
The modelling framework for a floating PV system needs a new pattern of thought towards interactive
systems thinking, dynamic thinking, causal thinking, stock-and-flow thinking and thinking endogenously to
close the knowledge gaps around planning-support systems for implementing floating PV systems. This
section details the thesis postulate in terms of defining a model construct of a sustainability assessment
framework that meets the requirements of the systems dynamics pattern of thought mentioned above.
While data analysis modelling helps to analyse the data system hierarchy visually, the data analysis
model provides a conceptual representation of the data objects, associations between the data objects, and
the data-driven rules. The implementation of the data modelling technique, through the entity-relationship
modelling schema in the conceptual framework depiction of Figure 3.18a, helps to clarify the modelling task
in terms of the degrees of modelling the institutionalised system’s abstraction of a floating PV ecosystem.
According to the syndicated 3E framework definition of sustainability (Figure 3.18a), the systemic interven-
tion in floating solar energy systems modelling requires that all three sustainability dimensions are assessed
together with their respective feedback responses in a mutualistic symbiotic assessment arrangement.
The conceptual framework translation of Figure 3.18a illustrates overlapping segments of individual
operational elements, an abstract representation of implementing mutualistic and synergetic technical, en-
vironmental, and economic system interactions in a functioning floating PV installation project ecosystem.
Based on the hierarchical systems thinking modelling concept proposed by (Strahler, 1980), Figure 3.18b
depicts the hierarchical meta-model for realising the floating solar PV synthesis model. This hierarchy
integrates the conceptual sustainability framework as computer programme logic in a physical computer
synthesis model to establish a data model suitable for experimental data capturing.
In model-based reasoning, the development of an information systems model based on the systems
thinking approach is generally organised into a hierarchy of levels. Applying systems theory to translate
the postulated reference framework as a programme-logic architecture and an electronic computer model
requires the definition of a hierarchical meta-modelling strategy (Strahler, 1980). Such a meta-modelling
strategy would rest on applying systems thinking skills in a hierarchical modelling pyramid, as depicted in
Appendix G (Orgill et al., 2019).
110
Technological
impacts
Economical
impacts
Environmental
impacts
Integrated
Analytical Framework
for floatovoltaic
Sustainability Profiling
techno-econo
responses techno-enviro
responses
econo-enviro
responses
(a) Conceptual framework ontology
Dynamic Experimental Data
Physical System Synthesis Model
Logical System Model
Conceptual Model
(b) Meta-modelling architectural hierarchy
Figure 3.18: Proposed new (a) conceptual framework ontology, and (b) meta-modelling architectural hierarchy, for
empirical characterisation of floating PV systems (source: author).
To establish the meta-model according to the modelling hierarchy defined by Figure 3.18b, an architec-
tural meta-model representing an enterprise guides the computer model design towards the realisation of
a floating PV ecosystem enterprise. Towards the definition of a meta-model suitable for meta-analysis in
an agile programme planning process, enterprise information technology architectures, such as The Open
Group Architecture Framework (TOGAF) standard (Gerber et al., 2010), offer a generic guiding structure for
a hierarchical analytical process, broken down and generally depicted in the layers defined in Figure 3.18b.
From the architectural perspective of an enterprise, the meta-model’s development classification layers,
depicted by Figure 3.18b, provide the definitions of all the building blocks of the system, as well as the type,
in a layered meta-analytical modelling hierarchy, according to a tiered systems model of development activity
(Strahler, 1980). Comprising different levels on the hierarchy, Figure 3.18b models the classification of the
data modelling development layers in terms of the conceptual model, the logical systems model, the physical
systems model, and the dynamic experimental data-type model, each developmental layer with its specific
modelling abstraction purpose. In the conceptual framework schema of Figure 3.18a, the conceptual model
typically represents a theoretical framework model at the top of the hierarchy in Figure 3.18b. While the
general function of the conceptual model is to help understand, scope and define ecosystem concepts
and rules, the conceptual model aims to establish the elements of the object entities of the system and to
determine the attributes and associative relationships of these object entities (Schlesinger, 1979; Xia et al.,
2017). When designing an information system, conceptual modelling describes the activity that describes
and elicits the strategic knowledge or reality the system needs to represent. It thus focuses on how to
define the floating PV ecosystem as a system of systems, with operations that are cognitively abstracted to
describe the real-world technology ecosystem (refer to Prinsloo (2020) in the context of Kotov (1997)).
Therefore, at the top layer of the modelling hierarchy of Figure 3.18b, the conceptual model layer (con-
ceptual framework layer) offers an ecosystem or an institution on an organisational level with comprehen-
sive coverage of the operational ecosystem concepts, described by the critical functional system’s tenants
of the physical entity (real-world floating PV ecosystem), the system’s entity attributes (entity properties or
characteristics), the system’s entity relationships (associations with or dependencies on other entities), and
the system’s entity cardinality (one to many relationship occurrences with other entities) (Strahler, 1980;
Thacker et al., 2004). Moving towards the end goal of establishing the theoretical framework in a physical
data or computer model, the logical system model in Figure 3.18b defines how the study should implement
111
the floating PV system. Adding further information and functional detail to the conceptual model’s functional
object entities helps to explain the systemic topology as a technical map of ontological structure and rules
as a foundation for the physical computer model and objects. As the base for the model of the physical sys-
tem, the logical systems model defines the object components and the structure of the floating PV system,
together with the set of relationships among the FPV system’s object entities. With the logical abstraction
modelling layer of the system in Figure 3.18b which helps to generate the modelling schema, the logical
system model is used to implement and encode the structure of the data object elements in the physical
system’s synthesis model. As the implementation of the analytical model is geared to specific applications,
the synthesis model describes how the system is implemented using a particular system and programming
format and how to integrate the logical system objects based on the model’s scope.
By translating the framework of Figure 3.19a into a logical computer model, as depicted by Figure 3.19b,
the system can define the digital system traits that represent aspects of coordination, unification, diversifi-
cation, replication of the generic systems model (in a type of mixed-effects modelling approach). The goal
is to deliver a more integrated target architecture, or systems construct consisting of component building
block types. The meta-model describes the component types in terms of component functions, properties,
and relationships to structure architectural thinking in an analytical architectural enterprise or an analytical
ecosystem domain that aims to harmonise relationships and inter-systemic cooperation.
Technological
impacts
Economical
impacts
Environmental
impacts
Integrated
Analytical Framework
for floatovoltaic
Sustainability Profiling
techno-econo
responses techno-enviro
responses
econo-enviro
responses
(a) Conceptual model for FPV ecosystem
Energy
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
(b) Logical model for FPV ecosystem
Figure 3.19: Proposed new (a) conceptual framework model ontology, and (b) logical programmed architectural
model topology, for the empirical characterisation of floating PV ecosystems (source: author).
In the modular design and systemic integration of the floating PV synthesis model in Figure 3.19b, the
synergetic theory foundations established by Dori (2011), Forrester (1958), and Habermas (1981) provide
communicative network action theory as a means to realise a system dynamics permutation of a system
in terms of interacting software objects. The object-process methodology for this 3E data fusion model
can apply communicative network theory to define an appropriate systemic structure that best enables the
simulation and optimisation of an integrated floating PV system. It can advance the principles from sys-
tems thinking to system dynamics thinking in modelling through numerical techniques. By clustering the
model objects of the three knowledge domain sectors (3E), as shown in the stock-flow topology of Fig-
ure 3.19b, the CAS-model of this thesis can define an integrated analytical computer simulation model for
floatovoltaic technology. After that, an object-oriented approach integrates the energy, environmental and
economic ecosystem objects in the framework of Figure 3.19a into a dynamic system-of-systems configu-
ration through the integrated closed-loop, object-process methodology, as depicted in Figure 3.19b.
112
As a digital template or abstract model of a real-world floating PV system, the system dynamics meta-
model of Figure 3.20a enables the meta-modelling systems architect to define the properties, rules, con-
straints, and relationships of the floating PV synthesis systems construct suitable for experimenting with
analytical assessments. As such, the (floating PV) meta-model can establish an overall model as a mod-
ular solution composed of elements, which are themselves models, to establish a system consisting of
sub-models to perform several specific processing functions in a broader ecosystem (Nisbet et al., 2018).
Energy
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
(a) Logical systems model for FPV ecosystem (b) Physical systems metrics for FPV model
Figure 3.20: Systems thinking applied to floating solar PV modelling, integrating functional intelligence components
into a synergetic systems-oriented digital synthesis model (source: author).
While Figure 3.20a deals with the empirical and analytical aspects of the study, the decision-support pro-
filing exercise requires the survey to conceptualise a means for quantitatively charting a decision-support
dashboard. To this end, the project sustainability profile mapping display in Figure 3.20b involves multiple
parameters/criteria to support analytical decision-making and to guide the decision-making process regard-
ing the specifics of the decision goals. In this thesis, the profile mapping display device, as depicted by Fig-
ure 3.20b, is designed to visually articulate the techno-economic, enviro-economic and techno-ecological
viability determinants in a profiling method that involves multiple parameters/criteria.
A comparative multiscale analytical assessment method must provide statistical consistency on the
profile mapping scale to enable decision support through the quantitative decision-support charting and
-dashboard means of Figure 3.20b. In this thesis, the AHP technique of Saaty (2008) is considered an
ideal means to statistically weigh and compare indicators obtained from the modelling endeavours, as
depicted by Figure 3.20. The AHP technique offers an integrated data-processing approach built around
the statistical comparison of a proposed decision-making support model in the decision-profile mapping
display of Figure 3.20b. As such, the spider diagram display (also known as a radar chart, web chart,
polar chart or Kiviat diagram) supports analytical decision-making and guides the decision-making process
regarding the specifics of the project’s decision goals.
This section describes implementing the sustainability assessment framework in a quantitative com-
puter simulation model expression. The following section details the methodological implementation of a
performance-profiling solution for floating PV systems by embedding the computer simulation model and the
decision-support system as layers in a geo-informatics research instrument and decision-support system.
113
3.6.3. Implementation flowcharts for analytical model and decision model
Keeping the philosophical level goals in the conceptual illustration of Figure 3.3 in mind, this section nav-
igates the reader through the final strategic part of implementing the research model in the tool develop-
ment process of the geographical application. It presents a narrative overview of the implementation steps
towards establishing the digital floatovoltaic analytical technique and computer synthesis model as foun-
dational parts of the toolset and decision-support system in the proposed geoinformatics analysis. It also
presents the flowchart diagrams for the step-by-step toolset implementation of the geoinformatics systems
model detailed in the next chapter.
To conceptualise the geoinformatics toolset as a virtual instrument suitable for quantitative data collec-
tion in research experimentation, the conceptual cascaded system dynamics representation of Figure 3.21
integrates the physical systemic components into the geo-information system as GIS layers. The base
layer of the geo-information system of Figure 3.21 includes the system dynamics simulation integration
of the Energy-, Environmental- and Economical- simulation models. In this analytical configuration, the
sustainability framework logic underpins the system dynamics simulation model as an archetype for real-
time floating PV energy system assessments in an analysis-by-synthesis methodology. The analytical GIS
layer further incorporates the composite intelligence of the floating PV ecosystem characterisation model
in a dynamic modelling ontology. The next layer in developing the research instrument model, as depicted
in Figure 3.21, integrates the metrics obtained from the analytical floating PV sustainability assessment
model into a heuristic-type multi-criteria decision-support display model. As such, the floatovoltaic model
in Figure 3.21 reflects the model design philosophy described in Section 3.2.3 and the model design goals
depicted by Figure 3.3 in a combined descriptive, predictive and prescriptive analytical model.
Country-specific geospatial, environmental, legislative, economic, technical context
△Floatovoltaic Planning Decision Indicator Model Layer
⊕Floatovoltaic Ecosystem Synthesis Model Layer
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
Figure 3.21: Proposed coherent system dynamics simulation model for floating PV systems, fundamentally embed-
ding decision support into a parametric synthesis model ontology (source: author).
The conceptual meta-design model of the floating PV time series analysis model in Figure 3.21 repre-
sents the collective multiscale modelling of the integrated modelling and simulation of a floating PV system
as a floating PV system simulation toolkit called the FPV-GIS toolset. True to the philosophical goals for a
systemic theoretical sustainability modelling concept in Figure 3.3, the theoretical conceptualisation of Fig-
ure 3.14 in a coherent system dynamics simulation model of Figure 3.21 can deliver a computer model that
integrates behaviour synthesis, performance evaluation, and impact assessment all in a single consolidated
114
sustainability modelling synthesis. The next chapter has the task of physically establishing the systemic
components and governing dynamics of the integrated geoinformatics model as a computer software pro-
gramme. With the analytical and decision-support models embedded as layers in the digital geo-informatics
platform layers, and the sustainability indicators consolidated in a profile mapping representation, the inte-
grated FPV-GIS toolset model enables the user to study the predicted performances and benefits of planned
FPV systems according to the sustainability framework hypothesis.
With the conceptual tridentate ISA-framework establishing the pragmatic philosophical underpinning of
the thesis, and the CAS-model construct providing a computerised model representation of the environ-
mental profiling ISA-framework, the final step is to formulate the workflow for the implementation of the con-
ceptual geo-informatics toolset of Figure 3.21 in terms of flowcharts. As such, the flowcharts of Figure 3.22
direct the workflow required to integrate all the model components systemically. While the flowcharts offer a
step-by-step implementation of the experimental GIS mode, it also shows how systems thinking is applied
to floating solar PV modelling in terms of: (a) modelling the systems-oriented digital synthesis model; and
(b) modelling the decision-support algorithm based on the analytical hierarchical project goals.
Research Objective 1: Conceptualise and de-
sign a floating PV system modelling framework
and systems structure (Sections 3.2 to 3.7)
Research Objective 2a: Define a dynamic
floating PV systems structure, components,
interactions and parameters (Section 4.2)
Research Objective 2b: Characterise float-
ing PV model object components using a
model-based design process (Section 4.3)
Research Objective 3a: Characterise sustain-
ability decision-support indicators (Section 4.4.2)
Research Objective 3b: Characterise
decision-support model using AHP
decision hierarchy (Section 4.4.3)
Research Aim: Embed analytical and decision
models as geoinformatic system layers for
data collection and processing (Section 4.5)
Research Experiments: Design an
experimental case study for frame-
work theory evaluation (Section 4.6)
(Section 4.4)
Determine objective goals
of project (sustainability)
expertsystem
Set up the decision criteria
Construct the decision hierarchy
Make pairwise metric comparisons
Calculate indicator weights
update
model
Pass consistency
inspection?
Aggregate indicator weights
Apply indicator weights to
analytical model metrics
no
yes
(a) (b)
Figure 3.22: Implementation design process, showing flowcharts for implementation of (a) computer modelling and
simulation model and method, with (b) expanded view of the development of a decision-support algorithm model
based on analytical hierarchical process goals (source: author).
Assisting with the implementation of Figure 3.21 as a broader-spectrum floatovoltaic sustainability as-
sessment model, according to the conceptual, theoretical analysis and modelling framework of Figure 3.18,
the modelling flowcharts in Figure 3.22 implement the sustainability assessment framework as a driving-
force engine or ontological blueprint at the heart of the analytical system in Figure 3.21. In research method-
ology, the research aim defined in Section 1.3.4 sets a clear purpose for answering the research questions
through the conceptualisation and engineering of a quantitative computer and simulation systems model
or floatovoltaic digital twin model (FPV-geomodel). Therefore, within the context of the research aim and
115
objectives described in Chapter 1, the implementation of the system in Figure 3.21 in Chapter 4 achieves
the complete set of research aim and objectives for answering the research questions.
While the methodological strategy of this study can be classified as a computerised quantitative simu-
lation research method, this section presents the conceptual geo-informatics and data science techniques
to support theory-driven thinking and statistical modelling through computer software programming codes.
These conceptualisations enable the implementation of the integrated geomatic toolset and model with
practical functionality to serve as a research tool or research instrument for application in data collection
and processing according to the experimental requirements of this thesis.
3.7. Theoretical Substantiation of the Research Methodology
This section familiarises the reader with the theoretical underpinning that supports the implementation of
the methodological framework as an expression of a quantitative computer simulation model. The discus-
sion starts by substantiating the theoretical and practical value of the computer modelling and simulation
methodology engaged by the FPV-GIS toolset and moves towards the theoretical foundations around the
model-based design and system dynamics modelling methods. Finally, it ends with the theoretical under-
pinnings of the relevant analytical decision-support approaches in geoinformation.
In the global sustainable development context (Lee et al., 2022; Maka, 2022), the geographical data sci-
ence philosophy offers methodological options for knowledge integration through mathematical and com-
puter science models that support the geographical science mission as a vital part of the geographical
science engagement (Batty, 2011; NSF, 2011). Such organisational management theories are often repre-
sented as formal axioms to define theoretical frameworks and model ontologies in data analytics solutions
as a key part of real-life problem-solving (Ma, 2019; MuSigma, 2016; Vazquez, 2018). The methodological
approach details the proposed descriptive quantitative analytical process in a computer synthesis algorithm
(Escobedo et al., 2017; Kim et al., 2016a), giving special consideration to framework formulation in support
of experimental modelling for floating PV systems.
The solution proposed by this thesis is based upon empirical analysis-by-synthesis modelling towards
quantitative data collection, quantitative data processing, and quantitative data analysis (NSF, 2006). In
this context, computer modelling and synthesis entail the creation of a computer analogy of a technology
system to predict certain behaviours in estimating performances or impacts or to help understand particular
phenomena or impact effects in a simplified, versatile representation (Khaiter and Erechtchoukova, 2007).
Built on laws of physics, digital mathematical modelling offers the theoretical possibility to engage Indus-
try 4.0 type 4IR modelling in the design of a floating PV system as a virtual powerplant system capable
of delivering a specific digital synthesis technique for understanding the dynamics of a technology system
(Goossens, 2021; Martinelli et al., 2020). This technique is further suitable for predicting future outcomes
of an installation within the sphere of operation of the floating PV ecosystem. Because the characterisa-
tion technique of the floating PV system proposed by this thesis uses a computational model to describe
systems behaviour in a particular experimental design and setup configuration, it can ensure contextual
intelligence in location-based data architectures (Graser and Olaya, 2015; Jakhrani et al., 2014).
The modelling and simulation of dynamic systems, such as the (floating) photovoltaic information ecosys-
tem, stems from the roots of the model-based systems engineering framework (Estefan, 2008; Macedo
et al., 2021). The technique engages arithmetic dynamics as it amalgamates concepts from two areas of
mathematics, namely dynamic systems and number theory, to model the discrete dynamics of an integrated
ecosystem in studying the iteration of operational maps in the complex space-time plane (Silverman, 2007).
As such, the modelling aspect of multimethod modelling in a system dynamics modelling technique refers
to the (geo-aware) engineering of a mathematical representation of a physical system and its subsystem in-
116
teractions as a virtual multi-subsystems model in a system-of-systems configuration (Anylogic, 2020; Kotov,
1997; Suslov and Katalevsky, 2019). Secondly, the simulation aspect of the technique refers to computer
synthesis procedures for running the descriptive equation formulations and relationships of the integrated
model (temporal mathematical solver aspect) (Coyle, 1997). Thirdly, the floatovoltaics enterprise modelling
and knowledge representation efforts should combine the reference framework principles in reusable core
ecosystem ontologies for integration into a digital twin model (Barachini and Stary, 2022; Madni et al., 2019).
From a cyber-physical energy system of systems perspective (Goossens, 2017; Jainandunsing, 2019), the
usefulness or quality values of such a numerically integrated (floatovoltaic solar) system dynamics model
can be measured in terms of the model’s ability to arithmetically encapsulate the physical governing fea-
tures of a physical floatovoltaic (eco)system in the associated context of the scenario-planning problem of
the decision analytics project. This thinking makes the system dynamics thinking pattern ideal in terms of
integrating the technical, economic and environmental performance aspects of the development projects in
a causal system-thinking structure (refer to Figure 3.13 in the context of the theoretical feedback systems
frameworks set by Dori (2011)).
In modelling the knowledge and network dynamics in the three-sided floating PV model topology, the
conceptualised analytical model expression, or integrated CAS-model for the ISA-framework conceptuali-
sation, uniquely combines generic geospatial modelling in digital twinning techniques (Clark and Kulkarni,
2021; Wainwright and Mulligan, 2013). By engaging model-based engineering/environmental design (Do-
natelli, 2008; Estefan, 2008) and complex system dynamics modelling approaches (Forrester, 1958; O’Brien
and O’Brien, 2014), the thesis can affect the characterisation of a floatovoltaic ecosystem technology as an
expression of a computer model. By exploring the prospects for modelling systemic object interdependence
in Figure 3.16, the thesis extends the philosophical idea to fill the observed knowledge gaps in geo-aware
analytical floatovoltaic evaluations. Being an ideal administrative strategy or governing mechanism to drive
the model workflow in Figure 3.16, the thesis’s proposed contextually sensitive conceptualisation of the
sustainability evaluation framework of Figure 3.18 serves as a blueprint for an analytical computer model.
As a method for model-driven engineering (Bernardo et al., 2012), Figure 3.16 thus establishes the basis
for engaging an object-process methodology process (Dori, 2011; Kohen and Dori, 2021), together with
a system dynamics methodology (Forrester, 1958; Stanton et al., 2019), as a combined methodology for
translating the proposed analytical framework into a computer model expression. The object-process mod-
elling technique generally offers an ISO-accredited programming paradigm for designing systems models
and dynamically capturing knowledge within ecosystem models in an object-oriented model-based systems
engineering environment (Dori, 2011; Graser and Olaya, 2015). As an intuitive holistic systems design
method for embedded implementation into Figure 3.16, the object-oriented system dynamics processing
techniques of Dori (2011) and Forrester (1958) are ideally suited to formally specify the structure, func-
tion and behaviour of natural and artificial ecosystem model stereotypes as data-processing computer-
programme-building blocks from a macroscopic perspective (As opposed to a microscopic perspective,
functional discipline object blocks run individually or sequentially in isolation; refer to Figure 3.9).
Systems thinkers generally promote the systems-thinking perspective as a beneficial way to expand
systemic behavioural perspectives (Midgley et al., 2014). In this context, the proposed systems think-
ing approach to floating PV systems modelling reminds us, to a certain degree, of relationship-seeking in
people-orientated Soft Systems Methodology (Burge, 2015), or associative agency forming in machine-
oriented Soft System Dynamics Methodology (Rodriguez-Ulloa and Paucar-Caceres, 2005). The design
strategy advances from systems thinking to systems dynamics to take the computational model depicted in
Figure 3.16 as a synergetic floating PV systems design one step further. This approach enables the thesis
to create a feedback model that better supports closed-loop architectures and complex decision-making
(Borshchev and Grigoryev, 2021). System dynamics models in sustainable development missions would
allow developers to model essential inter-system linkages and help them to make better decisions when
117
confronted with complex dynamic systems. In this context, the application and relevance of Figure 3.19b
offer systems dynamics in a systems thinking approach that is quantitative by nature and uses engineering
equations and computer models to investigate complex systemic behaviours (Forrester, 1958). The system
dynamics methodology extends the intuitive and holistic systems-design method to facilitate enterprise ar-
chitectural sequencing, thus making it capable of delivering attractive, dynamic temporal features suitable
for floatovoltaic ecosystem modelling. In theoretical terms, the synergistic system dynamics thinking model
towards characterising the floating PV system in Figure 3.19b is based on communicative coordination and
a macro-theory of the systemic integration of ecosystem elements through a network mechanism topology
(Habermas, 1981).
With the integrated FPV-GIS toolset as a dynamic system capable of providing macroscopic and micro-
scopic perspectives on the adoption of FPV technology, the properties of the departmental elements in the
proposed FPV analytical system of Figure 3.20 can affect system-dynamics-type cause-and-effect analyses
in sustainability assessment narratives for floating solar technology project valuations. With system dynam-
ics capability of accommodating multi-physics simulations in multi-systemic environments, system dynamic
modelling is founded on the system dynamics concepts of stocks and flows, with stocks describing quan-
titative physical entities and flows describing how stocks change in time (Borshchev and Grigoryev, 2021).
In this way, the category of systems dynamics considers feedforward/feedback interconnections and influ-
ences and acknowledges mutual interactions between constituent object parts as reciprocal interactional
processes (Wiener, 1949). As such, dynamic cybernetics systems mainly operate as self-adaptable mech-
anisms, similar to cellular elements in a body, resembling the characteristics of self-regulating homeostatic
mechanisms (Ashby, 1961; Molenaar, 2001). Towards the age of digital transformation, Wiener (1954) de-
fined the systemic organisational attributes of interrelatedness and interconnectedness as cybernetics-type
systemic qualities or properties whereby the study can model the sequence of events over time with inter-
connected feedback loops. These ideas were advanced further by Ashby (1961) when computer science
and data science research started to exploit cybernetics in a complex adaptive system as foundations of
purposeful behaviour in decision-information systems and robotics automation. Ashby’s work pioneered
cybernetics and systems theory as a theory of organisation in which the introduction of reinforcing and
feedback loops addresses the problem of system dependability. This concept inspired the essential idea
behind the approach of this thesis toward a closed-loop environment and a modelling-based requirement
for a floating PV software system (Prinsloo, 2020).
Finally, a statistical technique is required to weigh or balance the set of multi-indicators to provide a fair
heuristic comparison between the sustainability indicators toward common project-decision goals (Prinsloo,
2020). Since the decision metrics output by the information system model of Figure 3.21 are reported in
different units of measure and scales (watt, m2, R, kg, etc.), a statistical technique is further required to
normalise the set of multi-indicators (as probability ratio variables) as a means to provide a fair compari-
son between the sustainability indicator profile dimensions in a visual floatovoltaic decision support system
display (Prinsloo, 2020). Theoretically, decision-making supports using modern integrated data-processing
approaches to statistically weigh and compare indicators obtained from the modelling endeavours, as pre-
sented in Figure 3.20b. In the context of model-based reasoning, probabilistic goal-driven algorithms, such
as analytical hierarchical network decision modelling (AHP/ANP) (Saaty, 2008), are an ideal technique for
statistically comparing multi-indicators to provide floatovoltaic decision support. The computer technique
for order of preference by similarity to ideal solution (TOPSIS) (Hwang and Yoon, 1981) offers similar multi-
criteria decision-making benefits, while other options include GIS-MCDA algorithms in multi-criteria decision
analysis (MCDA), the weighted sum method (WSM), weighted-product methodology (WPM), multi-attribute
theory (MAUT), elimination and choice translation reality technique (ELECTRE), compromise programming
(CP), the weighted linear combination method (WLC) (Pohekar and Ramachandran, 2004), or the decision-
making trial and evaluation laboratory (DEMATEL) method (Si et al., 2018). Another flexible and robust
118
decision-making tool and dashboard diagram suitable for sustainability assessments and appraisals has
been developed by ARUP (2019). This empirical diagram is an extension of the spider diagram and was
originally developed to plot quality-of-life counts in sustainability projects on a global scale. Coined the
Sustainable Project Appraisal Routine (SPeAR), this geo-environmental software tool presents a scientific
diagram that operates as a flexible evaluation- and decision-support tool geared to ensuring appropriate
systemic performances in terms of project goals.
The AHP technique was developed by Saaty (2000, 2008) as a quick and simple method for making
decisions in prioritising projects and in selecting options. As a normative modelling framework for decision
weighing, AHP enables the designer to capture the outcome goals of strategic projects as a set of weighted
criteria to score project performances. While the AHP technique is an effective way to deal with qualitative
decision areas in operations management, the technique is often applied to problems in operations man-
agement as a structured technique for organising and analysing complex decisions (Partovi et al., 1990).
As mentioned under Figure 3.20b, this thesis selected the AHP technique of Saaty (2008) as a statisti-
cal means to process indicators obtained from the modelling endeavours, as presented in Figure 3.20a.
As a probabilistic means of weighing and comparing FPV project sustainability indicators in the decision-
profile mapping display of Figure 3.20b, it offers valuable opportunities to plot WELF-related parameters in
a vector-borne sustainability diagram (Gamarra and Ronk, 2019; Sarkodie and Owusu, 2020).
This section concludes the discussion on the theoretical substantiation of the research methodology
and processes to model the proposed framework into a quantitative computer simulation model template
referred to as the FPV-GIS toolset. It generally shows how a robust theoretical underpinning in scientific
literature supports the proposed research methodology. The discussion further supports the proposal that
systems thinking, as an integrative paradigm, promotes broader analytical approaches as a beneficial way to
expand planning perspectives or improve decision-making skills around aspects such as project vulnerability
and goal setting in the modelling and assessment of floating PV systems.
3.8. Summary
In working towards answering Research Question 1, this chapter deals with Research Objective 1 as it
examines the design of the research process and the research methodology towards the realisation of the
research aim through this objective. It introduces the philosophical narrative around the goals of Research
Objective 1, and works towards establishing a sustainability reference framework for assessing the sus-
tainability qualities of future planned floating PV installations. It further dictates the design of the analytical
floating PV model elements and the decision-support methodology of the computer synthesis model, pro-
viding the flowcharts for implementing the analytical and decision-support models in the next chapter. The
chapter closes by theoretically substantiating the research methodology and implementing the computer
model from a geographical sustainable developmental supporting perspective.
This next chapter describes the step-by-step modelling and simulation of the floating solar photovoltaic
system by elaborating on the execution aspects of the research methodology to empirically collect data
in terms of the Research Aim and Research Objectives 2 and 3. Furthermore, it works toward the tac-
tical implementation and experimental evaluation model for a proposed experimental digital floating solar
ecosystem in terms of Research Objective 2. As such, the next chapter dictates the model-based design
steps towards implementing the subsystem elements of the FPV model that constitute the floating solar-,
energy-, environmental-, and economic- subsystem objects and narrates the decision-support methodology
of the computer synthesis model. This approach enables the thesis to further deal with integrating the object
layer of the decision model and the indicator-processing functionality of this GIS layer, as final preparation
for scientific experimentation in Chapter 5, in terms of evaluating Research Objectives 1 to 3.
119
4. Parametric Modelling Methodology
4.1. Introduction
This chapter deals mainly with implementing the research methodology to realise the research aim through
Research Objectives 2 and 3. This chapter focuses on model-building and design research efforts from the
philosophical theory-building action narrated in the previous chapter. By actively engaging in geoscientific
model development, it focuses on implementing the workflow models described in Chapter 3 towards estab-
lishing the research instrument as integrated Floating PhotoVoltaic Geographical Information System toolset
(FPV-GIS toolset). The primary function of the FPV-GIS model is the experimental real-time characterisa-
tion of the space-time geodynamical responses of floatovoltaic technology systems. The chapter presents
the proposed theoretical framework conceptualised in Chapter 3 as execution method logic in an analysis-
by-synthesis simulation model for quantitative data collection and analysis required to answer the research
questions. It navigates the reader through implementing the simulation model, indicator selection, and de-
cision support to establish a computer synthesis-based quantitative research model and instrument. While
the previous chapter focused on Research Objective 1 to help answer the first research question around
developing a novel integrated analytical framework for the theoretical assessment of floating PV projects,
this chapter focuses on Research Objectives 2 and 3 as a means to help answer the second research ques-
tion. It focuses on developing and realising an analytical computer simulation model for the South African
agricultural context. It establishes a contextually-geo-sensitive sustainable development assessment and
decision-supporting (eco)system that integrates the operational dynamics of floatovoltaic technology into a
systems-thinking topology based on the integrated theoretical modelling framework conceptualised in the
previous chapter.
As such, Chapter 4 describes the step-by-step modelling and simulation of the floating solar photovoltaic
system called the FPV-GIS toolset in terms of the parameter descriptions and notation definitions listed in
Table N.1 (Appendix N). Section 4.2 details the system dynamics modelling and computer-synthesised
methodology in an experimental interactive geoinformatics toolset to emulate an experimental digital float-
ing solar ecosystem. Section 4.3 subsequently dictates the model-based design of the subsystem elements
of the FPV model constituting the co-simulated floating solar, energy, environmental, and economic compo-
nents or subsystemic objects and narrates the decision-support methodology of the computer-synthesised
model. This floatovoltaic model establishes the analytical technique to empirically collect data in terms of
the Research Aim and Research Objectives 1 and 2. Towards Research Objective 3, Section 4.4 deals
with the integration of the object layer of the decision model and the indicator-processing functionality of
this GIS layer in preparation for scientific experimentation. Section 4.5 summarises the integrated research
instrument as a functional geoinformatics toolset for studying a floating PV system, highlighting its GIS
functionality as a research instrument for application in scenario-driven case-study experiments according
to the research objectives. Section 4.6 introduces the experimental procedures for evaluating the proposed
model in scenario-based empirical floating solar synthesised analyses in hypothetical case-study narratives.
Section 4.7 details the ethical considerations associated with the research project and experiments.
120
4.2. Research Model: Systems-level Design and Implementation
With the research aim set on the re-conceptualisation of the floatovoltaic technology enterprise, the focus
is on theoretically implementing appropriate floating model parameters suitable for theoretical performance
modelling and profiling in a proposed system dynamics model for floating PV systems. In a geo-sensitive
information systems context, this endeavour essentially comprises building sustainability-assessment in-
telligence onto a GIS decision-support platform. This section describes the process of systematising the
floating PV analytical ecosystem according to the framework template introduced in the previous chapter.
It addresses the systems-level specifics of the proposed floating PV modelling expression according to the
first part of Research Objective 2. As such, the thesis moves toward dictating the systems-based design-
element specifics in terms of the constituent component interactions of the floatovoltaic systems model, as
defined in the implementation process of Figure 3.22a.
Towards establishing the design framework for a virtual floating PV power plant as a software-configurable
digital twin model for the installation of a generic real-world floatovoltaic system, the algebraic objects and
ontology of the computer model define the individual energy, environmental and economic domain elements
as computer software objects that interact among one another. In this context, Section 4.2.1 performs a
causal analysis procedure within the sustainability assessment model construct. Section 4.2.2 uses the
object-process diagram to guide the floating PV systems model as a set of interacting software objects to
implement the FPV model as a system dynamics technique. Section 4.2.3 integrates the floatovoltaic sys-
tems model and subsystems objects, while Section 4.5.1 details the specifics around the implementation of
the Python software simulation model.
4.2.1. Systems-based causal analysis in an object-process methodology
This section narrates the customisation of the systems-thinking methodology to suit the floatovoltaic en-
terprise and ecosystem. Using basic object-process methodology principles, a cause-and-effect diagram
tool enables the systems designer to unpack the ontological systemic properties. This approach allows a
designer to logically organise possible causes in the form of a causal-type diagram used in an analytical
application design and to identify potential factors causing an overall effect in a representation of a digital
FPV systems model.
Starting with the conceptual model depicted in the rationalisation candidate of the computer model in
Figure 3.19, the research investigation can strategically apply the Object-Process Methodology (OPM) ap-
proach to geographically model the energy, environmental, and economic model objects toward establishing
a baseline synthesised model of a floatovoltaic system. As an extension of the systems-modelling capabil-
ities of the current object-oriented methods, the OPM approach to systems-solution engineering generally
focuses on integrating function, structure and behaviour into a single unifying model (Dori, 2011). The OPM
methodology supports systems thinking by providing causal loop diagrams as a data-flow visualisation tool
that lets the systems designer visualise the relationships among the object variables over time. It also en-
ables the designer to map the systemic structure responsible for producing recurring patterns of events over
time, thus contributing to the visualisation of the software architecture during the project-planning stages.
In this way, causal analysis can assist in the framing and reframing of reference analyses in the systems-
thinking context and in adopting another framework that would include environmental and enviro-economic
systems in an integrated systemic structure. From a systems synthesis perspective, the OPM methodology
can model the floating PV ecosystem as a whole entity structure that requires interwoven or interconnected
systems-thinking network-analysis modelling, as opposed to reductionist isolationist-type analytical think-
ing, built up of individual systemic components placed in series.
121
Causal analysis within the sustainability-assessment model construct of Figure 3.19 can provide a vo-
cabulary of systemic model-object interactions. This holistic vocabulary system describes the object in-
teractions in the context of an ecosystem and the ontology (functionality, feedback and communicative
interactions) of the proposed floating PV cybernetics utility. While the proposed theoretical SIA framework
can serve as a driving force/engine or an ontological blueprint in a proposed object-process methodological
modelling framework, the causal vocabulary can drive a circular logic of the modelling framework definition
depicted in Figure 3.19. In this respect, the design-thinking processes behind geographical engineering in
the expression of any holistic systems model create exciting opportunities to interlink the transdisciplinary
knowledge domains into an integrated network topology (Pourdehnad et al., 2016). As such, an integrated
systems-thinking framework and modelling solution can perform causal loop analysis on the generic struc-
tures of the identified and formulated problems pertaining to the energy ecosystem.
Consistent with the fundamental framework theory defined in Section 3.5.4, the phenomenological
model, as defined by Roman and Hartmann (2014), can be depicted in terms of the simplified causal
diagram of Figure 4.1 to offer a scientific model that helps to describe and define the phenomenon of the
empirical relationships of the entity elements of the floating PV system concerning one other. In terms of
the ontological unpacking of the systemic properties, the systems designer can use the scientific evidence
described in Section 2.3.2 of the literature review as a means to help theoretically define the systemic
properties in terms of a computer synthesis model.
(a) Energy
Environment
Economic
Solar
Air
Land
W ater
(b)
Energy Environment
Economic
Solar
Air
Land
W ater
Figure 4.1: Causal loop diagram to aid in visualising how different object functions in the floatovoltaic ecosystem are
interrelated in a parametric synthesised model ontology, (a) structured in a linear topology, and (b) structured in a
system dynamics topology (source: author).
In the context of Figure 4.1, the conceptual framework of Figure 3.14 can further serve as a template
to decode, decipher and unpack the taxonomy and ontological abstractions in the networked analytical
topology narrative of Figure 3.16. In this way, as depicted in Figure 3.21, conceptual modelling construct
122
for exploring environmentally-linked floating PV systems behaviour can be achieved by deconstructing the
power-system dynamics to soft-programme the mechanics of the FPV-GIS into a real-world system dynam-
ics modelling research instrument.
In exploring the prospects for implementing the conceptual framework for the analysis of climate-smart
agricultural energy system proposed by Figure 3.14, the ecosystem value-chain analysis requires that the
power system dynamics be deconstructed as a multidisciplinary ensemble as part of the ontological unpack-
ing of the conceptual modelling constructs. As such, the modelling of the operational framework ontology
can benefit from the rudimentary causal diagram representation of the floating PV ecosystem shown in
Figure 3.13. The entity-relationship diagrams in Figures 3.13 and H.1 allow the floating PV system model
to inherently include the memory of previous events through inter-domain feedback loop functions. While
the integrated mind model of Figure 3.14 aims at systematically modulating the entourage effect of the
three key "E” knowledge fields (the energy, environmental, and economic object participants) in the floato-
voltaic ecosystem, the goal is to quantify the cumulative impact effects of the floatovoltaic ecosystem in an
integrated FPV characterisation and performance-assessment model.
As an entity relationship diagram, the causal charts in Figure 4.1 depict how the 3E (energy-economic-
environmental) functionaries coexist harmoniously as characteristics under the operating conditions of the
floating PV system. In that it depicts the deconstructing power system dynamics for exploring a conceptual
framework, based on the more extensive causal diagram of Figure H.1, the data-driven modelling approach
of Figure 4.1 helps to explore the relationships among the operational elements of the floatovoltaic ecosys-
tem. Moreover, in serving as a co-simulation reference network for the characterisation of floatovoltaic
technology, the reference framework in Figure 4.1 is uniquely staggering the analytical elements (in a circu-
lar network processing fashion) as part of the system-requirements analysis. Strategically, the floating PV
causality model depicted in Figure 4.1 appears to indicate that the entire closed-loop floating PV system
should instead be viewed as a type of circular economy that includes aspects of the environmental and
economic system as part of the organic growth structure for floating PV-model development.
Furthermore, Figure 4.1 underscores the value-laden nature of floating PV sustainability, in which holis-
tic cybernetics principles should incorporate an environmental system that also accommodates feedback
economics (environmental economics, conservation economics, climate economics), in addition to the typ-
ical energy economics, as part of the floating PV technical and economic performance modelling through
inter-domain feedback loops. Compared to a conventional reductionist-type framework for floating PV-
system modelling, depicted in the open-loop silo-based model of Figure 3.9, the closed-loop cause-and-
effect representation of Figure 4.1 offers an improved structure in that it provides for the various broad-
spectrum environmental and environmentally-derived technical and economic feedback effects (for exam-
ple, microclimatic impacts, land/water-surface albedo, land-water resource impacts, carbon footprint, carbon
taxation, carbon credits).
While the cascaded causal diagram of Figure 4.1 highlights the value of the environmental component
in an energy-systems type of thinking model, inherently, it also incorporates its contribution into the WELF
nexus and the value of the WELF nexus to the issue of integrated multi-criterial sustainability. As such,
Figure 4.1 underscores how the performance and impact effects produced by a floating PV system con-
stitute a dynamic process determined by multiple water-energy-land nexus factors originating from energy,
environmental and economic components, operations and interactions. These operational aspects sys-
temically interact with one another on multiple levels of causality and demonstrate intrinsic mechanisms of
system dynamic performance and impacts in temporal and spatial dimensions across diverse disciplines,
scales and metrics. Towards the development of a systemic solution in terms of the causal loop diagram
of Figure 4.1, the available dynamical system processing techniques include System Dynamics Computer
Modelling, Object Process Methodology, Interpretive Structural Modelling, Systemic Root Cause Analysis,
Main Chain Infrastructures, Systemigrams, and Stock and Flow diagrams (Monat and Gannon, 2015).
123
The causal diagram of the floating PV ecosystem in Figure 4.1 clearly shows that mono-causal ex-
planations are not able to fully account for the rich set of multi-directional interactions among the three
object elements, let alone for the interactions between the sub-elements within the elements themselves,
especially the environmental object interactions. The following section explains why the systems-thinking
viewpoint, progressing from mono-causality (linear model) to systems thinking (an integrated system dy-
namics model), is a complementary conceptual approach to floating PV modelling and characterisation.
4.2.2. Object-oriented system dynamics modelling solution
While the systemic interventions mentioned in the previous section explain how the natural environmental
system plays a crucial role in determining the project outcomes of floating PV systems models, this sec-
tion describes the multivariate analytical modelling method for assessing floating solar systems through
computer modelling and simulation methodologies. Such modelling activities for energy systems mainly
represent the computer model expression of the conceptual sustainability theory and analytical framework
proposed by the thesis.
Through the use of critical systems thinking and behavioural science principles (Dori, 2011; Jackson,
2010), the thesis recognises the opportunity to combine system dynamics and object-process methodolo-
gies to actualise the philosophical idea depicted by Figure 3.3. Such an integrated methodology combines
behaviour synthesis, performance evaluation, and impact assessment in a consolidated theoretical frame-
work for agricultural sustainability modelling synthesis for floatovoltaics. On an operational level, the object-
process methodology can play a vital role in the architectural synthesis of the entity-relationship model
shown in the simplified causal diagram of Figure 4.1 and its more detailed version depicted by Figure H.1.
The following discussion details the integrated geographical floating-PV simulation methodology and com-
puterised floatovoltaic-model abstraction. With the framework abstraction of Figure 3.18 as the governing
law, the entity relationship diagram of Figure 4.1 can guide the designer to characterise and emulate the
geographical operations of a virtual floatovoltaic power plant ecosystem down to the mathematical mod-
elling of system dynamics to predict the behaviour of floating solar PV operations. With the goal set in
a systems-design-thinking approach to agricultural floatovoltaic sustainability-performance characterisation
and impact-assessment profiling, the focus is on the architectural synthesis and integration of the framing
device of Figure 3.18 into a computer model arrangement. With an object-oriented computer software im-
plementation in mind, a detailed analysis of the functions and sub-functions of the proposed floating PV
assessment system is carried out to develop the physical structure of the analytical approach for floating
solar systems. From the perspective of a floating solar PV enterprise, the objective of the assessment
system is to represent the analytical floating PV ecosystem as a "system of systems” in a computer soft-
ware architecture capable of addressing the elements of the ecosystem as subsystems, components and
interconnections (Jackson, 2010; Kotov, 1997).
According to this understanding, the notion of systemic sustainability motivates the dynamic theoretical
characterisation of the operational ecosystem in a systemic network framing. This framing is represented
by stock and flows in a stock-flow diagram developed by the thesis in Figure H.1. This relationship diagram
highlights the requirement for new knowledge in a systemic framework ontology to capture systemic FPV
(eco)system imperatives in a system dynamics logic. As a complex interconnected system of elements
(Laszlo, 1972), the Habermas (1981) theory can be used to engage closed-loop feedback principles de-
picted by Figures 4.1 and H.1 to serve as a basis to operationalise the integrated assessment framework
in a system dynamics model in Figure 3.14 in a computer synthesis model. From a soft-systems-thinking
point of view, the proposed universal pragmatic approach aims to establish a holistic systems-level analyt-
ical expression for the floating solar operational and analytical assessment framework. Setting the goal to
create a holistic view of the enterprise, this dynamic systems-design process starts with each functional
124
operation component of the enterprise system. It defines how each operation supports the overall float-
ing PV ecosystem, as cause and effect in the systemic structure impact the system’s dynamic behaviour
according to the systems theory of Kotov (1997). Towards defining a digital twin model as a system of
systems (Barachini and Stary, 2022; Mitre, 2022), the thesis posits that the floating PV ecosystem can be
implemented in a networked system of systems architecture (refer to Figure J.1). The study can thus build
an improved characterisation of a floating PV enterprise upon the systemic architectural design process
in terms of information technology principles such as complex network theory (Clark and Kulkarni, 2021;
Klemm et al., 2012).
In this thesis, the underlying technical, environmental and economic operational elements and infras-
tructure of Figure 4.1, as well as the infrastructural relations that are interlinked among the operational
elements, are modelled using the system dynamics technique (refer (Laszlo, 1972)). In terms of the sci-
entific notation, this section describes the details of the system dynamics model for the floating PV digital
twin model in terms of networked stocks or state variables connected through inter-playing stockflows as
networking/communication variables/functions represented on a graphical model canvas (Prinsloo, 2020).
In terms of the notations mentioned above, the low-rate changes in the stock of the functional objects are
determined by the control variables or functions that define the rate of change for each flow in the system,
while on the other hand, the transforming variables or functions regulates the control variables over time.
On the execution level, an abstraction of a proposed system dynamics model is designed as a virtual
digital twin model of a generic real-world floatovoltaic ecosystem operating as a synthesised toolset in the
form of a portable desktop computer. In support of this Research Objective 2, the thesis proposed the
formulation of a Computer Analytical Simulation Model (CAS-model) to model the governing dynamics of a
floating PV system. The logical software programme of the model architecture proposed by this thesis is
depicted in the circularly cascaded framework model shown in Figure 4.2.
Energy
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro, ecologic)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
South African geospatial, environmental, legislative, economic and technical context
Figure 4.2: Proposed floatovoltaic system dynamics simulation model as an object-orientated software integration to
implement the proposed theoretical framework as a recurring empirical operation in a parametric synthesised model
ontology (source: author).
The model in Figure 4.2 enables the structured ISA-framework of Figure 3.14 to serve as a blueprint logic
in the software implementation and mathematical expressions that characterises the floating PV ecosystem
as digitalised FPV power plant. In the proposed ISA framework approach and CAS model, the recursive
nature of the floatovoltaic enterprise operates as an integrated processing ecosystem, actualised through
the circular-logic depiction of Systems Science in the form of the rudimentary system dynamics stock-flow
mapping diagram of Figure 4.1. It is a simplified version of the FPV ecosystem causal diagram shown in
Figure H.1. Both these relationship diagrams highlight the requirement for new knowledge in a systemic
framework ontology for capturing systemic imperatives for explaining how the system works and functions
125
differently from other PV systems. As such, it can help improve decisions around a PV project’s sustain-
ability, influence and sensitivity of a PV project and its associated systemic assessments. As a synergetic
object-process methodology permutation of communicative network action theory (Dori, 2011; Forrester,
1958; Habermas, 1981), the CAS model of this thesis defines an analytical computer simulation model for
floatovoltaic systems by clustering the model objects of the three knowledge-domain sectors in the rudi-
mentary causal (stock-flow) topology of Figure 4.2.
With the theoretical framework of Figure 3.14 enabling the characterisation of floating PV ecosystem re-
sponses as a 3E chain reaction (energy, environmental, economical chain reaction), the pragmatic system
dynamics processing solution proposed by Figure 4.2 involves sample-based serialisation in an iterative
sequence in a recurring cycle. In a sensible system dynamics model for FPV analysis, this means floating
PV performance and impact-orientated sequencing through serialisation in a repetitive sequence, with iter-
ative microscale serialisation in Figure 4.2 actualised in a circularly cascaded framework sequence based
on operational systems-thinking principles, as depicted in Figure 4.1. In this way, the integrated theoretical
analysis framework shown by the balancing trilemma in Figure 3.14 serves as a blueprint in the computer
programme logic for the FPV digital twin model in Figure 4.2. As such, Figure 4.2 defines a digital replica of
the operational objects and processes within an FPV ecosystem within the system dynamics architectural
configuration of a closed loop.
On a philosophical level, the integrated CAS-model illustrated in Figure 4.2 proposes a macroscopic
perspective of the assessment process whereby the interdisciplinary object blocks run in an integrated cas-
caded network, as opposed to a microscopic perspective, with the functionally-disciplined object blocks
running individually, but in isolation. In this context, the framework in Figure 4.2 includes three aspects,
namely (a) the architectural content or metamodel; (b) the process activities and their order of sequence;
and (c) the organisation and roles of the systemic elements (systemic actors). In this context, the imple-
mentation of the conceptual geographical framework model of Figure 3.18 in the computerised model of
Figure 4.2 serves to reveal the deeper systemic and theoretical foundations in a new methodological ap-
proach toward realising and validating the research aim and objectives in a scientifically-founded valuation
paradigm for experimental modelling.
In support of the research methodology, the proposed framework model in the diagrammatic represen-
tation of Figure 4.2 is realised as a dynamic system in a computer model expression and is, according to
the conceptual, analytical framework of Figure 3.18, implemented as a typical stock-flow diagram to model
the reciprocal system dynamics. At the onset of the description of the parametric FPV model design, it
is of crucial importance to take cognisance of the fact that the model-based design methodology to es-
tablish the model of Figure 4.2 is structured around the context of the research assumptions defined in
Section 1.5. The consolidated system dynamics model of Figure 4.2 describes an integrated system en-
capsulating three systemic nodes or elements, namely the energy, environmental, and economic nodes, or
software object elements consolidated into one analytical system. As such, the construction of the simu-
lated software model of the cascade system, namely, Figure 4.2, implements the sustainability reference
framework of Figure 3.18 as a logical architectural programme. Notably, according to the integrated CAS
model of Figure 4.2, systems thinking enables the use of object-orientated data-processing techniques to-
wards specifying the structure, function and behaviour of floating solar PV ecosystem models in terms of
computer programme objects (Prinsloo, 2020).
As a system dynamics archetype for causal systems diagnosis of floatovoltaic impacts, the responsive
model design of Figure 4.2 is intended to model the static and dynamic responses of a real-world floating
solar PV installation process in response to how the FSPV system reacts to variations in parametric inputs
over time. As such, the CAS model of Figure 4.2 incorporates the component synthesis of the subsystemic
object functionaries (energy, environmental, economic), as well as the definition of the potential mutually
beneficial/detrimental relationships and bilateral synergistic data flows across the suite of clustered sub-
126
systemic network objects, either as continuous sustainability transactions or as a resource or cost-data
movements. In this dynamic context, this synthesised parametric systems model establishes a contextu-
ally geo-sensitive sustainable assessment model, which integrates the innate systemic intelligence of the
floating solar ecosystem into the sphere of interdependent operational dynamics. The domain object inter-
connections in the system dynamics model represent the complex interactions of the systemic components
and the domain relationships derived from the tri-dentate modelling framework topology and archetype, as
conceptualised in Figure 3.18.
With the modelled architecture of Figure 4.2 accommodating the organising principles of a smart ana-
lytical assessment framework in a contextually-sensitive computer simulation model, the following section
presents the accomplishment of the mission for designing a computer model framework. The designer
is subsequently tasked with giving analytical expression to the transdisciplinary ISA framework within a
CAS-model system dynamics environment. Accordingly, and as described in the next section, the thesis
formulates the object functions and interactions of the computer model as analytical model expressions in
terms of arithmetic variables and functional procedures in a model-based design approach.
4.2.3. Multiscale system dynamics simulation model for floating PV
Based on a coherent system dynamics simulation model for floatovoltaic technology installed on open-air
water reservoirs, the previous sections laid the theoretical foundations for the components of the floatovoltaic
information system. In preparation for scientific experimentation, this section describes the integrated floa-
tovoltaic synthesised model components and their integrated functionality.
From a mathematical modelling perspective, this thesis’s scientific computer modelling and simulation
methodology can now use a geographical model-based design approach to define the floating photovoltaic
ecosystem components as software domain objects in a system dynamics model structure, as shown in the
model layout of Figure 4.3. The proposed floating photovoltaic emulator architecture in Figure 4.3 advances
a simulation tool specific to floating solar photovoltaic systems.
Technological
Energy
Simulation
Model
Object
Economical
Simulation
Model
Object
Environmental
Simulation
Model
Object
f(x)e(t)f(y)e(t)
f(z)e(t)
Power Pm (kWh)
Conversion eff Pη(%)
Volts potential Vm (V)
Ampere current Im (A)
Panel temps fTpanel (◦C)
Project lifetime specifications
Site location configurations
Geometric design & surface area
Technical floater design parameters
Technical FPV design configurations
Agro yield (kg)
Avoided CO2(kg)
H2O saved (liter)
Avoided SOx(kg)
Avoided NOx(kg)
Avoided aPE (kg)
Avoid Coal fuel (kg)
Avoid Ash waste (kg)
Solar irradiation
Weather data
Climatic data
Cloud modulation
Agronomic setup
Revenue/Savings (R)
LCOE, LACE (R/kWh)
Depreciation value (R)
NPV, Payback (yrs)
ecoTax credits (R)
Value of H2O (R)
Value agro prod (R)
Capex & Opex costs (R)
Incentives, funding & subsidies (R)
Water, land & agro economics (R)
Electricity tariff rates (R/kWh)
Carbon credit pricing (R/ton)
Discount & interest rates (%)
ecoTaxes savings & rates (%)
Country-specific geospatial, environmental, legislative, economic and technical context
Figure 4.3: Coherent system dynamics simulation model for floatovoltaic technology, embedded in a parametric
synthesised model ontology for dynamic system characterisation (source: author).
127
Using a compositional approach and spatio-temporal modelling, the FPV simulation model includes an
assembly of the operating system’s elemental components (technical, economic and environmental model
objects) in a compositional model structure to effectuate a dynamic parametric time-series analysis toward
simulation-assisted performance modelling. The integrated recursive system dynamics simulation model
for floatovoltaic technology, as depicted in the multivariate model of Figure 4.3, can thus be formulated as a
multivariate analytical system to orchestrate the analytical process according to the topology of Figure 4.2.
The interactive FPV synthesis model in Figure 4.3 defines the integrated model of the floating PV system
in terms of process-level data, namely temporal and spatial scales, while defining its context in terms of the
agricultural niche to which it applies. Note that the data flow between the model object inputs and outputs in
Figure 4.3 are not unidirectional. Instead, the object outputs are characterised by various systems’ dynamic
feedback loops in a non-linear fashion, as regulated by the causal diagram in Figure H.1. For example, the
Economic-object outputs are functions of the Energy-object and Environmental-object inputs plus outputs.
Similarly, the Energy-object outputs are functions of the Energy-object inputs and the Environmental-object
inputs (refer Figure H.1).
The FPV’s multiplexed co-simulation model in Figure 4.3 represents the proposed system dynamics
model and application kernel embedded into a proposed FPV-GIS toolset. As a lumped-element model, the
lumped-parameter model depicted in the schematic block diagram of Figure 4.3 illustrates the closed-loop
workflow processes for analysing the multicore floatovoltaic project ecosystem operations in a networked
model ontology comprising multiple feedback and feedforward loops. As a co-simulated system dynamics
model, its applications in Figure 4.3 define the FPV system identifiers and model input parameters for any
user-defined FPV project variant. The parameters organised by the floating PV system model reveal the
collaborative framework simulation integration of the 3E = Energy, Environmental and Economic simulation
model objects. While Figure 4.3 details the graphical description of the integrated model, it also gives an
overview and insights into the object couplings together with the subsystemic object inputs and outputs of
the systemic model in preparation for software implementation.
The proposed model in Figure 4.3 helps focus on the physical implementation of the dynamic system
for sustainability assessments for floating PV systems and the extraction of the parameters of interest. The
model-based design process associated with developing the FPV system model in Figure 4.3 from a purely
conceptual model into a concrete model of indicators is presented in Sections 4.3 and 4.4. The detailed
discussion in these sections includes advice on which indicator metrics to select, algorithm development to-
wards indicator computing, and formulations to aggregate the chosen metrics in vectors or nexus-indicator
sets. These integrational and synthesis aspects form part of a floating PV system’s collective multiscale
simulation modelling, thus capturing the multisector dynamics of the FPV ecosystem in a digital twin as per-
formance, diagnostic and analysis tool model. For the benefit of real-world practitioners, a critical reflection
on the selected parameters and indicators of the FPV simulation model in Figure 4.3 is presented in Sec-
tion 4.3.2, while Sections 4.3.3 to 4.4.2 detail the description and modelling of each FPV model parameter
and indicator sets more comprehensively.
The mathematical definition of system causality aims at mapping from the time-dependent inputs to
the time-dependent outputs in Figure 4.3, with the floating PV ecosystem incorporating the collective in-
telligence of the composite ecosystem in a responsive model-design configuration. The multi-attribute
discrete-time algebraic simulation model for the floating PV systems of Figure 4.3 therefore implements
a fully integrated and iterative analysis-by-synthesis model for floatovoltaic technology. Defined as an
energy-economy-environmentally-modelled configuration, the internal dynamic model expressions driving
Figure 4.3 essentially replicate the operations and interacting dynamic responses of a real-world floato-
voltaic ecosystem. Multivariate model-building exercises can, as such, fundamentally assimilate the inte-
grated theoretical framework as a recurring empirical operation in a parametric synthesised model topology
and ontology. While the cross-functional model of Figure 4.3 encapsulates the functional energy, envi-
128
ronmental and economic elements in an integrated applied mathematics modelling approach, the self-
regulating dynamic system is choreographed by the solar-driven PV-lib Python energy-generating simulator
(Holmgren et al., 2018; Sandia, 2020). As such, the integrated performance simulation model of the floating
photovoltaic ecosystem is designed to use an hourly solar resource and meteorological data set to perform
discrete-time or real-time simulations and lifecycle sustainability assessments or risk assessments for future
planned floating PV-system installations.
From a real-time simulation processing perspective, the recurrent real-time synchronous analytical float-
ing PV-systems model of Figure 4.3 performs discrete-time digital electronic data processing according to
the sliding-window procedure illustrated in Figure 4.4. The temporal pipeline process depicted by Figure 4.4
automates the repetitive simulation cycles of the co-simulated 3E floating PV ecosystem elements in com-
puter programme software. As such, data-processing activities in the sequence depicted in Figure 4.4 can
be divided into the broad steps of data inputs, data processing, data collection, data storage, the sorting of
data, data analysis, and data presentation. Being a solar-driven process, time-series forecasting of floating
solar PV operations starts with processing dynamic solar radiation and weather station data at each time
increment of the iterative execution stack.
t−2t−1t t + 1 time horizon
ensuing 3E analysis iteration sample periods
next 3E analysis sample iteration
Figure 4.4: The progressive analytical processing steps of the iterative operations of the discrete-time synchronous
system dynamics simulation model during data-analysis/processing (source: author).
Each discrete-time spatiotemporal processing step in Figure 4.4 follows the reciprocal workflow in pars-
ing through the nested sub-model sequence of Figure K.1. As such, Figure 4.4 illustrates how the computer-
automated real-time co-simulation of the interactive floating PV ecosystem in Figure 4.3 is governed and
orchestrated by the repetitive processing cycle of the discrete-time energy simulator of the real-time PVlib
model. As such, the responsive repetition of the interactive floating PV ecosystem in Figure 4.3 can co-
simulate and output an environmental assessment of floating solar PV technology, together with the asso-
ciated predictive energy yield and economic performance estimates, both indicated instantaneously and in
life-cycle format.
Following the pre-processing of time-series data at each simulated time increment, the recursive data-
processing method and the co-simulation cycles illustrated in Figure 4.4 involve a shifting-time window
involving the conversion of the raw input data into useful floatovoltaic ecosystem performance information
over the complete array of the data-sampling occurrences. As such, the responsive geo-aware model pro-
posed in this thesis (see Figure 4.3) effectively models the intrinsic behavioural properties of the floatovoltaic
energy system (a business enterprise) in discrete time intervals of a specific frequency of analysis as time-
varying functions and in terms of the temporal resolution level of the model’s input data and the prevailing
environment around the ecosystem.
This section summarises the skeleton of the synthesised floatovoltaic system and its functionality as
an integrated systemic model. With the systemic details of the soft-simulation model already introduced,
the following section accordingly formulates the computerised-model object formulations and interactions
as analytical model expressions in terms of arithmetic variables and functions.
129
4.3. Research Model: Component-level Design and Implementation
Towards addressing the object specifics of the model in terms of the final portion of the systemic defi-
nitions according to Research Objective 2, the thesis continues the implementation process, defined by
Figure 3.22a, towards dictating the model-based design element specifics in terms of the constituent FPV
subsystem parameters and indicator metrics that define the floatovoltaic systems model as subsystemic
components. As such, the implementation of the synthesised model dictates the model-based design
specifics to establish the constituent subsystem objects that define the floatovoltaic ecosystem. This con-
cept includes descriptions of the energy, environmental and economic domain objects, each with their own
set of constructs and algorithmic interactions to define the inter-operations among them. According to
the model-based systems-engineering design method, this model-design objective includes designing the
subsystemic components in a dynamic programming environment. It further comprises the parameterisa-
tion and software object details for the subsystems that characterise the technical performance of the FPV
system through the parametric floatovoltaics simulation model, as depicted in Figure 4.3.
The model-based design approach to establish the floating PV-system components is described in Sec-
tion 4.3.1, while Section 4.3.2 defines the model parameters, output indicators and user-configuration set-
tings of the overall floatovoltaic technology information system. Sections 4.3.3, 4.3.4 and 4.3.5 subsequently
detail the object-membership functions with algebraic equation formulations for the respective energy, envi-
ronmental and economic model objects and interactions.
4.3.1. Model-based design of floating PV systems components
Since the synthesised model of the floating solar system is programmed as a reconfigurable system dy-
namics network, the model needs to detail the mathematical notations and user-configuration specifications
available for reconfigurable FPV analytical syntheses through simulation formulae. Converging under the
real-world decision-support goal of Research Objective 2, the model-based design should include model
parameters for both the static and the dynamic behaviour of the floating PV ecosystem components for any
given geographical location (GPS location), according to the design parameters of the PV system, and in
terms of the contextually-sensitive framework of Figure 4.3.
Model-based design (from concept to code) entails the process of selecting essential features and defin-
ing the associated mathematical approximations to represent the reality of interest around the floating PV
system sustainability in a mathematically-synthesised model, in which the mathematical equations and pa-
rameters describe the physical reality in a sequence of modelling steps (Thacker et al., 2004). This calculus
is based on the theoretical FPV modelling framework and system dynamics modelling logic, and includes
the definition of the model calculations suitable for reconfigurable design specifications in a hierarchically
modelled approach. As such, the system, as well as the subsystems and their components, as depicted
in Figure 4.3, jointly represents the model to study the reality of interest (namely floatovoltaic system sus-
tainability). The data collected at each level of the integrated hierarchical model in Figure 4.3 comprise
component definitions to provide metric measurements that represent the ability of the floating PV model
to empirically predict the performance attributes and impact quantities of the phenomenology critical to the
accurate simulation of the integrated systemic performance of the model in Figure 4.3. The integrated hi-
erarchical model (Figure 4.3) uses computer-aided software tools to define and characterise the relevant
object and subsystemic classes, through which an object-oriented modelling approach allows subsystems
to be built, one-at-a-time, until the entire FPV system model is constructed entirely as an integrated model.
Towards the implementation of the component-level specifications according to model-based design
principles, the model implementation proposed in Figure 4.3 engages first theoretical principles around en-
gineering physics and model-based systems-engineering designs (Estefan, 2008; Wierzbicki et al., 2000).
130
Designing the component-level aspects of the integrated simulation model design from first principles,
the modelling enterprise establishes the three physical knowledge units or systemic submodel integrants,
namely the functionary energy, environmental and economic model integrants or components as object-
oriented software kernels in a predominantly physics engine calculus. From an intelligent cyber-physical
system of systems perspective (Barachini and Stary, 2022), the integrated digital twin model depicted by
Figure J.1 illustrates how the object-oriented system of systems architecture for the synchronous software
integration of 3E and WEL systems into a geo-informatically housed discrete-time analytical simulation and
decision-support model for floating PV characterisation and planning assessments.
In starting with the implementation of the base layer of the geoinformatics ecosystem, the rest of the
discussion narrates the model-based design of the floatovoltaic ecosystem components depicted in the
system dynamics model of Figure 4.3. The thesis defines and computes the performance parameters and
indicator metrics associated with the object components of floatovoltaic synthesised models by engaging
the first fundamental principles of science and expert subject matter inputs. This FPV model design pro-
cess is carried out in a model-based design process, discussed in detail in Sections 4.3 and 4.4 of this
chapter. In this context, model-based systems engineering provides a theoretical basis for developing each
systemic component. According to the theory, model-based systems engineering begins by defining the
problem statement and the systemic boundaries, followed by applying the modelling techniques. Through
this approach, the thesis can actualise the systematic implementation of the theoretical framework design
for the integrated floating solar synthesised model by applying data and techniques from the Geographical,
Mathematical and Physical Sciences (Schneider and Hutter, 2009). As such, FPV model design includes a
sequence of modelling steps in defining the analytical model expressions to establish a universal simulation
model capable of driving the predictive analytics mission in profiling a floatovoltaic project and appraising it.
While the problem statement and systemic boundaries have already been defined, and the model-
based design of the floating photovoltaic model and its elements have been described, the following section
informs of the specific notational aspects that determine the conceptualisation of the energy, environmental
and economic domain subsystems outlined in the model archetype of Figure 4.3.
4.3.2. Model-based design parameters, notations and configurations
This section details the model parameters and mathematical symbols, showing the standard symbols used
in modern mathematical notations and within the formulae as object descriptions to meet the design-
evaluation guidelines concerning international floating solar PV projects. It characterises the mathemat-
ical notations and user-configuration specifications for the terminology for the reconfigurable analytically-
synthesised FPV model and its simulation formulae towards reconfigurable design specifications.
A computer configuration model generally offers a definitional workspace provisioned by a configuration
file that can customise simulation operations for specific experimental purposes. Creating an assessment
configuration file permits the definition of the theoretical systems of the model components and their foun-
dations. It also allows for the specification of the model components in a hierarchically-based model devel-
opment configuration that can be personalised or customised around the required experimental specifics
(country, location, currency, legislation, etc.). Regarding geographical mathematical modelling principles,
the parametric time-series analysis model of Figure 4.3 defines universally-accepted mathematical nota-
tions and terminology for FPV model parametrisation, further allowing for user-configuration specifications
to establish a fully flexible and reconfigurable analytical floating PV synthesis model.
Regarding the mathematical notations for the predictive simulation of future planned floatovoltaic projects,
it is essential to standardise those used in the mathematical formulae programmed into the Python computer
code, which drives the simulation. The system variables, terminology, configurations, and defined identifier
metrics for the FPV model include input/output parameter data streams defining the techno-hydrological
131
attributes, techno-economic attributes, enviro-economic attributes and the techno-atmospheric attributes of
a physically-planned FPV installation. The parameter Afpv reflects the water PV plot size as a geometric
water surface area of the site covered by FPV panels on a recipient water reservoir (in m2). The parameter
fApd reflects the panel density of the engineered design with array spacing (panels/100m2). The remaining
parameters attributed to the model Figure 4.3 are defined and summarised in Table N.1 (see Appendix N).
The full set of default configuration parameters for the specification of the reconfigurable model in Figure 4.3
is comprehensively detailed in Appendix O.
All the designated parameter variables mentioned thus far are defined as reconfigurable floatovoltaic
enterprise/ecosystem settings. At startup, the system dynamics simulation model of Figure 4.3 reads the
user-configuration-file (FPVconfig.csv) to set up the various local-context configuration parameters for the
proposed floatovoltaic project under consideration (for example, see the experimental setup in Table O.1
in Appendix O). These FPV-system identifiers enable the user to pre-configure the geographical region,
monetary currency, carbon tax regimes, financial incentives, cost of farmland, agronomic production rates,
interest rates, tariff structures, inflation rate, equipment Capex, system Opex and maintenance details, FPV
floater type, PV-system type, equipment type, surface albedo, systemic degradation rate, geometric design,
geometric description, floatovoltaic layout and orientation, country-specific or region-specific environmental
offsets, the growth or escalation rates of the parameters as mentioned above, as well as the lifetime of
the project. The panel density is determined by the inter-module spacing (water or ground coverage ratio,
GCR), meaning the ratio of the total water area covered by the photovoltaic modules, given the row-to-row
panel spacing.
Since each of the key model parameter terms (refer to refTabtab:param, Appendix N) references an
individual time-series point/instance in time in a varying data-flow sequence, the discrete-time Python model
processes and stores these parameters as time-varying arrays in a data matrix. Therefore, for each of the
above parameter terms, a subscript (y, t) is added, notation-wise, to reference that parameter at time tfor a
year y. Universally, the matrix subscript "y” defines the project year (1,2,· · · , Y ), while the array subscript
"t” represents the time-marker (minute, hour, day, month) within that year y. For example, the notation,
Pm(y,t), refers to floatovoltaic power generated at time instance tfor project year y. Note the temporal
aspects of the model outcomes for all FPV model indicators in Table R.1 (Appendix R).
Moreover, each of the model parameters for the FPV-GIS model also has a spatial attribution, thus mak-
ing the FPV-GIS model outputs truly spatiotemporal. As such, the FPV-GIS model parameter output terms
(refer to Tables R.1 to T.1, Appendices R to T) relate to the spatial attributions for the model input parameters
(specified in Table N.1, Appendix N). These input parameters, in turn, associate with the spatial attributions
of the FPV model configuration (specified in Table O.1, Appendix O), as defined in terms of the system con-
figuration parameters for a particular spatial location (latitude, longitude, altitude) as a geo-reference point
on the earth’s surface. As such, the spatiotemporal FPV model parameter outputs are a function of the
spatial location of the FPV unit, the landscape character of the installation site, the surface solar irradiance
data for that location, and the meteorological and climatic data associated with that geographical location.
A final aspect of the spatial arrangement of the model attributes relates to the spatial scale of the FPV-GIS
model parameters, which is a direct function of the spatial resolution of the meteorological and climatic
TMY (typical meteorological year) input dataset that drives the FPV model. This climate model resolution
is typically 3 to 15 km, depending on the geo-station data resolution (number of weather stations in the
geographical installation area) or the resolution of the remote sensing measurements (satellite, radar, lidar)
the data service provider employed in building the TMY dataset (chosen for the FPV system installation
experiments) (refer HelioClim (2022)).
The rest of this discussion details the skeleton implementation of the floatovoltaic synthesised model
towards its system dynamics implementation as a data-gathering research instrument for application in
scenario-driven case-study experiments. The following sections dictate the model-based design specifics
132
to establish the constituent subsystemic object components that define the floatovoltaic ecosystem and its
sub-system interactions. It details the model-based design of the Energy model object in Section 4.3.3,
while Section 4.3.4 details the design of the Environmental model object. The Economic model object
model-based design is detailed in Section 4.3.5, while the cross-domain collaborations and performance
metrics are detailed in Section 4.3.6.
4.3.3. Energy simulation model object design
This section describes the energy characterisation and profiling of the floatovoltaic system as described
in mathematical terms by the simulation of the energy domain model object based on physics simulations
in an object-orientated calculus environment. While the energy model is located at the nerve centre of
the integrated floating PV ecosystem, this subsystem model is concerned with the technical assessment
of floating photovoltaic power generation under varying weather profiles. While it defines suitable techno-
metrics to describe the energy behaviour of the FPV system, it also provides the dynamic techno-economic
and techno-enviro membership functions and other interactions among the FPV ecosystem objects of the
FPV energy model. With floatovoltaic systems being solar-event-driven production systems, and electricity
generation being one of the main cost and income drivers of floatovoltaic installations, this model deserves
special attention. The energy model object hosts the recipient reservoir’s solar position and time-zone
sub-model, the PV panel thermal model, the PV panel microclimate thermal-impact model, and the energy
production sub-model, each with its respective description of the mathematical simulation details.
4.3.3.1. Reservoir FPV solar-positioned time-zone model
In the simulation evaluation of a floating photovoltaic power system, the PVlib ModelChain initialisation
functions are used to set up the locational details of the reservoir at the simulation startup of the solar PVlib
model (Sandia, 2020). As such, the PVlib Python model chain is set up with the PVlib location library class
to start solar geometry processing using the Solar Position Algorithm (SPA) for solar radiation applications
(Reda and Andreas, 2008). While the SPA procedure stores the floatovoltaic location site details as PVlib
array data parameters for a given PV systems topology, the SPA modular code computes the hourly position
of the sun in the sky and stores this information in the SPA data frame (Holmgren et al., 2018; Sandia, 2020).
This procedure includes defining the variables (defaults denoted in brackets) associated with the reservoir’s
geospatial configuration as follows: longitude (fln =18.8 degrees E), latitude (flt =33.3 degrees S), time
zone (ftz =GMT+2) and altitude (falt =200m) (for parameter descriptions, refer Table N.1 in Appendix N).
The user also specifies the optimal panel array orientation (for maximum solar energy absorption, optimal
panel spacing and to counteract open-surface wind-drag forces) as the optimal surface tilt angle (fptilt =12
degrees), panel orientation and site surface azimuth (0 =North), solar panel density (a function of the PV
plot size, panel tilt angle, optimal panel spacing, panel shading/shadowing) (fApd =60 panels per 100m2),
the geometry of the floatovoltaic system and the plot size or FPV system’s surface area (Afpv =1000m2),
plus the light-polarising water reflectivity function as the albedo of the geographical terrain topography and
the water surface (falbedo =0.15) in the user configuration file. While arguments exist for a dynamic albedo
function for water (Patel and Rix, 2020), such a function would depend on the site geographics. To facilitate
the optical characterisation of the PV system, the PVlib Python pseudo-code syntax given in Function F.1
calls on the PVlib solar position algorithm (solarposition.SPA) function object (together with the normal PVlib
ModelChain.init-system functions) (King et al., 2004) to initiate the PVlib object variables, and to set up
data imports that facilitate TMY2 or TMY3 data transfers (from earth observation platform archives) (Solar
Energy Lab, 2017; Ascencio-Vasquez et al., 2021). The historical TMY geosensor data are loaded from the
selected world-readable meteo-database. It represents the in-situ measurement observations from dynamic
133
weather station data or remote sensing satellite weather data that drives the FPV model at the heart of the
geospatial digital twin simulator. The SPA furthermore configures the irradiance module, initiating array
functions to model global horizontal irradiance, direct normal irradiance, diffuse horizontal irradiance, total
irradiance, the on-panel plane-of-array irradiance, and the climatic conditions for the installation location site
under various calendar dates and timestamps associated with the given geospatial configuration context.
Note that according to the convention used in this thesis, all Sandia PVlib functions are denoted by (F.x) as
illustrated below1:
SP Af pv =pvlib.solarposition.spa(f tz, f lt, f ln)(F.1)
In irradiance-to-power modelling, the solar irradiance incidence-angle dependency defined by the SPA
Function F.1 is critical to the accurate global irradiance and efficiency characterisation of the PV power
system, as detailed later in Section 4.3.3.4. In this context, the SPA return-data frame columns specify
the zenith (degrees), azimuth (degrees), apparent zenith (degrees), apparent elevation (degrees), elevation
(degrees), and equation-of-time (minutes, hours) of the floatovoltaic system. These parameters are passed
to the PVlib SAPM model to estimate the geospatial energy yield under the various timing, operational and
configurational conditions of the floatovoltaic system.
4.3.3.2. PV panel thermal model
Because, environmentally-wise PV-power generation is highly temperature/climate-sensitive (Nisar et al.,
2022; Micheli, 2022), the temperature properties of an operative floating photovoltaic module in Figure 4.5
need to be modelled in terms of the characteristic temperature-performance curves of a PV module (King
et al., 2004). This solar-cell temperature and heat vulnerability effect is explained by the power curves
of the PV panel, as depicted in Figure 4.5. It illustrates the inverse irradiance temperature/power output
relationship in the temperature properties and the power/temperature coefficient graphs (Chen et al., 2017;
Rahaman et al., 2023). The panel’s proposed PV thermal model also accounts for the ambient temperature,
thermodynamic heat exchange and wind effect on the floating PV panels.
Figure 4.5: PV-panel power curves: an inverse irradiance temperature/power output relationship in temperature
properties as power/temperature coefficients (source: Chen et al. (2017), page 7).
The PVlib photovoltaic performance modelling thus includes a temperature-dependent PV-efficiency
function or solar illumination-dependant temperature coefficients. These coefficients model the negative
and positive effects of the heat loss coefficient and panel-cell operating temperature on the yield perfor-
mances of a ground-mounted PV system (Driesse et al., 2022).
1In the thesis’s convention, the function numbers thus differ from the mathematical equation numbers which are denoted by
numerical equation numbers only.
134
In conventional land-based PV performance modelling, the PVlib software library Function F.2 includes
a transient thermal model to characterise the module temperature effects of Figure 4.5. This routine invokes
the Sandia module temperature model and the Sandia-cell temperature model software routines to estimate
the internal PV-cell-operating temperatures (gTpanel) for a given commercial photovoltaic module (Sandia-
Module). In the conventional PVlib software, the PV-operating temperature estimates are time-dependent
functions of the effective on-panel irradiance as the transmitted values for on-panel irradiance (ITR), the
ambient temperature (Tdry), the wind speed (Wspd), and the photovoltaic-module-specific thermal coeffi-
cients (amod, bmod , δTmod ) specified by the Sandia-module manufacturer for a chosen PV panel (obtained
from the PVlib library catalogue) (Sandia, 2020) (for parameter descriptions, refer Table N.1 in Appendix N).
gT panel(y,t)=pvlib.temperature.sapm.cell(I T R(y,t),
T dry(y,t), W spd(y,t), amod, bmod, δTmod )(F.2)
However, experimental investigations into the hydroclimate effects of floating solar panels on cooler wa-
ter surfaces have shown substantial decreases in the panel operating temperatures of FPV systems due
to the heat loss coefficient factor. As such, solar panel operating temperature anomalies occur because of
the hydroclimate impact effects, which causes the water-panel energy balance to result in a lower equilib-
rium operating temperature within the solar module. Through the characteristic heat loss coefficient of the
solar panels, these dynamic cooling and moisture effects, in turn, signify substantial improvements in FPV
power-plant efficiencies, effects that are not yet fully accounted for floatovoltaics modelling in existing PVlib
libraries (Karatas and Yilmaz, 2021; NREL, 2019a). This performance variability is primarily attributed to
the microclimatic effects on floating solar power-plant production, causing signature variations in the be-
haviour of solar panel temperatures under different environmental operating conditions within the floating
PV microhabitat (Mahdi et al., 2020; Prinsloo, 2019). For floating PV systems, the less temperate FPV mi-
croclimate primarily causes a reduced ambient operating temperature (f T ambient(y,t)), which is engaged
in the floating PV panel temperature estimation procedure of Function F.3.
f T panel(y,t)=pvlib.temperature.sapm.cell(I T R(y,t),
f T ambient(y,t), W spd(y,t), amod , bmod, δ Tmod)(F.3)
Project appraises should rather not engage Function F.2 in floating PV assessments, as it would cause
inaccurate energy yield assessments, which in turn impacts environmental offset predictions while hamper-
ing project bankability forecasts. Function F.3 solves these problems through the help of the environmental
simulation model object described in Section 4.3.4. It fundamentally relies on a newly proposed microcli-
mate model, developed within the environmental model object, to estimate the time-varying floatovoltaic
ambient temperature (f T ambient(y,t)) and relative humidity (RH(y,t)) as detailed in Section 4.3.4.2.
4.3.3.3. PV panel microclimate thermal-impact model
While field studies confirm the positive effect of the cooler and more humid surrounding environment on the
energy produced by a floating PV system (Karatas and Yilmaz, 2021; Kjeldstad et al., 2021), FPV research
is often specifically interested in modelling the on-module temperature variations internal to floating solar
PV systems (Peters and Nobre, 2020, 2021; Sutanto and Indartono, 2019). Therefore, to compute the
floating PV module efficiency, the efficiency of heat loss to the environment as well as the operational cell
temperature needs to be modelled for a given floating PV technology (Lindholm et al., 2021). This thesis
proposes a more general Geographical Sciences approach in defining a microclimate model to support the
thermal modelling of floating PV panels. In this context, the thesis argues that panel operating temperatures
135
(defining the heat signature of the PV panels) should be a dynamic function of both the mesoscale and
microscale environmental conditions. This environmental feedback consideration is visually depicted by the
environmental aspects in Figures 3.10 and 3.12.
Regarding the thesis’ microclimate impact modelling concept during each time instance of the sim-
ulation, causality feedback caused by hydroclimatic variability should be facilitated by the environmental
abstractions shown in Figures 3.11 and 4.1. Since the hydroclimatic conditions play a crucial role in de-
termining the micro-climate impact on the PV panel parameters (i.e. thermal, humidity, wind speed), a
climate reformation algorithm is developed to pre-process the input parameters to the PVlib algorithm. In
this context, the thesis proposes a climatically-motivated floating PV habitat-modelling procedure (detailed
in Section 4.3.4.2) as the basis for the estimation of parameters (such as the cell temperatures) for floating
PV panels installed over open-water basins. Figure 4.6 depicts the pre-processing procedure for the hy-
droclimatic conditioning of the PVlib model input parameters for floating PV system operations. It includes
climate reformation built around hydroclimatic pre-processing as a function of the meteorological weather
station data and PV floater design type.
Meteorological
TMY data
Data repository
Hydroclimatic
Pre-processing
Floater-specification
PVlib engine
FPV-configuration
PVlib engine
GPV-configuration
ut r e y
y
Figure 4.6: A conceptual model for conditioning the PVlib model inputs (purple) for floating PV system operations
based on FPV microhabitat climate reformation, environmental sub-model solution (green) built around hydroclimatic
pre-processing as a function of TMY weather station data and PV floater design-type (source: author).
As part of the floating solar photovoltaic array performance model object, modelling the microclimate-
impact on PV panel thermal model in Figure 4.6 helps to decipher the impact of the environmental micro-
habitat on the thermal behaviour of floating photovoltaic installations. In this context, special consideration
and efforts are needed in the modelling of the surrounding microclimate (the floater hydroclimate), towards
the iterative conditioning of the thermal model of the PVlib model chain to help predict the internal operating
temperatures of the floating solar panel as a function of the prevailing environmental conditions (recur-
rently received from the environmental model object). Estimating the module temperature for water-based
photovoltaic systems is a crucial requirement, as the sensitivity of the module temperature-power function
determines the module’s optic-to-electricity energy conversion efficiency (light-productivity-factor). Such an
FPV modelling solution requires consideration and a significant modelling effort to condition the PVlib pa-
rameters to differentiate between the microclimate and the moisture regimes for water-mounted floating PV
(FPV) and ground-mounted photovoltaics (GPV).
In the panel temperature thermal model of this thesis, the mathematical model for calculating or esti-
mating the FPV module temperature from weather station data accounts for the feedback effects from the
natural environmental system (microscale meteorological information) on the technical floating PV system
(refer to Figure 4.1). It conditions the contrasting PVlib performance procedure for conventional GPV sys-
tems (Function F.2) to that for FPV systems (Function F.3), to ensure that the climatic data input to the
PVlib procedure is conditioned for FPV and is adjusted to estimate the time-varying cell temperatures of a
floating photovoltaic system. Based on the environmental microclimate model, described in Section 4.3.4.2,
136
the modified function or procedure models the humidification of the floating PV microhabitat towards the
conditioning of the temperature profile for floating FPV solar panels as a means to estimate the PV cell
temperatures.
For the proposed microclimate simulation tool to function accurately to model and predict floating so-
lar panel parameters in open water basins, a new meteorologically-motivated floating PV-habitat modelling
procedure is defined by this thesis in Section 4.3.4.2. The proposed microclimate modelling procedure
conditions the floating PV-system meteo-input parameters for the microhabitat. It caters to a dynamic op-
erational temperature and humidity environment in which a moisture-rich water body surrounds the PV
panels (to act as a natural coolant and natural heatsink for the floating PV panels (Allen and Prinsloo,
2018)). Since such a cooler ambient environment or enveloping microhabitat effectively improves the ab-
solute solar-to-electricity conversion efficiency of the FPV system, the panel temperature model accounts
for the time-varying increases in the FPV photovoltaic module outputs. These increases in effectiveness
relate to decreasing panel temperatures in an inverse power/temperature relationship (King et al., 2004).
As a guideline, it is known from field experiments that floating PV systems typically deliver around an eight
to 16% efficiency gain in yield (fgPmη) compared to the output of GPV systems (Allen and Prinsloo, 2018;
World Bank, 2019b).
In this context, the microclimate model proposed by Section 4.3.4.2 can condition the local-scale mete-
orological microclimate parameters of the FPV microhabitat (input to the PVlib procedures) as a means to
account for the feedback effects from the natural environmental system (microscale meteorological data) on
the technical floating PV system operations and performances. Practically, the environmental microclimate
modelling parameters of the floating PV system detailed in Section 4.3.4.2 are used to condition the PVlib
model parameters for the surrounding FPV-microhabitat recurrently. As such, the energy object applies the
floatovoltaic ambient temperature-conditioning factor, ∆Td(y,t)or ∆Td’(y,t)in Section 4.3.4.2, to the panel’s
internal temperature predictor of PVlib, then using Equation 4 to calculate the estimated ambient tempera-
ture (fTambient) around the floating PV arrays. Given the predicted ambient temperatures for the anticipated
FPV microhabitat, the hydro climatically pre-conditioned meteorological data inputs to the Sandia procedure
ensure more accurate operational temperatures of the FPV panels (fTpanel) to better estimate the floating
PV system power output through Function F.4.
f T panel(y,t)=pvlib.temperature.sapm.cell(I T R(y,t),
(T dry(y,t)−∆T d′
(y,t)), W spd(y,t), amod, bmod, δTmod )(F.4)
Offering a floatovoltaic temperature offset calculation (fT ambient(y,t)=T dry_(y, t)−∆T d′
(y,t)), the
microclimate model enables the PVlib procedure to condition the floatovoltaic simulation model panel tem-
perature calculations in Function F.4. As such, the environmental model object of Section 4.3.4.2 defines
Equation 2 to render the ambient temperature reduction logically adapted to floatovoltaic operating condi-
tions. As one of the predictor variables in the energy production model of Function F.5, the hydroclimatic
temperature factor in Function F.4 should help to predict the internal operating temperatures of the floating
PV system (fTpanel) more accurately.
Apart from being used as a predictor of the energy yield of the floating PV system, the panel temper-
ature Function F.4 can also be applied to model advanced cooling technologies for floating PV systems.
PV panel cooling involves artificial enhancements and reductions in PV efficiency, which can also be mod-
elled through the microclimate parameters (Equations 2 and 3 in Section 4.3.4.2). Such examples include
experimental cases where FPV is equipped with technology-assisted radiative cooling-, active evaporative
cooling- or effective forced water-cooling systems to improve floatovoltaic performance intentionally (Dizier,
2018; Kjeldstad et al., 2021; Perrakis et al., 2021; Tina et al., 2021a). Such passive or active cooling in-
terventions (ACi =1) describe cases where the capacity of the FPV system is intentionally enhanced to
137
improve the FPV’s techno-economic performance. This cooling can typically improve the FPV power yield
to an average nett gain of six to 13.5% in yield efficiency (i.e. by sprinklers spraying irrigation water over
photovoltaic panels at regular intervals to disperse the excessive heat). In such technical panel washing
and cooling interventions, the ∆T d and ∆T d′floater ambient offset coefficients (FCα,FCβ) in Equations 2
and 3 can be applied more aggressively (values [0.00 · · · 3.00] to emulate cooling by ice-cold water or even
snow). In this way, the coefficients can serve as a means to condition the ambient operational environment
for natural or active cooling situations.
4.3.3.4. Energy production model
Finally, to gauge FPV performance and sustainability in energy production, the energy model can now pre-
dict the power output of a floating solar-powered energy-generation system (monofacial panels) at any given
GPS location (for which meteorological weather station data are available). The energy model requests a
continuous stream of data inputs from the environmental model object to perform the empirical predictions
of the energy yield profile in a sequence of processing steps.
To model the software energy object, according to the real-time modelling approach of Figure 4.3, the
idea is to condition a universal engineering model for PV to accommodate floating PV energy modelling
calculations. The benefit is that the open-source engineering performance model PVlib suite is already in
the public domain (given its creative commons attribution) (Gurupira and Rix, 2016; Sandia, 2020). Further-
more, since the energy object is an integral part of the framework-integrated FPV system modelling logic,
the open-source PVlib algorithm is also used as an integrated simulation synchronisation mechanism for
marshalling the iterative analytical process. The PVlib platform thus offers an iterative synchronous real-
time platform to enable software developments (such as the system dynamics network of this thesis) to be
choreographed in the step-by-step or iteration-by-iteration PVlib simulation run as depicted in Figure 4.4.
While PVlib maintains a database of various solar power plant components, the model development needs
to condition the PVlib simulation engine to suit floatovoltaic systemic properties to theoretically resemble
the real-world performance of floatovoltaic energy systems within the local landscape. As such, the en-
ergy model object relies extensively on the technical design philosophy of the PV system libraries and the
attributes formulated within the Sandia PV Array Performance Model (SAPM) as an engine to perform time-
series power-flow analysis for photovoltaic energy generation in the technical analysis phase (Holmgren
et al., 2018; King et al., 2004; Stein et al., 2017).
In terms of the PVlib toolbox procedures (engaged by the FPV-GIS toolset), an analysis of the lifetime
power performance for the PV system is determined for the first full-year period of the project. It is based
on the typical meteorological year (TMY) sensor data compiled from solar irradiation data (engaging solar
emulation or heliophysics models) modulated by spatiotemporal statistics of weather and cloud models
(engaging climate, aerosol, precipitation, humidity, wind and cloud physics) for the given installation site.
After that, the lifetime power performance is cyclically determined through the application of annual systemic
degradation and derating factors over the entire lifetime of the project (CSIR, 2022). Note that the PV
degradation factors are programmed into the user configuration file of the FPV model. While GPV solar
panels are subjected to higher temperatures that cause various power-derating effects over time (Kim et al.,
2021a; Stein, 2017), a cooler FPV ambience counteracts the harmful and deleterious impact on solar panels
and, in doing so, extends the system’s lifetime while increasing its power outputs. An FPV aqua-voltaic
counter-degradation function is therefore included in the simulation model’s configuration as an efficiency
improvement factor (fPVd =0.6%/year) to ensure that the warranty for the annual decline in performance is
corrected upwards (IEA PVPS, 2020).
The solar irradiance and meteorological time-series estimates are subsequently used as inputs to the
PVlib solar FPV power modelling algorithm to compute the solar FPV power-output estimates based on
138
the PVlib SAPM procedures (Sandia PV Array Performance Model procedures). This procedure estimates
energy delivery as a function of the accrued SPA parameter set, the temperature feedback climate sensi-
tivity, the chosen Sandia module (default=Canadian-Solar-CS5P-220M-2009), the operating temperatures
on the SAPM panel, panel density, and the Sandia-module component library parameter set (panel-power
density, electrical-signal profile, coefficients, etc.). It is assumed that the panel shading losses due to the
effects of trees are negligible in open-air water-mounted FPV systems. The panel-density parameter (fApd)
specified in the model user’s configuration file relates to the row-to-row shading. The Sandia PVlib model
calculates the diffuse radiation incidence on the solar FPV panels at the specified inclined angle. The solar
on-panel irradiance is estimated as Plane-of-Array (POA) irradiance, using a sequence of models internal
to PVlib (the decomposition of GHI into DNI and DHI). With the plane of array irradiance analysis, the solar
irradiance data is transposed to the PV panel plane of the array (Sandia, 2020). Following the transposition
of GHI, DNI and DHI into total irradiance and POA (as defined in Table N.1), the internal panel temperatures
of the floating PV system (fTambient) need to be determined as a key step toward floating PV power yield
estimation.
In this context, the model-output estimates of energy yield as a time series are calculated continuously
throughout the simulation run, simulating the solar-to-electricity conversion process. Model outputs are
stored in the time series Pm(y,t)(W or kW) of the energy-output performance indicator, and the energy-
conversion-efficiency time series Pη(y,t)as solar irradiance fill-factor % (light-productivity-factor, or solar-to-
electric conversion %).
Figure 4.7: PV module I-V performance curve showing five key points on the curve stored by the Sandia SAPM model
(source: King et al. (2004), page 9).
In terms of the parameterisation of the performance model towards the implementation of FPV energy
performance modelling, the subsystem engages open-source PV-Lib Python solar photovoltaic simulation
software for the remote-site analysis of any particular systems configuration and the associated perfor-
mance metrics (Sandia, 2020). This model applies the PVlib plane-of-array (POA) irradiance model to
transpose the solar irradiance data to the effective plane of the array (ITR on solar panel surface). It thus
determines the effective on-panel/in-plane irradiance (ITR) from the sun path diagram, by combining the
DNI with the sky diffuse and ground/water-reflected irradiance through the direct and diffuse irradiance
components in the plane of the floating photovoltaic array (modelling the transposition gain in the GHI to
POA transposition) (Sandia, 2020). The Python syntax in Function F.5 then calls upon the universal Sandia
PVlib Python SAPM procedure towards computing the FPV power output (Pm) for the defined solar array,
with a specified nameplate rating (in watts), given the determined cell/panel temperatures (fTpanel) and the
transmitted values on panel irradiance (ITR) for the reservoir site under consideration.
f P m(y,t)=pvlib.pvsystem.sapm(I T R(y,t), f T panel(y,t), SandiaM odule)(F.5)
139
gP m(y,t)=pvlib.pvsystem.sapm(I T R(y,t), gT panel(y,t), SandiaM odule)(F.6)
While the internal operating temperatures of the floating PV system (fTpanel) are predicted by Function F.4,
it serves as key identifiers in predicting the time-varying energy yield performance of the floating PV system
through Function F.5. From the routine in Function F.5, the theoretical energy production potential is stored
in the SAPM module output data frame (DF). It renders an instantaneous maximum power-point current, Im
(A), voltage Vm (V), and power Pm (W), as a real-time series array associated with the data sample of each
solar weather station (concurrently and recurrently provided by the environmental model object). Note that
the energy production for a co-located GPV system (gPm) is calculated by using the gTpanel parameter in
Function F.6, where the gTpanel parameter is obtained from Function F.2. Also, note that by further pre-
calling the PVlib.pvsystem.pvwatts-losses function, the Sandia procedure in Function F.5 can automatically
factor in system capacity fade and the specified performance losses for factors such as soiling, shading
attenuation, component mismatches, wiring, connections, nameplate-rating, snow losses, age degradation
and so forth (Sandia, 2020). The calculations described above cater for the most commercially available
solar panels, which Sandia catalogues as module parameters for registered photovoltaic energy systems in
the catalogue library (typified by cell-type polycrystalline, dye-sensitised, multi-crystalline silicon, thin-film,
quantum dots, organic, perovskites), as the PVlib.bifacial routine also models bifacial modules, etc. (Scavo
et al., 2019; Tina et al., 2021b). The Sandia (2020) library further allows for the involvement of balance of
systems (BOS) components (solar micro-inverter, cabling, storage, transmission, distribution, etc.) to be
modelled in addition to the SAPM direct current model. In simplified terms, the DC to AC conversion factor
(Bosη) can also cater for workaround modelling (Pmac =Pm*Bosη) towards energy conversion and power
conditioning equipment (including a power inverter, alternating current power and systems losses).
While the following two sections detail the proposed environmental and economic model objects, it is
essential to note from the outset that environmental and economic models assume the optimised schedul-
ing of the energy-demand profile at the energy delivery site. Generally known as the optimal economic
dispatch in the scheduling supply-and-demand balances, this means that all the power generated by the
floating PV system computed above is, in fact, consumed on-site. This supposition is classed under Thesis
Assumption 2 (refer to Section 1.5), according to which the computer model synthesis for the environmental
and economic model objects provides for a breakdown of the environmental impacts and financials for a
future planned floatovoltaic project in terms of the most likely scenario for optimal energy usage and context.
4.3.4. Environmental simulation model object design
This section describes the FPV environmental model object with its membership functions in support of
sustainability and environmental planning processes. By bringing environmental understanding into the
trilogic picture of sustainability theory, the model is able to frame environmental contributions in an in-
tegrated environmental assessment facility that can quantify the technology’s benefits to the environment.
The environmental domain model object forecasts the operational environmental impact profile for a planned
floatovoltaic design regarding geospatial environmental offset metrics towards the WELF-nexus parameter
metrics (excluding the embodied impacts). The environmental model object accommodates the submodel
for solar weather and climate, that for floatovoltaic platform microclimate, for avoided water evaporation, for
avoided grid emission, that for land-surface reservation, and for soil-health carbon sequestration, each with
a description of its respective mathematical simulation and with provision made for the interactions between
the dynamic enviro-economic and techno-enviro components of the objects.
140
4.3.4.1. Model for solar weather and climate data
As a crucial part of the theoretical and experimental analysis of the performance of a floating photovoltaic
design, the resource model for the solar, weather and climate of an environmental object continually defines
the fuel (solar resource) for the floatovoltaic system. The parametric data for solar weather in Figure 4.3 are
generally obtained from a synthetic solar irradiance generator or numerical weather prediction assimilation
from a geostatistical data model, which delivers statistically-compiled data from historical data harvested
from meteorological weather stations and satellite data. The climatic data patterns (irradiation, cloud, rain,
weather data parameters) for any geographical case-study location are delivered in the form of the typical
meteorological year (TMY) data sets compiled from remote-sensing applications (at pre-selected climate
model resolution and weather prediction model tolerances) (McGuffie and Henderson-Sellers, 2001; Wag-
ner et al., 2009). In terms of applying solar resource data to solar energy projects, there are, in fact, specific
best practices for the collection and use of solar resource data for solar energy applications (Sengupta et al.,
2021).
Descriptive statistics for satellite-sourced, meteorologically-driven TMY time-series data are contextu-
ally sensitive to geographical location (altitude, latitude, longitude), and form a vital portion of the power-
generation sustainability of any PV or FPV project (Dellosa and Palconit, 2022; Holmgren et al., 2018). The
model downloads the TMY data from the solar resource and weather dataset repositories at given sub-
scription fees for the solar resource input data provider. Such supplier-based solar-simulator datasets may
typically include location-based time-series data for solar irradiation patterns (DNI, DHI and GHI) such as
cloud and clear-sky models, atmospheric cloud-cover prediction models, ambient temperature time series
(Tdry and Twet), relative humidity (RH), airmass (AM), and wind speed (Wspd) data series (HelioClim, 2022;
Sauran, 2021). Certain dataset providers also conduct applied research in meteorology and climatology to
publish extended time-series data for air-pollution meteorology, atmospheric aerosol properties, columned
ozone aerosol gases, cloud properties, atmospheric profiles, atmospheric pressure, surface albedo, snow
albedo, rain/snow precipitation, and index values as a proxy for cloud cover. In the geo-context of the
South African agricultural landscape, the forecast provider, SolarGIS, offers relevant downloadable resource
datasets compiled in collaboration with Stellenbosch University (SolarGIS, 2018) on a paid meteorological
data-subscription basis.
4.3.4.2. Model for floatovoltaic platform hydroclimate and microclimate
Getting to the heart of floating PV-system modelling, this environmental-microhabitat sub-model object de-
serves special attention. The modelling framework of this thesis considers localised environmental intelli-
gence as a fundamental concept in accurately characterising FPV technology behaviour within the aquatic
theatre of operation. While solar panels floating on water create an altered microhabitat, a virtual sensor
model is needed for reconstructing the hydroclimate and hydroclimatic conditions from a proxy (refer to
(Cahill et al., 2022; Liu, 2022)). It proposes a computational algorithm for the climatological modelling of
the hydroclimate induced by the floating PV microhabitat. In the conditioning of the PVlib model climate
parameters to suit the floatovoltaic hydroclimate bubble enveloping the FPV system reservoir, the model
estimates the ambient dry-bulb FPV temperature, the relative humidity and the wind speed variations from
knowledge of the floater mechanisms and local weather station data.
Starting with the temperature variations, the habitat-specific temperature scaling or a weighted temper-
ature offset offers a pragmatic solution to the modelling of the microclimate problem. In this approach, the
wet-bulb temperature, Twet (measured in the context of increased humidity and cooling in the same envi-
ronment), was logically regarded as the absolute extreme in the naturally ambient operational environment
for floating PV systems. As such, the proposed approach could exploit the dynamic wetbulb/drybulb tem-
141
perature references as anchor points, and then subsequently marginally scale the FPV ambient operational
temperature along the Tdry and Twet temperature differential scale with an adjustable ambient offset coeffi-
cient. Since the floater design type influences the airflow, temperature and evaporation rate (refer to (Bellini,
2021; Peters and Nobre, 2021)), it is posited that a set of floater design coefficients (FCα, FCβ, FCγ, FCδ)
can be specified by the floater manufacturer to reflect the extent to which the floater type impacts on the
modulated ambient microclimate of the floater through a set of custom-designed climatic floater coefficients
In this context, a proposed logical solution was derived through algorithmic means, whereby a modified
version of a polynomial function presented by Sullivan and Sanders (1974) is used to estimate FPV wet-bulb
temperatures from dry-bulb temperatures. The psychrometer’s wet-bulb temperature measurement gener-
ally reflects the quasi-equilibrium between evaporative heat loss and the heat gained from the surrounding
air. The Sullivan and Sanders (1974) psychrometric approximation formula Equation 1 can be used to deter-
mine the wet-bulb depression (difference between the dry-bulb temperature and the wet-bulb temperature)
when meteorological wet-bulb temperatures are not available (or when the wet-bulb moist dried up during
measurements).
∆T d(y,t)= 6.6+2T dry(y,t)−20
10 −3RH(y,t)−50
20 −T dry(y,t)−20
10 RH(y,t)−50
20 (1)
The proposed modified expressions to determine the ambient temperature difference for an FPV system
are given in Equations 2 to 4, producing a probabilistic reconstruction of the hydroclimatic variability that
impacts the dynamical FPV system behaviour and responses. The thesis designed these expressions to
estimate the ambient wet-bulb temperatures from the ambient dry-bulb temperatures to deliver a credible
FPV-specific meteorological-type solution to determine the ambient temperature surrounding the floating PV
system. As a modified version of the polynomial temperature function in Equation 1, the modified polynomial
function formulated by Equation 2 defines a set of climatic floater coefficients (FCa, FCb, FCb) to serve as the
proxy basis for a proposed temperature offset-based approach to estimating the FPV ambient operational
temperature. In Equations 2 and 3, the proposed FPV floater coefficients (F Cα= 0.5, F Cβ= 1.0) scale the
ambient temperatures according to the thinking described above. As such, the ambient temperature Tdry
(relevant to a GPV system), could be offset downwards by an adjustment factor, ∆Td, defined in Equation 2,
to condition the ambient operating temperature of the floating solar system marginally closer towards the
wet-bulb temperature point (Twet).
∆T d(y,t)= 6.6F Cα+ 2 T dr y(y,t)−20
10 −3RH(y,t)−50
20 −T dry(y,t)−20
10 RH(y,t)−50
20 (2)
∆T d′
(y,t)=F Cβ∗∆T d(y,t)(3)
f T ambient(y,t)=T dry(y,t)−∆T d′
(y,t)(4)
The expressions in Equations 2 to 4 realistically scale the relative ambient operating temperature of a
floating PV system through an adaptive adjustment factor (∆T d and ∆T d′) to imitate floatovoltaic opera-
tional environments. The expression in Equation 2 serves to estimate the ambient operational temperature
offset (∆Td) of the FPV system from routine weather station data as a function of relative humidity and
ambient temperature variations (even if the weather station’s Twet parameters happen to be incorrect or
unavailable). While the calculation to determine the ambient temperature adjustment or conditioning factor
in the FPV System is presented in Equation 2, the use of ∆Td in Equation 4 offers an empirical model
for estimating floating PV-module temperature on the minute or hourly time scales under time-varying field
conditions.
142
Furthermore, according to Equations 2 and 3, the proposed FPV floater coefficients (ambient-offset
coefficients F Cα= 1.0, F Cβ= 0.8) make provision for floater-specific coefficient factors respectively
(values [0.00 · · · 1.00]). These coefficients tune the ∆Td for the FPV ambient offset relative to the Tdry/Twet
scale as a function of the region (tropics, sub-tropic, arid area, semi-desert (Peel et al., 2007)) in which the
site is located, and also of the floater type and potential active cooling features (tubular buoyancy, canal top,
flexible float, entire cover (see (Scavo et al., 2020)). As such, the FPV floater substructure manufacturer
should define floater-specific ambient-offset coefficients and temperature impact specifiers in the future.
This approach may also include manufacturer specifications through a set of custom-designed climatic
floater coefficients (FCγ,FCδ) respectively to model the microclimate variations in terms of floater impact
on relative humidity as in Equation 5, and the floater impact on wind speed as in Equation 6.
fRH′
(y,t)=F C γ ∗RH(y,t)(5)
f W spd′
(y,t)=F C δ ∗W spd(y,t)(6)
One of this thesis’s strategic conclusions is that modelling the floating PV microhabitat in microclimatic
and hydroclimatic terms deserves much more attention in future research (see Section 6.7). For now, as
detailed above, the microclimate model associated with floating photovoltaic modules can confidently en-
gage the observed solar insolation and meteorological data to predict the expected operating conditions
of the floating PV system. While this section offers a set of dynamic microclimate parameters suitable for
plug-and-play in the PVlib library, the proposed microclimate model can be used to study the effects of the
microclimatic behaviour of PV panels under time-varying field conditions. Furthermore, the microclimate pa-
rameters are also used to determine the hydroclimatic impact on PV panel thermal model temperatures for
floating PV systems as a means to perform a more accurate energy yield assessment (refer Sections 4.3.3.3
& 4.3.3.4). Moreover, the parameters of the environmental object’s floatovoltaic platform microclimate sub-
model, defined in this section, are further recurrently exchanged with the energy production model (PVlib
model climate parameters) in Section 4.3.3.4, the microclimate-impact on PV panel thermal model in Sec-
tion 4.3.3.3, and the avoided-reservoir-evaporation model in Section 4.3.4.3.
4.3.4.3. Avoided-reservoir-evaporation model
From an integrated water systems research perspective, water preservation and hydrologic impact effects
are leading environmental cost drivers of floating PV systems (Armstrong et al., 2020; Farrar et al., 2022;
Khalifeh Soltani et al., 2022). Since floating PV directly impacts local water rights, freshwater budgets and
irrigation water requirements, accurate open-water evaporation and water-retention scenario modelling for
FPV technology are crucial. Such modelling involves reservoir simulations and water-surface modelling
toward developing a water evaporation model for installing floating photovoltaic covers. In the context of
the microclimate model detailed in the previous submodel, this part of the FPV model deserves special
attention in terms of evaporation modelling and predicting the water basin evaporation benefits emanating
solar screening through the installation of a floating PV cover over the water surface (Armstrong et al., 2020;
Cagle et al., 2020). Impacting the water economy, a floating PV system can effectively reduce evaporative
loss from areas of a limited area dedicated to water storage. However, that excessive experience insolation
is similar to how artificial monolayer materials impact micro-meteorological conditions and, in so doing, on
evaporation from water storage areas (Gallego-Elvira et al., 2013; Gebrekiros, 2015).
While surface evaporation rates from open water surfaces have been studied and parameterised in the
literature through the application of different models, to theoretically quantify the counteracting effects of
evapotranspiration from floating PV systems on freshwater sources means to estimate the evaporation rate
143
in a water reservoir partially covered by a floatovoltaic power plant (Scavo et al., 2020). Floatovoltaic en-
ergy systems installed on open-air water reservoirs reduce water evaporation from their below-the-surface
reservoirs by reducing thermal airflow rates and absorbing radiation ordinarily absorbed by the reservoir
water. A floating PV system thus offers attractive environmental resource conservation properties in terms
of water preservation and improvements in water quality of the water (Spencer et al., 2019; World Bank,
2019b). However, an estimate of the reduction factor in the evaporation losses still needs to be determined
(Majumder et al., 2021).
To support the experimental assessment of evaporation rates associated with water basis accepting
floating PV installations, the environmental model in this thesis was found to account for avoided on-site
water evaporation (eH2O) in a modified floatovoltaic evaporation model for flexible floats, as proposed by
Scavo et al. (2019, 2020), was used. The research group in question adapted the Penman-Monteith equa-
tions to derive a linear regression formula, whereby routine weather data is used to estimate the evaporation
rate in water basins partially covered by the FPV system based on the daily solar irradiation factor Rsday
(Scavo et al., 2020). In their mathematical expression, given in Equation 7, the daily evaporation rates for
suspended photovoltaic covers in free-water basins are primarily based on the horizontal solar irradiation
flux (the energy flux rate incident on the water surface) with the relative humidity and the wind speed at the
floating PV system installation site. The model expression in Equation 7 thus expresses the direct estimated
volume of water evaporation avoided by an FPV cover.
Emm/day = 2.421 + 0.012Rs/day + 0.159T dry/day −0.056RH/day + 0.122W spd/day (7)
However, to determine the hourly evaporation rate (eH2Ot), the evaporation equation in Equation 7 for
daily evaporation needs to be adapted to use hourly solar irradiation parameter samples obtained from
the meteo-sensor data. Certain intuitive normalisation factors needed to be developed as an interpolation
method. Interpolation is required to fit in with the thinking behind the simulation model’s hourly resolution,
thus modifying Equation 9 to express the hourly preserved evaporation for an FPV system as an FPV per-
formance metric in mm/hour (or litres/m2/hour). Firstly, Equation 9 applies a W/m2/hour to M J/m2/hour
conversion factor (W/m = J/s -> MJ/hour -> W/m x 60sec x 60mins / MJ = W/m x 0.0036) in the Python
simulation, to engage the hourly global horizontal irradiation on water data (GHIhour) in the place of the
daily Rsday parameter of Equation 7. Secondly, the total daily horizontal solar irradiation factor Rsday in the
daily evaporation equation in Equation 7 is multiplied with a solar daylight factor (ts) to convert the radiation
incident GHI’s MJ/m2/hour into M J/m2/day, Thereafter the total evaporation normalised with the same
daily sunlight hours factor (ts) to get the average hourly evaporation per square meter (eH2Otin mm/δt).
This procedure offers an improved floating PV evaporation model compared to the first version documented
in (Prinsloo et al., 2021).
eH2O′
(y,t)=2.421 + 0.012 ∗GHI(y,t)∗0.0036 ∗ts + 0.159 ∗(T dry(y,t)+ 0.5)
δts (8)
eH2O(y,t)=eH2O′
(y,t)+−0.056 ∗RH(y,t)+ 0.122 ∗W spd(y,t)
δts (9)
The evaporation simulation model for FPV (given by Equation 9) includes a temperature factor to repli-
cate field measurements in terms of energy budget dynamics, wherein the modelled surface water tem-
perature conditions show an average increase of around 0.5◦C when compared to uncovered open water
temperatures (Yang et al., 2021). These field measurements compared open-water conditions for a shallow
tropical reservoir. The altered solar flux beneath floating PV panels generally results in a slightly higher equi-
librium temperature near the water surface. The proposed modified real-time simulation model in Equation 9
thus estimates the hourly floating PV reservoir evaporation as a sampled data time series, with a discrete-
time sample rate equal to the sampling frequency of the weather station data (samples taken by the hour
144
and the minute, etc.). The adaptation towards an hourly temporal resolution reflected by the expression
of Equation 9, is different to the Scavo et al. (2020) model in that the fraction normalises the daily pan
evaporation rates to hourly evaporation rates if the hourly solar and climatic data parameters are used in
the numerator. As such, the fraction equation for evaporation normalises the model’s predicted evaporation
input and output data concerning the FPV model’s simulation frequency period (δts), relative to the 24-hour
daily evaporation period. The time-normalised expression thus predicts the avoided evaporation of an FPV
reservoir as a time series (expressed in mm/δt or litres/m2/δt equivalent) from weather station climatic data
variables (irradiance, ambient air temperature, relative humidity, wind-speed). Overall, the eH2Ometric
reflects the water preservation potential of an FPV, which not only reflects the avoided water evaporation
thanks to the islanded floating PV canopy, but also reflects the quantitative amount of water resources saved
to supplement groundwater resources underground.
4.3.4.4. Atmospheric impact model
Amidst the global climate emergency, there is a growing interest in engaging climate science and atmo-
spheric science principles in atmospheric impact modelling as part of the energy, water, land and atmo-
spheric interactions and feedback modelling. The aim is to characterise environmental drivers and as-
sociated atmospheric simulations, including the interactive representation of tropospheric aerosols and
atmospheric chemistry (Collins et al., 2017; IEA, 2021). As an atmospheric chemistry transport model,
the environmental offsets of floating PV imply that the technology can act as a catalyst for environmental
remediation. In reducing the coal-based grid pollution load, the technology makes producing clean, renew-
able energy possible and offers a range of beneficial atmospheric offsets and avoided pollution benefits
important to the carbon science field (Prinsloo, 2019).
For an FPV model to quantify the environmental offsets of a floating photovoltaic project (at optimal
economic power dispatch condition, assuming all power consumed as per research Assumption 2), specific
diagnostics need to make future projections about the chemical composition of the avoided grid pollution
quantities. Regarding environmental economics and environmental footprint improvement parameters, en-
vironmental offset predictions are significant to our country’s national air pollution management strategy. In
grid-substitution applications, estimates of variations in air pollutant concentrations and other FPV technol-
ogy impacts capable of counteracting the environmental price paid for the emissions issued from grid-power
generation are critical. It is crucial to model FPV’s climate pollution reduction impacts, as it counteracts the
anthropogenically-sourced atmospheric emissions, ambient air pollution and the chemistry of pollutants
dispersed into the atmosphere through the burning of fossil fuels by grid providers (Prinsloo, 2019).
In the context of this clean energy production, the floatovoltaic grid substitution impact model for the en-
vironmental object in question predicts the environmental offsets for a future planned floating solar energy
system, with the focus on country-specific resource conservation, the mitigation of atmospheric emissions
and improvements in air quality (Prinsloo et al., 2021). With the FPV philosophy supporting the just transi-
tion to clean renewable energy, the attention is focused on an unbiased analysis of fossil fuel and natural
resources in terms of their chemical and physical stressors and their impacts on humans and the environ-
ment (EPA, 2022). Concentrations of atmospheric pollution through aerosol emissions impact air quality
standards, as some chemical compounds are the primary components of air pollution. The main reason
behind this severe problem is that some react with volatile organic compounds in the presence of sunlight
and contribute to forming an ozone layer at ground level. Others play a significant role in forming smog and
acid rain (Prinsloo, 2019).
Exposure assessment modelling considers environmental offset attributes in terms of avoided sub-
stance emissions, the reduced use of coal fuel, declining accumulations of ash waste, and water conserva-
tion, thanks to the energy substituted by FPV systems to reduce dependence on water-cooled, coal-fired
145
power station electricity. Due to the dominance of fossil fuel combustion in grid power generation in South
Africa, energy production offers a reliable proxy for a large portion of the environmental impacts of interest
in Table 4.1 (DFFE, 2021b). The mathematical model for the environmental subsystem impact calculations
incorporates SA’s national air quality model, providing data typically used in decisions about emission re-
quirements recorded in Eskom’s "Carbon Monitoring System”. The country-specific substance equivalence
factors are detailed in Eskom’s model for grid-substituting environmental-impact calculations, as presented
in Table 4.1 (DFFE, 2021b; Eskom, 2017; SAIT, 2020). By engaging these factors, the environmental
subsystems model accounts for the fuel-reform impacts of the floatovoltaic solar-powered generator, es-
sentially quantifying the environmental rider impact benefits (accelerator benefits) that would emanate from
any proposed floating solar system to be incorporated in South Africa’s national environmental system.
Table 4.1: Eskom country-specific impact factor offsets defining environmental implications for generating or using or
saving grid electricity in South Africa (source: Eskom (2017), page 2).
According to the green chemistry parameters depicted by the organic element compositions of the cli-
matic hazards in Table 4.1, the mathematical environmental profiling of the thesis model characterises the
country-specific substance-equivalence factors attributed to the grid-provider model using country-specific
grid-substituting environmental impact factors (DFFE, 2016; Prinsloo, 2019). As such, the selected environ-
metric indicators attributed to floatovoltaic energy-yield performance profiles are articulated as grid-avoided
aerosol emissions of carbon dioxide offsets (xCO2), sulphur-oxide offsets (xSOx), nitrogen-oxide offsets
(xNOx), airborne-emission or particulate-matter offsets (xPEa), reduced coal-fuel offsets (xCoal), avoided
ash-waste offsets (gAsh), as well as the preservation of national water resources (xH2O) as an offset. The
predictions for the reduction in the atmospheric emission of carbon and other volatiles are respectively cal-
culated through Equations 10 to 16, for which the associated multiplier impact factors or functions (xfacCO2,
xfacNOx, etc.) are defined as volumetric weights in the FPV model’s user configuration file.
xCO2(i,j)=P m(i,j)∗xfacCO2∗10−3(10)
xH2O(i,j)=P m(i,j)∗xfacH2O∗10−3(11)
xSOx(i,j)=P m(i,j)∗xfacSOx ∗10−3(12)
xNOx(i,j )=P m(i,j)∗xf acNOx ∗10−3(13)
xP Ea(i,j )=P m(i,j)∗xf acP E a ∗10−3(14)
xCoal(i,j)=P m(i,j)∗xfacCoal ∗10−3(15)
xAsh(i,j)=P m(i,j)∗xfacAsh ∗10−3(16)
146
In the case of floating PV installations on farms in South Africa, the outputs from this model object spec-
ify predictions concerning the environmental offsets and impacts, especially the carbon offsets measured in
tonnes of carbon dioxide equivalent (CO2e). With environmental impacts quantified in terms of the physical
and/or chemical compounds in question, and based on the energy and environmental economics detailed
under the data requirements for calculating the Carbon Emission Factors (CEF) for the South African grid
(Eskom, 2021), these parametric indicators are associated with the environmental-impacts regime and pro-
tocols of the national power generation utility (Eskom, 2017, 2021). The empirical outcomes quantitatively
establish the environmental offset mitigation (local and national environmental-impact profile characterisa-
tion) for a floating solar system (or land-based solar system) in terms of Eskom’s parameter narratives for
grid-power substitution (DFFE, 2021b; Eskom, 2021). The above calculations provide scientific evidence (in
terms of NEMA legislation (RSA, 2008)) concerning the projected ability of the installed system to mitigate
the effects of the power generation process. Eskom cools her coal-fired power stations through copious
freshwater usage, which in turn results in significant volumes of wastewater; land-use changes; the ne-
cessity for environmental remediation, atmospheric aerosol loading, biochemical flows, and the dispersal
of pollutants and contaminants into the national atmospheric and ecological environmental systems in the
national South African region.
4.3.4.5. Land-surface-reservation model
In dealing with methods to mitigate negative land-use issues, the efficient use of land to meet sustainable
energy needs is a vital offset indicator in respect of environmental impact studies (Hernandez et al., 2014;
Kim et al., 2021b). By erecting FPVs on water surfaces, the FPV project saves/reserves land surfaces for
other applications. Regarding the impact metrics of FPV projects, this requires metrics to reflect the land-
surface-saving or land-surface-reservation benefits of floatovoltaics in terms of the direct water surface area
covered by PV panels. Determined in respect of the PV plot size or the actual site or geometric area of the
water surface covered by FPV panels on a recipient water reservoir (Afpv in m2), this metric is defined as
the nominal area of farmland preserved by an FPV project (n-FLp). However, scientists have recognised
that an essential factor, namely, the land-gearing factor, that is at play in preserving actual real farmland
(n-FLp) should be determined in terms of the energy output of an FPV system relative to a GPV system.
Given the land-use intensity of solar power production and tomorrow’s energy landscape (Lovering
et al., 2021), the FPV land-sparing model estimates the effective number of farming hectares saved by an
FPV system in terms of land required by an equivalent GPV system to realise the same power generation
capacity. In this approach, the FPV/GPV generation efficiency gain ratio (fgPmη) is of key importance
when determining the area in square metres of farmland required for a GPV system to generate an amount
of energy equal to the capacity of the FPV system under consideration. As such, a farmland gearing
ratio (FLg) is formulated in Equation 17 to equate the power generation of an FPV efficiency improvement
with its land-sized equivalent and relative to the total water area supporting the site of the FPV on the
recipient water body (Afpv). This estimation enables the model to formulate the expression in Equation 17
to estimate a generation-based farmland-preserving indicator (FLp) from the geometric water surface area
(Afpv) covered by the plot area (in m2) of a floating PV system multiplied by the land-gearing ratio (%) and
divided by a space-loss coefficient (FLgϵ=0.10 or 10%). The loss coefficient of the productive farmland
space (agronomic land-loss coefficient) (values [0.0, · · · , 0.95]) makes provision for additional sacrifices in
agronomic land space in terms of access roads to the GPV, open peripheral firebreak zones around the
GPV system, underground cabling and access routes to the GPV).
F Lg = 1 + PP m(Y ,T )
PgP m(Y ,T )∗(1 −F Lgϵ)(17)
147
F Lp = (Af pv ∗F Lg )(18)
While the Afpv variable defines the geometric water surface area covered by floating solar panels, the
FLg metric gives insight into the land equivalent ratio for a comparable PV system of the same production
capacity (considering the enhanced efficiency of an FPV system).
The FLp metric then defines the ground coverage (ratio) of an equivalent land reclaimed by an FPV
system (compared to a GPV installation). This FPV land-sparing indicator FLp in Equation 18 as such
portrays the predicted area (in m2) of farmland required (food crops, fruit orchards, vineyards to be uprooted)
to install a ground-mounted PV system of the same power-output capacity as an equivalent FPV system,
which geometrically covers a water surface area (Afpv) of square metres on the reservoir.
4.3.4.6. Soil-health model for carbon sequestration
From an ecosystem farming perspective, low-impact floating solar developments can indirectly improve soil
health, biosphere carbon storage, preserve soil micro-biomes and nurture native species, supports land re-
habilitation, help to produce more food, and deliver cleaner energy at a lower cost (Armstrong et al., 2020;
Exley et al., 2021b; Torhan et al., 2022). With sustainable soil amendment, organic matter and agronomic
plant life cultivated on farmland around water reservoirs also sequester a significant portion of inflowing
carbon during plant respiration to help reduce the CO2efflux rates by slowing the carbon cycle and delay-
ing the return of carbon to the atmosphere (Pretorius, 2020; Scandellari et al., 2016). Thus, the preserved
farmland factor (FLp) has further dimensions to consider in support of the natural restoration of farmland, as
floating PV projects more readily allow for natural land regeneration in support of regenerative farming. As
such, floating solar energy development can impact the soil health and carbon sequestration benefits ema-
nating from the land-cover change on farms and in protected areas (and indirectly on the volumetric water
content soil porosity) (Hernandez et al., 2015; Prinsloo, 2019). Furthermore, towards accounting for the
potential role of rescued cropping and vineyards in mitigating the impacts of climate change, FPV systems
further support sustainable food production. For example, floating PV systems circumvent microclimatic
interferences in orchards and vineyards, such as the ripping out of healthy fruit trees or vines or detrimental
intrusions into productive vineyards to install GPV systems. While a long-term commitment to protecting the
vegetation and managing soil health is essential, FPV systems can contribute to the ecological economics
and sustainability positives of reduced soil erosion and greenhouse gas emissions from agricultural soils.
As such, the contribution of FPV systems to soil health assimilates in the carbon cycle, hydrological cycle,
and nutrient cycle can help to ensure healthy farmland soils with the continued capacity to function as a
vital living ecosystem able to sustain plants, animals and humans (Attar et al., 2016; Decock, 2020).
In terms of one of the impacts of the land-sparing benefit on the farm’s biodynamic carbon budget
(terrestrial C cycle) for vineyards, the study termed a new metric, known as the rescued-plant atmospheric-
carbon sequestration into green pastures, grapes and cover crops (pCO2). This impact benefit is predicted
in terms of the crop-location specific nett CO2sequestration capacity by Equation 19. The CO2sequestra-
tion is specified by the vegetation-territory emissions factor (pfacC =0.3-0.6 kg/m2/yr), typically determined
by carbon-budget studies of orchards (refer to (Decock, 2020; Pretorius, 2020; Scandellari et al., 2016)).
pCO2=F Lp ∗pf acC (19)
By not altering the orchard or vineyard microclimate, the CO2(and N2O) sequestration potential (2.8 kg/m2/yr
for grapevines above- and below-ground biomass preservation) for both the healthy soil and the perennial
crop vegetation is preserved (Nistor et al., 2018). The atmospheric carbon-sequestrated coefficient (pfacC
= 2.8 kg/m/yr) in the plant pCO2 calculation for the cover crop is a specific location-cropping factor that
accounts for at least the climate-mitigating loss factors of the avoided bare soil and green vegetation. This
148
factor is because the ground soil is actively sequestering carbon, especially in regions identified for solar
PV development that are known to be significant carbon sinks (Tavassoli and Hernandez, 2012). Since the
rescued amount of carbon sequestration resulting from the avoided uprooting of plants or green vegetation
is a function of the type of plant life, this factor should be determined precisely with the farmland to be
uprooted to install a conventional PV system of size (FLp) around the same area as a floating PV system.
In the best interests of conservation stewardship, numerous opportunities exist to model further such
downstream co-benefits around the potential role of vineyards in farmland-preserved areas with the imple-
mentation of climate change mitigation measures in wetland and dryland agricultural systems. Scientific
biochemistry and ecological research around the definition of healthy vineyard soils and soil health man-
agement offers valuable information about the carbon sequestration potential of orchards and vineyards
and their potential to sequester atmospheric carbon (Lazcano et al., 2020; Scandellari et al., 2016). While
outside the scope of this thesis, FPV systems may, as such, affect a range of multivariate assessment
indicators towards biological, ecological, and environmental soil health indices or indicators. These may
include, for example, crop yield, organic carbon, cation exchange capacity, soil texture, soil respiration,
microorganism activity, electrical conductivity, and nutrient and hydrological cycles (Cardoso et al., 2013;
Goodwin, 2018).
4.3.5. Economic simulation model object design
Regarding the economics of floating PV technology, this section describes the characterisation and econo-
metric analysis performed by the economic model object. By bringing the economics of climate health into
the trilogic picture of sustainability theory. The model can connect environmental and economic changes
and incorporate climate economy membership functions that support business analytics and the associ-
ated financial planning processes for floating PV systems. In economic terms, an appropriate cost-benefit
analysis helps to gain insight into the project value proposition and to gauge the financial investment wor-
thiness of the floating solar project. Furthermore, a combined economic and environmental economics
cost-benefit analysis becomes essential in a small-scale application of an embedded power generation en-
terprise. It supports the sustainability business case through green economy indicators. The economic
model is designed to accommodate financial modelling with system dynamics, using feedback economics,
climate change economics, feedback environmental economics, and feedback water economics with the
conventional parameters in Figure 4.8 towards environmental modelling with system dynamics.
Figure 4.8: Cost component and sensitivity analysis (in US $) for internationally evaluated floating solar PV projects
(source: World Bank (2019b), page 112).
149
By incorporating research findings on energy and environmental finance as business analytics indi-
cators, this financial model object focuses on the metrics of microeconomic impact analyses. As such,
this domain object predicts the economic performance of future planned floating solar stations in terms
of its geo-contextually-sensitive financial parameter metrics. The economic evaluation model focuses on
the business intelligence case for floating PV systems whereby the cost-modelling methodology defines
the mathematical functions to describe the operations of the model object. By applying mathematical eco-
nomics algorithms, the model provides the dynamic enviro-economic and techno-economic interactions
relevant to the agricultural context and the associated economies of scale. This model object accommo-
dates the economic-performance-scoring model, the income-revenue-streaming model, project profitability,
the agronomic land-gain revenue/value model, and the optional agrivoltaic/aquavoltaic production income
value model, each with its respective description details of the relevant mathematical simulation.
For the cost figures for floating solar systems, and considering the related proprietary and confidentiality
constraints, installers such as New Southern Energy Africa (Pritchard et al., 2019) have been approached
to document cost/watt-peak figures for the local SA FPV systems context. The renewable energy analysis
techniques and methodologies recently devised could thus offer a better understanding of the optimisation of
the design and planning problems commonly related to installation (Barbuscia, 2016; Prinsloo, 2019). The
broad spectrum of practical and modelling experiences worldwide offers a valuable base for the present
research in developing methodologies and factors to consider in a decision-support system.
Regarding financial modelling, the World Bank (2019b) Report, as presented in Figure 4.8, provides
valuable cost breakdowns and guideline benchmark figures with sensitivity parameters for international FPV
projects. From an economic perspective, the financial figures in Figure 4.8 provide benchmark figures to
guide model configurations in decision-support studies in international floating solar systems. The guideline
financial figures in Figure 4.8 also provide valuable insights into the case of economic investment in FPV,
as the configuration parameters can guide decision-makers or investors around the justification towards
technology diffusion and the adoption of clean-technology-type floating solar methods (NREL, 2019a; World
Bank, 2019b).
4.3.5.1. Economic performance-scoring model
Econometric scoring aspects in solar system modelling define the cost of generation and financial supply-
chain modelling parameters for predicting the economic performance profile for a planned floatovoltaic
system in terms of empirical economic indicators. These economic determinants of floatovoltaic projects act
as financial indicators to portray project due diligence and project bankability through the main cost-driving
financials, applied economics ratios and metrics as monetary value indicators.
Such cost considerations of FPV systems are related to the capital expenses, cost of capital and energy
production costs of FPV projects as cost per unit of electricity or generically-levelled electricity tariffs for
energy accrued over the design life of the floatovoltaic power plant. The instituted set of economic param-
eters or econometric indicators for the FPV cost model includes the pre-tax base-levelled cost of electricity
(LCOE), nett present value (NPV), return on investment (ROI) and the internal rate of return (IRR). In ad-
dition, a future version of the model could make provision for Production Tax Credits (PTC) and Investment
Tax Credits (ITC) which are financial offset measures applicable to some countries abroad. In addition,
project financiers depend on project dynamics, such as third-party ownership, the renting of water space,
or considerations regarding retail electricity rates.
The key economic indicators in the digital economy of floatovoltaic systems start with the Levelised Cost
of Electricity (LCOE) metric, which predicts the effective cost of electricity generated by the floatovoltaic
system (Prinsloo et al., 2021; Sutterlueti, 2015). By international convention, the illustration in Figure 4.9
presents the floating photovoltaic system cost benchmark factors considered in determining the LCOE
150
(Ramasamy and Margolis, 2021; Sutterlueti, 2015). As such, the financial formula for calculating the LCOE
is given as the (present value of the total lifetime costs of the project)/(present value of the aggregated
electricity generated during the lifetime of the project) (Short et al., 1995). According to Figure 4.9, the
inflation-discounted LCOE calculator is implemented through the numerical expression in Equation 20 (in a
currency value per kWh).
Figure 4.9: Factors to be considered in LCOE cost determinations (source: Sutterlueti (2015), page 26).
The actual monetary/energy values are inflation and the discount rate decremented to the present year
values (over the sustained lifetime operation of the system). Considering energy costs, Equation 20 de-
termines the LCOE as a financial capital recovery metric to evaluate project viability and calculate project
break-even periods and margins.
LCOE =PY
y=0 (fCpvy+fCrmy+f I cyf Omy)∗Iy/Dy
PY
y=0 (1 −δy)y∗P m(y,T )∗Iy/Dy(20)
where:
Iy/Dy=(1 + inf ln)y
(1 + disconto)y(21)
The variables in the above equations include the area-unitised FPV investment costs per year (Capex
toward the floatovoltaic asset inventory), costs including general sales tax and value-added tax GST/VAT).
It defined the total PV system price as PV solution Capex (fCpvy=600.00 R/m2), plus the racking/mooring
solution (fCrmy=300.00 R/m2); the financing cost of capital, loan interest, bank commitments; leased-
water-space rental to install the FPV (fIcy=0); the installation’s annual variable costs (Opex expenditures,
operational and maintenance costs including insurance charges) per unit area (fOmy=30.00 R/m2/yr); the
prevailing inflation rate (infln =3.5%); the project discount rate (disconto =7.5%); and the accrued sum of
all the electricity generated per unit area for every year y (Pm(y,T )in kWh/m2). Considering the annual
losses (δy) in terms of fPVd, this factoring includes losses such as soiling, shading, snow, mismatch, wiring,
connections, nameplate rating, and age degradation.
These values are also computed for the complete geometric area (Afpv) over the project lifetime or the
life of the floatovoltaic system (Y =20yrs), as specified in the model’s configuration file. While up-front costs
are estimated to be slightly higher for a floating solar system than for a traditional ground-mounted system,
the floatovoltaic costs over time are at par with those of the ground-based system because of the higher
energy yield of floatovoltaic systems on account of the cooling effect of water.
Opinions still vary about the all-in-all systems cost variations (planning, commissioning and installation
cost variations) between the utility-scale and agricultural-scale costs. The prevailing view dictates that the
costs of the system for both land- and water-based systems are supposed to be on par, except for a Capex
cost disparity between the project-approval processes, as well as the steel-framing and floating-pontoon
support structures and associated civil works (Cox, 2019a; Verbruggen, 2018). However, the all-in-all costs
for the construction and installation of the floatovoltaic system on an agricultural scale are often slightly lower
151
due to the opportunity for local innovation (Allen and Prinsloo, 2018). For utility-scale systems, structural
floatovoltaic costs are generally higher due to the manufacturing costs, the import logistics and expenses of
proprietary floating-pontoon modules (Ciel et Terre, 2022; Gazdowicz, 2019; Pritchard et al., 2019).
The Levelised Avoided Cost of Electricity (LACE) is the second economic performance indicator pre-
dicted by the FPV synthesised model. The LACE metric gives project developers an indication of the
relative savings value or marginal benefit of FPV energy generation relative to the grid costs (in the case
of grid substitution) in terms of electricity tariffs (the nett-metering value of a renewable energy project to
the grid, if a feed-in tariff scheme were to be available). While there are many formulations to compute
LACE, in the base model, it is determined through the simplified expression in Equation 22, where the grid
tariff (xTr =1.875 R/kWh, daytime dynamic time-of-use rating) reflects the surplus of the prevailing retail-grid
electricity tariffs avoided. A negative value reflects the marginal energy bill cost savings per kWh, indicative
of the excellent bankability of an FPV project. Note that, while the model processes data on a discrete-time
base, the model can in future make provision for complex time-of-use tariff ratings. landowner application
of water-space zoning regulations for farming-scale floating solar.
LACE =LCOE −xT ry(22)
The third economic performance indicator is the predicted project-investment breakeven point (Qbe) in
years, computed from the LCOE through the breakeven formula in Equation 23. This equation factorises
the LCOE nett current tariff value with the current grid tariff rating that was avoided as a function of the
project’s lifetime.
Qbe =YLCOE
xT ry(23)
Note that the study chose the universal Qbe pay-back computation to align with the financial indicator
norms. However, a floatovoltaic formulation may also include carbon credits and water-saving and land-
saving incomes to present a more appropriate or realistic income stream specific to floatovoltaic technology.
This study project also elaborates on the model calculations for Nett Present Value (NPV), Internal Rate of
Return (IRR), Discounted Cash Flow (DCF) and Asset Depreciation Costs (ADC).
Qbe =LCOE
W SE t (24)
The fourth economic performance indicator calculated in this submodel is the Nett Present Value (NPV) of
the FPV project. This financial indicator is simply computed as the income from the denominator minus the
numerator of Equation 20. The NPV financial metric generally captures the predicted current value of the
potential investment opportunity (floatovoltaic project) as a function of the inflation rate and the expected
rate of return or project discount rate. As a measure of project profitability, the universal NPV calculation in
Equation 25 represents the energy-related time value of money as the future FPV project cash flows from
energy are discounted at the given discount rate. The Nett Present Cost (NPC) component is the current
value of all the costs of installing and operating the FPV system over the lifetime of the project, and is given
as the numerator of Equation 20, as computed by Equation 26.
N P V = Y
X
y=0
xT ry∗(1 −δy)y∗P m(y,T )∗Iy/Dy!− Y
X
y=0
(fCpvy+fCrmy+f I cy+f Omy)∗Iy/Dy!
(25)
N P C = Y
X
y=0
xT ry∗(1 −δy)y∗P m(y,T )∗Iy/Dy!(26)
152
The Return On Investment (ROI) percentage factor is an important parameter used in profitability assess-
ment and the project sustainability priority ranking of a floating PV installation. The ROI is determined as
the ratio of the electricity income (or electricity plus land-saving agro income) relative to the initial installation
investment (year zero) into the PV project, as given in the ROIycalculation provided by Equation 27.
.ROIy= 100 ∗N P V
fCpv0+fCrm0
(27)
Next, the Internal Rate of Return (IRR) for the FPV project is determined as the measure of the profitabil-
ity of the FPV system over the project’s lifespan. The higher the IRR of the FPV project, the more profitable
the installation. As such, the IRR represents the discount rate that makes the FPV project profitable, the
same discount rate that makes the sum of all cash flows over the project lifetime equal to zero. IRR is thus
determined by solving for X in Equation 28.
N P V =
Y
X
y=0
xT ry∗(1 −δy)y∗P m(y,T )∗Iy/Xy
−
Y
X
y=0
(fCpvy+fCrmy+f I cy+f Omy)∗Iy/Xy= 0.00
(28)
where:
Iy/Xy=(1 + infln)y
(1 + X)y(29)
The projected discounted cash flow (DCF) is determined as the worth of the FPV over the year y’, given the
projected income and expenditure streams, specified interest rate and the rate of return. Through the CF
calculation in Equation 30, future FPV project cash flows are discounted at the discount rate to reflect the
cash at hand up to the year y’.
DCFy′=
y′
X
y=0
xT ry∗(1 −δy)y∗P m(y,T )∗Iy/Xy
−
y′
X
y=0
(fCpvy+fCrmy+f I cy+f Omy)∗Iy/Xy
(30)
The discounted cash flow analysis is thus a valuable method for evaluating the FPV asset or FPV enterprise
asset at any point in the project’s lifetime. In terms of Asset Depreciation Costs (ADC), the year zero
investment costs (fCpv0+fCrm0) minus the residual or reserve value of the asset (Cres) over the useful
lifetime of the asset (Ult) determines the straight-line depreciation for the FPV power plant, as calculated
through Equation 31. Generally, asset depreciation costs (ADC) in Equation 31 are considered an expense,
reducing the project’s taxable income.
ADCy= (f C pv0+f Crm0) (1 −Dr ∗y)(31)
From a business innovation perspective, project financing for renewable energy in BRICS countries
includes options for the long-term leasing of unused water bodies and water reservoirs. This makes solar
leasing or solar rentals for turnkey floating solar solutions a viable commercial floatovoltaic technology option
(World Bank, 2019b), an aspect that can be accounted for as an added running expense.
153
4.3.5.2. Economic income revenue streams model
The economic revenue submodel predicts the income revenues of the floatovoltaic project in cash-flow
projections for at least three income streams, namely, the electricity resource revenues. First, it includes
avoided electricity purchases (Rgs), reducing the out-of-pocket expenses for grid electricity as the self-
generated power is exchanged for grid power. With Rgs revenue based on the prevailing utility rate struc-
tures, it is considered a self-service type of on-demand electricity consumption by the farming prosumer at
the prevailing generator selling price. Income further includes tradable environmental levy incentives or rev-
enues such as e-Tax or eco-Tax carbon credits (Rcc) and renewable or solar-Tax credits (Rtc) as a monetary
value to justify the saving of avoided carbon emissions. In addition, the value of the preserved evaporative
water resources (Rwe) is accounted for as avoided purchases or water-pumping costs, the income from
agri-fresh produce or fresh food produced by agronomic crop farming (Rxp), and commodity sales from
processed fresh food (Rcs) such as wine or olive oil. In combination, all FPV application-specific income
streams (Rzt), agrivoltaic farming activities, aquaculture farming-income activities (aquavoltaic aquatics),
waste-water processing and treatment income, increased hydropower generation (from water savings), and
other income streams or values associated with the sphere of floatovoltaic installations or land-sparing,
floating solar island and resource utility impact attributes.
Starting with the grid-substitution applications modality, the parameter, Rs, for the floating PV revenue
model reflects the avoided cost of electricity purchases from the grid supplier. An avoided grid-electricity
purchase constitutes a type of electricity bill calculation to determine the deemed energy sales income
according to Equation 32. The electricity grid-price policy tariff (xTry=1.465 R/kWh) and the dynamic tariff
escalation growth rate (xTre =15.09%) can be read from the model’s user configuration file.
Rgs(y,t)=P m(y,t)∗10−3∗xT ry∗(1 + xT re)y(32)
Note that the factor xTr can be used to model crypto-currency tie-in, by substituting xTr with cryptocurrency
units for tapping into other energy-trading or solar-sharing marketplace narratives (i.e. IPP grid feed-in-tariff
FIT, nett metered credits, export power wheeling, cooperative micro-grid sales, cryptocurrency transfers,
blockchain wallet units, or related e-commerce currency denominations).
The next revenue parameter (Rcc) is the value of environmental impact in terms of the avoided effec-
tive CO2e(carbon offsetting), which involves funding a low carbon-energy project through carbon credits or
carbon trading. Carbon-offset income offers a financing mechanism to compensate a floating PV project
for reduced emissions by funding equivalent carbon-dioxide emissions elsewhere. Regarding carbon ac-
counting (say in the wine economy), the carbon-credit income model defines the potential carbon credits the
tax beneficiary trades as project-revenue income (Rcc =Rcc1+Rcc2 · · · Rcn). This carbon-offset income
(Rcc) from CO2credits is obtained by grid-substitution emission reduction (xCO2) and carbon sequestra-
tion through the avoided crop disturbance (pCO2) of perennial cover crops/vineyards on preserved farmland
(FLp), and defined according to the expressions in Equations 33 and 34. The environmental-levy cost factor
(Cty=144.00 R/ton) is specified as a factor of the carbon-credit pricing value at which carbon credits can be
sold on the country’s open market (the positive economic consequences of counteracting climate change)
(RSA, 2019a). This carbon-offsetting income factor and its growth rate are defined and read from the model
configuration file and its dynamic annual escalation (Cte =2.00%).
Rcc1(y,t)=xCO2(y,t)∗10−3∗Cty∗(1 + Cte)y(33)
Rcc2(y,t)=pCO2(y,t)∗10−3∗Cty∗(1 + Cte)y(34)
The third income metric, considered floatovoltaic water preservation income, is defined as the value of
water (Rwe). This indicator refers to the value of on-site water savings attributed to FPV in terms of avoided
154
water purchases or avoided-water pumping costs. The water revenue model determines the value of this
environmental impact in terms of avoided on-premise evaporation (eH2O), considered a resource-saving
income according to the expression in Equation 35. The water consumption tariff or water-pumping cost
factor (Wty=0.027 R/ltr) is the value of the reservoir water preserved, a factor specified in the model’s
configuration file, with the anticipated annual escalation (Wte =10%) in the water tariff/cost.
Rwe(y,t)=eH2O(y,t)∗W ty∗(1 + W te)y(35)
The fourth income metric is the total floatovoltaic income revenue (Rzt) generated from all the sources of
FPV’s application-specific activities, such as agronomic income, aquatic farming income, water treatment
income, added hydropower income, or similar revenue streams. Added income revenue (Rzt) includes
agronomic income (Rxp), described in the next sub-model, comprising the value of the preserved production
capacity (Rxp) of the farmland as a function of the agricultural commodity and profitability per bearing
hectare of fruit trees/grapevines.
4.3.5.3. Agronomic land-gain value model
With land saving being one of the critical environmental attributes and sustainability cost drivers of float-
ing PV systems in the agricultural sector, this model (subsystem) estimates the predictable income from
future preserved crop harvests. The model deserves special attention because it improves agronomy for
sustainable development and requires that it be viewed from a market-analysis, an agricultural-economic,
a water-economic, and a surface-transforming perspective. The agronomic land-gain metric relates to the
monetary value of the farmland as a result of the impact effect of land saving, thus offering a treasure trove
of options to monetise this co-benefit of FPV in terms of agricultural potential. It also accounts for agrarian
economic factors, including the production and sale of agricultural commodities on preserved farmland.
This thesis views the proposed model from the perspective of commercial agriculture, where floating-
solar land preservation effectively increases the area cultivated to add fertile, productive farming hectares to
an already established agricultural unit. Thus, agronomics-wise, this model can account for the productive
income value of FLp, either by appropriating FLp as a real estate asset value, or as a value associated with
preserved food-crop plantings (conventional vineyards/vineyard plantings) or agro-vegetable production (at
the fresh produce market, commodity trading at farm-gate prices), as post-processed merchandise, as a
commodity-generating income revenue for growers’ produce (from a retail price point for wine grape sales),
or as a value measure of a gross farming income determined in terms of industrial benchmarks.
As such, the thesis proposes to translate farmland conserved or saved into an economic efficiency
metric for land. The model incorporates four modes of cost-benefit analysis methodologies to unbundle
the monetisation of the farmland that the FPV has preserved (FLp) (Prinsloo et al., 2021). Regarding the
proposed capital-saving modality of the model, the financial valuation procedure in Equation 36 capitalises
the value of FLp as a financial asset liquidated as a monetary injection factor (Rxi). This cash injection can
be effectively redeemed as an offset to the initial capital investment in the FPV project by defining a user-
configurable cost-of-farmland factor (Rcfl = 85.00 R/m2or =850k R/ha). The land-saving contribution factor
(Rxi) thus reflects the real-estate value of farmland (realtor valuation), or the landowner’s avoided liability in
terms of any land acquisition to install over-water FPV systems (instead of a conventional over-land GPV
system installation, which generally sacrifices valuable farmland).
Secondly, in terms of the proposed model’s modality option for rescued crop production (grain, wheat,
maize, rye, fruit, grape harvests), the thesis estimates that the effective agronomic production rescued by
FPV through the preserved farmland (FLp) can be reconciled, monetary-wise, in terms of crop cultivation
from green pasture as an equivalent annual agro-produce income/revenue stream (Rxp). To assign the
income reserve of farmland savings to the FPV market-clearing price, the thesis considers the crop yield
155
estimate and harvest capacity (tons per acre, or tons per hectare) together with the average commodity
sales price, such as in Figure 4.10.
Figure 4.10: Average sales price in R/tonnes of grapes on SA markets (source: Galal (2020), page 3).
According to this procedure, Equation 37 performs a land-saving (FLp) cost-benefit analysis using a
user-configured sales-unit price (Xxy=15.288 R/kg or =15288 R/ton) for agro-crop produce modulated by
a predictive annual farmland food/fruit production capacity (harvest productivity factor). As a cropping-
specific nett primary production factor (FLx =1.6 kg/m2/yr or =16.11 ton/ha/yr), this harvest productivity
factor is cultivar-dependent for grapes, fruits, cereal, or berry harvests.
The crop yield estimate is an averaged metric because the harvest capacity and vineyard yield estima-
tion is a function of space utilisation (based on row spacing, cluster spacing), berry weight, plant density,
and soil health (Komm, 2015). As such, Figure 4.10 shows the survey records of the SA Wine Industry’s
Information Systems SAWIS (2021) which were used to compile the Galal (2020) records for the average
sales price (in South African Rand per metric ton) of grapes on local markets between 2000 and 2019.
In the third modality of the model, which pertains to rescued commodities, the processed commodity-
rescued modality, the annual food/fruit production capacity (FLx), is translated into a downstream commodity
(FLx2). It represents fresh food produce (maize cereal, grapes, olives) processed into a commodity (wine,
olive oil, jams, botanicals). The processed commodity sales translated into the commodity revenue stream
(Rxp) monetises the potential future commodity income assigned to the farmland preserved (FLp) by the
FPV system (Allen and Prinsloo, 2018). In the default case for wine production from the grapes harvested
(refer to Figure 4.11), it is estimated that for every 1000 kg yield of grapes, about 756 bottles of wine (FLx2
= 63 cases of wine/ton grapes, at 12 bottles per case) are produced (Gerling, 2011).
In this modality, the thesis can also use data such as Equation 38 to estimate the anticipated sales
income/revenue (Rcs) as a crop value commodity addition (wine, olive oil, etc.) in terms of a predictive
multiplier factor for grapes as processed wine commodity (FLx2 =1.6 units/m2/yr or =1611 units/ha/yr) (Gbe-
jewoh et al., 2021), modulated by the vintage sale price estimate for the commodity such as bottled wine
(Xcy=300 R/unit). Once again, these factors are set by the user in the configuration file of the FPV model.
Rxi(y=0,t=0) =F Lp ∗Rcfl(y=0,t=0) (36)
Rxp(y,T )=F Lx(y,T )∗F Lp ∗Xxy(37)
Rcs(y,T )=F Lx2(y,T )∗F Lp ∗Xcy(38)
In the fourth modality option of the model, to monetise the impact effect of FPV land-saving, the benchmark
farming-income modality uses the agri-science profitability approach of the annual production plan survey
(Rabie, 2020). The industry-coded regional benchmarks (profit margin, gross margin, operating margin),
represented as GFI, NFI, and FLx, are applied in Equations 39 and 40, towards translating the FLp metrics
156
(a) (b)
Figure 4.11: Annually sampled (a) production capacity (tonnes/ha) and (b) gross profitability margins (R/tonnes) for
grape growers on the SA market (source: Rabie (2020), pages 3 and 5).
into the gross farming income (Rxg) and nett farming income revenue (Rxn) streams respectively. This
conversion is conducted by engaging the agricultural economics figures for the local winemaking industry,
as recorded in Figure 4.11, as a measure of the profitability parametric for the production income on the
local wine-producing markets (Rabie, 2020).
In the agri-science profitability approach proposed in this thesis, the procedure computes the Rxg and
Rxn revenue indicators from the annual average crop yield capacity per bearing hectare (FLx). The calcu-
lation includes the modulation with the relevant monetary factors, namely the gross- and nett farm income
and gross-margin setup factors (GFI =7.2439, NFI =2.0617, GMI =3.3816 R/m2), subsequently encoded in
the user-configuration file under the relevant industrial benchmark protocols.
Rxg(y,T )=F Lx(y,T )∗F Lp ∗GF Iy(39)
Rxn(y,T )=F Lx(y,T )∗F Lp ∗N F Iy(40)
With the description presented above, the economic simulation model extends the planning analysis of float-
ing solar systems into the field of the green economy and the evaluation of the bankability of the planned
system as an investment deal or business case. Finally, the discussion concludes with the techno-economic
analysis and model-based design aspects of the floatovoltaic-synthesised model from a business engineer-
ing and modelling perspective.
4.3.5.4. Agrivoltaic and aquavoltaic production income value model
This model makes provision for co-locating photovoltaics with agriculture and aquaculture (double harvest
voltaics). In such PV configurations, the PV panel cover may be mounted above green crops in agricultural
settings, or the FPV may facilitate aqua-cultural farming underneath the PV panels. As such the model can
predict the agronomic income from hybridised agrivoltaics (land farming activities underneath PV panels)
and aquavoltaic (water farming activities underneath floating PV panels) systems Gorjian (2022); Pringle
et al. (2017). While the main focus of the FPV model is on conventional performance and the impact of
floating PVs, the user configuration file includes a binary toggle parameter (y/n factor) to switch on/off the
agrivoltaic production and income options. The model further includes configurable parameters to account
for agrivoltaic hybridisation in the model’s income predictions. If agrivoltaic production is enabled, 3E analy-
ses parallel integrates agrivoltaic-related agricultural economics into energy economics. As such, the model
additionally engages Equation 41 and Equation 42 to determine the aqua/agrivoltaic production income
157
(avRxp) and agrivoltaic processed commodity income (avRcs) stream respectively. The aqua/agrivoltaic
production factors (avFLx and avFLx2), with their respective annual growth factors, can be configured in the
user configuration file for the aqua/agrivoltaic scenario at hand.
avRxp(y ,T )=avF Lx(y,T )∗F Lp ∗avXxy(41)
avRcs(y ,T )=avF Lx2(y,T )∗F Lp ∗avXcy(42)
From an agrophotovoltaic systems viewpoint, it is recognised that future research efforts may need to
be extended to include improved agrivoltaic applications. Especially in the case of agrivoltaic GPV, a future
model should allow for plant photosynthesis and evapotranspiration to help cool off the PV panel array and
raise the PV yield efficiency. While hybridised agri-PV systems are suitable for crop planting, livestock
farming and animal agriculture, animals can graze on grassland or green pastures beneath the panels, or
plants can grow and produce yields below the elevated solar panels of the GPV installation systems installed
for solar energy generation. Furthermore, the colocation of solar energy and agriculture in agrivoltaics can
generate electricity while allowing additional income streams for farmers from agronomy (Dupraz et al.,
2011) or aquavoltaics (Pringle et al., 2017) to grow food for selling or cattle farming, for example. While
such factors are not the main focus of this thesis, they may be valuable to consider as a direction for future
research and avenues for the computer modelling of analytical prognoses for GPVs in PV-site-remediation
applications.
While Sections 4.3.3 to 4.3.5 characterise the constituent floating solar, energy, environmental, and
economic subsystem model objects, respectively, the following section deals with the model-based design
of the metrics and decision-support indicators of the floatovoltaic systems model to establish the next layer
of the FPV analytical instrument.
4.3.6. Cross-object interactions and stock flow linkages design
This section deals with the cross-disciplinary domain interactions and systemic stock flows that is defined
as cross-cutting metrics and performance ratios. In Decision Science terms, decision-making variables are
crucial in computational decision-support models. This section thus supports the model-object specifics in
support of decision analysis (required by Research Objective 3). Generally constituting the FPV model ob-
ject interactions and exchanges (potential systemic covariance links), defining cross-cutting metrics and per-
formance ratios implies deriving performance metrics based on cross-referencing data exchanges across
model object boundaries. Similarly, with the WELF nexus indicators, these indicators are modelled across
model domain performances (modelled into a geoinformatics decision support model on a higher-order
level). This model starts with the terminology and metrics of the WELF-nexus scoping submodel, the
project-efficiency submodel, and the economic surface-transformation submodel, each with a description of
its respective mathematical simulation and with provision made for the dynamic interactions of the object. As
such, this section attends to the technical FPV vs GPV power generation gain model in Section sec:FPVeff,
the technical FPV project efficiency model in Section 4.3.6.2, the integrated FPV project performance ratio
model in Section 4.3.6.3, the EST-water-economic surface transformation model in Section 4.3.6.4, and the
FPV climate-economic water-energy-land-food nexus-scoping model in Section 4.3.6.5.
4.3.6.1. Technical FPV versus GPV power-generation gain model
In terms of PV system efficiency measurements based on an analysis of operational data, Kwon et al.
(2019) worked on an energy-generation efficiency coefficient to compare the output of a floating PV system
with that of a ground-mounted GPV system. In this context, the simulation was able to determine the
158
average improvement in the efficiency of the FPV in terms of the magnitude of power that could be gained
over power generated by a ground-mounted photovoltaic (GPV) system through Equation 43. The Pythonic
code for Equation 44 is further used to compute the solar-to-electricity conversion-rate efficiency (Pη) for the
FPV system on an hour-by-hour or minute-by-minute basis, depending on the selected model simulation
frequency and the frequency of the weather data. These technometrics are estimated as the efficiency
improvement fgPmηin Equation 43 and the solar-to-power conversion efficiency Pη(%) in Equation 44.
f gP mη =DF m(y,t)
DGm(y,t)(43)
P η(y,t)=P m(y,t)
I T R(y,t)
(44)
Equation 43 thus predicts the performance of PV systems installed on water instead of on land. The Python
object further employs Equation 44 to compute the efficiency of the solar-to-power conversion Pη(%) for
every time period in the run-time simulation. In solar-energy technology systems, the Fill Factor (P η(y,t)),
determined in Equation 44 is essentially a measure of the efficiency of the PV module. This measure
determines how closely a solar cell acts as an ideal source in converting optical solar energy to electrical
energy (filling irradiation energy into power, the ratio of the actual highest achievable power from sunlight).
4.3.6.2. Technical FPV project-efficiency model
The project-efficiency submodel determines the energy, environmental, and monetary or economic values
of the efficiency-related factors for a future planned FPV system. These indicators will help to indicate if the
water surface is effectively used (Alam, 2022; Cagle et al., 2020). As with the WELF-nexus indicators, these
indicators are modelled as across-model domain performances (modelled into a geo-informatics decision-
support model on a higher-order level). This model starts with the terminology and metrics defined by Cagle
et al. (2020) to determine the capacity-based water-surface transformation (WST) for FPV (in m2/MWh/yr).
Here, the areal size of the water plot (Afpv) or of the effectively preserved farmland area (FLp) on which the
floating PV system (Afpv) is based, can be used to determine the capacity-based WST. This factor can be
computed as an expression of the average surface capacity factor in Equation 45.
capacity.W S Tfpv =Af pv
PP m(1,T )∗10−6(45)
Another related FPV indicator metric defined by Cagle et al. (2020) computes the efficiency of the direct
water-surface-use metric (WSUe) for FPV (in W/m2or kWh/m2), computed through Equation 46. While
direct WSUe in Equation 46 requires no a-priori knowledge of the reservoir, a future modality may be added
to include the reservoir and floatovoltaic design metrics to determine the so-called total WST.
direct.W S Uef pv =f P rc ∗Afpd ∗10−3
100 (46)
WSUe computations may also include measuring the perimeter of the geometric object and the surface
areas for the water body by outlining the perimeter with Google’s polygon tools to enable computations
towards the convexity spatial metrics of roundness, compactness and eccentricity, as devised by Cagle
et al. (2020), from the major and minor axes of a proposed FPV installation site.
159
4.3.6.3. Integrated FPV project performance-ratio model
Performance Ratio (PR) defines the ratio between the FPV AC power output and the theoretical energy
produced, given that all incoming solar irradiation is transformed relative to the standard PV efficiency at
STC, with systemic and electrical losses (mismatch, wiring, etc.).
P R =P m ∗Bosη
Afpv (47)
The annual specific production capacity (SPC), is defined by the generated power quantity over a particular
period per unit of installed power (unit MWh), which is indicative of production capacity relative to installed
capacity, given cloud and weather intermittences (Dizier, 2018). The annual specific production is deter-
mined as the ratio of generated power factorised by systemic capacity relative to the nominal power rating
of the PV panels, as specified in Equation 48. As FPV systems do not operate at a fully-rated power level,
the specific energy production is a function of the climatic and weather inputs in combination with the panel
orientation factor.
SP C =PP m ∗100
f P rc ∗Af pd ∗Afpv (48)
A techno-economic value indicator is further needed to quantify the financial benefits of the energy-yield
efficiency gains delivered by FPV when compared to a similar-sized and configured co-located GPV system
(currency per m2). While numerous indicators can be coined from this relationship, the thesis defines the
nett gain ratio and the real raw nett gain value as the income gain of FPV over GPV per m2as calculated
through the following expressions.
NettGain =P m
gP m (49)
RawN ettGain =xT ry(P m −g P m)
Afpv (50)
eW ST =P m
gP m ∗W SE tfpv ∗W S Tfpv (51)
The above mention FPV project performance ratios can further be combined with the WELF nexus
indicators and the water surface transformation (WST) and water surface use efficiency (WSUe) indicators
defined by Cagle et al. (2020) for use in an analytical decision support capacity.
4.3.6.4. w-EST-water-economic surface-transformation model
This thesis proposes new performance indicators: the economic-surface-transformation indicator (EST) and
the water-economic surface-transformation indicator for FPV (w-EST or f-EST). This geo-economic surface
transformation indicator computes the direct economic water surface transformation efficiency metric for
FPV (in currency/m2/year). This novel performance water surface economic transformation metric (w-SET)
measure is defined as econometrics to indicate the extent to which every square metre of previously unused
water surface area is transformed into an income-producing, revenue-generating space thanks to floato-
voltaic technology. Being part of the broader geospatial, financial surface transformation (FST, or geo-FST),
such factors may become increasingly important in the case of turnkey floating solar facility rental solutions,
where income estimates are needed for leasing out water surface areas in rural areas or on farmland for
floating solar installations in solar-leasing or solar-rental agreements.
Within the econometric income context seen from a water-economic surface transformation perspec-
tive, these new performance metrics start with defining the water-economic surface-transformation (w-EST)
160
indicator for FPV. It is formulated to serve as a quantitative metric of the extent to which every square
metre of formerly unused water surface area is transformed into an income-production space, or a revenue-
generating space, owing to floatovoltaic technology. As such, the newly defined WEST indicator is calcu-
lated as the annual joint land-normalised sum of all potential direct and indirect revenue streams per unit of
FPV area per year, w-EST = (Rgs + Rcc + Rwe + Rxp + Rzx )/Afpv (currency/m2/year). The w-EST metric
for a floating PV project is thus calculated from Equation 52 as the sum of all revenue streams accrued over
a period of a full year and normalised with respect to the floatovoltaic plot size for which the unused water
space can now be considered a productive annual income revenue-generating space.
w−EST(1,T )= PT
t=0 Rgs(1,t)+Rcc(1,t)+Rwe(1,t)+Rzx(1,t)−Rxl(1,t)
Afpv !(52)
Thus, in agricultural economics, the geo-sensitive water-economic surface-transformation metric serves
as a water-economic surface-transformation (WEST) or hydro-economic surface transformation (HEST) to
describe the value-proposition of floating PV in terms of water surface transformation. While the WEST
metric offers an economic-efficiency metric used in water use, it reflects the annual economic value per
square metre of a previously unused water surface area, which the floatovoltaic project can transform into
a yearly productive income and revenue-generating space. The economic surface transformation indicator
(EST) defined by Equation 53 can also be applied to conventional land-based PV systems. In the GPV
context, the EST metric gives an economic indicator of the land-use change from agricultural production to
renewable energy production in association with the sacrifice of the agro-crop output.
g−ES T ′
(1,T )= PT
t=0 Rgs(1,t)+Rcc(1,t)−Rxp(1,t)+Rzx(1,t)
Afpv !(53)
The thesis defines the GPV-orientated g-EST’ metric, formulated in EST’ in Equation 53, offers a type
of land-use-change economic efficiency metric that reflects the annual economic value per m2of formerly
used agricultural land converted into an energy farming area. In this context, the EST indicator serves as an
enviro-economic efficiency metric that predicts the annual economic value per square metre (total deemed
income/m2) of a previously unused water surface area, which the floatovoltaic project can transform into a
productive income or revenue-generating geometric space (on a farm, at a utility, facility, etc.). Therefore, to
condition the EST value for ground-mounted PV, Equation 53 omits the value of the water evaporation (Rwe)
and subtracts the now sacrificed potential agronomic income factor (Rxp) to render EST’ as an indication of
the new income value rendered by the GPV agro-replacement project.
4.3.6.5. FPV climate-economic water-energy-land-food nexus-scoping model
With floating PV as a natural resource type of technology providing for the preservation of energy and
touching on issues around the interaction of the water-energy-land-food nexus dimensions (WELF-nexus)
(Hamidov et al., 2022; Torhan et al., 2022), a robust and efficient validation of the co-benefits and suggested
impacts of the technology on the nexus of the local energy-water-food (EWF) system is lacking (Gadzanku
et al., 2021b). In this thesis, the WELF nexus stands central to an integrated approach to the governance
of and accounting for natural capital, especially since this resource system needs to be absorbed into the
nexus-decision research and decision-policy fields.
An algorithmic impact assessment of the floatovoltaic WELF-footprint submodel needs to effectively
integrate the FPV sustainability metrics into a WELF-nexus decision-profile-mapping display in support of
decision-making. By modelling the water-energy-land-food nexus elements and by coupling these artic-
ulated indicators for floating PV systems, it is possible for this selective combination of resource-related
161
envirometrics with technometrics cross-coupled indicator set to essentially model the across-domain per-
formance for incorporation into a geoinformatics decision-support model on a higher level. The WELF
sustainability indicator series in the decision space for project planning offers a means for ensuring bal-
anced sustainability within the photovoltaic ecosystem. In this annualised WELF-indicator model, the Water
element of the WELF-nexus metric set includes the local preservation of water through avoided evapora-
tion (eH2O), plus the water preserved through coal-fired grid substitution (xH2O) (refEqsandeq:evapoeq:ho).
The power generation yield (Pm) constitutes the Energy element of the WELF-nexus metrics (Equation F.5).
The Land-preserved element of the WELF-nexus metric set is defined by the impact metric for the preserved
farmland (FLp), which is calculated by the land-saving submodel (Equation 18). As an indicator related to
the mitigation of land-use issues, a land-factor gearing ratio relates to the geometric water surface area
(Afpv) covered by the photovoltaics plot (in m2). The Food element of the WELF-nexus metric set is associ-
ated with the annual fruit/food production capacity (FLx) or the processed commodity product units (FLx2)
agro-produced from the preserved farmland as attributes of the FPV installation (Section 4.3.5.3 and Fig-
ure 4.10). As a result of the deployment of the floatovoltaic power plant, this metric is indicative of assured
access to food security and the preservation of food.
W elf(y=1,T )= T
X
t=0
eH2O(1,t)+xH2O(1,t)!(54)
wElf(y=1,T )= T
X
t=0
f P m(1,t)!(55)
weLf(y=1,T )=F Lp (56)
welF a(y=1,T )=F Lx(y=1) ∗F Lp (57)
welF b(y=1,T )=F Lx2(y=1) ∗F Lp (58)
The thesis defines the new climate-economic water-energy-land-food index (ce-WELF index) as a PV
project sustainability index. This index combines the annual carbon sequestration footprint (c) and economi-
cal surface transformation (e) metrics for the year of project commencement with the WELF-resource indica-
tors, thus uniquely integrating technometrics and econometrics with natural resource-related envirometrics.
It thus uniquely combines a set of six ce-WELF attribute metrics to define a composite sustainability index
for floating PV projects. Having already defined the formulations for the WELF-nexus calculations above,
the ce-factor that makes up the ce-WELF index is defined around the CO2footprint (c) and the economic
EST (e) of the FPV system. The ce-factors are formulated by Equations 59 and 61 respectively. The total
lifetime avoided fCO2(c = fCO2) includes two components, as defined in Equation 59, namely, the grid-
substitution emission reduction (xCO2) and the carbon sequestrated by avoided crop disturbance (pCO2)
owing to the perennial crop/vineyard covering the preserved farmland (FLp). The total lifetime income from
the economic transformation of the water surface EST (e = fEST) includes the EST defined in Equation 52
for the total FPV plot area and the annual performance of the project (in year y=1). For a GPV system coun-
terpart, the EST factor (e = gEST) is given by Equation 62 (note that the annual agro-production income
Rx(1,t)is treated as an expense, as the GPV system sacrifices agro-production).
fCO2(y=1,T )=pCO2∗F Lp + T
X
t=0
xCO2(1,t)!(59)
gCO2(y=1,T )= T
X
t=0
xCO2(1,t)!(60)
162
fEST(y=1,T )= PT
t=0 Rgs(1,t)+Rcc(1,t)+Rwe(1,t)+Rxp(1,t)
Afpv !(61)
gEST(y=1,T )= PT
t=0 Rsg(1,t)+Rcc(1,t)−Rxp(1,t)
Afpv !(62)
The ce-WELF sustainability indicator series for floating solar developments offers assistance in project
planning, as it provides the decision-maker with a sense of awareness as to the extent to which a planned
PV project would threaten the farm’s WELF equilibrium or the coordinated WELF-nexus performance and
impact balances within the planned sustainable energy ecosystem. Note that for the GPV system analyses,
by definition, the WELF factors the eH2O (W) while land (L) and food (F) all equate to zero as the GPV has
zero contribution to land and food production as well as from avoided evaporation.
This discussion concludes this section’s narrative view of the dynamics simulation modelling and inter-
active geoinformatics ontology toward meeting the empirical, analytical goals of Research Objective 2. The
next section attends to specifics of the analytical decision support required for Research Objective 3.
4.4. Research Model: Decision-level Layering Design and Implementing
Progressing towards implementing the next layer of the geoinformatics ecosystem, the model-design pro-
cess advances towards integrating the object layer into the decision model and establishing the indicator-
processing functionality of this GIS layer. Focusing on the implementation of Research Objective 3 accord-
ing to the flowchart of Figure 3.22b, the decision-model layer establishes a basic framework for mapping
the quantified decision-indicator portfolios. This section describes the integration procedures for modelling
the geoinformatics decision-support profile map in preparation for the scientific experimentation and exper-
imental evaluations whereby the analytical and decision-analysis models are applied.
The goal is thus directed towards processing the quantitative data-sample outputs from the floatovoltaic
eco-system model according to a suitable data-analysis technique that allows for comparative decision sup-
port on a quantitative floatovoltaic-performance-profile decision-support map. To this end, Section 4.4.1
narrates the integrated system dynamics model architecture to accommodate the framework and analyt-
ical technique and to articulate the results in a decision-profile mapping display. Section 4.4.2 defines
the sustainability goal of the project as a decision-support index to work in support of the computational
decision-support classification capabilities of the model towards sustainability. Section 4.4.3 presents the
implementation of the analytical hierarchical process (AHP), a computational technique to weight the indi-
cators according to the project goals statistically and to drive the profile-mapping display dashboard as one
of the layers of the geoinformatics decision-support toolset device.
4.4.1. Integration of decision-support systems layering
The previous chapter introduced the strategic aspects of integrating the floatovoltaic-synthesised decision-
support model, the conceptual model of Figure 3.21, leaving this section to deal with the specifics of im-
plementing Research Objective 3. This section thus narrates the foundational aspects of the FPV-GIS
profile-mapping mechanism to enable project decision support through sustainability-viability assessments.
In the data-driven decision-making process, the decision-support model and the indicators determined
by the floatovoltaic system dynamics simulation model of Figure 4.3 need to be integrated into a proposed
analytical geoinformatics toolset for floating PV systems. The integration is made possible by embedding
the simulation model in a decision-support model of the project, as illustrated in Figure 4.12. The layered
163
systems model of Figure 4.12a describes the integration of the synthesised FPV model in an analytical hi-
erarchy process, according to the layered hierarchical data-processing strategy shown in Figure 4.12b. Fig-
ure 4.12 thus establishes the foundational aspects around the FPV-GIS profile-mapping mechanism towards
decision support around sustainability-viability assessments in a systems-oriented planning-environment
paradigm.
Country-specific geospatial, environmental, legislative, economic, technical context
□Floatovoltaic Geoinformatics Information System Platform
△Floatovoltaic Planning Decision Indicator Model Layer
⊕Floatovoltaic Ecosystem Synthesis Model Layer
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
(a) Geo-informatical platform components
⊕Floatovoltaic Ecosystem Synthesis Model Layer
(Primary-level-synthesis)
△Planning Decision Indicator Model Layer
(Secondary-level-processing)
□FPV Geoinformatics IS
(Tertiary-level-contextualisation)
Project Planning
(Supervisor-level-optimisation)
(b) Geo-informatical platform layers
Figure 4.12: Conceptual geo-informatics analysis and the decision-support system as a research instrument to study
floating PV systems, integrating (a) the floatovoltaic synthesised model with an analytical hierarchical process accord-
ing to (b) a layered hierarchical data-processing strategy (source: author).
In the case of science-based decision-making for floating PV project support, the sustainability assess-
ment system of Figure 4.12 should resonate with decision-makers and allow them to articulate the sus-
tainability performance of the system in terms of a multi-criterial profile-mapping mechanism. Furthermore,
integrating analytical and visual methods toward the goal of decision support contributes to the exploration
of complex datasets and patterns in geo-related multivariate datasets to help understand the significance
of the performance data relevant to an agricultural floating PV system (Sips et al., 2007).
As such, an appropriate multidimensional display mechanism can serve as a means to support project
decision-making in terms of displaying the sustainability profile status of candidate projects. To articulate the
performance metrics as sustainability indicators, the thesis implements a multidimensional profile-mapping
solution to visualise the sustainability results by displaying its multiple attributes. The model of Figure 4.12
uses the multidimensional spider diagram dashboard of Figure 4.13a to plot the response indicators of the
floatovoltaic system.
The project profile display in Figure 4.13a can accommodate a tight integration of the project goals
through a carefully selected set of indicator metrics while accommodating interactive visualisations with
automated decision analysis. As such, the profile-mapping solution proposed by Figure 4.13a uses a mul-
tidimensional spider diagram dashboard to plot the response indicators of the floatovoltaic system in terms
of the so-called sustainability decision-support index or indicator set.
The chosen subset of 3E indicators in Figure 4.13a is then plugged into the forward propagation network
model in Figure 4.13b to determine the decision probability of the project meeting the project goals. The
input variables to the numerical AHP decision-support model Figure 4.13b is described in the next section
and is referenced as representative of the decision-support indicators in respect of sustainability. While the
decision-support indicators in Figure 4.13a establish a dynamic model presentation of a composite set of
164
Indicator 1Indicator 2
Indicator 3
Indicator 4 Indicator 5
Indicator 6
(a) Decision-indicator components
Indicator #1
Indicator #2
Indicator #3
Indicator #4
Indicator #5
Indicator #6
Prob D#1
Prob D#2
AHP
weights
Input
layer
Decision
layer
(b) Decision-processing layers
Figure 4.13: Floating PV sustainability decision space, showing (a) selected metric elements or indicator pillars plot-
ted on a profile mapping display, and (b) the application of AHP-decision criterial weights to the indicator dimensions
in terms of computational decision support (source: author).
generic sustainability indicators, the sustainability indicator metrics in this index serve as input indicators in
the multi-criteria computational project decision-support, as depicted in Figure 4.13b.
As a statistically-driven means for multi-criterial decision support, Figure 4.13b essentially processes the
decision indicators in a feed-forward multi-criterial weighting scheme to determine the success probabilities
for sustainability for various candidate projects.
The conceptual profile-mapping projection for sustainability, as depicted in Figure 4.13, establishes a
dynamics model presentation of a composite set of generic indicators in terms of Cartesian coordinates or
points on a decision plane on a radar-type diagram. From a theoretical perspective, the display can, as such,
articulate information to project decision-makers that might lead them to identify planning challenges and
also to actions towards improved sustainability or to choose between other project alternatives (Moldavska
and Welo, 2015). The sustainability profile-mapping display of Figure 4.13a can thus present and articulate
the status of a floating PV project as a sustainability index in terms of multidimensional profile-mapping vec-
tors. These vectors represent the decision-space variables of a floatovoltaic project on a two-dimensional
plane. The dashboard diagram of Figure 4.13a thus serves as a dynamic decision-support tool, whereby
the data-visualisation dashboard template depicts a project’s essential sustainability performance values as
decision-space variables. The sustainability profile of candidate projects is further weighted statistically in
terms of the decision goals of the project.
With the emphasis on the virtualisation of the decision space, the analytical model, with its various
systems components and variables, and a set of mathematical equations, uses metrics to describe the
processes responsible for systems behaviour in terms of the performance indicators (Jakhrani et al., 2014).
As such, the proposed sustainability display diagram, Figure 4.13a, allows for the mapping of the sustain-
ability profile of one or more candidate projects to be superimposed on the decision-support dashboard
diagram. In this respect, the red and blue project-sustainability profile markers in Figure 4.13a serve as
sustainability-profiling examples for two plotted demonstrative projects. Furthermore, for the sustainability
profile of candidate projects to be statistically weighted in terms of the decision goals of the project, a sta-
tistical processing technique such as that depicted in Figure 4.13b is needed to process the performance
metrics of the floating PV system into weighted project-decision goals. One such weighting and the nu-
merical decision-support scheme is proposed by the AHP-data-processing object structure, as depicted in
Figure 4.13b. The procedural AHP-weighting of metrics as such can accommodate relative comparisons
between multiscale parameters or on a visual device such as Figure 4.13a, while enabling the indicators
to be further fed into the probabilistic decision-level algorithmic layers capable of advising on potential
165
decision-outcome probabilities as outputs, and as depicted in Figure 4.13b.
While the system dynamics model and its application in Figure 4.12, include the simulation integration
of the Energy-, Environmental- and Economic simulation models, or the three "E” models, a set of chosen
decision indicators to drive the profile-mapping display of Figure 4.13 can statistically be calculated from the
3E performance metrics output from each of the object models and their interacting interfaces. In this way,
the floatovoltaic planning-decision indicator model layer in Figure 4.12 engages the integrated analytical
framework theory of Figure 3.18 in a quantitative life-cycle sustainability assessment model framework
in the integrative real-time simulation mode. In the decision mode, the geoinformatics toolset can thus
perform life cycle processing on the performance and impact-offset metrics as input parameters to support
the functionality of the toolset in planning decision support.
While the profile-mapping display mechanism of Figure 4.13 establishes a dynamics model presen-
tation of a composite set of generic indicators that defines floating PV-project sustainability, the following
section focuses on the project-sustainability goal as a decision-supporting definition for advisement towards
decision-making of sustainability around the project in question.
4.4.2. Defining sustainability performance indicators for FPV projects
This section focuses on the construction of a sustainability decision-support index for floating PV to serve as
project sustainability performance indicators. It introduces the definition of the project sustainability criteria
and goal to be used in the computational decision-support system described in the following section. To
this end, an appropriate set of decision metrics are selected from the 3E dataset to define and classify the
computational sustainability-decision goal in terms of a sustainability-decision index.
This thesis conceptualised the analytical 3E framework (technical/energy, economic, environmental
framework) as a computer-logic blueprint for assessing floating PV system sustainability. As such, it should
be clear that the study can combine one or more FPV performance metrics under each 3E performance
category into a custom-designed decision-support framework for sustainability. When one or more of these
3E attribute metrics are array integrated into a composite sustainability index, the index array can serve as
an indicator set for statistical decision analysis and support (refer to Figure 4.13). As such, the selected
indicator metrics can form part of user-customised decision criteria that define the sustainability goal of the
project as a means of managing decision trade-offs in sustainability assessment (Morrison-Saunders and
Pope, 2013). If processed in an appropriate decision algorithm, the sustainability priority profile vector out-
comes can drive computational decision-support towards probabilistic advisement regarding sustainability
decisions around the FPV project.
While the decision-support model of Figure 4.12 requires cryptic performance metrics to assess various
project-planning situations, there is a specific requirement to choose a set of decision indicators to serve
as decision criteria in quantitative numerical decision support. It establishes the criteria framework for
effective decision-making in a multi-context problem through a numerical decision matrix or model (the
basis for SWOT analysis) (NAC, 2013). One option is to digitally translate the decision metrics through the
mathematical calculation of the indicators from the performance-output metrics through the object models
detailed in Figure 4.3, or from the floating PV performance factors and ratios defined by Section 4.3.6.
These indicators can even be combined in the cross-object project efficiency decision submodel, to translate
the efficiency-related factors or the energy, environmental and monetary or economic values into decision
indicators. In this context, the cross-cutting metrics and performance ratios defined in Section 4.3.6 are
helpful in the plan to implement the analytical hierarchy process in floatovoltaic decision-support planning
underlying floatovoltaic system dynamics model analytics and metrics as per Figure 4.12.
Thus, the decision-support aim is set to define a holistic environmental-sustainability index for floating
PV projects to measure and evaluate the quantitative attributes of sustainable development in terms of a
166
composite sustainability index. The 3E framework offers a sound basis for principal sustainability com-
ponent analysis regarding pertinent performance metrics based on the systemic energy, economic and
environmental responses of a floatovoltaic system. While this thesis defines the 3E-framework as the basis
for sustainability assessments, individual elements of each of the 3E-dimensions in Figure 4.14a need to
be selected to form part of a sustainability-decision index as a means to express the sustainability status of
a project. As such, a sustainability value index is comprised of a metric value for each of the 3E disciplines,
with the index developed to reflect its degree of influence over the sustainability index of the overall project.
Energy
Economic Environmental
(a) 3E-category indicator components
Land
Energy
Water
EST CO2
Food
(b) Enviro-Econo-WELF-nexus components
Figure 4.14: Floating PV sustainability decision space, showing (a) the potential 3E metric dimensions and (b) the
enviro-economic WELF-indicator index set as sustainability pillars plotted on the profile map (source: author).
The sustainability framework of Figure 3.18, in conjunction with its indicator dimensions in the 3E
trilemma of Figure 4.14a, generally offers the option for selecting dimensions to accomplish optimal sus-
tainability in terms of a composite indication of project sustainability. However, from a WELF-systems per-
spective (Goswami and Sadhu, 2021a; Zarei et al., 2021), the question boils down to whether a floating PV
project can reliably meet the energy demand while also addressing the pressure on the local WELF system
(Dorgo et al., 2018; Gadzanku et al., 2021b).
In this context, the WELF-nexus concept offers a valuable measuring tool to study sustainability scenar-
ios around the application of PV (and floating PV) technology in the agricultural sector (Lopes et al., 2020;
Sohofi et al., 2016), especially in terms of the optimisation of the water-resource, energy-resource, land-
resource and food-resource interactions (Hoff, 2011; Ringler et al., 2013). Furthermore, while advancing
the clean energy economy in an integrated manner (Advanced Energy Economy, 2019; Rizzoli et al., 2008),
an integrated systems approach to modelling the WELF nexus through integrative analytical models offer
valuable WELF-nexus sustainability indicators for assessing natural resource management (Amadei, 2019;
Nhamo et al., 2020).
Fortunately, the WELF resources and its associated nexus are already encompassed within the 3E sus-
tainability framing, as shown in Figure E.1, at the intersection of the Energy and Environmental domains.
With the WELF metric attributions being a condensed subset of metrics within the set of 50) 3E framing
metrics, it enables the thesis to mathematically select or translate the WELF indicator parameters from
the 3E metrics. In this context, Figure 4.15 depicts the related process to move from 3E model data to
WEF-nexus-index-driven project decision-making. According to the process of Figure 4.15, the study can
build the criteria for sustainability upon the WELF-nexus concept and its components. In terms of formu-
lating condensed sustainability criteria as a subset of the comprehensive set of 3E, the thesis presents an
innovative approach to defining a decision-support framing that relies on constructing a composite index
167
by combining WELF-nexus parameters with other 3E metrics. This concerns blending WELF metrics with
project-performance attributes such as economic and climate-change parameters in a unified sustainability
criteria index. The concept meets this goal by defining a composite sustainability indicator set compris-
ing six indicator metrics, as defined by the metric indicator dimensions depicted in the spider diagram of
Figure 4.14b.
Figure 4.15: Data-driven decision-making based on WEF Nexus Index, moving from model data to decision making
(source: Simpson et al. (2022), page 4).
Based on the principles depicted by Figures 3.11 and 3.15, and the empirical formulations in Sec-
tion 4.3.6.5, the study digitally translates a novel condensed set of climate-economic-WELF-nexus indica-
tors (ce-WELF) from the 3E (energy, environmental and economic) performance metrics. With the com-
prehensive set of 3E metrics obtained from the analytical model in Figure 4.12, the study can compile this
composite set of ce-WELF-nexus sustainability indicators to define the proposed new resource-integrated
sustainability decision-support index for FPV projects. In defining this composite ce-WELF-nexus indicator
model, the thesis combines the carbon sequestration footprint and economical surface transformation with
the WELF nexus in a novel sustainability index for floating PV projects. As described in Section 4.3.6.5,
this composite ce-WELF-nexus indicator combines the economic surface transformation with the founda-
tional elements of the WELF-nexus parameters to define an integrated set of project key performance areas
(KPI’s) (Prinsloo et al., 2022a). The notion of the combinatory metrics in key performance indicator target-
setting requires sustainability criteria components in the decision model to be foundational components of
the on-site WELF nexus. As principal resource components, the sustainability index should be working
in support of the WELF nexus. While Equations 54 to 58 define the WELF-resource components for the
on-site or on-farm resource determinations, the climate-carbon component ("c”) is defined in Equation 59
and the economic component ("e”) is defined in Equation 61.
The next step is to evaluate the anticipated benefits of each metric contribution to the project’s sustain-
ability objectives or goals. The following section describes a statistical means whereby the metrics can be
weighted as indicators according to the project goals.
4.4.3. Analytical hierarchically-driven decision-support model
One of the defining characteristics of sustainability impact assessment is the forum it provides for the explicit
consideration of project trade-offs in an inherently complex decision-making process. It brings the discus-
sion to the implementation procedures, described by the workflow defined in the flowchart of Figure 3.22b,
describing the implementation of the hierarchical analytical process as a statistical means to weigh the
metrics in an indicator index according to the specific goals of an FPV project.
The study explored various options for the mathematical processing of model parameters in support
of the profile-mapping and data-driven decision-making process (refer to Section 3.7). It opted for the
168
development of a decision-supporting model based on the principles of the Analytical Hierarchical Pro-
cess (AHP) towards Multi-Criterial Decision Analysis (MCDA) and multi-criterial decision-making (MCDM)
(Prinsloo, 2020; Prinsloo et al., 2022a). As one of the probability-driven mathematical models available to
support the decision theory in a Systems Science context, the digital signal processing procedure of the
AHP methodology offers an ideal decision-making framework in the form of a multivariate and multiscale
decision problem (Anbari et al., 2010; Saaty, 2008). In this solution, the decision metrics and indicators se-
lected in the previous section are processed into a decision index supporting the definition and classification
of the computational sustainability decision goal.
In general, a probabilistic reasoning model describes a model in which the project decision-maker or
stakeholder engages with information theory principles in an information-gathering process, and has to face
the challenge of a trade-off between the results of the modelling solution and the associated observational
cost of the results (Reverberi and Talamo, 1999). As an operations research problem aimed at digital
decision-making support, the AHP is identified as a valuable normalisation technique for the statistical
weighing of floating PV appraisals. However, although it offers a helpful approach, the multidimensional
decision-element units might be difficult to quantify, as would also be the case when it must compare
different units of measurement (Vargas, 2010).
In preparing to engage the AHP process in empirical data-driven decision-making around floating PV
projects, the model can now use the proposed portfolio measures around the ce-WELF sustainability index
to establish a suitable decision matrix according to the hierarchical tree of Figure 4.16. The overall goal of
the analytical hierarchical-network decision diagram in Figure 4.16 is shown at the top level of the hierarchy.
In this thesis, the goal is to select the best photovoltaic project to support sustainability farming empirically.
In the second tier, the criteria pertaining to the indicator attributes drive the decisions in the third layer
towards selecting the optimal probability project (say FPV or GPV) as the optimal solution in support of a
suitable multi-tier farming solution that is sustainable.
Sustainability
Welf
Prob D#1 Prob D#2
wElf
Prob D#1 Prob D#2
weLf
Prob D#1 Prob D#2
welF
Prob D#1 Prob D#2
CO2
Prob D#1 Prob D#2
EST
Prob D#1 Prob D#2
Figure 4.16: Project decision hierarchy to solve an integrated multi-criterial sustainability-planning and investment-
decision problem in respect of a floatovoltaic system (source: author).
Figure 4.16 depicts the decomposition of the investment-decision problem in respect of multi-criterial
floatovoltaics sustainability into a project decision hierarchy, including the target layer (top), criterion layer
(middle), sub-criterion index layer (second bottom layer, not shown), and the assessment unit probability or
decision layer (bottom layer). Note that each of the ce-WELF metrics in Figure 4.16 is a parent element that
might be further subdivided into smaller elements in the future. However, its contribution always remains
linked to the priority of the parent element.
With the decision-performance portfolio defined in Figure 4.16, the main goal of the AHP is to empirically
determine the mathematical ratio scale measures of each of the performance measures of the project for
each of the performance components and based on the user preferences. Each sustainability index metric
criterion is assigned a probability score (defined as a metric indicator weight) such that the summation of
these probabilities adds up to 100%. The priority hierarchy established according to Figure 4.16 evaluates
the criteria in pairs to assign importance to the attributes in the middle layer of a decision matrix. This matrix
empirically quantifies the relative importance of the attributed characteristics as opposed to the relative
weight contribution of each to the sustainability decision goals relating to the global project (Saaty, 2008).
169
This approach, in turn, helps to determine the ratio-scale measures of the relative importance of each of
the performance components in respect of the goal of the entire floating PV project.
While the decision tree processes the chosen ce-WELF categories belonging to the decision indicator
set defined in the previous section, the rest of the discussion details the mathematical calculations for
digitally translating the sustainability indicators from the performance-metrics outputs by the object models,
as obtained from Figure 4.3. From the established decision hierarchy in Figure 4.16, the priorities in respect
of the indicator-selection criteria for the ce-WELF metric elements are evaluated in pairs to determine the
relative importance of the indicator metrics as the relative weight of each metric in respect of the overall
project goal (sustainability goal). This exercise aligns with the project’s strategic objectives by determining
priorities for each sustainability objective (Vargas, 2010), as illustrated in Figure 4.17.
W E L F CO2EST
W1 1/2 2 3 1/2 1/2
E2 1 2 3 1 2
L1/2 1/2 1 2 1/3 1/3
F1/3 1/3 1/2 1 1/2 1/4
CO2 2 1 3 2 1 4
EST 2 1/2 3 4 1/4 1
(a) Pairwise comparison matrix
Intensity of Importance (scale
for pairing aij)
Numeric
Rating
Reciprocal
(decimal)
Extreme Importance 9 1/9 (0.111)
Very strong to Extremely 8 1/8 (0.125)
Very strong Importance 7 1/7 (0.143)
Strong to very Strong 6 1/6 (0.167)
Strong Importance 5 1/5 (0.200)
Moderately to Strong 4 1/4 (0.250)
Moderate Importance 3 1/3 (0.333)
Equally to Moderately 2 1/2 (0.500)
Equal Importance 1 1/1 (1.000)
(b) Criteria-rating scale
Figure 4.17: Determining the relative weights of the decision criteria through (a) a comparison matrix for a multi-
criterial floatovoltaic sustainability goal as an investment-planning decision problem; using (b) the fundamental weight
scale for pairwise indicator comparisons (after: Saaty (2008)).
Defined as a dependability matrix, the dependency matrix specified in Figure 4.17a, serves as a re-
lational map for deriving ratio scales from paired comparisons to finally reach a point where the model
can process a decision-ranking output. The scale of the pairwise comparisons in Figure 4.17b offers a
standardised nine-point scale for the pairwise comparisons in the AHP matrix of Figure 4.17a, while ob-
taining the normalised ratio scales in Figure 4.17a by normalising the comparative indicator scales with the
eigenvectors of the dependency matrix (Wierzbicki et al., 2000).
It is ideal for determining the pair-wise comparison matrix for the multi-criteria floatovoltaic sustainabil-
ity goal in Figure 4.17a with the help of an environmental impact practitioner, preferably in consultation
with other project decision-makers or participants. The impact practitioner engages the numerical weights
defined by the evaluation criteria in Figure 4.17b to define the index-based sustainability assessment in
the AHP framework matrix of Figure 4.17a. To this end, the scale of the matrix elements in Figure 4.17
can tactically and strategically determine the contribution of each decision-index metric determinant to the
project’s sustainability objectives according to the calculations in Figure 4.18. This process requires thor-
oughly structuring the relative contribution of each sustainability component or criterion. As demonstrated
in Figure 4.18, this process requires the participative understanding and contributions of the impact practi-
tioner to define the relative priorities in Figure 4.17a on the basis of the rating scale defined by Saaty (2008)
in Figure 4.17a. As a geographical impact assessment practitioner, Prinsloo et al. (2022a) determined the
relative weights of the initial criteria sets that contribute to project-sustainability decisions, as well as to the
relative weight data between the criterial alternatives from the normalised comparison matrix, as shown in
the dependency matrix of Figure 4.17a. As demonstrated by Figure 4.18, the AHP Software Tool, developed
by Goepel (2018), supported the process to define the dependency matrix of Figure 4.17a.
170
Figure 4.18: Metric weighting results of the Analytical Hierarchical Process determined through the user-defined
criteria of the Goepel AHP Software Tool (after: Goepel (2018)).
In this process, the contribution of each criterion to the sustainable project investment goal was deter-
mined by the priority vector (or eigenvector) as relative weights between each attribute criterion.
In this context, the calculation of the maximum eigenvalue (λmax) is determined from the eigenvector
values of the interdependency matrix (Goepel, 2013). The determined consistency index for the matrix in
Figure 4.17a is determined as CI=0.27, with a computed consistency ratio of CR=0.074 or 7.4%. Since the
consistency ratio value is less than 10%, the matrix for comparing the sustainability criteria is considered
consistent with and suited to AHP decision-making (Goepel, 2013; Saaty, 2008). The priority criteria results
for the second-level layer of the AHP hierarchy can be seen in the weight graph exhibit of Figure 4.18.
As such, the impact practitioner defines the project goals as criteria judgement ratings in the pairwise-
comparison matrix as rankings, whereby pairwise comparisons are based on empirical scores or subjective
opinions about the goal criteria for the project as defined in the comparison matrix of Figure 4.17a. The
numerically-normalised columns and average rows are used to define the sought-after criterial weights. In
this way, the results of Figure 4.17a are computed by way of Figure 4.18 so that the target layer in Figure 4.16
defines the project goal (to balance the pillars of sustainability). It engages the criterial layer as goal-driving
criteria (performance and impact categories, sustainability metrics, WELF criteria, strategic criteria, energy
criteria, environmental criteria, economic/financial criteria), optional subcriteria (sustainability submetrics,
profiles, conditions, status, properties, quantities), and the decision layer, to represent the selected decision
alternatives (probabilities of sustainable project options probabilities).
The thesis can now apply the AHP weights of Figure 4.18 to the individual ce-WELF metrics to table
the relative AHP decision weights for the CO2, EST and WELF-nexus pillars. The AHP criterial weights
pertaining to the project goal decisions for each indicator dimension, defined by Table 4.2, serve as indicator
metric-weight priorities. The fourth column in the AHP matrix results of Table 4.2 shows the normalised
principal eigenvectors or weight factors for the decision-support elements for each floating PV project, as
depicted in Figure 4.13b. Finally, the fifth element in Table 4.2 represents the importance of the ranking
score for each of the chosen metric elements for ce-WELF sustainability.
The AHP criterial weights for each indicator dimension of the project sustainability decision goal define
the decision-metric priorities in Table 4.2. The sustainability weight profile defined by the AHP weights in
Table 4.2 essentially defines the AHP project goal policy. The AHP goal policy essentially defines an impor-
tance ranking to scale or prioritise the (ce-WELF) decision criteria indicators to help the user determine the
extent to which the project outcomes meet the project goals. The project goal policy weights in Table 4.2
171
Table 4.2: AHP results for the CO2, EST and WELF-nexus pillars (source: author).
Category Criteria Probability Weight Ranking
Water 13.3% 0.133 4
WELF Energy 23.7% 0.237 2
Land 8.7% 0.087 5
Food 6.6% 0.066 6
Climate CO229.6% 0.296 1
Income EST 18% 0.18 3
are applied to the 3E performance metrics obtained from the 3E floating PV simulation system to determine
the goal-scaled (ce-WELF) sustainability indicators as project sustainability profile vectors.
Numerically, the AHP performs this processing action by engaging the decision-weighting mechanism
(depicted in Figure 4.13b) and by displaying the processed indicators on the spider diagram dashboard
as sustainability priority profile vector estimations. Indicator processing entails multiplying the performance
metrics with the normalisation weighting factors, thus applying the weighting or ratio-scale measures of
the anticipated project sustainability benefits toward the project sustainability objectives and the ratio-scale
measures of each of the project metrics in the performance portfolio. As such, the AHP decision algorithm
essentially modulates the ce-WELF priority goals of a project on the ce-WELF decision criteria probabili-
ties to determine the sustainability priority profile vector estimates to be displayed on a radar-type profile
mapping display. As such, the decision space for the sustainability of the floating PV system, as depicted in
Figure 4.13b, shows the selected metric elements or indicator pillars plotted on the profile mapping display.
An important aspect that influences the AHP multi-criteria decision method concerns the normalisation
procedure (Vafaei et al., 2016). The mathematical expression in Equation 1 applies the linear normalisation
formula to the decision criteria metrics (the ce-WELF criteria metrics) to translate the raw FPV performance
data into criteria-specific beneficial probability scores (Vafaei et al., 2020).
X(i,j)=X(i,j)
PN
j=1 X(i,j)
(1)
The normalisation of criteria data is key to making decisions in an environment where performance
profile metrics are expressed in different/incompatible units and scales (Rands, litres, watt/hour, m2, kilo-
grams, etc.). This aspect applies explicitly to the normalisation of the ce-WELF criteria metrics where they
are translated into project probabilities, or when the raw 3E model outcome performance data for an FPV
project may need to be plotted directly on the profile mapping display for comparison purposes.
The metric priorities for decision support, as presented in Table 4.2, are applied to the decision-support
means of Figure 4.13b establishes that the universal simulation model can drive the analytical and decision-
support exercises in geoinformatics towards meeting Research Objective 3. Note that the nature of the cho-
sen metrics is such that if the parent metrics are further subdivided into smaller elements, each subdivided
element will conserve the total priority of the parent element so that the subdivision fraction of the global
priority is, as such, calculated through multiplication.
This section details the design and development of the decision-support layer of the proposed integrated
floating solar ecosystem model. The following section summarises the entire structure of the integrated
geoinformatics toolset for project planning and decision support for floating solar PV energy systems as a
research instrument to help answer the research questions associated with Research Objectives 1 to 3.
172
4.5. Integrated Data-collection and Data-processing Instrument
In finalising the development of the FPV-GIS application tool, this section summarises the implementation of
the Python-coded multiscale numerical simulation model and the entire structure of the computer simulation
model as an integrated data collection and data-processing unit. It delivers an integrated geoinformatics
toolset for design stage floatovoltaic sustainability analyses and assessments. It functionally serves as
a research instrument to accomplish integrated data-collection and -processing according to this thesis’s
proposed integrated processing theory and framework. Towards establishing an integrated multiscale nu-
merical simulation model suitable for experimental floatovoltaic analytics, Section 4.5.1 encodes the cyclic
workflow ontology of the floating PV system in a geographical software engineering model. Section 4.5.2
finally provides an overview of the integration of the research model, where the systemic components are
incorporated into the geoinformatics research systems model and its component layers.
4.5.1. Integrated discrete-time Python floatovoltaic synthesis model
This section describes the Python-coded discrete-time floatovoltaic synthesis model. As a GIS systems
analyst and systems programmer, the author selects the object-orientated Python language to encode
the FPV model’s cyclic co-simulation processing workflow ontology in a geographical software-engineering
method. From a geospatial sciences perspective, the software development dynamically establishes the
integrated multiscale numerical simulation model as an FPV-GIS toolset, an integrated dynamic opera-
tional/behavioural model and a data-processing unit suitable for experimental FPV project assessments.
From the onset, it is essential to note that the software development for the FPV-GIS toolset is integrated
into the backbone of the conventional PVlib Python-prototyping simulation engine (Sandia, 2020). In this
way, the PV simulation model is programmed to switch operations in any one of two modes, namely in the
ground/land-mounted mode and in the floating/water-mounted mode. In the ground/land-mounted mode,
the FPV-GIS toolset performs GPV simulation analyses according to conventional PV analysis procedures.
In contrast, the floating/water-mounted mode of the FPV-GIS toolset performs FPV simulation analyses
according to the procedures postulated in this thesis. This design feature enables the implemented FPV-
GIS toolset to switch between the FPV and GPV modes, thus enabling the experiments to compare the
performances of any FPV system variant and its GPV counterpart, which inherently also provides a measure
of the error that the user would encounter if employing the conventional GPV analysis in an FPV project
assessment.
In this context, the proposed universal pragmatic software implementation of the proposed integrated
floatovoltaic synthesised model is set to implement the 3E (energy, economic, environmental) software
objects in Figure 4.3 as FPV-GIS model kernels. The goal is thus to physically implement the model in
Figure 4.3 as Python-prototyping source code to dovetail the newly proposed economic and environmental
software kernels (economic and environmental objects) of this thesis with Sandia-developed PVlib Python
software development kit procedures (energy object) (Andrews, 2018; Enthought, 2017). It aims at a soft-
ware implementation suitable for integration into a descriptive floating-PV technology information system.
As such, the Python code establishes a holistic systems-level analytical expression (Habermas, 1981) by
creating a software-implemented digital twin model, equivalent to a floating PV system, in terms of Python
geo-engineering software procedures. From an operational and analytical assessment-framework perspec-
tive, the research aim and objectives have been set for the floating PV simulation model of Figure 4.3
to be built upon the sustainability reference framework of the system dynamics and logical architectural
programme of Figure 4.2. As an integrated geospatial data science tool for floating PV systems, the ar-
rangement of the system dynamics model of Figure 4.3 must be collectively cross-coupled in a triangular
arrangement of software objects, an agile programming aspect for which Python’s object-oriented program-
173
ming capabilities and code architecture offer valuable options for software implementation in object/agent
orientated programming (Flores-Mendez, 2014; Vlacic et al., 2000).
Towards the computer modelling and synthesised methodology of the research, Python language is
engaged to implement the transdisciplinary mathematical-simulation formulae for the model object of each
subsystem in its particular knowledge domain (energy, economic, and environmental subsystem objects).
As such, the custom-designed computer software code to implement the floating PV-synthesised model,
Figure 4.3 is tailor-made and soft-encoded to condition the model software to synthesise floating PV-
ecosystem operations uniquely. One of the strategic goals of the Python model-development strategy is
to actualise a contextually-sensitive floating-solar synthesised tool that can universally operate in any coun-
try and different farming, financial, currency, and legislative contexts. Design thinking helped to deliver
contextual sensitivity through a user-configurable dynamic conditioning strategy. The user-configuration file
allows around 40 parameters to be set up for the floating PV system installation, spanning the spectrum of
the energy, environmental and economic system parameters. This configuration file option for FPV allows
for a more resilient floating PV model capable of ensuring that contextually-sensitive floating solar system
simulations can comply with the regulations, practices and policies of various countries, regions, districts,
and application contexts.
From a Systems Science organisational perspective, it is essential for the Python software development
implementation to numerically model each of the systemic object functions of the system in the enterprise
kernel outlined in Figure 4.3. According to Research Objective 2, it is further crucial for the Python imple-
mentation to numerically model all the systems components, together with all the bi-directional transactive
interactions and information exchanges in the structural architecture in the topology of Figure 4.3. Mod-
elling the floatovoltaic ecosystem’s static and dynamic behaviour in terms of the collective intelligence of
Figure 4.3 is crucial to improving its performance-prediction integrity because it emulates real-world interac-
tions. As such, because the model inherently includes the memory of previous events through inter-domain
feedback loops in a co-simulation fashion, the spatiotemporal modelling framework of Figure 4.3 accommo-
dates the successional dynamics of the system. Memory benefits are derived from the model’s system dy-
namic network structure as it treats the object-based representations as time-based entities in a closed-loop
format. As a space-time operational phenomenon within the digital model representation of the floatovoltaic
ecosystem, the location-based representation offers a complementary geospatial-modelling approach.
To convert the FPV design concept of Figure 4.3 into reality, the simulation of the extract of the Python
code, as depicted in Figure 4.19, implements the discrete-time floatovoltaic synthesised model. While the
Python code for the system dynamics model is developed around the PVlib Python software development
kit (SDK) (Holmgren et al., 2018; Klise and Stein, 2009), continuous software calls regularly interface with
the online applications programming interface (API) for PVlib to perform algorithmic solar-positioned calcu-
lations, together with complex simulations of photovoltaic PV systems.
While PVlib Python offers an object-oriented computational physics implementation of land-based PV-
modelling functions, descriptions of the use of the PVlib simulation engine as PVlib SDK and API calls are
available online from the GitHub platform catalogue (Sandia, 2020). Sandia Labs specially developed the
PVlib Python for the Pythonic language to interface with user-configurable applications like those developed
in this thesis. With the floatovoltaic ecosystem being a solar-event-driven production system, the Python
implementation requires inputs from the solar simulator of the environmental model at every time-instance
of the real-time simulation procedure. For this purpose, the environmental software model in Figure 4.3 uses
Python semantic constructions to import online meteorological data from the chosen resource dictionaries.
This aspect constitutes part of the supply chain logistics of the floatovoltaic ecosystem. The energy model
software procedure of Figure 4.3 then continually processes this data by engaging programme libraries or
exchanging resources. This process includes interacting with the Sandia PVlib simulation libraries, which
model a PV system as a fundamental and applied physics problem (Sandia, 2020). Other interactions
174
Figure 4.19: Screen-print extract of the integrated Python-coded discrete-time floatovoltaic synthesised model imple-
mented as recurring system dynamics model expressions (source: author).
include the importation of environmental dynamics and weather station metadata baskets from a chosen
provider database (such as PVGIS-SARAH) through the relevant providers (HelioClim, 2022; NREL, 2019b;
PVGIS, 2021), or by accessing solar-resource assessment data as the typical meteorological year (TMY)
weather data forecasted by stochastic weather models in mixed-effects modelling (SolarGIS, 2018; PVGIS,
2021).
From a real-time mathematical modelling perspective, it is important to note that the Python floatovoltaic
ecosystem simulation model depicted in Figure 4.3 operates as a discrete-time system. This statement is
another way of saying that the simulation runs on an hour-by-hour (or minute-by-minute) basis over periods
of a full day, full week, full year, or the operating lifetime of the project (depending on the analysis fre-
quency set in the model configuration, and the availability of weather station data). It’s dynamic processing
essentially implements the temporal dynamics of the cognition framework of Figure 3.18 in a computer-
synthesised model to produce hour-by-hour analytical predictions about the performance of a floating PV
ecosystem. As a real-time simulation, for which the implementation of the Python-code simulation is de-
picted in Figure 4.19, the analytical floating PV systems model of Figure 4.3 performs discrete-time elec-
tronic data processing according to the cyclical sliding-window procedure illustrated earlier in Figure 4.4.
Note that, through the procedure presented in Figure 4.4, orchestrated data processing is executed concur-
rently within and among all three of the 3E domain objects, and according to the recurring system dynamics
and stock-flow diagram of Figure 4.3 and Figure H.1 respectively. In this way, the responsive model of this
thesis models both the steady-state and the transient behavioural properties of the energy enterprises of
floatovoltaic systems in discrete time intervals of a specific frequency of analysis as time-varying functions.
During the Python model’s discrete-time-simulation run-time operations, the FPV-GIS model iteratively
processes the 3E workflow by stepping through each of the 3E sub-system object components during each
time instance of the simulation synthesis, as shown in Figure K.1. In this iterative 3E network processing
procedure, the operations and interactions of the floatovoltaic model objects mimic the steady-state and
transient state responses of the floatovoltaic ecosystem (Figure 4.3) in digital terms (the so-called digital twin
model). As such, the digital twin model (the virtual floatovoltaic power-plant model) of Figure 4.3 emulates
a real-world floatovoltaic ecosystem, providing real-time outputs as time-varying parameters that represent
or mimic the respective energy, environmental and economic domain performances of a planned real-world
floatovoltaic project. Important to note is that the FPV synthesised model offers a multi-scale numerical
simulation model based concurrently on loop processing. It means that the processing of the quantitative
technical energy, economic and environmental performance processing of the performance efficiencies of
the project elements are determined in the loop or in tandem for every discrete time cycle of the simulation
175
according to Figure 4.4. In this discrete data-processing methodology of the model in Figure 4.4, the data-
processing cycle performs a series of simulation-processing steps. As such, the raw geographical data and
technical systems data (inputs) are recurrently fed into the processing unit (the FPV simulation model) in
each time interval to produce results (techno-enviro-economic outputs) and actionable insights (decision-
profile maps). Each analytical processing step by the synthesised model of the floatovoltaic system depicted
in Figure 4.3 is thus processed in a specific order, with one cycle of a data-processing output being stored
and also being fed as an input into the next processing cycle (refer to Figure K.1). Data processing is
performed for each data-sampling occurrence. At the same time, the entire process is repeated cyclically
over the project’s lifetime to provide real-time and life-cycle performance data about the floating PV system’s
predicted impact effects and performance. The simulated performance-output parameters and values are
continually written into a "Results.csv” file as the execution of the Python model progress (refer to Figure 4.4)
through the simulated time horizon set for the simulation experiment or over the entire lifetime of the project.
With the systemic and component-level details of the computer coding and the soft-simulation model
described, the following section summarises the functionality of the integrated geographical model-driven
simulation instrument as a research tool suitable for conducting experimental desktop evaluations of the
floating PV systems model in real-world case scenarios.
4.5.2. Geoinformatics PV-analysis model and decision-support toolset
This section provides an overview of the integration of the research model, where the system components
are incorporated into the geoinformatics research model and its component layers. In characterising this
digital model, it can likewise use the strategic decision-support properties of the proposed geoinformatics
model to evaluate contextually-sensitive environmental scoping and economic due-diligence advisements.
From a software development point of view, the Geography and GeoInformation Science focus of this
thesis offer the opportunity to prepare the software as kernels for Figure 4.3 to be suitable for the ultimate
implementation of the proposed FPV-GIS model in a geospatial cloud computing environment (Danger-
mond, 2018; Yang et al., 2011). The integrated model would thus support project developers and allow
them to win regulatory approval for new projects, especially in terms of the potential integration of hypothet-
ically planned floating-solar/land-based systems in the South African agricultural sector. To this end, the
research methodology would be able to deliver an integrated floating PV simulation and decision-supporting
toolkit to study the reality of interest, namely floatovoltaic system installation sustainability.
The goal is for the integrated FPV-GIS model to operate as a desktop GIS-enabled floatovoltaic synthe-
sised toolset. As such, the digital desktop model integration would essentially also offer a novel multivariate
analytical tool as a geomatics solution in the form of a digital twin model or FPV-geomodel. The entire
structure of the desktop research tool, with software objects making up its internal architecture, is defined
and implemented according to the model archetype of Figure M.1, thus offering a combined physics-based
data-driven performance-forecasting model for pre-planning assessments of floating PV technology instal-
lations. The geoinformatics sustainability performance and impact indicators output by the analytical layers
in Figure M.1 are passed to the decision layer in Figure M.1. In the decision layer the quantitative 3E
performance metrics are translated into ce-WELF sustainability indicators, which are consolidated into a
sustainability index, and input into the AHP-based numerical decision-support processing. The ce-WELF
index vectors are then projected on a profile-mapping display, delivering a sustainability representation to
study the performances and benefits of future planned floatovoltaic systems according to the proposed
sustainability theory and tentative framework as the thesis’s hypotheses. True to the philosophical goals
for a systemic theoretical sustainability modelling concept in Figure 3.3, the theoretical conceptualisation of
Figure 3.14, operationalised in a geospatial digital twin model of Figure M.1, delivers a computer model that
176
integrates behaviour synthesis, performance evaluation, and impact assessment all in a single consolidated
sustainability modelling synthesis.
From an analytical perspective, the geographical floatovoltaic system simulation methodology and im-
plemented model abstraction in Figure M.1 characterise and emulate the recurring dynamic operations of
a virtual floatovoltaic power plant ecosystem in a portable desktop tool model (virtual digital twin model). In
its analytical modality, the discrete-time simulation model mimics the operational dynamics of a real-world
floatovoltaic ecosystem in the digital realm. As such, the virtual floatovoltaic equivalent imitates the geo-
graphical performance characteristics and impact qualities of an analogous floatovoltaic energy system in
a pre-defined geospatial context. This feature makes the model ideal to function as a desktop research in-
strument to study the performance profiles and impact effects of future planning about floating PV systems.
The geospatial tool model of Figure M.1 further offers functionality in multi-objective decision support
in the floatovoltaic system planning domain. In the decision modality, the tool computes a composite set
of hierarchical metric values in the analytics domain articulated on a geo-aware profile-mapping interface
(spider diagram) to aid project decision-making around the WELF nexus to aid in the understanding of the
agricultural impacts of a future planned floatovoltaic installation. As such, sustainability indicators can be
processed into sustainability indices on an electronic sustainability scorecard, while the multi-criteria profile
mapping approach can be applied concomitantly, the multi-criteria profile mapping approach (ARUP, 2019;
Collins et al., 2010) can plot customary floatovoltaic sustainability indicators and goal priorities as auxiliary
units on Cartesian coordinates. This visualisation can serve as a means to compare various sustainability-
scenario narratives in FPV planning.
This section summarises the features and functionality of the proposed geomatics desktop as a research
instrument for application in scenario-driven case-study experiments according to Research Objectives 1 to
3. The following section addresses the experimental design aspects in preparation for the model evaluation.
It describes the experimental valuation procedures whereby the geoinformatics toolset is used in quantita-
tive data collection and data gathering towards answering research questions through scenario evaluations
of experimental projects.
4.6. Experimental Design, Data Collection and Evaluation
This section provides an overview of the research validation process regarding the methodological approach
proposed in this study. It is intended to provide a framework for a set of research experiments designed
around theoretical framework evaluation. The overarching goal of the experimental evaluation is to answer
the research questions by referring to empirical case-based scenario exercises as the basis. The experi-
mental testing phase generally takes the research project through the process of ascertaining whether the
empirical evidence in floatovoltaic planning exercises can answer the research questions.
With the core functionality of the digital geoinformatics model and the integrated FPV-GIS model estab-
lished as a system dynamics simulation instrument, the point is reached where the sampling of case data in
a responsive desktop-computerised environment can be conducted. To this end, the experimental section
of this study engages the prototype of the conceptualised geoinformatics model for data sampling and data
analytics processing to demonstrate the accelerated life-cycle testing and impact effects analysis of the pro-
posed floating PV model. The judicial perimeter for the SA Renewable Energy EIA Application Database
on the EGIS platform DFFE (2021b) serves as a demarcated study area for experimental FPV-GIS desktop
applications. While the operations of the FPV-GIS model need to be evaluated within this geospatial area,
a hypothetical pilot project is defined to test and evaluate the research questions in a baseline case-based
scenario analysis. A comparative decision-support experiment is also conducted for a pre-configured floa-
tovoltaic system and a land-based photovoltaic system in a behind-the-meter grid-substitution application.
177
With the development of the predictive model completed, case studies must then be drafted and con-
ducted to test the design stage floating PV characterisation method and model as a means to elucidate the
issues relating to systems performance and impact effects in real-time reliability studies. In the first experi-
ment, a quantitative model evaluation methodology is followed to test and confirm that the computer model
expressions have been correctly encoded into the computer programme. In the first part of the evaluation
procedure, the validity of the floating PV modelling endeavours is evaluated mainly through an analysis of
the predicted energy of the model. The goal of this model-corroborating experiment is to determine any
potential deviation that the results might be subject to when the empirical model outputs obtained from the
computer-synthesised test runs with the empirical data extracted from the module datasheet loaded from
the Sandia Module Database.
In the second experiment, the focus is on a quantitative project evaluation methodology executed in an
operational dynamics environment. In this experiment, the researcher (as an environmental impact assess-
ment practitioner) defines a typical use-case scenario to use the functionality of the conceptualised geoinfor-
matics analytical model to collect empirical data for floating PV systems characterisation and impact-effects
profiling. The experiment further intends to demonstrate the data collection and data archiving capabilities
of the FPV-GIS model and toolset. A sample test case is defined to quantitatively establish the baseline
sustainability performance of a hypothetical pilot floating solar system at an appropriate and arbitrarily se-
lected location in the Western Cape region of South Africa. This experiment aims to use the designed
FPV-GIS toolset and the data-capturing and analytical capabilities of this geoinformatics toolset in a de-
sign stage FPV performance evaluation synopsis experiment. This analysis-by-synthesis experiment is
conducted to determine the project’s indicative impacts and the savings and return-on-investment impacts.
Another aspect is to compare the analytical capabilities of the conceptualised model in terms of the abil-
ity of the model to empirically predict the projected sustainability indicators (energy yield, environmental
offsets, and economic impacts) for a baseline floatovoltaic project plan. In this context, the goal is to com-
pare the simulated sustainability-performance outputs for the planned floatovoltaic system with those of a
hypothetically-planned and related land-based photovoltaic solar system of similar size and proximity to the
exact geographical location (adjacent to the reservoir, on the same farm).
In the third experiment, the decision-evaluation methodology uses the geoinformatics decision-support
portion of the conceptualised model to collect data around the quantitative variability of decision support for
the test-case scenario from the integrated geoinformatics model. This experiment step aims to demonstrate
the decision-support properties of the conceptualised model. It intends to demonstrate the flexibility of
the dynamic geoinformatics model configuration, as well as its portfolio-mapping capabilities. The goal
is to empirically justify the validity of the proposed new framework and the computer model expression
in applying decision-information management. This experiment, therefore, aims to provide quantitative
insights into the impacts of decision-making factors on the variations in the sustainability performance of
the floating-solar/land-based photovoltaic systems, given certain variations in the planning parameters of
the system or the baseline study and as part of the process to validate/refute the hypothesis. Relying on
the ability to identify generic leverage points where effective interventions can help to create balance in
sustainability, the analysis-by-synthesis approach finds decision areas where small design efforts may lead
to bigger payoffs. The profile-mapping aim is to find criteria areas to intervene or relieve technological
uncertainty and help decision-makers with decisive planning actions or strategic recommendations.
While this section briefly introduced the experimental procedures for evaluating the conceptual modelling
and simulation of analytics and decision support in floating solar projects, the following section deals with
the ethical considerations for this research and the relevant experiments to be conducted.
178
4.7. Ethical considerations
In the experiments covered by this project, the researcher endorses the Ethical Codes and Ethics Policy of
the University of South Africa (Unisa, 2013). These ethical considerations (under the project’s ethics clear-
ance number 2020/CAES_HREC/125) require that research be conducted in an ethically-sensitive manner,
that respect is shown to the research participants, and that their basic human rights be honoured. It further
states that no conflicts of interest or financial benefit should materially affect the investigation’s outcome or
jeopardise the name of the University. Data collection and analysis should take place under the appropriate
controls, while all results, including any adverse findings, should be reported accordingly (Cossio, 2012;
Unisa, 2013). Regarding the Unisa Ethics Policy, participation in the project is voluntary, and any participant
has the right to withdraw at any time. While the research naturally draws attention to the range of attractive
salient benefits of floating solar technology, it simply aims to support the intervention of the South African
DFFE (2019a) National Climate Change Adaptation Strategy towards generating technology-domain aware-
ness in the South African landscape. In this context, the proceeds from this research may naturally support
this country’s climate-change goals through the diffusion of floatovoltaic technology in SA and the 4IR vision
of the Fourth Industrial Revolution Partnership for South Africa (4IRSA, 2020).
It is emphasised that this study is driven primarily by the knowledge gap in the literature and not through
any commercial or marketing need. Furthermore, the study does not serve as a marketing exercise for
any commercial GIS tool or floating solar panels; neither is the study directly nor indirectly sponsored for
commercial benefit. Moreover, as an independent researcher, the candidate is in no way involved in selling
floating solar devices or marketing or branding any floating solar labels or equipment.
4.8. Summary
In working towards answering Research Question 2, this chapter pronounces details of the research method-
ology and procedures toward realising the research aim through Research Objectives 2 and 3. Then,
guided by the scope of the research-design layout portrayed in Figure 1.7, it details the specifics of the
underlying model-based design-engineering and system dynamics modelling techniques as fundamental
building blocks in establishing the synthesised floatovoltaic model in terms of the aim and objectives of
the research. The discussion provides a step-by-step guide to implementing a synthesised model of the
floatovoltaic ecosystem and establishes a research method that facilitates the research experiments.
In this context, this chapter sits at the heart of the thesis as it navigates the reader through the main
details of the philosophical approach and the underlying theoretical concepts in establishing a synthesised
digital floatovoltaic model as part of the proposed geoinformatics decision-support system. While detailing
the philosophical approach, the chapter also describes the methodological approach and research method-
ology, together with the design and development of the proposed FPV analytical framework and model. Fi-
nally, the chapter further narrates the decision-support methodology and computer analysis model, as well
as the integration method and systemic structure of the GIS-based geoinformatics platform. This contextual
approach narrates the methodological design and modelling towards establishing the FPV geoinformatics
model in terms of the Research Aim and Research Objectives 1, 2, and 3, towards the tactical implementa-
tion and experimental evaluation model for a proposed experimental digital floating solar ecosystem.
The methodological design of this research instrument prepares the thesis for the experimental eval-
uation of the FPV model and its internal mechanics (theoretical framework and computer model) that will
be used to test the research objectives and answer the research questions in the next chapter. The next
chapter details the experimental validation of the proposed technique by applying the ethical considerations
associated with the research project and the planned experiments.
179
5. Experimental Results and Discussion
5.1. Introduction
This chapter deals mainly with evaluating the research methodology and the implementation of the model,
refocusing on a value assessment of the research outcomes emanating from the Research Aim and Re-
search Objectives 1 to 3. The research philosophy and methodology in Chapters 3 and 4 respectively deal
with the analytical considerations and design basis to actualise an evaluation toolset for an experimental
simulation-based floating solar energy system operating as a digital twin model. This chapter presents
quantitative experimental results based on the case and simulation-based project enactment and data col-
lection procedures regarding the FPV-GIS toolset of Figure M.1. The experimental results obtained in
the context of the research assumptions, detailed under Section 1.5, as well as the ethical and experi-
mental considerations outlined in Sections 4.6 and 4.7 respectively, offer insights into the spatiotemporal
geo-dynamic responses of a scenario-based floatovoltaic installation.
Section 5.2 introduces the experimental layout, design and configuration procedures for responding to
the questions in the context of scenario-based experimental analyses for floating solar case-study narra-
tives. Three analysis-by-synthesis experiments are then conducted in Sections 5.3 to 5.5. They are based
on applying the proposed computer modelling and simulation methodology. The experiment in Section 5.3
deals with the corroborative verification of a floating solar PV systems model. The experiment in Section 5.4
then deals with performance analysis in comparing equivalent floating solar and ground-mounted PV sys-
tems. Towards making data-driven decisions through descriptive and inferential statistics, the decision-
support analysis experiment in Section 5.5 compares the economic WELF performances of the FPV and
GPV installations in terms of project-decision criteria. Finally, Section 5.6 summarises the chapter.
5.2. Experimental Evaluation Layout, Design and Configuration
In guiding the research through the geographical systems methodology and experimental procedures de-
scribed in the previous chapter, this section lays out the experimental grounds for engaging the project
design-stage performance prediction capabilities of the FPV geoinformatics analytical toolset in quantita-
tive experimental desktop-based analysis-by-synthesis studies. Following Section 4.6, it presents a layout
and overview of the experiments to study the reality of interest, namely floatovoltaic system installation
sustainability. This activity is realised by digitally conducting and evaluating sustainability impact assess-
ment studies for hypothetically-planned agricultural floating solar systems supplying renewable power in
agricultural grid-substitution applications in South Africa.
While the goal of this thesis was to develop a unique GIS platform, named the FPV-GIS toolset, as a
novel analytical tool to assess floatovoltaic projects in a dynamic geomatic FPV toolset environment, this
digital-synthesised model (the FPV-GIS toolset) engages in a quantitative experimental research methodol-
ogy to empirically characterise and assess the performance and impact profiles of floatovoltaic technology.
180
As such, the product of the research around this thesis offers an integrated geospatial floatovoltaic system
in the form of a digital twin model, as depicted in Figure M.1, as a dynamic research instrument capable of
acting as a digital proxy equivalent to a real-world floating PV installation. The FPV-GIS model a digital com-
puter model representation of floatovoltaics capable of mimicking the real-time responses of a real-world
floatovoltaic system on a geoinformatics platform. The FPV-GIS toolset is now ready for application as a
research instrument in scenario-driven case-study experiments. Having functionality as a transdisciplinary
FPV data-gathering research instrument, the digital twin model effectively establishes a system dynamics
and predictive analytical tool as a desktop research instrument that can be appropriated to conduct scien-
tific analysis-by-synthesis experimentation. Being driven by the novel philosophically-designed theoretical
modelling framework of Figure 3.14, the relevant experiments demonstrate the collective strength of the
interlocked disciplines according to the theoretical modelling framework of the FPV-GIS toolset proposed
by this thesis.
Regarding the strategic aspects of the experimental methodological approach, Figure 5.1thus provides a
layout for each experimental evaluation process. In terms of the specific experimental design, data collection
and evaluation procedures, Section 4.6 provides a methodological overview of the model-validation and
data-collection processes of the experimental research. Figure 5.1 defines the three tactical experiments.
Experiment 5.3 starts with the model’s evaluation and verification process to ensure that the model is, in
fact, representative of its intended application in terms of the theoretical accuracy of its performance results.
The experiment performs the simulation model’s quality tests using universal industrial protocols to verify
model accuracy while the test results are verified under de facto standard laboratory test conditions. The
experiment performs the simulation model’s quality tests using universal industrial verification protocols
to reference model accuracy in datasheet specifications, while the test results are verified under de facto
standard laboratory test conditions.
Experiment 2:
Floating PV Project Assessment
(Section 5.4)
Experiment 1:
Floating PV Model Verification
(Section 5.3)
Experimental Layout Design and Configuration
(Section 5.2)
Experimental Case-study
FPV Project assessments
Experiment 3:
Floating PV Project Decision Support
(Section 5.5)
Experimental Conclusions
(Section 5.3.4, Section 5.4.4 and Section 5.5.4)
Responding to Research Aim and Objectives
(Chapter 6)
Answering Research Questions,
Strategic Conclusions and Recommendations
(Chapter 6)
Figure 5.1: Experimental evaluation design of procedures and steps towards answering the research questions
around floating PV sustainability assessments (source: author).
Subsequently, the goal of Experiment 5.4 in Figure 5.1 is to build upon Experiment 5.3 through the en-
gagement of the digital twin capabilities of the FPV model in the operation of a real-world system and under
meteorological test conditions. This experiment uses the floating solar PV computer-synthesised model as
a desktop tool in the quantitative data collection and data analysis procedures to assess a user-configured
181
floating PV system at a selected real-world installation site during the project’s design stage. In this ex-
periment, the evaluation methodology enables the researcher, as an environmental impact assessment
practitioner, to collect experimental performance data for sample FPV test cases in arbitrary floating solar
system scenarios selected during the experimental analysis stage of the research.
Next, Experiment 5.5 in Figure 5.1 is defined to run the implemented FPV-GIS geo-informatics toolset
in the analytical decision-supporting mode. As such, the FPV-GIS model responses and site data collected
for the project design stage in Experiment 5.4 serve as input parameters to the desktop floating solar PV
decision-synthesised model. Remember that all of the experiments are conducted within the scope of the
research assumptions, as detailed under Section 1.5, Figure 5.1 demonstrates how the conclusions from the
three experiments serve as scientific evidence. In this capacity, the results are instrumental in responding
to the research aim and objectives; they answer the research questions; and present strategic conclusions
and recommendations for these experiments in the next chapter.
The ultimate aim of the three desktop computer experiments is to evaluate the extent to which the study
succeeded in establishing a new analytical methodology and analytical data science technique for predicting
the outcomes of an agricultural floating PV project installation compared to that of a ground-mounted PV
project. In addition, the experiments evaluate the extent to which the model could successfully integrate the
proposed theoretical framework and computer model expression into a geo-informatics decision-supporting
platform. Section 5.3 is concerned with floating PV energy model validation, Section 5.4 performs an FPV
project performance assessment, while Section 5.5 performs an FPV decision-support analysis.
5.3. Experiment 1: Corroborative Floating PV Model Validation
The first experiment aims to generate scientific test results to help answer the research questions around the
success of floating PV systems modelling. The goal is to create experimental results around the operational
accuracy of the PV power model for the implemented analytical FPV-GIS toolset. While the experiment
requires a reliable and straightforward theoretical method to evaluate the performance of the FPV model,
this experiment engages standardised PV modelling test conditions and industrial-standardised protocols
to determine, as intended, the extent to which the implemented computer model and code of the FPV-
GIS toolset are correct and functioning. As such, the experiment details the selected quantitative model
evaluation methodology to confirm whether key computer model expressions for the floating PV model have
been correctly encoded into the FPV-GIS toolset computer program code.
According to the EPA (2009), the process of theoretical model evaluation sets the goal to determine
whether a computer model and its analytical results sufficiently agree with the known data in specific di-
mensions and can resolve the analytical problem to a degree suitable for informed decision-making. As
such, the Modelling Guidance Document of the EPA (2009) recommends that the concept of model valida-
tion be based on theoretical principles for model corroboration. This recommendation is precious in cases
where outdoor field measurement data is inaccessible, under proprietary ownership, or the statistical sam-
ple size is insufficient (as is the case with the short lifetime of proprietarily-installed agricultural floating PV
systems in South Africa). While field-validated models are those that have been shown to correspond to a
specific set of field-measured data for a particular procedure, theoretical model corroboration focuses on the
processes and techniques for theoretical model data evaluation. It is based on laboratory-based evaluation
tests to be compared with scientifically-verified reference data.
In theoretical terms, corroborative model validation is a quantitative process for evaluating the degree
to which the PV model corresponds to reality by confirming other theoretical results referenced in credible
industrial research databases (EPA, 2009; Thacker et al., 2004). While the primary function of a PV module
is to convert solar radiation directly into electrical power, Sandia National Labs (SNL) provides a credi-
182
ble scientific procedure for simplified theoretical model validation that evaluates the accuracy of predicting
photovoltaic model performance based on power performance predictions for test modules at a given op-
erating point (De Soto et al., 2006; Stein et al., 2011; Whitaker and Newmiller, 1998). In this context, the
universally-used Sandia PV model-evaluation procedure follows a set procedure for evaluating the energy
performance of photovoltaic systems to determine the impact of performance accuracy on project risk as-
sessments (Kurtz et al., 2013; Stein et al., 2011). In terms of this theoretical evaluation procedure, the power
generation of a photovoltaic system is evaluated and documented by way of an energy capacity test (Klise
et al., 2013; Sandia, 2022c). In such a test, the experiment quantifies the power output of a photovoltaic
systems model under set conditions defined as the standardised photovoltaic test conditions (Abdullahi
et al., 2017; Hukseflux, 2021; Kurtz et al., 2013). The rating performance of a selected photovoltaic cell
is typically confirmed via the name-plate power rating on the panel, as specified by the manufacturer or
supplier of the PV module, and typically under specifically referenced test conditions.
The goal of this corroborative validation experiment is described in Section 5.3.1, while the experimental
method and procedures are outlined in Section 5.3.2. Section 5.3.3 details the results of the empirical case
study, which, as set out in Section 5.3.4, are used to draw the research conclusions.
5.3.1. Goal of experiment
The corroborative verification process for the model can confidently engage the predictive capabilities for
power production of the FPV-GIS model to evaluate the energy model’s accuracy in the standardised verifi-
cation tests for power production. In this context, the evaluation of this FPV performance model experiment
aims to quantify any potential deviation of the power-performance results when comparing the empirical
model outputs obtained from the computer-synthesised test runs for a selected photovoltaic module with
those of the benchmark (Marion et al., 2014; Whitaker et al., 1997). The benchmark refers to scientifically-
verified reference data for the chosen standardised test module at maximum power-point performance
(Reza Reisi et al., 2013). In operational terms, this technical evaluation experiment sets the goal to fol-
low the validation guideline procedures of NREL, for which SNL, as the photovoltaic research institution,
developed an internationally-recognised evaluation procedure and standards to validate PV systems de-
signs and PV power-performance models in terms of their nominal performance and potential impact on
project risks (IEA, 2017; Stein et al., 2011).
In this context, this experiment aims to provide consistency by following the SNL procedure that uses
the technical performance feature of maximum power/energy yield as a critical metric directly used as the
validation metric for the PV performance model (Sandia, 2022c). The maximum power point energy yield is
a valuable reference metric in model verification, especially since the power/energy metric feature is often
further processed into other derived metrical forms (such as the derivative economic and environmental per-
formance features (Prinsloo et al., 2021)). As such, the model-based verification goal and approaches aim
to evaluate model prediction accuracy and computational performance by comparing the model outputs
of the FPV-GIS analytical toolset with that of scientifically-verified reference data. This process involves
comparing the empirical model energy outputs obtained from the computer-synthesised test runs for a se-
lected photovoltaic module with published reference data for the chosen PV test module under standardised
SNL-defined operating conditions.
5.3.2. Experimental method and procedure
This section outlines the experimental method and procedures to evaluate whether the fidelity of the FPV-
GIS model is sufficient for its intended use as an experimental toolset and whether the qualitative evaluations
of the model corroborate its application (EPA, 2022). Regarding the research methodology, the experimen-
183
tal method and procedures to substantiate the FPV model in this experiment are listed in Figure 5.2. In
navigating the model-validation process, the experiment starts with experimental definitions in Steps 1 & 2,
continues to validate the model in Steps 3 to 7, concluding with experimental findings in Step 8.
step 1 step 2 step 3 step 4 step 5 step 6 step 7 step 8
Select the test
protocol and
define the test
conditions
Select test
PV module
and tabulate
datasheet
reference data
Configure
FPV model
to simulate
test conditions
FPV model
uploads
the PV test
module
parameters
Perform
runtime
simulations for
test module
under test
conditions
Tabulate
model
simulated
outputs for
comparison
with verified
reference data
Compute
relative error
deviation
for model
simulated vs
reference data
Derive
experimental
conclusions
Figure 5.2: Experimental procedure for corroborative floating PV performance model validation, according to the
defined experimental methodology (source: author).
The first step in the experimental synthesis method of Figure 5.2 is to select the test protocol and to
define the referenced test conditions under which to evaluate the FPV-GIS model in terms of the nominal
performance of the test models. In terms of the test protocol, the NREL/SNL model verification procedure
operates as a consensus-based approach, in which the FPV performance model is validated for a specific
photovoltaic module in a procedure intended to address the de facto standardised module power rating for
solar PV cells at Standard Test Conditions (STC) (Stein et al., 2011). A similar procedure can validate a PV
model based on a PV project rating according to the so-called PVUSA Test Conditions (PTC), whereby the
PTC procedure specifies the ambient conditions rather than the module conditions (Whitaker and Newmiller,
1998). Under the STC, performance testing of the PV model is conducted to determine its performance in
terms of power output under a laboratory-specific set of fixed module conditions (solar irradiance of 1000
W/m2, an ambient temperature of 25◦C, an airmass of 1.5, and a wind speed of 0 m/s). Under the related
PTC conditions, PV-performance testing is conducted to determine power-output performance at the set of
fixed laboratory-specific ambient module conditions (solar irradiance of 1000 W/m2, an ambient temperature
of 20◦C, an airmass of 1.5, and a wind speed of 1 m/s).
While the procedure of this experiment intends to implement simulated FPV-GIS model-evaluation pro-
cedures under both STC and PTC conditions, Table 5.1 summarises the definitions for the corroborative
validation of the FPV-GIS photovoltaic systemically-synthesised model for a chosen test PV module under
the laboratory-specific test conditions for an STC and PTC ambient module. These test conditions ensure
consistency in model testing and reduce the effects of the so-called environmental derate factor of the PV
model.
Table 5.1: Comparative test condition parameters for the STC and PTC model evaluation tests of the model evaluation
experiment (after: Sandia (2022b)).
Experimental Test Conditions
Cell Parameter STC Test Conditions PTC Test Conditions
Number of PV panels 1.00 1.00
Solar cell Temp 25.00 ◦C 45.00 ◦C
Ambient Temp 25.00 ◦C 20.00 ◦C
Wind speed 0.00 m/s 1.00 m/s
Air mass 1.50 1.50
184
Moving to module selection, the reference test is the basis for the comparative theoretical evaluation
of the model, as in Step 2 of Figure 5.2. This experiment selects the universally-used Sandia Module,
tagged "Canadian Solar CS5P-220M”, as the test module (Sandia, 2022b). While the online Sandia PV
module library hosts a software warehouse for all scientifically-verified solar modules, together with their
technical configurations and performance data, this specific module is used in this experiment to compare
the predictive accuracy of the FPV-GIS model’s performance relative to the scientifically-verified referenced
data for this test module. This solar module design has a nominal nameplate rating of 220W. Within the
online Sandia database, the datasheet for this reference model is tagged under the PV catalogue reference
label, Sandia-Module = "Canadian_Solar_CS5P_220M_2009_” (Sandia, 2022b). The Sandia PVlib library
database lists the physical characteristics of the solar panel and the PV performance data for operating the
solar panel of this test module under STC and PTC test conditions in the datasheet extract, as in Table 5.2
(Sandia, 2022b).
As such, the scientifically-verified solar cell datasheet parameters in Table 5.2 serve as scientifically-
verified power-referenced characteristics for the test module, "Sandia Module Canadian Solar CS5P-220M”,
to serve as reference data in this FPV-GIS model verification experiment. Note that the selected PV module
is not the universal standard used in model verification, although it is commonly used as a reference in PV
performance evaluation experiments. The user-configuration file for the FPV-GIS model allows the selected
PV module to be replaced with any module listed on the online Sandia Library database. Note that the
Sandia PV Performance Modelling Collaborative can lab-test any (locally) manufactured PV module for
datasheet referencing on the downloadable Sandia database (Sandia, 2022c), meaning any generalised
South African PV module can be created by Sandia provided that sufficient manufacturing samples/data
are made available for lab testing.
Table 5.2: Scientifically-verified datasheet referenced characteristics for the experimental Sandia Module Canadian
Solar CS5P-220M (after: (Sandia, 2022b)).
Scientifically-verified Solar Cell Datasheet Parameters
Cell Parameter Reference Rating
Single Panel area 1.70 m2
STC Power Rating 220.00 W
PTC Power Rating 200.10 W
Power Tolerances 0 %/+2 %
STC Power per unit of area 129.40 W/m2
STC Peak Power Efficiency 12.94 %
PTC Peak Power Efficiency 11.77 %
Imp 4.68 A
Vmp 47.00 V
Isc 5.01 A
Voc 58.80 V
NOCT (nominal operating cell temp.) 45.00 ◦C
Temp. Coefficient of Power -0.45 %/K
Moving on to the execution steps in the experimental procedure listed under Figure 5.2, Step 3 con-
figures the FPV-GIS toolset model to simulate the STC and PTC test conditions, as defined in Table 5.1,
for a given photovoltaic systems panel as a test module. In Step 4 of this experiment, the FPV-GIS model
uploads the PV test module parameters and technical solar cell details for the experimental solar module,
"Sandia Module Canadian Solar CS5P-220M”, from the online Sandia-shared database library. In this way,
the FPV-GIS model is programmed and configured with the parameter details of the solar cell for the se-
lected solar test module operating in a stand-alone module configuration. The test configuration of the solar
PV model is thus set up to determine the maximum power point performances (Reza Reisi et al., 2013) in
185
Step 5 of the experiment, namely to perform a run-time simulation of the "Sandia Module Canadian Solar
CS5P-220M” module under STC and PTC test conditions and to record the simulated maximum power point
data-output equivalents for the evaluation parameters of the FPV-GIS power model for the test module. With
the laboratory-confirmed referenced output data available for the PV test module in Table 5.2, Experimental
Step 6 compares the simulated output of the maximum power point data of the FPV-GIS model for the test
module under STC and PTC test conditions, with the scientifically-verified referenced data under STC or
PTC test conditions, as listed in Table 5.2. Experimental Step 7 then determines the relative error at the
maximum power point prediction between the module-simulated PV-performance data and the datasheet
module-referenced PV-performance data in Table 5.2. As such, the accuracy of the FPV-GIS model is
investigated by evaluating the percentage relative error (RE) between the FPV-simulated results and the
Sandia-referenced datasheet results under STC/PTC conditions at the maximum power point performance
scale.
The process mentioned above allows for the execution of Experimental Step 8, which is to derive scien-
tific conclusions from the experimental results. Note that the method and procedures for the corroborative
validation of the model, as listed under Figure 5.2, do not need a selected specific geographical location or
geographical region for testing the validity of the model. This exemption is valid because the SNL model
validation procedure is based on a lab-based analysis-by-synthesis procedure for a results-based technical
evaluation around the model’s performance parameters at the specific STC/PTC ambient operating condi-
tions applied under laboratory conditions. Also, note that while the measurement of the PV module’s power
is the most common testing method and metric to validate the accuracy of the model, the maximum power
point (MPP) or maximum power production (Pmax or Pm) for a given temperature profile for the solar cell
or test module serves as a key validating metric reference (Sandia, 2022c). Finally, the FPV-GIS model
predictions can be compared with the empirical data extracted from the test module datasheet and stored
in the Sandia module database for the selected module at the maximum power point. The following section
presents the experimental results.
5.3.3. Experimental case-study results
With the FPV-GIS toolset model configured to simulate the STC and PCT test conditions, as defined in
Table 5.1, the FPV-GIS energy model is then able to load the module parameters for the selected referenced
solar PV test module, "Sandia Module Canadian Solar CS5P-220M”, from the online Sandia (2022b) library.
The FPV-GIS model is configured and programmed to execute the run-time simulation for the "Sandia
Module Canadian Solar CS5P-220M” module under STC/PTC test conditions. With the run-time execution
of the synthesis model, the PV module characterisation procedure within the analytical layer of the FPV-GIS
toolset profiles the output characteristics of the selected PV module and logs the predicted experimental
results, or nominal performance of the PV system model, for the test module running under STC and PTC
conditions in Table 5.3.
Table 5.3 records the simulated outputs of the model as a means to gauge its predicted performance in
comparison with the verified reference data, as presented in Table 5.2. With Table 5.3 depicting the power
output results of the PV module simulated for the default Sandia-Module in the test configuration of the
STC/PTC PV system at a given time interval, the experiment is then able to investigate the accuracy of
the FPV-GIS model by evaluating the percentage relative error (RE) between the simulated results of the
FPV-GIS and the Sandia-referenced datasheet results under STC/PTC conditions at the maximum power
point performance scale. As such, the accuracy of the PV power-output prediction model is evaluated based
on the performance values for the Sandia datasheet for the selected PV module under the STC/PTC test
conditions, as defined in Table 5.1.
186
Table 5.3: Experimental simulated power characteristics of the test results for the Sandia Module, Canadian Solar
CS5P-220M, under STC and PTC test conditions (source: author).
Experimental Test Results
Cell Parameter STC Test Conditions PTC Test Conditions
Number of panels 1.00 1.00
Solar cell Temp 25.00 ◦C 45.00 ◦C
Ambient Temp 25.00 ◦C 20.00 ◦C
Wind speed 0 m/s 1 m/s
Pmp 219.66 W 198.96 W
Isc 5.09 A 5.13 A
Vmp 48.32 V 43.61 V
Imp 4.55 A 4.56 A
Voc 59.26 V 54.92 V
Efficiency (η) 12.92 % 11.70 %
While the error of prediction of the PV model quantifies the overprediction or underprediction of energy
production for the timing-interval snapshot presented in Table 5.3, the experimental evaluation can subse-
quently determine the degree to which the FPV-GIS model is accurate in predicting the state of a physical
system under STC/PTC operating conditions at the given MPP. From the run-time execution data of the
PV model, as recorded in Table 5.3, the comparative results in Table 5.4 can then compute the numerical
performance accuracies and error deviations of the FPV-GIS model.
Table 5.4: Comparative error deviations between the model-simulated and referenced module-performance values at
MPP under STC and PTC test conditions (source: author).
Experimental Test Comparisons
Parameters Datasheet Simulated %RE relative error Model accuracy
STC Pmax (W) 220.00 W 219.66 W 0.15% 99.85%
STC Efficiency (η) 12.94% 12.92% 0.16% 99.84%
PTC Pmax (W) 200.10 W 198.96 W 0.57% 99.43%
PTC Efficiency (η) 11.77% 11.70% 0.60% 99.40%
The recorded model performance accuracies and error margins are based on comparisons between
the Sandia-verified referenced module values and the FPV-simulated MPP values derived under simulated
STC and PTC operating/ambient conditions. In terms of the relative error of maximum power output (Pmax)
in Table 5.4, the FPV-GIS simulation results offer acceptable error margins, and the projected data find
good agreement for the characteristic performance of the PV solar module, "Canadian Solar CS5P-220M”,
under both STC and PTC test conditions. The FPV-GIS model achieves remarkably accurate results of
99.85% and 99.43% under STC and PTC test conditions, respectively, both sufficiently accurate to confirm
a corroborative verification of the model. The following section summarises the experimental findings and
conclusions.
5.3.4. Conclusions and critical findings of the experimental case study
This section summarises the experiment’s findings and conclusions on the accuracy of the FPV-GIS geo-
informatics analytical toolset from corroborative test results detailed in this synthesis-type experimental
validation assignment.
With the FPV energy model built upon standardised geographical-engineering calculations for PV-
project representations, the synthesised results for estimating the performance of the test module, as de-
fined in Table 5.2 are presented in Table 5.3. Benchmarked against the datasheet energy-performance
187
expectations in Table 5.2 for the selected PV module in Table 5.4, the accuracy of the simulated PV sys-
tem energy performance results predicted by the FPV performance model show a good correlation with the
datasheet reference performances under both STC and PTC test conditions (maximum error deviation of
0.15 % and 0.57 % respectively). The computational representations of the implemented FPV performance
model (the FPV-GIS model) thus achieve remarkable accuracies of 99.85% and 99.43% under STC and
PTC test conditions, respectively. While a typical accuracy of 98.5% (error tolerances of <1.5%) under STC
conditions can be deemed as an adequate deviation to confirm corroborative verification of the PV model,
the performance-prediction accuracy of 99.85% of the FPV-GIS toolkit under the STC conditions for the
experiment are presumed to be sufficient to confirm the corroborative verification of the model as this devi-
ation falls within the realistic error boundary. Although the FPV-GIS model slightly underestimates the test
performance of the PV module for both the STC and PTC test conditions, the results should lean somewhat
toward the conservative side to ensure that they would not generate over-performance expectations. While
the FPV-GIS model can predict the modelled PV energy yield within acceptable tolerance margins, the FPV-
GIS model can be considered validated, and the FPV-GIS computer model is ready for engagement in the
use-case experiments described in the rest of this chapter.
While both the experimental STC and PTC test results in Table 5.4 offer perfect accuracy scores in
terms of the corroborative verification of the model, namely 99.85% and 99.43% under STC and PTC test
conditions, respectively, the differences in the performance accuracies of the model between the results
under STC and PTC test conditions are attractive. While these differences in accuracy are minimal, we
should keep in mind that the set of reference STC test conditions is based fundamentally on internal mod-
ule operating conditions (Stein et al., 2011). This STC test stands in contrast to the more functional PTC
test conditions, which are based primarily on external module conditions related to the ambient operating
conditions of the test module (Whitaker and Newmiller, 1998). The PTC test conditions are more sensitive
to the PV-model outputs for specific PV modules and PV-array designs with varying thermal-transfer charac-
teristics (Whitaker and Newmiller, 1998). It may imply that the differences in the accuracy of the simulated
STC/PTC-based FPV model may be attributed to the characteristic thermal transfer coefficients engaged
by the online PVlib library (refer to the model-based designs for energy modules in Section 4.3.3). These
are technical aspects that are not of crucial importance to this geographical study.
Irrespective of the slight deviations between the STC and PTC performance results, the accuracy of the
simulated FPV model under both STC and PTC conditions is sufficient for this thesis to accept the FPV-GIS
performance model as one offering an accurate representation of the real-world power-output performances
of the PV system. Since the Standard Test Conditions (STC) is often the de facto standard for the accu-
racy of both actual and simulative evaluations and verifications of the module power rating universally, it is
significant that the maximum relative power-error percentage is 0.15% under STC conditions. This value
indicates a good prediction agreement between the simulated performance values and the expected refer-
enced performance values.
Strategically, the experiment applied an analytical framework and a systems-based FPV digital twin
model to answer a portion of the second research question. As such, this experiment generated scientific
results to validate the implemented FPV model and confirmed the power-production accuracy of the imple-
mented approach to theoretical PV systems modelling. While the accuracy of energy production serves
as a key metric in the accurate prediction of economic and environmental metrics (Prinsloo et al., 2021),
the validated model can help explore the combined effects of the technical, economic, and environmental
tridentate aspect as practical narratives of floating-solar development plans in the South African agricul-
tural context. In the next section, the tested FPV-GIS model is applied to a practical, real-world test-case
scenario which can answer the research questions around the integrated operational aspects of the imple-
mented FPV model more fully.
188
5.4. Experiment 2: FPV Project Performance Assessment
Having concluded the corroborative validation procedures for the model in Experiment 5.3, the assess-
ment task of this experiment focuses on generating scientific test results across the broader, theoretically-
integrated assessment of the energy, environmental and economic performances of a future planned FPV
system. These FPV systems model responses and results are essential to responding to the research
questions around integrated floating PV modelling. From a climate-smart agricultural perspective, agricul-
tural project engineers and energy project developers are often faced with the dilemma of assessing the
feasibility of a newly-planned floating photovoltaic project in a reservoir setting dedicated to agricultural ir-
rigation. Such a freshwater floating PV deployment requires a computer-aided assessment to investigate
the impacts of the floating solar platform on the water and to conduct a performance analysis of the floating
photovoltaic cover installed on the farm reservoir. Through these means, we can analyse the agricultural
industry’s climate-risk processes to capture the potential value of floating PV systems. The process involves
the electronic characterisation of floating photovoltaic performance and environmental impact profiling by
using contextual information to predict the site-specific power-generation capacity, environmental-offset pro-
file, and the economics of this FPV technology. In this context, Experiment 2 demonstrates the application
of the proposed digital twin model, in the FPV-GIS toolkit of Figure M.1, in a virtual design-stage project-
rehearsal scenario, whereby a scenario-based simulation of the FPV project is used to perform predictions
of the integrated energy, economic, and environmental project outcomes in a desktop-based digital-twin-
operating environment. While the experimental goal is described in Section 5.4.1, the practical method and
procedure are recorded in Section 5.4.2. Section 5.4.3, on the other hand, details the real-world-driven
experimental case-study results, which are used in Section 5.4.4 to provide reasoned conclusions from the
experimental results.
5.4.1. Goal of experiment
While the FPV-GIS toolset essentially offers a virtual prototyping platform for the design and assessment of
a real-world floating PV-project-installation variant, the focus of this experiment is on performance diagnosis.
As a digital twin model, the strategic goal is to respond to the first and second research questions around an
assessment of the enactment and performance quality of a floating PV project. The goal is thus to schedule
a real-world simulation project and to engage the FPV-GIS desktop toolset in a quantitative data-collection
methodology for a typical use-case project in a design stage narrative. Experimentation with the FPV-GIS
geo-informatics analytical toolset as a digital twin model or a virtual prototype model of an FPV system is
based on the research assumptions defined under Section 1.5.
In terms of the operational goals of this experiment, the aim is to use the FPV synthesis model as a
geospatial desktop tool in a virtual project-rehearsal scenario to synthesise FPV system operations for a
reference-cased wine farm and a project narrative for a winery-type installation. While the FPV-GIS toolset
operates as a digital twin proxy for the configured real-world floating PV systems variant, the desktop FPV
systems-synthesised simulation collects data by mimicking the real-time operations of the reconfigurable
real-world FPV project in a digital virtual-reality realm. In this way, the exercise sets the goal of demon-
strating a predictive estimation of the energy, economic and environmental (3E tridentate) performance
capabilities of the FPV-GIS toolset for a hypothetically planned floating PV system, and specifically for a
configured FPV project variant at an appropriate inland water body site at a selected agricultural farming lo-
cation in South Africa. The first objective is to test run a pre-defined floating PV system in a case-based test
scenario over a limited time horizon to demonstrate the discrete-time simulation capabilities of the model.
With this objective, the geospatial FPV-GIS model/digital twin engages real-world geodata, environmental
sensor data retrieved from a real-time meteorological repository for sourcing such data, to determine and
189
plot the predicted 3E metrics for the FPV case-study system and its site location as simulated time-graph
plots. The second objective of the experimental assignment is to run the floating PV system in the same
test scenario over the project’s extended lifetime horizon to determine the comprehensive operational per-
formance capabilities of the floating PV system over the entire lifecycle period - a long-term analysis. It
also demonstrates the integrative simulation capabilities of the model in assessing the cumulative lifecy-
cle performance of the floating PV system. The experiment will configure the desktop-computer toolset to
perform a multi-year analysis of the FPV system to evaluate its performance as an essential focus in the
reference-cased narrative of the above-mentioned local winery.
While the virtual prototyping capabilities of the FPV-GIS toolset offer an immersive workspace for digital
FPV project design in virtual dimensions, the goal is to configure the FPV-GIS toolset to operate as a
virtual FPV prototype tailored to the specifications of a proposed project narrative. The FPV model can thus
perform project enactments regarding the experimental objectives.
5.4.2. Experimental method and procedure
The methodology for a scenario-driven investigation in the experimental method and procedures defined
in this section represents a typical use-case study exercise to engage the FPV-GIS toolset in appraising a
proposed FPV pilot-project installation plan. The experimental FPV modelling evaluation thus constitutes a
scenario-driven case study of an analysis-by-synthesis assessment of a hypothetically-planned floatovoltaic
energy installation envisaged for the Boland grape/wine-producing and farming region of South Africa.
The experiment is guided by the quantitative research methodology outlined in Section 4.6 regarding the
material and method. In the following, the researcher (as an environmental-impact-assessment practitioner)
defines a typical use-case project scenario to test the geoinformatics-type analytical functionality of the con-
ceptualised FPV-GIS model in a real-world configuration setting under variable climatic and meteorological
conditions. The main objective is to engage the FPV model of Figure M.1 in an analysis-by-synthesis capac-
ity for empirical data collection in an authentic FPV pilot-project scenario. In this capacity, the experiment
reaffirms the philosophical idea of Figure 3.3, as the geospatial digital twin mimics the real-time behaviour
of a typical floating PV system to characterise its performance, impact effects and sustainability profile in
a selected project narrative. To help navigate the experimental method through its setup and execution
procedures, the processing diagram in Figure 5.3 outlines the prototypical design steps of a virtual FPV
system in terms of the envisaged site selected for the FPV installation and configures and simulates the
execution steps for this experiment towards the realisation of this analytical FPV-GIS model.
step 1 step 2 step 3 step 4 step 5 step 6 step 7 step 8
Choose FPV
installation
site water
body and
geographical
parameters
for the
case study
Define FPV
variant as
technical
specification
environmental
conditions
and financial
parameters
Configure
FPV model
with country
and system
specific
conditioning
parameters
Retrieve
standardised
solar and
weather
forecast data
for FPV site
Execute
runtime
simulations
for FPV and
GPV system
performance
metrics
generated by
FPV model
Plot time
varying FPV
and GPV
performance
and real time
impact effects
data output
generated by
FPV model
Create project
report graphs
from lifetime
accrued
FPV vs GPV
performance
and project
impact data
Derive
experimental
conclusion
Figure 5.3: Experimental method and procedure for an assessment of the performance of an integrated floating PV
project, to be executed according to the defined experimental methodology (source: author).
190
More specifically, Figure 5.3 shows that the methodology starts with defining the experimental case
study and outlining the model definitions in Steps 1 and 2. Figure 5.3 continues to execute the simulation
via Steps 3 to 7, and concludes with the experimental summary and findings in Step 8. To use the FPV-
GIS synthesis model as a quantitative research instrument in a scenario-based case study, the definition
and parametric tuning of the identifiers of the FPV systems model allow the user to pre-configure the ge-
ographical region, project assumptions and weather conditions required for the authentic evaluation of this
use-case model. As such, Step 1 of the experimental procedure in Figure 5.3 calls on the research to define
a floating PV installation site and a suitable water body with its geographical parameters.
In terms of the demarcated study area for the experimental desktop application, as defined in Sec-
tion 4.6, the experimental FPV installation site is chosen within the service area of the EGIS interactive
map service for SA REEA EIA Application Database, namely the blue georeferenced area, as shown in
Figure 5.4. As such, the experimental floating PV pilot project in the Cape Winelands is planned for the
wine farm labelled RWE in the Paarl area, a district within the Boland grape/wine-producing and agricul-
tural region of South Africa. As marked on the EGIS interface map of Figure 5.4, the selected solar-voltaic
farming site is located in the so-called Agter-Paarl region.
Figure 5.4: Demarcated study area for the South African Renewable Energy EIA Application Database on the EGIS
platform (after: DFFE (2021b)), showing the selected RWE wine farm site for the FPV project experiment.
The proposed FPV project installation site (defined by the GPS coordinates presented in Figure 5.4) falls
within South Africa’s Central Renewable Energy Corridor, a geographical region that forms one of South
Africa’s renewable development corridors for carbon-tax beneficiation (refer to Figure 2.4). At the same
time, the proposed site is located within an agricultural region that can be considered an ideal use-case
area for South Africa (refer to Research Assumption 3 in Section 1.5).
With the location of the geographical FPV installation site specified, the next step revolves around adding
the dimensions to the floating PV project design and configuring the technical, environmental and economic
aspects of the proposed FPV project variant. In this context, Step 2 of Figure 5.3 defines and configures the
FPV systems variant in terms of its geometric and technical specifications, the site-specific environmental
sensor data, the environmental conditions, and the financial conditions for the proposed floating PV pilot
191
project. For the selected experimental site, the experiment envisages a specific water-integrated floating
PV system, such as the variant of the system illustrated in the installation scene in Figure 2.13a.
Regarding the technical conditioning of the floating PV system, the dimensioning of the FPV systems
variant can be specified for the location of the selected FPV project, namely RWE. The open water basin on
this farm provides a sufficiently large geometric water surface area to comfortably install an FPV system with
a geometric PV design/area of approximately 1000m2. In this context, the energy and water-evaporation
performance of the FPV system within the hydrological and environmental microhabitat is a strong function
of this type of photovoltaic floating substructure specifically designed to harvest solar energy and the per-
centage of water surface covered. This realisation implies that parameters such as water-surface clearance
and the moisture plus air-ventilation properties of the floater-type design are specified in terms of the cool-
ing model coefficients (as outlined in Section 4.3.4.3) of the floating PV floater. The floater parameters for a
hypothetical floater-type design are defined in Appendix O.
In terms of the specifications of the electrical floating PV system for the proposed project variant, this
experiment continues to deploy the Canadian Solar PV module chosen for the FPV-model-validation pro-
cess in Experiment 5.3 for its design configuration variant. As such, the technical conditioning configuration
is based on the details for the Sandia Module (Canadian Solar CS5P-220M module), with a nominal name-
plate rating of 220W, and fully characterised by the technical datasheet attributes, as defined in Table 5.2.
The Canadian Solar panels each cover a surface area of 1.7m2. For such an area, the project can in-
stall approximately 333 Canadian Solar PV panels in a geometric FPV design that covers the available
1000m2water area. As such, the configured floater-design type, panel-packing density, panel-surface tilt
angle, and equipment allow sufficient spacing for the panel shading and the maintenance walkways be-
tween the panels. This layout means that the required panel/power-density configuration is set to a value of
33 panels/100m2. (Keep in mind that the user can reconfigure this parameter for any alternative panel size
or layout design.)
Step 2 specifically needs to take cognisance of the model framework, as presented in Figure M.1, and
the model parameters for each of the energy and environmental model objects (3E objects), as shown
in Figure M.1 as (re-)configurable floatovoltaic ecosystem settings (refer to user-configuration parameters
in Appendix O). Thus, having defined most of the settings to account for the technical conditions for the
proposed project variant, Step 2 needs to define further the environmental and financial condition set-
tings for the envisaged floating PV project variant for this FPV assessment experiment. The environmental
conditions focus on terrain-based conditions for the environment around the modelled FPV systems array.
In terms of the environmental condition setting, it is important to consider the concurrent meteorological
year data (TMY-based date-tagged solar insolation and weather data for the RWE site) as a key aspect
in this research study as these data actually define the real-time environmental sensor data that describe
the operating conditions under which PV energy systems can successfully function. For the proposed ter-
rain and geographical location, the study selected the cost-efficient HelioClim (2022) weather station data
repository to import the solar/weather data files associated with the selected farming area in the Paarl
region. A selected sample of the meteorological TMY geosensor input dataset is shown as an example
in Table P.1 (Appendix P). Other environmental model-conditioning settings include the country-specific
emissions-equivalence factors such as the coal-fired grid-generation substance emissions, as defined by
the Eskom model for grid-substituting environmental-impact calculations, as presented in Table 4.1. In terms
of the financial condition settings for the FPV project, the model configuration is set to reflect the installation
costs of South-African-based PV equipment according to the new energy financing cost parameters and
FPV system benchmark costs (Bloomberg Energy, 2022; Ramasamy and Margolis, 2021), as well as the
financial measures and agricultural production costs, plus production income, as presented in Section 4.3.
From an agro-economic perspective, the wine farm selected for this research engages in viticulture and
oenological activities to produce grapes and bottled wine. In this case, the grapes are sold at prevailing
192
market prices while the wines are sold at cellar prices tagged at the median-price point among the medium-
tier wine-price range. In Step 3, the FPV systems specification settings defined above are programmed into
the FPV model’s user-configuration file to allow the simulation-driven proposal-generation process. While
the user-configuration file is programmed to accommodate the above specifications and conditions, the re-
maining configurations for the operating conditions of the FPV-GIS simulation toolset are defined in terms
of the default values, as detailed in Section 4.3.2.
In methodological terms, Experimental Step 4 prepares for the simulation execution by retrieving the
standardised solar and weather forecast data for the selected site area of the FPV project from the repository
selected under the configuration of environmental conditions, as described above. This setup enables
Step 5 of Figure 5.3 to execute run-time simulations to determine the FPV systems performance metrics
generated by the FPV model in a 20-year window period for the life-cycle of the project. The geospatial
digital twin model is driven by meteorological geosensor data for the RWE reference site. TMY-format
weather station data are real-time geosensor meteo-data measured and modelled for the Paarl arterial
(see geo-data sample in Table P.1). This data serves as input to the data processing activities of the
geospatial digital twin model, thus driving the simulated output predictions through the period 2023 to 2043
(20-year project lifetime). During this execution phase of the model, the user-configured digital floating PV
analogue conducts an analysis-by-synthesis procedure for the lifetime of the project as a run-time simulation
in the virtual realm to derive equivalent real-world time-series data-flow outputs for the proposed floating
PV project site, given the project variant and floating PV business ecosystem.
The FPV-GIS toolset is Python-programmed to explore the techno-economic and techno-environmental
externalities of a parallel ground-mounted GPV system variant of the exact specifications in individual or PV
cogeneration formats (refer to Section 4.5.1). The operational methodology and architecture of the FPV-GIS
toolset would simultaneously animate the operations of a co-located land-based GPV system of the exact
dimensions using the conventional PVlib toolset (considered to be planned for installation adjacent to the
scheduled FPV system). Furthermore, in the FPV/GPV project comparisons, it is essential to note that the
solar panels in an FPV configuration can be more densely packed and aligned at different panel/orientation
angles when compared to the conditions required for the operation of a GPV system (Gadzanku et al.,
2021b). This aspect means that the energy, environmental and economic performance margins for the
FPV system may, in practical terms, be much higher than those of a GPV system for the same PV plot area.
However, for this experiment and experimental comparisons, the same PV design configuration and systems
specifications are programmed into the configuration file for the FPV variant and its GPV counterpart.
While FPV project simulation performs the dynamic modelling and simulation analysis at an hourly tem-
poral resolution, it enables Step 6 of the methodology, as presented in Figure 5.3, to plot the real-time data
outputs for the FPV-GIS model generated FPV and GPV performance and impact data as time-varying
arrays. Step 7 is subsequently able, through mathematical integration, to accrue the time-varying perfor-
mance metrics annually and over the project’s entire 20-year lifetime. The setup enables the experiment to
deliver data relating to relevant FPV performance and impact effects required to generate life-cycle-reporting
graphs for the project. These performances and graphs would then represent the anticipated lifetime data
accrued in terms of FPV vs GPV performance and the impact data for the proposed project variant at the
selected farming location.
The floating PV installation site and floating PV systems configuration are defined and tailored in terms
of the model’s user configuration file (FPVconfig.csv) (refer to configuration setup in Appendix O), and
the FPV-GIS toolset offers the systems designer or EIA practitioner the ability to perform the simulation-
based testing of floating PV project variants in the configuration of a real-world geographical, environmental,
economic and technical setting (as defined above). By engaging this configuration in the digital FPV-GIS
model-synthesising exercise, the following section presents the experimental results predicted by the virtual
FPV prototype in terms of the design specifications of the floatovoltaic model, as defined in the associated
193
configuration file (FPVconfig.csv). Furthermore, screenshots of the FPV-GIS Python model interface in
Figure Q.1 (Appendix Q) show GUI snapshots where the model reads the configuration data and weather
station data as input variables on the display screens of the Python model.
5.4.3. Experimental case-study results
In an experimental digital computer synthesis trail run, this section demonstrates the data-processing and
data-acquisition capabilities of the FPV-GIS toolkit depicted in Figure M.1. Running as a virtual prototype of
a real-world FPV system, the scheduled desktop experiment digitally synthesises the operations of the pre-
configured floating solar project variant planned for the irrigation reservoir in a vineyard block on the wine
farm RWE. The results provide a synopsis of the quantitative project performances and outcomes predicted
from the analytical data captured during the project’s design stage layout selected for the topographical
map layout of Figure 5.5a. With the site parameters defined, the associated satellite image of Figure 5.5b
presents a superimposed hybrid image of the terrain-based two-dimensional geometric layout (polygon
shaped system layout) of the experimental FPV system (and its GPV counterpart) as definitional inputs to
the geo-spatial FPV digital-twin simulation.
(a) (b)
FPV
GPV
Figure 5.5: Digital terrain model for the RWE wine farm water reservoir site on (a) a topographical map and (b) a
satellite image, the virtual project terrain showing the superimposed 2D geometric layouts for neighbouring FPV and
GPV system designs (source: author; after: Google Maps 2022).
In the context of the experimental floating PV pilot project for the geo-referenced site, as depicted in the
hybrid image of Figure 5.5, the first part of the digitally synthesised experimental results demonstrates the
ability of the FPV-GIS model to analyse the quantitative performance of a pre-defined floating PV systems
variant in the proposed case-based test scenario over a limited time horizon of one week only. In terms of
the first objective of the experiment, this part of the experimental results determines and plots the predicted
3E metrics for the case-study site as simulated time-graph plots. In terms of the second objective of the
experiment, the second part of the experimental results demonstrates the ability of the FPV-GIS model to
quantitatively predict the anticipated performance of a pre-defined floating PV systems variant in the same
case-based test scenario over a time horizon period configured at 20 years - a long-term life cycle. Note
that real-world meteorological geosensor data serves as input driver to the FPV-GIS geospatial digital twin
model, which weather station data serves as input to the FPV-GIS model as a data processing unit to
perform the 3E sustainability performance predictions for the period 2023 till 2043 (20 year project lifetime)
as model outputs. The geospatial weather station input data is downloaded as structured online TMY input
datasets for the GPS location of the selected RWE experimental reference site (see sample in Table P.1).
194
Screenshots of the FPV-GIS Python model interface itself, shown in Figure Q.2 (Appendix Q), provide
GUI snapshots of some of the output variables and display screens for the Python model. Then, to elaborate
on the empirical results, Section 5.4.3.1 presents the real-world performance results for the time-varying
analytical performance data simulated and predicted by the FPV-GIS toolkit in real-time over an arbitrary
one-week period. Section 5.4.3.2 then processes the real-time data to accrue the experimental results
for the FPV-GIS toolkit’s analysis of the project’s performance over its entire life cycle, thus estimating the
sustainability performance metrics over the project’s entire lifetime period of 20 years.
5.4.3.1. Data-acquisition results for real-time performance analysis
This subsection details the experimental results for the defined FPV-project assessment task to demon-
strate the analytical data-capture and data-acquisition capabilities of the FPV toolset during the project’s
design stage. It reports on the performance results for the real-time analytical performance data for the
digital project rehearsal, with an FPV systems variant digitally simulated through the FPV-GIS analysis-
by-synthesis model. Note that in this exercise, the FPV-GIS toolset operates as a geospatial digital twin
model, driven by real-world meteorological geosensor data shown in the extracted sample of Table P.1 (Ap-
pendix P). It thus mimics the real-world operations of the pre-configured floating PV system in virtual reality
as if it is operating at the RWE installation site. Further note that the FPV-GIS toolset uniquely applies the
3E tridentate framework of Figure 3.14 as a theoretical framework in a system dynamics configuration to
simultaneously predict the FPV system’s output energy/power, environmental, and economic performances
(the 3E sustainability metrics) iteratively on a sample-by-sample basis (throughout the 20 year lifetime of
the project). It enables the pertinent visualisation of the project results for any chosen time window.
Following the systems dynamics simulation procedure (Step 5 of Figure 5.3), the experiment plots the
time-varying FPV performance outputs and real-time impact effects (digital data outputs) generated by the
FPV model over a particular time range on a real-time graphic display. Through the execution of the run-time
simulation at an hourly simulation-processing frequency, the live stream results are depicted in the temporal
propagation patterns reflected by the multigraph timelines shown in Figure 5.6. The digitally synthesised live
stream results project the digital rehearsal outcomes for the proposed floating PV system over the selected
project dates for the zoomed window period for the project. Note that the temporal propagation of the time-
varying graphs in the visual sustainability metric representations of Figure 5.6 depicts only some of the key
selected output parameters of the FPV model. While the FPV model in Figure M.1 generates around 60
different 3E metric data streams on a sample-by-sample basis, the number of plotted graphs is limited to
six key parameter data groups to keep the presentation concise.
The multi-scale live-stream graphs of Figure 5.6 depict The perpetuation of the FPV project behaviour
throughout the project timeline. As a time-lapse of the floating PV installation operations, this multigraph
offers a 3E co-simulation overview of the FPV project parameters and expected FPV-project deliverables.
The sample frame performances are determined iteratively for each PV model simulated cycle throughout a
time-zoomed window. As the 3E simulation propagates through time, the projected performance and impact
results are laid out chronologically throughout the demonstration window for a specific weekly snapshot
period extending over the first week in May 2023. The synthesised project animation exercise operates the
FPV model in a digital real-time synthesis mode, configured to produce time-varying project-performance
outputs at the pre-configured hourly resolution level. The results thus demonstrate the digital-desktop FPV-
GIS model’s multi-scale discrete-time analytical capabilities.
In the FPV performance-evaluation task depicted by Figure 5.6, the FPV model is driven by site-specific
meteorological (climatic) and environmental sensor data. The energy model’s input data arrays are graphi-
cally shown in Figure 5.6a. It depicts a snapshot window of the geo-sensor measured solar irradiation data
and patterns received by the solar voltaic panels at the site location of the FPV reservoir.
195
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
0
0.5
1
(a)
Irade(kWh)
POA direct-on-panel ITR effective-on-panel
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
20
40
(b)
Te(◦C)
Tdry ambient FPV ambient fTpanel operate temp
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
0
50
100
(c)
Pe(kWh)
FPV-system power-out Pm
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
72
73
74
75
76
(d)
ΣCO2(Ton)
FPV grid substitution emissions avoided xCO2
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
100
150
200
250
(e)
ΣH2O (kL)
FPV-avoided evapo eH2O FPV grid avoid xH2O
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
0
50
100
150
(f)
ΣRe(ZAR)
FPV electr cost save (Rgs) FPV carbon tax credits (Rcc)
Figure 5.6: Time-varying FPV model-input parameter streams and future predicted (model-simulated) real-time per-
formance and impact-effect metric outputs, as predicted for the first week of May 2023 in respect of the installation of a
planned agricultural floating PV project variant and its systemic case-study scenario at the RWE farming site (source:
author).
196
These geo-referenced data arrays include the direct-normal, direct-horizontal, and global-horizontal ir-
radiation (DNIt, DHItand GHItrespectively), which determine the effective on-panel irradiation samples
(ITRt). Next, the time-varying dataset arrays in Figure 5.6b portray the dry-bulb (Tdryt) and wet-bulb (Twett)
input ambient temperature samples retrieved from the meteorological TMY data for the selected FPV instal-
lation site.
From the solar irradiation and TMY climate data, the FPV model determines the time-varying operating
temperatures of the floatovoltaic panels (fTpanelt) by processing the real-time ambient temperature (Tdryt)
and FPV ambient parameter values at the simulated time instance of each system dynamics model. The
FPV-GIS model predicts the operating temperatures for the floatovoltaic array (fTpanelt) in Figure 5.6b.
The sample values for the internal operating temperatures (fTpanelt) for the FPV panels in Figure 5.6b are
determined from the data in Figure 5.6a and b. Next, the power-yield curve samples (fPmt) are simulated
through Function F.5, as shown in the time-series graph of Figure 5.6c. This power curve portrays the
predicted values for the FPV-GIS model for the energy yield for the FPV system at every hourly instance,
thus reflecting the dispatchable instantaneous power (fPmt) potentially generated by the FPV system during
every hour.
During the same simulation step, the environmental model object determines the concurrent environ-
mental offset profile for the FPV system. It is specified in terms of the environmental performance param-
eters, according to the description in Section 4.3.4 at the optimal dispatch conditions of economic power
(assuming that all energy consumed, Research Assumption 2). The avoided CO2pollutant emissions data
(CO2t) depicted in Figure 5.6d is mathematically time-integrated to reflect the compounded CO2impact
accrued over time. These specific results reflect the accrued (SA Eskom) grid-substitution CO2-pollutant
emissions avoided at each time instance over the selected window period. Given the display space con-
straints, the real-time plot in Figure 5.6d depicts only one of six elements of the environmental-footprint
benefits (avoided pollutants: xSOx, xNOx, xPEa, xCoal, xAsh) gained by the FPV project as a result of
its displacement of coal-based grid power with renewable solar energy. The FPV-GIS model predicts the
carbon emission (CO2) together with the five remaining environmental offset metrics (avoided pollutants:
xSOx, xNOx, xPEa, xCoal, xAsh). These metrics form part of the hydrocarbon footprint of the PV system
and are plotted later, in Figure 5.9, as lifecycle impacts.
The environmental model of the FPV-GIS toolset further predicts the accrued reserve margin for water
resources preserved by the FPV system (eH2O + xH2O) at each discrete time sample, as depicted in
Figure 5.6e. The forecasted water-preservation metric reflects the volume of water evaporation avoided
by FPV (eH2O), owing to the local FPV panel shading on the irrigation reservoir, and the water savings
(xH2O), owing to the grid-power cooling avoided. In addition, the H2O environmental-offset metric for FPV
is pro-rata mathematically integrated over the snapshot window period in Figure 5.6e, thus portraying the
accrued FPV water-saving benefit as a temporally-accrued resource.
During the next step of the same simulation cycle, the economic model object of the FPV-GIS toolset
predicts the concurrent financial performance profile for the future planned FPV system in terms of the eco-
nomic performance parameters, as detailed in Section 4.3.5. Thus, in tandem with the instantaneous power-
yield and environmental-offset performance-simulation samples for the floating PV system (Figure 5.6a-e),
the economic model of the FPV-GIS toolset predicts the effective carbon credits value (Rc as an effective
carbon tax benefit), and the avoided grid-power billing purchases (Rs) as savings in terms of the electricity
bill. Figure 5.6f thus predicts the cumulative accrual-of-finances metric, computed from the pro-rata aggre-
gation of the income revenue streams potentially generated by the FPV systems variant over the selected
display window.
While the capabilities of the FPV-GIS toolset further include programming to explore the stepwise
techno-economic-environmental externalities of a parallel ground-mounted GPV systems variant of the ex-
act same specifications, Step 5 of the analysis-by-synthesis methodology of Figure 5.3 allows Step 6 of
197
the experimental procedure to plot the digital time-varying FPV and GPV performances and the real-time
impact-effects data in the form of real-time graphs. As such, the synthesised live stream results in the
visual sustainability metric representations of Figure 5.7 compare the geo-referenced floating PV-systems-
performance data, shown earlier in Figure 5.6, with that of a co-located land-based solar GPV model,
hypothetically hybridised and considered to operate in tandem with the referenced FPV system. In deter-
mining the GPV model performances in Figure 5.7, the live stream PVlib simulation engine uses the same
geo-referenced environmental sensor data for the scenario in the context of the FPV systems installation
(shown in Figure 5.6a). This data is engaged to determine the effective live stream on-panel irradiation
(ITRt) and the operating temperature for the GPV array (gTpanelt), as shown in Figure 5.7a.
The FPV-GIS toolset in the GPV model engages conventional PVlib performance-prediction procedures
to determine the predicted power-yield curve samples for the GPV-simulated time series, as shown in
Figure 5.7b. Through the pro-rata mathematical integration of the live stream energy production samples,
Figure 5.7c compares the cumulative power generated by each of the FPV and GPV systems over the
duration of time covered by the dated weekly display window.
The economic and environmental-performance data comparisons between FPV and GPV systems are
portrayed in Figure 5.7d to f, while the pro-rata accrual of these metrics shows the output performances
as temporally-accrued resources. The comparative GPV performances can also be determined live stream
on a sample-by-sample 3E-processing basis by the FPV-GIS model during each simulation cycle. The pre-
dicted environmental impact offset for the CO2pollutant in Figure 5.6d precisely reflects the accrued (SA
Eskom) grid-substitution (xCO2) pollutant emissions avoided by the GPV system over the window period
of the project. Regarding water preservation benefits, the ground-mounted GPV systems counterpart de-
livers no locally-avoided water-evaporation component (eH2O). Therefore, the GPV environmental-impact
graph for H2O preservation reflects only the accrued volume of national water savings (xH2O) ascribed to
the avoided coal-fired/water-cooled grid substituted with solar electricity. Compared to the income predic-
tions for the FPV project, the financial projections of the economic model of the FPV toolset for the GPV
counterpart in Figure 5.6e show lower values for both the effective carbon credits value (Rc as an effec-
tive carbon-tax benefit) and the avoided grid-power billing purchases (Rs). Logically, these benefits can be
ascribed to the lower power output (gPm) of the GPV system and the reduced number of carbon credits
earned from the lesser volume of avoided carbon emissions (xCO2), respectively.
This section illustrates the data-acquisition capabilities of the FPV-GIS toolset in characterising the real-
time dynamics of a physical floatovoltaic-power-plant ecosystem. Furthermore, it determines the predicted
performance profiles for a specific FPV systems variant that are determined by digitally synthesising the
project with site-specific real-time environmental sensor data. The next part of the experimental results
reports Section 5.4.3.2 on the capture of analytical data pertaining to the life cycle and the data-acquisition
capabilities of the FPV toolset.
5.4.3.2. Data-acquisition results for life-cycle performance analysis
This subsection reports on the experimental results for the estimated lifetime performance empirically pre-
dicted by the FPV-GIS toolkit, as defined in Step 7 of the testing methodology and presented in Figure 5.3.
This characterisation exercise about the profiling of the PV system presents empirical performance results
regarding an analytical evaluation of the lifetime performance of the proposed FPV systems variant per-
formed during the project’s design stage.
Under this emulation-based paradigm, a behavioural re-enactment exercise analyses the PV project’s
long-term performance. As such, the model execution of the FPV project simulation is allowed to run its
entire course (lifetime) to generate live stream hourly output samples over the whole 20-year configured
period. (This is similar to Figure 5.6, but is now for the entire lifetime).
198
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
20
40
60
(a)
Te(◦C)
FPV operate temp (fTpanel) GPV operate temp (gTpanel)
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
0
50
100
(b)
Pe(kWh)
FPV-system power-out PmtGPV-system power-out Pmt
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
68
70
72
74
76
(c)
ΣPe(MWh)
FPV-system power-out GPV-system power-out
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
68
70
72
74
76
(d)
ΣCO2(Ton)
FPV grid avoided xCO2GPV grid avoided xCO2
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
100
200
300
400
(e)
ΣH2O (kL)
FPV-evap + grid avoid (eH2O+xH2O) GPV grid avoid (xH2O)
2023-05-01 2023-05-02 2023-05-03 2023-05-04 2023-05-05 2023-05-06 2023-05-07
140
145
150
155
160
(f)
ΣRe(ZAR)
FPV income (Rgs+Rcc) GPV income (Rgs+Rcc)
Figure 5.7: Time-series comparisons between the future predicted (simulated) performance outputs and impact
streams of the PV model, as predicted for the first week of May 2023 in respect of the planned agricultural floating FPV
systems scenario and its adjacent land-based GPV systems counterpart at the RWE farming site (source: author).
199
The lifetime performance analysis for FPV and GPV systems accrues the configured real-time data sam-
ple values for each performance and impact metric series through ordinary mathematical integration. There-
fore, it suffices to say that life-cycle performance analysis engages the same data-simulation-processing
procedures described in Section 5.4.3.1 over the project’s entire 20-year lifetime. As such, the case-study
experiment can visually report on the sustainability performance metrics for the lifetime energy, environmen-
tal and economic (3E) performances predicted by the FPV-GIS toolset for the land-based and water-based
floating PV project variants driven by the same real-world geo-data (actual real-world solar irradiation, me-
teorological weather and climatic sensor data prevailing at the selected farming site).
In this context, reporting on the comparative lifetime performance starts with the digitally synthesised
energy yield performances projected for the FPV and GPV systems. As such, Figure 5.8 reports on the
annual energy-yield profile plots predicted by the FPV-GIS toolset for the selected RWE Paarl farming site.
The annual energy-yield quantities accrue to the cumulative lifetime energy-production capacity (GWh) for a
future planned floatovoltaic system (5.22 GWh). The study can compare these outputs with the cumulative
lifetime energy-production capacity for an adjacent ground-mounted GPV system (4.60 GWh) for the same
project site. The increased energy production over time highlights the leverage of the FPV design over the
GPV design in terms of power or Energy generation aspects of sustainability.
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2049
2040
2041
2042
180
200
220
240
260
280
Year
Solar power Pe(MWh)
Annual FPV energy yield Annual GPV energy yield
Figure 5.8: Annual energy-yield profile plots predicted by the FPV-GIS toolset for the selected farming site and
accruing to the cumulative lifetime energy-production capacity (MWh) for the FPV system (5.22 GWh), as opposed to
an adjacent GPV system (4.60 GWh) (source: author).
Over the project’s lifetime, the digital FPV-GIS simulation synthesis predicts the generation capacity of
the floating PV system at 73.3 kWh, while Figure 5.8 predicts accrued energy output of the floatovoltaic
system in a total of Pt=9.12 MWh. On the other hand, a co-located ground-mounted GPV system (of the
same size and specification) is estimated to deliver a total of Pt=9.12 MWh. This result means that the aver-
age lifetime efficiency improvement of the FPV in terms of power generation gained over a ground-mounted
PV installation is estimated to be fgPmη=11.8%. This efficiency improvement or output gain is attributed
to the hydroclimatic cooling effects of the FPV system’s microhabitat. While the cooling effect of the water
helps to cool down the panels to make them more efficient, the cooler operating environment also helps
to counteract PV module degradation over time. Furthermore, in terms of the cross-object performance
ratios of a floating PV system, as defined in Section 4.3.6, the capacity-based water-surface transformation
factor, as defined by Cagle et al. (2020), equates to WST=9.12 m2/MWh/yr, while the efficiency factor for
water-surface use amounts to WSUE=73.2 W/m2.
200
In terms of the distinctive lifetime, environmental-offset benefits of the floating PV project for the param-
eters defined in Section 4.3.4, Figure 5.9 displays the empirically projected impact-effect values amassed
by the PV projects in terms of the avoided environmental-impact factors, H2O, CO2, SO2, NOx, and aPE.
The legend displays the bars for the hydrocarbon footprint impact values articulated for both the planned
floating FPV and the land-based GPV systems. In terms of the carbon-sequestration footprint of the pro-
posed FPV system, the model estimates a total of 5 million metric tons of CO2emissions avoided through
utility grid-power substitution over the lifetime of the project (under the assumption that energy supply is in
balance with energy demand). The most salient aspect of avoided CO2emissions represents the potential
decarbonisation contribution of the planned FPV energy plant, with the model-predicted lifetime reduction
of 2 140 tons of effective CO2aerosol emissions into the earth’s atmosphere.
Coal
Water
Carbon
Ash
FLp
SOx
NOx
aPE
101
103
105
107
Impact effect
Quantity (kg/m2/litre)
FPV environmental performance GPV environmental performance
Figure 5.9: Environmental signals in a decision-support dashboard, detailing the environmental-offset profile of the
operational lifecycle for the floatovoltaic installation (base-10 log scale) (source: author).
Note that the GPV system does not preserve any farmland, which explains why the estimate for the
preserved farmland factor (FLp factor) for the GPV counterpart in Figure 5.9 outputs at zero value (according
to the convention of this thesis). This factor, together with the remaining environmental factors wherein FPV
has an advantage, highlights the FPV’s design leverage over the GPV design in terms of the Environmental
offset aspects of sustainability. These zero values in Figure 5.9 evidence but some of the holes and gaps
left by conventional PV performance models are features that signify the so-called "technology unknowns”
of PV technology when it floats on water (refer to Section 1.2.1.1). The results highlight how conventional
GPV performance modelling outcomes are unreliable when used in floating PV assessments. It also shows
how the 3E system dynamics framework fills gaps and holes in floating PV performance assessments in
the energy, environmental and economic dimensions as evidenced throughout Figures 5.8 to 5.10.
Continuing with Figure 5.9, another distinctive lifetime environmental contribution by the FPV system is
the estimated total of 2 961 kL of freshwater preserved on account of avoided on-site evaporation (eH2O),
plus the avoided water cooling (xH2O) at grid-generating power stations. In that coal-free power plants can
conserve a significant amount of water, installing a floating PV system in an agricultural grid-substitution
application would mean a greener solar energy installation that would generate clean energy as a substitute
to grid power. Water conservation through the installation of a floating PV system at the installation site
shows potential annual water savings of, on average, approximately 400 Kilolitres of water (per year per
site). As illustrated in Figure 5.9, this is one of the distinctive resource-preservation implications of floating
201
PV. Furthermore, as determined by the FPV-GIS toolset, the environmental conservation scoring values
in Figure 5.9 represent the cumulative direct and indirect water-conservation components as the impact
effects of the floating PV plant. Indirect water saving represents the predicted volume of national water
resources conserved (gH2O) on account of the substitution of coal-based power from an Eskom water-
cooled grid-serving power plant with locally-generated energy from a floating PV power plant. The direct
water-resource preservation value (fH2O) represents the predicted volume of water preserved as a result
of a reduction in on-site water evaporation, mainly due to the shadowing effect of the floating PV system’s
structure on a local water body such as an irrigation reservoir or pond.
In addition to the operational characteristics of the empirical environmental-offset and impact profile in
Figure 5.9, preserved farmland is yet another critical distinctive environmental sustainability offset offered
by floating PV technology. While the sparing of farmland needed for agriculture and conservation is a crucial
sustainability factor of FPV technology, the model must adequately quantify the distinctive territorial benefit.
With the pre-configured floating PV system covering an area of 1000 m2, the effective land area of farmland
preserved by not installing a GPV system of the same output is calculated to be FLp=1283 m2. According
to the land-sparing modelling definition in Section 4.3.4.5, the land-gearing saving factor for the floating PV
system is calculated as 28.36% (land gearing = a 1.2836 increase). This factor implies that installing a
floating PV power plant preserves 1283 m2of productive farmland for agricultural production.
In terms of the economic effects of the lifetime performance of a solar photovoltaic system, the dash-
board display presented in Figure 5.10 reflects the income amassed from the PV project installations in
terms of the South African rand currency. It provides a breakdown of the income amassed and revenue
collected as financial metrics for the floatovoltaic installation. The bar graph reflects the most likely income
scenario prediction for the main cost-driving economic metric dimensions.
Operating
Electricity
Commodity
Land-value
CO2-credits
Agronomic
Water
101
103
105
107
Economic indicator
Value (ZAR)
FPV economic performance GPV economic performance
Figure 5.10: Econometric signals in a decision-support dashboard detailing the economic-impact profile of the life
cycle for the floating solar PV system (base-10 log scale) (source: author).
In the results of Figure 5.10, the model notably predicts the operating life-cycle LCOE of the power
plant at ZAR 0.25/kWh (in ZA Rand currency). At the same time, the LACE metric is determined at a
marginal benefit gain of ZAR -26.84/kWh. The LACE metric signals that the electricity generated by the
FPV installation would be at a cost saving of 48 cents/kWh below the grid tariff. The FPV project NPV is
estimated at a current NPV cost value of ZAR +1.298 million and an NPV income value of ZAR +2.044
202
million, NPV=0.746 million ZAR (a positive value indicative of investment worthy of consideration). The
model predicts that the project will reach an investment break-even point in Project Year 6.
The FPV system results in Figure 5.10 further shows an increase in the agronomic farmland value
gained for the productive agricultural farmland preserved (FLp=1283m2), plus a reduction in the land-cost
capital burden of ZAR 109,055.00, approximately 10% of the FPV asset cost. In addition, the income
generated through solar farming on this unused farm space can potentially result in a distinctive economic
surface transformation (EST) of approximately EST=614.26 ZAR/m2/year (in the first year of the project).
Note that the visual metric representations in Figure 5.10 portray how the GPV system counterpart does
not preserve any one-site water or farmland, while the farmland it occupies does not contribute to agricultural
production or processed commodity income. It explains why the estimates for the "Water” income (value
of avoided evaporation), the "Agronomic” income (grapes), the agricultural "Commodity” income (wine),
and the monetary value for the "Land-value” for the GPV counterpart in Figure 5.10 outputs at zero value
(according to the convention of this thesis). These factors, together with the remaining economic impact
factors wherein FPV has an advantage, highlight the FPV’s design leverage over the GPV design in terms
of the Economic impact aspects of sustainability.
Given the preserved farmland (FLp=1283m2), the model estimates the wine farm’s rescued agricultural
production at 1.98 tons of grapes per year (sufficient produce for approximately 2000 bottles of wine per
year). In terms of retail wine-sales revenue to be potentially gained from 2000 bottles of wine rescued
(model agro modality 3), the FPV land saving translates into an income of ±0.6 million ZAR per year. By
NPV cost-comparison standards, this aspect could accrue enough gross income over the project’s lifetime
to pay for more than ten new FPV power plants.
An essential and unique aspect of the thesis research is the ability of the FPV-GIS model to support
the EIA assessment process. Such results are significant, given the current challenges concerning EIA
assessments for floating PV project proposals (refer to Sections 2.3.2&2.4). In this context, the FPV model’s
analytical data archiving capability supports an EIA assessment for the project site presented in Figure 5.5.
To this end, the FPV-GIS model’s quantitative energy, environmental and economic performance outputs
and environmental dimensions can be directly used to complete the EIA reporting scorecard, as shown
in the redacted version of the sample EIA project report given in Figure U.2. Furthermore, EIA project
authorisations specifically require credible scientific evidence of future lifetime performance data and the
impact effects of a planned energy project before installing a PV system. To this end, the application of the
empirical outputs of the FPV model in the sample EIA report in Figure U.2 provides duly required scientific
evidence of the EIA-related impact effects for the proposed FPV project.
This part of the experimental results demonstrates the analytical prediction capabilities of the life cycle
in the implemented FPV model in terms of the second objective of the experiment, as defined in Step 7 of
the experimental methodology and depicted in Figure 5.3. Finally, Step 8 of this testing methodology calls
for the derivation of empirical conclusions, as narrated in the next section.
5.4.4. Conclusions and critical findings of the experimental case study
Given the techno-geographical context backdrop of the pilot project system, the study can draw conclusions
about the reference case study results in the context of the FPV system’s geo-environmental application
within the Paarl arterial. The setting provides strategic value and significance to the geo-location-tagged
experimental results for the local RWE wine-farm installation in the visual sustainability metric represen-
tations of Figures 5.6 to 5.10. In this agricultural arena, electricity prices are high, productive agricultural
land is scarce and expensive, the agro-cultivation intensity is high, and water bodies are more readily avail-
able than fertile agricultural land to potentially install floating photovoltaic panels. This geographical area
of influence constitutes a wine-producing and olive-growing region renowned for its mosaic of famous vine-
203
yards and olive-tree orchards. This so-called southerly Agter-Paarl precinct of the Drakenstein Municipality
of South Africa is, as such, famous for its internationally recognised brand range of wines, olive oils and
exotic cheeses (produced from goats roaming the lush green vineyards). In this experimental reference-
cased pilot-project system, as depicted in Figure 5.5, the techno-geographical context includes fruit and
wine-processing facilities close to a farmyard that offers the consumption capacity to uptake all of the solar
electricity to be potentially generated by an FPV system during the daytime. In the broader context of the
spatial development framework for the Drakenstein Municipal Council, the investigation into the establish-
ment of an agricultural energy infrastructure through the concept of energy farming could aid the required
EIA processes in support of the environmental management framework and environmental management
review processes of the local council. This aspect is demonstrated in the redacted version of the sample
EIA project submission shown in Figure U.2.
While the geo-referenced simulation results Figures 5.6 to 5.10 can help to answer the research ques-
tions around floating PV-project enactments and performance-quality-assessments, the 3E metric set also
serves as floatoplanometrics (an integrated set of 3E metrics to help plan floatovoltaic projects). Sec-
tion 5.4.4.1 presents more detailed experimental conclusions around this thesis’s real-time analytical and
data-acquisition aspects and results on the assessment of a planned operational FPV project. In the same
context, Section 5.4.4.2 attends to the experimental conclusions around the results of a more strategic
assessment of the lifetime of an FPV project in preparation for extended data analysis procedures for the
future use of the results in decision-support capacity. Finally, viewed from a strategic FPV-GIS framework
and modelling perspective, Section 5.4.4.3 presents summarised conclusions around the strategic value of
the experimental evaluation results generated by the FPV toolset.
5.4.4.1. Real-time operational FPV project-assessment conclusions
With the PV projects rehearsed in cyberspace for the proposed RWE project context, the map and an aerial
image of the RWE wine farm in Figure 5.5 depict the virtual project terrain for the proposed real-world
FPV installation. As such, the selected experimental case study appraises and predicts the real-world
geographical performance of the FPV system at the farming reservoir referenced case within the Boland
theatre of operation. The real-time results also deliver physical evidence of the FPV model outputs in that
they deliver a multitudinous mix in the form of a 3E-integrated technical, economic and environmental set of
parameters and metrics during every simulation step or period. While the live stream FPV-GIS model output
results in Figures 5.6 and 5.7 records the underlying data in a time-index format pertaining specifically to
dates, the archived output in the acquisition data snapshot in Appendix R presents a sample of the model
outputs as real-time performance progression graphs. It practically assists the reader in developing a better
appreciation of the research conclusions within the context of the real-time FPV simulation operation and
outcomes delivered during the FPV-GIS system and operating processes of the digital twin model.
The spatio-temporal patterns of the integrated set of quantified 3E-performance results, as articulated
in Figure 5.6, exhibit some key output benefits of the closed-loop circular 3E systems-thinking integration
improvements offered by this thesis. While the analytical model is uniquely driven by an integrated closed-
loop analytical framework and a co-simulated system dynamics model logic, the analytical model has suc-
cessfully delivered the integrated set of 3E results in a newly established methodology for the theoretical
characterisation of floatovoltaics. The results indicate that the proposed integrated 3E modelling frame-
work can capture the broader and more diverse spectrum of anticipated floatovoltaic project outcomes
(refer to Section 2.3.2), covering all three sustainability criteria bases required for the integrated require-
ments for FPV performance and impact assessments. The discrete-time FPV model iteratively delivers
an integrated set of 3E output signals, parameters and metrics in real-time (refer to the philosophical goal
in Figure 3.3), providing evidence of the unique capabilities offered by the integrated closed-loop circular
204
analytical methodology proposed by this thesis for assessing floating PV technologies. The predicted out-
comes for the FPV system stand in sharp contrast to those comparative outcomes in Figure 5.7 predicted
for the GPV system with a conventional open-loop linear processing framework and model typically used by
conventional PV-performance-simulation methodologies. Compared to the conventional closed-loop frame-
work, the results can only ascribe credit for the superiority of the 3E modelling methodology to the analytical
simulation model’s underpinning theory, integrated closed-loop analytical framework, and system dynamics
model logic.
In the real-time performance progression results for an equivalent GPV system, as presented in Fig-
ure 5.7, the experimental results specifically acknowledge the fact that some of the greenhouse-gas offsets
and financial income contributions of floating solar FPV units in Figure 5.7 also apply to some extent to land-
based PV systems. In this context, Section 4.5.1 details how the FPV-GIS toolset is programmed to explore
the techno-economic and techno-environmental externalities of an equivalent co-located ground-mounted
GPV systems variant of the same specifications, in tandem with the proposed floating PV project assess-
ment. As such, the results in Figure 5.7 present the real-time analytical performance data predicted by the
FPV-GIS toolkit for an equivalent GPV system potentially to be co-located with the proposed FPV system.
These experimental results enable the environmental impact practitioner to directly compare the real-time
output performance curves and impact benefits of a counterpart GPV system, co-located with the FPV
system in a type of hybridised FPV and GPV systems configuration. Referring back to the date-indexed
results data in Appendix R, note how the three simulated time slices of the sampled data acquired from
the combined floating FPV and its ground-mounted GPV counterpart are continuously being logged during
each simulation period. In this way, the date-indexed model outputs form part of a larger structured data
frame (a two-dimensional matrix), whereby the key performance indicators for every FPV project simulated
over time are stored under the date-stamped index of the data-frame matrix.
An exciting result offered in the unique conditioning of the FPV-GIS model for the floating PV microhab-
itat concerns the estimation of the internal time-varying operating temperatures of the panels of the floating
PV system that, as presented in the results of Figure 5.6b, are based on floater-specific coefficients. When
comparing the model-predicted FPV panel temperatures in Figure 5.6b with the GPV panel temperatures
in Figure 5.7a, the FPV panel temperature estimations show a significant reduction in the PV panel temper-
atures for the floatovoltaic system. This effect occurs due to the cooling of the FPV system by a moisture
regime associated with a more humid enveloping microclimate. This benefit is, in fact, a hydroclimate tem-
perature phenomenon whereby the FPV model uniquely estimates the deemed ambient temperature and
humidity around the FPV system in terms of a novel meteorologically-based procedure and the solution
proposed by this thesis (refer to Section 4.3.3.3).
In terms of the integration of the dynamic time-varying microscale FPV hydroclimate variability into the
FPV-GIS simulation model of Figure M.1, we should observe that the model-predicted panel temperature
graph comparisons, as depicted in Figure 5.7a, show that especially during high levels of solar irradiation,
the FPV systems array consistently runs significantly cooler than its GPV counterpart. These lifestream
temperature reductions are enduring effects that can be directly ascribed to the higher humidity and cooler
aquatic microhabitat around the open water reservoir surface on which the FPV system is mounted. Since
the operating temperatures in the internal time-varying panels in the PV thermal model relate to the char-
acteristic power-temperature coefficients of the chosen PV panels, the PV panel-operating temperature is
one of the critical factors determining the prediction accuracy of the FPV model’s power-generation profile.
As presented in Figure 5.7b and c, the above-mentioned cooler microclimate impact effect implies that
the FPV system delivers a notably improved energy-yield profile (higher energy outputs) compared to the
GPV system. This cooling effect further helps to improve equipment integrity and ensure prolonged project
sustainability. The characteristic reservoir-induced microhabitat of FPV, and the hydroclimatic impact on the
FPV microhabitat, are unique to floatovoltaic technology. As such, the FPV-GIS model makes provision for
205
modelling the characteristic FPV microhabitat phenomenon (refer to Section 4.3.4.2), wherein the cooler
photovoltaic panels of the FPV system benefit from the signature power/temperature coefficients of the
selected solar panels/cells (refer to the PV inverse temperature/power output relationship in Figure 4.5).
While the energy model object of the FPV toolset predicts the real-time energy production yield samples
(fPmt) in tandem with the solar input data and module temperatures in Figure 5.7a, the impact of the
characteristic microclimate around the FPV ecosystems environment manifests itself in the characterisation
of the energy-yield profile of the FPV system, as presented in Figure 5.7b and c.
As such, the characteristic profile representing the magnitude of the power output for floatovoltaics (Pmt)
in Figure 5.6c is determined as a function of the physical characteristics of the solar panel, in combination
with the weather-modulated clear-sky solar radiation and the dataset samples of the ambient temperature-
predictor variables in each simulation iteration of Figure 5.6a and b. In this context, note how the cor-
responding hourly and daily cloud-cover impact and weather patterns are translated into the DNI pattern
in Figure 5.6a, and how this weather modulation impacts the generated power-output variations and the
energy-performance profile of the floating PV system in Figure 5.6c. Also, observe the prolonged FPV
efficiency gain (FpvGpv-Eff-gain) in the date-time stamped model-simulated results of Appendix R, thus
showing how, as opposed to its GPV counterpart, the cooler microclimate associated with the FPV system
offers an improvement of up to six per cent (6%) in the hourly energy-yield gain for the FPV system.
The results in Figures 5.6 and 5.7 continue to highlight the key aspects that help to improve project
sustainability in the composite environmental footprint profiles for the pre-configured FPV and GPV sys-
tems. The experimental results highlight the unique ability of the FPV-GIS toolset to forecast the projected
reductions in emissions as the simulated FPV/GPV project progresses through each period of iteration
over time. For the FPV-GIS toolset, the reductions in pollutant emissions through grid substitution (refer
to Table 4.1) are modelled in terms of the fact that the power-yield curves for the FPV and GPV systems
in Figure 5.6b and c signal the ability of the FPV system to avoid carbon emissions (refer to modelling in
Section 4.3.4.4). As such, the environmental model object of the system dynamics model proposed by
this thesis can uniquely predict the environmental-offset profile for the FPV and GPV systems during each
discrete-time processing instance of the simulation. In this context, as depicted in Figure 5.7d, the data-
array metric values, recorded as outputs of the model, predict the relative decarbonisation contribution of
the FPV and GPV systems variants in terms of the avoided carbon metric (xCO2). Note that the empirical
prediction for the improvement in the air quality as a result of the avoided CO2pollutant constitutes one of
the six environmental-offset metrics (avoided pollutants: xCO2, xSOx, xNOx, xPEa, xCoal, xAsh) predicted
by the FPV-GIS model (refer to the logged values for all of the pollutants in Appendix R).
As one of the most significant environmental impacts to help improve the sustainability of the project
delivered by the floating PV power-generating system, the cooler and more humid microclimate around the
FPV system, together with the solar flux screening offered by the floating PV canopy, is a further manifes-
tation of its ability to deliver unique water preservation benefits (refer to modelling in Section 4.3.4.3). The
water preservation results in Figure 5.7e demonstrate the unique ability of the FPV-GIS toolset to forecast
the total volume of water preserved by the FPV as opposed to that preserved by the GPV system. As a
result, the aggregated H2O resource preservation benefits offered by the FPV system (locally and nation-
ally) are significantly greater than those of the GPV system. The FPV system’s water resource preservation
benefit is quantified in the environmental impact graph for H2O preservation. The FPV system models
both the accrued volume of water evaporation avoided by the FPV (eH2O), on account of the local FPV-
panel shading on the irrigation reservoir and the accumulated national water savings (xH2O) ascribed to
coal-fired/water-cooled grid-electricity substitution.
Finally, the results in Figures 5.6 and 5.7 also highlight the delivery by the cooler microhabitat in the
immediate environment of the FPV system of improved financial prospects to help improve FPV-project fea-
sibility and sustainability. With increased power outputs in the energy-yield forecast of Figure 5.7c, the FPV
206
system can offer superior revenue income in terms of savings on grid-power purchases (grid substitution,
Rgs). In addition, the improved power-yield benefits further translate into increased carbon-taxation benefits
(carbon credits, Rcc) due to the improved carbon footprint profile, as presented in Figure 5.7d. At the same
time, additional economic value is generated by the FPV system in terms of the avoided evaporative water
savings, as presented in Figure 5.7e. Together with the agricultural-production income (Rxp), rescued in
terms of the land-saving benefit offered by the FPV system, all of the income mentioned above factors, as
shown in Figure 5.7f, are included in the comparisons between the total revenue generated by the FPV and
GPV systems respectively.
While the quantified real-time results for the empirical performance outputs of the FPV systems variant in
Figure 5.6 demonstrate the discrete real-time simulation capabilities of the implemented FPV co-simulation
model, the performance results for the PV systems in Figure 5.5b show how the experiment responds to
the first objective of the investigation, as defined in the experimental methodology of Figure 5.3, namely
to assess and plot the real-time performance-profile curves of the FPV and GPV systems over a real-
time window period. With the real-time FPV/GPV performance and impact comparisons, as portrayed in
Figure 5.7, the following section deals with the second objective of the experiment, namely to assess and
plot on a graphic dashboard display the lifetime systems-performance profiles for the proposed FPV systems
variant and its GPV systems counterpart.
5.4.4.2. Lifetime operational FPV project-assessment conclusions
During the data-analysis phase, the visual sustainability metric representations for the real-time sliced 3E-
simulation datasets in Appendix R are integrated over time, thus accumulating the sampled 3E-simulation
outcomes to determine the lifetime performance results of the FPV system under consideration, as given
in Appendix S. The redacted version of the EIA scoping assessment report shown in Figure U.2 shows
how the FPV model’s analytical data archiving capability supports the environmental due-diligence and EIA
assessment processes for the project site in Figure 5.5. It illustrates how the FPV-GIS model is uniquely
able to quantify the 3E lifetime-performance results for the FPV system, and how it can be uniquely com-
pared to the 3E performances it’s GPV counterpart adjacent to the water reservoir. In this context, the
cascaded results portrayed by Figures 5.8 to 5.10 define the performance signatures for FPV and GPV. It
generally shows that, as opposed to its GPV counterpart, the FPV microhabitat ensures critical enhance-
ments in the performance and impact benefits that endure throughout the project lifetime of the FPV. These
include, amongst others, increased efficiency in panel-power generation, increased carbon sequestration
benefits, reduced evaporation from the water reservoir, and reduced farmland usage, all due to the spe-
cific aquatic surface and environmental microhabitat associated with the FPV installation. Another of the
leading tertiary benefits of accommodating photovoltaics on the water is that the energy system can serve
as an on-facility power source for irrigation for pumping water onto the lands/vineyards or driving winery
operations, such as refrigeration or cooling processes. While an illustrative set of time-sliced snapshots of
the annualised simulation outputs produced through the operations of the experimental floating PV systems
model is presented in Appendix S, the lifetime results derived from these data, as explained in Figures 5.8
to 5.10, meet the second objective of the experiment in that they demonstrate the integrative capabilities of
the FPV model, in terms of assessment and simulation, in a cumulative life-cycle version of a floating PV
performance-assessment model.
In support of the sustainability of the project, the conclusions reached in respect of the real-time perfor-
mance results (under Section 5.4.4.1), and the quantified floatovoltaic performance and impact results for
the empirical-performance outputs of the FPV systems variant and presented in Figures 5.8 to 5.10, demon-
strate the value of a federalism-based framework as a model concept in a lifetime 3E costing methodology.
The experimental results thus confirm the lifetime-simulated performance capabilities of the implemented
207
floating PV model in terms of the objectives defined in the testing methodology of Figure 5.3. In this way,
the empirical performance and impact results, as recorded in Figures 5.8 to 5.10, confirm many of the
salient impact attributes of floating PV technology (discussed under Section 2.3.2). Furthermore, the geo-
referenced contextualities embedded in the integrated set of 3E-performance results highlight some of the
critical benefits and co-benefits of the floating PV technology variant. With the results delivering a broad
overview of the project’s scope, the experimental results demonstrate the unique ability of the FPV-GIS
toolset to co-simulate and predict a broad spectrum of 3E responses and benefits. By engaging novel
FPV-GIS tool features, the co-simulation framework and closed-loop feedback model help alleviate many
real-world project-planning challenges currently experienced with conventional sequential open-loop PV
performance models.
A key conclusion drawn from the experimental results in Figures 5.8 to 5.10 is that the decrypted life-
time 3E-performances and impact effects of FPV technology serve as a valuable data resource that can
directly apply for due-diligence investigations into projects, business case investment analysis, and EIA-
based project-licensing authorisations. From an EIA perspective, an extraction from the provisional EIA
report completed from the experimental results in Figures 5.8 to 5.10 are given the EIA scoping report at-
tached in Figure U.2. The experiment succeeds in offering scientifically-motivated, rapidly generated data,
offering the possibility of making predictions by tentatively profiling the performance characteristics and im-
pact effects of the proposed FPV project variant over its entire lifetime period (in this case, 20 years). From
an EIA perspective, the results in Figures 5.8 to 5.10 complete the circle between the FPV-GIS model,
proposed by this thesis. In this context, the FPV-GIS model’s empirical results in technology assessment
results constitute Figures 5.8 to 5.10 provides duly required scientific evidence to document the EIA-related
impact effects for the proposed FPV project and its GPV counterpart as given in the EIA scoping report of
Figure U.2.
In the context of the energy budget dynamics of a floatovoltaic system, the lifetime technical power-
generation performance results presented in Figure 5.8 offer a valuable insight into the technical benefits
provided by the floating PV systems’ reservoir microhabitat. While real-world power system simulation
measurements require transducer measurements of 20-25 years to determine the reliability of the module
(Kim et al., 2021a), the FPV-GIS model can forecast and compare the long-term outdoor performances
of a floating PV module under diversified terrestrial conditions. In terms of the energy-yield assessment,
the lifetime-yield predictions in Figure 5.8 show that an FPV system’s energy production outperforms the
GPV system’s power-production profile during every year of operation. The energy-performance profile in
Figure 5.8 further shows how the FPV power-efficiency gain continues to surpass GPV performances over
time. This efficiency gain can be ascribed to the lower performance loss rate and temperature degradation
owing to the impacts of the cooler microclimate on the FPV system that propagate into future systemic
performances over time. For the GPV case, the performance profile in Figure 5.8 highlights the continuing
derating effects of the temperature-related power on the system, and an escalation in the process whereby
the power outputs are reduced, thus resulting in a gradual decline in the power output over time. In terms
of the numerical energy-efficiency gain of the FPV system, the results of Figure 5.8 indicate that the cooler
microclimate associated with the FPV system ensures an improvement of up to six per cent (6%) in the
average hourly energy-yield gain in the first year of operation when compared to that of its land-based GPV
systems counterpart, while this average FPV efficiency gain increases to up to 19% at the conclusion of the
project’s lifetime (see parameter Eff-gain-FPV in Appendix S).
While it is notoriously difficult to quantify the environmental impact effects of floatovoltaics (Gadzanku
et al., 2021b; Hernandez et al., 2019), incorporating an understanding of environmental principles into the
FPV-GIS toolset assists in overcoming this difficulty. In terms of the environmental budget dynamics, the
logic of the model framework can help deliver a broad spectrum of environmental-impact performances in
the environmental profile shown in Figure 5.9. As part of the broader environmental-offset profile of an
208
FPV system, these environmental performance metrics directly impact project sustainability as empirically-
quantified impact results in the environmental profile for the FPV system. From an environmental-offset
perspective, the ability of the proposed FPV model to forecast and quantify the water-saving potential of a
floatovoltaics power plant, as in Figure 5.9, is considered one of the most exciting outcomes. In this environ-
mental impact model prediction, the results reflect the total amount of water preserved (local evaporation
and national grid cooling H2O) by FPV as 26.82 ML, compared to that of 6.31 ML by GPV for grid cooling
avoided (national H2O). These water-preservation savings for FPV include both the accrued volume of water
evaporation (eH2O=19.67 ML) over that which FPV avoids over the systems’ lifetime, and the accumulated
national grid cooling water savings (xH2O=7.15 ML) ascribed to the substitution of coal-fired/water-cooled
grid electricity. Since only the grid-cooling water-preservation factor (xH2O=6.31 ML) applies to GPV sys-
tems technology, the lifetime water-preservation results, as presented in Figure 5.9, confirm that the FPV
system offers superior water-preservation benefits over its land-mounted GPV systems counterpart. From
an environmental-offset perspective, the FPV water-saving results provide an interesting WELF-nexus per-
spective on the water-retention qualities and the water-security aspects associated with an FPV system (in
comparison to GPV).
Furthering the environmental-offset perspective (refer to the SA context in DFFE (2015)), the experi-
mental environmental-offset footprint results in Figure 5.9 provide valuable insights into the environmental-
offset value propositions offered by FPV. The results provide scientific evidence of the unique ability of the
toolset to quantify the broad spectrum of environmental-impact performances delivered by the FPV system
as opposed to those of its GPV counterpart. The lifetime environmental footprint of the FPV system, as
depicted in the lifetime environmental-offset report card in Figure 5.9, empirically quantifies the grid-energy
substitution in terms of the avoided emission of the pollutants xCO2, xSOx, xNOx, xPEa, as air-quality im-
provements predicted by the FPV-GIS model, in addition to the avoidance of large volumes of coal fuel and
ash pollution, namely xCoal and xAsh respectively, emanating through grid substitution. The avoided life-
time carbon-pollutant metric (xCO2) offers a significant empirical result since it results reflects the carbon-
sequestration footprint credentials for the FPV system, which are superior to those for its GPV systems
counterpart (refer to the numerically-logged values in Appendix S). In the context of the South African grid-
energy mix (DFFE, 2021b; Prinsloo, 2019), the improved reduction in carbon emissions for FPV over GPV
(5.18 kton vs 4.5 kton) can be ascribed to the evaporative cooling and the reduced operating temperature
of the FPV panels. As such, the FPV system delivers higher power outputs to further translate, as part of
grid substitution, into increased levels of avoided carbon-polluting emissions (xCO2). This reduction is in
addition to the carbon-drawdown potential of the agricultural crop rescued (i.e. grapevines) owing to the
preservation of farmland and the health of the soil (refer to Section 4.3.4.6). These results exemplify the
predicted carbon-sequestration footprint credentials of the FPV system for the configured project scenario.
In addition, the farmland-surface reservation factor (FLp) quantifies another one of the crucial aspects
as environmental-offset effects quantified by FPV technology. This model estimates this value to be worth
around R2.8 million at current land prices in the area. From a sustainability-nexus perspective, this metric
quantifies one of the salient attributes unique to floatovoltaics. It is also most valuable since it impacts project
sustainability regarding the Land and Food aspects of the WELF nexus. It also has an associative economic
contribution factor (Rxi = R2.8m) to make it that it reflects the farm land’s real-estate value. Regarding any
land acquisition, the Rxi metric can also be viewed as an avoided liability for the landowner should he install
an FPV system over a water body. This saving contrasts with the expense of conventional over-land GPV
systems, where the PV installation would generally result in the landowner having to sacrifice the same
amount of valuable agricultural farmland. The predicted size of the area of farmland preserved by the FPV
system (FLp-FPV) amounts to 1120.75m2in the first year of operation, increasing to 1332.18m2in the final
year of operation. In terms of the real estate asset value, the FLp factor thus translates into a farmland-
savings value of R2.8 million to R3.3 million (LandValue-FLp-FPV). In terms of project income during the
209
first year of FPV installation, the FLp factor translates into a vineyard-harvesting value of R27414 per year
derived from the sale of the estimated saved grape harvest of 1793 kg (FreshFood-FPV). Alternatively, the
agronomic value of FLp can be seen to translate into a winery-commodity production value of R0.4 million
in terms of income derived from the value of 1355 bottles of wine (ProcdFood-FPV) that can be produced
from the rescued farmland harvest.
Regarding the broader financial feasibility aspects of FPV-project sustainability, the prediction of the
economic budget dynamics of the floatovoltaic system in Figure 5.10 provides valuable insights into the
economic value propositions offered by FPV. The monetary offset results of the FPV project portrayed by
Figure 5.10 highlight important quantitative performance-prediction novelties provided by the 3E framework
and the FPV model proposed by this thesis. The FPV toolset shows that an FPV system can deliver a range
of economic offsets in terms of water savings/preservation (Water), increased energy/power outputs (En-
ergy), agricultural land preservation (Land), agricultural produce (Food), all of the components that make
up the WELF-resource-system segments as pillars of floatovoltaic sustainability. Note that the agrivoltaics
option for both GPV and FPV is disabled in the configuration file, thus meaning that food production is
generated only by the preserved farmland. From a business-enterprise perspective, the lifetime financial-
performance results in Figure 5.10 demonstrate how all of the WELF-resource parameters in Figures 5.8
to 5.9 can also help to generate revenue and additional income for the floatovoltaic power-plant installation.
To highlight some of the important sustainability indicators, the levelled cost for FPV (LCOE-elect-Yr-FPV)
is determined at 25c/kWh, which is comparable with that of 0.29c/kWh for GPV (LCOE-elect-Yr-GPV). The
water economic surface-transformation value (w-EST) is determined at R614.26/m2. The EST or WEST
indicator is uniquely defined by this thesis to indicate the normalised annual monetary value of the trans-
formed water surface as a recipient of the FPV system (previously unutilised farming space). In the exper-
iment, the value of R614.26/m2for this indicator means that the farmer would generate around R614 for
every square metre of water surface area covered by FPV panels when installing the FPV systems variant
proposed by this experiment. In general, the results in Figure 5.10 can help to answer important questions
that may need to be addressed in terms of the project’s financing and revenue-generating schemes before
the implementation phase of the photovoltaic power plant (IEA, 2021).
Finally, this experimental investigation is largely framed around the broad paradigm of sustainable devel-
opment in agricultural farming, which begs for quantifying the natural resource impacts of the FPV system.
To this end, the empirical characterisation of the diverse impact-effects profile as environmental offsets for
the FPV system in Figure 5.9 helps to clarify the WELF-resource impacts underlying the operational en-
vironmental offset of the floating PV systems variant relative to its associated GPV counterpart. In this
context, the empirical lifetime 3E-performance and impact results in Figures 5.8 to 5.9 offer a unique oppor-
tunity to translate the predicted floatovoltaic systemic outputs into other derivative parameter sets, such as
the WELF-nexus parameters. The lifetime-performance results in Figures 5.8 to 5.9 thus directly support
the derivation of the WELF-nexus systemic resources in terms of water savings/preservation (Water), in-
creased energy/power outputs (Energy), agricultural land preservation (Land) and the agricultural produce
on the preserved farmland (Food), in fact, the pillars of the WELF system.
On a decision-making level, the results in Figures 5.8 to 5.9 strategically serve as a testament to the
unique ability of the FPV-modelling framework, or the 3E-tridentate framework in Figure 3.14, to translate
the lifetime-performance results for a floating PV project into WELF-nexus parameters, according to the
philosophical principles conceptualised by this thesis, as in Figure 3.15. More generally, the overall results of
this experiment demonstrate the value of the integrated 3E framework and FPV model in terms of translating
the floatovoltaic 3E-model outputs into decision metrics/indices suitable for goal setting in a decision-support
framework, such as the WELF-resource framework, geared to derivative-type energy sustainability, and to
be demonstrated in the next experiment.
210
5.4.4.3. Strategic FPV project assessment conclusions
Strategically, the digitally synthesised experimental results for this scenario-driven FPV installation at the
proposed RWE wine-farm location give a practical exhibition of the tactical efficacy of the co-simulated tri-
dentate 3E-modelling framework theory, methodology and technique. The results are a testament to the
improved sustainability theory and FPV-GIS-modelling framework postulated by this thesis in Figure 3.14.
The main conclusion inferred from the experimental evidence, presented through the operational visual
sustainability metric representation of the results of this experiment, is that the purposively-designed inter-
active framework (modelled around the energy, environmental and economic trilogy) provides a valuable
quantitative methodology geared to experimental synthesis to characterise the sustainability attributes of
the floatovoltaics technology ecosystem digitally. The theory helps to actualise the quantitative assessment
of the key sustainability aspects of floating PV technology through integrated diagnostic analyses. For the
proposed RWE project context, the experimental case-study results in the visual sustainability metric rep-
resentations of Figures 5.6 to 5.10 provides a valuable exhibition of the unique geo-performance-profiling
capabilities of the integrated sustainability assessment theory posited by the thesis. With the 3E sustain-
ability theory underpinning the processing operations of the FPV-GIS toolset for a real-world PV operating
environment in South Africa, the empirical results provide a sound basis to support sustainability profiling
of floating PV technology projects in terms of key sustainability performance indicators (as is shown in the
following experiment).
The desktop-driven laboratory experiment further demonstrates the viability of a new theoretical frame-
work and method for quantitatively assessing the attributable characteristics of floatovoltaic technology un-
der actual real-world operating conditions. As such, the experimental results show that the new 3E assess-
ment theory and framework provide more realistic predictions of FPV project outcomes than the anticipated
outcomes for an operational FPV system concerning a GPV system as given in Section 2.3.2. In terms of
the practical utility of the experimental results, the results portrayed by the metric representations in Fig-
ures 5.6 to 5.10 meets the reporting requirements for carbon taxation as well as the National Treasury reg-
ulatory requirements in terms of the Carbon Tax Act (DavisTaxCommittee, 2015; National Treasury, 2020).
From a regulatory environmental authorisation perspective (ELaw, 2022), the results further uniquely meet
most of the complex operational EIA directives in terms of enabling electronic E-GIS submission-reporting
requirements towards virtual or real-world PV project registration and application for PV project certifica-
tion (EGIS, 2022; RSA, 2021). Therefore, the experiment overall demonstrates the prospective benefits of
programmatic-type economic due-diligence assessment as part of environmental impact assessment and
fully integrated technical energy potential assessment as part of the broad-spectrum sustainability impact
assessment theory posited by this thesis.
Quite notably, the sustainability profiling results of this experiment serve as scientific evidence of the ex-
tent to which 3E theoretical postulate of this thesis advances the state-of-the-art in PV performance profiling.
The PV performance modelling outcomes of conventional performance models are unreliable in floating PV
assessments, as the results are considered full of holes and gaps. Incidentally, these holes and gaps are
recognised as the so-called technology unknowns of floating PV systems, the exact same holes and gaps
filled by the FPV model as evidenced in the comparisons between the FPV and GPV model predictions in
Figures 5.8 to 5.10. Highlighting the mismatch of a conventional PV performance model in assessing float-
ing PV performances, the distinction between the synthesis results for the FPV and GPV system models
in the visual sustainability metric representations of Figures 5.7 to 5.10 reflects the performance profiling
errors or model inaccuracies to be expected when inconsiderately applying a conventional GPV tool in the
quantification of the broad spectrum of sustainability attributes of the FPV system variant. Remember that
the GPV results are digitally synthesised with a conventional open-loop PVlib model, while the FPV results
are digitally synthesised with the proposed open-loop 3E framework-driven FPV model. In this context,
211
the distinctive performance profile differences between FPV and GPV project outcomes in Figures 5.7 to
5.10 inherently provide valuable scientific evidence of the lack of precision and measures of inaccuracy
or error that would be encountered when employing a conventional GPV performance analysis tool in the
quantitative assessment of an FPV installation project. While the impact of the mismatch is escalatory over
time, both the real-time and lifetime experimental results in Figures 5.6 to 5.10 highlight the imprecision that
would be encountered as a result of the knowledge gaps emanating out of the use of a conventional GPV
performance analysis model in the quantitative sustainability assessment of an FPV installation project vari-
ant. As such, the comparative FPV/GPV results in Figures 5.6 to 5.10 highlight the value of the quantitative
methodological postulations of this thesis to empirically quantify the broad spectrum of (3E) sustainabil-
ity attributes of floatovoltaics through an integrated theoretical framework that can overcome performance
assessment imperfections of GPV models in the assessment of FPV projects.
In summary, this experiment demonstrates the value proposition offered by the analytical layer in the
geospatial FPV toolset in creating situational awareness to deliver a range of interdisciplinary 3E results
based on the integrated analytical and quantitative data-collection capabilities of the model. In terms of
further data analysis, the following experiment is concerned with the value proposition offered through
an analysis of the 3E-framework results in quantitative decision-making. In this respect, it engages the
decision-support layer of the FPV toolset to demonstrate its WELF-resource-related decision-support capa-
bilities.
5.5. Experiment 3: Floating Solar Decision-support Analysis
The previous experiment successfully demonstrated the theoretical geo-sensitive approach for exploring the
combined effects of the technical, economic, and environmental aspects of the 3E or tridentate narratives
of floating solar development plans in a local agricultural context. This experiment uses the 3E metrics from
the previous experiment to generate scientific test results to help answer the research questions around
integrating WELF-resource parameters into a floating PV decision-support model.
Integrating the analytical principles offered by the FPV-GIS toolset into the functional planning layers
of a geo-informatics platform establishes functional decision-support capabilities in an integrated empirical
project assessment and decision-support platform for sustainable development. As such, this experiment
aims to uncover the role of a theoretical geo-sensitive approach in a technology-specific floating PV toolset
in demonstrating the ability of the FPV model to explore the combined effects of the technical, economic,
and environmental (3E; tridentate) narratives in a decision-support model. The research questions have
been formulated around the project’s essential requirements for due diligence and its assessment within
the regulatory context of the scientific EIA framework requirements. Towards answering the question, this
experiment engages the analytical framework and systems-based FPV digital twin model to present sci-
entific results that demonstrate the use of decision mechanisms in the planning context of an agricultural
floating solar development.
While the FPV-GIS toolset is capable of exploring the techno-economic and techno-environmental exter-
nalities of a parallel ground-mounted GPV systems variant of the exact specifications (refer to Section 4.5.1),
the operational methodology of this experiment allows for the set of results generated in Experiment 5.4,
and as depicted in Figure M.1, to be integrated into the AHP decision-support layer and architecture of the
FPV-GIS toolset. Furthermore, concerning the FPV-GIS-driven decision-support model, this experiment fur-
ther compares the animated performance and impact results for the FPV systems variant FPV-GIS-toolset
of Experiment 5.4, with a co-located land-based GPV system of the same dimensions, the latter considered
to be installed adjacent to the planned FPV system. In this way, when presented with given project alterna-
212
tives in development and the EIA process requirements, the experimental results demonstrate the FPV-GIS
toolset’s ability to evaluate the selection options of the empirical project (refer to Figure 2.7).
The experiment’s goal is discussed in Section 5.5.1, while the practical experimental method and pro-
cedures are recorded in Section 5.5.2. Section 5.5.3 details the experimental case-study results, while
Section 5.5.4 provides reasoned conclusions and critical findings of the experimental results.
5.5.1. Goal of experiment
The previous experiments on performance diagnosis have laid a solid foundation by presenting digital com-
puter synthesis results. While it helps to answer the research questions regarding the success of conceptu-
alising a novel theoretical modelling framework and an empirical model expression, this experiment focuses
on FPV project scenario analysis to uniquely characterise a floating PV system in terms of the WELF nexus
parameters computed from the 3E-tridentate framework (refer to Figure E.1a). By building on the functional
value of the practice-orientated FPV-GIS computer-synthesised 3E framework and model, this experiment
aims to use the quantitative FPV performance results of the previous project design stage experiment to em-
pirically predict and profile the sustainability qualities of the floating photovoltaic project outcomes in terms
of an extended set of climate-economic water-energy-land-food resource indicators. The strategic goal is to
investigate the success of digitally translating the assessment metrics of the 3E-tridentate framework into
on-site WELF-resource indicators (refer to Section 4.4.2, and the concept of Figure 3.15). The goal is to use
mathematical reasoning principles to weigh decision criteria statistically to make optimal project decisions
or to optimise project design according to the procedures provided by the balanced scorecard decision
theory that drives the AHP technique (Section 4.4.3). While the WELF-nexus indicators are determined an-
nually, the WELF data for year one of the project illustrate sustainability profiling in this experiment. Profiling
the sustainability qualities of a floating photovoltaic project in terms of a set of sustainability performance
indicators, this experiment determines the enviro-econo-WELF index indicators as sustainability portfolio
framework described in Section 4.4.2. The experiment uses this on-site ce-WELF data as sustainability
criteria to drive visual and empirical decision support.
The operational goal is to determine the enviro-econo-WELF sustainability index (see ce-WELF metrics
in Section 4.4.3) and to demonstrate the use of the data-driven decision-supporting capabilities of the
implemented FPV-GIS toolset. This goal engages the defined ce-WELF index to officiate sustainability
rating rankings in s comparison between a scenario-based FPV variant and a GPV counterpart in a typical
use-case scenario. In terms of statistical decision support (Saaty, 2000), this experimental assignment
thus sets two objectives: Firstly, to demonstrate the visual articulation of the sustainability-profiling results
to evaluate plausible project options on the decision-supporting profile-mapping display generated by the
FPV-GIS toolset during the project design stage. Secondly, in using the AHP process in the statistical
processing of the 3E-driven WELF-nexus performance measures, the experiment aims to demonstrate the
mathematical and numerical prioritisation of plausible project options among a potential portfolio of PV
energy projects for a given case scenario around their installation and associated narrative.
5.5.2. Experimental method and procedure
To help navigate the experimental method in terms of the procedures for this project design stage scenario-
based investigation, the analytical decision-support processing for this experiment is listed in Figure 5.11.
Steps 1 and 2 prepare for the experiment by defining the decision goal, selecting the criteria, and deter-
mining the criterial weights according to AHP procedures. The AHP multi-criteria decision weights, already
determined in Section 4.4.3 are defined in Table 4.2. With the preparatory steps already completed in
Section 4.4, the AHP decision goal is to predict the PV project sustainability performance in terms of the
213
ce-WELF nexus metrics. As presented in Section 4.4.2, these metrics constitute a PV sustainability index
with metric components/elements to serve as the AHP decision-support criteria.
step 1 step 2 step 3 step 4 step 5 step 6 step 7 step 8
Define the
project
performance
decision goal
Select the
decision
criteria
metrics and
determine
AHP metric
weights
Determine
the criteria
metrics
values from
simulated
FPV and GPV
performances
Normalise
criteria
metric values
and store
as criteria
priority results
Run AHP
decision
sensitivity
analysis to
synthesise
project profile
decision
priorities
Record the
performance
scores for the
project regime
options
Articulate
results on
graphical
decision
support profile
display map
Derive
experimental
conclusion
Figure 5.11: Experimental method and procedure for AHP-based decision support and selection of alternatives,
according to the defined experimental methodology (source: author).
With the decision criteria defined, and their relative weightings established, Step 3 determines the an-
nual performance values for the selected ce-WELF decision criteria metrics in the FPV and GPV project
variant options from the results in Experiment 5.4. Proceeding onwards to Steps 4 to 7, the decision-support
layer of the FPV-GIS toolset (Section 4.4.1) engages the implemented AHP algorithm to process the criterial
metric values into normalised probability scores for the project criteria (Step 4); to configure and run the
AHP decision model. In so doing, the AHP criteria weights are applied to the normalised criteria metrics
and are thus able to determine the sustainability scores and preference options for the candidate projects
in the AHP sensitivity analysis (Step 5); to record the performance scores for each of the AHP criteria for
each project-regime option in a table (Step 6); and to plot the results on a spider diagram as a means of
articulating the results on a decision-supporting profile-mapping display. Finally, the experiment concludes
by reporting the experimental findings in Step 8. These processing steps are followed to determine the
results detailed in the next section.
5.5.3. Experimental case-study results
This section presents the experimental results of applying the principles and techniques of the analytical
AHP process in the evaluation, articulation, prioritisation and selection of project profiles in a portfolio of
project options during the project’s design stage.
Figure 5.12 details how the decision problem of the PV project is structured in terms of the decision
goal (top layer, sustainability goal layer), the decision criteria in terms of ce-WELF metrics (middle layer,
ce-WELF input layer), and the decision options (bottom layer, FPV/GPV outcomes layer).
Sustainability
CO2
FPV GPV
EST
FPV GPV
Welf-nexus
FPV GPV
wElf-nexus
FPV GPV
weLf-nexus
FPV GPV
welF-nexus
FPV GPV
Figure 5.12: Project-decision hierarchy to solve an integrated multi-criterial planning and investment decision problem
for sustainability in a floatovoltaic system (source: author).
214
In this context, the first task is for the decision layer of the FPV-GIS toolset to translate the 3E-tridentate
assessment outcome metrics into WELF-nexus parameters. These calculations are performed on the basis
of the formulations presented in Section 4.3.6.5.
The first step is ce-WELF sustainability priority vector estimation. As such, experimental processing
Step 3 of Figure 5.11 determines the ce-WELF criteria metrics from the simulated FPV/GPV performances.
Since the FPV-GIS toolset has already explored the techno-economic and techno-environmental externali-
ties of the integrated floatovoltaic ecosystem, the annual performance profile results for the FPV and GPV
systems, as defined and tested in Experiment 5.4, can be translated into FPV-sustainability indicators for
articulation as ce-WELF-nexus parameters.
By processing the projected annual project outcome results in terms of the WELF-nexus calculation
procedures, as presented in Equations 54 to 58 and detailed in Section 4.3.6, the procedure can empirically
determine the annual WELF-nexus footprint of the floatovoltaic system. As such, the WELF-nexus parame-
ter elements for the proposed FPV installation case study of Experiment 5.4 are determined respectively as
505 kL of Water preserved, 116 MWh of Energy generated, 1283 m2of farmland preserved (owing to FPV
land reclamation), and 39.6 tons of grapes as fresh Food preserved (owing to FPV farmland preserved). In
terms of the remaining "ce” components of the ce-WELF index (Section 4.3.6.5), Equations 59 to 61 provide
for the CE parameter calculations. For the proposed FPV systems variant, the total volume of CO2pollution
avoided (owing to the clean energy generated by the FPV system) amounts to 114 tons, while the total
economic value of the FPV water-surface transformation is 196.00 ZAR/m2. With all of the components of
the annual ce-WELF sustainability index determined, the ce-WELF sustainability priority vectors portrayed
in the visual ce-WELF sustainability indicator representations of Figure 5.13 to articulate the performance
parameters and environmental-resource impacts for the floating solar system as ce-WELF nexus parame-
ters in a conventional analytical decision-supporting display. Note again that the GPV system counterpart
does not preserve any farmland at the site. In contrast, the farmland it occupies does not contribute to agri-
cultural food production, which explains why the estimates for the WELF resources for "Land preservation”
and the "Food preservation” are at zero value in the indicator representations of Figure 5.13.
FPV
GPV
262
264
266
268
270
272
274 273.35 ton
262.45 ton
Tons (t)
Climate
(a) CO2offset
FPV
GPV
540
560
580
600
620 R614.26/m2
R537.55/m2
Rands/m2(ZAR/m2)
Economic
(b) Income transform
FPV
GPV
400
600
800
1,000
1,200
1,400 1361.91 kL
363.06 kL
Kilolitres (kL)
Water
(c) Water preservation
FPV
GPV
264
266
268
270
272
274
276
276.21 MWh
265.00 MWh
Mega Watt hours (MWh)
Energy
(d) Energy generation
FPV
GPV
0
200
400
600
800
1,000
1,200 1120.75 m2
0 m2
Square meters (m3)
Land
(e) Land preservation
FPV
GPV
0
0.5
1
1.5
1.8 ton
0 ton
Kilograms (kg)
Food
(f) Food preservation
Figure 5.13: FPV-GIS toolset determining the climate-economic-WELF performance-parameter profiles engaged as
decision scorecard metrics input to the AHP decision-support process (blue=FPV, red=GPV) (source: author).
Similarly, according to calculation procedures in Equations 54 to 62, the PV performance toolset trans-
lates the projected annual 3E metrics for the PV installation case study of Experiment 5.4 into the empirical
WELF-nexus parameters for the ground-mounted GPV systems counterpart. As such, the WELF-nexus
footprint of the GPV system is empirically determined respectively as 153 kL of Water preserved, 114 MWh
of Energy generated, 0.0 m2of farmland preserved (farmland forfeited), and zero tons of grapes as fresh
food preserved (agricultural land/produce forfeited to install GPV). In terms of the remaining ce components
of the ce-WELF index (Section 4.3.6.5) for the GPV counterpart, the total volume of CO2pollution avoided
215
(owing to clean energy generated by the GPV) amounts to 110 tons, while the total economic value of the
GPV land-surface transformation is 158.00 ZAR/m2(economic surface transformation). For a comparison
with the ce-WELF index, metrics for the GPV counterpart are also plotted in respect of the ce-WELF graphs,
as presented in the visual sustainability indicator representations of Figure 5.13.
With the ce-WELF sustainability-index metrics digitally synthesised in Figure 5.13, the profile-decision
priorities of the project are the next to be determined. These values are predicted through the forward
sequencing of the steps in the propagation network of the AHP process, as depicted in Figure 5.14 and
Steps 4 and 5 in Figure 5.11. The ce-WELF sustainability-index metrics are first normalised (refer to Equa-
tion 1) to express the ce-WELF metrics as criterial probability scores (ce-WELF PV probability scores for the
project), as presented in Table 4.2. Next, the AHP method in Figure 5.14 digitally synthesises the decision
priorities of the project profile by applying the AHP criteria weights (the AHP project goal policy, defined by
the project goal weights in Table 4.2) to the normalised criterial metrics of the project. Known as the com-
parative project-decision roll-up process in the "apply AHP weights layer” of Figure 5.14, which involves the
multiplication of each of the criteria weights with each of the priorities of the components of the performance
measures of the PV project (normalised ce-WELF metrics). Finally, the weighted priority scores for each
project are totalled in Table 4.2 to determine the individual FPV/GPV project-priority scores, both in terms
of the unscaled/unweighted and scale/weighted project-sustainability balancing-outcome display.
Welf (Water)
wElf (Energy)
weLf (Land)
welF (Food)
CO2 (Climate)
EST (Economic)
.
.
Sustainability rating FPV
Sustainability rating GPV
Apply
AHP
weights
Normalise
index
metrics
AHP
decision
score
Figure 5.14: AHP decision processing in a forward propagation network to apply AHP criterial weights for decision-
making to normalised sustainability index criteria as ce-WELF nexus metric probabilities to determine relative
weighted sustainability scores for FPV and GPV candidate projects (source: author).
The performance scores for the project-regime options in each of the sequences in the AHP process-
ing steps in Figure 5.14 are recorded in the empirical indicator representation of Table 5.5 (Experimental
Step 6). This table shows the AHP criterial weights for decision-making the normalised principal Eigenvector
weight factors for each indicator dimension in the third column, and the normalised ce-WELF sustainability-
index metrics as probability scores for the sustainability criteria performance of the project (ce-WELF PV
project-performance probability) in the fourth column. The fifth column of Table 5.5 lists the weighted prob-
ability scores for project criteria to emphasise the importance of the ranking scores for each of the chosen
ce-WELF sustainability metrics to serve as decision criteria for each of the support elements of the decision
indicators for the floating PV project. The sixth column in Table 5.5 summarises the weighted probability
scores for the project criteria and presents the project-performance scores for the project-regime options,
as recorded in the empirical indicator dataset of Table 5.5, and evaluated for project likelihood or project-
ranking. The sustainability profile probabilities for each of the candidate projects (FPV and GPV) are statis-
216
tically weighted in terms of the user-defined project decision goals to assign a mathematically meaningful
sustainability score to each project as ratio scale measure.
Table 5.5: Performance-probability scores for sustainability in respect of the FPV and GPV projects, and expressed
in terms of climate-economic-WELF-nexus pillar probabilities (source: author).
Analytical hierarchy of weighted project-decision scores: FPV project
Category AHP Criteria Criterial weight FPV prfmce prob FPV weighted prob FPV score
Water 0.133 0.79 0.11
WELF Energy 0.237 0.51 0.12
Land 0.087 1.00 0.09 Σ0.63
Food 0.066 1.00 0.07
Climate CO20.296 0.51 0.15
Economic EST 0.180 0.53 0.01
Σ1.00 Σ0.72 Σ0.63
Analytical hierarchy of weighted project-decision scores: GPV project
Category AHP Criteria Criterial weight GPV prfmce prob GPV weighted prob GPV score
Water 0.133 0.21 0.03
WELF Energy 0.237 0.49 0.12
Land 0.087 0.00 0.00 Σ0.37
Food 0.066 0.00 0.00
Climate CO20.296 0.49 0.14
Economic EST 0.180 0.47 0.08
Σ1.00 Σ0.28 Σ0.37
Throughout the decision-making process, the digitally synthesising technique of the AHP decision pro-
cess generates a range of statistical measures to articulate the decision profile results at different stages
of the AHP process (refer to Table 5.5). These statistical data calculations offered empirical profiling data
suitable to populate a comparative display of empirical results on the multi-dimensional profile-mapping
display (refer to Sections 3.6.2 & 4.4). In terms of articulating the sustainability-performance results of the
PV project on a graphical decision-support profile-display map (Experimental Step 7), the experiment can
overlay the statistical AHP results, as listed in Table 5.5, on the proposed sustainability profile-display map
of Figure 5.15.
The ce-WELF sustainability profile display in Figure 5.15 allows for the superimposed mapping of sus-
tainability profiles for both candidate projects (FPV=blue and GPV=red traces) on the sustainability profile
display as ce-WELF sustainability criteria priority vectors. From the AHP decision results logged in the
fourth and fifth columns of Table 5.5 respectively, the red and blue FPV/GPV sustainability-performance
traces overlaid on the visual sustainability indicator representations Figure 5.15 reflect the ce-WELF sus-
tainability priority vectors to establish integrated project sustainability profiles for the FPV and GPV systems
in Figure 5.15a&b respectively. The profile-mapping display schemes in Figure 5.15a and b respectively
overlay the performances for both the FPV (blue) and GPV (red) project options in the portfolio in terms of
the weighted and unweighted priority results for the portfolio of project options. In this respect, Figure 5.15a
presents the raw FPV/GPV performances as unweighted sustainability profiles for the two project options,
while Figure 5.15b presents the AHP goal-weighted sustainability profiles for the two project options. In
Figure 5.15b, the AHP decision algorithm is essentially modulating the project’s ce-WELF priority goals (Ta-
ble 4.2) on the ce-WELF decision criteria probabilities (Figure 5.13), to determine the sustainability priority
profile vector estimates (Table 5.5) for display on the radar-type sustainability profile mapping display. As
such, the sustainability profile probabilities for each of the candidate projects (FPV and GPV) are statistically
weighted in terms of the desired or user-defined project decision goals.
217
Land [0.0, 1.0]
Energy [0.49, 0.51]
Water
[0.21, 0.79]
EST [0.47, 0.53] CO2[0.49, 0.51]
Food
[0.0, 1.0]
Land [0.00, 0.09]
Energy [0.12, 0.12]
Water
[0.03, 0.11]
EST [0.08, 0.01] CO2[0.14, 0.15]
Food
[0.0, 0.07]
(a) (b)
Figure 5.15: Profile-mapping display portfolio for PV project sustainability as decision space, showing the profiles
for FPV and GPV candidate projects as relative sustainability indicator vector probabilities (Table 5.5) for the (a)
raw/unweighted probability profile and (b) goal-weighted probability profile, on a spider/radar-type decision-support
profile-mapping display (blue=FPV, red=GPV) (source: author).
This is done by assigning a mathematically meaningful sustainability score to each project as a ratio
scale measure. In this context, the experiment’s conclusions are presented in a discussion on using the raw
and scaled project-performance dimensions to render a re-ordered ranking in terms of the AHP weighting
in support of project decision-making comparisons, as depicted in the graphs of Figure 5.15a and b.
5.5.4. Conclusions and critical findings of the experimental case study
With the 3E-trilemma as the centrepiece of ce-WELF-based decision support, the theoretical framework
conceptualisation of Figure 3.14 highlights its value by simultaneously considering a sustainability rating
in all six ce-WELF-integrated dimensions. In this context, The results of this decision-supporting experi-
ment give a practical exhibition of the tactical efficacy of the sustainability decision framework and decision
methodology posited by this thesis. It enables the study to make operational level project assessment
conclusions, as presented in Section 5.5.4.1, while strategic level project assessment conclusions are dis-
cussed in Section 5.5.4.2.
5.5.4.1. Operational FPV project assessment conclusions
Given the scenario-driven techno-geographical context of the experimental system depicted in Figure 5.5,
the project-specific outcomes for the 3E-derived set of ce-WELF performance metrics in Figure 5.13, to-
gether with the sustainability priority profile vectors in the sustainability profile plots of Figure 5.15. The
sustainability profile probabilities in the dashboard plots in Figure 5.15 specifically serve as visually and
numerically-balanced decision scorecard indicators. While the FPV model records the ce-WELF decision
indicator data and its statistical AHP processing in a date-index format, the empirically reported values for
the ce-WELF data for three of the 20 project years is shown as examples in Table T.1 (Appendix T).
Both the empirical (Table 5.5) and visual (Figure 5.15) results thus offer a synopsis of the anticipated
performances and quantitative relationships between the ce-WELF sustainability sectors for the proposed
FPV project and its GPV alternative. Viewed from an observer perspective, the further the distance of
each metric-plot coordinate from the origin of the spider display on each axis, the greater the level of
218
sustainability for the project in that criterial dimension. In the same context, the closer each metric-plot
coordinate is to the axis origin, the greater the level of unsustainability contributing to that dimension of
the sustainable development project. By re-running the FPV analytical and decision-support model with
different FPV systems-design configuration settings, the user can numerically compare the outcomes for
the various projects on the decision-support profile spider map and the statistically-derived numerical AHP-
sustainability decision scorecard. As such, the sustainable-development spider diagram creates project-
decision options regarding trade-off decisions and balancing decisions in the context of sustainability.
From a sustainable development perspective, these spider diagram plots in Figure 5.15 essentially
provide a future perspective on the ce-WELF sustainability prospects of the two PV project options in
the decision portfolio. The sustainability profile mapping displayed in Figure 5.15 thus creates project-
decision options regarding sustainability trade-off decisions and sustainability balancing decisions. To en-
able project decision-makers not to be biased in terms of arbitrarily-set project goals, Figure 5.15a presents
the performance-profile results in terms of the PV system’s sustainability potential that is unscaled by the
pre-set AHP goals defined by the environmental impact practitioner/project developer. As such, the un-
weighted criterial priority results, as presented in Figure 5.15a, reflect the raw sustainability criteria ratings
for the portfolio of project options. The FPV/GPV sustainability profiles in Figure 5.15b represent a goal-
scaled version of the profiles in Figure 5.15a. This figure incorporates the pre-determined goal priorities of
the project (refer to Section 4.4.3), which goal weights are used to scale the project-performance dimen-
sions into a re-ordered ranking in terms of the AHP decision weighting. The project overlays in Figure 5.15b
reflect the goal-weighted priority results or criterial sustainability ratings for the portfolio of project options. In
the latter profile, the FPV/GPV project profiles are scaled according to the pre-defined criteria goals for sus-
tainability for the development of the PV project, as defined by the environmental impact practitioner/project
developer.
From a strategic decision-evaluation perspective, the predicted sustainability profile presented in Fig-
ure 5.15b demonstrates how the AHP technique can prioritise measurable goals by creating a criterial
profile-conditioning vector in the response space. The sustainability profile articulation means of the graphic
project in Figure 5.15b can, as such, help to solicit an opinion on the degree to which a specific project-
design configuration meets the sustainability goals of the project plan. Regarding the results for the sus-
tainability indices for the FPV and GPV projects in Figure 5.15, the analytical AHP-driven decision-profile
results for the project confirm that the FPV project variant outperforms its GPV counterpart in virtually all of
the sustainability dimensions. Moreover, while the performance profiles in Figure 5.15 paint a picture of the
comparative sustainability profiles for the FPV and GPV project variants, the performance margins in the
ce-WELF profile of the FPV power plant epitomise the energy sustainability in terms of the WELF resources.
The ce-WELF spider sustainability framework profile highlights the efficiency in the use of the water surface,
the farmland-sparing benefits, the economic transformation of the water surface, and the energy-water sur-
face transformation capabilities as superior sustainability qualities of the floating photovoltaic solar energy
installation.
In the sustainable development decision-support context, the AHP-driven decision-support component
further offers a vital result in automated decision-making in terms of the numerical decision-making support
principles, as presented in the fourth column of Table 5.5. As regards the AHP weighted numerical results,
depicted in the spider diagram of Figure 5.15b, the analytical AHP decision score puts the FPV project
score at 63%, while its GPV project counterpart is scored at 37%. With a 63% sustainability score, the
AHP results numerically confirm that the FPV project variant outperforms its GPV counterpart in terms of
the decision framework for sustainability offered by the ce-WELF composite sustainability index chosen for
this experiment. In terms of the raw ce-WELF data and probabilities for the FPV/GPV project performances
based purely on the data in Figure 5.13, as listed in the third column of Table 5.5, the associated sustain-
ability profile is depicted in the spider diagram of Figure 5.15a. These results show that, with the AHP
219
goal weighing, the FPV project scores a 63% sustainability rating. In the absence of AHP decision goal
weighting, the relative probability score for sustainability, namely 72%, for the FPV project is higher, while
its GPV counterpart scores only 28% in terms of sustainability. In terms of the FPV project sustainability
rating scores of 63% (AHP goal weighted) and 72% (raw ce-WELF data), the decision-supporting experi-
ment confirms visually (Figure 5.15) and numerically (Table 5.5) that the FPV project variant outperforms its
GPV counterpart in virtually all of the sustainability criteria dimensions.
The remainder of the statistical probabilities, as recorded in Table 5.5, reflects the critical strengths in
multi-criteria decision-making, namely the application of the user-defined AHP goal weighting procedure
and its ability to host a set of weighted statistical measures as a reflection of the sustainability qualities
of the FPV and GPV project options in terms of attaining to a numeric statistical sustainability score. As
a likelihood that a probability score of the respective project performances in a portfolio would meet the
required pre-defined project goals, the numerical decision value in the last column of Table 5.5 serves
as a type of goal-benefit plausibility score in terms of requirements for each of the PV projects under
consideration. In project-decision-support terms, the project’s sustainability score serves as an empirical
measure for deciding upon PV projects within a portfolio of options.
5.5.4.2. Strategic FPV project assessment conclusions
Overall, the climate-economic-WELF resource system indicators in Figure 5.13 well as the probabilistic PV
system sustainability profile plots in Figure 5.15 provide valuable insight into the value propositions offered
by floatovoltaic technology in terms of the photovoltaic panel efficiency gains, avoidance of land-energy
conflicts, improving water preservation qualities, favourable carbon sequestration potential, and reducing
the water-energy conflicts. As such, the floating solar decision-support analysis further stresses the water
surface transformation benefits whereby unused water space now generates revenue in terms of electricity
sales/savings, climate health or carbon credit trading, and the sale of raw or processed agricultural produce.
From an EIA perspective, the tactical value of the results of this experimental investigation is the quan-
titative and visual reporting on the effectiveness of floating solar PV-sustainability drivers and their capacity
to help achieve project sustainability. Operationally, the project-appraisal results reflect the ce-WELF sus-
tainability priority vectors and profiles in the decision space of Figure 5.15, enabling the user (the EIA
practitioner, energy developer) to visually and quantitatively appraise the sustainability attributes of the
floatovoltaic planning scenario in heuristic analytical decision space. In this context, the FPV-GIS model’s
empirical results in Figures 5.13 and 5.15 as well as the ce-WELF indicators in Table 5.5 provides duly
required scientific evidence to document the EIA-related impact effects for the proposed FPV project and
its GPV counterpart as given in Figure U.2.
Strategically, the results demonstrate the operations of the decision-support layer, offering a geo-sensitive
visualisation of the performance and resource impacts of a floatovoltaic ecosystem in the proposed ce-
WELF-driven visually analytical decision-support framework. With the experiment thus delivering an in-
tegrated set of multi-criteria sustainability-performance results in support of computational thinking, it can
comparatively evaluate energy projects concerning project ranking in design project portfolio selection or
the likelihood that any of the projects would meet the required sustainability goals. From the perspective of
supporting smart climate agriculture, the results portrayed in the visual sustainability indicator representa-
tions of Figure 5.13 and Figure 5.15 illustrate the value of the model’s decision-supporting type display and
computational intelligence, especially in terms of its ability to support the evaluation of sustainability trade-
offs and synergies, sustainability design optimisation, and analytical cost-benefit decision-making. From an
operational perspective, this means that the results would allow for project performance and impact com-
parisons in terms of "what-if” scenarios commonly required in project development and risk-management
exercises.
220
The user can further exploit the decision-making capacity of the model by re-running the FPV analytical
and decision-support model with different FPV systems-design configuration settings. The model can thus
numerically compare the outcomes for the various project configurations on the decision-support profile
spider map or through the statistically-derived numerical AHP-sustainability decision scorecard. The results
exhibit the quantitative data-driven decision-support capabilities of the FPV-GIS toolset in terms of visual
and numerical decision support. In the decision-support context of sustainable development, the AHP-
generated statistical probabilities of Table 5.5 reflect one of the key strengths in multi-criteria decision-
making. In project decision-support terms, the project-sustainability score is a valuable empirical measure
for deciding upon PV projects within a portfolio of options based on a bigger-picture likelihood or best-fit
project-decision score. In this decision-support context, the AHP-driven decision support offers significant
empirical decision-support results for automated decision-making in terms of the numerical decision-making
support principles.
Quite notably, the sustainability profiling results of this experiment serve as evidence of the extent to
which the 3E theoretical postulate and the ce-WELF sustainability framing of this thesis advance the state-
of-the-art in terms of PV performance and sustainability profiling. While Section 5.4.4 also highlighted this
conclusion, this experiment’s comparative GPV/FPV results give more profound insight into the deficiency
of the conventional open-loop framework to predict the sustainability attributions of an FPV system. In this
context, comparing the results for the GPV model predictions with that of the FPV model predictions in
Figure 5.13 and Figure 5.15, both illustrate the inherent imperfections that are encountered when predicting
the sustainability of an FPV project with a conventional GPV performance model. As such, the resource-
based ce-WELF sustainability profiles for FPV and GPV systems portrayed in Figure 5.13 and Figure 5.15
strategically illustrate the value of the integrated 3E assessment framework of this thesis (Figure 3.14) as a
theoretical basis for overcoming the imperfections of a GPV model when assessing the distinctive and more
diverse (resource-type) sustainability profile of an FPV project.
5.6. Summary
With the advanced analytical capability to empirically assess the performance and impact profile of floato-
voltaic technology, this chapter describes practical case-study experimentation with the proposed theoretical
research framework, analytical model, and project assessment methodology. When driven with the relevant
meteorological data, the FPV-GIS model’s 3E and ce-WELF data forecasts are universally applicable and
can be used anywhere in South Africa. As such, the FPV-GIS model automatically adapts to the envi-
ronmental conditions associated with the location-specific meteorological weather station data that drives
the model at runtime. In the context of the thesis experiments, the meteorological data was measured by
weather stations around the experimental installation location in the Suid Agter-Paarl region of South Africa.
The FPV-GIS model and simulation data by configuration thus fit perfectly with the environmental and other
conditions highlighted for a proposed agricultural FPV system installation in South Africa.
In this context, the chapter conduct experiments with the implemented FPV-GIS toolset in a typical use-
case scenario to determine the extent to which the geo-informatics-based FPV-GIS toolset can contribute
to an assessment of the respective energy, environmental and economic performances of a particular float-
ing solar configuration in an agricultural area in South Africa. It presents quantitative experimental results
related to case-based simulation-based project enactments based on the experimental modelling consider-
ations detailed in Chapter 4, and within the scope of the research assumptions, as presented in Chapter 1.
The experiments demonstrate the processing capabilities of the geomatics tool to affect long-term operat-
ing life-cycle predictions on the impact of such systems installations. With the three tactical experiments
defined in Figure 5.1 concluded, the results can help to answer the research questions in Section 6.3.2 of
221
the next chapter.
The results demonstrate the extent to which the predictions of the outputs of the FPV-GIS toolset can
be used to evaluate the degree to which a specific pre-configured floating-PV design configuration can
meet the desired energy, environmental and economic performances in respect of the qualification criteria
for due diligence and environmental sustainability. The experiments further provide evidence-based em-
pirical results whereby the newly proposed systems thinking method and dynamic simulation model of this
research can be applied to articulate the performance profiles of floating PV systems in the analysis of a
case-based scenario. While the experiments acknowledge that many of the performance and offset con-
tributions of floating solar FPV units apply to land-based GPV units, they also consider the output benefits
of a counterpart GPV system in a hybridised-type FPV and GPV systemic configuration. These results
demonstrate particular features of the FPV-GIS toolset, which allows it to empirically compare the quantita-
tive performance and impact-effect results between associative-type FPV and GPV systems configured for
the comparative evaluation of performances at an envisaged installation-site setting.
Overall, these experiments succeeded in evaluating the extent to which the proposed theoretical frame-
work and the framework’s computer modelexpression integrated onto a geo-informatics decision-supporting
platform has succeeded in establishing a new analytical methodology and analytical data science technique
for predicting the outcomes of an agricultural floating PV project installation when compared to that of a
ground-mounted PV project. In this context, the research conclusions articulate an opinion on the value
of the experimental evaluation of the sustainability assessment framework and model, testing the extent to
which the Research Aim and Objectives 1, 2, and 3 were met. The experimental synthesis phase guides
the research project through evaluation procedures to ascertain whether empirical evidence in floatovoltaic
planning exercises either supports or refutes the postulated tentative hypothesis around the theoretical
and framework conceptualisations presented herein. The experimental results and conclusions enable the
thesis to respond to the research aim and objectives and answer the research questions in the next chapter.
222
6. Summary, Conclusions and Novel Contributions
6.1. Topical Research Overview
Topically, the geographical research of this thesis is framed in a broad paradigm of sustainable development
(Lee et al., 2022; Maka, 2022). Through this discourse, the research investigation develops new theoretical
foundations to support the sustainable energy development theme while making quantitative assessment
contributions to substantive areas of geographical information, energy, environmental and economic data
sciences. The study solves an intricate problem around floating PV project valuations by developing a geo-
informatics approach to the sustainability assessment of floatovoltaic technology in South African agricul-
tural applications. The research topic and theme are highly relevant, given the context of the global energy
transformation, the ongoing need for sustainable energy production, the recent discovery of floatovoltaic
renewable energy generation technology, and the requirement for improvement in sustainable development
practice and computer-aided EIA praxis.
Within this topical context, this thesis’ scientific, geographical engagement establishes niche interdisci-
plinary research by engaging in novel theory-building research to solve a current real-world geographical
problem around agricultural energy system sustainability classifications. Contextually, the focus is on the
circumscribed area of PV technologies, wherein the investigation and solutions of this thesis address the
real-life issues of how to assess and model the more complex and diverse impact effects hierarchies ema-
nating from the recent discovery of floating PV technologies (Allen and Prinsloo, 2018; Essak and Ghosh,
2022). Faced with the looming environmental degradation, water stress, freshwater shortages, and agri-
cultural land scarcity for new energy developments (Choi et al., 2020; Lovering et al., 2021), farmers and
especially agronomists value the development of small water footprint and low carbon fingerprint energy
systems to limit other adverse impacts on farmland, irrigation water and land-based resources (Armstrong
et al., 2020; Havrysh et al., 2022). As such, this technology’s data-driven energy modelling and charac-
terisation expose critical theoretical knowledge and methodological gaps in the fields of environmental-,
information-, economic-, and engineering- sciences (Armstrong et al., 2020; Cagle et al., 2020). These
gaps are causing deeply problematic challenges worldwide, failing to meet ever-stricter regulatory require-
ments for approving and licensing such PV system installations amidst global climate change imperatives
(Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). These knowledge and methodology gaps call
for objective academic and knowledge contributions to fill critical knowledge gaps through geographical re-
search in applied systems analysis and design-scoping methodologies (Exley, 2022; Pouran et al., 2022).
The challenge is of sufficient complexity and motivation to stimulate dialogue toward new geographically-
motivated theoretical conceptualisations and simulation solutions to solar energy system assessment for
sustainability. Furthermore, current barriers and problems in the development and installation of floating
solar energy systems create challenges and opportunities for geographical sustainability analysis through
data-driven analytics in an informative inference tooling solution conceptualised and built around improved
geographical intelligence.
The scope of the problem is particularly challenging in the modern agricultural industry (Allen and
223
Prinsloo, 2018; Prinsloo and Lombard, 2015b), whereby extended knowledge gap challenges warrant new
research and academic contributions to deal with the complexity and diversity of the sustainability criteria
defined in a climate-smart agricultural context (refer to Figure A.1). Since agricultural sustainability goals
govern the conditions for new energy installations on farms as part of the green energy production move-
ment (Cagle et al., 2020; WWF, 2021), this thesis argues that floating PV sustainability should instead
exploit the science of the total environment in the search for a new sustainability assessment paradigm
(Alaoui et al., 2022; Luca et al., 2015; Santiago-Brown et al., 2015; Simpson et al., 2022). Within an agri-
cultural context, the complexity and diversity of the impact assessments for floatovoltaics escalate, and the
sustainability assessment becomes more problematic because the installation of newly planned PV systems
permanently floats on life- and food-sustaining water reservoirs (to impact positively on farmland) (Hoffacker
et al., 2017; Vema et al., 2022). Furthermore, since floatovoltaic technologies directly interact with water
reservoirs and water resources are managed as a national resource in most countries, the technology faces
a lingering threat of new international licensing regulations anticipated on the regulatory horizon for future
floatovoltaic technology installations (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021). This threat
arises because water is a sensitive and critical resource in global food security, and as a scarce resource,
it is under constant national and international climate change monitoring.
In the interest of environmental justice, stakeholders such as government administrators and impact
practitioners worldwide are urgently seeking new scientific solutions and geographical tooling models to
overcome the dilemma of determining the sustainability status of newly discovered FPV systems (Chand
et al., 2022; Gadzanku et al., 2021a; Mathis and Huber, 2018). Equally important, sustainable energy
project developers and investors are looking to improve bankability through improved solar energy impact
modelling towards scientifically measuring or predicting the impact effects of these technologies on environ-
mental resources and the economic feasibility (Armstrong et al., 2020; Weigl, 2021; IFC, 2015). While plac-
ing inadequate emphasis on integrating environmental impacts into the valuation of the performance profiles
of prospective photovoltaic projects, Section 3.5 presented theoretical level challenges around overcoming
an inherent problem of conventional PV performance assessment in a digitally enabled environment. More-
over, formulating decarbonising energy systems has led to the development of integrated impact assess-
ment scenarios where traditional modelling approaches strongly focus on hydro-carbon emission impact.
However, when assessing the "sustainability” of energy systems, one must consider more aspects than
climate change risks and decarbonisation to address the complexity in the transformation to sustainability
(Almeida et al., 2022b; Hottenroth et al., 2022; Naegler et al., 2021). Therefore, multi-attribute diagnostic
and forecast models and decision-making methods should facilitate a broader conceptualisation framework
for sustainability assessments (Al-Subhi et al., 2020; Buchmayr et al., 2021). Furthermore, the notion of floa-
tovoltaic installation sustainability is a core real-world geographical problem poorly understood at present
(Gadzanku et al., 2021b; Kumar et al., 2022), and strategically confronting scientists with the challenge of
cultivating new ideas toward defining a fit-for-purpose sustainability framing for floating PV technology.
Before the sustainability assessment problem can be addressed, scientists must overcome a more
fundamental problem, namely, quantifying the peculiar performance phenomena of FPV technologies, com-
monly known among scientists as "technology unknowns” of floatovoltaics (Banik and Sengupta, 2021; Ca-
gle et al., 2020; Gadzanku et al., 2021a). PV tooling and modelling deficiencies are currently experienced
because the siloed execution processing philosophy followed by conventional modelling frameworks is inca-
pable of dealing with the complex systemic interaction of the FPV system with the underlying aquatic habitat.
The problem persists because floating PV technology causes a diversified range of obscure performance
and impact unknowns, seemingly caused by incremental inconsistencies that appear to perpetuate over the
project lifetime. These challenges are considered the main deployment barriers to new energy projects as
they jeopardise regulatory project approval authorisations (Gadzanku et al., 2021a; Gorjian et al., 2022c).
The operational modelling and empirical characterisation of floatovoltaic technologies present a core real-
224
world geographical problem understood poorly at present. Being symptomatic of reductionist thinking in
existing photovoltaic modelling techniques, the inability of conventional models to decrypt these "technology
unknowns” is also an inherent problem of conventional PV performance assessment theories. The funda-
mental disjunctures between the diverse sustainability performances and impacts of GPV and FPV systems
must enjoy priority as they cause serious sustainability assessment challenges. These systems concern
the correlative conjunctions and cascading nature of the "technology unknowns” of newly discovered floato-
voltaic technologies, which increase the complexity of floating PV performance modelling. The technology
model’s layered and interconnected analytical complexity represents a significant research problem area, to
the extent that it warrants new research within the parameters of physical geography and the geographical
discipline of environmental science. This research topic has caught the attention of major universities and
large energy research organisations worldwide (Friel et al., 2019; NREL, 2019a; Tsuji, 2021).
In working towards the creation of new knowledge and tradecraft (frameworks, models or indicator
metrics) that can help to understand and predict floating PV installation sustainability at the project design
stage, this thesis argues in favour of solving this trans-disciplinary geographical problem in a philosophic-
theoretic level through a theory-building facilitation methodology. On a strategic philosophical level, the
thesis posed the argument, based on the consideration of Figure 3.3), that a strategic focus on solving the
sustainability modelling problem could inherently encompass solutions for operational behaviour synthesis,
performance analysis, and impact assessment in a single consolidated sustainability synthesis.
An inductive research methodology inspired the idea to draft an improved theoretical framework based
on the emerging knowledge development disciplines of the smart systems thinking principles of the 4IR
paradigm. Posting a new framework to measure/predict the sustainability efficiency, effectiveness and re-
liability of FPV power systems in an ecosystem context helped to improve the predictability of floating PV
sustainability from a theoretical perspective. The proposed 3E theoretical construct and framework de-
vice offers a new program logic in a computer synthesis model for synthesising FPV behaviour in a digital
twin toolset. The proposed geospatial digital twin enables the experimental characterisation of floating
photovoltaic power plants while profiling the broad-spectrum sustainability attributions of the technology in
real-time and lifetime project analysis assessments. Therefore, with the help of geographical and systemic
principles, this scientific study could think strategically and engage geospatial technology principles in the
cognition of the problem and the drafting of a new theoretical framework and toolset to challenge the state-
of-the-art conventional approaches to help solve complex analytical challenges such as location sensitive
floating PV sustainability.
Solving the FPV sustainability assessment problems in the sustainable development context is of sig-
nificance because deep-seated theoretical imperfections with conventional assessment frameworks and
tooling are proving to be among some of the greatest project-planning hindrances and regulatory obstacles
to development investors, government institutions, environmentalists, and impact practitioners worldwide
(Cuce et al., 2022; Hernandez et al., 2019; WRC, 2016). An enabling policy environment around support
for adopting and deploying floatovoltaics has already created scope for this nascent technology in the in-
ternational policy landscape (Cohen and Hogan, 2018; Gadzanku et al., 2022b). However, most regulatory
policy considerations hinge on the requirement for the scientific capability to quantify the sustainability of the
technology in real-world settings duly. In this policy context, the International Energy Agency further sees
opportunities to support the transition to sustainable PV energy systems by enhancing digital performance
forecasting quality in computer-aided assessments (IEA, 2017). It argues that reliable photovoltaic perfor-
mance prediction models can help lower the barriers and costs of technology diffusion as more accurate
performance forecasting reduces PV system planning and investment costs.
By solving this problem through a geographical operations research approach, this thesis breaks new
ground by drafting a new theoretical framework to define sustainability in floatovoltaics while concomitantly
helping to unravel the intricate mysteries of the complex and diversified set of uncertainties that constitute
225
the so-called "technology unknowns” to floatovoltaic technologies. It provides new foundational theoretical
insights into the generic sustainability drivers of FPV technology, which uniquely helps to provide a core un-
derstanding of the quantitative impact effects of the technology on the environment and natural resources.
With the theoretical framework as the underpinning logic in a novel computer model expression solution to
assess floatovoltaic projects, the academic contribution advances the state-of-the-art as it provides a more
advanced and sophisticated quantitative ecosystem modelling regime to support an improved understand-
ing of the impact effects of the technology.
Considering the substantial transformative potential of 4IR and cellular mobile connectivity to support e-
Agriculture and the knowledge economy in Africa (4IRSA, 2020; FAO, 2022), the geoinformatics orientation
of the geographical software implementation of the theory in analytical and decision-support models can
support sustainable development on multiple levels of society. Strategically, this research approach runs in
tandem with our national government directives regarding access to information through the E-GIS platform
(DFFE, 2022, 2019b; EGIS, 2022). This government platform is driving the digital migration of information
sharing and EIA data transparency to an electronic platform level (refer to Figure U.1). In this digitisation
context, geographical problem-solving creates an opportunity to provide experimental geospatial technolo-
gies to develop next-generation digital twin modelling solutions in purpose-built geographical information
system toolsets (ESRI, 2022; Maplesoft, 2022a).
Based on the observations and conclusions inferred from comprehensive coverage of the topic by the
relevant scientific scholarship, a systematic and critical assessment of the literature helped to make novel
contributions with the establishment of new philosophical ideas to advance the state-of-the-art through im-
proved geo-analytical data science theories. With this posited sustainability theory for floating PV serving
as the underpinning theory for the proposed further analytical contributions described herein (analytical
framework, procedures, methodologies, models, and techniques), the significance of this thesis’s contribu-
tions especially relates to the tactical and strategic value propositions. The theory supports sustainability
policy modelling and the development of new regulatory standards or impact assessment procedure proto-
cols for floating PV technologies. The approach thus serves as computer logic in a combined descriptive,
predictive and prescriptive model that improves location-aware performance and sustainability profiling in
floating PV technology project examinations. The significance of this thesis’ contributions is that it offers an
automated diagnosis system that models system behaviour in the cyber world of digital experiences. These
contributions offer future opportunities to create enhanced policy, improved practice, and novel academic
research around the sustainability status of PV energy technologies towards the potential regularisation of
agricultural floating photovoltaic energy system installations.
Supporting the sustainable development discourse towards sustainability assessments through process-
based modelling synthesis, the postulates of this thesis provide unique value propositions by establishing
a more holistic and comprehensive methodology for the appraisal of PV projects. As such, it delivers a
theoretically driven method, assessment methodology and data-science technique by uniquely advancing
predictive PV modelling capabilities to ramp the analytical capabilities up the hierarchy from the conven-
tional performance assessments, to more advanced geodynamical integrated evaluations, to even more
advanced sustainability assessments required to support the sustainable development complex (refer to
the hierarchical abstraction depicted by Figure 3.3).
6.2. Research Summary and Synthesis of Critical Findings
In a sustainable development practice context, the voice of reason in project planning toward smarter renew-
able energy portfolios calls for developing relevant geospatial tool models that can account for hierarchical
impacts in scientific sustainability assessments for newly-planned renewable energy project developments
226
(Hernandez et al., 2019; Luderer et al., 2019). While sustainability assessment offers a broader perspective
on environmental impact (Hottenroth et al., 2022; UN, 2022), sustainability assessment is a more com-
plex appraisal method that requires a systemic framework for sustainability appraisals and decision-making
(Binder et al., 2010; Sala et al., 2015; Villeneuve et al., 2017). This challenge is true in principle for pho-
tovoltaic energy production in agricultural settings (Binder et al., 2010; de Olde, 2017), where floatovoltaic
power-generation technologies interact more intensely with the natural aquatic environmental system to
provide a broader and more diverse range of value-laden benefits and co-benefits (refer to Section 2.3.2.
As detailed in this thesis, the discovery of floating PV technology brought a variation of the photovoltaic
theme, wherein floatovoltaics represents an alternative energy technology solution defined by a new type
of solar PV technology installation in which solar PV panels are sited directly on open water surfaces (Fig-
ure 1.2) (Allen and Prinsloo, 2018; Spencer et al., 2019). The regulatory environment (Figure 2.8) however,
experiences serious sustainability assessment challenges with the cascaded nature of the disjunctures be-
tween the diverse sustainability performances and impacts of GPV and FPV systems (refer to Figure 2.7a).
Regulators are therefore calling for a scientific resolution to the problem of quantifying the "technology
unknowns” of floatovoltaic technologies (Banik and Sengupta, 2021; Gadzanku et al., 2021b). From a
technologically-driven sustainable energy development context and EIA viewpoint (refer to Figure 2.7b), the
geographical investigation of this thesis succeeded in conceptualising and actualising a geoinformatics ap-
proach to design-stage sustainability assessments of floatovoltaic technology in South African agricultural
applications.
Towards solving this real-world geographical problem, Chapter 1 narrates the academic merits of the
research project. It sets the research agenda and delineates the geographical research process to re-
solve the identified problem and the knowledge gaps around floating PV project assessments. Putting the
spotlight on the scope for new innovation to solve the problem on a theoretical level, Chapter 2 offers an
extensive literature review on the real-world challenges associated with broad-spectrum performance and
impact assessments for floating PV projects, while systematically reviewing the state-of-the-art philosophy
around such systems, by exploring the scientific literature focused on investigations into floating PV sys-
tems. Section 1.1 delineates the topic of the study and the research process directed at resolving the
identified real-world geographical problem around the theoretical characterisation of floatovoltaic technol-
ogy. Section 1.2.1 defines the research problem statement from the current critical knowledge gaps and
methodological gaps identified in the prevailing state-of-the-art floatovoltaic philosophy and best practices
around the modelling of floatovoltaic systems in terms of sustainability performances, as described in the
research literature. The discussion of the purpose of the study in Section 1.2.3 sets the goal of challeng-
ing the state-of-the-art philosophy in the field of the geographical sciences by contemplating an improved
theoretical solution to the problem from a systems-thinking narrative viewpoint.
Setting the research agenda in a quantitative research-design process, Section 1.3 defined the respec-
tive steps of identifying the research problem, defining the problem statement and formulating the purpose
of the study. The research-design framework (Figure 1.6) and research procedures (Figure 1.7) jointly aim
at defining a quantitative research methodology in an inductive methodological research framework (Fig-
ure 3.8), through which an inductive reasoning process to strategically improve the prevailing wisdom. It
navigates the scientific investigation towards creating new knowledge that can fill critical knowledge and
research gaps around theoretically-based floatovoltaic performance and sustainability assessments. This
methodology uniquely supported the exploration of the scientific literature (refer to Chapter 2) in terms of
engaging analytical, behavioural geography principles in a systems-based approach to make crucial obser-
vations that allowed the identification of critical knowledge gaps from a Geographical Information and Data
Science perspective.
To fill these knowledge gaps, the thesis conceptualised a prospective theoretical solution as the study
hypothesis. The proposed theoretical framework leverages the concrete agglomeration benefits of the
227
FPV’s techno-environmental-economic ecosystem in a holistic 3E domain ensemble. Such a corpora-
tion of heterogeneous operational domains in an ecosystemic context advances a new philosophical idea
on the profiling of the sustainability of floatovoltaics through cross-disciplinary ecosystem domain connec-
tion modelling. To this end, Section 1.3.3 drafted a tentative research hypothesis and appropriate research
questions to fill existing knowledge gaps and to address limitations observed in the operational response
patterns in current floating PV assessment models. The research hypothesis and questions as such reflect
the author’s ideas around the conceptualisation of a more holistic sustainability assessment strategy and
reference framework as underpinning theoretical basis and analytical methodology to improve simulation-
type floatovoltaic project valuations in terms of the research aim and objectives, as defined in Section 1.3.4.
The significance of the research, detailed in Section 1.4, highlights the urgency and the value of the re-
search within the context of South Africa’s National Climate Change Adaptation Strategy (DFFE, 2019a)
and the global sustainable development goals dictated by the current transformation in global energy (UN,
2020a). In its turn, Section 1.5 deals with the definition of research terms and the research assumptions
adopted by the thesis.
The geographical research study adopts a quantitative research methodology. Chapter 3 describes the
philosophical and methodological foundations for postulating a new holistic evaluation strategy and theoret-
ical framework arrangement for sustainability to improve the formalised assessment of floatovoltaic systems
sustainability under Research Objective 1. While Section 3.2 summarises the goal of the study in terms of
the broad philosophical principles of systems thinking, Section 3.3 details the methodological approach to-
wards the formulation of the methodological framework and research procedures, as defined in Section 3.4.
The choice in Section 3.5 is for an inductive research process as a methodological framework, whereby
the chronological research activities provide the essential milestone steps (as shown in Figure 3.8). Both
the research Aim and Research Objective 1 advance the notion of a holistic quantitative integrity-focused
evaluation strategy as a sustainability portfolio theory in a novel systems-based sustainability assessment
framework for evaluating FPV installation project proposals. While the postulated theoretical construct for
the Integrated Sustainability Assessment Framework (ISA-framework), graphically represented by the sys-
temic mind model abstractions in Figures 3.14 and E.1, the proposed ISA-framework for FPV stands central
to the thesis’ quantitative theoretical characterisation of floating PV ecosystem sustainability. Owing to the
inductive reasoning processes of this chapter, the archetypal quantitative framework for modelling sustain-
ability, as depicted in Figure 3.14, is purposively designed around a consolidated alliance between the
heterogeneous system’s components (refer to Figure 4.1), in which the operational trilogic energy, environ-
mental and economic domain objects (cohesive 3E ensemble, 3E trilogy), together with their networking
interactions, constitute a proposed composite floatovoltaic assessment ecosystem.
Towards the implementation of the theoretical framework in a quantitative research methodology in the
computerised model, Chapter 4 deals with the key modelling imperatives in appropriating the conceptu-
alised theoretical framework as an underpinning of blueprint logic in the development of a user-configurable
computer synthesis model. As a programmable computer model expression of the underpinning theoreti-
cal framework, sustainability theory thus forms the foundation to realise and actualise the characterisation
of floatovoltaic sustainability and decision-support models according to Research Objectives 2 and 3, re-
spectively. To meet Research Objective 2, Chapter 4 describes the research methodology in terms of
parametric modelling and the implementation of the experimental FPV-type Computer Analytical Simulation
model (CAS-model). To help realise this research aim, the thesis adopts the latest design-thinking, service-
thinking and theoretical-modelling principles to deal with the implementation of the research methodology.
Section 4.2 defines the unique systems-level implementation of the user-configurable FPV research model,
concurrently proposing a systemic intervention based on the application of theoretical modelling techniques
and system dynamics thinking in the object-process methodology (refer to Figure H.1). Section 4.3 sub-
sequently defines the component-level implementation of the research model objects as a multivariate cal-
228
culus of function theory. This process engages first theoretical geographic- and physics- principles in a
model-based design approach to define the operational properties of the FPV systemic components and
interactions. In this way, the newly postulated ISA modelling framework construct (shown in Figure 3.14)
serves as a consolidated computer-programmed blueprint logic in the newly postulated system-dynamics
thinking-based simulation model expression for the FPV digital twin model (shown in Figure 4.3).
Chapter 4 further progresses toward implementing the heuristic decision-support aspects of the re-
search methodology according to Research Objective 3. Towards the implementation of a proposed geospa-
tial decision support system in Section 4.4, the thesis implements a hierarchical, statistically-based, ana-
lytical process (AHP) in the decision-level simulation layering of the FPV assessment toolset (refer to Fig-
ure 4.12). As described in Section 4.4.2, this user-configurable AHP-based decision model is founded on a
proposed resource-based climate-economic WELF-nexus-driven index as a decision-supporting framework
for project sustainability assessments in FPV projections. Also, as an index-based sustainability assess-
ment framework for decision support in the planning stage of the floating PV portfolio and based on the
goal-oriented AHP decision-support principles of the project, as presented in Section 4.4.3, the climate-
economic WELF index is mathematically translated from the 3E data of the FPV-GIS model and integrated
into the decision model. While the integrated FPV synthesised model characterises the dynamic behaviour
of floatovoltaic systems in the 3E trilogy and the climate-economic WELF-sustainability pillars, the decision-
support framework and model serve the purpose of supporting project decision-making in appraisals of a
future planned floating solar project.
In Section 4.5, the developments above (analytical and decision models) establish the basis for the
geoinformatics-type systems-level integration of the FPV systems-model elements into a digital twin-type
model-driven FPV-GIS simulation model. With the FPV-GIS toolset as a research instrument, as depicted by
Figure M.1, Section 4.6 sets out the experimental design toward analysis-by-synthesis type data collection
and toolset evaluation, while Section 4.7 details the ethical considerations for the experiments. Finally, a
representative case study review in Chapter 5 engages the implemented custom-designed FPV-GIS toolset
to appraise a proposed real-world FPV project in a computer-based analysis-by-synthesis fashion. These
simulation experiments use the FPV-GIS model to predict the behaviour of a proposed floating PV systems
variant in terms of 3E metric data through simulation analysis, while translating the 3E results to ce-WELF
nexus metrics to profile PV project sustainability in an agricultural case-based wine-farming scenario exper-
iment. In each of the experiments, the results are examined to provide reasoned conclusions.
Essential research findings demonstrate using the new framework and modelling concept in a real-
world case-study experiment. The experiments engage the geospatially-aware FPV-GIS toolset in three
digitalised desktop experiments conducted in the PV installation narrative for a hypothetically-planned float-
ing PV system on a wine farm on the Paarl arterial. With the geo-informatics toolset implemented as
the designated quantitative research instrument, the user-configurable FPV-GIS toolset in Figure M.1 is
applied in the design-stage evaluation of the FPV project outcomes to support future sustainable develop-
ment scenarios. While the FPV-GIS toolset offers a simplified digitalised parallel to installing a real-world
FPV system, the composite geo-spatial floatovoltaics twin model toolset in Figure M.1 effectively estab-
lishes a holistic predictive analytical tool in a dynamic desktop research instrument. Since the performance
model predictions for the geospatial FPV-GIS project model are based on geo-sensitive solar-irradiation and
dynamic weather-data inputs, this instrument’s appropriation enables the user to enact scientific analysis-
by-synthesis experimentation for any pre-configured FPV project. As such, the design-stage performance
and impact calculations for all three E domains are typically performed in parallel, synchronising with the
meteorological data at an hourly time interval on a sample-by-sample basis.
The experimental evaluations in Chapter 5 follow the quantitative methodological guidelines for research
experimentation defined in Section 5.2. It offers a virtual reality experience by engaging the custom-
designed FPV-GIS toolset as a sensor-driven geospatial digital twin to appraise a proposed real-world FPV
229
project in a computer-based analysis-by-synthesis fashion. In the first experiment, the Floating PV Model
Validation Experiment in Section 5.3 engages in a corroborative PV model-verification process for standard
test conditions (STC) and PV test conditions (PTC), as developed by Sandia (Sandia, 2022c), to evaluate
the technical quantification aspects of the accuracy of the energy model in terms of laboratory-based and
referenced datasheet values. As such, the energy simulation test results in Table 5.4 reflect the datasheet-
referenced accuracy of the PV performance model for a selected PV module. In the GPV reference model,
the test results show that the simulated FPV-GIS model achieves energy prediction accuracies of 99.85%
and 99.43% under STC and PTC test conditions, respectively. The predicted energy performance results
in Table 5.4 thus correlate well with the datasheet energy-performance results under both STC and PTC
test conditions (maximum error deviation of 0.15 % and 0.57 %, respectively). As such, the error tolerance
under STC conditions (<1.5%) falls within a realistic error boundary, and the deviation margin is presumed
sufficient to confirm the corroborative verification of the PV energy model.
The second simulation experiment in Chapter 5 concerns the experimental characterisation of a floating
PV power plant in a real-world agricultural setting. It follows the guidelines in Section 5.2 to perform a
feasibility study of a floatovoltaic system at a given geographical location. The proposed FPV-GIS model
is used in its analytical modality to evaluate a realistic real-world scenario-driven case study by predicting
the broad-spectrum performance and impact behaviour of a proposed floating PV systems variant through
simulation analysis. The map and aerial image in Figure 5.5 depict the virtual project terrain for the proposed
real-world FPV installation, where the FPV Project Performance Assessment Experiment in Section 5.4
applies the FPV-GIS toolset to an actual real-world design-stage case study at the RWE wine-farm. With
the FPV-GIS model specification settings for the FPV systems variant, as listed in Table O.1, the FPV
modelling experiment is configured to determine and compare the geographical 3E-trifecta performance
and impact effects for a proposed FPV systems variant with that of its equivalent GPV systems counterpart
at the given geo-location. Figures 5.6 to 5.10 shows this experiment’s synthesised analytical 3E-framework
results. Unlike the equivalent GPV counterpart, the FPV results portray the water-based PV installation as a
synergistic system with an increased energy-output profile and an improved carbon-footprint impact profile.
The study primarily ascribes the superior performance qualities of the FPV system over its equivalent GPV
counterpart to the hydro-climatic cooling effects associated with the floating PV micro-habitat. The FPV
system variant delivers a range of additional natural resource conservation and WELF-nexus benefits, owing
to the land-surface preservation and water-evaporation reduction benefits. Based on the conceptual data-
translation principles, the digital FPV-GIS model can uniquely predict and quantify the water-energy-land
resource benefits of a proposed FPV installation variant (WELF-nexus benefits). As depicted in Figures 3.12
and E.1, the Energy and Environmental domain intersection facilitates the WELF nexus translation as part
of the unique 3E framework output projections.
In the third experiment, the study evaluates the decision-support capabilities of the FPV-GIS toolset in
the decision-supporting modality during the design stage of the FPV project. The Floating Solar Decision-
support Analysis Experiment in Section 5.5 digitally translates the simulated 3E-trifecta data of the previous
experiment into a climate-economic-WELF indicator index. It applies the AHP technique in the proposed
digitalised analytical-hierarchy model, as depicted in Figure 4.13, to map the ce-WELF sustainability indica-
tor index on a decision-supporting profile-mapping display. By digitally translating the assessment metrics
of the 3E-tridentate framework into WELF-nexus parameters (refer to the concept in Figures 3.15 and
E.1), the experiment profiles the sustainability qualities of the floating photovoltaic project according to the
enviro-econo-WELF sustainability portfolio framework. The graphical depiction of the synthesised decision-
supporting results in Figure 5.13 and Figure 5.15 thus act as a barometer for FPV and GPV sustainability
performance while driving visual and empirical decision support on a comprehensive dashboard display. In
this display, the user-defined AHP criteria weights for decision support, together with the experimentally-
determined decision scorecard in Table 5.5, offer an improved goal-driven numerical synopsis of the an-
230
ticipated performances and quantitative relationships between the ce-WELF sustainability sectors for the
proposed FPV project and the alternative GPV options. The raw ce-WELF sustainability indicators, as fea-
tured in Figure 5.13 and Figure 5.15a, define the respective sustainability profiles of the FPV and GPV
projects as raw ce-WELF indicator data. At the same time, the AHP goal-weighted in Figure 5.15b shows
the processed ce-WELF indicator data. The latter helps the user to graphically and numerically compare the
systems, or to determine the extent to which the proposed FPV systems variant and its GPV systems coun-
terpart can practically meet specific environmental sustainability and impact criteria as qualifying criteria for
the installation of future planned PV systems.
Overall, this thesis’s philosophical theory-building and design research advance the state-of-the-art phi-
losophy in floating photovoltaic performance modelling by breaking new ground in postulating a holistic
sustainability strategy and theoretical framework as fundamental imperatives in floating PV performance
modelling. The thesis further advances new theoretical foundations in an improved systems-based tech-
nique and theory-driven methodology to study the geo-dynamic behaviour of floating PV installations in a
digital computer-synthesised modelling and prototype environment. The following section deals with the
fundamental research conclusions. It presents the responses to the research aim and objectives and the
strategic research conclusions from the experimental findings summarised in the research results above.
6.3. Fundamental Research Conclusions
This section details the study conclusions regarding the research aim and objectives while responding to the
research questions. While deriving the research aim and objectives from the research hypothesis and ques-
tions, they help to conceptualise and define a computer model for finding analytical and decision-planning
support in an FPV geomodel. They all work harmoniously to propose a more 3E-system sustainability
valuation methodology underpinned by a conceptually integrated sustainability assessment framework for
floatovoltaic technology proposed by this thesis. The research aim and objectives progressively focus on
implementing and applying a holistic computer modelling and simulation methodology capable of realising a
cause-effect analysis in a whole-systems-based operational floating PV ecosystem to turn FPV project per-
formance prediction results into actionable insights. To this end, Section 6.3.1 revisits the research aim and
objectives individually. Section 6.3.2 responds to the research questions and answers, each within the con-
text of the literature review, the research methodology and the experimental research results. Section 6.3.3
reflects on proving the research hypothesis.
6.3.1. Revisiting the Research Aim and Objectives
This section revisits the research aim and objectives of the study and responds to the successful imple-
mentation of Research Objectives 1 to 3 towards achieving the Research Aim. The discussion reflects
on the measure of success to which the conceptualised theoretical framework and the performance of the
analytical and decision-support layers of the FPV-GIS model can establish an integrated geoinformatics in-
formation systems platform and research instrument suitable for analytical sustainability assessments and
decision support in an environment geared to the implementation of a floatovoltaic technology.
The research aim and objectives are presented in Section 1.3.4 to help conceptualise and define new
theoretical sustainability criteria in a theoretical systems framework for modelling and measuring sustain-
ability in floating PV project assessments. The defining criteria of the sustainability theory serve as the
underpinning theoretical base and computer programme logic in an analytical computer-simulated mod-
elling methodology suited for floatovoltaic technology. Implementing the proposed new sustainability theory
in a computerised model for finding analytical and decision-planning support in an FPV geomodel, they
231
promote sustainability by conceptualising and appropriating a potentially more holistic and WELF resource-
friendly evaluation framework. As such, the research objectives progressively focus on conceptualising,
implementing and applying a holistic computer-modelled and simulation methodology capable of realising
a cause-effect analysis of a holistic systems-based operational floating PV ecosystem. To help guide the
thesis to meet the respective research aim and objectives, Figure F.1 in Appendix F presents a simplified
flowchart to summarise the research actions and activities concerned with the systematic development of
an underpinning theoretical framework, a computer-modelling method, and a simulation model, through
the staged implementation of each of the relevant research objectives towards achieving the research aim.
Guided by the research aim, the three research objectives closely align with the three research questions,
as they define the goals the research sets out to achieve. The following considerations are relevant in this
case:
Research Objective 1 sets the goal to conceptualise and evaluate an analytical sustainability-referenced
framework design that includes a functional energy system model with the operation of components of the
environmental and economic domains as systemic integrants. This objective is achieved in Chapter 3,
wherein a novel geographic-theoretic conception of a predictive ecosystem sustainability framework for
floating PV technology is explored and improved. The framework hypothesis for FPV sustainability is for-
mulated from an operations research perspective in Section 3.5.4, where the postulated ISA-framework is
graphically depicted as a theoretical construct in terms of the geographical mind models of Figures 3.14
and 3.15. The conceptualised framework can serve the appropriation purpose of Objective 1 in terms of
which the contextual intelligence provided by the cooperative framework logic can serve as software logic
in the development of coherent network-type FPV geomodel for the design-stage assessment of floating
PV sustainability within the context of the local agricultural landscape. To this end, Section 3.6 advances
conceptual ideas around the design and realisation of the research model in the graphical depictions of
Figures 3.18 to 3.20. The conceptualised framework can thus be appropriated as computerised software
logic to function as an analytical ecosystem for the theoretical characterisation of the behaviour of a floating
PV ecosystem in the geographical context of local agricultural landscapes.
Research Objective 2 is concerned with applying the sustainability reference framework (defined under
Research Objective 1) as computer programme logic in a computer-simulated modelling methodology. This
objective is achieved as Sub-objectives 2a & 2b in Chapter 4 where a novel real-time simulation model is
defined as a digital expression of the posited sustainability framework theory for FPV installations. Objec-
tive 2a is achieved in Section 4.2, where the systems-level implementation of the parametric research model
succeeds with appropriating the sustainability reference framework as computer logic for developing a dy-
namic systems-thinking modelling and simulation methodology. Section 4.2.3 details the implementation
of this responsive multiscale simulation model arrangement for floating PV in Figure 4.3 (FPV geomodel).
This systems-based 3E co-simulation model dynamically integrates the real-time simulation models for the
functional energy, environmental and economic component objects converging as operational integrants in
a systems network ontology. Objective 2b relates to the component-level implementation of the 3E sys-
tems model, which is achieved in Section 4.3. It followed a model-based design in an object-oriented
environment to transform a large set of systemic properties and inter-connections into the correspond-
ing set of mathematical equations as a compact parametric representation of the FPV ecosystem. As
such, the model-based design approach dictated the development of the 3E floatovoltaic subsystemic com-
ponents and relational properties from first principles (geographical, mathematical, engineering, physics,
economics). This data-processing pipeline also formulated an integrated set of 3E sustainability indicator
metrics and the interactive networking element-coupling definitions and parameter dynamics associated
with the constituent object parameters of the 3E subsystem object domains. Research Objective 2 thus
delivers a prototype digital twin model as a responsive analytical FPV geomodel to command automated
sustainability assessments in a reciprocal model ontology. The proposed model can digitally synthesise
232
and theoretically predict the pre-installation performance and sustainability impact profiles of future planned
floating PV projects on the South African landscape.
Research Objective 3 addresses design aspects around the implementation of the decision-level layer
of the research model. Towards achieving this objective in Section 4.4, the thesis has devised and de-
signed an integrated analytical and decision-supporting profile mapping display in a multi-criterial spider di-
agram (also known as a radar diagram). This profile mapping display enables the integrated FPV geomodel
to serve as a visual geoinformatics decision-supporting interface for floatovoltaic project-planning assess-
ments. The research objective is achieved in Section 4.4.2 through the development of a climate-economic-
WELF framework (ce-WELF) as a decision-support index for driving sustainability in FPV project profiling
(research Objective 3a). As a purpose-formulated sustainability index for floatovoltaics, the set of ce-WELF
sustainability indicators drive computer-aided reasoning in a statistically-based analytical-hierarchy (AHP)
decision model arrangement developed in Section 4.4.3 (research Objective 3b). The layered integration be-
tween the numerical decision-support model with the analytical FPV geomodel, as depicted in Figure 4.12,
facilitates the digitalised translation of the ce-WELF indicators directly from the 3E performance metrics
of the FPV model. The ce-WELF indicators facilitate the move from raw data to actionable insight during
project decision-making. The ce-WELF project goals in Figure 4.2 support active project decision-making
in terms of the user’s ranking and weighting of sustainability indicators for the hierarchically-presented
analytical process model, as depicted in Figure 4.13. Under Research Objective 3, user-directed decision-
making regarding project conditionalities or project goals (e.g. a comparison of sustainability performance
in projects) is supported by the stochastic model through the criterial importance of decision support in the
ranking index of the AHP technique. In this case, decision-making is championed in terms of the user’s defi-
nition of the decision goal for the project, as stated in its policy, the selection of the decision criteria (decision
metrics and indicators), and by determining the criterial weights (importance in the ranking of probabilities
or goal weights), according to the computational intelligence provided by the AHP procedure (refer to Fig-
ure 5.12). The implemented FPV decision model in Figure 4.12 adopts the 3E framework results pertaining
to the FPV model to define a balanced set of ce-WELF performance attributes as evidence-based indicators
in sustainability portfolio mapping. This format of sustainability articulation is suitable for heuristic planning
support in grid-substitution floatovoltaic applications in the South African agricultural context.
The research aim sets the goal to conceptualise and implement a geo-sensitive information system for
quantitative floating PV sustainability profiling. In attaining the research objectives in a computer-aided pro-
cess engineering model for the case of floating PV systems, the research aim succeeded with the geomatic
engineering and integration of the conceptualised theory and simulated developments into a geo-sensitive
information system for sustainability profiling and decision support. While the implemented geoinformatics
system for the processing of sustainability assessments is graphically depicted in Figure M.1, the universal
FPV-GIS model meets the research aim in terms of the implemented geospatial digital twin model that offers
the analytical ability to perform theoretical performance and impact assessments based on a novel analyt-
ical framework, expressed as a computerised model. The first part of the research aim is achieved under
Research Objective 1 in Chapter 3, where the thesis conceptualises a novel theoretical and holistic systems
framework (Figure 3.14) for the theoretical modelling and characterisation of FPV systems behaviour and
the relevant responses to it. The second part of the research aim is achieved under Research Objective 2
in Chapter 4, where this thesis conceptualises and implements a responsive system-dynamics simulation
solution for floatovoltaics in an FPV computer-synthesised model. As such, the research aim is achieved
in terms of the appropriation of the postulated framework that serves as a software blueprint or programme
logic in realising new computer-simulated model expressions in Figure 4.3, through which the synthesised
model contextually predicts/measures floating PV project sustainability in terms of the 3E pillars of sustain-
ability. As part of the research aim, the thesis has successfully expressed the geographical and geomatic
engineering of this real-time computer modelling and simulation solution as a geospatial digital twin model
233
for floating PV systems (FPV geomodel in Figure 4.3). This geospatial digital twin model is driven by actual
real-world geo-sensor data (raw geo-data obtained from satellites and local weather stations) to numerically
animate or mimic the actual (real-world) operating conditions of a floating system in real-world project "re-
hearsals” during the conceptualisation and design stages of the project. Research Objective 3 supports the
research aim by processing the geospatial digital twin model results in an AHP model towards plotting and
articulating the profiled sustainability results on an integrated decision-supporting display. As the final part
of the research aim, the analytical and decision-support models are further embedded in a geoinformatics
platform in Figure M.1, thus establishing an integrated analytical and heuristic-type decision-support system
for floatovoltaics.
The integrated analytical and decision-support solution designated as the FPV-GIS toolset in Figure M.1
can stand as a proxy for sustainability assessments and decision support in guiding future planned FPV in-
stallations. The integrated FPV-GIS toolset serves as a geoinformatics research instrument, with its FPV
project-synthesising capabilities to support 3E and WELF-nexus-oriented project planning in the theoret-
ical appraisal of future planned sustainable floating PV energy projects earmarked for off-grid and grid-
substitution applications in the South African agricultural context. In this context, the synthesised analytical
results of the desktop experiment in Figure 5.4 demonstrate the quantitative theoretical characterisation
capabilities of the FPV model in terms of predicting the performance and impact profiles for future planned
floating PV projects, as depicted in Figure 5.5, within the geographical context of a local agricultural land-
scape. At the same time, the synthesised decision-support results in Figure 5.5 engage the FPV decision
model in Figure 4.12 to expressly demonstrate the sustainability-profiling capabilities of the FPV-GIS toolset
towards quantitative data-driven decision-making support within the geographical context of a local agricul-
tural landscape as geographically depicted in Figure 5.5.
This section summarised the research conclusions by revisiting the research aim and objectives. It
provides a basis for the next section to strategically answer the research questions in terms of the extent to
which strategic research conclusions could be derived from the use of the product of the research aim in
the analytical account of a new floating PV installation planning scenario narrative.
6.3.2. Answering the Research Questions
This section answers the research questions of this thesis within the context of having attained the research
aim and objectives of this thesis, as described in the previous section. It is guided by the method and
experimental synthesising phase of the thesis, where the research methodology and results contribute to
the answering of the research questions.
Research Question 1 asked how systems-thinking principles can be used to conceptualise and formu-
late an integrated tri-dentate framework to theoretically characterise the combined technical, economic and
environmental (3E) responses of floating photovoltaic technology in support of the sustainable development
discourse. While this research question uniquely advances the notion of sustainability portfolio theory in
a holistic 3E floating PV ecosystem, the question is answered in Chapter 3 through applying philosophical
positivist-type and systems-thinking theoretical principles. As such, the thesis posits an integrative theoret-
ical framework alliance in which the mutual technical energy, economic and environmental (3E) responses
of floating PV technology as the tenets of the operational energy ecosystem are recognised as key pre-
dictors of sustainability in agricultural floating PV projects. Under the associated Research Objective 1,
the research examines how philosophical system-dynamics-thinking principles can be applied in a design-
thinking methodology to conceptualise an integrated geographical mind model. This model can serve as
a theoretical reference framework or archetypal logic for the 3E characterisation of the dynamic behaviour
of floatovoltaic technologies within the agricultural theatre of operation. As part of the literature study, the
thesis performed a descriptive analysis to make observations from field experiments in studying the be-
234
haviour of floating PV systems in real-world situations, in terms of which the development process of the
theoretical framework in Section 3.5 examined various options to identify critical sustainability imperatives
at the component and (eco)systemic levels. Descriptive observations made from studying the dynamic
cause-effect behavioural patterns in the responses of real-world floating PV systems and project opera-
tions in Section 3.5.1 permitted Section 3.5.2 to identify and capture the fundamental imperatives to floating
PV identity in the rudimentary holistic systems-thinking diagram of Figure 3.13, thus showing the relation-
ship between the respective entities. Thus, answering the "how” of the research question, the coherent
network-type 3E theoretical framework construct, in conjunction with the 3E causal diagram in Figure H.1,
jointly supports philosophical theory building, with the opportunity to establish new theoretical principles
behind the assessment of floating PV through integrated systemic causality principles in an integrated 3E
tri-dentate framework. While the mind model of the tri-dentate framework in Figure 3.14 reflects the theo-
retical constructs postulated by this thesis to explain FPV systems behaviour and interactions, it represents
only one perceived way of integrating the combined 3E responses of FPV technology in the theoretical
characterisation of FPV.
Research Question 2 asked how a geo-sensitive analytical computer simulation model can be es-
tablished using an integrated theoretical framework logic to characterise the operational dynamics of floa-
tovoltaics in a tri-dentate systems-thinking model for empirical project assessment and decision support
towards sustainable development narratives. In this context, the question is progressively concerned with
the examination of ways in which a modular simulation model can be established as a representation of
the cooperative operations theory posited under Research Objective 2. In answering this question, Chap-
ter 4 is concerned with the application of the postulated theoretical framework archetype/device, particularly
as to how the model can be applied, as an expression of the new theory, to characterise the operational
dynamics of floatovoltaic installations. This research question is answered by applying the first theoreti-
cal principles in Chapter 4 to establish the digital twin as a virtual powerplant design in an adaptive system
model. As such, Section 4.2 details a potentially optimal way through which an object-process methodology,
with model-based design principles, ensures a smarter workflow design to ensure that a comprehensive
object-integrated analytical computerised simulation model is established in terms of an integrated theo-
retical framework logic, as presented in Figure 3.14. The proposed object-oriented system-dynamics-type
digital twin FPV model solution in Figure 4.3 reflects one deterministic solution as a perceived effective way
of defining a quantitative geo-sensitive FPV computer-synthesised model for empirical project assessment
and decision support towards sustainable development narratives.
Corroborating the answer to Research Question 2, the empirical results of the synthesis derived by
the FPV model in a real-world agricultural case study are further presented the theoretical experiments of
Sections 5.3 and 5.4. The quantitative project pre-installation "dress-rehearsal” results of this case-study
experiment make for a practical exhibition of and an explanation for the quantitative performance predictions
delivered by the FPV model. These evaluations are based on the computational performance, and analyt-
ical functionality of the conceptualised FPV-GIS model in a real-world configuration setting under static
and variable climatic and meteorological conditions. The synthesised results of the experiments illustrate
the extent to which the proposed FPV computerised model can characterise the nominal performance, the
real-time performance progression and the operational dynamics of floatovoltaics, especially according to
the proposed tri-dentate systems-thinking model developed for empirical project assessment and decision
support towards sustainable development narratives. The conclusions inferred from the experimental val-
uations in Chapter 5 highlight the value and significance of the experimental FPV-GIS solution, principally
in terms of the framework and the unique ability of the model to decipher the broad spectrum of positive
sustainability traits attributed to the FPV systems variant for a proposed FPV systems configuration.
Research Question 3 asked to what extent an integrated tri-dentate assessment framework and com-
puterised model expression can support the sustainability profiling of floating photovoltaic projects in terms
235
of its water-energy-land-food (WELF) nexus qualities as a means to drive visual/empirical project decision-
making in a sustainable development context. In answering the third research question (in the context of
insufficient in-situ field data, see Section 6.7), the thesis focused on comparative performances between an
FPV and GPV system (as stated in the experimental goal for Experiment 3). This location-specific scenario
analysis and performance comparison enabled the thesis to quantify "the extent” to which the proposed
FPV framework and model can quantify the performance of an FPV system relative to that of an equivalent
co-located GPV system.
The question is associated with the progressive application of AHP decision theory and algorithmic
thinking towards natural resource-based sustainability valuations based on critical Resource Geography
principles. The approach falls in line with the translation, aggregation and weighting of decision indicators
in Research Objective 3, in which the thesis examines the extent to which the FPV model’s 3E metrics can
drive WELF-nexus-type decision support in floating PV project-planning exercises. To this end, Section 4.4
describes the extent to which the empirical 3E performance metrics, derived from the integrated tri-dentate
assessment framework (Research Question 1) and expressed as a computerised model (Research Ques-
tion 2) can be mathematically translated into a quantitative set of climate-economic-WELF sustainability
indicators. While Section 4.4.2 describes the proposal of the thesis to translate 3E metrics into ce-WELF
indicators to help establish a consolidated sustainability index as a basis for AHP-based visual/numerical
decision-support processing, Section 4.4.1 details the extent to which the ce-WELF nexus qualities can
support the articulation of the sustainability profile of floating photovoltaic projects in the visual profile map-
ping display of Figure 4.13a. While delivering an integrated set of multi-criterial sustainability-performance
results in support of computational thinking in a stochastic-type AHP model, this statistical procedure allows
for PV energy projects to be comparatively evaluated with respect to project ranking or the likelihood that a
project would meet the pre-defined operational obligations or sustainability goals defined by the user/impact
practitioner.
Corroborating the answer to Research Question 3, the empirical synthesised and visual results achieved
by this decision-supporting model in a real-world agricultural case study are presented in Section 5.5. It
demonstrates the extent to which the FPV decision model is able to visually deliver quantitative statistical
results, thus illustrating the extent to which the proposed ce-WELF decision indicators are able to compile
the sustainability profile of a floatovoltaic PV systems variant for visual decision making or comparison
with its ground-mounted PV systems counterpart. Considering the multi-criterial spider graph as a profile
mapping display, the environmental sustainability indicators can be ranked in terms of an Importance Index,
specified by the user or impact practitioner through the AHP criterial matrix. In this way, since the framework
combines the project measures in Figures 5.14 and 5.15 as ratio scale measures, the sustainability profile
probabilities for each of the candidate projects are statistically weighted in terms of the user-defined project-
decision goals to give a mathematically meaningful relative sustainability score for each project. The results
of Experiment 3 highlight the broader-spectrum sustainability attributes delivered by an FPV project, with
the values of the ce-WELF decision-criterial vectors as a means to visually support project decision-making
for the purpose of sustainable development narratives.
In terms of Research Question 3, many other decision-processing techniques (see Section 3.7) and
research options can be examined to consider the extent to which ce-WELF-nexus qualities can be statisti-
cally or numerically processed into decision-support indicators for visual spider diagram displays to support
project decisions. However, Section 4.4.3 presents the solution proposed by the thesis as a perceived effec-
tive way in which an empirical AHP solution can effectively be established through the model in Figure 4.13b
to serve as a means to statistically prepare the ce-WELF index indicators as ce-WELF decision-criterial vec-
tors for driving visual project profiling in support of project decision-making and furthering the sustainable
development discourse.
While this section summarised the fundamental research conclusions in terms of answering the re-
236
search questions, the following section responds to the value proposition of the research hypothesis and
towards substantiation of its theoretic principle through experimental data.
6.3.3. Reflecting on the Research Hypothesis
With the scientific hypothesis being the product of the philosophical methodology, and the experimental
results substantiating the theoretical framework principle derived from this conjecture, the thesis can give
an account of hypothesis testing/proving in terms of the experimental results. Consideration is, as such,
given to the role and value of the hypothesis in the conceptualisation of a novel theoretical framework
for quantifying the energy generation and sustainability potential of photovoltaic systems, considering the
impacts of geography, climate, and aquatic installation habitat. The inferences made from the literature
review and philosophical discussions in Chapters 2 and 3 were used to develop the tentative research
hypotheses defined in Section 1.3.3. The drafting of this scientific research hypothesis as such ensured
alignment to the purpose of the study, by stating that: The sustainability attributes and natural resource
impacts of agricultural-type floatovoltaics can be derived from the tridentate responses of the technical-
energy, economic, and environmental domain operations and their collective ecosystem-level interactions.
This hypothetical research statement served as a research compass to guide this philosophical research
investigation (INNS, 2021), to scientifically advance a testable proposition about the possible outcome of this
scientific research study. The hypothesis ultimately offers a meaningful representation in a model capable
of predicting the evolution and evolutionary outcomes of an FPV project in real-time.
As a conceptual antecedent to the tri-node sustainability theory, the hypothesis as an outcome of philo-
sophical innovation advanced the prospective solution towards determining the sustainability repertoire of
FPV systems, to the extent that the interactive techno-economic-enviro ecosystem ensemble established
a new theory for defining sustainability and for explaining the dynamical behaviour of floating PV ecosys-
tems. This theory established the foundation of a sound theoretical framework to formalise the assessment
of floatovoltaic technology sustainability (Research Question 1 and Objective 1). The hypothesis-derived
theoretical framing and logic played a decisive role in the digital development of a geomatics model as a
quantitative systemic modelling intervention (Research Question 2 and Objective 2). The framework further
drove a statistically driven decision-supporting algorithm to support natural resource-based sustainability
decision-making in floating PV project development plans (Research Question 3 and Objective 3). Since
the case study results obtained from the sustainability assessment exercises of Experiments 1 to 3 pro-
vide an exhibition of the value of the philosophical framework, it indirectly serves as a substantiation of a
research hypothesis through the theoretical principles derived from the prepositions of this hypothesis.
While the scientific hypothesis served as a guiding philosophy, the experimental observations could
not disprove the theoretical idea of leveraging concrete agglomeration benefits offered by the techno-
environmental-economic ecosystem ensemble to improve cross-disciplinary ecosystem domain connection
modelling through agency theory. The experimental results rather quantitatively substantiated the tentative
hypothesis as the first scientific evidence that a new sustainability theory for sustainability in floatovoltaics
could not be disproved. The research can conclude that the hypothesis contributed positively to the goal
of defining a new integrated theory for sustainability in FPV that can challenge the state-of-art in the ge-
ographical science challenge of characterising the sustainability profile of floatovoltaic technologies. The
scientific hypothesis and theory and its derivative framework and methodology postulations are of strate-
gic significance as they advance the state-of-the-art to ensure improved coherence between sustainability
assessment frameworks for agricultural-type PV energy systems and agricultural sustainability valuations.
After this section summarises the fundamental research conclusions in terms of revisiting the research
aim and objectives as an extension of the research questions and hypothesis, the following section deals
237
with the strategic research conclusions and recommendations by spotlighting the strategic value of the
research in terms of theoretical and policy implications.
6.4. Strategic Research Conclusions and Recommendations
This section summarises some strategic research conclusions and recommendations around the advance-
ment of knowledge through original research in an applied research value-creation context that ensures
a more realistic portrayal of performance features. First, from a sustainable development standpoint, it
presents strategic interpretations intended to help the reader to understand why the thesis’s multi-disciplinary
research efforts and results are essential. It interprets the value propositions of the posited sustainability
theory and framework derivatives while discussing further academic research toward real-world praxis and
policy implications. Finally, it provides evidence of new knowledge contributions to advance new knowledge
while making research recommendations to guide and highlight the implications of the research for theory,
policy, and practice.
6.4.1. Creation of new theory and knowledge to support floatovoltaic sustainability policy
While the notion of sustainability in floatovoltaic technologies is poorly understood, this paper maintains that
scientists should view this real-world sustainability criteria problem through the lens of cross-disciplinary
sustainability. In this regard, the thesis identifies theory-building prospects in creating a new holistically
integrated theoretical policy that can infuse sustainability and climate change principles into operational PV
performance modelling. From a philosophical theory-building perspective, the thesis sets the goal of geo-
graphical theory-building, in which the study formulates an inductive methodological framework to spear-
head the research towards philosophical theory-building under the sustainable development discourse. In
marrying the energy, environmental and economic dynamics of the FPV ecosystem, the thesis poses the
following Floating PV Sustainability Theory Development Methodology Contribution as an inductive method-
ological framework to remedy the deficiency of traditional models and to study the operational principles and
the geodynamical performance behaviour phenomena of floatovoltaic technologies:
New FPV Sustainability Theory Development Methodology Contribution
The thesis contributes an inductive research methodology in a quantitative methodological research
framework to study the principles of the performance behaviour phenomena of floatovoltaics in a
multi-sector system thinking perspective. The development of the proposed FPV sustainability the-
ory development methodology depicted in Figures 3.7 and 3.8 forms the basis in working towards
the conceptualisation of the proposed sustainability framework in Figure 3.14 and the integrated
behaviour synthesis depicted in Figure 3.3, both summarised in Figure E.1.
In this floatovoltaitry context, this study is rethinking the relationship between the PV energy system
and its interaction with the natural environmental system as well as climate health and resource economics
responses. It proposes a holistic 3E sustainability resolve in terms of the cascading responses between
the techno-economic-environmental or TEE operational domains as well as the coordinated micro-macro-
feedback links and dynamics among the three TEE domain components. The thesis thus makes novel
contributions to the sustainable development discourse by developing a new 3E sustainability theory as
bedrock logic to characterise the floatovoltaic complex and to uniquely develop Energy and Environmental
Economic modelling and analytical capabilities for floating PV installations.
238
The theory’s logic is conceptualised around the holistic integration of environmental and economic do-
main indicators in floating PV energy systems modelling, thus meeting the critically necessary and sufficient
conditions and systemic properties to characterise sustainability in floatovoltaics theoretically. The posited
3E systems theory integrates energy, environmental and economic sustainability in a cascaded coordination
mechanism driven by system dynamics theory. Under the posited theory, dynamic systems principles can
explain the emergent behaviour of a system as a product or outcome of the cross-disciplinary interactions
of heterogeneous ecosystem processes. This integrated set of 3E metrics signifies the cross-correlating
performances and impact effects, offering a unique indicator set to help plan floatovoltaic projects. The
proposed 3E framework and programmatic FPV solution provide a blended set of multi-domain floatoplano-
metrics, including technometrics, envirometrics and econometrics as relevant indicators and situational
intelligence required for breaking the floating PV performance cipher by decrypting, predicting and tracking
the sustainability status and attributions of a floatovoltaic installations.
The main findings of the research are that: (a) the posited holistically integrated techno-enviro-economic
sustainability theory can establish a new methodology for assessing floatovoltaic technologies; (b) the
methodology can be parametrically modelled as a context-sensitive digital computer synthesis technique
in a network-type system dynamics modelling environment; (c) the WELF nexus parameters are valuable
by-products of the techno-enviro-economic sustainability theory; while (d) 3E and WELF metric indicators
can provide FPV project decision support in an AHP-type analytical numerical thinking technique. In terms
of the posited theory, as such, dynamic 3E systems principles can explain the emergent behaviour of an
FPV (eco)system as a deterministic outcome of the cross-disciplinary interactions of heterogeneous pro-
cesses at different levels of operation thanks to the communicative action of the constituent techno-enviro-
economic sub-system domains (Figure E.1). Overall, the posited techno-enviro-economic sustainability
theory and methodology offer a remarkable opportunity to establish an integrated sustainability policy for
the future regularisation of floating PV project valuations as it brings together crucial climate change imper-
atives under a single analytical framework architecture and constructs.
6.4.2. Floating PV sustainability theory as shared vision for PV energy sustainability
With the 3E sustainability expression, the thesis advances the current state-of-the-art of floatoplanometry by
creating new knowledge and understanding by postulating a unique holistic sustainability theory to appraise
and predict FPV project outcomes. With the advancement of the posited 3E system dynamics theory for
floatovoltaic sustainability, logical derivatives of the theorem axiomatically include the development of a sys-
temic framework, computer model and simulation methodology conceptualisations purposefully designed
to synthesise sustainability behaviour as part of design-stage floatovoltaic project planning. This thesis’s
advancement of foundational academic and theoretical principles breaks new ground in floatovoltaic under-
standing and the contextual profiling of the technology’s sustainability credentials. The posited theory also
helps to build a shared vision for floating photovoltaic technology behaviour and sustainability valuations
in an agricultural context, but with widespread relevance to applications in most other industrial sectors.
Furthermore, it strategically advances novel theoretical and framework foundations for regularising float-
ing photovoltaic project performance and impact assessments under eminent stricter future appraisal policy
stipulations for floating PV systems on the international front (refer to (Cohen and Hogan, 2018; Norton Rose
Fulbright, 2021)), potentially offering trailblazing research opportunities in terms of policy implications.
With scientists confronted with the question: "What makes a floatovoltaic installation sustainable?”,
the thesis created new knowledge to answer the question by positing an integrated sustainability theory
and appropriating this theory as a foundation for an integrated multi-criteria framework and novel assess-
ment methodology to provide a unique holistic understanding of floating PV technology sustainability. By
re-imagining the concept of sustainability in farming energy systems from a philosophic-theoretic basis,
239
this thesis’s theoretical sustainability postulate inherently aligns with the classic scientific view of resource-
based agricultural floatovoltaic sustainability. Defined as a function of the collaborative energy sustainability,
environmental sustainability, and economic sustainability (3E) pillars in a nexus network context, it estab-
lishes the basis for deriving the WELF resource interactions for the 3E performance metrics. The thesis,
therefore, advances the following Sustainability Theory Contribution statement as an explanatory theoret-
ical resolution capable of logically explaining the fundamental underpinning principles behind the peculiar
performance behaviour phenomena of floatovoltaics:
Floating PV Sustainability Theory Contribution
The thesis contributes a new contextually-sensitive analytical sustainability theory that systemically
integrates multi-disciplinary operational domains to numerically derive the quantitative sustainabil-
ity attributes and natural resource impacts of agricultural-type floatovoltaics from the tridentate re-
sponses of the technical-energy, economic, and environmental domain operations and their collec-
tive ecosystem-level interactions. The posited sustainability theory is depicted in Figures 3.14 and
E.1, while its outcomes are evidenced in Figures 5.6 to 5.10.
The FPV sustainability philosophy and theoretical contribution to the sustainable development discourse
is shown in the conceptual abstractions of Figure E.1, and its logic operationalised in geo-data-driven
geospatial digital twin, as shown in the desktop geo-informatics simulation instrument in Figure M.1. Draw-
ing on the understanding of the unified 3E, the creation of an enhanced theoretical policy for sustainability
underscores the value of sustainability goal interdependence. The posited 3E theory offers remarkable
flexibility in that the water-energy-land-food resource indicators are an inherent enclave of 3E sustainability
metrics and, as such, derivatives of the tridentate 3E ecosystem responses. In this context, the posited sus-
tainability theory emphasises the vital role of the system dynamics regime in the agricultural sustainability
policy. The postulation of a system dynamical 3E sustainability theory goes a long way towards helping to
build a shared vision for the integrated operational behaviour and sustainability status of floating PV tech-
nologies in agricultural settings. Ultimately, enhancing the quality of digital performance forecasting through
improved computer-aided assessments and digital performance analyses during project design stages can
lower PV systems planning and investment costs (IEA, 2017).
6.4.3. Significance of sustainability theory in sustainable agricultural development
While the sustainability of agricultural floating PV and its spatial sensitivity is a core real-world geographical
problem poorly understood at present, this geographical investigation from the onset recognised that agri-
cultural sustainability fundamentally describes a multidisciplinary goal in a holistic-systems approach (im-
pacting on energy development, environmental protection, economic feasibility and food security) (IRENA,
2015; Luca et al., 2015). The study drawing on recent theoretical advancements in agricultural sustainabil-
ity geography saw prospects in aligning floating PV sustainability assessment in a holistic policy framework
conception that matches that of agrarian sustainability frameworks and agricultural business goals (refer to
(FAO, 2021; Velten et al., 2015)). This agri-vantage point was crucial, both from an environmental and eco-
nomic point of view, specifically because floatovoltaics float on environmentally-sensitive water resources
while directly impacting agricultural farmland by relieving the fertile agricultural land from the energy burden.
These impact effects are directly linked to the critical natural resources imperative to agricultural production.
As such, the technology delivers an economic injection of the farm’s operating budget thanks to the energy
saving-offsets from grid substitution, the climate change carbon taxation/carbon credits of the renewable
energy installations, as well as the asset value of the farmland preserved, all bolstering the income of the
entire sustainability-oriented farming enterprise.
240
This vantage point provided the impetus for the literature study (Chapter 2) and philosophical discussion
(Chapter 3) to contemplate a more holistic agriculturally-aligned evaluation strategy for the sustainability
of Agri-floatovoltaic to help deal with the complexity and multiplicity of the layered levels of "technology
unknowns” of FPV technology (refer to Figure 2.7). In this context, the thesis posited an agri-oriented
framework theory for energy sustainability assessments that advances the notion of unified performance
modelling. Furthermore, the proposed assessment theory serves as a recourse against classic tooling
deficiencies framework imperfections (refer to Figure 3.9), while permitting more reliable and more inte-
grated energy system sustainability projections that better align with agricultural sustainability measures
(refer to Figure 3.15). The thesis can therefore advance the following Agricultural FPV Sustainability Align-
ment statement to emphasise the underpinning synergetic principles that distinctively dovetail with the 3E
operational production components of agricultural production in a theoretical resolution:
Agri-Floatovoltaic Sustainability Alignment Contribution
The posited sustainability theory contributes a new sustainability conceptualisation for farming en-
ergy generation systems based on the unified dynamic interactions between the heterogeneous
technical-energy, economic and environmental operational domains, as a means to ensure broad-
spectrum resource-based scorecard metrics for energy project developments associated with the
sustainability business goals and certification requirements for sustainable agricultural production
operations. The agricultural sustainability alignment is reflected in the comparison between the sus-
tainability concept in Figure 2.8 with the theoretical sustainability concept in Figures 3.14 and 3.15.
The performance profiling theory is defined to command analysis across the three spheres of sustain-
ability, namely the energy-environmental-economic domains. This concept creates space for a systemic
intervention whereby cross-disciplinary floatovoltaic energy ecosystem sustainability modelling leverages
concrete agglomeration benefits offered by the agricultural PV system’s techno-environmental-economic
domain ecosystem ensemble. While the statement underlines the value proposition in unified network-
ing behaviour as a critical component of agri-based sustainability assessments, these fundamentals also
dovetail with the holistic-systemic pillars of agricultural production effectiveness (refer to (Kumar and Mani,
2022)). The posited theoretical framing thus ensures improved alignment with Sustainable Agriculture Stan-
dards and Certifications as it inherently can report on the WELF-nexus system of farming production prac-
tices. Under this agricultural sustainability alignment statement, the posited sustainability theory derivatively
supports the certification for sustainable agriculture products, which is a crucial requirement in the modern-
day export of farm produce and commodities such as fruit and wines (refer to (Moscovici and Reed, 2018;
Sonke, 2008; WWF, 2021)).
On an operational level, the agriculturally-inspired theory advances an integrative 3E systemic interven-
tion to better measure and evaluate the broad and diversified spectrum of sustainability attributes deliv-
ered by water-based floatovoltaic systemic installations for agricultural applications. On a tactical level, the
proposed framework theory addresses the general lack of coherence between sustainability assessment
frameworks for agricultural-type (floating) PV energy systems. Even on a strategic level, the same theoreti-
cal construct uniquely encompasses both the 3E and the WELF-nexus pillars of sustainable development,
through which the posited sustainability theory helps to connect the crucial SDG sustainability goals (refer
to (Bazzana et al., 2022; UN, 2022)). Furthermore, the WELF nexus outcomes capacitate even more signif-
icant alignment with agricultural sustainability valuation/certification frameworks (Alaoui et al., 2022; FAO,
2021; WWF, 2021).
241
6.4.4. The role of Geography and GIS in new sustainability framing for FPV
The thesis addresses current real-world challenges around sustainability appraisals for floatovoltaics through
a newly proposed theoretical framework opportunity conceptualised from geographical foundations. It
uniquely tackled the problem from a geographical sustainable development perspective to bring an envi-
ronmental informatics perspective into agricultural energy system modelling (refer to Figure E.1a), to en-
sure a unique and novel geographic information systems integration of environmental understanding into
the concepts of energy informatics. Furthermore, this inspired the philosophical idea to combine behaviour
synthesis, performance evaluation, and impact assessment in a consolidated theoretical framework for agri-
cultural sustainability modelling synthesis for floatovoltaics (refer to Figure E.1b), an Economic Geography
perspective inspired the opportunity to further bring environmental economics into the integrative equitation
of environmental and energy sustainability prospectives.
This unique geographical perspective enabled the thesis to provide an improved understanding of the
broad-spectrum behaviour of agricultural-type floating PV technologies and to help make sense of the kalei-
doscope of often contradictory performance and impact effects. Traditionally, these performance attributes
have left assessment gaps and voids that have not been adequately accounted for by the conventional
technically-oriented linear assessment techniques (Armstrong et al., 2020; Gadzanku et al., 2021b; Kumar
et al., 2022). This tooling vulnerability problem in itself demanded more advanced tradecraft (frameworks,
models or indicator metrics) to improve predictive evaluation analyses of PV system performances. From a
geographical perspective, traditionally technical PV performance assessment frameworks further need to be
re-modelled from the ground up because sustainability analysis needs more criteria than the conventional
CO2emissions and energy system costs (Hottenroth et al., 2022). By definition, normative sustainability de-
fines a broader geographic-sensitive concept that demands more complex and integrated assessments of
the performance timelapse progression and spatio-temporal dynamics of a floating PV project development
proposal.
In this context, sustainability research required location-sensitive models as fundamental geospatial
tools to support new theoretical and empirical insights into the precepts of sustainability (refer to (Hansen
and Coenen, 2015; Truffer et al., 2015)). Sustainability science research is vitally needed To solve the prob-
lem by exploring promising new avenues to sustainability. Geography’s methodologies and technologies
helped to bring different perspectives and new insights into the sustainability assessment problem through
a theoretical synthesis (refer to (Fu, 2020)).
Geographical Perspective on FPV Sustainability Contribution
The posited sustainability theory contributes a new sustainability conceptualisation for farming en-
ergy generation systems based on the unified dynamic interactions between the heterogeneous
technical-energy, economic and environmental operational domains, as a means to ensure broad-
spectrum resource-based scorecard metrics for energy project developments associated with the
sustainability business goals and certification requirements for sustainable agricultural production
operations. The role of Geography in the proposed sustainability contribution is depicted by the
methodological options in the geographical science engagement shown in Figure 1.8, while the ge-
ographical information system implementation for FPV is evidenced in Figure M.1.
Various fields within the Geographical sciences discipline, particularly Geoinformatics, Climatology
and Sustainable Development, provided a sound theoretical base, numerical platforms and computational
means for exploring plausible 4IR solutions by framing sustainability conceptualisation in an agricultural
context. This view is precious for projecting its implications through the theoretical assessments required
to sustain local energy-food systems through sustainable governance. This thesis argued and has proven
242
that context-sensitive geographical synthesises can give special meaning to the understanding and framing
of agricultural floating PV sustainability.
The theory challenges the state-of-art from an agricultural sustainability context viewpoint by hypothesis-
ing that the infusion of environmental knowledge and climate change economics into a unified sustainability
modelling policy is crucial to better analytical alignment with WELF resource-based agricultural business
sustainability goals (refer to Figure H.1). The thesis further hypothesised that micro-climatic conditions and
hydroclimatic variations in tandem with floating PV design specifications for the aquatic theatre of operation
would help to give fresh insights into the broad and diversified spectrum of location-sensitive sustainability
attributes of these new floating PV energy systems in the future.
6.4.5. Analytical framework for floating PV sustainability assessments
The posited 3E sustainability theory of floatovoltaics goes beyond the risk paradigm of the classical inte-
grated assessment paradigm (refer to (Hottenroth et al., 2022; Naegler et al., 2021)). The thesis defines
a framework for deriving sustainability from collective networking behaviour as the key to on-farm energy
generation sustainability. In line with the sustainability theory, the framework derives sustainability from
the cross-disciplinary coalition between energy sustainability, environmental sustainability, and economic
sustainability to ensure cross-network sustainability consensus. With this theoretical conceptualisation, the
study, in principle, suggests the 3E sustainability theory as an alternative framework for defining a new ana-
lytical sustainability regime to assess a favourable sustainability policy for future floating PV energy project
assessments in agriculture and non-agricultural applications.
The WELF nexus provides a resource-based policy to support sustainable development by preserving
energy, water and environment systems (Buonomano et al., 2022). To support this goal, the thesis derives a
sustainability assessment framework from the underpinning principles of the integrated sustainability theory
by advancing the following Sustainability Framework Contribution statement as a framework resolution for
the assessment of the sustainability phenomenon of floatovoltaics:
FPV Sustainability Analysis Framework Contribution
The thesis contributes a new theoretical sustainability assessment framework for agricultural-type
floating photovoltaic technologies, which states that context-sensitive sustainability attributions and
natural resource impacts for FPV can be generated from the collective system dynamics responses
of the technical-energy, economic and environmental domain operations through systems dynam-
ical exchange interaction analysis. This sustainability analysis framework concept is depicted in
Figures 3.18 and 3.19, its implementation is depicted in Figures 4.1 and 4.4, while its modelling
outcomes are evidenced in Figures 5.6 to 5.10.
According to the proposed framework theory for sustainability modelling and assessments, the context-
sensitive sustainability assessment framework above underscores the practical value of drawing on the
understanding of the unified multidisciplinary 3E sustainability goal interdependence. As a criterial frame-
work for understanding holistic system sustainability, the analytical framework comprises a total of fifty-six
operational performance and impact metrics as potential sustainability indicators classified in the three sus-
tainability categories, namely the energy, environmental and economic disciplinary domains. Refer to the
framework-driven modelling results for the real-time analysis of a floating PV operational ecosystem in the
model synthetic data rendering format given in Table R.1 and Table S.1.
243
6.4.6. WELF-nexus driven decision framework for FPV sustainability assessments
Strategically, the 3E theoretical framing was designed around water-energy-land resources from a statutory
framework perspective. The integrated natural resources continuum aims to develop WELF sustainability
performance indicators for agricultural energy systems. The posited 3E sustainability framing uniquely
allows for the context-sensitive natural resource impacts (WELF indicators) to be numerically derived from
the comprehensive set of 3E metrics because the WELF nexus is seated at the intersection of the Energy-
and-Environmental domain objects as shown in Figure E.1. The environmentally-focused 3E and climate-
economic-WELF resource-related frameworks posited by this thesis aim to move from raw data to actionable
insight. To this end, the thesis selects and ranks the ce-WELF as sustainability performance indicators
(SPIs) for FPV-type agricultural energy systems through the analytical hierarchy process (AHP). The 3E
and ce-WELF indicators can thus serve as a new regulatory framework foundation for regularising the
EIA code of conduct for the performance and impact assessments of floating PV-type renewable energy
systems. In this sustainability cost-benefit analysis context, the thesis documents the following Floating PV
Sustainability Decision-support Framework Contribution for the theoretical characterisation of sustainability
in floatovoltaics in terms of climate-economic WELF resource parameters:
FPV Sustainability Decision-support Framework Contribution
The thesis contributes a novel method for determining and deriving the WELF resource system pa-
rameters from the 3E sustainability attributions. It further defines a new climate-economic-WELF sus-
tainability decision support criteria as composite sustainability index for appraising agricultural floa-
tovoltaic projects. This sustainability decision support framework concept is depicted in Figures 3.20
to 3.22, its implementation is described in Figures 4.12 to 4.14, while the modelling outcomes are
evidenced in the results shown in Figures 5.13 and 5.15.
This thesis’s climate-economic-WELF indicator set engages an AHP algorithm to establish a software
system advisory model supporting FPV project sustainability advisement. While the FPV model generates
the 3E metric outcome sequence in the analytical layer of the GIS system in Figure M.1, the decision-
support layer of the model mathematically translates the 3E metric sequence into a set of customised sus-
tainability indicators. These sustainability data indicators represent the climate-economic-WELF resource
impacts. It serves as a decision-supporting framework index in evaluating the natural resource-based sus-
tainability of FPV projects. The "economic” indicator of climate-economic-WELF indicator set comprises a
unique EST (economic surface transformation) indicator invented by this thesis, while the "climate” indicator
defines the hydro-carbon sequestration attributions of floatovoltaics (refer Section 4.4.2). The thesis posits
the novel water economic surface transformation indicator (w-EST) as a novel ratio-based indicator that is
uniquely designed with the idea to reflect the income potential of an FPV system per square meter (also
refer Prinsloo et al. (2021) article).
Seen as a fitting decision-supporting framework for EIA, the posited ce-WELF decision-support frame-
work comprises six sustainability performance and impact metrics as potential sustainability indicators
classified in the three sustainability categories. With the climate-economic-WELF sustainability indicators
mathematically derived from the geographical performance metrics as sustainability indicators (refer to re-
search hypothesis in Section 1.3.3), these indicators were custom-designed to empirically model, evaluate
and graphically map the sustainability profile of floating PV projects. By shaping the narrative of WELF
resource-use sustainability, the proposed geomatic metrics and indicators were all designed to support an
improved understanding of the performance and resource-use sustainability profile of FPV technology in
climate-smart agricultural applications.
With this contribution, this decision support framing constitutes a resource-driven sustainably indicators
244
climate-economic-WELF decision support sustainability index or framework comprising six indicators in the
three categories. The ce-WELF sustainability indicator index delivers computational decision advisement
capacity whereby the user can consider project decision options delivered by a stochastic AHP model on
a numerical or visual basis. These aspects are illustrated in framework-driven modelling results shown
in Table T.1 (see Appendix T). Based on the premise that sustainability is a multi-disciplinary ecosystem
quality, the proposed geographic-theoretic construct and associated framework serve as logical foundations
for the cross-disciplinary modelling of the collective sustainability intelligence attested to an FPV ecosystem
in a dynamic methodology and computational architecture.
6.4.7. Fundamental principles of sustainability assessment methodology for FPV
Overall, the thesis’ systematisation of technical expert knowledge offers an integrated sustainability as-
sessment methodology in a dynamical systems networking technique. The method provides the unique
capability to perform scientific sustainability evaluations in a computer-aided environment. It has a service-
oriented objective in addressing the broader critical project decision-making elements required to profile PV
project sustainability attributions that are required by, and focused on, by domain experts such as impact
practitioners (Sala et al., 2015). Sustainability assessment is as such proposed as a complex scientific
project appraisal method to support sustainable development through decision-making and policy imple-
mentation in a collective energy, environmental, and economic (3E) context (refer to the mental model of
Figure 3.14). From a systemic standpoint, the proposed theory asserts that silo-type floating PV project sus-
tainability analysis at one operational domain, in isolation from other functional domain parts, is incomplete
and provides unsatisfactory sustainability assessment results.
The theoretical 3E framework abstraction of this thesis is operationalised on the ecosystem level thanks
to the application of system dynamics principles in an object-process methodology (Figures 3.14 and M.1.
The theoretical framework ensures that the operational techno-enviro-economic (3E) network nodes or net-
work elements are interconnected and mutually influence one another (the so-called coopetition stakeholder
theory, or theory of cooperation and competition in relationships). In this systemic framework intervention,
the proposed theoretical framework ensures improved connectivity between modern climate-change prin-
ciples, environmental understanding and resource economics as part of the holistic multi-scale theoretical
characterisation of floating PV technology ecosystem sustainability. Furthermore, the thesis delivers a
responsive FPV simulation model that renders an array of more than 50 3E performance metrics to pro-
file the technology’s behaviour and performances through real-time and lifetime synthesis results (refer to
Tables R.1 and S.1). As a systemic innovation supporting multi-disciplinary sustainability ecosystem syner-
getics (Lindhult et al., 2022; Meynhardt et al., 2016), the practical implementation of the newly concreted 3E
theoretical principal excels the state-of-art in sustainability assessment practice beyond the conventional
technical evaluation practices of performance assessment and integrated environmental assessments.
Advancing the state-of-the-art through the practical implementation of a newly proposed theoretical
principle accentuates that the thesis created new knowledge to fill critical methodological gaps presently
jeopardising regulatory project approvals in the industry. The 3E cross-disciplinary concept of sustainability
includes operationalising normative sustainability assessments through computer synthesis, thus enabling
the study to advance the following FPV Sustainability Methodology Contribution statement:
245
Integrated FPV Sustainability Assessment Methodology Contribution
The thesis contributes a new methodology for the predictive assessment of sustainability in FPV
energy generation systems developed as an object-process methodology wherein the technical-
energy, economic and environmental domains are treated as software objects (network agents).
The object entities operate as nodes in an operational network that continuously exchange infor-
mation and resources to deliver metrics representing project performance and impact outcomes in
integrated assessment profiles as real-time operations data during 3E co-simulation. The integrated
floatovoltaic sustainability assessment methodology is depicted in the simplified data-driven object-
process modelling methodology procedures shown in Figure 4.1 and in the more extensive system
dynamic causal diagram depicted in Figure H.1. The outcomes of its execution in Figure 5.3 is
evidenced in the experimental results of Figures 5.6 to 5.10.
In establishing a newly proposed sustainability assessment methodology for floating PV, the posited
3E/EEE (Energy, Environmental, Economic) sustainability theory, in effect, warrants the externalisation of
the theoretical proposal as the foundation of a new methodology. Guided by the posited framework as a
computer logic in an assessment tool, the methodology can duly predict or measure sustainability outcomes
in terms of cross-disciplinary interactions, transdisciplinary sustainability metrics, and indicator dimensions.
The conventional performance assessment framework for PV energy systems decouples the technical and
economic aspects in a simplified silo-based assessment approach. This approach has a drawback in that
insufficient consideration is given to environmental causation or environmental impact aspects (refer Fig-
ures 3.9 and 3.10). With the void, the conventional framework implicitly finds it difficult to remedy the toolset
deficiencies towards complying with the sustainability assessment profiling protocols required of sustainable
development initiatives. The posited theory thus provides an edge over conventional analytical frameworks
for PV project assessments as the 3E framework remediates core imperfections in current tooling models.
It also provides a holistic systems-based approach as recourse against the foundational deficiencies left by
the silo-based framework approach engaged by conventional techno-economic PV performance models.
6.4.8. Sustainability framework and methodology as computer model logic
As a fundamental theorem of the sustainability calculus for floating PV energy generation, the posited the-
ory intrinsically aligns the broader sustainability modelling with the grassroots-level systemic operations
behaviour of floatovoltaics. In this context, the theoretical sustainability policy regime was operationalised
in a deterministic computer model wherein the machine logic integrates the technical energy, environmental
and economic responses as software objects in an object-oriented programming environment. Further-
more, as a framework for efficient automation of computational scientific workflows (Liu et al., 2019), this
3E system model conceptualisation supports evaluation scenarios in the development of pervasive sus-
tainable energy development projects in the agricultural sector based on the principles of climate change
sustainability (refer to (Madurai Elavarasan et al., 2021; Nakicenovic, 2022)). The integrated geospatial 3E
sustainability assessment framing intrinsically supports project assessments in terms of the water-energy-
land-food-resource (impacts typically required in environmental impact assessments).
Seen from a sustainable energy development viewpoint, the thesis’s sustainable development policy
theory uniquely defines an energy development that is energy efficient, economically feasible, and environ-
mentally viable. Since the theoretical 3E sustainability policy framework is descriptive of both sustainability
policy and function, the postulated theory is numerically integrated into the underlying functioning of the
theoretical characterisation of floating PV operations at an ecosystem level. By the premise of the 3E
theoretical sustainability framework being true, the externality of it serving as computer program logic in
246
characterising floating PV sustainability behaviour is emulated through the operationalisation of the theory
in a computer model implementation expression. In this context, the thesis documents the following Floating
PV Operations Modelling Contribution as an operational behaviour statement in support of the theoretical
characterisation of the operations behaviour of floatovoltaics:
Floating PV Operations Modelling Contribution
The thesis contributes a system dynamics computer model expression as a novel geospatial digital
twin for floatovoltaics to emulate the contextual tridentate responses of the technical-energy, eco-
nomic and environmental domain operations and its dynamical causal interactions with the theoreti-
cal characterisation of the real-time geodynamical behaviour of agricultural-type floating PV technolo-
gies. The FPV operations modelling process is depicted in Figures 3.17 to 3.19, its implementation
is shown in Figures 4.1 to 4.3, and the model outcomes evidenced in Figures 5.6 to 5.10.
From a modelling regime governance perspective, the sustainability valuation framework and method-
ology precepts provide a theoretical foundation for establishing a novel GIS-housed geospatial digital twin
model for floatovoltaics (refer to (ESRI, 2022; Stoter et al., 2021; Jones et al., 2020)). The posited sus-
tainability framework thus supports real-time floating PV synthesis modelling since its logic underpins the
participatory system dynamics modelling to establish the geospatial digital twin model for FPV as a sustain-
ability profiler (Figure M.1). With the posited theory integrated into a responsive analytical FPV-geomodel, it
operates as a geospatial digital twin environment to mimic real-world floating PV system behaviour to com-
mand automated sustainability assessments. Since the geospatial digital twin implements a digital floating
PV emulator, it can stand as a proxy in design-stage floating PV project prototyping. It can thus perform
time-series forecasting of project outcomes to ensure an improved understanding of the sustainability and
operations of FPV technology.
The posited sustainability theory, framework and methodology are integrated into a geospatially-driven
model. This generic digital prototype modelling appropriation brings a different perspective to the problem
of PV sustainability assessments, as the theory is operationalised in a digital twin to provide synthesised
insights into real-world project assessments. The simulation solution enhances integrated FPV enterprise-
level operations understanding, as featured in the parametric study and experimental performance profiling
outcomes of Figures 5.6 to 5.10.
6.4.9. Sustainability decision supporting framework as computer model logic
Since the thesis’ postulated 3E sustainability theory is inherently built upon the synergistic fundamentals
of the underlying water-energy-land-food ecosystem enclave, the 3E elements of the proposed approach
further capacitate numerical modelling in WELF resource-based sustainability profiling. The WELF decision
framework provides a set of sustainability indicators to facilitate and enhance decision-making in an ecosys-
tem context. Associated with the land, air and water impacts of an FPV energy system, the ce-WELF index
provides a conceptual structure and principles for integrating the critical resource elements of the technical
(energy), environmental, and economic metrics as institutional dimensions of project sustainability deci-
sions. The ce-WELF decision framework forms part of the comprehensive PV project ecosystem appraisal
methodology, as featured in the well-balanced scorecard sustainability appraisal outcomes portrayed by
Figures 5.13 and 5.15. In this context, the thesis documents the following Floating PV Decision-support
Modelling Contribution as a sustainability advisory decision platform in support of the theoretical character-
isation of the sustainability profiling of floatovoltaics:
247
Floating PV Decision-support Modelling Contribution
The thesis contributes an analytical hierarchy model to process the tridentate 3E responses to-
wards the decision-supporting characterisation of climate-economic WELF sustainability profiles of
agricultural-type floatovoltaic systems. The floating PV operations modelling process is depicted
in Figures 3.20 to 3.22, its implementation is reflected in Figures 4.12 to 4.18, while the model’s
outcomes are evidenced in Figures 5.13 and 5.15.
With the numerical floating PV decision-support modelling contribution, the thesis developed an ana-
lytical hierarchy model as an empirical decision-support model driven by the multivariate output data sets
(3E or WELF data) of the FPV model. The AHP numerical decision support model is layer integrated
into the GIS toolset as depicted in Figure M.1 to serve as an empirical decision-support model. The FPV
model’s multivariate output data sets (3E or WELF data) drive the AHP decision-supporting model to serve
as multi-criteria numerical decision support to help to improve the user’s decision-making capabilities dur-
ing the planning phases of a floatovoltaic project’s commissioning and certification phases. In this context,
the thesis postulates the engagement of the climate-economic-WELF (ce-WELF) data framework (derived
from the 3E output data metrics) to serve as a decision-supporting index in the valuation of the natural
resource-based sustainability of FPV projects.
6.4.10. Environmental and resource/climate-health economics into FPV modelling
This thesis maintains the philosophical argument that an improved understanding of the theoretical and
empirical aspects of the natural environmental system and its associated environmental economic account-
ing and natural resource economics should undoubtedly play a more prominent role in conceptualising an
ecosystem-based theoretical sustainability framing for floatovoltaic technologies. To this end, the posited
theoretical framing of this study recognised the importance of environmental understanding, resource-
economics and climate change/health economics in modelling the operational behaviour and sustainability
attributions of floatovoltaics. The posited theory affords considerably more prominence to environmental un-
derstanding and environmental-economic understanding as they are considered necessary and sufficient
elements of floatovoltaic sustainability. Naturally, as key drivers of PV project sustainability, the environ-
mental and economic simulation functionality is needed to improve the design-stage characterisation of a
floatovoltaic system in a multi-disciplinary research effort and model. From a regulatory perspective, this
philosophical idea can also help to prove to regulators that creating floatovoltaic projects does not incur
unnecessary costs to the climate, rather than the technology offers highly promising environmental offset,
resource and climate-change benefits that are also financially rewarding thanks to climate-health economic
interventions promoted under climate change principles.
To this end, the thesis exploited the unique opportunity offered by FPV to integrate environmental un-
derstanding and environmental-health economics into the holistic modelling of sustainability (refer to Fig-
ures 3.10 to 3.13. It is essential to incorporate environmental understanding into the sustainability criteria
because FPV energy technology touched on issues concerning the interplay among the water-energy-land-
food nexus dimensions, a critical assessment issue lacking in FPV valuations thus far (Gadzanku et al.,
2021b; Prinsloo, 2017a). In this context, the EPA’s sustainable development interpretations further helped
to inspire the philosophical arguments and motivations in Chapter 3 to uniquely incorporate environmental
and climate-health understanding into the physical, operational modelling of the floating PV ecosystem. As
one of the critical contributions of this thesis derived from the posited theoretical sustainability construct, the
thesis documents the following Environmental-Economic Sustainability Integration Contribution statement:
248
FPV Environmental-Economic Sustainability Integration Contribution
The thesis makes a novel theoretical contribution by systemically fusing technical Energy thinking
with an understanding of the natural Environmental system (environmental offsets, impacts and mi-
croclimate) and environmental-Economics (resource-economics, climate-health economics, energy
economics), key determinants of sustainability and drivers of the distinct variance and diversity of
FPV system performances. This 3E trilogic fusion uniquely succeeded in uncovering the hidden
WELF resource impact offsets and, by implication, revealed the broad-spectrum sustainability reper-
toire of floatovoltaics in a single theoretical construct. The incorporation of environmental and eco-
nomic understanding into floating PV sustainability modelling is reflected in the mind model depic-
tions of Figures 3.14 and E.1, in the FPV causal diagrams depicted by Figures 4.1 and H.1, as well
as in the FPV computer software modelling charts shown in Figures K.1 and M.1. Environmental-
economic integration significance is evidenced in the environmental footprint shown in Figure 5.9, in
the economic performance profile of Figure 5.10, as well as in the climate-economic-WELF footprint
for a proposed FPV system given in Figures 5.13 and 5.15 .
With this postulate, the infusion of the natural environmental system and the economic system into
the posited 3E sustainability theory gives the FPV-GIS synthesis model a contextual window into the envi-
ronmental influences and offset effects together with its associated climate change financial ramifications
on installation sustainability. This opportunity inspired a more robust framework for validating the tech-
nology’s suggested co-benefits and impact effects concerning the local energy-water-food (EWF) system
and the economic benefits of natural resource preservation as an assessment requirement not adequately
addressed in current models (Gadzanku et al., 2021a). While these real-world challenges currently es-
tablish crucial knowledge gaps and research problems. The thesis could address these issues thanks to
investment approval and EIA commissioning inputs. The framework was, as such, partly inspired by the
challenges experienced with the EIA processes, which demanded a solution to the FPV assessment prob-
lem as a means to support project licensing and commissioning authorisations. Regarding support for EIA
practice, tactical improvements require provision for computer-aided EIA support in compiling FPV planning
reports to be submitted concerning project licensing and commissioning authorisations. In this context,
the investigation viewpoints of this thesis emerged from the realisation that critical knowledge gaps are not
isolated but also layered in their own right, thus surfacing as layered levels in respect of the interrelated
"technology unknowns” in floating PV system assessments (Armstrong et al., 2020). Although, at the same
time, these layered knowledge gaps form the basis of the "technology unknowns” that remain the most
significant barriers to floatovoltaic installations worldwide (Gadzanku et al., 2021a; Pritchard et al., 2019;
Spencer and Barnes, 2020), the research advances the notion that most of these knowledge gaps are
associated with a lack of understanding in terms of environmental impacts and climate-change economics.
With the various critical knowledge gaps and technological unknowns remaining as the main barriers to
entry into the floatovoltaic arena in many regions of the world (Cuce et al., 2022; Gadzanku et al., 2022b;
Spencer et al., 2019), the motivation for this thesis to consider new avenues to develop a geographically-
motivated tool to quantify the broad spectrum of environmentally-oriented impact effects of prospective
floatovoltaic installations is underlined. To this end, the thesis makes novel contributions underlying the
notion that could address the "technology unknowns” in FPV project sustainability assessments by incor-
porating an understanding of modern climate change in the systemic modelling of floating PV ecosystemic
operations. While advancing the notion of incorporating an understanding of the environment and environ-
mental economics into the theoretical characterisation of floating PV technology to support such processes
and sustainability modelling, the thesis recommends that efforts to improve existing PV performance mod-
els should ideally place greater emphasis on an understanding of the environment and of environmental
249
economics to facilitate the modelling of floating PV performances.
6.4.11. Geographical modelling of floating PV natural environmental micro-habitat
Water reservoirs are known to create their own microclimate footprint, conceptually as a type of dynamic
bubble hemisphere around the reservoir. With floating PV systems deployed as hydrone platforms in this
aquatic theatre of operation, it is unsurprising that field-installed floating photovoltaic projects report clear
signs of distinct (water reservoir-induced hydroclimatic) variability. These effects also remind of grape vine-
yards, where the micrometeorology of the vineyard canopy impacts fruit production (Hunter et al., 2021). It
emphasises the role of micrometeorology and microclimate in the spatiotemporal behavioural and sustain-
ability responses of floatovoltaic energy systems. The challenge relates to the science behind the technol-
ogy’s geodynamics and modelling of the cumulative microclimatic effects. In the FPV context, the thesis
considers microclimatic effects to be water reservoir-induced hydroclimatic variability impacts, in terms of
which these environmental interactions are evidently the key drivers of the distinct variance and diversity
in the dynamical behaviour and sustainability attributions of floatovoltaic power generating systems. The
geographic mind model conceptualisation of Figure 3.11 suggests that the role of the ambient hydroclimate
in the energy system’s sustainability value chain deserves commensurate attention. Because the reservoir
evidently creates a microclimate or hydroclimatic bubble around the floating PV system, the aquatic habitat
impacts quantitatively the PV system’s operational performance and impact effects. The challenge is akin
to modelling of microclimatic effects in vineyard fruit production (Hunter et al., 2021), where solar-driven
grape production is impacted by evapotranspiration and the micrometeorology within the vineyard canopy.
The thesis advances the academic argument that many of the distinctly varying and diversified perfor-
mances and impact effects of floating PV systems can be explained in terms of the (closed-loop-type) en-
vironmental feedback effects from the water reservoir and its hydroclimatic bubble. Since the hydroclimatic
bubble elucidates the distinct dynamical behaviour of floatovoltaics, modelling the micro-environmental habi-
tat forms an integral part of the 3E integrated analytical paradigm. Seeing opportunity in the fact that the
apparent hydroclimatic effects have not been fully modelled from a PV microclimatic perspective before,
the hydroclimatic variability is systemically modelled in this thesis as part of the environmental domain
object of the operational floatovoltaic ecosystem. Modelled by the hydroclimate’s virtual sensor model in
Section 4.3.4.2, the observed effects of environmental hydroclimatic causality are characterised in climatic
terms. The FPV model further includes systemic-type reinforcing feedback loops to the energy and eco-
nomic domain objects to support characterising the impacts on energy and financial performances. The
profiling of the consequences of these hydroclimatic effects is visually reflected in the experimental results
of Figures 5.8 to 5.10, wherein the power generation efficiency curve of floating PV in Figure 5.8 especially
shows a significant lifetime sustained efficiency gain in power production yield when compared to that for
an exact ground-mounted PV system counterpart.
The thesis provides an interesting geographical perspective on the peculiar behaviour of floating PV as it
theorised about the role and impact of the natural resource interactions and aquatic reservoir habitat (refer
to Sections 4.3.3.3&4.3.4.2. It proved to be vitally important to incorporate a climatology understanding
of the aquatic microclimate into the floatovoltaic sustainability model because the hydroclimatic variability
reforms the microhabitat (refer to Figures 3.10 and 3.11). As such, the dynamic hydroclimatic feedback
within the aquatic theatre of operation directly impacts the power generation efficiency of the PVlib model,
making it crucial to condition the input parameters to the PVlib model before simulating the FPV energy
outputs (refer to Figure L.1). This thesis, therefore, includes a hydroclimatic pre-processing sub-model
between the TMY weather station meteo-data inputs and the PVlib model, thus empirically accommodating
hydroclimatic understanding as part of the environmental sustainability criteria (refer to Section 4.3.4.2).
250
These conceptualisations helped to incorporate an aquatic micro-habitat with micro-scale and mesoscale
hydro-climatic variability into the 3E framework’s environmental system. As another of the key contributions
of this thesis, the new micro-habitat modelling hypothesis is therefore defined to quantitatively assess and
evaluate the sustainability features of generic floating PV project installation as summarised as follows:
Floating PV Microhabitat Modelling Contribution
The thesis makes a novel contribution with modelling of the microhabitat housing the PV energy
system, specifically the microclimate and the hydroclimatic variations of the natural environmental
system that are drivers of the distinct variance and diversity of floating PV system performance and
impact effects and, by implication, drivers of the integrated sustainability of floating PV installation
projects. Floating PV microhabitat modelling is depicted in the FPV ecosystem causal diagrams of
Figures 4.1 and H.1 and in the hydroclimate conditioning model object processing of Figures K.1
and L.1. FPV microclimate modelling significance is evidenced in environmental footprint results of
Figure 5.9, the energy yield profile of Figure 5.8, the economic performance profile of Figure 5.10,
and the FPV system’s climate-economic-WELF footprint results of Figures 5.13 and 5.15.
This postulate is based on the premise that the local hydroclimatic conditions around the water body
that receives the FPV system alter the climatological and meteorological conditions around a floating PV
installation. The model abstractions in Figures 3.10 and 3.12 guided the modelling of the hydroclimatic
impact aspects as a part of the PV microhabitat model. Furthermore, the FPV entity diagram in Figure H.1
provided more insight into the logical thinking on how to infuse the hydroclimatic conditions of the enveloping
ambient environment into the environmental domain object of the integrated 3E analytical paradigm (refer
to Section 4.3.4). The thesis thus proposed a computational algorithm for the climatological modelling of
the hydroclimate induced by the floating PV microhabitat. The idea follows a novel approach to tackling the
microhabitat modelling problem from a climatological and meteorological angle as an integral part of floating
PV modelling. The proposed geographical science approach defines the microclimate and mesoclimate
impact as numerical climatological characteristics that condition the meteorological weather data from local
dynamic weather station data to emulate the higher humidity and moisture content at the air-water interface.
As such, the water-integrated sustainability conceptualisation gives greater prominence to the microscale
environmental complex and micro-climate prediction, which provides an all-important tactical advantage in
the realistic modelling and characterisation of the real-world floating PV ecosystem.
Predicting and comprehensively understanding the spatial extent requires computational algorithms for
dynamic climatological and hydrological sensitive applications proved crucial. Such a model compensates
for how these hydroclimatic conditions impact, for example, generation system sustainability as it improves
the energy performance efficiency profile, the environmental effects offset profile, the economic potential of
the floating PV system, and the life expectancy of the floating PV panel and equipment. The hydroclimatic
impact within the aquatic microhabitat for FPV can be so dramatic that the climate rating for photovoltaic
modules begins to take effect (refer (IEA PVPS, 2020)), to the extent that modelling an FPV system in
an arid climate location may need to engage a sub-tropic PV panel rating because of the hydroclimate’s
altered ambient humidity regime. Further support to the 3E sustainability theory is thus given in terms of
improved energy performance profiling to ensure more effective due diligence and broad-spectrum sus-
tainability assessments. In this context, a new meteorologically-motivated floating PV habitat modelling
procedure and panel thermal model could further be defined in Section 4.3.3.3 to estimate the time-varying
cell temperatures for PV panels installed over open water basins.
As a key recommendation of this research, advanced in more detail in Section 6.7, the thesis suggests
that efforts towards improving existing/conventional PV performance models should place greater emphasis
251
on modelling the micro-climatic dynamics and hydro-climatic conditions specific to the floating PV floater de-
sign. The numerical modelling of the floating PV microhabitat embodiment’s localised ambient hydroclimatic
conditions can further help to improve conventional PV assessment models. In addition, it offers an oppor-
tunity for externalising mutual environmental impact wisdom to address the effects of analysis uncertainty
in current floating PV sustainability modelling schemes.
6.4.12. Geo-informatics-based data-science technique for predictive FPV assessments
With ontological intelligence being one of the foundations of information extraction and multidisciplinary
digital modelling in Data Science environments (Buitelaar et al., 2008; Vazquez, 2018), the proposed new
4IR-type 3E theory forms a valuable ontology-based logic for information retrieval functionality for the sus-
tainability appraisal of PV projects. The proposed geo-informatics approach concerns the general applica-
tion of tools of computation and analysis to the capture and interpretation of energy system data. Together
with the associated method and methodology, the thesis essentially establishes a new analytical Data Sci-
ence technique for FPV project assessments in a geospatial digital twin environment (refer to (Ayani et al.,
2018; Semeraro et al., 2021)). According to this narrative, a geospatial digital twin can act as a digital
proxy equivalent to a real-world floating PV installation to digitally synthesise and predict the real-world out-
come of an FPV installation system’s performance and impact effects in terms of 3E project outcomes and
WELF resource indicator outcomes. Driven by real-world geo-sensor data, the implemented FPV model
as a geospatial digital twin for floating PV essentially mimics the geodynamical behaviour of floatovoltaics
through discrete-time computer simulations in the real-time synthesis of lifetime digital project enactment
exercises. With real-world operational data driving the geospatial digital twin model, analysis-by-synthesis
experiments can engage the FPV-GIS toolset to help draw inferences from digital project rehearsals in
cyberspace during the project’s pre-installation phase.
Thus, as a further part of the novel contributions of this geographical assignment, the geospatial digital
twin sits at the heart of a new analytical data science technique established for the sustainability assessment
of floatovoltaic technologies. By following the physical Geography principles in the geographical science
engagement, depicted by Figure 1.8a, the analytical data science philosophy in Figure 1.8b drives the
simulation tool development for floating PV. In the abstract theoretical modelling representation for FPV, the
rudimentary mechanistic data-flow model of Figure 4.1 offers an improved methodology to orchestrate data
flow in the newly proposed analytical data science technique for floating PV. With this new methodology
serving as an underpinning theoretical basis for PV project assessments, it supports the implementation
of the analytical data science technique and method in the decision-supporting geoinformatics platform
of Figure M.1. Furthermore, it highlights the multi-disciplinary information technology value proposition of
the newly proposed analytical data science technique in terms of its integration of geographical/geospatial
sciences, environmental sciences, decision sciences and mathematical sciences with knowledge integration
from computer sciences and engineering disciplines to support the geo-intelligence principles in Figure 1.8.
As part of the broader digital earth application initiatives in support of digital geography (Scharl, 2007;
Yang et al., 2011), our government engaged in the development and expansion of digital platforms such as
the online E-GIS platform (Environmental Geographic Information System). It is a digital platform for accom-
modating tailored GIS solutions and improving contextualised science-based location insight (DFFE, 2018;
EGIS, 2022). In this context, the thesis’ postulation on the 3E cross-disciplinary concept of sustainabil-
ity includes operationalising normative sustainability assessments through computer synthesis and project
playbook narratives. It thus advances the following Analytical FPV Data Science Technique Contribution:
252
Analytical FPV Data Science Technique Contribution
The thesis makes a new contribution by advancing a new analytical data-science technique for proac-
tive sustainability analysis in FPV energy generation systems. The geo-informatics-based data-
science processing model for predictive FPV assessments is reflected in the model framework of
Figure 3.21, in the implementation flowcharts given by Figure 3.22, as well as the analytical and de-
cision layer implementations in Figure M.1. The performance of the analytical FPV data processing
technique is evidenced in the experimental results given in Figures 5.6 to 5.10 as well as in the WELF
footprint shown in Figures 5.13 and 5.15.
Competent authorities can support new energy technology usage, provided that its impact benefits are
quantifiable in a scientifically compiled environment and project plan (IFC, 2020). The academic research
work of this thesis advances a new analytical data science technique for floatovoltaic technologies. It
provides an improved EIA-justifiable analytical methodology and data-science technique to address the
current real-world problems around floating PV installation characterisation. Resting on the pillars of the
posited theoretical framework for the theoretical characterisation of floating PV system sustainability, the
context-specific geo-sensitive sustainability valuation technique of this thesis supports digital Earth and EIA
initiatives (Digital Earth, 2022; ELaw, 2022). The appropriation of this theoretical framework in a digital twin
modelling for geomatic project rehearsals can decode and quantify the more diverse and broader spectrum
of performance and impact effects prediction aspects of floating PV technologies.
6.4.13. Comparison of FPV model with existing PV performance models and techniques
Despite floatovoltaic’s impressive technological performance features, installation growth rates and techno-
logical developments in installed floating PV capacity in recent years, analytical modelling developments
and studies on the broad-spectrum performance, impact effects and reliability of floating PV are as yet
scarce (Gadzanku et al., 2021b; Kjeldstad et al., 2022; Kumar et al., 2022). Although critical knowledge-
and methodological- gaps in conventional PV assessments currently serve as a barrier to the deployment of
floating PV systems, the fundamental principles behind 3E floating PV sustainability assessment resolution
have proven their capacity to address these gaps.
According to the postulations of this thesis, the state-of-art PV performance assessment practice needed
to go beyond the conventional technical evaluation practices of performance assessment and integrated
environmental assessments to accomplish and accommodate the broader criteria of sustainability assess-
ments. The posited theory and framework provide an edge over conventional analytical frameworks as it
remediates core imperfections in current tooling models (refer to Sections 2.6 and 3.5). With the modelling
abstractions shown in Figure 3.15, the theoretical postulation of this thesis overcomes severe structural im-
balances existing in open-loop modelling frameworks of existing GPV systems shown in Figure 3.9. In this
context, the thesis advances the following PV performance modelling improvement as Novel Contribution:
Floating PV Modelling Improvement as a Novel Contribution
The thesis’ novel contributions improve conventional analytical PV performance capabilities by ad-
vancing predictive PV performance modelling capabilities from conventional performance assess-
ments, to more advanced integrated assessments, to even more advanced sustainability assess-
ments required to support the sustainable development complex. The floating PV modelling improve-
ment is evidenced in the comparative experimental results for conventional GPV and the posited FPV
modelling paradigms portrayed in Figures 5.6 to 5.10.
253
This contribution is backed by scientific evidence to enlighten the details of how the holistic systems-
based closed-loop approach of the thesis’ knowledge contributions serves as recourse against the founda-
tional deficiencies left by conventional open-loop frameworks engaged by most universal techno-economic
PV performance models (refer to the model abstractions in Figures 3.9 to 3.15). While current best practice
models in PV performance assessment mostly explore the simplification benefits of reductionist thinking,
such models conventionally engage an open-loop analytical framework (depicted by Figure 3.9). However,
with this open-loop framework being illustrative of the conventional tooling deficiencies, such framework
logic is principally unable to cope with the analytical processing complexities required by causal-responsive
floating PV operations (requiring closed-loop causality). The conventional open-loop modelling framework
in Figure 3.9 in principle does not meet the modelling requirements to handle environmental feedback and
systemic dynamism of causal-type interactions observed in the floating PV technology-inspired analyses
of this thesis (i.e. depicted by Figures 3.13 and H.1). The result is that ground-mounted photovoltaic
(GPV) systems such as conventional PV modelling tools show wide-spectrum deficiencies because they do
not account for the hydroclimatic variations that occur as environmental feedback effects. These deficien-
cies, therefore, arise because the hydro-embedded FPV system interfaces directly with underlying aquatic
ecosystems to provide a more diverse range of superiority traits in terms of technical benefits, system
degradation counter-activity, natural capital benefits and environmental co-benefits (Gorjian et al., 2021;
Goswami and Sadhu, 2021b; Randle-Boggis et al., 2020; Spencer and Barnes, 2020).
As such, the synthesised experimental results in Chapter 5 corroborate the problem addressed by this
theoretical contribution of this research, while explaining and justifying the value of the improvements offered
by the 3E framework postulated by this thesis. In this context, the reader is reminded that the experimental
results in Experiments 2&3 for the GPV counterpart have been synthesised with a conventional open-loop
PVlib model for ground-mounted PV configurations. At the same time, the closed-loop FPV model has
been used to synthesise the comparative floatovoltaic behaviour according to the thesis closed-loop 3E
framework postulations. Therefore, for a PV system of the exact same specification, the comparative syn-
thesis from the two dichotomously different FPV/GPV frameworks/toolsets signifies the mismatch between
the conventional GPV performance model outputs compared to that from the new FPV model. Even for
the exact same PV equipment, the conventional open-loop GPV model’s assessment fails to account for
the broad and diversified spectrum of performances and impact effects of a PV system floating on water.
In this sense, the convention GPV performance model predictions are illustrative of conventional tooling
vulnerability and its inability to handle the complexity and diversified range of floating PV performances and
impacts if used blindly to assess a floating PV project.
Offering a distinction between synthesised performances for the FPV/GPV performance modelling vari-
ants, the synthesised results for the GPV system in Figures 5.6 to 5.10 and Figures 5.13 and 5.15 therefore
not only exhibit the distinctive FPV performance benefits and diverse set of 3E performance gains of the FPV
technology variant (compared to its GPV counterpart), but also serves as an illustration of the performance
prediction errors that can generally be expected when using a conventional open-loop GPV performance
analysis model in the synthesis of an FPV installation project (refer to (Cagle et al., 2020; Kumar et al.,
2022)). A comparison between the synthesised results for GPV and FPV systems is, as such, indicative
of the limitations of current performance assessment methodologies in their inability to assess the sus-
tainability of floatovoltaic technologies. The comparative FPV model results provide scientific evidence of
this thesis’ PV system modelling improvements. On a strategic research level, the distinctive differences in
the performance synthesis from the innately different FPV and GPV performance models reflect the deep-
rooted critical knowledge gaps that this thesis had to fill towards the development of an inherently different
technology-specific framework, model and methodology for assessing the integrated 3E sustainability status
of floatovoltaic projects.
Therefore, while the literature review studied the underpinning best practices in existing state-of-the-art
254
GPV models that drive most present-day techno-economic PV performance modelling tools, it questioned
the effectiveness of existing conventional methods and prior-art paradigms in providing inclusive answers
to real-world analytical and modelling problems. The comparative results from the FPV and GPV models
in Figures 5.6 to 5.10 and Figures 5.13 and 5.15 essentially serves as confirming evidence of the correct
interpretation of the literature review (refer to Section 2.6), as the results confirm how the de facto linear or
open-loop framework (Figures 2.18 and 3.9), fail to provide an adequate theoretical foundation for assessing
floatovoltaic technology sustainability landscape. The GPV synthesis results as such serves as quantita-
tive scientific evidence of how that standardised metrics for GPV models/projects can further not properly
account for FPV technology’s complex range of distinct and diverse range of resource-use-efficiencies and
impact-effect-positives (the FPV’s "technology unknowns”, refer to (Armstrong et al., 2020; Cagle et al.,
2020)). Because of the structural imbalances in the conventional framework model, which signifies the set
of currently standardised photovoltaic appraisal metrics, the traditional model’s outputs cannot align with or
account for the precarious performance profile of floating PV systems.
Giving credit to the thesis’s postulated contributions, the same synthesised results in Figures 5.6 to
5.10 and Figures 5.13 and 5.15 by contrast signifies the ability of the holistic systems-based approach of
the posited theoretical 3E framework theory remediates core imperfections in current tooling models. The
results provide empirical evidence of how the posited theory secures recourse against the foundational de-
ficiencies left by the conventional serial-type open-loop processing framework approaches. It also confirms
how the extended range of performance and impact metrics formulated by this thesis can uniquely account
for the exceptional scope and hierarchical layering in the performance and impact benefits delivered by
waterborne floatovoltaic. Since floating PV deployment barriers are at present generally caused by a lack
of reliable and consistent evidence-based scientific appraisals in regulatory project permissions mandated
by law (Hernandez et al., 2019; Spencer and Barnes, 2020), the significance of the experimental results of
Figures 5.6 to 5.10 and Figures 5.13 and 5.15 derived from the theoretically-based FPV modelling improve-
ments offered by this thesis is self-evident. The posited theory thus provides an edge over conventional
analytical frameworks for PV project assessments and state-of-art sustainability assessment practices, de-
livering improved results beyond the traditional technical evaluation practices of performance assessment
and integrated environmental assessments.
6.4.14. Effective EIA assessments in environmentally-based FPV sustainability
Conventional GPV assessment tools create a practical predicament for impact practitioners because the
results contrast the impact effects expected for FPV systems. It inspired the development of the 3E frame-
work to support computer-aided environmental impact assessments (EIA) in an improved customer adop-
tion model dedicated to project validation practices under carbon tax legislation, policy and protocols (RSA,
2019a; Siweya, 2021). With a range of FPV "technology unknowns” being the most significant barrier to
floatovoltaic deployments and approvals worldwide, scientists are trying to tackle the research question
of quantifying the enormous technical and environmental performance potential of floating solar projects
from an EIA assessment standpoint (Gonzalez Sanchez et al., 2021; Hernandez et al., 2019; Spencer and
Barnes, 2020). However, conducting an EIA assessment for floating PV projects is a laborious, compli-
cated, time-consuming and computationally intensive process. Nevertheless, computer-aided EIA support
through FPV-GIS principles can dictate analytical policy to help simplify EIA process flow in carbon taxation
assessment protocols, environmental offset analysis, and hydro-carbon footprinting analysis. Such com-
pliance analysis and assessment in project validation practices with the help of the computer-driven EIA
assessments robotically conducted by the FPV model (semi-intelligent EIA process automation).
With the development of a new geographical tool for the integrated assessment of the feasibility, ef-
ficiency and sustainability of floating PV technologies, this thesis’ toolset and application considerations
255
improve the automation and authenticity of EIA results (refer to FPV/GPV environmental impact effect com-
parisons in Figure 5.9). The results in Figure U.2 also illustrate how the FPV model outputs meet the prereq-
uisite requirements of governmental policies and initiatives (refer to (Grand View Research, 2020; Hernan-
dez et al., 2019; Prinsloo, 2019)). In support of environmental project plans per government legislation and
regulatory requirements, the thesis’ postulation of an integrative theoretical framework recognises technical
energy, economic and environmental (3E) alliance responses of floating PV technology as key agricultural
FPV sustainability imperatives. As such, the consummation of the 3E coalesce through 3E causal system
dynamics established the drivers of accurate predictions of FPV project impact assessments. The thesis
can therefore advance the following EIA-type Environmental Impact Assessment Contribution statement to
emphasise the significance of the research in terms of automated EIA support:
Environmental Impact Assessment Contribution
The thesis contributes a new Environmental Impact Assessment (EIA) supporting conceptualisation
for farming energy generation systems based on integrating environmental understanding into the
unified energy, economic and environmental sustainability assessment analysis. The EIA supporting
capabilities of the FPV-GIS model is evidenced in the experimental results shown in Table 5.5, in the
graphs of Figures 5.9 and 5.13 as well as in the EIA reporting scorecard shown in Figure U.2.
This holistic EEE or 3E-modelling framework is the advancing principle in offering a holistic 3E systems
methodology in establishing an improved 3E analytical sustainability modelling hallmark for broad-spectrum
floatovoltaic ecosystem valuations. The 3E framing inherently includes prescriptive EIA-supporting Envi-
ronmental Impact and Offset Analysis, as illustrated in the environmental footprint impact results for FPV
and GPV in Figure 5.9. The postulated sustainability framework founded on the 3E pillars of performance
sustainability permits improved assessments for floating PV by delivering a broader and more diverse set of
sustainability metrics and indicators. These metrics and indicators align closely with EIA reporting require-
ments, thus establishing a foundation for electronically-aided EIA reporting and comparisons with science-
based emissions, carbon accounting and impact targets. It includes EIA compatibility with the WELF met-
rics and indicators used to measure agricultural-ecosystem and WELF-nexus sustainability in agricultural
resource-based methodologies. With the potential worldwide implementation of new EIA directives, proce-
dures and protocols towards the standardisation of floating photovoltaic system installation assessments
(Cohen and Hogan, 2018; Norton Rose Fulbright, 2021), the 3E assessment framework may support the
worldwide uniformity of FPV project reviews as a potential code of practice standard or paving the way for
the potential regularisation of floating PV project assessments around the globe.
6.4.15. Closing the circle between FPV-model outputs and EIA reporting
One of the principal motivations for developing the digital FPV-GIS model was to lay the groundwork to sup-
port EIA reporting quantitatively. To this end, the model parameter outputs can be applied to EIA reporting
in future planned FPV projects. EIA footprint reporting has been part of the discussions in the experi-
mental conclusions and critical findings of the case study in Section 5.4.4, as well as the strategic FPV
project assessment conclusions in Section 5.5.4.2. With this information available, this section highlights
the link between the FPV-GIS model outputs, environmental audit requirements, climate-related information
disclosure and EIA scoping report parameters portrayed in Figure U.2.
From an EIA perspective, the results in Figures 5.8 to 5.10 complete the circle between the FPV-GIS
model proposed by this thesis and the challenges of required EIA assessment. In this context, the the-
sis engages the confluence of the energy, environmental and economic parameter streams to exploit the
symbiosis between the 3E and EIA frameworks. A vital aspect of the experimental results in Figures 5.8 to
256
5.10 is the predicted lifetime 3E-performances and impact effects of FPV technology. These climate data
parameters can serve as a valuable data resource that directly applies to environmental due-diligence in-
vestigations, carbon accounting ledgers, sequestered carbon certifications and EIA-based project-licensing
authorisations. To this end, the lifetime 3E-performances and impact effects of FPV technology assessment
of the empirical input to each activity are shown in the sample EIA scoping assessment report of Figure U.2.
It is essential to understand the structure of the EIA scoping report in Figure U.2 before explaining how
the FPV-GIS model outputs provide input to each EIA report activity. The EIA scoping report consists of
three sections, namely (a) the energy project operations phase EIA report summary, plus two energy project
activity sections, (b) the parameters for the proposed main energy project (Water-mounted Floating Pho-
tovoltaic System), and (c) the proposed energy project alternative (Ground-mounted Photovoltaic System).
To complete Section a of the EIA scoping report in Figure U.2, the technical energy system parameters and
capacity, the FPV-GIS model outputs in Figure 5.8 provide the relevant energy generation capacity param-
eters. Sections b & c of the EIA scoping report in Figure U.2 each consist of three subtopics, namely (i) the
space and land transformation parameters, (ii) the environmental and air quality impacts, and (iii) the natural
water resource impacts. To complete Section b of the EIA scoping report, the environmental impact dash-
board in Figure 5.9 is of particular importance, as it quantitatively details the environmental-offset profile of
the operational lifecycle for the floatovoltaic installation (FPV system). Similarly, to complete Section c of
the EIA scoping report, the environmental impact dashboard in Figure 5.9 details the environmental-offset
profile of the operational lifecycle for the ground-mounted PV installation as a PV project alternative.
The FPV-GIS model’s empirical results in Figures 5.8 and 5.9 thus provide most of the scientific ev-
idence required to document the official EIA-related impact effects for the proposed FPV project and its
GPV counterpart as given in Figure U.2. While the experimental results report on the integrated 3E-project
outcomes and systems-performance/impact profiles for the FPV system variant predicted by the FPV-GIS
toolkit, the FPV model also offers the results for a so-called EIA alternative (see Alternatives Development
in the EIA process, Figure 2.7b). To this end, the model presents the comparative profiles for the lifetime
performance characteristics of an equivalent GPV counterpart as an EIA alternative for the exact site loca-
tion and setting. In this regard, Figure U.2 serves as evidence that the thesis addresses the problems and
purpose of the study as discussed in Section 1.2).
Within the context of the above discussion, the thesis can advance the following Automated EIA Scoping
Report Generation Contribution statement to emphasise the significance of the research in terms of the
scientific evidence generated by the FPV-GIS model to support EIA reporting documentation:
Automated EIA Scoping Report Generation Contribution
The 3E sustainability theory of this thesis supports the completion of EIA and EIA scoping documen-
tation to overcome one of the significant current limitations in floating PV project licensing approvals.
The EIA supporting capabilities of the FPV-GIS model is evidenced in particular in using the exper-
imental energy results in Figure 5.8 and the environmental impact results in Figure 5.9 towards the
completion of the EIA scoping report given by Figure U.2.
In this automated EIA reporting support context, Figure U.2 shows that the thesis delivers a valuable
model for computer-assisted EIA project scoping assistance. It further offers this capacity in a decision-
supporting geoinformatics platform in Figure M.1 that embeds integrated analytical and decision-supporting
models in an FPV-GIS toolset. The cascaded results portrayed by Figures 5.8 to 5.10 define the perfor-
mance signatures for FPV and GPV, while the resource-based climate-economic-WELF sustainability pro-
files for FPV and GPV systems portrayed in Figure 5.13 and Figure 5.15 inherently serve the EIA mission
and underscore the value of the integrated assessment framework of this thesis as a theoretical basis for
257
computer-aided EIA decision support for reporting on carbon taxation protocols, solar tax incentives, envi-
ronmental offset analysis, and hydro-carbon foot-printing assessments in project assessment and validation
practices.
6.4.16. Defending FPV sustainability in classical sustainability theory context
Strategically, the thesis can advance strong arguments to motivate the move from the internationally ac-
cepted framework for sustainability to that of the 3E framework for agricultural energy system sustainability
posited by this thesis. It asks how the posited framework aligns with the international norm for concepts,
methodologies and knowledge management approaches in sustainability science. Most scholars empha-
sise that the traditional three-pillar conception of sustainability, described in terms of classical social, eco-
nomic and environmental criteria and outcomes (triple bottom line), has become the universal international
standard for sustainable development and sustainability governance (Purvis, 2019; Sala et al., 2015). This
classical sustainability conceptualisation is defined in terms of the typical three "P” relationship pillars (3P
= People-Profit-Planet pillars) (Gbejewoh et al., 2021). This philosophy underscores the classical thinking
behind the relationships in the 3P sustainability definition, whereby it facilitates the study of human civilisa-
tion taking resources from nature to sustain a modern way of lifestyle, while natural systems at the same
time function in response by attempting to help the ecology remain in balance (DFFE, 2008; Wainwright
and Mulligan, 2013).
As part of its support for socioeconomic and environmentally-friendly industrialisation initiatives, the
US Environmental Protection Agency’s (EPA) Science and Technology for Sustainability Program sanctions
broader insights into the conceptualisation of sustainability (EPA, 2022; NRC, 2013). Therefore, the EPA
encourages new ideas and theories for sustainability to accommodate contextual variations to the charac-
terisation of sustainable development as support of innovative technogenic developments (of which floato-
voltaic technology is a good example). In this context, the EPA defines two baseline foundations for newly
proposed sustainable development frameworks/theories, namely: (a) the requirement to emphasize an un-
derstanding of mutual environmental interaction towards environmental justice while at the same time (b)
emphasizing and enhancing the connections among energy-water-land-food resource vulnerabilities as a
means of making sustainability operational (National Research Council, 2011; NRC, 2013). Crucially impor-
tant to note is the role of the water-energy-land-food nexus, as the leading framing for resource management
policies is crucial in the EPA’s understanding of sustainability because the EPA considers the WELF nexus
as the golden thread that connects all of the Sustainable Development Goals (SDGs) in Figure 2.2 (Bazzana
et al., 2022; UN, 2022). In this way, the WELF nexus is the key that directly supports social development
goals in development projects towards improved economic health and vitality (NRC, 2013; Raviv et al.,
2022). In working towards making a visible difference in communities through the SDGs, the EPA framing
of sustainability permits practitioners and researchers to define novel actionable frameworks and context-
specific sustainability concepts in ways that are compliant and consistent with the international sustainable
development goals through environmental impact and natural resource implications. This encouragement
of the EPA includes making sustainability frameworks operational through developing new tools and indi-
cators for building resilient and sustainable development paradigms (Bazzana et al., 2022), as a means
to better engage sustainable development stakeholders through modern paradigms such as the fourth in-
dustrial revolution philosophy. In this context, the model development protocols of the EPA, portrayed in
the methodological modelling process framework shown earlier in Figure 3.17, offer a practice-orientated
extension of the sustainable development supporting goals of the EPA (EPA, 2022).
Strategically, the EPA frame of reference encourages scientists to cultivate new ideas around sustain-
ability and explore novel approaches to help achieve environmental management goals and processes
in innovative theoretical frameworks that can justify attaining maximum benefits of (project) decisions. In
258
particular, the EPA standpoint on sustainability permits new approaches that contribute to the sustainabil-
ity discourse through new theoretical frameworks that recognise the natural resource interactions and the
need for ecosystem thinking to help achieve resilient project outcomes (NRC, 2013). In the absence of
a solid theoretical conception to measure floating PV sustainability (refer Section 2.5 and 2.6), this thesis
puts an environmental management goal and sustainability assessment process in place to help ensure
that the installation of floating PV systems supports sustainable agricultural development goals. The study
succeeds in identifying the 3E pillars as the genesis and theoretical foundations of a new sustainability
conception for FPV technology that offers agricultural goal-based foci on floating PV technology outcomes
(refer Figure 3.14). Emphasising the mutual dependencies of the roots-driven 3E operational elements,
the 3E framework is consolidated around the environmental WELF resource nexus to close the loop be-
tween energy production and environmental interactions while increasing the overall resource efficiency
on the FPV ecosystem level. The proposed 3E sustainability framing further supports the understanding
that environmental resources are inextricably intertwined with the sustainable energy and economic crite-
ria dimensions, thus helping to decrease environmental risks and resource scarcities while supporting the
SDGs as conditions to aid sustainable development and environmental management. These aspects of
sustainability framework formulation align with the EPA requirements in built-for-purpose sustainable devel-
opment framework guidelines (EPA, 2022). In this context, the thesis documents the following Sustainability
Theory-Building Methodology Contribution statement:
Sustainability Theory-Building Methodology Contribution
The thesis contributes to creating new knowledge by developing a sustainability theory-building
methodology, which delivered a novel 3E sustainability theory for FPV that meets EPA requirements
regarding mutual environmental interaction and emphasises the WELF resource vulnerabilities. The
floating PV performance assessments in the energy, environmental and economic sustainability di-
mensions are evidenced in the experiments throughout Figures 5.8 to 5.10.
The EPA’s sustainability interpretations were the backbone of the philosophy of instituting a sustainability
framework for floating PV technologies through the philosophical arguments and motivations in Chapter 3.
To this end, Figure 3.6 emphasised how the EPA (2022) defines the four stages of new model developments
in a four-stage life-cycle. The EPA process thus guided the floating PV sustainability assessment framework
and model development process based on the EPA steps of taking the conceptual understanding of an en-
vironmental process to a fully analytical model in terms of problem identification, framework development,
model development, and model application in practice. Thus engaging the EPA modelling philosophy in
the model implementation process (shown in Figure 3.17), the EPA steps guided model development and
testing through the steps of: (a) studying FPV sustainability as the reality of interest; (b) defining a concep-
tual systems model for FPV sustainability as computer logic; (c) implementing a computerised sustainability
model (Schlesinger, 1979); and (d) evaluating the conceptual model framework for sustainability.
Therefore, taking guidance from the EPA’s sustainability interpretations of the classical 3P criteria for
sustainability, the thesis offers a theoretically rigorous description of PV energy project sustainability as a
design-for-purpose PV technology sustainability theory in terms of the three E pillars. At the same time,
in line with the EPA guide, the conceptualised 3E framework and process methodology recognise water-
energy-land-food nexus interactions in an agricultural application context (refer Figure 3.15).
6.4.17. Reflecting on role of 3E sustainability vs classical sustainability theory
This section reflects on the possible constraints and opportunities of floatovoltaic technologies from a social
or ecological point of view. These aspects can be added to the 3E framing to meet the standardised or clas-
259
sical sustainability definition requirements. In this context, it also reflects on the harmony between the 3E
framework and classical sustainability theory and the opportunity to extend the proposed 3E sustainability
framing to incorporate aspects of classical sustainability assessments.
The availability of data on the field performances of floatovoltaic technology development projects is
currently deficient due to the novelty of the technology and the immaturity of the techniques and processes
to predict system performances and impact effects (Kumar et al., 2022; NREL, 2019a). This is especially
true for data and publications on the long-term social and ecological impacts of deploying FPV power plants
on freshwater reservoirs in local environments, which is poorly known (Acharya and Devraj, 2019). To prove
this point, Bax et al. (2022) recently attempted to explore the social feasibility of a floating solar energy
PV pilot project in the Netherlands, upon which stakeholders expressed concerns that the deployment
of FPV remains subject to many unknowns and uncertainties. While Armstrong et al. (2020) provided a
valuable framework to study the ecological impacts of floating PV systems, studies to quantify the ecological
impacts of floating photovoltaic plants are mostly in their infant stages (Grodsky et al., 2022), and more often
conducted in the context of hydropower plants (Haas et al., 2020). The lack of long-term data and academic
studies on the social and ecological impacts of FPV in South Africa makes it extremely difficult for the
thesis to scientifically reflect on the risks of ecological damage or cultural acceptance of floatovoltaics. The
same argument holds for reflecting on the possible constraints and opportunities of floatovoltaic technology
deployments from a social or ecological point of view (refer to Research Assumption 4 in Section 1.5).
To study the human-environmental interaction of FPV, the thesis explicitly defined the quantification
of social and ecological sustainability impacts of floatovoltaic systems as important directions for future
research. In this context, the thesis offers new downstream research opportunities into the extension of
the 3E sustainability framing towards the incorporation of aspects of the classical sustainability definition
into future FPV project assessments. While the classic definition and form of sustainability are defined in
terms of the social, economic and ecological criteria and outcomes (triple bottom line) (Purvis, 2019; Sala
et al., 2015), this definition remains at the centre of the broader geographical and theoretical sustainability
discussion. The future issue thus centres around integrating the thesis’s 3E framework into the theoretical
framework of classical sustainability. By adding social and ecological features to the 3E assessment for
FPV systems, the 3E assessment effectively supports the traditional three-pillar conception of sustainability
(refer to the model extension of Figure V.1 in Directions for Future Research, Section 6.7).
By integrating the 3E assessment into the classical sustainability triangle, this conventional social, eco-
nomic and ecological criteria (triple bottom line) remain the universal international standard for sustainable
development and sustainability governance. From a wider geographical perspective, the 3E/TEE sustain-
ability theory proposed by this thesis thus offers an alternative route to / preparing for the full spectrum
sustainability assessment for PV energy systems. The 3E framework thus paves the way towards a more
extended definition of PV system sustainability whereby the results of the 3E synthesis could, in the future,
be extended to include the social and ecological aspects associated with the standard/traditional definition
of sustainability in a Social, Ecological, Energy, Environmental, Economic (SE-3E) sustainability definition.
Important to note that the classical sustainability definition rests on the interpretation and context-
specific understanding of the impact of humans on planet Earth (NRC, 2013; Purvis, 2019). This is also the
case for evaluating mutually beneficial renewable energy systems (Burke, 2018). In the context of floating
PV power system installations, FPV technology is known to offer unique conservation attributes that are con-
sidered to be "giving back” to nature (reclamation of water, energy, and land as offsets, each associated with
ecological impacts), while nature in turn "contributes” to attenuate technical, natural resource and economic
efficiency to help the technology improve project profitability (energy efficiency, environmental offsets). On
the philosophical level, as discussed under Section 2.3, the derivative technology-environment-economic
relationships in the case of floating PV thus help to resemble more of a two-way relationship (geographi-
cally dependent bilateral give-and-take relationship). As such, the 3E framing effectively supports aspects
260
of the classical sustainability theory, a relationship that should be characterised by a broader framework
that accommodates the distinct and diversified set of circular-type feedback effects between these three
operational elements. In such a context, the 3E framing helps the classical 3P conceptualisation to cover all
the bases as a reference regulatory framework for sustainability. It means an improved theoretical frame-
work can be offered to support the practical operationalisation of floating PV project 3E assessment into the
broader operational enterprise or ecosystemic business context. As such, the thesis contributes by prepar-
ing the way for incorporating the proposed 3E and WELF resource metrics into Classical Sustainability
Assessments, as reflected in the statement below:
Preparing the Way for Classical Sustainability Assessments
While the thesis’s 3E sustainability theory and WELF resource indicators for FPV projects meet EPA
requirements regarding mutual environmental interaction, the 3E sustainability theory helps to link
the SDG goals by making provision for future integration of social and ecological impact options to
dovetail with the classical 3P sustainability criteria. The sustainability theory and modelling frame-
work posited by the thesis, as evidenced in Figures E.1 and M.1, can thus be extended through
modelling additions depicted in Figure V.1 to meet classical 3P sustainability criteria requirements.
In this context, the postulated 3E sustainability conceptualisation for floating PV technology as such
arises from a broadly different school of thought that aligns with the EPA’s view of technogenic sustainability
rather than the conventional three P pillars (People, Profit, Planet=3P) that predate the 3E sustainability
objectives. The absence of "people development” and "ecological impact” as sustainability pillars in the
proposed 3E theoretical conception for floating PV assessments is by no means limiting the theoretical
operationalisation of 3E sustainability. In the classical sustainability sense, the "social development” and
"ecological” dimensions still have a role to play as the fourth and fifth pillars of sustainability under the
sustainable development jurisdiction (as proposed future research topic in Section 6.7). At the same time,
in line with the EPA’s understanding of social sustainability, consideration for the WELF nexus resources
in the 3E sustainability theory for agri-based energy systems further helps to connect the various SDGs
shown in Figure 2.2 and by implication, can cover the socio-technical and ecological-technical dimensions
to support justify the requirements for classical sustainability (refer (Bazzana et al., 2022; Jackson, 2010)).
While multi-criteria sustainability and environmental impact assessments would benefit significantly from
ecological impact modelling, these impacts are, to a large extent, contextually sensitive, depending on the
regions and subregions within a country or region. The same argument goes for social impact studies and
for studying the possible constraints and opportunities for floatovoltaic technologies from a social point of
view. To overcome the context-sensitivity barriers from an ecological (and social) perspective, a substantial
body of novel FPV system impact research would thus need to be conducted beforehand as a prerequisite
to modelling sensitive marine and aquatic factors in ecological subsystems in different subregions within
South Africa. The same argument holds for impacts on global citizenship, social capital and human beings,
all of which are the primary focus areas of corporate social responsibility and social acceptability of FPV
technology within the social impact space. While these aspects are currently beyond the scope of this thesis,
the study opted to include such aspects and the measures of classic sustainability as potential directions
for future research (see options in Section 6.7).
6.4.18. Unexpected contribution in support of ground-mounted photovoltaics
Taking a step back, it was realised that the built-for-purpose technology-specific 3E sustainability integrity
valuation theory developed by the thesis for floating PV technologies equally offers a sound theoretical value
proposition for the sustainability valuation of ground-mounted PV technologies. This unexpected realisation
261
can have significant academic utility implications regarding the improved modelling and assessments of
GPV system sustainability. In principle, this implies that the closed-loop 3E theoretical modelling framework
can apply to GPV assessments as it adds environmental variables to the project analysis mix (Figure 3.15),
while at the same time overcoming severe structural imbalances existing in the open-loop modelling frame-
works of existing GPV systems (Figure 3.9). Logically the conclusion makes sense, as the GPV system
also operates within the natural environmental system and also benefits from climate change economics,
although not benefiting to the same extent as floating PV technologies. In this context, the thesis documents
the following GPV Sustainability Assessment Methodology Contribution statement:
Sustainability Assessment Methodology for GPV Contribution
The thesis contributes new knowledge to the integrated performance/sustainability assessment of
any agricultural photovoltaic system, land-based or water-based, for which the sustainability profile
for GPV determined by the 3E sustainability theory would map a less impressive sustainability profile
specific to that of a GPV system. The GPV-specific environmental and economic impact effect pro-
files are evidenced in the experimental results for GPV shown in Figures 5.9 and 5.10 respectively.
The sustainability theory, framework, methodology, and technique posited by this thesis can, as such,
serve as a sustainability assessment strategy to enhance the sustainability valuation of ground-mounted PV
technologies. All that needs to change to accommodate GPV sustainability assessments is that the impact
practitioner must configure the FPV-GIS model with the parameters for the GPV system under consideration.
This FPV/GPV system sustainability compatibility realisation incidentally came when the FPV-GIS model
was programmed so that the user-configuration file catered to the user’s configuration of both FPV and
GPV systems. It led to the realisation that the GPV system counterpart is just a more basic or rudimentary
configuration of the slightly more complex FPV system configuration. In this context, this thesis’s holistic 3E
theoretical conceptualisation inherently creates the same virtual-reality environment for the sustainability
valuation of GPV systems, just with a different (non-floating) PV design configuration.
On a meta-level, a genomics analogy can explain the realisation of the FPV variant using a DNA
metaphor. Anatomical genetic codes for the DNA of sustainability in the case of the FPV technology pheno-
types seem to have a more complex DNA code structure than that of the GPV system phenotype (different
DNA strand or opposite orientation). The sustainability DNA alphabet, therefore, needed to be extended
to include environmental and environmental economic letters in a more advanced sustainability DNA se-
quencing tool to ensure improved decryption of the FPV sustainability DNA based on the new theoretical
conceptualisation of the sustainability DNA framework. With the extended DNA alphabet, the sustainability
DNA strain of the GPV phenotype can still be characterised or sequenced by the same DNA sequencing
tool developed for the FPV system variant. However, the sustainability DNA for GPV would show a less im-
pressive DNA sustainability strain as an outcome (refer for example Figure 5.15). In this figurative context,
the theoretical contribution of this thesis thus creates broader research opportunities for investigating the
incorporation of the 3E theoretical framework model in conventional PVlib and PV system-type quantitative
performance modelling methodologies, especially in terms of the potential conditioning of conventional PV
performance models based on the theoretical 3E framework and microhabitat modelling procedures dis-
closed in this thesis. Regarding recommendations, the above ideas can inspire future academic contribu-
tions to the sustainable development discourse in water-based PV and land-based photovoltaic technology
modelling.
262
6.4.19. International relevance and application potential for the research
While sustainability is a concept with international relevance, the thesis’s EIA-inspired research offers the
generic-type sustainability valuation strategy for holistic floatovoltaic propositions as part of a uniform global
sustainability strategy. Serving as a climate partner in providing climate action solutions, the 3E project
assessment framework theory and model can form the basis of an internationally relevant balanced EIA
electronic scorecard approach to ensure broader global relevance. As an improved framework for the per-
formance assessment of FPV as an agricultural energy technology system, the proposed 3E assessment
framework uniquely amalgamates the coalescence of energy, economic and environmental (3E) funda-
mentals as whole-system pillars of sustainability. Furthermore, the previous section discussed how the 3E
assessment theory for floating PV could be extrapolated to the analysis of ground-mounted photovoltaics.
While the 3E framework dovetails with the international environmental and natural resource-oriented agri-
cultural sustainability standards, the 3E assessment theory for FPV certainly has international relevance
and application potential as a floating PV system sustainability valuation strategy.
While the universal Sandia reference model is technically generic (function of PV panel characteristics
and meteorological data), the FPV-GIS model is correspondingly generic by nature (function of the instal-
lation’s spatial attributes, landscape character, and meteorological data defined in the model’s client/user
configuration file). When given the relevant meteorological data, the FPV-GIS model’s 3E and ce-WELF
data forecasts are universally applicable and suitable for FPV assessment applications worldwide. As a re-
sponsive model, the FPV-GIS model automatically adapts to the environmental conditions associated with
the location-specific meteorological weather station data the model is driven with at runtime. In the context
of international experiments (US, Canada, Germany, Spain, etc.), the meteorological data from weather
stations around the experimental FPV installation specified in the user-configuration file inherently aligns
with the environmental and economic conditions for any proposed installation site around the globe.
The data science technique’s definitional workspace, provisioned by the model’s client configuration file
of the digital twin simulation engine, therefore gives impact practitioners and project developers worldwide
control and flexibility over how the application is configured to run. The flexible software configuration file
permits FPV project template processing for most countries in the world, by allowing for flexible design
architecture through reconfigurable project parameter specifications. In this context, the thesis documents
the following FPV Sustainability Model Flexibility Contribution statement:
Sustainability Model Flexibility Contribution
The thesis contributes a flexible floating PV sustainability analysis and assessment model that
can be user-configured to perform the integrated performance/sustainability assessment task for
agricultural-type floatovoltaic systems based on geo-intelligence and meteorological data from coun-
tries over the world. International relevance is ensured through the FPV-GIS synthesis model’s user-
programmable configuration file, which is suitable for the location and country-specific PV system
installations parameters as evidenced in Table O.1.
The posited theoretical principle, framework, model methodology and technique thesis ensure interna-
tional technological relevance and application potential, as it constantly focuses on context sensitivity as
a critical hallmark of global significance. Thanks to the model’s reconfiguration capabilities, the proposed
prototype dynamical (eco)system simulation model provides flexible system specification options for the
desired design reconfiguration of the floatovoltaic (eco)system plans. Therefore, global resilience in func-
tionality has been ensured at the design stage, wherein the adaptive capacity of the FPV-GIS model ensures
broader international relevance (i.e., country-specific funding mechanisms, incentive schemes, fiscal poli-
cies, carbon taxes, technology types, environmental conditions, weather data, etc.). As such, the thesis
263
successfully created new and generic academic knowledge with international practical application oppor-
tunities for further development or application. Furthermore, by offering novel philosophical EIA ideas in a
proposed electronic meta-modelling concept, the geographical information system can support agricultural
"solar over water” projects and floating PV technology diffusion in the local and international arena.
This section provided a detailed summary of the strategic conclusions and recommendations of the
research in terms of the various aspects of novel contributions of the thesis in terms of the topological
and ontological architecture behind the philosophical framework of the theory-building research, the posited
theoretical principle, the appropriation of the principle in a new assessment framework and methodology,
as well as the establishment of a digital twin model and data science technique to quantitatively derive
experimental results from which could infer conclusions around the operational, tactical and strategic value
of the contributions. The following section briefly summarises these contributions in sequence, expressing
an opinion on the significance of these academic research contributions from a geoinformatics perspective.
6.5. Summary of Novel Contributions of the Study
In terms of making a meaningful novel academic contribution within the context of the geographical infor-
mation science field, this research investigation began exploring opportunities to support the environmen-
tally conscious knowledge economy by making original contributions through the geo-informatics discipline.
Drawing valuable lessons from field observations in addressing strategic deficiencies in the modelling and
characterisation of new floating PV technology systems, the research has been set up structurally in antici-
pation of creating new knowledge in this operational field, discipline and related practices.
6.5.1. Contextualisation of novel contributions of the study
Expressions around the novelty and originality of academic contributions in this section take cues from the
work of Baptista et al. (2015), who defines originality in doctoral studies in terms of its relationship with
creativity and innovation. According to this definition, the concept of creativity is introduced as a focus on
the production of knowledge, while the concept of innovation centres around the relevance of the research
for application, economic, or legal purposes in a knowledge-based economy context.
Helping to contextualise the original academic contributions and creation of new knowledge of this re-
search investigation, it is essential to note that the thesis’ quantitative research effort has followed a unique
geographical operations research approach to build and extend the research work of other researchers. In
this sense, the study and FPV model representation build onto the conventional theoretical concepts and
paradigms of traditional technical and techno-economic PV performance modelling efforts, as detailed un-
der State-of-the-Art in Floating PV Sustainability Profiling in Section 2.5. The study succeeds in creating
new knowledge through a new theoretical basis, an essential imperative because conventional models and
frameworks failed to provide an adequate integrated and contextual foundation for solving pressing reg-
ulatory and certification requirement challenges in the sustainability landscape of floatovoltaic technology
(Hernandez et al., 2020; Prinsloo, 2019, 2020). The thesis further contributes to current debates around
FPV sustainability as it helps to advance the knowledge frontier of the sustainable development discourse.
In this context, the geographical research investigation helps transform the prevailing understanding
of floatovoltaics by bringing together various research perspectives at multiple temporal, spatial and tech-
nical scales. It thus contributes to a better scientific understanding of floating PV system behaviour and
dynamics towards attaining an improved formulation and communicative articulation of FPV-system sus-
tainability. Consolidated in a sound 3E ensemble as theoretical contribution underpinning sustainability
logic, the posited sustainability theory enabled the thesis to offer layered levels of aggregated contributions
264
on the strategic-, tactical -and activity-layer levels. The theoretical contribution lies at the root of these con-
tributions as it sparked an aggregation of contributions to provide new insights into the systemic variables
and relationships between FPV ecosystem sustainability variables. It also generates actionable insights in
an operational system that supports evidence-based decision-making. Notably, the posited theory not only
explains "why” these relationships exist by also "how” model interactions impact integrated sustainability.
To elaborate on the layered levels of aggregated contributions, the integrated discrete-time FPV syn-
thesis model presented in Figure M.1 offers at least six layers or aspects of novelty. The first is on the
systemic level, where the sustainability theory holistically integrates the cohesive 3E domain ensemble as
object actors in an object-process topological architecture to ensure cascading effects processing in a sys-
tem dynamic ontology. In this circular-type system dynamics execution network, the second layer of novelty
lies on the model component level. On this level, the FPV simulation model develops the 3E domain ob-
jects from first theoretical mathematics and physics principles, further realising the operational 3E object
mechanics in a geo-engineering model-based design approach. In the third instance, on the processing
level, Figure M.1 portrays the parallel-type 3E domain and feedback data processing during each simula-
tion time-step, which contrasts the conventional siloed domain processing execution (refer to Figure 3.9).
The fourth contribution is on the ecosystem abstraction level, whereby the theory and model consolidate
the environmental functions under a single Environmental domain object. As conceptually illustrated in
Fig. 3.11, the framework concept strategically moves the solar-meteorological processing functions out of
the traditional energy model processing to consolidate all the meso-micro habitat climate, natural resource
processing, interactions, and feedback effects/functions under a single Environmental operational domain
object. The fifth novelty concerns the algorithmic modelling of the FPV microclimate, whereby the Environ-
mental model object in Figure M.1 performs reservoir climate parametrisation in a hydroclimatic submodel.
This computational micro-climate modelling submodel modulates the micro-scale hydroclimate signals onto
the meso-scale TMY climate signals loaded from the weather station geo sensors for the PV system GPV
vector (observational dataset from meteo data cloud repository). The sixth contribution lies at the WELF
resources level, in terms of which the WELF nexus is inherently seated at the intersection of the Energy-and-
Environmental domain objects, as illustrated in Figure E.1, allowing for the WEL indicators to be empirically
extracted from the metrics at this intersection to articulate the quantitative natural resource nexus indicators
on a multi-indicator profile mapping display.
Regarding the impact and significance of the conceptual and theoretical interventions to FPV’s sustain-
ability policy-related analyses, the posited theory and its downstream contributions provide better expla-
nations as it allows for more realistic predictions of the diverse and complex spectrum compared to field
installation outcomes. Thanks to the theory being operationalised in an analytical framework and numerical
simulation model, the theoretical contribution offers an improved understanding of the complexities of floa-
tovoltaics behaviour, performances, and impact effects. Furthermore, operationalising the posited theory in
a heuristic analytical hierarchy multi-objective decision-making paradigm extends the deterministic geospa-
tial digital twin synthesis model to determine an integrated sustainability rating for FPV projects based on
WELF nexus in a multicriteria sustainability classification.
6.5.2. Summary of novel contributions of the study
The novel contributions, discussed under strategic research conclusions in Section 6.4, are summarised
in the brief synopsis of the novel contributions provided below. The research investigation worked towards
a predictive floating PV characterisation methodology by tackling the real-world problem of integrated sus-
tainability assessments from an analytical operations research perspective. It saw a hypothetical change in
reframing the current basis in PV performance assessments by postulating a new quantitative sustainability
assessment framework for remodelling floating PV characterisation through a holistically integrated system
265
dynamics thinking and modelling process (refer to Figures 3.3 and 3.4). Based on a systematic review of
the relevant literature, the thesis theorises about the dynamical interplay between the energy, environmen-
tal and economic aspects of the floatovoltaic complex in operational narratives. Having formulated a novel
theory development methodology in Figure 3.8, the thesis engaged in advancement in existing theory by
developing the cognitive conceptualisations in Figures 3.11 to 3.13 as philosophical interpretations of the
gaps and opportunities in the theoretical characterisation of real-world floatovoltaics. These philosophical
conceptualisations emphasised the value of environmental understanding and environmental economics as
the basis for a new integrated theoretical foundation (Figure 3.14). The posited systems-thinking logic elu-
cidates the complex interactive relationships and collective identity within an agricultural energy ecosystem,
enabling it to imitate and explain the complex behaviour of floatovoltaic technology in real-world operating
contexts. Being able to explain the emergent behaviour and intrinsic value of floatovoltaics, the posited
theoretical logic can decode the broad and diverse spectrum of sustainability attributions offered by the
technology in real-world applications.
With this paradigm change, the thesis exploits consortium principles in the symbiosis between the en-
ergy, environmental and economical operations as an operational 3E ecosystem archetype. This pioneering
research motivates a novel system dynamics theoretical abstraction for sustainability, by conceptualising
and mobilising dynamic ontological insights from the broader ecosystem operations in the theoretical syn-
thesis of floatovoltaics. The theoretical framing, depicted in the sustainability triangle of Figure 3.14b, posits
an improved underpinning theoretical policy and logic for computer modelling of the floatovoltaic complex.
The theoretical framework serves as an underpinning theory in a unique floating PV sustainability modelling
policy and a methodological framework for developing a proposed integrated computer modelling expression
of the sustainability theory (Figure 4.3). Proof of concept is established predictive mathematical modelling
approach and digital tool for floating PV project sustainability assessments. In this tool, the theory serves
as computer logic in a floating photovoltaic emulator architecture based on systemically choreographed
energy, economics and environmental science analytics and synthesis principles. Operating as an FPV
energy information system, it establishes a novel geospatial digital twin for floatovoltaics. The digital twin
enables rapid prototyping and narrative evaluations in a pre-production environment as the model repli-
cates a floating PV powerplant design and operating conditions in the cyber-physical domain. The digital
twin allows users to perform customisable time-series analyses and mapping visualisation of floating PV
performance synopses under real-world operating conditions.
The FPV-twin computer animation model thus provided an FPV-GIS desktop tool as an analytics en-
gine for analysis-by-synthesis experimentation with future planned PV systems to be deployed in an aquatic
operation theatre. True to the philosophical goals for a systemic theoretical sustainability modelling con-
cept for the exploration of solar PV system performances in Figure 3.3, the theoretical conceptualisation
of Figure 3.14 is operationalised in a geospatial digital twin model of Figure M.1. It delivers time series
analyses in a computer synthesis model with a theoretical underpinning that integrates real-time behaviour
synthesis (Figure 5.7), performance profiling and evaluation (Figure 5.8), environmental offset assessment
(Figure 5.9), WELF resource impact assessment (Figure 5.13), economic assessment (Figure 5.10), and
integrated sustainability assessment (Figure 5.15). The FPV-GIS model’s performance projection capability
and empirical results thus provide the scientific evidence required to document the official EIA-related im-
pact effects for the proposed FPV project and its GPV counterpart as given in Figure U.2. With the capability
of predicting future performance, the geospatial digital twin model ensures system dynamics sustainabil-
ity assessments and virtual commissioning of floating PV projects in an integrated 3E and WELF nexus
value-chain analysis.
In principle, consolidated sustainability modelling synthesis is driven by the newly posited system dy-
namics sustainability theory postulated by this thesis. In 3E performance-based projections, the posited the-
ory maintains that economic and economical understanding are necessary and sufficient components/conditions
266
to characterise all sustainability signals in a full value chain sustainability analysis. It also maintains that the
sustainability performance of FPV as an agricultural energy system should be a multi-disciplinary objective
related to the agricultural sustainability goals of water, land, energy, and food production or savings. The
thesis believes that sustainability criteria and benchmarks are crucial for comparing the criteria dimensions
in sustainable system performance assessment and management. The study and posited contributions
thus ensure an enhanced understanding of floating PV systems by posing the philosophical idea of defining
a complex holistic systems thinking solution as an improved theoretical paradigm to bridge the real-world
knowledge gaps through a newly proposed philosophical system thinking and modelling approach. In this
context, the novel contributions of the study can briefly be summated as follows: (a) the conceptualisation of
a new theoretical sustainability principal; (b) the practical implementation of the posited theoretical principal
in an analytical computer model expression; (c) contextualisation of the theoretical principal in analytical
and decision support frameworks; (d) substantiation of theoretic principle through experimental data; (e)
providing an analytical account of theoretical principle in an exemplary case study; (f) providing a decision-
supporting discretionary account of the theoretical principle in an officiating case study; and (g) providing
evidence of new contributions to advance the state-of-the-art and furthering of existing knowledge. The
synthesis experiments testify to how the postulated theoretical principle is an underpinning foundation for
improving situational awareness and predictive understanding of photovoltaics through the quantification
of sustainability of FPV technology in a reflective assessment. The experimental results for this reflective
assessment are presented in real-time 3E portfolio snapshot timeline sequences (Figures 5.6 to 5.7) and
lifetime 3E integrated assessment articulations (Figures 5.8 to 5.10). The reflective assessment also pro-
vides a sound basis to account for natural resource impacts on the WELF nexus in Figures 5.13 and 5.15
as part of the research outcomes.
Judging from the outcomes delivered by the experimental results, this thesis makes meaningful and
relevant contributions to the body of international knowledge on sustainable development in floatovoltaic
systems in a discrete-time multi-methodology that ensures real-time geo-tracking of project performance
information. On a strategic level, the thesis succeeds in posing a philosophical idea of combining geody-
namical behaviour synthesis, performance evaluation, and impact assessment in a consolidated theoretical
framework for agricultural sustainability modelling framework for floatovoltaics through a geospatial digi-
tal twin model. In this context, the thesis uniquely delivers a unique discrete-time simulation model for
floatovoltaics because it simultaneously simulates the energy, environmental and economic operations in
an integrated model. To this end, the spatiotemporal processing steps in Figure 4.4 follow the reciprocal
workflow in parsing through the nested sub-model sequence of Figure K.1 during each simulation iteration.
This integrated systems dynamic perspective of this thesis’s modelling philosophy shown in Figure 4.2 is a
novel approach as it contrasts conventional models that perform processing in a linear silo-based process
according to the reductionist thinking approach of Figure 3.9.
The dynamic systems thinking sustainability theory, framework logic, computer model and analytical
technique in a geospatial digital twin enable the user to fast-forward the real-time simulation of floatovoltaic
operations to predict the broad-spectrum sustainability performances ahead of time quantitatively (on a
theoretical basis and ahead of installation). While comprehensive impact assessment in PV project analy-
sis tasks would traditionally be complicated, time-consuming and computationally intensive, the proposed
technique cuts down on time spent in complex floating PV. As a result, this technique can accelerate ana-
lytical PV project performance and impact assessments in a desktop environment. Moreover, the proposed
dynamical system interactive modelling methodology improves the theoretical floating PV model’s perfor-
mance profiling and prediction accuracy. The thesis’s populations help to overcome the current limitations
of environmental impact assessments of floatovoltaic project plans (i.e. methodological choices, perfor-
mance metrics, data transparency, etc.) that lead to poor comparability and inconsistency among reports
and studies, as it provides a more consistent framework for comparative project assessment approvals.
267
In terms of supporting future sustainable development scenarios through the production of new knowl-
edge, the thesis’ philosophical theory-building and design research advance the state-of-the-art in floating
photovoltaic performance modelling. Postulating a holistic sustainability strategy and theoretical framework
as fundamental imperatives in floating PV performance modelling breaks new research ground. The thesis
offers new theoretical foundations in an improved systems-based technique and theory-driven methodol-
ogy to study the geodynamical behaviour of floating PV system installations in a digital computer synthesis
modelling and digital prototyping environment. The proposed theory, methodology, simulation model and
technique advance the state-of-the-art as it offers an enhanced improvement over conventional photovoltaic
analysis techniques. The sustainability status or integrity of a floating photovoltaic project is characterised
distinctively in terms of its whole-project energy, economic and environmental sustainability pillars and its
whole-project water-energy-land-food resource-nexus sustainability pillars.
The results provide compelling scientific evidence that the thesis has advanced the state-of-the-art in
PV performance modelling and that the system dynamics sustainability theory and the FPV-GIS model
framework posited by this thesis outperform conventional PV performance models. In providing evidence
in this regard, the reader can compare the project outcomes predicted by the open-loop GPV model to
that of the closed-loop system dynamics FPV model in the experimental results posted in Figures 5.8 to
5.10. For the exact same PV equipment at the exact same location, the FPV-GIS model mathematically
projects more predictable project outcomes and comprehensively covers the broader diversified spectrum
of distinct sustainability attributions observed to be delivered by real-world floatovoltaic field installations.
The sustainability results offered by the theoretical contribution of this study contrast those obtained from
conventional PV performance models (the GPV results in Figures 5.8 to 5.10), in terms of which scientists
rightfully judge that conventional PV assessment models deliver dubious performance outcomes when it
comes to floatovoltaics. The project outcomes predicted by the conventional GPV performance model are
unreliable for floatovoltaics as it is considered full of holes and gaps (examples in prominent zeros or holes
in Figure 5.9). While these holes and gaps have become known as the so-called "technology unknowns”,
it is exactly these holes and gaps that have been filled and plugged by the 3E sustainability framework and
theory of this thesis, as evidenced in the open-loop GPV model predictions throughout the more realistic
energy, environmental and economics predictions for floatovoltaics shown in Figures 5.8 to 5.10.
The thesis in principle makes strategic and functional knowledge contributions by exploring and im-
proving a geographic-theoretic conception of a predictive ecosystem sustainability framework for floating
PV technology. It explores an improved theoretical postulate from the drivers of FPV system sustainability
postulation of a theory as academic contribution forms an integral part of creating new knowledge. The the-
oretical contribution of the thesis advances the state-of-the-art as it identifies the key drivers of floating PV
system performance and impact effects to provide a geographic-theoretic basis as an enhanced theoretical
principal and framework able to give rational scientific explanations for the distinct variance and diversity of
floating PV system performance and impact effects.
The research contributes to the sustainable development discourse through a sustainability theory-
building methodology by advancing fresh and novel philosophical ideas with theoretical principles that may
have far-reaching international implications for the development of floatovoltaic and ground-mounted PV
performance models worldwide. Strategically, in terms of the current worldwide challenges with the as-
sessment of the complex and diverse set of benefits and co-benefits offered by PV systems (Cohen and
Hogan, 2018; Gadzanku et al., 2021b; Norton Rose Fulbright, 2021), the philosophical research postulates
may further offer valuable policy-related opportunities for the potential regularisation of ground-mounted or
floating PV project assessments in the international arena. Even in the current regulatory space, regulatory
authorities require compelling scientific evidence of the anticipated project outcomes.
In line with the geographical science engagement of this thesis, the proposed theoretical framework and
computer model expression of the framework is GIS-layer integrated on a geo-informatics novel platform to
268
establish a new project analytical methodology for the predictive assessment and evaluation of agricultural
floatovoltaic installation project outcomes. Strategically, the thesis conceptualises and implements a new
data science technique for the holistic characterisation of floating PV installations in a context-sensitive
digital twin model based on environmental and climate change economics as sustainability principles. The
experimental results serve as evidence of the new knowledge created and how the theoretical principle
posed by this thesis relates to the improvement of current analytical frameworks and models used in the
assessment of PV system performances.
With the geographical rendition of a new theoretical formulation and holistic framework construct to
better describe the dynamics of floating PV system sustainability, the thesis makes original academic con-
tributions at the frontiers of the fields of sustainable development and geographical information technology.
The thesis further makes relevant functional and strategic modelling contributions that advance the cur-
rent state of art in floating PV performance modelling by developing a new geographical research tool by
practically implementing the 3E theoretical principle in an experimental research model.
An additional key contribution of the thesis is the development of a comprehensive set of quantitative
performance analysis metrics formulated to enhance the sustainability inference capabilities of the designed
digital model. In addition, the thesis posited a novel ratio indicator, the water economic surface transforma-
tion (w-EST) indicator (Prinsloo et al., 2021), as a ration indicator that is uniquely designed by the thesis to
reflect the income potential of an FPV system per square meter. The strategic and functional knowledge
contributions of the geographical tooling development were significantly more complex, and created new
knowledge that helped to significantly advance the current state-of-art FPV assessment modelling.
The novel contributions of this study are able to explain the limitations/deficiencies in the current state of
the art. The thesis’s theoretical contribution is able to answer the questions as to why the newly proposed
theoretical conceptualisations and trans-disciplinary system modelling framework of this thesis add to the
body of knowledge, and how the research findings offer substantive evidence to confirm how the theoret-
ical contribution overcomes the above difficulties and limitations in the current state-of-the-art by adding
environmental sustainability principles and environmental economic principles to the modelling equation.
In answering the question "why” the newly proposed theoretical conceptualisations and trans-disciplinary
system modelling framework of this thesis add to the body of knowledge, the concise explanation in Sec-
tion 6.5.2 presents important synopsis of the fundamental requirements for constructing the collective intel-
ligence imperatives of the floating PV ecosystem. It also answers the critical question as to "how” do the
research findings offer substantive evidence to confirm how theoretical contribution overcomes the above
difficulties and limitations in the current state-of-the-art, by adding environmental sustainability principles to
the cooperative FPV system operations model.
Overall, the presented theoretical framework, methodology and model, together with the reported re-
sults, provide an improved understanding of floating PV renewable energy system behaviour and highlight
the extent to which the successful integration of operating system components provides an enhanced the-
oretical basis for floating PV system modelling and sustainability assessments. Furthermore, the research
advances fresh philosophical ideas with novel theoretical principles that may have far-reaching international
implications for developing globally-relevant floatovoltaic and ground-mounted PV performance models.
6.6. Academic and Practical Utility of the Research
The study’s research holds strong academic and practical potential. It offers a range of opportunities
for further downstream academic study and research into practical applications supporting agriculturally-
integrated photovoltaic installations. Since the continued growth and increase in the penetration of floating
solar power plants would trigger multifaceted regulatory challenges that could hinder progress, the research
269
could help to overcome such challenges. Furthermore, with anticipated advances in floating solar energy
installations in the emerging solar power market, the research could benefit systems designers, developers,
investors or development organisations.
Although the sustainability concept emanating from this thesis’s geography research offers the generic-
type holistic Agri-floatovoltaic system sustainability valuation strategy from an indigenous knowledge ap-
plication perspective, the context-sensitive 3E sustainability conceptions of this thesis have international
relevance. They can serve as an underpinning basis in context-sensitive designs for uniform floating PV
sustainability assessment research worldwide. Considering the worldwide challenges with conflicts related
to land and water resources with the development and implementation of solar PV power plants, the im-
portance of scientifically-based environmental analytics in project-scenario planning can certainly guide
strategic thinking. In the context of studying the sustainability efficiency, effectiveness and reliability of PV
power systems amidst climate change vulnerabilities, the digitally-oriented FPV-GIS toolset offers a new
methodology and technique contributing to better-informed decisions in the floatovoltaic energy sector.
To this end, other geographical researchers or energy and environmental practitioners can use the re-
search outcomes and the data-driven analysis toolset to perform valuable tasks, such as installation project
sustainability assessments and environmental impact evaluations. Without this geoinformatics computer
model, as conceptualised and implemented in this thesis, the tasks mentioned earlier would be compli-
cated, time-consuming and computationally intensive. Therefore, in terms of its research utility value, as
conceptually depicted in the research-use model in Figure 6.1, Huang et al. (2015) highlights the impor-
tance of information usage relationships among the (floatovoltaic) simulation model data variables, (3E)
indicators, and (WELF) indices as a means to help understand the academic and practical utility of the
research model and its associated results.
Figure 6.1: Relationships among the data variables, indicators, and indices of a simulation model contributing to
an understanding of the academic and practical utility of the research model and its results in terms of stakeholder
engagement, policy formulation and further research (source: Huang et al. (2015), page 1185).
Regarding Figure 6.1, agricultural FPV sustainability indicators can be essential in advancing the sci-
ence and practice of energy and landscape sustainability in agricultural resource systems. They can also
support public/stakeholder engagements or policies while delivering the information needed to implement
sustainability actions (i.e. to support the drive towards 100% renewable energy technologies). Viewed from
the angle of research utility in crucial application fields, floatovoltaic technology can ensure security in the
context of sustainable local energy resulting from acceptable power-generation processes. While rapidly
emerging as a novel decentralised form of agricultural-energy technology, it is becoming a favoured agro-
renewable clean energy worldwide (World Bank, 2019b). In this context, the user-configurable digitalised
270
geo-analytics model and FPV-GIS toolset offer a robust regularisation model and a validated installation
technique to help solve business-planning problems in a decision-support paradigm.
Because the thesis developed Python code for PV electrical and meteorological data analysis, there
are opportunities for applying classical and statistical/machine learning techniques (i.e. Python Scikit-learn
(Hao and Ho, 2019)) to improve the floating PV model. Furthermore, developers can interface the balanced
scorecard results output by the (F)PV-GIS toolset with the Department of Forestry, Fisheries and Environ-
mental Affairs E-GIS platform. This interface can help to provide web-based GIS online support for PV
project assessments in energy development projects in an energy cloud environment through the E-GIS
interface portrayed in Figure U.1. Moreover, in terms of data collection, the succession of the research and
the practical application of the findings of the research model, this model can help elucidate how scientists
can develop integrated sustainability measures to support farming practices reliant on sustainable energy
sources and integrated sustainable production certifications to support agricultural product exports.
From a practical research succession perspective, the thesis highlights the relevance of tooling
development and policy challenges for sustainability assessment, with due attention to applicability in real-
world decision contexts. In this context, the study lists the following new research utility and application
opportunities: The ability to: (a) make assessments of the continental and national capacity for FPV tech-
nology (i.e. spatial statistics on ArcGIS Global Mapper) (Gangavarapu and Kulkarni, 2022; Guaita-Pradas
et al., 2019; Gonzalez Sanchez et al., 2021), similar to the USA’s geospatial analytics workflows in Float-
ing Solar Explorer studies Bourgeois and Seman-Graves (2022); Heeter et al. (2021), paying particular
attention to systematically assessing the technical potential of agricultural FPV installations on man-made
water bodies in South Africa, on the African continent, or any other virtual project terrain (Digital Terrain
Model) (VTP, 2014); (b) rehearsing floating PV projects in cyberspace to support outdoor testing facilities
where real-world performance monitoring and model validation experiments (CSIR, 2022), or climate rat-
ing tests for PV modules (IEA PVPS, 2020) can be supplemented by the FPV-GIS toolset operating as
a PV Data Acquisition System; (c) study climate hazards, climate change vulnerabilities and reliability of
power systems in terms of the impact of climate variability extremes and climate change scenarios, as
climate-induced risk assessment is an essential topic in weather risk identification and PV energy failure
mitigation research (Jackson and Gunda, 2021), thus using the FPV-GIS toolset to test sustainability impact
narratives caused by climate change effects or climate shocks to counteract the causes of such climate
change impacts through improved sustainability designs (Altamimi and Jayaweera, 2021; Groundstroem
and Juhola, 2021); (d) study energy production and water savings from floating solar PV on national/global
reservoirs, to provide estimations of the daily or annual water evaporation losses from reservoirs FPV can
prevent thanks to the installation of floating PV renewable energy systems (Abdelal, 2021; Baradei and
Sadeq, 2020; Majumder et al., 2021); (e) provide estimations of the power supplementation capacity by
pairing the installation of floating PV systems with hydro-power generation facilities in the country or on the
continent (Sterl et al., 2022; Turner and Voisin, 2022; Quaranta et al., 2021); (f) site-selection procedures
for GIS-based FPV installations, analysis in support of finding and selecting of suitable or optimal locations
for the installation and deployment of floating PV from a set of potential locational alternatives (Chaves and
Bahill, 2010; Noorollahi et al., 2022; Tercan et al., 2022); (g) support investment decision-making, where
the regulators and investors require scientifically-founded impact analyses as a mandatory component in
considering statutory and regulatory project approvals (Goh et al., 2022; Piana et al., 2021); (h) conduct
sustainability assessments in the analysis of future planned floating PV projects, and apply the support
tool in integrated EIA project assessments by integrating the hierarchy of 3E impact effects into an un-
derstanding of the environment in floatovoltaic deployments (Armstrong et al., 2020; Lee et al., 2020); (i)
perform programmed due-diligence performance assessments for future planned floatovoltaic installations
in a desktop environment, to compile a project playbook for an FPV installation during the conceptualisation
271
and design phases of the project; (j) support improved EIA practice, with the FPV-GIS toolset delivering
an impartial assessment, with transparent data used in support of analytical EIA imperatives that meet the
requirements of EIA directives and reporting; (k) support the sustainability certification of farming in the
case of the fruit-farming and wine-making industry, with particular attention being directed to certification
advantages for sustainable farming activities such as FPV- and farm-based energy generation (Moscovici
and Reed, 2018); (l) determine the decarbonisation potential for FPV in support of carbon management ac-
tivities by studying the ability of FPV to sequester carbon, to reduce the grid pollution load, to displace fossil
fuels (Pouran et al., 2022), and to improve the hydrocarbon footprint through indigenous energy production
in each fiscal year (Ntombela, 2019); (m) support reporting requirements in investigations into carbon-tax
compliance, with the FPV-GIS model simplifying complex carbon taxation and environmental assessment
predictions for floatovoltaics by using a computing tool that concurrently evaluates the sustainability qual-
ities of floatovoltaic solutions (Cohen and Hogan, 2018; Fereshtehpour et al., 2021; Siweya, 2021); (n)
support reporting requirements in floating PV technology installation authorisations (Cohen and Hogan,
2018; Norton Rose Fulbright, 2021) as well as applications for sustainable farming certifications (Kuschke
and Cassim, 2019; Moscovici and Reed, 2018); (o) apply as an instructional teaching tool to improve solar
PV literacy and problem-solving skills, using the FPV model as an inquiry-based digital learning tool to train
a current workforce on sustainability opportunities, or to avoid technogenic risks with the help of the FPV
model as an e-learning tool (Anker et al., 2008), with the instructional design learning objectives incorpo-
rating the FPV-GIS system dynamics tool’s workflow and data-driven decision support capabilities to teach
and learn about the intricate associations and interactions between technology and the environment, and
in an understanding of how a PV system design can be reconfigured to ensure more sustainable projects
for developments incorporating strict environmental constraints; (p) to serve as a solar PV design tool, us-
ing the FPV-GIS model’s user configuration file to design various floating PV configurations in performance
testing experiments; (q) to analyse factors leading to floating PV system underperformance, such as climate
factors, module type or mounting configurations; (r) to support the global PV industry with a scalable, robust
Python analysis tool in a programmable computer analysis environmental to identify project vulnerabilities
and to help reduce the uncertainty of floating PV system performance and loss factor calculations; (s) to
support SA’s National Climate Change Adaptation Strategy through digital climate solutions (DFFE, 2019a)
in modelling the potential impact on climate change by FPV and PV installations in the agricultural sector
through probabilistic modelling; (t) to study anthropogenic climate change issues by engaging E3SM cou-
pled climate-energy models and E3SM data modelling techniques (Leung et al., 2020) with the FPV-GIS
model to study the role of floating PV as a future climate change mitigation solution; and (u) finally, compu-
tational engineering of the digital twin as digital representation of an FPV project can further find application
in the post installation service and maintenance phases of physical floating project installation life cycle.
The efficacy of the geospatial digital twin and the experimental results it produced in FPC analysis
spotlight how this research’s novel framework contribution created new knowledge as a fitting sustainability
framework with international relevance that helped to fill critical knowledge gaps that currently hamper com-
pliance in floating PV deployment applications worldwide (refer to (Gazdowicz, 2019; World Bank, 2019b)).
The ability to scientifically forecast energy supply generated from renewable resources and its impacts on
the environment and the economy are crucial for effective energy planning, policy modelling, and policy
support. From a policy support perspective, the ultimate aim is to appropriate the 3E sustainability metrics
and indicators in a regulatory EIA assessment protocol for FPV. In the context of return on innovation, the
thesis anticipates that the posited theoretical framework can help to alleviate current regulatory project de-
lays and, as such, command a more favourable technology acceptance rating on the scales in Figure 2.14,
particularly on the technology’s TRR (regulatory readiness) rating scale, the IRR (investment readiness),
and on the TMR (market readiness) rating scale. This statement serves as a valuable recommendation
272
in respect of two areas for future research policy, namely to use the theoretical principles posited by this
research in the development of TRL, TRR, and TMR metrics (refer to (Kobos et al., 2018)), as well as in the
development of fit-for-purpose EIA protocols for the regulatory assessment of floatovoltaics in South Africa
(which would also have international relevance).
From a academic research succession perspective, the theoretical constructs, models and data are
of significant relevance and utility in sustainable development research. As the present study contributes
to the sustainable development discourse, it supports meaningful academic research of value in that it is
concerned with developing further knowledge that can directly improve the theoretical concepts posited by
this thesis. Furthermore, it offers anticipated opportunities for downstream research in terms of extended
academic, theoretical and policy research opportunities. As such, the study lists the following potential op-
tions in respect of the academic value of the research: (a) the development of technical and sustainability
standards for floating PV projects around the proposed 3E sustainability framework; (b) the adoption of the
3E sustainability assessment framework theory as a legal framework for the regularisation and/or uniform
standardisation of floating PV project assessments; (c) considering the 3E sustainability framework as a
means to overcome the challenges around the legislation and framework policies for floating PV systems
as environmental models (Cohen and Hogan, 2018; Norton Rose Fulbright, 2021); (d) considering the im-
plications of using the 3E sustainability framework in supporting EIA assessments and meeting regulatory
EIA directives for sustainability assessments and resource conservation impacts; (e) considering the impli-
cations for using the 3E framework in support of energy sustainability certification for farming enterprises
based on the technology’s benefits arising from agricultural resource conservation Moscovici and Reed
(2018); (e) considering studies on the implications for FPV in water-resource modelling to relieve pressure
on the already-limited water supplies around the world (Cohen and Hogan, 2018); (f) considering studies on
the potential climate-change benefits of FPV technologies in terms of the modelling of the decarbonisation
potential of the energy system as an essential policy focus in the face of global warming and its threatening
effects (Cohen and Hogan, 2018); (g) investigating opportunities for extending the PV performance assess-
ment model to incorporate processing further to support PV project-validation practices under carbon-tax
legislation protocols (RSA, 2019a; Siweya, 2021); (h) investigating opportunities for the theoretical frame-
work and (F)PV-GIS toolkit to support the sustainable development goal seven (SDG7) of the United Nations
SDG Acceleration Toolkit initiative (UN, 2022); (i) investigating opportunities for linking the (F)PV-GIS toolkit
to the E-GIS platform of the Department of Forestry, Fisheries and Environmental Affairs, to provide online
support for PV project assessments in energy development projects; (j) improving the modelling of float-
ing PV through extended research on floater-specific FPV micro-habitat (micro-climatic and hydro-climate)
modelling procedures for FPV systems; and (k) investigating opportunities for incorporating the 3E frame-
work model in conventional PVlib and PV systemic-type performance-modelling methodologies, including
the conditioning of traditional floating PV performance models based on the theoretical 3E sustainability
framework theory and micro-habitat modelling procedures disclosed in this thesis.
This section offers an opportunity to study the practical applications of the research outcomes and
develop new academic research opportunities around these outcomes. The following section provides
directions for future research toward improving the theoretical modelling concepts introduced in this thesis.
6.7. Study Limitations and Directions for Future Research
Regarding the direction for future research, the future aim is to roll out the floating solar PV design tool as
cloud-based energy modelling software, thus estimating the terrain-based performance of floating PV power
plants, predicting its environmental impacts, and accessing its economic potential through a digital gateway
or online application. Research into Solar Futures studies from reliable institutions, such as the NREL
273
and Sandia Corporations (Gadzanku et al., 2021b; Spencer and Barnes, 2020; Woodhouse et al., 2021),
have identified future floating PV research opportunities where subject matter expertise and knowledge can
be incorporated into the multi-benefits analysis modelling to support the digitalisation and automation of
project assessments. They specifically emphasise the value of engaging more disciplines and investigating
potential research areas where further efforts are needed to perform root cause analysis, to enable model
order reduction through deep learning, while ensuring reduced uncertainty in floatovoltaic performance pre-
dictions. In this context, the development of the proposed FPV-GIS model can be improved concerning
different spatial interfacing scales and application-specific improvements such as integrating hydropower
systems, hydrogen generation and ecological/social impact knowledge bases into the modelling framework
and computer-modelling architectures (Prinsloo et al., 2021). In addition to studying the whole-system
FPV sustainability, economic & carbon ledger information, displacing fossil fuels, human-environmental in-
teraction, and the long-term durability and recyclability of floatovoltaics, further research may also consider
striving for deeper insights into economic development, the financing of renewable energy projects, increas-
ing investment opportunities, adopting the 3E framework in the modelling procedures of the PVlib model
chain, modelling multi-orientation fields with the mixed-array orientation of panels on water (Derevyanko,
2022; Princloo, 2023), the modelling of diversity and energy storage options in floating PV power gener-
ation (Kumar et al., 2022), accommodating water-submerged PV panels, mono-facial and bifacial floating
PV power plants (Aghaei et al., 2022; Röhr et al., 2022; Tina et al., 2021a), and improving the accommo-
dation of the dual use of the water area in agrivoltaic or aqua voltaic-type aquaculture systems (Gorjian,
2022; Pringle et al., 2017). Finally, the FPV-GIS model can be deployed in a virtualised deployment por-
tal, thus engaging cloud computing infrastructure and climate engine models to serve in a virtual assistant
environment making the FPV model application available to users worldwide. Such improvements would
strengthen the application of the model in terms of the development and integration of additional compo-
nents into the systemic model to overcome certain study limitations as stated in the research assumptions
(refer Section 1.5).
There are several opportunities for future research work around improving predictive accuracy and
calibrating the FPV-GIS model through model-parameter uncertainty quantification and sensitivity anal-
yses. However, to ensure floating PV model calibration and data assimilation with actual real-world 3E
performance data, a comprehensive multi-year research effort would be required to take unique in-situ field
measurements over several years. To establish such reference datasets, dedicated 3E sensors would thus
have to be physically installed and audits taken at various test sites to quantitatively collect real-world field
data from actual operating FPV systems (refer to Kjeldstad et al. (2022)). An adequately-sized multi-year
sample of comparable energy, environmental and economic field data would need to be collected from out-
door FPV installation projects. Data from distributed field processing and acquisition techniques can help
identify potentially unavoidable imperfections in the floating PV model or model processing algorithms (or in
the chosen model inputs or outputs). It may help to identify model prediction uncertainty or the propagation
of errors in lifetime FPV project assessments. However, floating PV energy system modelling performance
calibrations come with the challenges of 3E data shortages and a lack of comparability of 3E data among
real-world installations for different regions (not to mention data for other countries). It may therefore take
several years to collect adequate field data regarding the energy amount of energy, economic and environ-
mental performance components (3E field data) from operational floating PV systems, especially in South
African agricultural applications. Instead of evaluating the long-term performance of floating PV modules
under diversified terrestrial conditions, such collection of outdoor performance data would lead to the highly
undesirable situation of waiting 10 to 20 years to empirically determine the reliability of the model (Kim et al.,
2021a). While the geospatial cyber-infrastructure offers powerful spatial cloud-computing platforms to help
improve computer models (Kjeldstad et al., 2022; Yang et al., 2011), future research towards automated
model calibrations and uncertainty quantification may develop unsupervised machine-learning methods or
274
faster floating PV surrogate modelling methods to analyse lifetime or time-series performance data in com-
parison to real-world observational records (Khan et al., 2022; Meyers et al., 2020).
Towards next-generation FPV modelling and performance forecasting, future research could develop
Python machine logic and machine learning algorithms in an FPV-GIS machine learning layer. In
such a generative design context, the geoinformatics-driven predictive digital twin already established a
digitalisation architecture to enable the transition from physics-based modeling to scientific machine learn-
ing (refer to (Willard et al., 2020)). Building mathematical intelligence and machine learning into renewable
energy systems forecasting for floatovoltaics would aim to use AI-generated synthetic data from regression
or surrogate data models to analyse floating PV model outcomes, parameter dependencies or model input-
output associations (refer to (Abdelmoula et al., 2022; Gunda and Cai, 2019; Raschka and Vahid Mirjalili,
2019)). Such a generative AI feature in digital twin-type FPV power systems analyses can be based on
model-based reasoning to offer valuable system design tuning benefits to support the synergistic optimi-
sation of renewable energy-based FPV energy systems (refer to (Dellosa and Palconit, 2022; Krechowicz
et al., 2022)). While digital twin modelling integrates high-fidelity simulations to operate at the forefront of
the 4IR digital transformation (ESRI, 2022; Jainandunsing, 2019), the next step in the development would
logically be to add an artificial intelligence software layer to integrate machine learning API layers into the
hierarchy of the floating photovoltaic ecosystemic model. With hindcasting as an extended model capa-
bility in the synthetic intelligence layer, statistical data mining can assist in model parameter tuning. The
FPV-GIS model could thus be set up to run numerous configuration options to build up feature stores,
enabling artificial intelligence algorithms to search through the data output sets for ideal patterns of perfor-
mances, to study the sensitivity of parameterizations, or to find perfect system configurations that best align
with operational goals of a planned FPV project. Such prescriptive analytics can train a machine learning
model/layer (decision tree classifier, regression model, classification and regression tree or artificial neural
network ANNs) to identify idealistic project outcomes or optimal configuration performance clusters.
To this end, the geospatially-enabled FPV-GIS digital twin model of this thesis has, in effect, already
established the foundations for a more advanced Data Science approach to floating solar PV planning
assessments through an extended machine learning model layer depicted in Figure 6.2. The geospatial
digital twin model extension for digital project re-enactment in Figure 6.2 can thus operate as a self-learning
Data Science tool.
Country-specific geospatial, environmental, legislative, economic, technical context
□Floatovoltaic Geoinformatics Information System Platform
∗Floatovoltaic Artificial Intelligence Analytical Model Layer
△Floatovoltaic Planning Decision Indicator Model Layer
⊕Floatovoltaic Ecosystem Synthesis Model Layer
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
(a)
⊕Floatovoltaic Ecosystem Synthesis Model Layer
(Primary-level-synthesis)
△Planning Decision Indicator Model Layer
(Secondary-level-processing)
△Artificial Intelligence Analysis Layer
(Machine-level-optimisation)
□FPV Geoinformatics IS
(Tertiary-level-contextualisation)
Project Planning
(Supervisor-level-optimisation)
(b)
Figure 6.2: Adding augmented analytics capability in an artificial intelligence layer to the FPV digital twin model as (a)
an extension to the geoinformatics planning decision-support system for floating PV projects through (b) a geospatial
AI driven GIS layer for computerised PV project analysis (source: author).
275
In this capacity, the FPV-GIS digital synthesis model in Figure 6.2 may be set to run hundreds of project-
configuration scenarios to create a library inventory of meta datasets from which machine learning algo-
rithms could explore design tradeoffs, optimise operational configuration, or discover statistical correlations
in cause-effect diagnostics. In such a proposed future data engineering development and digital twin’s AI
mode of operation, the floating solar digital twin can serve as a form of intelligent computer-aided decision
support. The big-data AI assessment results from Figure 6.2 can, as such, be helpful in a model-parameter
sensitivity analysis, an uncertainty analysis, an automated model calibration, or a multi-criterial optimisation
of project design and configuration parameters. To this end, the AI model layer in Figure 6.2 can, in the fu-
ture, contribute to more dynamic and accessible ways of conveying knowledge and discovering patterns or
trends in operational characteristics by employing machine-learning and deep-learning artificial-intelligence
algorithms. AI analysis on multiple FPV project performance prediction datasets can provide actionable
business intelligence to support sustainable project planning improvements. While GIS offers vital tool fea-
tures for environmental impact assessment and mitigation, the GIS model in Figure 6.2 may ultimately help
deliver a scalable FPV EIA assessment service on a flexible software-as-a-service (SaaS)-based platform,
an appropriate service-oriented architecture. This may incorporate the FPV model as an application-specific
geo-solution with geospatial data drawn directly online from the relevant big data repositories.
Further research is also required around the more focused development of the FPV assessment tech-
nology towards a model extension for pairing floating PV with natural gas-fired power plants or hydro-
power plants (Farfan and Breyer, 2018; Gadzanku et al., 2022a). While hydro-electric hybridisation with
floating PV establishes a modern trend in FPV development and greenhouse gas control (Kakoulaki et al.,
2022; Quaranta et al., 2021), the integration of floating photovoltaics into hydropower reservoirs would offer
remarkable benefits in terms of the generation of power that can attain an increase in hydropower capacity
and efficiency thanks to the FPV system reducing water evaporation losses (Gupta, 2021; Sterl et al., 2022;
Turner and Voisin, 2022). To this end, decision engineering scientists can use the Google Earth Engine,
geographical information science (GISc) or ArcGIS Pro platforms, UAV remote sensing, drone photogram-
metry or image processing expert systems to model the land geomorphology, identify geomorphic features,
study water landscapes, or identify appropriate water zones to install floating PV systems (Yilmaz et al.,
2022). As highlighted in Figure W.1, hybrid FPV system configurations could further help boost energy
capacity on a continental scale, build new revenue attribution models, and help reopen rivers by replacing
hydropower by solar energy facilities (Sharma and Waldman, 2021). In this respect, future improvements
to the FPV-GIS model are proposed, with the hybridised floatovoltaic-hydropower sustainability framework
model variant, as depicted in Figure V.1, being the primary means to achieve such an objective. Further-
more, the application-specific incorporation of a hybridised hydro-FPV model into the FPV-GIS platform
could, in fact, facilitate a study into the complementary operation of such a hybridised model in improving
water storage and energy generation in a multi-generational systems format.
The FPV model variant in Figure V.1 could also serve as an ecological sustainability framework el-
ement for floatovoltaics to assess the extent to which the technology protects human health and fragile
ecosystems in the environment. Such an extended 3E plus ecological framework extension can help to
study the environmental compatibility properties of FPV systems to help safeguard terrestrial ecosystems
and the natural environmental system associated with future FPV installations. This option may include eco-
logical impact modelling with sustainability credit score measures for floatovoltaics combined with the nexus
of the energy, water, land and atmospheric sectors. This approach would be offering exciting opportunities
for future research around an FPV’s impact on freshwater ecology, eco-hydrology, agro-ecology, biophysics
and biosystems engineering (DFFE, 2015). While islanded microhabitat for FPV technology interfaces di-
rectly with the underlying biologically-sensitive aquatic water body (Armstrong et al., 2020; Hernandez et al.,
2019), the modelling variant of the FPV system, as proposed in Figure V.1, offers the opportunity to mea-
sure or simulate biodiversity offsets in terms of bio-security metrics around counterbalancing the loss of
276
biodiversity or natural resources (Hamidov et al., 2022; de Lima et al., 2021a). Such FPV model can help
study photovoltaic consequences on fish ecology and photosynthetic organisms and the chemical, physical
and biological processes on drinking-water quality, terrestrial wildlife, aquatic biota, panel biofouling, algal
blooms or downstream ecosystems (Almeida et al., 2022a). The effects on nature, such as small animals
and bird species observed immediately after the installation of floating PV systems (Allen and Prinsloo,
2018; Andini et al., 2022), constitute part of the ecological footprint of floatovoltaics as they may impact di-
rectly and indirectly on natural processes and the mutually interactive techno-ecological field at play around
the FPV system’s aquatic theatre of operation.
The FPV model variant proposed in Figure V.1 can also serve as a social sustainability assessment
framework element for floatovoltaics to unravel the social accountability and social dynamics associated
with floatovoltaic technologies. In support of King IV reporting requirements for Corporate Governance
in South Africa, the study submits that social impact assessments should dovetail with climate risk and
response valuations (King, 2016). In the interest of corporate social responsibility, the social sustainability
framework measures may thus require conceptualising a social credit score to evaluate social accord and
vulnerability based on the Environmental, Social and Governance (ESG) social credit score (Alaoui et al.,
2022; Sustainalytics, 2021). In manoeuvring practical approaches to attain SDGs and socio-economic
development goals in the post-pandemic environment, understanding the economic and socioeconomic
impacts of floating solar in projects, programmes and computer modelling regimes is becoming more crucial
(Bax et al., 2022; Naegler et al., 2021). In an integrated assessment model context, human science and
human movement science stands central to project sustainability. As such, behavioural science aspects
of the Geography of human and social dynamics are essential elements that contribute to the successful
operation of the triple helix sustainability model. This model makes stakeholder engagement, environmental
protection behavior, and empowerment equally crucial elements to model into a floatovoltaic model. Such
socio-economic imperatives support the SDGs and create awareness of photovoltaic projects (Flammini
et al., 2014; Khan, 2021).
From a social impact modelling perspective, the proposed extension of social impact objects, as de-
picted in Figure V.1, offers a practical utility value regarding the technological transfers of new knowledge.
It provides an opportunity to study the dynamic of the sociotechnical aspects of FPV technology (refer to
(Geels, 2022; Jackson, 2010), thus adding to the knowledge base and supplementing the industrial prac-
tices that the floatovoltaics industry can employ. Social impact modelling for FPV offers opportunities for
incorporating a social system model into the climate-economic-WELF nexus. It may also combine environ-
mental and social impact assessments (ESIA) to evaluate and predict potential adverse social and environ-
mental impacts in developing suitable human settlements and their required mitigation measures (Prinsloo,
2017c; IUCN, 2020). The proposed FPV model social object can help to catalogue concerns around wa-
ter use, curbing of water recreation and undermining of food webs, including livelihood fishery, marring
of scenery and resistance from neighbouring landowners that may affect property prices (Almeida et al.,
2022a). Furthermore, bridging the social assessment gap with a social impact footprint for floatovoltaics
can undoubtedly help to improve understanding of the impacts of floatovoltaic installations’ on vulnerable
populations (DFFE, 2019b). This area is considered indispensable to empowering indigenous voices in floa-
tovoltaic EIA and dealing with the public health considerations in EIA and sustainable development systems
(Ernst and Fuchs, 2022; Colvin et al., 2014). Furthermore, the framework can incorporate aspects such as
modelling sociocultural influences on PV project decision-making. Finally, such a framework can address
issues of the European Green Deal, the EU’s action to attain climate neutrality amidst various geopolitical
problems and sustainable development goals (Fayomi et al., 2021; Kougias et al., 2021), and will be more
in line with the renowned Energy Trilemma Index tool developed by the World Energy Council to compute a
country’s energy ranking on an interactive energy index (WEC, 2021).
Finally, as one of the critical recommendations on the academic utility of the research of this thesis,
277
efforts towards improving the existing PV performance models should further emphasise the more compre-
hensive characterisation and modelling of micro-climatic variations within the floating PV ecosystem
micro-habitat. In this value proposition context, key research opportunities lie in further academic, theoreti-
cal and practical research around the contextual characterisation and recreation of the floating PV microhab-
itat in support of existing PV performance models (such as PVlib and PVsyst). To further inspire confidence
in floating PV project assessments, progressive analytical modelling solutions can direct further attention
to improving conditioning functions associated with floating solar PV energy systems suspended in marine
ecosystems or man-made freshwater ecosystems. Even in the case of ground-mounted PV systems, the
modelling of the microclimate habitat in terms of micrometeorology parameters and virtual sensor readings
can be based on soil moisture variations (especially in agrivoltaic systems). The geographically-motivated
microclimate model of this thesis postulated but one approach toward a meteorologically-motivated math-
ematical model for estimating the hydroclimatic conditions within the FPV microhabitat. This notion was
inspired by scientific FPV research, which determined that floatovoltaic systems mounted on a floating chas-
sis or open structures allow cool air breezes and wind flows to pass beneath the modules by offering higher
heat loss coefficients (Bellini, 2021; Sutanto and Indartono, 2019). It logically implies that the floatable pho-
tovoltaic mounting system architectural design type is crucial to enhancing the floatovoltaic cooling effects
on the water surface air interface (Scavo et al., 2020; Peters and Nobre, 2021). An appropriate floating PV
substructure design type can thus achieve improved hydro-elastic platform dynamics for different environ-
mental loads (Abbasnia et al., 2022), while at the same time ensuring an improved microhabitat to obtain
an appreciable enhancement in energy gain and water evaporation rates (Allen and Prinsloo, 2018; Bellini,
2021). The enveloping micro-climatic hydro-climatic effects could, as such, be a function of the floater ge-
ometry or structural design of the floater that houses the solar panels and PV canopy (floater carrying the
PV panel payload). The structural engineering of the floater design explains the floater’s mechanical, aero-
dynamical, and hydrological properties. In water photovoltaics development, the hydroclimate dynamics of
the floater design directly impact the surrounding microclimate. Under the supposition of this thesis, the
hydro-driven microclimate is numerically modelled as a function of the quantitative climate assessment. It is
predicated on the idea that the microclimate within the volumetric structure of the hemispheric hydroclimate
bubble can be represented as a schematic model (Figure L.1), whereby characteristic FPV system floater
coefficients help to modulate and filter the TMY data (see Section 4.3.4.2). In future research, floating PV
floater-system designers or manufacturers should instead study floater characteristics through field mea-
surements, laboratory experiments, or computational fluid dynamics to specify the governing dynamics of
an FPV floater design in terms of floater-specific thermal exchange thermodynamics or thermal convection
signature coefficients.
6.8. Summary
This chapter concludes the thesis with investigative research to help solve a contemporary real-world prob-
lem around the planning of contextualised floating solar PV energy systems through scientifically-driven
analytical decision-support mechanisms. The study addresses the issue of floating PV system character-
isation modelling and analysis, especially since traditional PV performance models do not reflect the full
complexity of the behaviour of floating PV systems and their real-world performances and impact effects.
From the philosophical perspective, the fundamental ideas of this thesis led to the conceptualisation of a
theoretical 3E solution that leverages the concrete agglomeration benefits in the convergence of the techno-
environmental-economic (3E) domain ensemble within the operational floating PV ecosystem. In proposing
a 3E theory for characterising the temporal and spatial interrelations among the floating PV ecosystem’s 3E
domain elements, the study maintains that system dynamics principles and arithmetics should choreograph
278
the coalescence of the energy, environmental and economic (3E) domain operations. The study thus offers
a new paradigm of PV system sustainability modelling and analysis, especially suitable to project planning
analysis and decision advisement during floating PV project commissioning phases. With 3E sustainability
qualities being the decisive requirement in creating a portrait of project outcomes and sustainability integrity,
the experimental results emphasise the significance of the thesis’s hypothesis and the usefulness of the 3E
theory in compiling a more realistic project sustainability playbook. With these theoretical conjectures, the
study contributes to the fields of earth and environmental sciences, wherein the engagement in scientific
research may continue to support the environment, energy and geoengineering strategies towards sus-
tainability and sustainable energy development in the South African agricultural and energy development
landscapes.
279
APPENDICES
280
A. Mind Map of the FPV Sustainability Narrative
Floatovoltaic
sustainability
assessment
ecosystem
benefits
cooler
system
>power
output
environment
offsets
water
preservation
land
preservation
ecological
impacts
planning
context
multiple
objectives
performance
goals
impact
appraisal
risk as-
sessment
viability
analyis
vulnerability
analysis
decision
context
complexity
inter-
dependencies
narrow un-
derstanding
diverse
sources
fragmented
info
time horizon
context
sensitivity
sustainability
geo-
spatiality
dam
availability
diverse
policies
compliance
require-
ments
investment
require-
ments
theoretical
foundations
geo-
informatics
energy
economics
environment
computer
modelling
systems
science
technical
equipment
type
equipment
sizing
geospatial
location
configu-
ration
meteoro-
logical
operations
Figure A.1: Mind map of the sustainability assessment narrative for floatovoltaics to highlight the complexity and
diversity of energy system sustainability in a climate smart agricultural context, while making critical observations to
identify systemic elements and behaviour patterns within the floating PV ecosystem to help reveal the sustainability
decision dimensions/complexities within the planning phase of a new information ecosystem for FPV (source: author).
281
C. State-of-the-art Tools: PV Performance Analysis
Figure C.1: State-of-the-art tools in PV performance analysis shown on Thematic Spatio-temporal Power Systems
Model Map, to highlight analytical simulation/optimisation tool models for energy system application domains and to
show the methodological gaps in floating PV performance assessment capabilities (after: NREL (2018b)).
283
D. Research Methodology Development Chart
Philosophical Approach (Section 3.2)
Methodological Approach (Section 3.3)
Methodological Framework (Section 3.4)
Theoretical Assessment Framework
Development Procedure (Section 3.5)
Computer Model Design towards Realisa-
tion of the Research Aim (Sections 3.6)
Theoretical Substantiation of the Research
Philosophy and Methodology (Section 3.7)
Figure D.1: Flowchart of the research framework and model development steps towards the goal of establishing a
systems-based quantitative research methodology (source: author).
284
E. Conceptual FPV Sustainability Philosophy/Theory
Technological
sustainability
Economical
Sustainability
Environmental
Sustainability
Agricultural
Energy
System
Sustainability
feasible WEL(F)
resources
viable
Operational
behaviour
Performance
features
Impact
effects
Floatovoltaic
Digital Twin
Sustainability
Synthesis
responses responses
indicators
(a) (b)
Figure E.1: Conceptual FPV sustainability philosophy as a new theoretical contribution to the sustainable develop-
ment discourse, showing (a) the theory integrating technological, economical and environmental sustainability in an
agricultural energy systems sustainability criteria framing, (b) to serve as underpinning theoretical basis and computer
program logic in the synthesis of floatovoltaics sustainability in a unique modelling concept that integrates operational,
performance and impact assessment in a consolidated model for agricultural floatovoltaic sustainability synthesis
(Note the energy-water-land EWL resource metrics nestled at the intersection of the Energy-Environmental pillars)
(source: author).
285
F. Floating PV Model Development Flowchart
Research Objective 1: Conceptualise and de-
sign a floating PV system modelling framework
and systems structure (Sections 3.2 to 3.7)
Research Objective 2a: Define a dynamic
floating PV systems structure, components,
interactions and parameters (Section 4.2)
Research Objective 2b: Characterise float-
ing PV model object components using a
model-based design process (Section 4.3)
Research Objective 3a: Characterise sustain-
ability decision-support indicators (Section 4.4.2)
Research Objective 3b: Characterise
decision-support model using AHP
decision hierarchy (Section 4.4.3)
Research Aim: Embed analytical and decision
models as geoinformatic system layers for
data collection and processing (Section 4.5)
Research Experiments: Design an
experimental case study for frame-
work theory evaluation (Section 4.6)
Figure F.1: Research design process, showing the flowchart for the model development of the computer modelling
and simulation model for application in new sustainability assessment methodology (source: author).
286
G. Systems-thinking Hierarchical Pyramid
Figure G.1: Development of a systems-thinking model in terms of the hierarchical model pyramid. The development
and implementation of a systems-thinking model requires that the systemic components be identified and analysed,
that these elements be mathematically synthesised, and that the relationship functions between the systemic elements
be synthesised as key ontological modelling activities (source: Orgill et al. (2019), page 7).
287
H. Causal-diagram for Floating PV Sustainability
Figure H.1: Analytical geo-information system perspective on floatovoltaic technology sustainability, using a system-
dynamics modelling approach with a causal-and-effect diagram to understand FPV system’s energy, environmental
and economic behavioural responses to reservoir-induced microclimate and radiation changes in a behaviour-over-
time graph. The causal-loop diagram offers an agency analysis or network cluster analysis oriented towards system
integration of sustainability performance characterisation in drafting the system boundaries, identifying the principal
components and principal agents, as well as the recurring dynamic interactions between the ecosystem actors as
these transients effects are causing reciprocity-type chain-reactions as feedback effects (to be modelled as cause-
and-effect relationships within the floating PV system’s computational workflows) (source: author).
288
I. Floating PV Model Simulation Synthesis Layers
Technological
impacts
Economical
impacts
Environmental
impacts
Integrated
Analytical Framework
for floatovoltaic
Sustainability Profiling
techno-econo
responses techno-enviro
responses
econo-enviro
responses
(a) Conceptual systems model (Theoretical framework)
Energy
Technology
Modelling
Economic
Modelling
Environmental
Modelling
f(techno, enviro)f(techno, economic)
f(enviro, economic)
Sustainability
Impact Assessment
(b) Logical systems model (Analytical assessment)
(c) Physical systems outcomes (Decision support)
Figure I.1: Development layers of an integrated systems model towards the implementation of a simulation model,
depicting its functions in a layered hierarchy comprising (a) conceptual framework modelling, (b) logical analytical
modelling, and (c) physical decision profile modelling (source: author).
289
J. Floating PV System-of-systems Conception
(a)
(b)
Figure J.1: Object-oriented system-of-systems architecture for the software integration of (a) an energy-
environmental-economic (3E) system framework and (b) a water-energy-land resource WEL systems framework into
discrete-time synchronous system-dynamics simulation models for floating PV characterisation and assessments
(source: author).
290
K. Iterative 3E System Dynamics Processing
Simulation
prototype
setup
Load user-
configuration
Modelling
site setup
Environmental
model Solar model Meteorology
model
Hydroclimate
model
Evaporation
model
Land save
model
Anti-pollute
model
Energy
Model
Reservoir
model
Thermal
model
Production
model
PVlib step
simulator
Economic
model
Electrcty
savings
Carbon
credits
Water
savings
Agri crop
income
Agri comm
income
System Dynamics
Controller
(a)
t−2t−1t t + 1 time horizon
ensuing 3E (Environmental, Energy, Economic) analysis iteration sample periods
next 3E (Environmental, Energy, Economic) analysis sample iteration
(b)
Figure K.1: Geospatial digital twin synthesis automation using (a) a nested multi-methodology simulation expres-
sion to ensure integrated system-dynamics workflow through step-by-step execution in a recurring 3E processing
sequence; and (b) the Energy, Economic and Environmental model sub-system component objects participate in
each time instance of the iterative networked simulation processing cycle to ensure reciprocal workflow in real-time
computational workflow simulation steps through an integrated system dynamical execution stack (source: author).
291
L. Conditioning of PVlib Model for Floatovoltaics
Meteorological
TMY data
Geo-data repository
Hydroclimatic
Pre-processing
Floater-specification
PVlib thermal-
model
FPV-configuration
PVlib simulator
engine
FPV-configuration
PVlib thermal-
model
GPV-configuration
PVlib simulator
engine
GPV-configuration
t r f yF
yG
Energy Model components
Environmental Model components
Figure L.1: A conceptual model for conditioning the PVlib model inputs for executing stepped floating PV system
simulations based on microhabitat climate reformation, using algorithm microclimate parameters for hydroclimatic
pre-processing as a function of the meteorological weather station data and PV floater design-type (source: author).
292
M. FPV-GIS Model Layers and Components
□Floatovoltaic Geoinformatics Information System Platform
Country-specific geospatial, environmental, legislative, economic and technical context
△Floatovoltaic Planning Decision Indicator Model Layer
⊕Floatovoltaic Ecosystem Synthesis Model Layer
Technological
Energy
Simulation
Model
Object
Economical
Simulation
Model
Object
Environmental
Simulation
Model
Object
f(x)e(t)f(y)e(t)
f(z)e(t)
Power Pm (kWh)
Conversion eff Pη(%)
Volts potential Vm (V)
Ampere current Im (A)
Panel temps fTpanel (◦C)
Project lifetime specifications
Site location configurations
Geometric design & surface area
Technical floater design parameters
Technical FPV design configurations
Agro yield (kg)
Avoided CO2(kg)
H2O saved (liter)
Avoided SOx(kg)
Avoided NOx(kg)
Avoided aPE (kg)
Avoid Coal fuel (kg)
Avoid Ash waste (kg)
Solar irradiation
Weather data
Climatic data
Cloud modulation
Agronomic setup
Revenue/Savings (R)
LCOE, LACE (R/kWh)
Depreciation value (R)
NPV, Payback (yrs)
ecoTax credits (R)
Value of H2O (R)
Value agro prod (R)
Capex & Opex costs (R)
Incentives, funding & subsidies (R)
Water, land & agro economics (R)
Electricity tariff rates (R/KWh)
Carbon credit pricing (R/ton)
Discount & interest rates (%)
ecoTaxes savings & rates (%)
Figure M.1: Geoinformatic model-driven simulation instrument implementing an empirical floatovoltaic ecosystem
synthesis and decision support system. Based on the integrated analytical framework and recurring system-dynamics
computer model expressions, it establishes a digital twin desktop instrument design. It develops prototype models for
real-world floating PV installation projects (source: author).
293
N. FPV-GIS Model Parameter Notations
Table N.1: FPV-GIS model parameter notations and units of measure (source: author)
Afpv water PV plot size m2
fApd panel density panels/100m2
fPrc nameplate-rated capacity of panels Watt
falt reservoir altitude meters
fln reservoir longitude degrees
flt reservoir latitude degrees
ftz reservoir timezone hr:min:sec
fptilt panel surface tilt angle degrees
falbedo surface albedo water unit
galbedo surface albedo land unit
Pm maximum electrical-power current outputs Watt
Vm maximum output voltage potential Volts
Im maximum output current Amperes
Pη(y,t)solar-to-electric power conversion rate %
fgPmηefficiency gain of FPV over GPV %
Pηsolar-to-power conversion efficiency %
Tdry, Twet dry-bulb-, wet-bulb- temps ◦C
fTambient floatovoltaic ambient temps ◦C
∆Td difference between FPV and GPV ambient temps ◦C
fTpanel, gTpanel operating temps for FPV and GPV panels ◦C
Pmax PV panel temperature coefficient watt/◦C
fPVd environmental degradation function of PV panel %/annum
DNI, DHI direct normal, direct horizontal irradiation W/m2
Notation Description Unit
Continued on next page
294
Table N.1: FPV-GIS model parameter notations and units of measure (source: author) (Continued)
GHI global horizontal irradiation W/m2
POA plane-of-array irradiance W/m2
ITR effective on-panel irradiance W/m2
Wspd location-specific windspeed m/s
RH location-specific relative humidity %
eH2O evaporation avoided by FPV mm/h or litres/h/m2
xCO2grid substituted carbon emissions kg/ton
pCO2pasture atmospheric-carbon sequestration kg/m2/yr
pfacC vegetation territory-emissions factor kg/m2/yr
xH2O grid substituted water savings litre
xSOx, xNOx, xPEa grid substituted atmospheric emissions g or kg
xCoal, xAsh grid fuel/ash reduction kg or ton
FLg farmland gearing ratio for FPV generation gain over GPV %
FLgϵroad space-loss coefficient for GPV %
FLp farmland preserved through FPV as avoided land-use m2
FLc harvesting yield, food-commodity capacity kg/m2/year
FLx annual food/fruit production capacity units/m2/year
FLx2 translated into a downstream commodity units/m2/year
avFLx, avFLx2 agrivoltaic production factors kg/m2/year
Xx, Xc agro-produce and retail sales price points Rand/ton
GFI, NFI gross and nett farming income factors Rand/m2/year
xTr electricity grid-price policy tariff Rand
xTre annual tariff escalation growth rate %
Rcfl cost-of-farmland factor R/m2
Rxi agricultural farmland cost saving Rand
Rgs avoided energy purchase costs Rand
Rcc potential eco-tax or carbon-credit incomes Rand
Rtc renewable-Tax credits or solar-Tax credits Rand
Rwe water-pumping/acquisition cost savings Rand
Rxp food-crop harvest income Rand
Rcs value-added income from processed commodity Rand
avRxp annual aqua/agrivoltaic production income R/year
Notation Description Unit
Continued on next page
295
Table N.1: FPV-GIS model parameter notations and units of measure (source: author) (Continued)
avRcs annual agrivoltaic processed commodity income R/year
Rzt total of all income streams Rand
LCOE levelised cost of electricity R/kWh
LACE levelised avoided cost of electricity R/kWh
NPV net present value Rand
NPC nett present cost Rand
Qbe project-investment break even point year
ROI project return on investment %
f-EST, w-EST water-economic surface-transformation R/m2/year
g-EST land-economic surface-transformation R/m2/year
ce-WELF carbon-economic-water-energy-land-food resources units
Yl lifetime of the project years
Notation Description Unit
296
O. Simulation Experiment Configuration
Table O.1: Simulation model config for FPV at Paarl RWE wine farm case study (source:
author)
TimeZone Africa/Johannesburg zone
Town Paarl place
Reservoir label RWE tag
Latitude -33 degree
Longitude 18 degree
Altitude 210 m
Site_Area_m2 1000 m2
Project_start_date 2023/02/01 date
GPV_space_loss_coeff 0.07 factor
Surface_Tilt_GPV 15 degree
Surface_Tilt_FPV 15 degree
Surface_Azimuth_GPV 0 degree
Surface_Azimuth_FPV 0 degree
Albedo_land 0.2 unit
Albedo_water 0.2 unit
Sandia_module Canadian_Solar_CS5P_220M name
Module_rating 200 watt
Panel_dense_100m2_GPV 51 panels/100m2
Panel_dense_100m2_FPV 51 panels/100m2
Panel_degrade_rate_GPV 0.015 factor/year
Panel_degrade_rate_FPV 0.006 factor/year
Panel_coeff_a -3.47 number
Panel_coeff_b -0.0594 number
Install Config Parameter Setting Unit
Continued on next page
297
Table O.1: Simulation model config for FPV at Paarl RWE wine farm case study (source:
author) (Continued)
Panel_coeff_deltaT 3 number
Pvfloat_type tubular buoyancy design type
PVfloat_coeff_a 0.8 microclimate factor
PVfloat_coeff_b 1.5 offset temp factor
PVfloat_coeff_c 0.6 rrad flux factor
PVfloat_coeff_d 1.5 RH factor
PVfloat_coeff_e 1.2 Wspd factor
Grid_Ltrs_Water_reduce 1.37 grid rate L/kW
Grid_Kg_Coal_reduce 0.54 grid rate kg/kW
Grid_gr_Ash_reduce 155 grid rate gr/kW
Grid_gr_SO2_reduce 7.93 grid rate gr/kW
Grid_gr_NOx_reduce 4.19 grid rate gr/kW
Grid_Kg_CO2_reduce 0.99 grid rate kg/kW
Grid_gr_aPE_reduce 0.31 grid rate gr/kW
Crop_seq_CO2_kgm2yr 0.5 kgm2yr
Land_m2_purc_price 2500 currency/m2
Land_m2_rentprice 0 currency/m2
Water_m2_rentprice 0 currency/m2
Capex_assetcost_GPV 1200000 currency
Capex_assetcost_FPV 1200000 currency
Opex_yrcost_GPV 35000 currency
Opex_yrcost_FPV 35000 currency
Opex_yrincr_GPV 0.02 factor
Opex_yrincr_FPV 0.02 factor
Subsidy_incent_GPV 0 currency
Subsidy_incent_FPV 0 currency
Bank_loan 0 currency
Interest_rate_loan 0.09 % per year
Interest_rate_earn 0.05 % per year
Project_IRRdisct_rate 0.075 rate
Inflation_rate 0.035 %/year
Install Config Parameter Setting Unit
Continued on next page
298
Table O.1: Simulation model config for FPV at Paarl RWE wine farm case study (source:
author) (Continued)
xGrid_Tariff 1.875 currency/kWh
xTariff_increase 0.1509 factor/year
ecoTax_Rate 0.1 factor/year
eTax_increase 0.01 factor/year
CO2credits_Price_ton 155 currency/ton
CO2credits_price_incrse 0.02 rate/year
Water_cost_ltr 0.027 currency/litre
Water_cost_incr 0.01 rate/year
Agro_crop_price_kg 15.288 currency/kg
Agro_crop_prod_kgm2yr 1.6 kgm2yr
Agro_crop_prod_incr 0.02 factor/year
Agro_fresh2comm_units 0.756 units/kg
Agro_procd_comm_price 300 currency/unit
Comdity_over_FreshFd 0 income: wine=1 grapes=0
Agrivolt_yn_GPV 0 binary yes=1,no=0
Agrivolt_GPV_crop_kgm2yr 0 kgm2yr
Agrivolt_yn_FPV 0 binary yes=1,no=0
Agrivolt_FPV_crop_kgm2yr 0 kgm2yr
Project_lifetime 20 years
AHP_wght_Climate 0.296 AHP probability
AHP_wght_Economic 0.18 AHP probability
AHP_wght_Water 0.133 AHP probability
AHP_wght_Energy 0.237 AHP probability
AHP_wght_Land 0.087 AHP probability
AHP_wght_Food 0.066 AHP probability
Install Config Parameter Setting Unit
299
P. Geosensor Meteorological Data Input Samples
Table P.1: TMY meteorological geosensor input dataset sample for Paarl RWE wine farm site
(source: author)
Date <··· 2022/05/01 2022/05/01 2022/05/01 2022/05/01 ··· >
Time <··· 11:00 12:00 13:00 14:00 ··· >
Direct Inclined <··· 641 601 306 286 ··· >
Diffuse Inclined <··· 183 183 227 172 ··· >
Reflected <··· 6 6 4 3 ··· >
Global Inclined <··· 830 789 537 461 ··· >
Direct Horiz <··· 471 440 221 197 ··· >
Diffuse Horiz <··· 155 157 208 154 ··· >
Global Horiz <··· 626 597 429 351 ··· >
Clear-Sky <··· 677 659 578 442 ··· >
Top of Atmosphere <··· 911 893 800 641 ··· >
Reliability Code <··· 2222··· >
Temperature <··· 290.29 290.38 290.38 290.38 ···>
Celsius <··· 17.14 17.23 17.23 17.23 ··· >
Relative Humidity <··· 79.73 78.89 78.66 78.51 ··· >
Pressure <··· 1015.75 1015.35 1015.12 1015.25 ··· >
Wind speed <··· 1.96 2.56 3.18 3.86 ··· >
Wind direction <··· 219.96 213.88 209.67 204.72 ···>
Rainfall <··· 0.003 0.002 0.001 0.000 ··· >
Snowfall <··· 0000···>
Snow depth <··· 0 0 0 0 ··· >
Geosensor Sample <··· 11:00 12:00 13:00 14:00 ··· >
300
Q. FPV-GIS Python Model GUI Interface Screens
(a) (b)
Figure Q.1: User Interface (UI) of the Python development platform showing the simulation runtime startup screen-
shots to illustrate (a) the reading of floating PV system user configuration from the Excel configuration file and (b) the
loading of the weather station data for the FPV site location from the meteorological data repository - both during the
FPV model’s simulation runtime startup (source: author).
301
Figure Q.2: Graphical User Interface (GUI) visual components of the Python development platform runtime simu-
lation screenshots to illustrate the model’s interaction through graphic icons and visual indicators as visual indicator
representations and visual metric representations (source: author).
302
R. Sampled Real-time Simulation Results Extract
Table R.1: Data extract of experimentally predicted real-time simulation results (source: au-
thor)
Time <··· 2023/05/01 2023/05/01 2023/05/01 ··· >
GeoSensor_DNI <··· 717 573 407 ··· >
GeoSensor_DHI <··· 527 414 284 ··· >
GeoSensor_GHI <··· 653 550 409 ··· >
GeoSensor_Tdry <··· 18.54 18.59 18.61 ··· >
GeoSensor_Twet <··· 15.5 15.91 16.1 ··· >
GeoSensor_RH <··· 109.4 113.15 114.99 ···>
GeoSensor_Wspd <··· 4.98 4.74 4.69 ··· >
T_ambient_FPV <··· 13.47 13.8 13.96 ··· >
T_wetblb_FPV <··· 15.5 15.91 16.1 ··· >
OnPanel_poa_direct_GPV <··· 533.07 470.15 340.34 ···>
OnPanel_poa_direct_FPV <··· 533.07 470.15 340.34 ···>
OnPanel_poa_global_GPV <··· 1134.18 927.25 644.21 ··· >
OnPanel_poa_global_FPV <··· 1134.18 927.25 644.21 ··· >
Panel_Temp_GPV <··· 49.52 44.19 36.43 ···>
Panel_Temp_FPV <··· 43.13 38.36 31.07 ···>
Panel_Power_GPV <··· 113223.57 94792.44 67910.28 ··· >
Panel_Power_FPV <··· 117080.35 97670.92 69771.27 ··· >
Int_Panel_Power_GPV <··· 70174899.4 70269691.84 70337602.12 ··· >
Int_Panel_Power_FPV <··· 73186972.2 73284643.12 73354414.39 ··· >
FpvGpv_Eff_gain <··· 3.29 2.95 2.67 ··· >
Solar2Electr_GPV <··· 19.45 19.96 20.59 ··· >
Solar2Electr_FPV <··· 20.11 20.57 21.16 ··· >
Python Indicator Metric <··· 11:00 12:00 13:00 ··· >
Continued on next page
303
Table R.1: Data extract of experimentally predicted real-time simulation results (source: au-
thor) (Continued)
mm_Evapo_Avoid_FPV <··· 0.21 0.19 0.17 ··· >
Ltrs_Evapo_Avoid_FPV <··· 205.91 186.03 173.19 ··· >
Sum_Ltrs_Evapo_FPV <··· 261225.51 261411.54 261584.73 ··· >
Ltrs_Water_reduce_GPV <··· 155.12 129.87 93.04 ··· >
Ltrs_Water_reduce_FPV <··· 160.4 133.81 95.59 ···>
Kg_Coal_reduce_GPV <··· 61.14 51.19 36.67 ··· >
Kg_Coal_reduce_FPV <··· 63.22 52.74 37.68 ··· >
gr_Ash_reduce_GPV <··· 17549.65 14692.83 10526.09 ··· >
gr_Ash_reduce_FPV <··· 18147.45 15138.99 10814.55 ··· >
gr_SO2_reduce_GPV <··· 897.86 751.7 538.53 ··· >
gr_SO2_reduce_FPV <··· 928.45 774.53 553.29 ··· >
gr_NOx_reduce_GPV <··· 474.41 397.18 284.54 ··· >
gr_NOx_reduce_FPV <··· 490.57 409.24 292.34 ··· >
Kg_CO2_reduce_GPV <··· 112.09 93.84 67.23 ···>
Kg_CO2_reduce_FPV <··· 115.91 96.69 69.07 ···>
gr_aPE_reduce_GPV <··· 35.1 29.39 21.05 ···>
gr_aPE_reduce_FPV <··· 36.29 30.28 21.63 ··· >
Int_Ltrs_Wtr_grd_GPV <··· 96139.61 96269.48 96362.51 ··· >
Int_Ltrs_Wtr_grd_FPV <··· 100266.15 100399.96 100495.55 ··· >
Int_Kg_CO2_reduce_GPV <··· 69473.15 69566.99 69634.23 ··· >
Int_Kg_CO2_reduce_FPV <··· 72455.1 72551.8 72620.87 ··· >
Revenue_Electr_GPV <··· 212.29 177.74 127.33 ··· >
Revenue_Electr_FPV <··· 219.53 183.13 130.82 ··· >
Sum_Revenue_Electr_GPV <··· 131577.94 131755.67 131883 ··· >
Sum_Revenue_Electr_FPV <··· 137225.57 137408.71 137539.53 ···>
Revenue_CO2credits_GPV <··· 17.37 14.55 10.42 ··· >
Revenue_CO2credits_FPV <··· 17.97 14.99 10.71 ··· >
Revenue_evapoWater_FPV <··· 5.56 5.02 4.68 ··· >
Python Indicator Metric <··· 11:00 12:00 13:00 ··· >
304
S. Sampled Life-time Simulation Results Extract
Table S.1: Data extract of experimentally predicted life-time simulation results (source: au-
thor)
Projct_Year 1 10 20
Pm_GPV 265006207.3 231303455.7 198858622.2
Pm_FPV 276216050.2 261653392.7 246371356.2
Eff_gain_FPV 4.06 11.6 19.29
Sum_Pm_GPV 265006207.3 2478153567 4608697629
Sum_Pm_FPV 276216050.2 2688762982 5220487016
Sum_Eff_gain_FPV 4.06 78.75 237.59
FLp_FPV 1120.75 1216.36 1332.18
H2O_evap_ltr_FPV 983501.99 983501.99 983501.99
H2O_grid_ltr_GPV 363058.5 316885.73 272436.31
H2O_grid_ltr_FPV 378415.99 358465.15 337528.76
Sum_H2O_evap_FPV 983501.99 9835019.89 19670039.78
Sum_H2O_grid_GPV 363058.5 3395070.39 6313915.75
Sum_H2O_grid_FPV 378415.99 3683605.29 7152067.21
FreshFood_GPV 0 0 0
FreshFood_FPV 1793.2 2325.86 3105.18
ProcdFood_GPV 0 0 0
ProcdFood_FPV 1355.66 1758.35 2347.51
Sum_FreshFood_GPV 0 0 0
Sum_FreshFood_FPV 1793.2 20492.8 47852.08
Sum_ProcdFood_GPV 0 0 0
Sum_ProcdFood_FPV 1355.66 15492.56 36176.17
kgCO2_plnt_GPV 0 0 0
Python Indicator Metric Year 2023 Year 2033 Year 2043
Continued on next page
305
Table S.1: Data extract of experimentally predicted life-time simulation results (source: au-
thor) (Continued)
kgCO2_plnt_FPV 560.38 608.18 666.09
kgCO2_grid_GPV 262356.15 228990.42 196870.04
kgCO2_grid_FPV 273453.89 259036.86 243907.64
kgCoal_GPV 143103.35 124903.87 107383.66
kgCoal_FPV 149156.67 141292.83 133040.53
grAsh_GPV 41075962.14 35852035.63 30823086.45
grAsh_FPV 42813487.78 40556275.86 38187560.22
grSOx_GPV 2101499.22 1834236.4 1576948.87
grSOx_FPV 2190393.28 2074911.4 1953724.85
grNOx_GPV 1110376.01 969161.48 833217.63
grNOx_FPV 1157345.25 1096327.72 1032295.98
gr_aPE_GPV 82151.92 71704.07 61646.17
gr_aPE_FPV 85626.98 81112.55 76375.12
Sum_kgCO2_plnt_GPV 0 0 0
Sum_kgCO2_plnt_FPV 560.38 5839.88 12235.83
Sum_kgCO2_grid_GPV 262356.15 2453372.03 4562610.65
Sum_kgCO2_grid_FPV 273453.89 2661875.35 5168282.15
Sum_kgCoal_GPV 143103.35 1338202.93 2488696.72
Sum_kgCoal_FPV 149156.67 1451932.01 2819062.99
Sum_grAsh_GPV 41075962.14 384113802.9 714348132.5
Sum_grAsh_FPV 42813487.78 416758262.3 809175487.4
Sum_grSOx_GPV 2101499.22 19651757.78 36546972.2
Sum_grSOx_FPV 2190393.28 21321890.45 41398462.03
Sum_grNOx_GPV 1110376.01 10383463.44 19310443.06
Sum_grNOx_FPV 1157345.25 11265916.9 21873840.6
Sum_gr_aPE_GPV 82151.92 768227.61 1428696.26
Sum_gr_aPE_FPV 85626.98 833516.52 1618350.97
LandValue_FLp_FPV 2801882.53 3040894.19 3330449.46
Rgs_GPV 496886.64 1536461.58 5385920.14
Rgs_FPV 517905.09 1738064.76 6672762.96
Rcc_GPV 40665.2 42418.04 44454.38
Python Indicator Metric Year 2023 Year 2033 Year 2043
Continued on next page
306
Table S.1: Data extract of experimentally predicted life-time simulation results (source: au-
thor) (Continued)
Rcc_FPV 42385.35 47983.82 55075.74
Rwe_GPV 0 0 0
Rwe_FPV 26554.55 29042.32 32080.79
Rxp_GPV 0 0 0
Rxp_FPV 27414.52 35557.69 47471.92
Rcs_GPV 0 0 0
Rcs_FPV 406698.85 527504.15 704253.7
Rzt_GPV 537551.84 1578879.61 5430374.52
Rzt_FPV 614259.52 1850648.59 6807391.41
Sum_Rgs_GPV 496886.64 9315585.82 41970487.22
Sum_Rgs_FPV 517905.09 10211713.5 49416434.37
Sum_Rcc_GPV 40665.2 415361.4 850662.9
Sum_Rcc_FPV 42385.35 451331.51 969369.03
Sum_Rwe_GPV 0 0 0
Sum_Rwe_FPV 26554.55 277819.38 584704.82
Sum_Rxp_GPV 0 0 0
Sum_Rxp_FPV 27414.52 313293.98 731562.62
Sum_Rcs_GPV 0 0 0
Sum_Rcs_FPV 406698.85 4647767.83 10852852.08
Sum_Rzt_GPV 537551.84 9730947.22 42821150.12
Sum_Rzt_FPV 614259.52 11254158.37 51702070.84
Capex_GPV 1200000 0 0
Capex_FPV 1200000 0 0
Opex_GPV 35000 41828.24 50988.39
Opex_FPV 35000 41828.24 50988.39
Sum_Capex_GPV 1200000 1200000 1200000
Sum_Capex_FPV 1200000 1200000 1200000
Sum_Opex_GPV 35000 383240.23 850407.94
Sum_Opex_FPV 35000 383240.23 850407.94
Costs_Yr_GPV 1235000 109297.17 387345.73
Costs_Yr_FPV 1235000 109297.17 387345.73
Python Indicator Metric Year 2023 Year 2033 Year 2043
Continued on next page
307
Table S.1: Data extract of experimentally predicted life-time simulation results (source: au-
thor) (Continued)
LCOE_GPV_numerator 1235000 1860146.42 4199684.57
LCOE_GPV_denomintr 265006.21 4142076.49 14495132.4
LCOE_FPV_numerator 1235000 1860146.42 4199684.57
LCOE_FPV_denomintr 276216.05 4529731.5 16929810.29
LCOE_elect_Yr_GPV 4.66 0.45 0.29
LCOE_elect_Yr_FPV 4.47 0.41 0.25
Grid_tariff_Yr 1.88 6.64 27.08
LACE_elect_Yr_GPV -2.79 6.19 26.79
LACE_elect_Yr_FPV -2.6 6.23 26.84
NPV_elect_inc_GPV -738113.36 3905476.16 40528108.44
NPV_elect_inc_FPV -717094.91 4432265.16 50303923.38
NPV_elect_Yr_GPV -738113.36 15617010.97 191391098.2
NPV_elect_Yr_FPV -717094.91 17428892.83 230386604.3
NPV_projt_inc_GPV -697448.16 4016314.47 40865816.98
NPV_projt_inc_FPV -620740.48 4726446.66 51326661.21
NPV_projt_Yr_GPV -697448.16 16323271.62 194249234.4
NPV_projt_Yr_FPV -620740.48 19220724.78 238394277.2
ROI_elect_Yr_GPV -161.51 1201.42 15849.26
ROI_elect_Yr_FPV -159.76 1352.41 19098.88
ROI_projt_Yr_GPV -158.12 1260.27 16087.44
ROI_projt_Yr_FPV -151.73 1501.73 19766.19
CashFlow_elect_GPV -738113.36 15617010.97 191391098.2
CashFlow_elect_FPV -717094.91 17428892.83 230386604.3
CashFlow_projt_GPV -697448.16 16323271.62 194249234.4
CashFlow_projt_FPV -620740.48 19220724.78 238394277.2
EST_GPV 537.55 1578.88 5430.37
EST_FPV 614.26 1850.65 6807.39
Python Indicator Metric Year 2023 Year 2033 Year 2043
308
T. Sampled Simulated AHP Results Extract
Table T.1: Data extract of empirical AHP decision-support indicators (source: author)
Projct_Year 1 10 20
C_welf_GPV 262356.15 228990.42 196870.04
C_welf_FPV 273453.89 259036.86 243907.64
M_welf_GPV 537.55 1578.88 5430.37
M_welf_FPV 614.26 1850.65 6807.39
W_welf_GPV 363058.5 316885.73 272436.31
W_welf_FPV 1361917.98 1341967.14 1321030.75
E_welf_GPV 265006207.3 231303455.7 198858622.2
E_welf_FPV 276216050.2 261653392.7 246371356.2
L_welf_GPV 0 0 0
L_welf_FPV 1120.75 1216.36 1332.18
F_welf_GPV 0 0 0
F_welf_FPV 1793.2 2325.86 3105.18
C_welf_prob_GPV 0.49 0.47 0.45
C_welf_prob_FPV 0.51 0.53 0.55
M_welf_prob_GPV 0.47 0.46 0.44
M_welf_prob_FPV 0.53 0.54 0.56
W_welf_prob_GPV 0.21 0.19 0.17
W_welf_prob_FPV 0.79 0.81 0.83
E_welf_prob_GPV 0.49 0.47 0.45
E_welf_prob_FPV 0.51 0.53 0.55
L_welf_prob_GPV 0 0 0
L_welf_prob_FPV 1 1 1
Softw Decision Indicator Year 2023 Year 2033 Year 2043
Continued on next page
309
Table T.1: Data extract of empirical AHP decision-support indicators (source: author) (Con-
tinued)
F_welf_prob_GPV 0 0 0
F_welf_prob_FPV 1 1 1
C_welf_AHPp_GPV 0.14 0.14 0.13
C_welf_AHPp_FPV 0.15 0.16 0.16
M_welf_AHPp_GPV 0.08 0.08 0.08
M_welf_AHPp_FPV 0.1 0.1 0.1
W_welf_AHPp_GPV 0.03 0.03 0.02
W_welf_AHPp_FPV 0.11 0.11 0.11
E_welf_AHPp_GPV 0.12 0.11 0.11
E_welf_AHPp_FPV 0.12 0.13 0.13
L_welf_AHPp_GPV 0 0 0
L_welf_AHPp_FPV 0.09 0.09 0.09
F_welf_AHPp_GPV 0 0 0
F_welf_AHPp_FPV 0.07 0.07 0.07
AHP_Yr_score_GPV 0.37 0.36 0.34
AHP_Yr_score_FPV 0.63 0.64 0.66
AHP_Life_score_GPV 0.37 3.66 7.15
AHP_Life_score_FPV 0.63 6.33 12.83
Enviro_H2O_prob_GPV 0.27 0.24 0.21
Enviro_H2O_prob_FPV 1 1 1
Enviro_CO2_prob_GPV 0.19 0.17 0.15
Enviro_CO2_prob_FPV 0.2 0.19 0.18
Enviro_Coal_prob_GPV 0.11 0.09 0.08
Enviro_Coal_prob_FPV 0.11 0.11 0.1
Enviro_Ash_prob_GPV 0.030 0.026 0.023
Enviro_Asv_prob_FPV 0.031 0.030 0.028
Enviro_SOx_prob_GPV 1.54 1.37 1.19
Enviro_SOx_prob_FPV 1.61 1.55 1.48
Enviro_NOx_prob_GPV 0.82 0.72 0.63
Enviro_NOx_prob_FPV 0.85 0.82 0.78
Enviro_aPE_prob_GPV 0.06 0.05 0.05
Softw Decision Indicator Year 2023 Year 2033 Year 2043
Continued on next page
310
Table T.1: Data extract of empirical AHP decision-support indicators (source: author) (Con-
tinued)
Enviro_aPE_prob_FPV 0.06 0.06 0.06
Enviro_FLp_prob_GPV 0 0 0
Enviro_FLp_prob_FPV 0 0 0
Econo_Rgs_Prob_GPV 0.96 0.88 0.81
Econo_Rgs_Prob_FPV 1 1 1
Econo_Rcc_Prob_GPV 0.08 0.02 0.01
Econo_Rcc_Prob_FPV 0.08 0.03 0.01
Econo_Rct_Prob_GPV 0 0 0
Econo_Rct_Prob_FPV 0 0 0
Econo_Rwe_Prob_GPV 0 0 0
Econo_Rwe_Prob_FPV 0.05 0.02 0
Econo_Rxp_Prob_GPV 0 0 0
Econo_Rxp_Prob_FPV 0.05 0.02 0.01
Econo_Rcs_Prob_GPV 0 0 0
Econo_Rcs_Prob_FPV 0.79 0.3 0.11
Softw Decision Indicator Year 2023 Year 2033 Year 2043
311
Figure U.2: Sample EIA scoping assessment report and PV project performance scorecard (source: author).
Energy Project EIA scoping report summary submitted in terms of the Environmental Impact
Assessment (EIA) queries, as per the National Environmental Management Act (No. 107 of
1998) and associated EIA Regulations.
(a) Energy project EIA report summary: Operations phase
Municipal jurisdiction Paarl, Drakenstein Municipality, Western Cape, South Africa
Property description Agricultural production facility, characterised by agronomic
food production, viticulture and wine farming activities
Project location 33◦46’37.6"S, 18◦54’03.7"E
Project description Proposed on-side installation of floating PV power generation
system on a local irrigation reservoir
Project lifetime 20 years
Technical capacity Technical capacity of PV power generation system is rated at
1112.20 kWh, to be connected in Eskom grid substitution mode
(b) Proposed Energy Project: Water-mounted Floating PhotoVoltaic system
Activities Adverse
effect
No effect Beneficial
effect
Quantify effect
Space transformation X Reclaiming fertile
agricultural land for PV
energy generation on unused
irrigation water-surface
Land compaction X Land compaction only during
installation process
Land use X Reclaiming 1283 m2of
productive food production
farmland
Soil quality X Disturbance of soil quality
is not foreseen
Pesticides/fertiliser X Use of chemical pesticides
and fertiliser not foreseen
Carbon footprint X Carbon sequestration
potential CO2e=5168 tons
over project lifetime
313
Activities Adverse
effect
No effect Beneficial
effect
Quantify effect
Air quality X In grid-substitution mode,
improvement in air quality
over project lifetime
includes COx=5168 ton,
NOx=21.87 ton, SOx=41.40 ton
of atmospheric emissions
displaced
Particulate matter X In grid-substitution
mode, avoided particulate
emissions aPE=1618.35 kg
over project lifetime
Water consumption X Grid substitution avoiding
7152 kL in water cooling at
power station over project
lifetime, insignificant
local water loss due to PV
panel cleaning
Water processing X FPV panel shading reducing
water evaporation by
19670 kL over project
lifetime
Water quality X Reducing poisonous algal
growth, impact to be
quantified
Noise & vibration X PV system operations silent,
noise and vibration free
Transport X No operational transport
impact, energy transport by
electrical cabling
(c) Energy Project alternative: Ground-mounted PhotoVoltaic system
Activities Adverse
effect
No effect Beneficial
effect
Describe & quantify effect
Space transformation X Reconfiguration of fertile
agricultural land into PV
energy generation on former
productive farmland
Land compaction X Land compaction both during
installation and 20 year
maintenance period
Land use X Sacrificing of 1283 m2
farmland for energy
production
Soil quality X Long-term adverse effects in
terms of concrete works,
concrete foundations,
removal of agronomic
rootstock and plant material
314
Activities Adverse
effect
No effect Beneficial
effect
Quantify effect
Pesticides/fertiliser X Use of chemical pesticides
required to control weed
growth under PV panels
Carbon footprint X Carbon sequestration
potential CO2e=4562 ton over
project lifetime
Air quality X In grid-substitution mode,
improvement in air quality
over project lifetime
includes COx=4562 ton,
NOx=19.31 ton, SOx=36.55 ton
of atmospheric emissions
displaced
Particulate matter X In grid-substitution
mode, avoided particulate
emissions aPE=1428.69 ton
Water consumption X Grid substitution avoiding
6313 kL in water cooling at
power station over project
lifetime
Water processing X Minimal local water usage
due to PV panel cleaning
Water quality X Minimal positive or adverse
impact of water quality
foreseen
Noise & vibration X PV system operations silent,
noise and vibration free
Transport X Light-weight vehicles in
system maintenance and panel
cleaning, energy transport
by electric cabling
DECLARATION BY THE PROPONENT / ENVIRONMENTAL PRACTITIONER
IFrederik Christoffel Prinsloo in my professional capacity or duly authorised hereby declare that:
• regard the information contained in this checklist to be true and correct;
• am fully aware of my responsibilities in terms of the National Environmental Management Act (NEMA) Act No.
107 of 1998), the Environmental Impact Assessment Regulations (EIA Regulations), 2010 in terms of NEMA
(Government Notice No. R.543 refers) and the relevant specific environmental management Acts, and that
failure to comply with these requirements may constitute an offence in terms of the environmental legislation;
• am fully aware that the Department’s determination of the applicability of the EIA Regulations, 2010 is based
on information at my disposal that is relevant to this request;
• aware that the response from the competent authority, to this request, is specific to the EIA Regulations, 2010
and does not exempt me from my legal obligations in terms of any other applicable legislation.
................................................................................ ...................
Signature of proponent / environmental practitioner: Date:
315
V. Proposed Model Additions for Future Research
Technological
Energy
Simulation
Model
Object
Economical
Simulation
Model
Object
Environmental
Simulation
Model
Object
Hydropower,
Ecological,
or Social
Model
Object(s)
f(x)e(t)f(y)e(t)
f(z)e(t)
f(h1)e(t)f(h2)e(t)
f(h3)e(t)
Power Pe−mp (kWh)
Volts Ve−mp (V)
Ampere Ie−mp (A)
Panel temps (C)
Project lifetime specifications
Site location configurations
Geometric design & surface area
Technical floater design parameters
Technical FPV design configurations
Reduced CO2(kg)
H2O saved (kg)
Reduced SOx(kg)
Reduced NOx(kg)
Reduced aPE (kg)
Reduced Coal (kg)
Solar irradiation
Cloud modulation
Climatic data
Weather data
Revenue/Savings (R)
LCOE, NPV (R/kWh)
Payback period (yrs)
ecoTax credits (R)
Spare (R)
Capex & Opex costs (R)
Incentives, funding & subsidies (R)
Electricity tariff rates (R/KWh)
Carbon credit pricing (R/ton)
Discount & interest rates (%)
ecoTaxes savings & rates (%)
Impact effect 1
Impact effect 2
Impact effect 3
Impact effect 4
Impact effect 5
Impact effect 6
Parameter 1
Parameter 2
Parameter 3
Parameter 4
Parameter 5
South African geospatial, environmental, legislative, economic and technical context
Figure V.1: Block diagram for proposed future model integration of empirical hydropower / ecological / social -impact
domain objects into a predictive multi-core framework model architecture for floating PV systems, to accommodate
hybrid hydropower-floatovoltaic sustainability integration, ecological sustainability modelling for floatovoltaic integra-
tion, and social sustainability modelling for floatovoltaic integration (source: author).
316
W. Potential for Floating PV Hydroelectrics in Africa
Figure W.1: Towards establishing hybrid FPV-hydro power plants throughout the African continent. The FPV-GIS
model is suitable for studying the technical, environmental and economic sustainability potential for installing floating
photovoltaic systems near hydropower generation stations and utility grid facilities as virtual energy batteries (source:
Gonzalez Sanchez et al. (2021), page 7).
317
References
4IRSA (2020). Digital Economy Summit towards the 4th Industrial Revolution Partnership for South Africa (Fourth
Industrial Revolution). Stimulating Dialogue, Understanding and Action to Shape a Coherent 4IR Plan for SA,
vol. 1, no. 2, pp. 1–7.
Ababneh, H. and Hameed, B. (2022). Electrofuels as emerging new green alternative fuel: A review of recent literature.
Energy Conversion and Management, vol. 254, no. 115213, pp. 1–23.
Abbasnia, A., Karimirad, M., Friel, D. and Whittaker, T. (2022). Fully nonlinear dynamics of floating solar platform with
twin hull by tubular floaters. Ocean Engineering, vol. 257, no. 111320, pp. 1–12.
Abd-Elhamid, H.F., Ahmed, A., Zelenakova, M., Vranayova, Z. and Fathy, I. (2021). Reservoir Management by Reduc-
ing Evaporation Using Floating Photovoltaic System: A Case Study of Lake Nasser, Egypt. Water, vol. 13, no. 6,
pp. 1–20.
Abdelal, Q. (2021). Floating PV: An assessment of water quality and evaporation reduction in semi-arid regions.
International Journal of Low-Carbon Technologies, pp. 1–8.
Abdelmoula, I.A., Elhamaoui, S., Elalani, O., Ghennioui, A. and El Aroussi, M. (2022). A photovoltaic power prediction
approach enhanced by feature engineering and stacked machine learning model. Energy Reports, vol. 8, pp.
1288–1300.
Abdullah, G. and Nishimura, H. (2021). Energetic, Economic and Environmental Analyses for Photovoltaic System in
Different Countries. IEEE 4th International Conference on Renewable Energy and Power Engineering, vol. 2, no. 1,
pp. 249–253.
Abdullahi, N., Saha, C. and Jinks, R. (2017). Modelling and performance analysis of a silicon PV module. Journal of
Renewable and Sustainable Energy, vol. 9, no. 33501, pp. 1–11.
Abid, M., Abid, Z., Sagin, J., Murtaza, R., Sarbassov, D. and Shabbir, M. (2019). Prospects of floating photovoltaic
technology and its implementation in Central and South Asian Countries. International Journal of Environmental
Science and Technology, vol. 16, no. 3, pp. 1755–1762.
Abid, M., Abid, Z. and Umer, M. (2021). Floating Photovoltaic System Technolog-Prospects of Its Implementation in
Central Asian, South Asian and South East Asian Region. In: Energy and Environmental Security in Developing
Countries, pp. 633–658. Springer.
Acharya, M. and Devraj, S. (2019). Floating Solar Photovoltaic (FSPV): A Third Pillar to Solar PV Sector. Tech. Rep.,
The Energy and Resources Institute, New Delhi.
Acker, O., Siri, N. and Mengue, C. (2020). COVID-19: A digital technology agenda driving an accelerated transition to
the new normal. Tech. Rep., PWC, Munich, Germany.
Advanced Energy Economy (2019). Putting Distributed Energy Resources to Work in Wholesale Electricity Markets:
Case Studies of Emerging Applications and Their Benefits for Customers and the Grid. Advanced Energy Economy
Institute, vol. 1, no. 2, pp. 1–13.
Afshar, A., Soleimanian, E., Akbari Variani, H., Vahabzadeh, M. and Molajou, A. (2021). The conceptual framework
to determine interrelations and interactions for holistic Water, Energy, and Food Nexus. Environment Development
and Sustainability, pp. 1–22.
Aghaei, M., KorevaarM, M., Babal, P. and Ziar, H. (2022). Bifacial Photovoltaic Technology: Recent Advancements,
Simulation and Performance Measurement. In: Solar Radiation: Measurements, Modeling and Forecasting for
Photovoltaic Solar Energy Applications, Reliability and Durability of Photovoltaics, pp. 1–133. IntechOpen.
318
Ahn, C.M., Joo, J.C., Kim, J.H., Choi, S.H., Jang, J.S. and Go, H.W. (2021). Review of Installation Status and Major
Environmental Issues of Floating Photovoltaic Power Plants (FPVs). Journal of Korean Society of Environmental
Engineers, vol. 43, no. 4, pp. 286–298.
Al-Subhi, A., El-Amin, I. and Mosaad, M.I. (2020). Efficient predictive models for characterization of photovoltaic
module performance. Sustainable Energy Technologies and Assessments, vol. 38, p. 100672.
Alam, M. (2022). Effective Utilization of Water Resources Through Floating Solar Photovoltaic (FSPV) Technology: An
Emerging Idea of Sustainable Development. Advanced Modelling and Innovations in Water Resources Engineering,
vol. 176, pp. 759–768.
Alaoui, A., Ferreira, C.S.S. and Hessel, R. (2022). An Overview of Sustainability Assessment Frameworks in Agricul-
ture. Land, vol. 11, no. 537, pp. 1–26.
Alkemade, R., Bouwman, L., Biemans, H., de Elzen, M., Hilderink, H., Klein Goldewijk, K., Kram, T., Lucas, P. and
Olivier, J. (2012). The IMAGE model suite used for OECD Environmental Outlook to 2050. Tech. Rep., PBL
Netherlands Environmental Assessment Agency.
Allen, G. and Prinsloo, F. (2018). Floating Solar System Site Visit and Meeting at the Far Niente Winery, Napa Valley,
USA. Personal Interactions and Correspondence with Far Niente Engineers, Micro-utility Owners and Operators,
vol. 1, no. 1, pp. 1–10.
Allen, M., Tanaka, K., Macey, A., Cain, M., Jenkins, S., Lynch, J. and Smith, M. (2021). Ensuring that offsets and
other internationally transferred mitigation outcomes contribute effectively to limiting global warming. Environmental
Research Letters, vol. 16, no. 7, pp. 1–10.
Almeida, R., Schmitt, R. and Grodsky, S. (2022a). Floating solar power: Evaluate trade-offs. Nature Sustainability,
vol. 606, no. 1, pp. 246–249.
Almeida, R.M., Schmitt, R., Grodsky, S.M., Flecker, A.S., Gomes, C.P., Zhao, L., Liu, H., Barros, N., Kelman, R. and
McIntyre, P.B. (2022b). Floating solar power could help fight climate change: Let’s get it right.
Altamimi, A. and Jayaweera, D. (2021). Reliability of power systems with climate change impacts on hierarchical levels
of PV systems. Electric Power Systems Research, vol. 190, p. 106830.
Altegoer, D., Hussong, J. and Lindken, R. (2021). Efficiency Increase of Photovoltaic Systems by Means of Evapo-
rative Cooling. In: Conference: 16th Conference on Sustainable Development of Energy, Water and Environment
Systems, pp. 1–20. Dubrownik, Croatia.
Amadei, B. (2019). A Systems Approach to Modeling the Water-Energy-Land-Food Nexus. In: Volume II: System
Dynamics Modeling and Dynamic Scenario Planning, pp. 1–115. Momentum Press, New York, NY, USA. ISBN
9781947083554.
Anbari, F.T., Cioffi, D.F. and Forman, E.H. (2010). Integrating performance measures to exert effective leadership
in managing project portfolios. In: Defining the Future of Project Management, pp. 1–10. Newtown Square, PA:
Project Management Institute, Washington, DC.
Andini, S., Suwartha, N., Setiawan, E.A. and Maarif, S. (2022). Evaluating the Effect of Floating Photovoltaic on
Trophic State using Mesocosm Experiments. In: International Conference on Modern Approaches in Science,
Technology and Engineering, pp. 70–80.
Andrews, R. (2018). Sandia-labs PVLIB Python Package: Online GitHub. Available: https://github.com/Sandia-
Labs/PVLIB Python, [Accessed 10 Sep 2019].
Anker, S., Lindee, S., Shanken, E.A. and Nelkin, D. (2008). Technogenesis: Aesthetic Dimensions of Art and Biotech-
nology. In: Lustig, B.A., Brody, B.A. and McKenny, G.P. (eds.), Altering Nature: Concepts of Nature’ and the Natural
in Biotechnology, pp. 275–321. Springer Netherlands, Dordrecht. ISBN 9781402069215.
Anylogic (2020). System dynamics modeling in AnyLogic. AnyLogic Simulation Software: Make Intelligent Decisions
with System Dynamics Modeling, vol. 1, no. 2, pp. 1–15.
Armstrong, A., Page, T., Thackeray, S.J., Hernandez, R.R. and Jones, I.D. (2020). Integrating environmental un-
derstanding into freshwater floatovoltaic deployment using an effects hierarchy and decision trees. Environmental
Research Letters, vol. 2020, pp. 1–20.
ARUP (2019). Decision-Support Tool: The Sustainable Project Appraisal Routine SPeAR. Available at:
https://www.arup.com/projects/spear, [Accessed 12 Aug 2019].
319
Ascencio-Vasquez, J., Osorio-Aravena, J.C., Brecl, K., Munoz-Ceron, E. and Topic, M. (2021). Typical Daily Profiles:
A novel approach for photovoltaics performance assessment. Solar Energy, vol. 225, no. 7, pp. 357–374.
Ashby, R. (1961). An Introduction to Cybernetics. 4th edn. Chapman and Hall LTD, Lodon.
Ashley, P. and Boyd, B.W. (2006). Quantitative and qualitative approaches to research in environmental management.
Australasian Jnl of Environmental Management, vol. 13, no. 2, pp. 70–78.
Attar, S., Thakur, S., Kumar, N., Deshmukh, S. and Solanki, P.D. (2016). Potential of Sequestering Carbon through
Fruit Orchards. In: Forest and Tree Based Land Use Systems for Livelihood, Nutritional and Environmental Security,
pp. 209–216.
Ayani, M., Ganebäck, M. and Ng, A.H.C. (2018). Digital Twin: Applying emulation for machine reconditioning. Procedia
CIRP on Manufacturing Systems, vol. 72, pp. 243–248.
Aznar, A.Y., Lee, N. and Booth, S.S. (2019). International Applications for Floating Solar Photovoltaics. National
Laboratory of the U.S. Department of Energy, vol. 2, no. 1, pp. 1–4.
Banik, A. and Sengupta, A. (2021). Scope, Challenges, Opportunities and Future Goal Assessment of Floating Solar
Park. In: 2021 Innovations in Energy Management and Renewable Resources, IEMRE 2021, pp. 1–22.
Baptista, A., Frick, L., Holley, K., Remmik, M., Tesch, J. and Âkerlind, G. (2015). The doctorate as an original con-
tribution to knowledge: Considering relationships between originality, creativity, and innovation. Frontline Learning
Research, vol. 3, no. 3, pp. 55–67.
Barachini, F. and Stary, C. (2022). System-of-Systems Thinking. In: From Digital Twins to Digital Selves and Beyond,
chap. 9, pp. 77–79. Springer, Cham.
Baradei, S.E. and Sadeq, M.A. (2020). Effect of Solar Canals on Evaporation, Water Quality, and Power Production:
An Optimization Study. Water, vol. 12, no. 8, pp. 1–22.
Barbuscia, M. (2016). Preliminary Study on Floating Photovoltaic Systems on Dams. Concept Paper, vol. 1, pp. 1–40.
Barbuscia, M. (2018). Economic viability assessment of floating photovoltaic energy. Working Paper, vol. 2, no. 1, pp.
1–11.
Baring-Gould, I. (2014). Project Evaluation Models. Alaska Native Village Energy Development Workshop, vol. 1,
no. 4, pp. 1–22.
Batty, M. (2011). Modeling and simulation in geographic information science: Integrated models and grand challenges.
Procedia Social and Behavioral Sciences, vol. 21, no. 1, pp. 10–17.
Baum-Gartner, F. (2017). Photovoltaic (PV) balance of system components: Basics, performance. In: Pearsall, N.
(ed.), The Performance of Photovoltaic (PV) Systems, pp. 135–181. Woodhead Publishing. ISBN 978-1-78242-
336-2.
Bax, V., van de Lageweg, W.I., van den Berg, B., Hoosemans, R. and Terpstra, T. (2022). Will it float? Exploring the
social feasibility of floating solar energy infrastructure in the Netherlands. Energy Research and Social Science,
vol. 89, no. 3, pp. 102569–102580.
Bazzana, D., Castellini, M., Comincioli, N. and Vergalli, S. (2022). SDGs and WEF Nexus: A Critical Review of the
Analytical Tools. In: Cavalli, L. and Vergalli, S. (eds.), Connecting the Sustainable Development Goals: Under-
standing the Role of the WEF Nexus in the 2030 Agenda, pp. 23–34. Springer International Publishing, Cham.
ISBN 9783031013362.
Bellini, E. (2021). System design key to enhancing cooling effect of water in floating PV. PV Magazine USA, vol. 2,
no. 1, pp. 1–2.
Benson, D., Gain, A. and Rouillard, J. (2015). Water Governance in a Comparative Perspective: From IWRM to a
Nexus Approach? Water Alternatives, vol. 8, no. 1, pp. 756–773.
Bernardo, M., Cortellessa, V. and Pierantonio, A. (2012). Formal Methods for Model-Driven Engineering. In: Int
School on Formal Methods for the Design of Computer, Communication and Software Systems, Bertinoro, Italy,
pp. 1–200.
Besenicara, J., Trstenjakb, B. and Setnikac, D. (2008). Application of Geomatics in Photovoltaics. The International
Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. 37, no. 4, pp. 53–56.
Binder, C.R., Feola, G. and Steinberger, J.K. (2010). Considering the normative, systemic and procedural dimensions
in indicator-based sustainability assessments in agriculture. Environmental Impact Assessment Review, vol. 30,
320
no. 2, pp. 71–81.
Blignaut, A. (2017). South Africa’s draft carbon tax legislation and carbon offset regulations: How will it work? Blue
North Carbon Tax Review, vol. 1, no. 3, pp. 1–3.
Bloomberg Energy (2022). Bloomberg New Energy Financing and Long-Term Costing Forecasting Models. Available
at: https://about.bnef.com [Accessed 2 Feb 2022].
Blue North (2017). Development of Sustainability Self-Assessment Tools. Performance Assessment and Monitoring,
vol. 1, pp. 1–3.
Boersma, T., van der Laan, J., Noorduyn, O. and Mesbahi, M. (2019a). A Comprehensive Overview of 200 + Global
Floating Solar Plants. First Global Floating Solar Conference, vol. 1, pp. 1–19.
Boersma, T., van der Laan, J., Noorduyn, O. and Mesbahi, M. (2019b). Floating Solar Mythbusting: Separating Fact
from Fiction. First Global Floating Solar Conference, vol. 1, no. 2, pp. 1–13.
Bond, A., Morrison-Saunders, A. and Pope, J. (2012). Sustainability Assessment: The State of the Art. Impact
Assessment and Project Appraisal, vol. 30, no. 1, pp. 53–62.
Bonkaney, A., Madougou, S. and Adamou, R. (2017). Impacts of Cloud Cover and Dust on the Performance of
Photovoltaic Module. Journal of Renewable Energy, vol. 1, no. 2, pp. 1–8.
Borshchev, A. and Grigoryev, I. (2021). Big Book of Simulation Modeling: Multimethod Modeling with AnyLogic.
Available at: https://www.anylogic.com/resources/books/big-book-of-simulation-modeling/ [Accessed 5 Jan 2021].
Bourgeois, T. and Seman-Graves, J. (2022). Floating Solar Explorer. In: Clean Energy Generation Capacity of
Floating Solar on Reservoirs, pp. 1–5. Availabe at: https://olyerickson.shinyapps.io/cohoespvapp/ [Accessed 28
Jun 2022].
Brent, A., Silinga, C. and Sanetra, N. (2016). Guidelines for energy management in the South African wine industry.
Journal of Energy in Southern Africa, vol. 27, no. 4, pp. 53–63.
Buchmayr, A., Verhofstadt, E., Van Ootegem, L., S Delmás, D., Thomassen, G. and Dewulf, J. (2021). The path to
sustainable energy supply systems: Proposal of an integrative sustainability assessment framework. Renewable
and Sustainable Energy Reviews, vol. 138, no. 110666, pp. 1–17.
Bui, P.A. (2019). Assessing the technical, economic, and environmental feasibility of floating solar power generation
on water reservoirs in Vietnam. Ph.D. thesis, Humboldt State University.
Buitelaar, P., Cimiano, P., Frank, A., Hartung, M. and Racioppa, S. (2008). Ontology-based information extraction and
integration from heterogeneous data sources. International Journal of Human-Computer Studies, vol. 66, no. 11,
pp. 759–788.
Buonomano, A., Barone, G. and Forzano, C. (2022). Advanced energy technologies, methods, and policies to support
the sustainable development of energy, water and environment systems. Energy Reports, vol. 8, no. 1, pp. 4844–
4853.
Burge, S. (2015). An Overview of the Soft Systems Methodology. Burge Hughes Walsh, pp. 1–14.
Burke, M.J. (2018). Mutually-Beneficial Renewable Energy Systems. Relations Beyond Anthropocentrism, vol. 6,
no. 1, pp. 87–116.
Burns, N. and Grove, S. (2005). The Practice of Research: Conduct, Critique and Utilization. 5th edn. Elsevier
Science Publishers, Saint Louis, U.S.A.
Caboz, J. (2019). South African Farmers have built the First Floating Solar Park in Africa. Business Insider South
Africa, vol. 5, no. 1, pp. 1–2.
Cagle, A., Shepherd, M., Grodsky, S., Armstrong, A., Jordaan, S.M. and Hernandez, R. (2022). Standardized Metrics
to Quantify Solar Energy-Land Relationships: A Global Systematic Review. SSRN Electronic Journal, vol. 2, no. 1,
pp. 1–35.
Cagle, A.E., Armstrong, A., Exley, G., Grodsky, S.M., Macknick, J., Sherwin, J. and Hernandez, R.R. (2020). The Land
Sparing, Water Surface Use Efficiency, and Water Surface Transformation of Floating Photovoltaic Solar Energy
Installations. Sustainability, vol. 12, no. 19, pp. 1–23.
Cahill, N., Campbell, M., Croke, J. and Hughes, K. (2022). A Bayesian Hierarchical Time Series Model for Recon-
structing Hydroclimate from Multiple Proxies. arXiv Vanity preprint, vol. 9383, no. 1, pp. 1–12.
Calvert, K. (2016). From energy geography to energy geographies: Perspectives on a fertile academic borderland.
321
SAGE Publications, vol. 40, no. 1, pp. 105–125.
Campana, P.E., Wästhage, L., Nookuea, W., Tan, Y. and Yan, J. (2019). Optimization and assessment of floating and
floating-tracking PV systems integrated in on- and off-grid hybrid energy systems. Solar Energy, vol. 177, no. 1,
pp. 782–795.
Canham, C., Cole, J. and Lauenroth, W. (2003). Models in Ecosystem Science. Princeton University Press, Princeton,
New Jersey. ISBN 9780691092898.
Cardoso, E., Vasconcellos, R., Bini, D., Miyauchi, M., Santos, C., Alves, P., Paula, A., Nakatani, A., Pereira, J. and
Nogueira, M. (2013). Soil health: Looking for suitable indicators. What should be considered to assess the effects
of use and management on soil health? Scientia Agricola, vol. 70, no. 4, pp. 274–289.
Carmichael, J., Talwar, S., Tansey, J. and Robinson, J. (2017). Where do we want to be: Making sustainability
indicators integrated, dynamic and participatory. In: Community Indicators Measuring Systems, 2, pp. 178–204.
Routledge. ISBN 9781351950213.
Cartelle, J.J., Lara, M., de la Cruz, M.P. and Del Cano, A. (2015). Indicators for assessing sustainability of power
plants: Environmental, social, economic and technical aspects. Renewable Energy and Power Quality Journal,
vol. 1, no. 13, pp. 139–144.
Cazzaniga, R., Cicu, M., Rosa-Clot, M., Rosa-Clot, P., Tina, G.M. and Ventura, C. (2018). Floating photovoltaic
plants: Performance analysis and design solutions. Renewable and Sustainable Energy Reviews, vol. 81, no. 9,
pp. 1730–1741.
Cazzaniga, R. and Rosa-Clot, M. (2021). The booming of floating PV. Solar Energy, vol. 219, pp. 3–10.
CGIAR (2020). CGIAR research is development: Catalyzing transformations to create better food systems, new
knowledge, tools and policies. Available at: https://www.cgiar.org/research/ [Accessed 5 Nov 2020].
Chand, A.A., Lal, P.P., Kushal, P. and Kumar, N. (2022). Economics and environmental impacts of solar energy
technologies. In: Solar Energy Advancements in Agriculture and Food Production Systems, pp. 78–92. Elsevier.
Chateau, P.A., Wunderlich, R.F., Wang, T.W., Lai, H.T., Chen, C.C. and Chang, F.J. (2019). Mathematical modeling
suggests high potential for the deployment of floating photovoltaic on fish ponds. Science of the Total Environment,
vol. 687, pp. 654–666.
Chaves, A. and Bahill, A.T. (2010). Locating Sites for Photovoltaic Solar Panels: Pilot study uses DEM derived from
LiDAR. ArcUser, University of Arizona, Tucson, vol. 1, pp. 24–27.
Chen, B. and Poquet, O. (2022). Networks in Learning Analytics: Where Theory, Methodology, and Practice Intersect.
Journal of Learning Analytics: Networks in Analytics, vol. 9, no. 1, pp. 1–12.
Chen, Y. and Cui, H. (2020). The abstract of doctoral dissertation: Some research on hypothesis testing and nonpara-
metric variable screening problems for high dimensional data. Statistical Theory and Related Fields, vol. 4, no. 2,
pp. 228–229.
Chen, Z., Wu, L., Cheng, S., Lin, P., Wu, Y. and Lin, W. (2017). Intelligent fault diagnosis of photovoltaic arrays based
on optimized kernel extreme learning machine and I-V characteristics. Applied Energy, vol. 204, pp. 912–931.
Choi, C.S., Siregar, I.Z. and Ravi, S. (2020). Reframing the Competition for Land between Food and Energy Production
in Indonesia. In: Walsh S.J., Riveros-Iregui D., Arce-Nazario J., P.P. (ed.), Land Cover and Land Use Change on
Islands, pp. 241–260. Springer, Cham.
Choi, S.M., Park, C.-D., Cho, S.-H. and Lim, B.-J. (2022). Effects of wind loads on the solar panel array of a floating
photovoltaic system: Experimental study and economic analysis. Energy, vol. 1, no. 124649, pp. 1–12.
Choi, Y. (2014). A Study on Power Generation Analysis of Floating PV System Considering Environmental Impact.
Journal of Software Engineering Applications, vol. 8, no. 1, pp. 75–84.
Choi, Y. and Lee, Y. (2014). A Study on Development of Rotary Structure for Tracking-Type Floating Photovoltaic
System. International Journal of Precision Engineering and Manufacturing, vol. 15, no. 11, pp. 2453–2460.
Choi, Y.K., Choi, W.S. and Lee, J.H. (2016). Empirical Research on the Efficiency of Floating PV Systems. Science
of Advanced Materials, vol. 8, no. 3, pp. 681–685.
Ciel et Terre (2022). Hydrelio Floating Solar Technology Expert Floating Systems. Available at: https://www.ciel-et-
terre.net/, [Accessed 2 Jan 2022].
CIF (2018). Note on Preliminary Financial and Economic Analysis for Energy Storage Solutions and Floating Solar
322
Photovoltaic for India: Innovation in Solar Power and Hybrid Technologies Project. Climate Investment Fund, vol. 1,
no. 2, pp. 1–5.
Clark, T. and Kulkarni, V. (2021). Adaptive Complex Systems: Digital Twins. In: Kaur, G., Tomar, P. and Tanque,
M. (eds.), Artificial Intelligence to Solve Pervasive Internet of Things Issues, pp. 211–238. Academic Press. ISBN
978-0-12-818576-6.
Clifton, D. (2022). Geospatial Digital Twin Explained: GIS Data Unlocked. Available at:
https://spaceiq.com/blog/geospatial-digital-twin/ [Accessed 13 Jan 2022].
Cohen, E. and Hogan, R. (2018). Made in the Shade: Promoting Solar over Water Projects. Idaho Law Review,
vol. 54, pp. 1–45.
Coletta, V., Pagano, A., Pluchinotta, I., Fratino, U., Scrieciu, A., Nanu, F. and Giordano, R. (2020). Causal Loop
Diagrams for supporting Nature Based Solutions participatory design and performance assessment. Journal of
Environmental Management, vol. 2, pp. 111668–11678.
Collins, E., Miller, S., Mundt, M., Stein, J., Sorensen, R., Granata, J., Quintana, M. and Laboratories, S.N. (2010).
A Reliability and Availability Sensitivity Study of a Large Photovoltaic System. 25th European Photovoltaic Solar
Energy Conference and Exhibition, vol. 2, no. 1, pp. 1–7.
Collins, J.W., Lamarque, J.F., Schulz, M., Boucher, O., Eyring, V., Hegglin, I.M., Maycock, A., Myhre, G., Prather, M.,
Shindell, D. and Smith, J.S. (2017). AerChemMIP: Quantifying the effects of chemistry and aerosols in CMIP6.
Geoscientific Model Development, vol. 10, no. 2, pp. 585–607.
Colvin, J., Blackmore, C., Chimbuya, S., Collins, K., Dent, M., Goss, J., Ison, R., Roggero, P.P. and Seddaiu, G.
(2014). In search of systemic innovation for sustainable development: A design praxis emerging from a decade of
social learning inquiry. Research Policy, vol. 5, pp. 1–7.
Comte, A. (2019). A General View of Positivism. General Press, London, U.K.
COP21 (2016). Report of the Conference of the Parties held in Paris from 30 Nov to 13 Dec 2015. United Nations
Framework Convention on Climate Change, vol. 10, no. 1, pp. 1–36.
Cossio, M.L.T. (2012). Research Ethics Checklist. University of Edinburgh, vol. 1, no. 3, pp. 1–5.
Cox, M. (2019a). Floating Solar Update: Bigger Projects and Climbing Capacity. Wood Mackenzie Power and
Renewables, vol. 1, no. 9, pp. 1–4.
Cox, M. (2019b). The Economics of Floating Solar. The First Global Floating Solar Conference, vol. 1, no. 2, pp.
1–18.
Coyle, R.G. (1997). System Dynamics Modelling: A Practical Approach. Journal of the Operational Research Society,
vol. 48, no. 5, p. 544.
CSIR (2022). The CSIR Photovoltaic (PV) Module Quality and Reliability Laboratory. Available at:
https://www.csir.co.za/pv-lab [Accessed 28 May 2022].
Cuce, E., Cuce, P.M., Saboor, S., Ghosh, A. and Sheikhnejad, Y. (2022). Floating PVs in Terms of Power Generation,
Environmental Aspects, Market Potential, and Challenges. Sustainability, vol. 14, no. 5, pp. 1–25.
da Costa, L.C.A. and da Silva, G.D.P. (2021). Save water and energy: A techno-economic analysis of a floating solar
photovoltaic system to power a water integration project in the Brazilian semiarid. International Journal of Energy
Research, vol. 45, no. 9, pp. 1–11.
Dallery, C. (2020). Techno-economic feasibility study of snow mitigation system adapted to floating photovoltaic power
plant. Ph.D. thesis, Royal Institute of Technology, Stockholm.
Dangermond, J. (2018). The Emergence of a Geospatial Cloud. Available at:
https://www.forbes.com/sites/esri/2018/09/24/the-emergence-of-a-geospatial-cloud/ [Accessed 12 Aug 2021].
DavisTaxCommittee (2015). A Carbon Tax For South Africa. Report an A Carbon Tax for South Africa for the Minister
of Finance, vol. 2, no. 1, pp. 1–43.
de Lima, R.L.P., Paxinou, K., Boogaard, F.C., Akkerman, O. and Lin, F.-Y. (2021a). In-Situ Water Quality Observations
under a Large-Scale Floating Solar Farm Using Sensors and Underwater Drones. Sustainability, vol. 13, no. 11,
pp. 1–18.
de Lima, R.L.P., Paxinou, K., Boogaard, F.C., Akkerman, O. and Lin, F.-Y. (2021b). In-Situ Water Quality Observations
under a Large-Scale Floating Solar Farm Using Sensors and Underwater Drones. Sustainability, vol. 13, no. 6421,
323
pp. 1–18.
de Olde, E. (2017). Sustainable Development of Agriculture: Contribution of farm-level assessment tools. Wageningen
University, Wageningen. ISBN 9789463430685.
De Soto, W., Klein, S. and Beckman, W. (2006). Improvement and validation of a model for photovoltaic array perfor-
mance. Solar Energy, vol. 80, no. 1, pp. 78–88.
De Villiers, S. and de Wit, M. (2010). H20-CO2-energy equations for South Africa: Present status, future scenarios
and proposed solutions. Africa Earth Observatory Network (AEON), University of Cape Town, Rondebosch, Cape
Town, vol. 1, no. 2, pp. 1–49.
DEA (2014). Environmental Impact Assessment Regulations of the NEMA Act, 2014. Government Gazette of South
Africa, vol. 19519, no. 1540, pp. 1–67.
Decock, C. (2020). Carbon sequestration and greenhouse gas emissions from soils. Tech. Rep., California Polytech-
nic.
Dellosa, J. and Palconit, E. (2022). Resource Assessment of a Floating Solar Photovoltaic (FSPV) System with
Artificial Intelligence Applications. Engineering, Technology and Applied Science Research, vol. 12, no. 2, pp.
8410–8415.
Derevyanko, V. (2022). PVsyst Manual: Muli-orientation PV arrays. Available at:
https://www.pvsyst.com/help/pvarray_heterogeneous.htm [Accessed 18 Feb 2022].
Deriu, C. (2021). Floating Photovoltaics, FPV: Important new technical methodologies. Tech. Rep., Universta di
Parma, Italy.
DFFE (2008). People-Planet-Prosperity: A National Framework for Sustainable Development in South Africa. Depart-
ment of Forestry, Fisheries and Environmental Affairs, SA, vol. 1, no. 6, pp. 1–94.
DFFE (2015). Discussion Document on Environmental Offsets. Department of Forestry, Fisheries and Environmental
Affairs, SA, vol. 1, no. 6, pp. 1–49.
DFFE (2016). The Calculation of Country-specific Emission Factors for the Stationary Combustion of Fuels in the
Electricity Generation Sector. Department of Forestry, Fisheries and Environmental Affairs, SA, vol. 3, no. 1, pp.
1–73.
DFFE (2018). Maps and graphics: Environmental geographic information systems (EGIS). Department of Forestry,
Fisheries and Environmental Affairs, SA, vol. 1, pp. 1–20.
DFFE (2019a). National Climate Change Adaptation Strategy for Republic of South Africa. Department of Forestry,
Fisheries and Environmental Affairs, SA, vol. 5, no. 42446, pp. 1–61.
DFFE (2019b). Requirement to submit a report generated by the national web-based environmental screening tool.
Available at: https://screening.environment.gov.za/screeningtool [Accessed 28 Jan 2022].
DFFE (2020). Sustainability in South Africa. Chapter 2 Contents. Department of Environment, Forestry and Fisheries.
Tech. Rep., Department of Forestry, Fisheries and Environmental Affairs, SA, Pretoria, South Africa.
DFFE (2021a). Integrated Environmental Management Information Series (IEM). Available at:
https://www.environment.gov.za/documents/strategies/integrated_environmentalmanagement_eim [Accessed
2 Nov 2021].
DFFE (2021b). Methodological Guidelines for the Quantification of Greenhouse Gas Emissions. Department of
Forestry, Fisheries and Environmental Affairs, SA, vol. 668, no. 44190, pp. 1–198.
DFFE (2022). REDZs and Transmission Corridors: Environmental Geographical Information Systems (Environmental
GIS). Available at: https://egis.environment.gov.za/redz [Accessed 9 Jan 2022].
Digital Earth (2022). Geospatial Information Science and Geomatics Technology: Context-specific virtual representa-
tion of the Earth. Digital Earth Africa Geoportal, vol. 2, no. 1, pp. 1–5.
Dizier, A. (2018). Techno-economic analysis of floating PV solar power plants using active cooling technique. A case
study for Taiwan. Master of Science Thesis, KTH School of Industrial Engineering and Management.
DMRE (2019). Outlining the Renewable Energy Independent Power Producer Procurement Programme Empower-
ment Imperative. Department of Mineral Resources and Energy, pp. 1–24.
DNV (2021). Design, Development and Operation of Floating Solar Photovoltaic Systems (DNVGL-0584). Tech. Rep.,
DNV-GL.
324
DOE (2018). Integrated Resources Plan for South Africa (2018). Department of Energy, South Africa, vol. 1, no. 2,
pp. 1–77.
Donatelli, M. (2008). Integrated Modelling Frameworks for Environmental Assessment and Decision Support: How
modeling extends the scientific method at the base of research. In: Jakeman, A.J., Voinov, A. (ed.), Developments
in Integrated Environmental Assessment, Environmental Modelling, Software and Decision Support, pp. 101–118.
Elsevier Science, London, U.K.
Dorgo, G., Sebestyen, V. and Abonyi, J. (2018). Evaluating the Interconnectedness of the Sustainable Development
Goals Based on the Causality Analysis of Sustainability Indicators. Sustainability, vol. 10, pp. 1–26.
Dori, D. (2011). Object-Process Methodology: Holistic Systems Paradigm. In: Knowledge Management: 2nd Edition,
pp. 1208–1209. Springer.
DPE (2019). Roadmap for Eskom in a Reformed Electricity Supply Industry. Department Public Enterprises. Tech.
Rep., Eskom.
Driesse, A., Theristis, M. and Stein, J.S. (2022). PV Module Operating Temperature Model Equivalence and Parameter
Translation. IEEE Photovoltaic Specialists Conference, PVSC 2022, vol. 1, no. 4, pp. 1–11.
DSI (2021). A Coordinated and Efficient National System of Innovation for South Africa. Department of Science, In-
novation and Technology. Available at: https://www.gov.za/about-sa/science-technology# [Accessed 22 Feb 2021].
Dunham, A. (2021). Exploring the Potential for Floating Photovoltaic Solar on Man-Made Reservoirs in the United
States. M.Sc. in Energy Policy and Climate, Johns Hopkins University, vol. 9, pp. 1–21.
Dupraz, C., Marrou, H., Talbot, G., Dufour, L., Nogier, A. and Ferard, Y. (2011). Combining solar photovoltaic panels
and food crops for optimising land use: Towards new agrivoltaic schemes. Renewable Energy, vol. 36, no. 10, pp.
2725–2732.
EGIS (2022). Sustainable Development Mandate for Environmental Geographical Information Systems (E-GIS): En-
vironmental Geospatial Data Products and Geo-referenced Environmental Datasets. Available at: https://egis-
southafrica.co.za/ [Accessed 1 Feb 2022].
El Hammoumi, A., Chalh, A., Allouhi, A., Motahhir, S., El Ghzizal, A. and Derouich, A. (2021). Design and construction
of a test bench to investigate the potential of floating PV systems. Journal of Cleaner Production, vol. 278, no.
123917, pp. 1–10.
ELaw (2022). Overview of the EIA Process. Environmental Law Alliance Worldwide - eLAW, vol. 2, no. 1, pp. 19–23.
Energia (2020). Empowering Women to Energize the World. International Network on Gender and Sustainable En-
ergy. Available at: https://www.energia.org/ [Accessed 15 Dec 2020].
Enthought (2017). Python Distribution and Integrated Analysis Environment: Python Package with a Graphical Pack-
age Manager. Enthought Canopy, Proven Scientific Python Distribution plus Integrated Analysis Environment,
vol. 1, no. 2, pp. 1–7.
EPA (2009). Guidance on the Development, Evaluation, and Application of Environmental Models. Tech. Rep., United
States Environmental Protection Agency, Washington, DC.
EPA (2022). The Model Life-cycle: Training Module for US Environmental Protection Agency. Available at:
https://archive.epa.gov/epa/measurements-modeling/model-life-cycle-training-module.html [Accessed 2 Feb 2022].
Ernst, A. and Fuchs, D. (2022). Power dynamics, shifting roles, and learning: Exploring key actors in participation
processes in the German energy transformation (Energiewende). Energy Research and Social Science, vol. 2, no.
102420, pp. 1–8.
Escobedo, G., Jacome, N. and Arroyo-Figueroa, G. (2017). Big Data and Analytics to Support the Renewable Energy
Integration of Smart Grids Case Study: Power Solar Generation. Proceedings for 2nd International Conference on
Internet of Things, Big Data and Security, vol. 1, no. 2, pp. 267–275.
Eskom (2017). Eskom Integrated Report, 2017. Eskom Annual Report, vol. 1, no. 3, pp. 1–68.
Eskom (2021). Eskom Sustainable Development Overview: Clean Development Mechanism (CDM) Calculations.
Available at: https://www.eskom.co.za/about-eskom/sustainable-development/ [Accessed 2 Nov 2021].
Esmaeili Shayan, M. and Hojati, J. (2021). Floating Solar Power Plants: A Way to Improve Environmental and
Operational Flexibility. Journal of Energy and Environment, vol. 12, no. 4, pp. 337–348. ISSN 20792115.
ESMAP (2021). Tracking SDG 7 Progress towards Sustainable Energy. Energy Sector Management Assistance
325
Program. Available at: https://trackingsdg7.esmap.org/ [Accessed 29 Jan 2021].
ESRI (2022). Digital twinning: A framework to create and integrate digital twins as GIS creates digital twins of the
natural and built environments. Available at: https://www.esri.com/en-us/digital-twin/overview [Accessed 14 Jan
2022].
Essak, L. and Ghosh, A. (2022). Floating Photovoltaics: A Review. Clean Technologies, vol. 4, no. 3, pp. 752–769.
Estefan, J.A. (2008). Survey of Model-Based Systems Engineering (MBSE) Methodologies. Jet Propulsion Laboratory,
California Institute of Technology, vol. 1, no. 2, pp. 1–70.
Eurek, K., Denholm, P., Margolis, R. and Mowers, M. (2013). Sensitivity of Utility-Scale Solar Deployment Projections
in the SunShot Vision Study to Market and Performance Assumptions. National Renewable Energy Laboratory
(NREL), vol. 2, no. 4, pp. 1–55.
European Commission (2008). Reform of the European Union wine market. European Commission, Agriculture and
Rural Development: Adopted by the Council of Ministers in April 2008, EC Regulation 479/2008 and 555/2008,
vol. 1, no. 2, pp. 1–2.
Exley, G. (2022). Environmental impacts of floating solar photovoltaics on their host water bodies: Opportunities and
risks. PhD Dissertation, Lancaster University.
Exley, G., Armstrong, A., Page, T. and Jones, I.D. (2021a). Floating photovoltaics could mitigate climate change
impacts on water body temperature and stratification. Solar Energy, vol. 4, no. 1, pp. 1–10.
Exley, G., Hernandez, R., Page, T., Chipps, M., Gambro, S., Hersey, M., Lake, R., Zoannou, K.-S. and Armstrong,
A. (2021b). Scientific and stakeholder evidence-based assessment: Ecosystem response to floating solar photo-
voltaics and implications for sustainability. Renewable and Sustainable Energy Reviews, vol. 152, no. 111639, pp.
1–14.
Eyring, N. and Kittner, N. (2022). High-resolution electricity generation model demonstrates suitability of high-altitude
floating solar power. iScience, vol. 56, no. 6, p. 104394.
FAO (1995). Environmental Impact Assessment (EIA) Process. In: Dougherty, T. and Hall, A. (eds.), Environmental
impact assessment of irrigation and drainage projects, pp. 1–12. Food and Agriculture Organization of the United
Nations. ISBN 92-5-103731-0.
FAO (2021). Dimensions of need: Sustainable agriculture and rural development. Available at:
http://www.fao.org/3/u8480e/u8480e0l.htm [Accessed 10 Jan 2021].
FAO (2022). e-Agriculture Strategy as Global Community of Practice, vol. 2. Available at: https://www.fao.org/e-
agriculture/ [Accessed 25 Jan 2022].
Farfan, J. and Breyer, C. (2018). Combining Floating Solar Photovoltaic Power Plants and Hydropower Combining
Floating Solar Photovoltaic Power Plants and Hydropower Reservoirs: A Virtual Battery of Great Global Potential.
12th International Renewable Energy Storage Conference, Düsseldorf, vol. 1, no. 3, pp. 1–12.
Farrar, L.W., Bahaj, A.S., James, P., Anwar, A. and Amdar, N. (2022). Floating solar PV to reduce water evaporation
in water stressed regions and powering water pumping: Case study Jordan. Energy Conversion and Management,
vol. 260, no. 1, pp. 115598–115610.
Fayomi, G., Mini, S., Fayomi, O., Odunlami, A. and Oyeleke, O. (2021). Sustainability and Clean Technology: A
Technological Perspective. IOP Conference Series: Materials Science and Engineering, vol. 1107, no. 1, pp. 1–8.
Fereshtehpour, M., Javidi Sabbaghian, R., Farrokhi, A., Jovein, E.B. and Ebrahimi Sarindizaj, E. (2021). Evaluation
of factors governing the use of floating solar system: A study on Iran’s important water infrastructures. Renewable
Energy, vol. 171, no. 2021, pp. 1171–1187.
Fischlin, A. (1990). Interactive Modeling and Simulation of Environmental Systems on Workstations. Systems Ecology
Group, Institute of Terrestrial Ecology, Dept of Environmental Sciences, Swiss Federal lnstitute of Technology,
Zurich, Switzerland, vol. 1, no. 5, pp. 1–18.
Fitch (2020). Floating solar going global with 10GW more by 2025. Institute for Energy Economics & Financial
Analysis, vol. 1, pp. 1–2.
FIVS (2008). International Wine Carbon Calculator Protocol Version 1.2. Tech. Rep., FIVS (International Alcohol
Beverage Industry Society), Paris, France.
Flammini, A., Puri, M., Pluschke, L. and Dubois, O. (2014). Walking the Nexus Talk: Assessing the Water-Energy-
326
Food Nexus in the Context of the Sustainable Energy for All Initiative. Climate, Energy and Tenure Division (NRC).
Food and Agriculture Organization of the United Nations, Rome, Italy.
Flores-Mendez, R.A. (2014). Towards the Standardization of Multi-Agent Systems Architectures: An Overview. ACM
Crossroads: Special issue on Intelligent Agents, vol. 1, no. 6, pp. 1–10.
Forrester, J. (1958). Industrial dynamics: a major breakthrough for decision makers. Harvard Business Review,
vol. 36, no. 1, pp. 37–66.
Freeman, J. (2022). Recent and Planned Improvements to the System Advisor Model (SAM) and PVWatts. Report
No. NREL/PR-6A20-74214). National Renewable Energy Lab.(NREL), Golden, CO, USA, vol. 1, no. 2, pp. 1–32.
Frenkiel, F. and Goodall, D. (1978). Simulation Modelling of Environmental Problems. 1st edn. John Wiley and Sons,
London, U.K. ISBN 0471991800.
Friel, D., Karimirad, M., Whittaker, T., Doran, W.J. and Howlin, E. (2019). A review of floating photovoltaic design
concepts and installed variations. 4th International Conference on Offshore Renewable Energy CORE 2019, vol. 1,
no. 1, pp. 1–10.
Fu, B. (2020). Promoting Geography for Sustainability. Geography and Sustainability, vol. 1, no. 1, pp. 1–7.
Gadzanku, S. (2020). Floating Solar: Overview of NREL Research and Initial Findings. ISES + GSC FPV Webinar,
National Renewable Energy Laboratory (NREL), vol. 1, no. 1, pp. 1–17.
Gadzanku, S., Beshilas, L. and Grunwald, U. (2021a). Enabling Floating Solar Photovoltaic (FPV) Deployment.
Review of Barriers to FPV Deployment in Southeast Asia. Tech. Rep., NREL.
Gadzanku, S., Lee, N. and Dyreson, A. (2022a). Enabling Floating Solar Photovoltaic (FPV) Deployment: Exploring
the Operational Benefits of Floating SolarHydropower Hybrids. Tech. Rep. 6, National Renewable Energy Labora-
tory (NREL, USAID).
Gadzanku, S., Lee, N. and Dyreson, A. (2022b). Enabling Floating Solar Photovoltaic (FPV) Deployment: Policy and
Operational Considerations. Tech. Rep., National Renewable Energy Laboratory (NREL, USAID).
Gadzanku, S., Mirletz, H., Lee, N., Daw, J. and Warren, A. (2021b). Benefits and Critical Knowledge Gaps in Deter-
mining the Role of Floating Photovoltaics in the Energy-Water-Food Nexus. Sustainability, vol. 13, no. 4317, pp.
1–17.
Galal, S. (2020). Sales average price of grapes in markets from South Africa between 2000 and 2019. Statista. Avail-
able at: https://www.statista.com/statistics/1148030/sales-price-of-grapes-in-markets-from-south-africa/ [Accessed
15 Dec 2020].
Gallego-Elvira, B., Martínez-Alvarez, V., Pittaway, P., Brink, G. and Martín-Gorriz, B. (2013). Impact of Micrometeo-
rological Conditions on the Efficiency of Artificial Monolayers in Reducing Evaporation. Water Resources Manage-
ment, vol. 27, no. 7, pp. 2251–2266.
Gamarra, C. and Ronk, J.J. (2019). Floating Solar: An Emerging Opportunity at the Energy-Water Nexus. Texas
Water Journal, vol. 10, no. 1, pp. 32–45.
Gangavarapu, M. and Kulkarni, P.S. (2022). Assessment of floating solar photovoltaic potential in India’s existing
hydropower reservoirs. Energy for Sustainable Development, vol. 2, pp. 1–12.
Gazdowicz, S. (2019). Floating Solar PV: A Sea of Opportunities. Multiconsult Global Floating Solar Conference,
vol. 1, no. 8, pp. 1–25.
Gbejewoh, O., Keesstra, S. and Blancquaert, E. (2021). The 3Ps (Profit, Planet, and People) of Sustainability amidst
Climate Change: A South African Grape and Wine Perspective. Sustainability, vol. 13, no. 2910, pp. 1–23.
Gebrekiros, M. (2015). Evaporation and Transpiration. In: Introductory Notes to Hydrology, pp. 49–75. GitHub.
Geels, F.W. (2022). Causality and explanation in socio-technical transitions research: Mobilising epistemological
insights from the wider social sciences. Research Policy, vol. 51, no. 6, pp. 1–14.
Gensol Engineering (2018). Floating Solar PV - An Introduction. Floating Solar Technology, pp. 1–3.
Gerber, A., Kotzé, P. and Van Der Merwe, A. (2010). Towards the formalisation of the TOGAF Content Metamodel
using ontologies. In: Proceedings of 12th International Conference on Enterprise Information Systems, pp. 1–11.
Gerling, C. (2011). Conversion Factors: From Vineyard to Bottle. Appellation Cornell, vol. 2011, no. 4, pp. 1–4.
Gilman, P. (2015). SAM Photovoltaic Model Technical Reference. National Renewable Energy Laboratory, NREL/TP-
6A20-6410, vol. 1, pp. 1–30.
327
Gilman, P. (2018). SAM Photovoltaic Model Technical Reference Update. National Renewable Energy Laboratory,
NREL P-6A20-64102, vol. 3, no. 1, pp. 1–93.
Global Legal Insights (2021). Energy Laws and Regulations in the South African Practice Area. Available
at: https://www.globallegalinsights.com/practice-areas/energy-laws-and-regulations/south-africa [Accessed 26 Jun
2021].
Goepel, K. (2018). Implementation of an Online software tool for the Analytic Hierarchy Process (AHP-OS). Interna-
tional Journal of the Analytic Hierarchy Process, vol. 10, no. 3, pp. 469–487.
Goepel, K.D. (2013). Implementing the Analytic Hierarchy Process as a Standard Method for Multi-Criteria Deci-
sion Making In Corporate Enterprises: A New AHP Excel Template with Multiple Inputs. In: Proceedings of Int
Symposium on Analytic Hierarchy Process, pp. 1–4. Kuala Lumpur.
Goh, H.H., Li, C., Zhang, D., Dai, W., Lim, C.S., Kurniawan, T.A. and Goh, K.C. (2022). Application of choosing by
advantages to determine the optimal site for solar power plants. Scientific Reports, vol. 12, no. 1, p. 4113.
Goldense, B.L. (2017). Technology Readiness Levels are widely adopted. Available at:
https://www.machinedesign.com/automation-iiot/article/21835443/technology-readiness-levels-are-widely-adopted
[Accessed 10 Feb 2021].
Gonzalez, A.M., Sandoval, H., Acosta, P. and Henao, F. (2016). On the acceptance and sustainability of renewable
energy projects: A systems thinking perspective. Sustainability (Switzerland), vol. 8, no. 11, pp. 1–22.
Gonzalez Sanchez, R., Kougias, I., Moner-Girona, M., Fahl, F. and Jäger-Waldau, A. (2021). Assessment of floating
solar photovoltaics potential in existing hydropower reservoirs in Africa. Renewable Energy, vol. 169, no. 2021, pp.
687–699.
Goodwin, J. (2018). Looking for Soil Health Indicators in the Field. Noble Research Institute, vol. 36, no. 10, pp. 1–3.
Goossens, P. (2017). Industry 4.0 and the Power of the Digital Twin: Adopt a Systems Approach to Machine Design
and Survive the Next Industrial Revolution. MapleSoft, vol. 5, no. 3, pp. 1–4.
Goossens, P. (2021). Industry 4.0 and the Power of the Digital Twin: Adopt a Systems Approach to Machine Design
and Survive the Next Industrial Revolution. Available at: https://www.maplesoft.com/ns/manufacturing/industry-4-
0-power-of-the-digital-twin.aspx [Accessed 12 Feb 2021].
Gorjian, S. (2022). Progress and Challenges of Crop Production and Electricity Generation in Agrivoltaic Systems Us-
ing Semi-transparent Photovoltaic Technology. Renewable and Sustainable Energy Reviews, vol. 158, no. 112126,
pp. 1–25.
Gorjian, S., Ebadi, H., Jathar, D. and Savoldi, L. (2022a). Solar Energy for Sustainable Food and Agriculture: Devel-
opments, Barriers, and Policies. In: Solar Energy Advancements in Agriculture and Food Production Systems, pp.
88–92. Elsevier.
Gorjian, S., Kamrani, F. and Ebadi, H. (2022b). Applications of Renewable Energy Sources in Agriculture from
Complementarity Perspective. In: Complementarity of Variable Renewable Energy Sources, pp. 1–15. Elsevier,
London.
Gorjian, S., Savoldi, L., Jathar, L. and Ebadi, H. (2022c). Solar energy technology in agriculture: Developments,
policies, and barriers. In: Gorjian, S. and Campana, P.E. (eds.), Advancements of Solar Energy Technology in
Agriculture and Food Production Systems. Elsevier.
Gorjian, S., Sharon, H., Ebadi, H., Kant, K., Scavo, F.B. and Tina, G.M. (2021). Recent Technical Advancements,
Economics and Environmental Impacts of Floating Photovoltaic Solar Energy Conversion Systems. Journal of
Cleaner Production, vol. 278, no. 124285, pp. 1–28.
Gorjian, S., Singh, R., Shukla, A. and Mazhar, A.R. (2020). On-farm applications of solar PV systems. In: Gorjian,
S. and Shukla, A. (eds.), Photovoltaic Solar Energy Conversion, pp. 147–190. Academic Press. ISBN 978-0-12-
819610-6.
Goswami, A., Sadhu, P., Goswami, U. and Sadhu, P.K. (2019). Floating Solar Power Plant for Sustainable Develop-
ment: A Techno-Economic Analysis. Environmental Progress and Sustainable Energy, vol. 2, pp. 1–12.
Goswami, A. and Sadhu, P.K. (2021a). Adoption of floating solar photovoltaics on waste water management sys-
tem: A unique nexus of water-energy utilization, low-cost clean energy generation and water conservation. Clean
Technologies and Environmental Policy, vol. 1, no. 2, pp. 1–22.
328
Goswami, A. and Sadhu, P.K. (2021b). Degradation analysis and the impacts on feasibility study of floating solar
photovoltaic systems. Sustainable Energy, Grids and Networks, vol. 26, no. 100425, pp. 1–20.
Grand View Research (2020). Floating Solar Panels Market Research Report by Type, Location, Technology: Global
Industry Analysis and Demand Forecast to 2026. Available at: https://www.grandviewresearch.com/industry-
analysis/floating-solar-panels-market/methodology [Accessed 31 Jan 2020].
Graser, A. and Olaya, V. (2015). Processing: a python framework for the seamless integration of geoprocessing tools
in QGIS. ISPRS International Journal of Geo-Information, vol. 4, no. 4, pp. 2219–2245.
Grieves, M. (2014). Digital Twin: Manufacturing Excellence through Virtual Factory Replication Whitepaper. White
Paper, vol. 1, no. 3, pp. 1–7.
Grieves, M. and Vickers, J. (2017). Digital Twin: Mitigating Unpredictable, Undesirable Emergent Behavior in Complex
Systems. In: Kahlen, F., Flumerfelt, S. and Alves, A. (eds.), Transdisciplinary Perspectives on Complex Systems,
pp. 85–113. Springer, Cham.
Grodsky, S., Holgerson, M., Schmidt, M. and Rice, A. (2022). Techno-ecological Impacts and Synergies of Floating
Solar. Available at: https://www.atkinson.cornell.edu/award/techno-ecological-impacts-and-synergies-of-floating-
solar/ [Accessed 2 Oct 2022].
Groundstroem, F. and Juhola, S. (2021). Using systems thinking and causal loop diagrams to identify cascading
climate change impacts on bioenergy supply systems. Springer Netherlands.
Guaita-Pradas, I., Marques-Perez, I., Gallego, A. and Segura, B. (2019). Analyzing territory for the sustainable
development of solar photovoltaic power using GIS databases. Environmental Monitoring and Assessment, vol.
191, no. 12, pp. 1–17.
Gunda, T. and Cai, M. (2019). Using Machine Learning, Climate and Site Characteristics to Predict PV Solar Plant
Generation. Sandia National Laboratories, Report No. SAND-2019-15241R, vol. 6, no. 29, pp. 1–6.
Gunjan, P. (2019). Solar Trend Alert: From Photovoltaics to Floatovoltaics. Navigant Research, vol. 3, no. 6, pp. 1–3.
Gupta, U. (2021). The case for floating PV on hydroelectric facilities. PV Magazine, pp. 1–2.
Gurupira, T. and Rix, A.J. (2016). Photovoltaic System Modelling using PVLib-Python. Fourth South African Solar
Energy Conference SASEC, Stellenbosch, South Africa, vol. 2, no. 1, pp. 1–10.
Haas, J., Khalighi, J., de la Fuente, A., Gerbersdorf, S.U., Nowak, W. and Chen, P.J. (2020). Floating photovoltaic
plants: Ecological impacts versus hydropower operation flexibility. Energy Conversion and Management, vol. 206,
no. 112414, pp. 1–8.
Habermas, J. (1979). What is Universal Pragmatics: Communication and the Evolution of Society. Beacon Press,
Boston, USA.
Habermas, J. (1981). The Theory of Communicative Action. Beacon Press, Boston, USA.
Haggerty, J. (2020). Floating solar nearing price parity with land-based US solar. PV Magazine USA, pp. 1–3.
Halog, N.C.O., Kucukvar, M. and Halog, A. (2017). Systems Thinking for Life Cycle Sustainability Assessment: A
Review of Recent Developments, Applications, and Future Perspectives. Sustainability, vol. 9, no. 5, pp. 706–731.
Hamidov, A., Daedlow, K., Webber, H., Hussein, H., Abdurahmanov, I., Dolidudko, A., Seerat, A.Y., Solieva, U.,
Woldeyohanes, T. and Helming, K. (2022). Operationalizing water-energy-food nexus research for sustainable
development in social-ecological systems: An interdisciplinary learning case in Central Asia. Ecology and Society,
vol. 27, no. 1.
Hansen, T. and Coenen, L. (2015). The geography of sustainability transitions: Review, synthesis and reflections on
an emergent research field. Environmental Innovation and Societal Transitions, vol. 17, pp. 92–109.
Hao, J. and Ho, T.K. (2019). Machine Learning Made Easy: A Review of Scikit-learn Package in Python Programming
Language. Journal of Educational and Behavioral Statistics, vol. 44, no. 3, pp. 1–8.
Haohui, L., Kumar, A. and Reindl, T. (2020). The Dawn of Floating Solar Technology, Benefits and Challenges. In:
Wang, C., Lim, S. and Tay, Z. (eds.), WCFS2019. Lecture Notes in Civil Engineering, Vol 41, pp. 373–383. Springer,
Singapore.
Havrysh, V., Kalinichenko, A., Szafranek, E. and Hruban, V. (2022). Agricultural Land: Crop Production or Photovoltaic
Power Plants. Sustainability (Switzerland), vol. 14, no. 9, pp. 1–23.
Heeter, J., Sekar, A., Fekete, E., Shah, M., Cook, J.J., Heeter, J., Sekar, A., Fekete, E., Shah, M. and Cook, J.J.
329
(2021). Affordable and Accessible Solar for All: Barriers, Solutions and On-Site Adoption Potential. Tech. Rep.
NREL/TP-6A20-80532, National Renewable Energy Lab.
Heisler, J., Glibert, P.M., Burkholder, J.M., Anderson, D.M., Cochlan, W., Dennison, W.C., Dortch, Q. and Gobler, C.J.
(2008). Eutrophication and harmful algal blooms: A scientific consensus. Harmful Algae, vol. 8, no. 1, pp. 3–13.
HelioClim (2022). Solar Radiation Data, HelioClim-3 solar radiation database. Available at: http://www.soda-pro.com/
[Accessed 10 Jan 2022].
HelioScope (2021). HelioScope: Mathematical Formulation. San Francisco, California. Tech. Rep., HelioScope LLC.
Hernandez, R., Armstrong, A. and Burney, J. (2019). Techno-ecological synergies of solar energy for global sustain-
ability. Nature Sustainability, vol. 2, no. 1, pp. 560–568.
Hernandez, R., Easter, S., Murphy-Mariscald, M., Maestree, F., Tavassolib, M., Allend, E., Barrows, C., Belnapg,
J. and Ochoa-Huesoh, R. (2014). Environmental Impacts of Utility-scale Solar Energy. Renewable Sustainable
Energy Reviews, vol. 29, no. 2, pp. 766–779.
Hernandez, R.R., Cagle, A.E., Grodsky, S.M. and Exley, G. (2022). Land use for United States power generation: A
critical review of existing metrics with suggestions for going forward. Renewable and Sustainable Energy Reviews,
vol. 143, no. 110911, pp. 1–6.
Hernandez, R.R., Hoffacker, M.K., Murphy-Mariscal, M.L., Wu, G.C. and Allen, M.F. (2015). Solar energy development
impacts on land cover change and protected areas. Proceedings of the National Academy of Sciences, vol. 112,
no. 44, pp. 13579–13584.
Hernandez, R.R., Hoffacker, M.K., Murphy-Mariscal, M.L., Wu, G.C. and Allen, M.F. (2016). Solar energy development
impacts on land cover change and protected areas. Proceedings of the National Academy of Sciences of the USA,
vol. 113, no. 12, pp. 13579–13584.
Hernandez, R.R., Jordaan, S.M., Kaldunski, B. and Kumar, N. (2020). Aligning Climate Change and Sustainable
Development Goals with an Innovation Systems Roadmap for Renewable Power. Frontiers in Sustainability, vol. 1,
no. 2, pp. 1–11.
Hira, A. and Swartz, T. (2014). What makes Napa Napa: The roots of success in the wine industry. Wine Economics
and Policy, vol. 3, no. 1, pp. 37–53.
Hoff, H. (2011). Understanding the Nexus. Bonn 2011 Conference on the water, energy and food security nexus.
Stockholm Environment Institute, Stockholm, vol. 1.2, pp. 1–18.
Hoffacker, M.K., Allen, M.F. and Hernandez, R.R. (2017). Land-sparing Opportunities for Solar Energy Development
in Agricultural Landscapes: A Case Study of the Great Central Valley, CA, United States. Environmental Science
& Technology, vol. 51, no. 24, pp. 14472–14482.
Hoffacker, M.K. and Hernandez, R.R. (2020). Local Energy: Spatial Proximity of Energy Providers to Their Power
Resources. Frontiers in Sustainability, vol. 1, pp. 1–7.
Holmgren, W.F., Hansen, C.W. and Mikofski, M.A. (2018). PVlib python: A python package for modeling solar energy
systems. Journal of Open Source Software, vol. 3, no. 29, pp. 1–3.
Holmgren, W.F., Lorenzo, A.T. and Hansen, C. (2017). A Comparison of PV Power Forecasts Using PVLib-Python.
Sandia National Laboratories, pp. 1–5.
Hottenroth, H., Sutardhio, C., Weidlich, A. and Tietze, I. (2022). Beyond climate change: Multi-attribute decision
making for a sustainability assessment of energy system transformation pathways. Renewable and Sustainable
Energy Reviews, vol. 156, p. 111996.
Huang, L., Wu, J. and Yan, L. (2015). Defining and measuring urban sustainability: A review of indicators. Landscape
Ecology, vol. 30, no. 7, pp. 1175–1193.
Hukseflux (2021). How to calculate PV performance ratio and performance index according to the latest IEC 61724
standard series. Tech. Rep., Hukseflux.
Hunter, J.J., Volschenk, C.G., Mania, E. and Castro, A.V. (2021). Grapevine row orientation mediated temporal and
cumulative microclimatic effects on grape berry temperature and composition. Agricultural and Forest Meteorology,
vol. 310, pp. 108660–108671.
Hwang, C. and Yoon, K. (1981). Methods for Multiple Attribute Decision Making. In: Multiple Attribute Decision Making.
Lecture Notes in Economics and Mathematical Systems, vol 186, pp. 58–191. Springer, Berlin, Heidelberg.
330
IEA (2017). PV Performance Modeling Methods and Practices: Results from the 4th PV Performance Modeling
Collaborative Workshop. In: International Energy Agency, pp. 1–98. Fraunhofer ISE, Freiburg, Germany.
IEA (2021). Best Practices Handbook for the Collection and Use of Solar Resource Data for Solar Energy Applications:
Third Edition 2021. Tech. Rep., International Energy Agency.
IEA PVPS (2020). Climatic Rating of Photovoltaic Modules: Different Technologies for Various Operating Conditions.
12. International Energy Agency. ISBN 9783907281086.
IFC (2015). Utility-Scale Solar Photovoltaic Power Plants: A Project Developers Guide. World Bank Group, Interna-
tional Finance Corporation, vol. 2, no. 1, pp. 1–216.
IFC (2020). Emerging Energy Solutions: Floating Solar Photovoltaic on the Rise. International Finance Corporation.
Tech. Rep., World Bank Group.
IFC (2021). Overview of Environmental Legislation in South Africa: Environmental and Social Risk Management Pro-
gram, International Financing Corporation. Available at: https://esrmqa.worldbank.org/program-countries/overview-
environmental-legislation-0 [Accessed 4 March 2021].
IISD (2016). Environmental Impact Assessment Training Manual. International Institute for Sustainable Development
(IISD), vol. 1, no. 2, pp. 1–150.
Ikhennicheu, M., Danglade, B., Pascal, R., Arramounet, V., Trebaol, Q. and Gorintin, F. (2021). Analytical method for
loads determination on floating solar farms in three typical environments. Solar Energy, vol. 219, pp. 34–41.
INNS (2021). Hypotheses Testing. In: International Network for Natural Sciences, pp. 1–7. Available at:
https://innspub.net/hypotheses-testing [Accessed 21 Feb 2022].
IRENA (2015). Renewable energy in the water, energy and food nexus. International Renewable Energy Agency,
vol. 2, pp. 1–125.
IRENA (2019). Global Energy Transformation: A Roadmap to 2050 (2019 edition). International Renewable Energy
Agency.
ISIGenere (2020). Solar Flotante by Isifloating saving the world by using water bodies in our planet to generate
renewable solar energy while preserving water and land. Available at: http://www.isigenere.com [Accessed 28 Sept
2020].
IUCN (2020). Environmental and Social Impact Assessment (ESIA): Biodiversity Conservation and Sustainable Use
of Natural Resources. International Union for Conservation of Nature, vol. 1, no. 3, pp. 1–6.
Jackson, M.C. (2010). Reflections on the development and contribution of critical systems thinking and practice.
Systems Research and Behavioral Science, vol. 27, no. 2, pp. 133–139.
Jackson, N.D. and Gunda, T. (2021). Evaluation of extreme weather impacts on utility-scale photovoltaic plant perfor-
mance in the United States. Applied Energy, vol. 302, no. 11, p. 117508.
Jainandunsing, K. (2019). Industry 4.0 Concepts: Cyber Physical System, IOT, Digital Twin. Available
at: https://www.aiscorp.com/blog/industry-4-0-concepts-cyber-physical-system-iot-digital-twin/ [Accessed 12 May
2019]., Horsham Township, PA.
Jakhrani, A., Samo, S., Kamboh, S., Labadin, J. and Rigit, A. (2014). An improved mathematical model for computing
power output of solar photovoltaic modules. International Journal of Photo-energy, vol. 2, no. 1, pp. 1–9.
Jashari, H. (2017). The Research Hypothesis in PhD Dissertation. Journal of Knowledge in Practice, vol. 20, no. 1,
pp. 507–512.
Jones, D., Snider, C., Nassehi, A., Yon, J. and Hicks, B. (2020). Characterising the Digital Twin: A systematic literature
review. CIRP Journal of Manufacturing Science and Technology, vol. 29, no. 1, pp. 36–52.
Jordan, D.C., Deline, C., Kurtz, S.R., Kimball, G.M. and Anderson, M. (2018). Robust PV Degradation Methodology
and Application. IEEE Journal of Photovoltaics, vol. 8, no. 2, pp. 525–531.
Judek, H.H. (2020). Chiba Yakamura water mega solar plant. Solar cell damage accident Council Third report. Kyocera
TLC Solar Limited Liability Company, pp. 1–27.
Kakoulaki, G., Gonzalez-Sanchez, R., Gracia, Amillo, A.M.G. and Jäger-Waldau, A. (2022). Benefits of Pairing
Floating Solar Photovoltaic with Hydropower Reservoirs in Europe (National and Regional Level). Social Science
Research Network Electronic Research Journal, vol. 22, no. 1, pp. 1–33.
Karatas, Y. and Yilmaz, D. (2021). Experimental investigation of the microclimate effects on floating solar power plant
331
energy efficiency. Clean Technologies and Environmental Policy, vol. 23, no. 41, pp. 2157–2170.
Karimi, H.A. and Karimi, B. (2020). Geospatial Data Science Techniques and Applications. CRC Press. ISBN
9780367572815.
Kaymak, M. and Sahin, A. (2021). Problems encountered with floating photovoltaic systems under real conditions: A
new FPV concept and novel solutions. Sustainable Energy Technologies and Assessments, vol. 47, no. 101504,
pp. 1–15.
Khaiter, P.a. and Erechtchoukova, M.G. (2007). From Complex to Simple in Environmental Simulation Modelling.
Modsim 2007: International Congress on Modelling and Simulation, pp. 2069–2075.
Khalifeh Soltani, S.R., Mostafaeipour, A., Almutairi, K., Hosseini Dehshiri, S.J., Hosseini Dehshiri, S.S. and Techato,
K. (2022). Predicting effect of floating photovoltaic power plant on water loss through surface evaporation for
wastewater pond using artificial intelligence: A case study. Sustainable Energy Technologies and Assessments,
vol. 50, no. 101849, pp. 1–13.
Khan, A.H., Omar, S. and Mushtary, N. (2022). Digital Twin and Artificial Intelligence Incorporated with Surrogate
Modeling for Hybrid and Sustainable Energy Systems. Handbook of Smart Energy Systems, vol. 2, no. 1, pp. 1–23.
Khan, I. (2021). Sustainability assessment of energy systems: Indicators, methods, and applications. In: Methods in
Sustainability Science, chap. 4, pp. 47–70. Elsevier, Jashore, Bangladesh.
Kim, H.M., Cheung, J.H.N., Laskowski, M. and Gel, I. (2016a). Data analytics using ontologies of management
theories: Towards implementing "from theory to practice". Proceedings of IEEE 17th International Conference on
Information Reuse and Integration, vol. 1, pp. 417–422.
Kim, J., Rabelo, M., Padi, S.P., Yousuf, H., Cho, E.C. and Yi, J. (2021a). A Review of the Degradation of Photovoltaic
Modules for Life Expectancy. Energies, vol. 14, no. 14, p. 4278.
Kim, J.Y., Koide, D., Ishihama, F., Kadoya, T. and Nishihiro, J. (2021b). Current site planning of medium to large solar
power systems accelerates the loss of the remaining semi-natural and agricultural habitats. Science of The Total
Environment, vol. 779, no. 1, p. 146475.
Kim, S., Yoon, S., Choi, W. and Choi, K. (2016b). Application of Floating Photovoltaic Energy Generation Systems in
South Korea. Sustainability, vol. 8, no. 12, pp. 1333–1342.
King, D.L., Boyson, W.E. and Kratochvil, J.A. (2004). Photovoltaic Array Performance Model. Photovoltaic System
R&D Department Sandia National Laboratories, Albuquerque, New Mexico. Tech. Rep., Sandia National Laborato-
ries.
King, M. (2016). King IV Report on Corporate Governance in South Africa. In: 46th World Congress for Finance
Executives, SA Institute for Business Accountants (SAIBA), pp. 1–120.
Kjeldstad, T., Lindholm, D., Marstein, E. and Selj, J. (2021). Cooling of floating photovoltaics and the importance of
water temperature. Solar Energy, vol. 218, no. 1, pp. 544–551.
Kjeldstad, T., Nysted, V.S., Kumar, M., Oliveira-pinto, S. and Otnes, G. (2022). The performance and amphibious
operation potential of a new floating photovoltaic technology. Solar Energy, vol. 239, no. 5, pp. 242–251.
Klemm, K., Serrano, M.Á., Eguíluz, V.M. and Miguel, M.S. (2012). A measure of individual role in collective dynamics.
Scientific Reports, vol. 2, no. 1, pp. 292–301.
Klise, G.T., Johnson, J.L. and Adomatis, S.K. (2013). Standardizing appraisals for PV installations. Appraisal Journal.,
Conference Record of the IEEE Photovoltaic Specialists Conference, vol. 2, no. 3, pp. 3454–3462.
Klise, G.T. and Stein, J.S. (2009). Models Used to Assess the Performance of Photovoltaic Systems. Sandia Corpo-
ration, Department of Energy, vol. 1, no. 12, pp. 1–67.
Kneifel, J., O’Rear, E., Grubb, S., Greig, A.L. and Suh, S. (2017). Building industry reporting and design for sustain-
ability (BIRDS) commercial database technical manual: update. Tech. Rep., National Institute of Standards and
Technology.
Kobos, P.H., Malczynski, L.A., Walker, L.T.N., Borns, D.J. and Klise, G.T. (2018). Timing is everything: A technology
transition framework for regulatory and market readiness levels. Technological Forecasting and Social Change, vol.
137, pp. 211–225.
Kohen, H. and Dori, D. (2021). Improving Conceptual Modeling with Object-Process Methodology Stereotypes. Ap-
plied Sciences, vol. 11, no. 2301, pp. 1–23.
332
Komm, B. (2015). Vineyard Yield Estimation. Washington State University, vol. 2, pp. 1–11.
Kothari, C. (2004). Research Methodology: Methods and Techniques. 2nd edn. New Age International Publishers,
New Delhi.
Kotov, V. (1997). Systems of Systems as Communicating Structures. Tech. Rep., Hewlett-Packard Company.
Kougias, I., Taylor, N., Kakoulaki, G. and Jager-Waldau, A. (2021). The role of photovoltaics for the European Green
Deal and the recovery plan. Renewable and Sustainable Energy Reviews, vol. 144, no. 111017, pp. 1–8.
Krechowicz, M., Krechowicz, A., Lichołai, L., Pawelec, A., Piotrowski, J.Z. and Ste¸pie ´
n, A. (2022). Reduction of the
Risk of Inaccurate Prediction of Electricity Generation from PV Farms Using Machine Learning. Energies, vol. 15,
no. 11, pp. 1–21.
Krug, F. (2021). Flexible Floating Solar System Design. Tech. Rep., Zimmermann PV Floating, Eberhardzell, Germany.
Kumar, J.C.R. and Majid, M. (2021). Floating solar photovoltaic plants in India: A rapid transition to a green energy
market and sustainable future. Energy & Environment, vol. 11, no. 1, pp. 1–55.
Kumar, M. and Mani, M. (2021). Sustainability assessment in manufacturing: Perspectives, challenges, and solutions.
In: Gupta, K. and Salonitis, K. (eds.), Sustainable Manufacturing, Handbooks in Advanced Manufacturing, pp.
287–311. Elsevier. ISBN 9780128181157.
Kumar, M. and Mani, M. (2022). Sustainability Assessment in Manufacturing for Effectiveness: Challenges and
Opportunities. Frontiers in Sustainability, vol. 3, no. 3, pp. 1–16.
Kumar, M., Niyaz, H.M. and Gupta, R. (2021). Challenges and opportunities towards the development of floating
photovoltaic systems. Solar Energy Materials and Solar Cells, vol. 233, no. 111408, pp. 1–43.
Kumar, N.M., Chakraborty, S., Yadav, S.K., Singh, J. and Chopra, S.S. (2022). Advancing simulation tools specific
to floating solar photovoltaic systems: Comparative analysis of field-measured and simulated energy performance.
Sustainable Energy Technologies and Assessments, vol. 52, no. 102168, pp. 1–14.
Kumar, P., Shukla, A.K., Sudhakar, K. and Mamat, R. (2017). Experimental exergy analysis of water-cooled PV
module. International Journal of Exergy, vol. 23, no. 3, pp. 197–209.
Kundi, M. (2006). Causality and the interpretation of epidemiologic evidence. Environmental Health Perspectives, vol.
114, no. 7, pp. 969–974.
Kurtz, S., Riley, E., Newmiller, J., Dierauf, T., Kimber, A., McKee, J., Flottemesch, R. and Krishnani, P. (2013). Anal-
ysis of Photovoltaic System Energy Performance Evaluation Method. Tech. Rep., National Renewable Energy
Laboratory.
Kuschke, I. and Cassim, A. (2019). Sustainable Agriculture: Market Intelligence Report 2019. Tech. Rep., Green
Cape.
Kwon, O., Lee, S.M., Kwon, J., Jo, H. and Cha, H. (2019). The deduction of generation efficiency coefficient of the
floating PV systems compared to ground-mounted PV systems based on their operation-data analysis. Transactions
of the Korean Institute of Electrical Engineers, vol. 68, no. 6, pp. 717–724.
Laing, N.J., Laing, I. and Hesse, A. (2011). Floating Solar Platform, WIPO PCT Patent Application WO2006027220A3.
Laszlo, E. (1972). The Case for Systems Philosophy. Metaphilosophy, vol. 3, no. 2, pp. 123–141.
Laubscher, N. (2022). Floating solar panels: Generating Power and Saving Water (Vredendal, Alzanne Fruit and
Vegetable Farm). Farmers Weekly, vol. 2, no. 8, pp. 1–3.
Law, A.M. (2014). Simulation Modeling and Analysis, Fith Edition. McGraw-Hill, New York, USA.
Lawford, R.G. (2019). A Design for a Data and Information Service to Address the Knowledge Needs of the Water-
Energy-Food (WEF) Nexus and Strategies to Facilitate Its Implementation. Frontiers in Environmental Science,
vol. 7, no. 56, pp. 1–11.
Lawson, H. and Martin, J.N. (2008). On the Use of Concepts and Principles for Improving Systems Engineering Prac-
tice. In: Proceedings of the 18th Annual International Council on Systems Engineering, International Symposium,
pp. 594–604. Wiley, Utrecht, Netherlands.
Lazcano, C., Decock, C. and Wilson, S.G. (2020). Defining and Managing for Healthy Vineyard Soils, Intersections
With the Concept of Terroir. Frontiers in Environmental Science, vol. 8, p. 68.
Lee, J. and Callaway, D. (2018). The cost of reliability in decentralized solar power systems in sub-Saharan Africa.
Nature Energy, vol. 3, no. 1, pp. 960–968.
333
Lee, K., Holland, A. and McNeill, D. (2022). Global Sustainable Development in the Twenty-First Century. Edinburgh
University Press. ISBN 9781474496902.
Lee, N., Grunwald, U., Rosenlieb, E., Mirletz, H., Aznar, A., Spencer, R. and Cox, S. (2020). Hybrid floating solar
photovoltaics-hydropower systems: Benefits and global assessment of technical potential. Renewable Energy, vol.
162, pp. 1415–1427.
Leung, L.R., Bader, D.C., Taylor, M.A. and McCoy, R.B. (2020). An Introduction to the E3SM Special Collection: Goals,
Science Drivers, Development and Analysis. Journal of Advances in Modeling Earth Systems, Special issue: The
Energy Exascale Earth System Model (E3SM), vol. 12, no. 11, pp. 1–11.
Liber, W., Bartle, C., Spencer, R., Jordan, M., Cagle, A. and Lewis, T. (2019). Statewide Potential Study for the
Implementation of Floating Solar Photovoltaic Arrays. Tech. Rep., Colorado Energy Office, Denver, CO, USA.
Lindholm, D., Kjeldstad, T. and Selj, J.K. (2021). Heat loss coefficients computed for floating PV modules. Progress
in Photovoltaics Research and Applications, vol. 29, no. 12, pp. 1–12.
Lindhult, E., Sankaran, S. and Midgley, G. (2022). Systemic innovation: Towards a new paradigm in systems thinking
and innovation. Systems Research and Behavioral Science, vol. 6, pp. 1–12.
Lindorfer, J., Rosenfeld, D., Annevelink, B. and Mandl, M. (2019). Technical, Economic and Environmental Assess-
ment of Biorefinery Concepts. Developing a practical approach for characterisation. Tech. Rep., IEA Bioenergy.
Liu, J., Braun, E., Dupmeier, C., Kuckertz, P. and Ryberg, D.S. (2019). Architectural concept and evaluation of a
framework for the efficient automation of computational scientific work flows: An energy systems analysis example
of a cyber-physical energy system. Applied Sciences (Switzerland), vol. 9, no. 4, pp. 1–22.
Liu, Y. (2022). Solar GHI Ensemble Prediction Based on a Meteorological Model and Method Kalman Filter. Advances
in Meteorology, vol. 2022, no. 1523198, pp. 1–12.
Lombard, A., Nicolau, M., Tshabalala, S., du Plessis, A., Pienaar, R., Anderson, R. and Khanye, B. (2021). The
Feasibility of an Environmental Awareness and Responsibility Programme in Rural South Africa: A Case Study
from the South. In: W., L.F., R., P. and L.O., d.S. (eds.), Sustainable Development in Africa, pp. 591–612. Springer,
Cham.
Lopes, P., Neto, S. and Branco, D. (2020). Water-energy nexus: Floating photovoltaic systems promoting water
security and energy generation in the semiarid region. Journal of Cleaner Production, vol. 273, no. 6, pp. 1–15.
López, M., Soto, F. and Hernández, Z.A. (2022). Assessment of the potential of floating solar photovoltaic panels in
bodies of water in mainland Spain. Journal of Cleaner Production, vol. 340, no. 130752, pp. 1–13.
Lovering, J., Swain, M., Blomqvist, L. and Hernandez, R.R. (2021). Land-use intensity of electricity production and
tomorrow’s energy landscape. Nature Climate Change, vol. 1, no. 1, pp. 1–24.
Luca, A.I.D., Molari, G., Seddaiu, G. and Toscano, A. (2015). Multidisciplinary and Innovative Methodologies for
Sustainable Management in Agricultural Systems. Environmental Engineering and Management, vol. 1, no. 7, pp.
1–12.
Luderer, G., Pehl, M., Arvesen, A., Gibon, A., Bodirsky, T. and De Boer, B.L. (2019). Environmental co-benefits and
adverse side-effects of alternative power sector decarbonization strategies. Nature Communications, vol. 10, no.
5229, pp. 1–13.
Lufkin, B. (2015). Solar Panels Floating on Water to power Japan. National Geographic, pp. 1–2.
Ma, X. (2019). Geo-Data Science: Leveraging Geoscience Research with Geoinformatics, Semantics and Open Data.
Acta Geologica Sinica, vol. 1, pp. 1–46.
Macedo, H.D., Sassanelli, C., Larsen, P.G. and Terzi, S. (2021). Facilitating model-based design of cyber-physical
manufacturing systems. Journal of Manufacturing Science and Technology, vol. 104, pp. 1936–1941.
Madni, A.M., Madni, C.C. and Lucero, S.D. (2019). Leveraging digital twin technology in model-based systems engi-
neering. Systems, vol. 7, no. 1, pp. 1–13.
Madurai Elavarasan, R., Pugazhendhi, R., Jamal, T., Dyduch, J., Arif, M.T., Manoj Kumar, N., Shafiullah, G.M.,
Chopra, S.S. and Nadarajah, M. (2021). Envisioning the UN Sustainable Development Goals (SDGs) through the
lens of energy sustainability (SDG 7) in the post-COVID-19 world. Applied Energy, vol. 292, no. 116665, pp. 1–26.
Magliocca, N.R., van Vliet, J., Brown, C., Evans, T.P., Houet, T., Messerli, P., Messina, J.P., Nicholas, K.A., Or-
netsmüller, C., Sagebiel, J., Schweizer, V., Verburg, P.H. and Yu, Q. (2015). From meta-studies to modeling: Using
334
synthesis knowledge to build process based land change models. Environmental Modelling and Software, vol. 72,
no. 10, pp. 10–20.
Mahdi, E.J., Resen, A.K. and Jawad, N. (2020). Behavior of the Solar Panel Temperature under Different Environ-
mental Conditions. Test Engineering and Management, vol. 83, pp. 7235–7243.
Majumder, A., Innamorati, R., Frattolillo, A., Kumar, A. and Gatto, G. (2021). Performance Analysis of a Floating
Photovoltaic System and Estimation of the Evaporation Losses Reduction. Energies, vol. 14, no. 24, pp. 8336–
8353.
Maka, A.O.M. (2022). Solar energy technology and its roles in sustainable development. Clean Energy, vol. 1, no. 6,
pp. 476–483.
Makhija, S.P., Dubey, S.P., Bansal, R.C. and Jena, P.K. (2021). Techno-Environ-Economical Analysis of Floating PV
On-Ground PV Grid Extension Systems for Electrification of a Remote Area in India. Technology and Economics
of Smart Grids and Sustainable Energy, vol. 6, no. 1, pp. 1–10.
Manoj Kumar, N., Ajitha, A., A. Chand, A. and Goel, S. (2021). Performance of Photovoltaics in Ground Mount-
Floating-Submerged Installation Methods. In: Bhoi, A. (ed.), Advances in Systems, Control and Automations, vol.
708, pp. 199–207. Springer, Singapore.
Maplesoft (2022a). Industry 4.0 and the Power of the Digital Twin. Adopt a Systems Approach to Machine Design
and Survive the Next Industrial Revolution. Available at: https://www.maplesoft.com/ns/manufacturing/industry-4-
0-power-of-the-digital-twin.aspx [Accessed 10 Mar 2022].
Maplesoft (2022b). Virtual Commissioning with a Model-Driven Digital Twin. Maplesoft Mathematical Modelling Sim-
ulations, vol. 1, no. 2, pp. 1–8.
Marion, W., Anderberg, A., Deline, C., Glick, S., Muller, M., Perrin, G., Rodriguez, J., Rummel, S., Terwilliger, K.
and Nrel, T.J.S. (2014). User Manual for Data for Validating Models for PV Module Performance. NREL National
Laboratory of the U.S. Department of Energy, vol. 1, no. 4, pp. 1–49.
Marshall, C. and Rossman, G.B. (2011). Designing Qualitative Research. Sage Publications, Thousand Oaks, CA.
Marstein, E.S., Mortensen, D. and Lindholm, D. (2020). Cooling of PV modules in floating PV power plants. 37th
European Photovoltaic Solar Energy Conference (EUPVSEC), vol. 1, no. 4, pp. 1–5.
Martinelli, A., Mina, A. and Moggi, M. (2020). The Enabling Technologies of Industry 4.0: Examining the Seeds of
Fourth Industrial Revolution. Laboratory of Economics and Management, vol. 2, no. 4, pp. 1–49.
Martins, B.P. (2019). Techno-economic evaluation of a floating PV system for a wastewater treatment facility. Master
of Science Thesis, KTH School of Industrial Engineering and Management.
Mathijssen, D., Hofs, B., Spierenburg-Sack, E., Vanasperen, R., Vanderwal, B., Vreeburg, J. and Ketelaars, H. (2020).
Potential impact of floating solar panels on water quality in reservoirs; pathogens and leaching. Water Practice and
Technology, vol. 15, no. 3, pp. 807–811.
Mathis, K. and Huber, B.R. (2018). Energy Law and Economics. Springer International Publishing, Berlin, Germany.
ISBN 978-3-319-74636-4.
Mavroidis, P.C. and Neven, D.J. (2019). Greening the WTO Environmental Goods Agreement, Tariff Concessions, and
Policy Likeness. Journal of Int. Economic Law, vol. 22, no. 3, pp. 373–388.
Mayer, M.J. and Gróf, G. (2021 feb). Extensive comparison of physical models for photovoltaic power forecasting.
Applied Energy, vol. 283, no. 116239, pp. 1–18.
Mazzola, S., Vergara, C., Astolfi, M., Li, V., Perez-Arriaga, I. and Macchi, E. (2017). Assessing the value of forecast-
based dispatch in the operation of off-grid rural microgrids. Renewable Energy, vol. 108, pp. 116–125.
McCarl, B.A., Yang, Y., Schwabe, K., Engel, B.A., Mondal, A.H., Ringler, C. and Pistikopoulos, E.N. (2017). Model
Use in WEF Nexus Analysis: A Review of Issues. Current Sustainable, Renewable Energy Reports, vol. 4, no. 3,
pp. 144–152.
McCartan, K. and Robson, C. (2016). Real World Research. Wiley.
McEwan, C. (2017). Spatial processes and politics of renewable energy transition: Land, zones and frictions in South
Africa. Political Geography, vol. 56, pp. 1–12.
McGuffie, K. and Henderson-Sellers, A. (2001). Forty years of numerical climate modelling. International Journal of
Climatology, vol. 21, no. 9, pp. 1067–1109.
335
McKuin, B., Zumkehr, A., Ta, J., Bales, R. and Viers, J.H. (2021). Energy and water co-benefits from covering canals
with solar panels. Nature Sustainability, vol. 4, no. 7, pp. 609–617.
McLeod, S. (2018). What is a hypothesis. Simply Psychology., vol. 2, no. 1, pp. 1–5.
Medinilla, A., Sergejeff, K. and Domingo, E. (2022). The geopolitics of African renewable energy european and
Chinese investments in a global green transition and climate change. Int Cooperation and Development Policy
Group in Europe and Africa, vol. 1, no. 316, pp. 1–42.
Mehren, R. and Rempfler, A. (2022). Assessing Systems Thinking in Geography. In: Bourke, T., Mills, R. and Lane,
R. (eds.), Assessment in Geographical Education: An International Perspective, pp. 31–54. Springer International
Publishing, Cham. ISBN 978-3-030-95139-9.
Meyerhof, R. and Prinsloo, F. (2019). Around Floating PV Opportunities. Discussion on LinkedIn, Senior Solar Expert
DNV GL, Hamburg.
Meyers, B., Deceglie, M., Deline, C. and Jordan, D. (2020). Signal Processing on PV Time-Series Data: Robust
Degradation Analysis without Physical Models. IEEE Journal of Photovoltaics, vol. 10, no. 2, pp. 546–553.
Meynhardt, T., Chandler, J.D. and Strathoff, P. (2016). Systemic principles of value co-creation: Synergetics of value
and service ecosystems. Journal of Business Research, vol. 69, no. 8, pp. 2981–2989.
Micheli, L. (2021). Energy and economic assessment of floating photovoltaics in Spanish reservoirs: Cost competi-
tiveness and the role of temperature. Solar Energy, vol. 227, no. 1, pp. 625–634.
Micheli, L. (2022). The temperature of floating photovoltaics: Case studies, models and recent findings. Solar Energy,
vol. 242, pp. 234–245.
Micheli, L., Talavera, D.L. and Tina, G.M. (2022). Techno-economic potential and perspectives of floating photovoltaics
in Europe. Solar Energy ·, vol. 243, no. 9, pp. 203–214.
Midgley, G. (2003). Science as Systemic Intervention: Some Implications of Systems Thinking and Complexity for the
Philosophy of Science. Systemic Practice and Action Research, vol. 16, no. 25, pp. 7–30.
Midgley, S., New, M. and Spelman, S.-s. (2014). The Food-Energy-Water-Land-Biodiversity (FEWLB) Nexus and
Local Economic Development in the Berg River Catchment : Framework and Description. Draft Research Report
for submission to the Foreign and Commonwealth Office, South Africa, vol. 1, no. 6, pp. 1–85.
Mingers, J. (2014). Systems Thinking, Critical Realism and Philosophy: A Confluence of Ideas. Routledge, London,
U.K.
Mitre (2022). Characterizing Systems of Systems and Systems of Systems Engineering. Mitre Systems Engineering
Guide, vol. 2, pp. 1–15.
Moghadam, S., Soloot, H.H., Soloot, A.H. and Jadhav, A. (2021). Floating Photovoltaics: A SWOT Analysis. EU
PVSEC Proceedings, European Photovoltaic Solar Energy Conference and Exhibition, Valencia, Spaine, vol. 2,
no. 1, pp. 1–7.
Moharram, K.A., Abd-Elhady, M.S., Kandil, H.A. and El-Sherif, H. (2013). Enhancing the Performance of Photovoltaic
Panels by Water Cooling. Engineering Journal, vol. 4, no. 4, pp. 869–877.
Moldavska, A. and Welo, T. (2015). Development of Manufacturing Sustainability Assessment Using Systems Think-
ing. Sustainability, vol. 8, no. 1, pp. 1–26.
Molenaar, P. (2001 jan). Systems Modeling. In: Smelser, N.J. and Baltes, P.B. (eds.), International Encyclopedia of
the Social & Behavioral Sciences, pp. 15423–15428. Elsevier.
Monat, J.P. and Gannon, T.F. (2015). What is Systems Thinking? A Review of Selected Literature Plus Recommen-
dations. American Journal of Systems Science, vol. 4, no. 1, pp. 11–26.
Moolman, S. (2019). Report 2019 update: Eskom tariff increases vs inflation since 1988 (with projections to
2022). Available at: https://www.poweroptimal.com/2019-update-eskom-tariff-increases-vs-inflation-since-1988-
with-projections-to-2022/ [Accessed 27 Oct 2020].
Morrison-Saunders, A. and Pope, J. (2013). Conceptualising and managing trade-offs in sustainability assessment.
Environmental Impact Assessment Review, vol. 38, pp. 54–63.
Moscovici, D. and Reed, A. (2018). Comparing wine sustainability certifications around the world: History, status and
opportunity. Journal of Wine Research, vol. 29, no. 1, pp. 1–25.
MuSigma (2016). State of Analytics and Decision Science. Mu-Sigma Analytics, Insights from key decision makers
336
on how their organizations approach analytics and problem solving, vol. 201, no. 1, pp. 1–19.
NAC (2013). Development of a Decision Framework. In: National Academies of Sciences, Engineering, and
Medicine; Sustainability for the Nation: Resource Connections and Governance Linkages, pp. 1–220. The Na-
tional Academies Press, Washington, DC.
Naegler, T., Becker, L., Buchgeister, J. and Hauser, W. (2021). Integrated Multidimensional Sustainability Assessment
of Energy System Transformation Pathways. Sustainability, vol. 13, no. 9, pp. 1–28.
Nakicenovic, N. (2022). Perspectives on the pervasive energy-systems. Oxford Open Energy, vol. 1, no. 2, pp. 10–12.
National Research Council (1997). Rediscovering Geography: New Relevance for Science and Society. National
Academy Press, Washington, DC. ISBN 0-309-05199-1.
National Research Council (2011). Sustainability and the U.S. EPA. The National Academies Press, Washington,
D.C.
National Treasury (2020). Carbon Offsetting under the Carbon Tax Act. Government Gazette of South Africa, vol. 22,
pp. 1–19.
Naushad, M., Kumar, P.S. and Suganya, S. (2019). Bromate Formation in Drinking Water and Its Control Using
Graphene Based Materials. In: A New Generation Material Graphene: Applications in Water Technology, pp.
239–260. Springer.
NDP (2019). National Development Plan 2030. Available at: https://www.gov.za/issues/national-development-plan-
2030 [Accessed 25 May 2019].
Nedcor Foundation (2004). Land Reform in the Wine Industry: The Wine Producer Programme. Document prepared
by VinPro in collaboration with the Nedcor Foundation, vol. 1, no. 3, pp. 1–44.
NERSA (2021). National Energy Regulator of South Africa. Available at: http://www.nersa.org.za/ [Accessed 5 Jan
2021].
New Southern Energy (2021). Floating Solar Farms, save money with clean energy, sustainably. Available at:
https://newsouthernenergy.com/solar-farms-and-floating-solar-farms/ [Accessed 15 Jan 2021].
Newman, B.M. and Newman, P.R. (2020). Dynamic Systems Theory. In: Behavioral Sciences: Theories of Human
Development, pp. 1–28. Psychology Press, New York. ISBN 9780203937174.
Nhamo, L., Mabhaudhi, T., Mpandeli, S., Dickens, C., Nhemachena, C. and Naidoo, D. (2020). An integrative analytical
model for the water-energy-food nexus: South Africa case study. Environmental Science and Policy, vol. 109, no. 5,
pp. 15–24.
Nisar, H., Janjua, A. and Hafeez, H. (2022). Thermal and electrical performance of solar floating PV system compared
to on-ground PV system: An experimental investigation. Solar Energy, vol. 241, no. 4, pp. 231–247.
Nisbet, R., Miner, G. and Elder, J.F. (2018). Model Evaluation and Enhancement. In: Handbook of Statistical Analysis
and Data Mining Applications, p. 215. Elsevier Inc.
Nistor, E., Dobrei, A.G., Dobrei, A., Camen, D. and Sala, F. (2018). N2O, CO2, Production, and C Sequestration in
Vineyards: a Review. Water, Air, and Soil Pollution, vol. 229, no. 9, pp. 1–9.
Noorollahi, Y., Ghenaatpisheh Senani, A., Fadaei, A., Simaee, M. and Moltames, R. (2022). A framework for GIS-
based site selection and technical potential evaluation of PV solar farm using Fuzzy-Boolean logic and AHP multi-
criteria decision-making approach. Renewable Energy, vol. 186, pp. 89–104.
Norton Rose Fulbright (2021). Are floating solar and wind installations the new frontiers of renewable energy. Available
at: https://www.projectfinance.law/publications/2020/august/floating-solar/ [Accessed 15 Feb 2021].
Novas, N., Mar, R., Camacho, J.M. and Alcayde, A. (2021). Advances in Solar Energy towards Efficient and Sustain-
able Energy. Sustainability 2021, 13(11), ;, vol. 13, no. 11, pp. 6295–6303.
Nowak, L.I. and Washburn, J.H. (2002). Building Brand Equity: Consumer Reactions to Proactive Environmental
Policies by the Winery. Intl. Journal of Wine Marketing, vol. 14, no. 3, pp. 5–19.
NRC (2013). Sustainability for the Nation: Resource Connections and Governance Linkages. In: Concepts, Method-
ology and Knowledge Management for Sustainability Science, pp. 1–142. National Academies Press.
NREL (2015). Reimagining what’s possible: How NREL Energy Analysis and Decision Support Capabilities are
Guiding Energy Systems Transformation at Home and Around the World. National Renewable Energy Laboratory
(NREL), vol. 1, no. 2, pp. 1–23.
337
NREL (2018a). Great Potential for Floating PV Systems: A Technology Already in Widespread Use Overseas, Espe-
cially in Japan. National Renewable Energy Lab Report, vol. 1, no. 2, pp. 1–7.
NREL (2018b). Solar System Modeling at NREL: Overview of NREL Solar Data and Tools. In: Proceesings of 2018
Solar Power International Conference, pp. 1–173.
NREL (2019a). Exploring Floatovoltaics: How Floating Solar PV could Influence our Changing Energy Landscape.
National Renewable Energy Lab Video Release, vol. 1, no. 2, pp. 1–15.
NREL (2019b). Viewer National Solar Radiation Database (NSRDB). Available at: https://nsrdb.nrel.gov/nsrdb-viewer
[Accessed 25 Sept 2019].
NREL (2020). NREL Solar Prospector. In: Layered GIS Interface for Solar Prospecting, pp. 1–7. Available at:
http://maps.nrel.gov/prospector [Accessed 23 Jun 2021].
NREL (2021). Geospatial Data Science Data and Tools. Available at: https://www.nrel.gov/gis/data-tools.html [Ac-
cessed 18 Feb 2021].
NREL (2022). System Advisor Model (SAM). Available at: https://sam.nrel.gov/ [Accessed 9 Feb 2022].
NSF (2006). Simulation-Based Engineering Science: Revolutionizing Engineering Science through Simulation. Re-
port of the National Science Foundation, Blue Ribbon Panel on Simulation-Based Engineering Science, vol. 2,
no. 1, pp. 1–88.
NSF (2011). Geography and Spatial Sciences Program, Strategic Plan, 2011-2015. National Science Foundation.
Tech. Rep., National Science Foundation.
Ntombela, S.M. (2019). A general equilibrium analysis of the effects of carbon tax policy on South Africa’s agricultural
industries. Doctor of Philosophy in Agricultural Economics, University of Pretoria, pp. 1–125.
O’Brien, J. and O’Brien, J. (2014). Modelling Knowledge Dynamics. Shaping Knowledge, pp. 65–88.
OECD (2001). Adoption of Technologies for Sustainable Farming Systems. Organisation for Economic Co-operation
and Development, Wageningen Workshop Proceedings, vol. 1, p. 149.
Oliveira-Pinto, S. and Stokkermans, J. (2020). Assessment of the potential of different floating solar technologies:
Overview and analysis of different case studies. Energy Conversion and Management, vol. 211, no. 112747, pp.
1–16.
Omrani, M.M. and Omrani, M.M. (2022). Techno-Economic and Environmental Analysis of Floating Photovoltaic
Power Plants: A Case Study of Iran. Renewable Energy Research and Applications, vol. 1, no. 5, pp. 1–14.
Orgill, M.K., York, S. and MacKellar, J. (2019). Introduction to Systems Thinking for the Chemistry Education Com-
munity. Journal of Chemical Education, vol. 96, no. 12, pp. 2720–2729.
Oudes, D. (2022). Landscape-inclusive energy transition landscape as catalyst in the shift to renewable energy. Ph.D.
thesis, Wageningen University.
Partovi, F.Y., Burton, J. and Banerjee, A. (1990). Application of Analytical Hierarchy Process in Operations Manage-
ment. Operations & Production Management, vol. 10, no. 3, pp. 5–19.
Patel, S. (2015). The research paradigm, methodology, epistemology and ontology. Available
at: http://salmapatel.co.uk/academia/the-research-paradigm-methodology-epistemology-and-ontology-explained-
in-simple-language/ [Accessed 15 April 2020].
Patel, S.S. and Rix, A.J. (2020). The Impact of Water Surface Albedo on Incident Solar Insolation of a Collector
Surface. In: 2020 International SAUPEC/RobMech/PRASA Conference, Cape Town, South Africa, 2020, pp. 1–6.
Pawlak, K. and Kołodziejczak, M. (2020). The Role of Agriculture in Ensuring Food Security in Developing Countries:
Considerations in the Context of the Problem of Sustainable Food Production. Sustainability 2020, Vol. 12, Page
5488, vol. 12, no. 13, pp. 1–20.
Peacock, B. (2021). Green roof improves solar panel efficiency by 3.6% on average. PV Magazine USA, pp. 1–3.
Peel, M.C., Finlayson, B.L. and Mcmahon, T. (2007). Updated World Map of the Koppen-Geiger Climate Classification.
Hydrology and Earth System Sciences, vol. 11, no. 5, pp. 1633–1644.
Pentland, W. (2011). Napa Winery Pioneers Solar Floatovoltaics. Forbes Business, vol. 8, pp. 1–3.
Perez, M., Perez, R., Ferguson, C.R. and Schlemmer, J. (2018). Deploying effectively dispatchable PV on reservoirs:
Comparing floating PV to other renewable technologies. Solar Energy, vol. 174, no. 1, pp. 837–847.
Perrakis, G., Tasolamprou, A.C., Kenanakis, G., Economou, E.N., Tzortzakis, S. and Kafesaki, M. (2021). Passive
338
radiative cooling for the temperature and efficiency control of photovoltaics. In: CLEO: Applications and Technology.
The Optical Society.
Peters, I. and Nobre, A. (2021). Deciphering the Thermal Behavior of Floating Photovoltaic Installations. Solar Energy
Advances, pp. 1–12.
Peters, I.M. and Nobre, A.M. (2020). On Module Temperature in Floating PV Systems. In: 47th IEEE Photovoltaic
Specialists Conference (PVSC), pp. 238–241. Institute of Electrical and Electronics Engineers Inc.
Piana, V., Kahl, A., Saviozzi, C. and Schumann, R. (2021). Floating PV in mountain artificial lakes: A checklist for site
assessment. Renewable Energy Environmental Sustainability, vol. 6, no. 4, pp. 1–20.
Pohekar, S.D. and Ramachandran, M. (2004). Application of multi-criteria decision making to sustainable energy
planning. Renewable and Sustainable Energy Reviews, vol. 8, no. 4, pp. 365–381.
Pouran, H.M., Lopes, M.P.C. and Ziar, H. (2021). Evaluating Floating Photovoltaics potential in providing clean energy
and supporting agricultural growth in Vietnam. Climate Compatible Growth, vol. 1, pp. 1–4.
Pouran, H.M., Padilha Campos Lopes, M., Nogueira, T., Alves Castelo Branco, D. and Sheng, Y. (2022). Environmen-
tal and technical impacts of floating photovoltaic plants as an emerging clean energy technology. iScience, vol. 25,
no. 11, p. 105253.
Pourdehnad, J., Wexler, E. and Wilson, D. (2016). Integrating systems thinking and design thinking. The Systems
Thinker, vol. 4, no. 1, pp. 1–6.
Pretorius, A. (2020). The carbon cost of wine. Available at: https://www.wineland.co.za/the-carbon-cost-of-wine/
[Accessed 26 Apr 2020].
Princloo, F. (2023). System Dynamics Characterisation and Synthesis of Floating Photovoltaics in terms of Energy,
Environmental and Economic Parameters with WELF Nexus Sustainability Features. Sustainable Energy Technolo-
gies and Assessments, vol. 55, no. 3, pp. 1–18.
Pringle, A.M., Handler, R.M. and Pearce, J.M. (2017). Aquavoltaics: Synergies for dual use of water area for solar
photovoltaic electricity generation and aquaculture. Renewable and Sustainable Energy Reviews, vol. 80, no. 5,
pp. 572–584.
Prinsloo, F. (2017a). Development of a Custom Designed GIS-Based Environmental Decision Support Tool to Address
Water Energy Nexus Issues in Floating Solar Environmental Impact Assessment. International Association of
Impact Assesment South Africa IAIAsa, vol. 8, no. 1, pp. 1–11.
Prinsloo, F. (2017b). Environmental Impact and Assessment for Floating Solar Systems on Wine Farms in the Western
Cape Wine Region. Bachelor of Science (Honours) Dissertation, Unisa.
Prinsloo, F. (2017c). Sustainable Human Settelements: Reflection on the Learning Experience. Unisa B.Sc (Honours)
Assignment Number 750897, vol. 2017, no. 4, pp. 1–5.
Prinsloo, F. (2019). Development of a GIS-Based Decision Support Tool for Environmental Impact Assessment and
Due-diligence Analyses of Planned Agricultural Floating Solar Systems. Master of Science Dissertation (Geograph-
ical Information), University of South Africa, vol. 1, pp. 1–121.
Prinsloo, F. (2020). PhD proposal towards digital geo-informatics approach to sustainability assessments for future
planned floatovoltaic technology in South African agricultural grid-substitution applications, PhD research study,
University of South Africa. University of South Africa.
Prinsloo, F. and Lombard, A. (2015a). Environmental Impact Factors in Floating Solar Systems: Lessons for the South
African Wine Industry. Academy of Sciences, Student Symposium in Natural Sciences 2015, at the University of
the Free State, South Africa, vol. 2, no. 1, pp. 1–15.
Prinsloo, F. and Lombard, A. (2015b). Floating Solar Systems in Sustainable Viticulture and Oenology: Environmental
Solutions in Support of Water-Energy-Food-Land-use Nexus Goals. International Winery Conference, Stias Centre,
Stellenbosch, South Africa, November 2015., vol. 2, pp. 1–9.
Prinsloo, F., Schmitz, P. and Lombard, A. (2021). Sustainability assessment framework and methodology with trans-
disciplinary numerical simulation model for analytical floatovoltaic energy system planning assessments. Sustain-
able Energy Technologies and Assessments, vol. 47, no. 101515, pp. 1–17.
Prinsloo, F., Schmitz, P. and Lombard, A. (2022a). A Geoinformatics Analytics and Decision-support Approach to-
wards Agricultural Floatovoltaic Energy System Sustainability Assessment and Planning Optimisation. Prepared
339
for Journal of Sustainable Energy Technologies and Assessments, vol. 1, pp. 1–12.
Prinsloo, F.C., Schmitz, P. and Lombard, A. (2022b). Integrated Energy, Environmental and Economical Analysis for
Floating Photovoltaics with WELF Nexus Sustainability Indicators. To be published: Sustainable Energy Technolo-
gies and Assessments (Social Science Research Network), vol. 2, pp. 1–18.
Pritchard, M., Klopper, R. and Prinsloo, F. (2019). Floating Solar System Site Visit and Meeting at Franschoek Valley,
South Africa. Site visit to the Boplaas Wine Farm, Marlenique Estate, Franschoek, South Africa.
Profloating (2021). Profloating Developer and Producer of Floating Solar Mounting Systems. Available at:
https://www.profloating.nl/ [Accessed 20 Feb 2021].
Purvis, B. (2019). Three pillars of sustainability: In search of conceptual origins. Sustainability Science, Concepts,
Methodology and Knowledge Management for Sustainability Science, vol. 14, no. 3, pp. 681–695.
PVGIS (2021). Photovoltaic Geographical Information System (PVGIS): Open access PV performance model tool.
Available at: https://ec.europa.eu/jrc/en/pvgis [Accessed 20 May 2021].
PVsyst (2022). Simulation Software Suite for Modelling and Performance Assessment of PV Technology Systems.
Available at: https://www.pvsyst.com/ [Accessed 12 Jan 2022].
Quaranta, E., Rosa-Clot, M. and Pistocchi, A. (2021). The role of floating PV in the retrofitting of existing hydropower
plants and evaporation reduction. In: Solar Hydro 2021, pp. 1–7.
Rabie, P.-A. (2020). Vinpro Production Plan Survey: The 2019 Vintage of SA Wine Making. Wineland Media, vol. 5,
no. 1, pp. 1–6.
Rahaman, M.A., Chambers, T.L., Fekih, A., Wiecheteck, G., Carranza, G. and Possetti, G.R.C. (2023). Floating
photovoltaic module temperature estimation: Modeling and comparison. Renewable Energy, vol. 208, no. 11, pp.
162–180.
Rahnama, E. and Aghbashlo, M. (2019). Spatio-temporal Solar Exergoeconomic and Exergoenvironmental Maps for
Photovoltaic Systems. Energy Conversion Management, vol. 195, no. 9, pp. 701–711.
Ramasamy, V. and Margolis, R. (2021). Floating Photovoltaic System Cost Benchmark: Q1 2021 Installations on
Artificial Water Bodies. National Renewable Energy Laboratory (NREL), vol. 1, no. 10, pp. 1–15.
Randle-Boggis, R.J., White, P.C., Cruz, J., Parker, G., Montag, H., Scurlock, J.M. and Armstrong, A. (2020). Realising
co-benefits for natural capital and ecosystem services from solar parks: A co-developed, evidence-based approach.
Renewable and Sustainable Energy Reviews, vol. 125, no. 109775, pp. 1–10.
Rangaraju, S., Vo, P.L. and Nguyen, K.V. (2021). Harvest the Sun on Water: A Review on Floating Photovoltaic
Solution For Vietnam. International Journal of Research Publication and Reviews, vol. 2, no. 3, pp. 169–173.
Ranjbaran, P., Yousefi, H., Gharehpetian, G.B. and Astaraei, F.R. (2019). A review on floating photovoltaic (FPV)
power generation units. Renewable and Sustainable Energy Reviews, vol. 110, pp. 332–347.
Raschka, S. and Vahid Mirjalili (2019). Python Machine Learning Second Edition Machine Learning and Deep Learn-
ing with Python, Scikit-learn and TensorFlow. Taiwan Review, vol. 69, no. 4, pp. 1–770.
Raviv, O., Palatnik, R.R. and Shechter, M. (2022). Review of the Economic Impact of Water Availability on Food Se-
curity and the Related Ecosystems. In: Cavalli, L. and Vergalli, S. (eds.), Connecting the Sustainable Development
Goals: Understanding the Role of the WEF Nexus in the 2030 Agenda, pp. 35–50. Springer Int Publishing, Cham.
ISBN 9783031013362.
Reda, I. and Andreas, A. (2008). Solar Position Algorithm for Solar Radiation Applications. Tech. Rep., National
Renewable Energy Laboratory.
Reeves, T. (2006). Design research from a technology perspective. In: den Akker, J.V., Gravemeijer, K., Nieveen, N.
and McKenney, S. (eds.), Educational Design Research, pp. 52–67. Routledge, London.
Refaai, M.R.A., Dhanesh, L., Ganthia, B.P., Mohanty, M., Subbiah, R. and Anbese, E.M. (2022). Design and Imple-
mentation of a Floating PV Model to Analyse the Power Generation. International Journal of Photoenergy, vol.
2022, no. 1, pp. 1–13.
Reverberi, P. and Talamo, M. (1999). Probabilistic model for interactive decision-making. Decision Support Systems,
vol. 25, no. 4, pp. 289–308.
Reza Reisi, A., Hassan Moradi, M. and Jamasb, S. (2013). Classification and comparison of maximum power point
tracking techniques for photovoltaic system: A review. Renewable and Sustainable Energy Reviews, vol. 19, pp.
340
433–443.
Richardson, G.P. and Pugh, A. (1981). Introduction to System Dynamics Modeling with Dynamo. Cambridge: MIT
Press.
Ringler, C., Bhaduri, A. and Lawford, R. (2013). The nexus across water, energy, land and food (WELF): Potential for
improved resource use efficiency. Current Opinion in Environmental Sustainability, vol. 5, no. 6, pp. 617–624.
Rizzoli, A.-E., Leavesley, G.H., Ascough, J., Argent, R.M., Athanasiadis, I.N., Brilhante, V., Claeys, F.H., David, O.,
Donatelli, M., Gijsbers, P., Havlik, D., Kassahun, A., Krause, P., Quinn, N.W.T., Scholten, H., Sojda, R.S. and Villa,
F. (2008). Integrated Modelling Frameworks for Environmental Assessment and Decision Support: State of the Art
and new perspective. In: Environmental modelling, software and decision support, chap. 7, pp. 101–118. Elsevier,
Amsterdam.
Rodriguez-Ulloa, R. and Paucar-Caceres, A. (2005). Soft System Dynamics Methodology (SSDM): Combining Soft
Systems Methodology (SSM) and System Dynamics (SD). Systemic Practice and Action Research, vol. 18, no. 3,
pp. 303–334.
Röhr, J.A., Sartor, B.E. and Duenow, J.N. (2022). Identifying optimal photovoltaic technologies for underwater appli-
cations. iScience, vol. 6, no. 1, p. 104531.
Roman, F. and Hartmann, S. (2014). Models and Theories in Science. In: Zalta, E.N. (ed.), The Stanford Encyclopedia
of Philosophy, pp. 1–26. Oxford Bibliographies Online, Stanford, CA.
Rosa-Clot, M. and Tina, G.M. (2017). The Floating PV Plant. In: Submerged and Floating Photovoltaic Systems:
Modelling, Design and Case Studies, pp. 89–136. Elsevier.
Rosa-Clot, M., Tina, G.M. and Nizetic, S. (2017). Floating photovoltaic plants and wastewater basins: An Australian
project. Energy Procedia, vol. 134, no. 10, pp. 664–674.
RSA (1998). National Environmental Management Act (NEMA), Act No. 107 of 1998. Government Gazette of the
Republic of South Africa, vol. 401, no. 19519, pp. 1–37.
RSA (2004). National Environmental Management: Air Quality Act No. 39 of 2004. Government Gazette, Republic of
South Africa, vol. 476, no. 27318, pp. 1–29.
RSA (2008). National Energy Act of the Republic of South Africa, Act No. 34 of 2008. Government Gazette of South
Africa, vol. 1, no. 34, pp. 1–15.
RSA (2019a). Carbon Tax Act of South Africa, Act No. 15 of 2019. Government Gazette, Republic of South Africa,
vol. 647, no. 42483, pp. 1–65.
RSA (2019b). Integrated Resources Plan 2030. Government Gazette, Republic of South Africa, vol. 652, no. 42778,
pp. 1–100.
RSA (2020). National Environmental Management Act: Procedures for assessment and minimum criteria for reporting
on identified environmental themes when applying for environmental authorisation. Government Gazette of South
Africa, vol. 20, no. 43110, pp. 24–62.
RSA (2021). Environmental Impact Assessment Regulations of the National Environmental Management Act (NEMA
Act). Government Gazette of South Africa, vol. 40772, pp. 211–276.
Rummler, G.A. (2007). Serious Performance Consulting according to Rummler. John Wiley & Sons.
Runge, J. (2018). Causal network reconstruction from time series: From theoretical assumptions to practical estima-
tion. APL Machine Learning: Chaos, vol. 28, no. 7, pp. 1–21. ISSN 10541500.
Saaty, T.L. (2000). Fundamentals of Decision Making and Priority Theory With the Analytic Hierarchy Process. RWS
Publications.
Saaty, T.L. (2008). Relative Measurement and Its Generalization in Decision Making Why Pairwise Comparisons
are Central in Mathematics for the Measurement of Intangible Factors The Analytic Hierarchy/Network Process.
Statistics and Operations Research, vol. 102, no. 2, pp. 251–318.
Sahoo, U., Kumar, R., Singh, S.K. and Tripathi, A.K. (2018). Energy, exergy, economic analysis and optimization of
polygeneration hybrid solar-biomass system. Applied Thermal Engineering, vol. 145, pp. 685–692.
Sahu, A., Yadav, N. and Sudhakar, K. (2016). Floating photovoltaic power plant: A review. Renewable and Sustainable
Energy Reviews, vol. 66, no. 1, pp. 815–824.
SAIT (2020). The South African Carbon Tax Act Explained. Available at: https://www.thesait.org.za/news/498153/The-
341
Carbon-Tax-Act-Explained.htm [Accessed 14 Jan 2021].
Sala, S., Ciuffo, B. and Nijkamp, P. (2015). A Systemic Framework for Sustainability Assessment. Ecological Eco-
nomics, vol. 119, no. 11, pp. 314–325.
Sandia (2016). Diagramming causal loops of PV system design, operations, and maintenance choices. Sandia
National Laboratories: Energy and Climate on Renewables, vol. 2, no. 1, pp. 1–11.
Sandia (2020). PV-LIB Toolbox for Software Modelling of Photolvoltaic Energy Systems. Available at:
https://pvpmc.sandia.gov/applications/pv_lib-toolbox/ [Accessed 15 Jan 2021].
Sandia (2022a). PV Performance Modeling Methods and Practices. In: PV Performance Modelling Collaborative
(PVPMC), pp. 1–15. Available at: https://pvpmc.sandia.gov [Accessed 22 Jan 2022].
Sandia (2022b). PVLIB Matlab: Software Modeling of a Solar PV System. Available at:
https://pvpmc.sandia.gov/PVLIB_Matlab_Help/html/PVL_TestScript1.html [Accessed 12 Jan 2022].
Sandia (2022c). Sandia Photovoltaic Performance Modeling Collaborative. Available at: https://pvpmc.sandia.gov
[Accessed 12 Feb 2022].
Santafé, M.R., Gisbert, F.P.S., Romero, F.J.S., Soler, J.B.T., Gozálvez, J.J.F. and Gisbert, C.M.F. (2014). Implementa-
tion of a photovoltaic floating cover for irrigation reservoirs. Journal of Cleaner Production, vol. 68, pp. 568–570.
Santiago-Brown, I., Metcalfe, A., Jerram, C. and Collins, C. (2015). Sustainability Assessment in Wine-Grape Growing
in the New World: Economic, Environmental, and Social Indicators for Agricultural Businesses. Sustainability, vol. 7,
no. 7, pp. 8178–8204.
Santiago Silvestre (2012). Review of System Design and Sizing Tools. In: McEvoy, A., Markvart, T. and Castañer,
L. (eds.), Practical Handbook of Photovoltaics Fundamentals and Applications, 2nd edn, chap. 4, pp. 673–694.
Academic Press.
Santos-Borja, A.C. (2021). Dealing with uncertainties: Floating solar farm in natural lakes. IOP Conference Series:
Earth and Environmental Science, vol. 789, no. 12036, pp. 1–10.
Sarkodie, S.A. and Owusu, P.A. (2020). Bibliometric analysis of water-energy-food nexus: Sustainability assessment
of renewable energy. Current Opinion in Environmental Science and Health, vol. 13, pp. 29–34.
Sathe, A. (2017). Floating Power Plant Market Analysis Report of Industry Trends, Insights and Forecasts 2017 to
2022. Tech. Rep., Markets and Markets.
Saunders, P. (2020). Land Use Requirements of Solar and Wind Power Generation: Understanding a Decade of
Academic Research. Energy Innovation Reform Project, pp. 1–49.
Sauran (2021). Southern African Universities Radiometric Network. Solar Radiometric Data for the Public. Available
at: https://sauran.ac.za [Accessed 12 Mar 2021].
SAWIS (2021). SA Wine Industry Information and Systems: South African Resources Portal, Statistics and Economic
Impact Studies. Available at: http://www.sawis.co.za/ [Accessed 20 Jul 2021].
Scandellari, F., Caruso, G., Liguori, G., Meggio, F., Palese Assunta, M., Zanotelli, D., Celano, G., Gucci, R., Inglese,
P., Pitacco, A. and Tagliavini, M. (2016). A survey of carbon sequestration potential of orchards and vineyards in
Italy. European Journal of Horticultural Science, vol. 81, no. 2, pp. 106–114.
Scavo, F.B., Tina, G.M., Gagliano, A., Merlo, L. and Bizzarri, F. (2020). Assessment of the evaporation rate in reservoir
partially covered by floating photovoltaic plants. In: 11th International Renewable Energy Congress (IREC), pp. 1–6.
IEEE, Hammamet, Tunisia.
Scavo, F.B., Tina, G.M., Merlo, L. and Nizetic, S. (2019). Modelling, experimental analysis and optimisation of floating
photovoltaic power plants. Tech. Rep., University of Catania, Sicily.
Scharl, A. (2007). Knowledge sharing and community building in contextualized information spaces. 5th International
Symposium on Digital Earth, vol. 1, no. 1, pp. 1–10.
Schelly, C., Lee, D., Matz, E. and Pearce, J.M. (2021). Applying a Relationally and Socially-embedded Decision
Framework to Solar Photovoltaic Adoption: A Conceptual Exploration. Sustainability, vol. 13, no. 711, pp. 1–18.
Scherr, S.J., Shames, S. and Friedman, R. (2012). From climate-smart agriculture to climate-smart landscapes.
Agriculture and Food Security, vol. 1, no. 1, pp. 1–12.
Schlesinger, S. (1979). Terminology for model credibility. Transactions of The Society for Modeling and Simulation
International, vol. 32, no. 3, pp. 103–104.
342
Schneider, L. and Hutter, K. (2009). Mathematical Preliminaries and Notations. In: Solid-Fluid Mixtures of Frictional
Materials in Geophysical and Geotechnical Context. Advances in Geophysical and Environmental Mechanics and
Mathematics. Springer-Verlag, Berlin, Germany.
Schubert, J., Seichter, L.T.C.S. and Huber, D. (2017). Data Farming Decision Support for Operation Planning. 11th
NATO Operations Research and Analysis Conference, vol. 1, no. 8, pp. 1–20.
SE4ALL (2017). Energizing Finance: Scaling and Refining Finance in Countries with Large Energy Access Gaps.
Tech. Rep. 2, United Nations, Washington DC, USA.
Seaflex (2020). SEAFLEX - The Mooring System. Available at: http://www.seaflex.net/ [Accessed 12 Oct 2020].
SEIA (2020). Solar Power & Renewables Market Insight Report 2020. Tech. Rep., Solar Energy Industry Association
(SEIA) and Wood Mackenzie.
Semelane, S. (2021). Just Energy Transition. Council for Scientific and Industrial Research, Parliament Webinar,
vol. 2, pp. 1–20.
Semeraro, C., Lezoche, M., Panetto, H. and Dassistincetta, M. (2021). Digital Twin Paradigm: A Systematic Literature
Review. Computers in Industry, vol. 130, no. 469, pp. 1–42.
Semeskandeh, S., Hojjat, M. and Hosseini Abardeh, M. (2022). Techno-economic-environmental comparison of
floating photovoltaic plant with conventional solar photovoltaic plant in northern Iran. Clean Energy, vol. 6, no. 2,
pp. 353–361.
Sen, A., Mohankar, A.S., Khamaj, A. and Karmakar, S. (2021). Emerging OSH Issues in Installation and Maintenance
of Floating Solar Photovoltaic Projects and their Link with Sustainable Development Goals. Risk Management and
Healthcare Policy, vol. 14, pp. 1939–1957.
Sengupta, M., Habte, A., Wilbert, S., Gueymard, C. and Remund, J. (2021). Best Practices Handbook for the Collec-
tion and Use of Solar Resource Data for Solar Energy Applications: Third Edition. Tech. Rep., NREL.
SERIS (2021). Floating PV R&D and Testbed. Solar Energy Research Institute of Singapore. Available at:
www.seris.nus.edu.sg/services/Floating-PV-R&D-and-Testbed.html [Accessed 5 Jan 2021].
Sharma, S. and Waldman, J. (2021). Potential Solar Replacement of Hydroelectricity to Reopen Rivers: Maine as a
Case Example. Fisheries, vol. 46, no. 8, pp. 383–390.
Sharp Corporation (2008). U.S. Vineyard uses Space-saving Floatovoltaics. Renewable Energy Focus, vol. 9, no. 4,
pp. 64–65.
Short, W., Packey, D. and Holt, T. (1995). A Manual for the Economic Evaluation of Energy Efficiency and Renewable
Energy Technologies. NREL.
Si, S.L., You, X.Y., Liu, H.C. and Zhang, P. (2018). DEMATEL Technique: A Systematic Review of the State-of-the-Art
Literature on Methodologies and Applications. Mathematical Problems in Engineering, vol. 2018, no. 1, pp. 1–33.
Silverman, J. (2007). The Arithmetic of Dynamical Systems. 1st edn. Springer-Verlag, New York, USA.
Sim, J. and Wright, C. (2000). Research in Health Care: Designs and Methods. Nelson Thornes.
Simpson, G.B., Jewitt, G.P.W., Becker, W., Badenhorst, J., Masia, S., Neves, A.R., Rovira, P. and Pascual, V. (2022).
The Water-Energy-Food Nexus Index: A Tool to Support Integrated Resource Planning, Management and Security.
Frontiers in Water, vol. 4, no. 3, pp. 1–17.
Singh, N.K., Goswami, A. and Sadhu, P.K. (2022). Technical and sustainable goal-based assessment of combined
floating solar PV systems with hydroelectric power station for cost-effective low-carbon clean energy generation.
Clean Technologies and Environmental Policy, vol. 23, no. 1, pp. 1–12.
Singleton, A. and Arribas-Bel, D. (2021). Geographic Data Science. In: Geographical Analysis, pp. 1–8.
Sips, M., Schneidewind, J. and Keim, D.A. (2007). Highlighting space-time patterns: Effective visual encodings for
interactive decision-making. International Journal of Geographical Information Science, vol. 21, no. 8, pp. 879–893.
Siweya, N. (2021). Carbon tax to lower emissions: the likely impact of carbon emissions tax on households in South
Africa. African Multidisciplinary Tax Journal, vol. 2021, no. 1, pp. 211–227.
Smith, T. (2021). Exploring floating solar, a first for Cape Town municipality. Available at: https://www.esi-
africa.com/industry-sectors/generation/solar/exploring-floating-solar-a-first-for-cape-town-municipality/ [Accessed
15 Oct 2021].
Sohofi, A., Melkonyan, A., Karl, C. and Krumme, K. (2016). System Archetypes in the Conceptualization Phase of
343
Water-Energy-Food Nexus Modeling. Proceedings of the 34th International Conference of the System Dynamics
Society, vol. 34, pp. 1–16.
Solar Energy Lab (2017). Weather Data: Monthly Average Data, Standard TMY (Typical Meteorological Year), TMY2
Data and Generated Hourly Weather Files. Available at: http://sel.me.wisc.edu/trnsys/weather/weather.htm [Ac-
cessed 2 May 2017].
SolarGIS (2018). Satellite-derived Solar Radiation and Meteorological Data. SolarGIS Database version 1.8, Geo-
Model Solar, vol. 1, no. 8, pp. 1–12.
Solomin, E., Sirotkin, E., Cuce, E., Selvanathan, S.P. and Kumarasamy, S. (2021). Hybrid Floating Solar Plant
Designs: A Review. Energies, vol. 14, no. 2751, pp. 1–25.
Solomon, B., Pasqualetti, M. and Luchsinger, D. (2003). Energy Geography. In: Gaile, G. and Willmott, C. (eds.),
Geography in America at the Dawn of the 21st Century, 1st edn, chap. Chapter 20, pp. 302–313. Oxford University
Press, New York. ISBN 0-19-823392-2.
Sonke, D. (2008). Brief History of Sustainable Certifications and the Protected Harvest Program. Tech. Rep., Sus-
tainable Solutions.
Soulemane Hayibo, K., Pierce Mayville and M. Pearce, J. (2022). The greenest solar power: Life cycle assessment
of foam-based flexible floatovoltaics. Sustainable Energy & Fuels, pp. 1–16.
Spencer, R.S. and Barnes, T.M. (2020). Betting on Floatovoltaics: Floating PV Opportunities in the U.S. Solar Summit,
vol. 2, no. 1, pp. 1–28.
Spencer, R.S., Macknick, J., Aznar, A., Warren, A. and Reese, M. (2019). Floating Photovoltaic Systems: Assessing
the Technical Potential of Photovoltaic Systems on Man-made Water Bodies in the Continental USA. Environmental
Science & Technology, vol. 53, no. 3, pp. 1680–1689.
SPG Solar (2010). Floatovoltaics Solar Power System. Overview and SPG Solar Statement of Qualifications, vol. 1,
no. 2, pp. 1–19.
Springer, R. (2013). A Framework for Project Development in the Renewable Energy Sector. National Renewable
Energy Laboratory. Tech. Rep., NREL.
Stanton, N., Salmon, P. and Walker, G. (2019). Systems Thinking in Practice: Applications of the Event Analysis of
Systemic Teamwork Method. Routledge. ISBN 9781138097872.
Staplin, N., Herrington, W.G., Judge, P.K. and Reith, C.A. (2017). Use of causal diagrams to inform the design and
interpretation of observational studies. Clinical Journal of the American Society of Nephrology, vol. 12, no. 3, pp.
546–552.
Stein, J. (2017). Challenges of PV Degradation Analysis: PVLIB and Performance Data Analysis. In: Sandia, Energy
Efficiency and Renewable Energy, pp. 1–7. Available at: https://www.osti.gov/biblio/1482294 [Accessed 15 Nov
2019].
Stein, J.S., Hansen, C. and Cameron, C.P. (2011). Evaluation of PV Performance Models and Their Impact on Project
Risk. In: IEEE Power and Energy Society General Meeting, pp. 1–24. IEEE Computer Society, Albuquerque, NM.
Stein, J.S., Holmgren, W.F., Forbess, J. and Hansen, C.W. (2017). PVLIB: Open source photovoltaic performance
modeling functions for Matlab and Python. Conference Record of the IEEE Photovoltaic Specialists Conference,
vol. 11, no. 1, pp. 3425–3430.
Sterl, S., Devillers, A., Chawanda, C.J., van Griensven, A., Thiery, W. and Russo, D. (2022). A spatiotemporal atlas
of hydropower in Africa for energy modelling purposes. Climate Information for Integrated Renewable Electricity
Generation, vol. 1, no. 3, pp. 1–29.
Sterman, J., Oliva, R., Linderman, K. and Bendoly, E. (2015). System dynamics perspectives and modeling opportu-
nities for research in operations management. Journal of Operations Management, vol. 39-40, pp. 1–5.
Sterman, J.D. (2003). System Dynamics: Systems Thinking and Modeling for a Complex World. Engineering Systems
Division Symposium, vol. 1, no. 1, pp. 1–31.
Stoter, J., Ohori, K.A. and Noardo, F. (2021). Digital Twins: A Comprehensive Solution or Hopeful Vision. Avail-
able at: https://www.gim-international.com/content/article/digital-twins-a-comprehensive-solution-or-hopeful-vision
[Accessed 17 Mar 2022].
Strahler, A.N. (1980). Systems Theory in Physical Geography. Physical Geography, vol. 2, no. 1, pp. 1–27.
344
Sudhakar, K. (2019). SWOT analysis of floating solar plants. MedCrave MOJ Solar and Photoenergy Systems Review,
vol. 3, no. 1, pp. 20–22.
Suh, J. and Brownson, J.R. (2016). Solar farm suitability using geographic information system fuzzy sets and analytic
hierarchy processes. Energies, vol. 9, no. 8, pp. 1–24.
Suh, J., Jang, Y. and Choi, Y. (2020). Comparison of Electric Power Output Observed and Estimated from Floating
Photovoltaic Systems: A Case Study on the Hapcheon Dam, Korea. Sustainability, vol. 12, no. 1, pp. 1–12. ISSN
2071-1050.
Sullivan, J. and Sanders, L. (1974). Method for Obtaining Wet-Bulb Temperatures by Modifying the Psychometric
Formula. Tech. Rep., Center for Experiment Design and Data Analysis, Washington.
SunGrow (2020). Floating Photovoltaic and Pontoon Systems. Tech. Rep., SunGrow Power Systems.
Suslov, S. and Katalevsky, D. (2019). Modeling and Simulation in Complex Project Management. AnyLogic Simulation
Software. In: Evolving Toolbox for Complex Project Management, pp. 1–86. Taylor & Francis, CRC Press.
Sustainable Energy Africa (2017). Sustainable energy solutions for South African local government: A practical guide.
Sustainable Solar Photovoltaic (PV) Systems, vol. 2, no. 1, pp. 1–34.
Sustainalytics (2021). ESG Risk Ratings: A consistent approach to assess material ESG risk. Available at:
https://www.sustainalytics.com/esg-data [Accessed 1 May 2021].
Sutanto, B. and Indartono, Y.S. (2019). Computational fluid dynamic (CFD) modelling of floating photovoltaic cooling
system with loop thermosiphon. In: American Institute of Physics Proceedings, vol. 2062, pp. 1–6.
Sutterlueti, J. (2015). Using advanced PV and BoS Modelling and Algorithms to Optimize the Performance of Large
Scale Utility Applications. 4th PV Performance Modelling and Monitoring Workshop, vol. 2, no. 1, pp. 1–30.
Sylvia, T. (2019). America’s largest floating solar project completed. PV Magazine USA, vol. 10, no. 1, pp. 1–3.
System Dynamics Society (2021). System Dynamics Society, Special Interest Group on Energy. Available at:
https://systemdynamics.org/energy/ [Accessed 23 Aug 2021].
Taboada, M.E., Clceres, L., Graber, T.A., Galleguillos, H.R., Cabeza, L.F. and Rojas, R. (2017). Solar Water Heating
System and Photovoltaic Floating Cover to Reduce Evaporation: Experimental Results and Modeling. Renewable
Energy, vol. 105, pp. 601–615.
Taguta, C., Senzanje, A., Kiala, Z., Malota, M. and Mabhaudhi, T. (2022). Water-Energy-Food Nexus Tools in Theory
and Practice: A Systematic Review. Frontiers in Water, vol. 4, no. 3, pp. 1–21.
Tavassoli, M. and Hernandez, R.R. (2012). Impacts of Soil Disturbance on Soil Respiration and Carbon Sequestration
in Desert Scrublands: Implications for Utility-Scale Solar Energy Development. In: 2012 Society for Advancement
of Hispanics/Chicanos and Native Americans in Science National Conference.
Tawalbeh, M., Al-Othman, A., Kafiah, F., Abdelsalam, E., Almomani, F. and Alkasrawi, M. (2021). Environmental
impacts of solar photovoltaic systems: A critical review of recent progress and future outlook. Science of the Total
Environment, vol. 759, no. 143528, pp. 1–12.
Tercan, E., Dereli, M.A. and Saracoglu, B.O. (2022). Location alternatives generation and elimination of floatovoltaics
with virtual power plant designs. Renewable Energy, vol. 2, no. 1, pp. 1–159.
Testori, G. (2021). Towards a sustainability assessment framework for floating photovoltaic systems in cities. Thesis,
University of Twente.
Thacker, B.H., Doebling, S.W., Hemez, F.M., Anderson, M.C., Pepin, J.E. and Rodriguez, E.A. (2004). Concepts of
Model Verification and Validation. Tech. Rep., Los Alamos National Laboratory (LANL), Los Alamos, NM.
The Guardian (2021). Renewable energy growth must speed up to meet Paris goals. Available at:
https://www.theguardian.com/environment/2021/mar/15/renewable-energy-growth-must-speed-up-to-meet-paris-
goals-agency-says [Accessed 16 Mar 2021].
Tina, G.M., Bontempo Scavo, F., Merlo, L. and Bizzarri, F. (2021a). Analysis of water environment on the performances
of floating photovoltaic plants. Renewable Energy, vol. 175, pp. 281–295.
Tina, G.M., Bontempo Scavo, F., Merlo, L. and Bizzarri, F. (2021b). Comparative analysis of monofacial and bifacial
photovoltaic modules for floating power plants. Applied Energy, vol. 281, no. 116084, pp. 1–14.
Tina, G.M., Cazzaniga, R., Rosa-Clot, M. and Rosa-Clot, P. (2018). Geographic and Technical Floating Photovoltaic
Potential. Journal on Thermal Energy Science, vol. 2, no. 8, pp. 1–12.
345
Tina, G.M., Ventura, C. and Rosa-Clot, M. (2016). Photovoltaic Floating systems: Design, construction and perfor-
mance analysis. Researchgate Project Summary, vol. 1, pp. 1–3.
Tina, G.M., Ventura, C., Sera, D. and Spataru, S. (2017). Comparative Assessment of PV Plant Performance Models
Considering Climate Effects. Electric Power Components and Systems, vol. 45, no. 13, pp. 1381–1392.
TNO (2021). Unique research into floating solar panels. Toegepast Natuurwetenschappelijk Onderzoek, Nether-
lands. Available at: https://www.tno.nl/en/focus-areas/energy-transition/roadmaps/renewable-electricity/solar-
energy/integration-solar-energy-environment/floating-solar-panels/ [Accessed 12 June 2020].
Toba, A.L. and Seck, M. (2016). Modeling Social, Economic, Technical & Environmental Components in an Energy
System. In: Procedia Computer Science, vol. 95, pp. 400–407. Elsevier B.V.
Torhan, S., Grady, C. and Ajibade, I. (2022). Tradeoffs and Synergies Across Global Climate Change Adaptations in
the Food-Energy-Water Nexus. Earth’s Future, vol. 10, no. 201, pp. 1–28.
Trapani, K. and Santafe, M. (2015). Review of Floating Photovoltaic Installations: 2007-2013. Progress in Photo-
voltaics: Research and Applications, vol. 23, no. 4, pp. 524–532.
Truffer, B., Murphy, J.T. and Raven, R. (2015). The geography of sustainability transitions: Contours of an emerging
theme. Environmental Innovation and Societal Transitions, vol. 17, no. 1, pp. 63–72.
Tsuji, S.R.J. (2021). Indigenous Environmental Justice and Sustainability: What Is Environmental Assimilation? Sus-
tainability, vol. 13, no. 15, pp. 1–30.
Turner, S.W.D. and Voisin, N. (2022). Simulation of hydropower at subcontinental to global scales: A state-of-the-art
review. Environmental Research Letters, vol. 17, no. 2, p. 23002.
UN (2020a). The 17 Goals to Sustainable Development: Agenda 2030. Available at: https://sdgs.un.org/goals [Ac-
cessed 12 June 2020].
UN (2020b). Transforming our world: The 2030 Agenda for Sustainable Development. Tech. Rep., United Nations.
UN (2021). UN Climate Change Conference (COP26) agenda, COP26 UN Climate Summit, Glasgow, Scotland.
Available at: https://ukcop26.org/ [Accessed 4 Oct 2021].
UN (2022). United Nations Sustainable Development Goals Acceleration Toolkit. Available at:
https://sdgintegration.undp.org/sdg-acceleration-toolkit [Accessed 2 Feb 2022].
UN Chronicle (2015). United Nations Goal 7: Ensure Access to Affordable, Reliable, Sustainable and Modern Energy
for All. Magazine of the United Nations, vol. 2, no. 42014, pp. 1–5.
UNESCO (2020). The United Nations World Water Development Report 2020: Water and Climate Change. Tech.
Rep., UNESCO.
Unisa (2013). Research Ethics Policy. Guidelines for Ethical Research, vol. 1, no. 2013, pp. 1–26.
United Nations (2019). Climate Action Summit 2019 New York, USA. Available at:
https://www.un.org/en/climatechange/ [Accessed 24 Sep 2019].
Urbaniec, K., Mikulcic, H., Rosen, M. and Duic, N. (2017). A Holistic Approach to Sustainable Development of Energy,
Water and Environmental Systems. Journal of Cleaner Production, vol. 155, no. 1, pp. 1–11.
USDoE (2020). The Five-Step Process Framework for Project Development. US Department of Energy. Tech. Rep.,
US Department of Energy.
Vafaei, N., Ribeiro, R.A. and Camarinha-Matos, L.M. (2016). Normalization techniques for multi-criteria decision
making: Analytical hierarchy process case study. IFIP Advances in Information and Communication Technology,
vol. 470, pp. 261–269.
Vafaei, N., Ribeiro, R.A. and Camarinha-Matos, L.M. (2020). Selecting Normalization Techniques for the Analytical
Hierarchy Process. In: Camarinha-Matos, L. (ed.), Technological Innovation for Life Improvement. DoCEIS 2020.
IFIP Advances in Information and Communication Technology, vol. 577, pp. 43–52. Springer, Cham.
Vairavamoorthy, K. (2019). Water and the SDGs. Available at: https://www.thesourcemagazine.org/water-and-the-
sdgs/ [Accessed 22 March 2022].
Van de Ven, J. (2016). Roles and application of carbon pricing in design of projects for the Green Economy Transition.
European Bank for Reconstruction and Development, vol. 2, no. 1, pp. 1–6.
van Niekerk, W. (2010). Solar Energy Technology Roadmap of South Africa (SETRSA). Department of Science and
Technology South Africa, vol. 1, no. 2, pp. 1–106.
346
Vargas, R.V. (2010). Using the analytic hierarchy process (AHP) to select and prioritize projects in a portfolio. Tech.
Rep., PMI Global Congress Newtown Square, PA: Project Management Institute., Washington DC, USA.
Vazquez, F. (2018). Ontology and Data Science. International Journal of Business Analytics and Intelligence, vol. 1,
pp. 1–23.
Velten, S., Leventon, J., Jager, N. and Newig, J. (2015). What Is Sustainable Agriculture: A Systematic Review.
Sustainability, vol. 7, no. 6, pp. 7833–7865.
Vema, V.K., Sudheer, K.P., Rohith, A.N. and Chaubey, I. (2022). Impact of water conservation structures on the
agricultural productivity in the context of climate change. Water Resources Management, vol. 36, no. 5, pp. 1627–
1644.
Verbruggen, S. (2018). Crunching the numbers on floating solar. PV-Tech Technical Briefing: Design and Build,
vol. 37, no. 9, pp. 1–3.
Vides-Prado, A., Camargo, E.O., Vides-Prado, C., Orozco, I.H., Chenlo, F., Candelo, J.E. and Sarmiento, A.B. (2018).
Techno-economic feasibility analysis of photovoltaic systems in remote areas for indigenous communities in the
Colombian Guajira. Renewable and Sustainable Energy Reviews, vol. 82, no. 5, pp. 4245–4255.
Villeneuve, C., Tremblay, D., Riffon, O., Lanmafankpotin, G.Y. and Bouchard, S. (2017). A Systemic Tool and Process
for Sustainability Assessment. Sustainability, vol. 9, no. 1909, pp. 1–29.
Vlacic, L., Engwirda, A. and Kajitani, M. (2000). Cooperative Behavior of Intelligent Agents: Theory and Practice. In:
Soft Computing and Intelligent Systems, pp. 279–307. Academic Press Series in Engineering, London.
Vos, P. and Shults, L. (2015). Learning about Simulations as a Research Approach. Mathematics Education Research
Group Agder, Department Mathematical Sciences, vol. 1, no. 2, pp. 1–3.
VTP (2014). Virtual Terrain Project: Foster the creation of tools for easily constructing any part of the real world in
interactive in 3D digital form. In: Synergetic Convergence of the Fields of CAD, GIS, Visual Simulation, Surveying
and Remote Sensing, pp. 1–20. VTerrain.
Wackernagel, M., Onisto, L., Bello, P., Linares, A.C., Falfán, I.S.L., García, J.M., Guerrero, A.I.S. and Guerrero, M.G.S.
(1999). National natural capital accounting with the ecological footprint concept. Ecological Economics, vol. 29,
no. 3, pp. 375–390.
Wagner, W., C. Verhoest, N.E., Ludwig, R. and Tedesco, M. (2009). Remote sensing in hydrological sciences. Hy-
drology and Earth System Sciences, vol. 13, no. 6, pp. 813–817.
Wainwright, J. and Mulligan, M. (eds.) (2013). Environmental Modelling: Finding Simplicity in Complexity. Wiley-
Blackwell, New York, NY.
WEC (2021). Energy Trilemma Index. Tech. Rep., World Energy Council, London, U.K.
Weigl, T. (2021). Exploring the Environmental Impact of Floating-PV Today and in the Future. Tech. Rep., Solarplaza
Floating Solar.
Whitaker, C., Townsend, T., Newmiller, J. and King, D. (1997). Application and validation of a new PV performance
characterization method. Conference Record of the Twenty Sixth IEEE Photovoltaic Specialists Conference, vol. 1,
no. 1, pp. 1253–1256.
Whitaker, C.M. and Newmiller, J.D. (1998). Photovoltaic Module Energy Rating Procedure Final Subcontract Report.
Tech. Rep., National Renewable Energy Laboratory.
Wiener, N. (1949). Cybernetics: Control and Communication in the Animal and the Machine. The Economic Journal,
vol. 59, no. 236, pp. 573–575.
Wiener, N. (1954). The Human use of Human Beings: Cybernetics and Society. Da Capo Press, New York, USA.
Wierzbicki, A.P., Makowski, M. and Wessels, J. (2000). Model-Based Decision Support Methodology with Environ-
mental Applications, vol. 2. Springer-Verlag, Berlin Heidelberg. ISBN 0792363272.
Willard, J., Jia, X., Xu, S., Steinbach, M. and Kumar, V. (2020). Integrating Physics-Based Modeling With Machine
Learning: A Survey. Computational Physics arXiv, vol. 1, no. 1, pp. 1–34.
Wines, V. (2019). The First Floatovoltaic Solar Array:Napa’s Far Niente Floats Solar Panels on Irrigation Pond to
avoid removing Vines. Available at: https://www.winesandvines.com/news/article/54755/The-First-Floatovoltaic-
Solar-Array [Accessed 2 Jul 2019].
Wood Mackenzie (2019). Floating Solar Landscape 2019. Wood Mackenzie, New York, USA.
347
Woodhouse, M., Feldman, D., Ramasamy, V., Smith, B., Silverman, T., Barnes, T., Zuboy, J. and Margolis, R. (2021).
Research and Development Priorities to Advance Solar Photovoltaic Lifecycle Costs and Performance. National
Renewable Energy Laboratory, vol. 1, no. 10, pp. 1–52.
World Bank (2013). Thirsty Energy: Securing Energy in a Water-constrained World. The Water-Energy Challenge,
vol. 8, no. 1, pp. 1–3.
World Bank (2017). State of Electricity Access Report 2017. In: International Bank for Reconstruction Development,
pp. 1–110. World Bank, Washington DC, USA.
World Bank (2019a). Where Sun Meets Water: Floating Solar Handbook for Practitioners. World Bank Report, pp.
1–146.
World Bank (2019b). Where Sun Meets Water: Floating Solar Market. World Bank Group, ESMAP and SERIS, vol. 1,
no. 2, pp. 1–131.
WRC (2016). Water-Energy nexus in the context of climate change: Investigating trade-offs between water use
efficiency and renewable energy options for South Africa. 2239. Water Research Commission, Cape Town, SA.
ISBN 9781431207534.
WWF (2021). World Wide Fund for Nature Conservation Champions in the South African Wine Industry, WWF South
Africa. Available at: https://www.wwf.org.za/our_work/initiatives/conservation_champions/ [Accessed 5 Feb 2021].
Xia, S., Zhou, X.-N. and Liu, J. (2017). Systems thinking in combating infectious diseases. Infectious Diseases of
Poverty, vol. 6, no. 144, pp. 1–7.
Xia, Z., Li, Y., Guo, X. and Chen, R. (2022). High-resolution mapping of water photovoltaic development in China
through satellite imagery. International Journal of Applied Earth Observation and Geoinformation, vol. 107, no.
102707, pp. 1–11.
Yadav, R., Singh, V. and Prabhu, J. (2021). A Novel Model Object Oriented Approach to the Software Design. In:
Satapathy, S. (ed.), Intelligent Data Engineering and Analytics, Intelligent Technologies and Robotics: Advances in
Intelligent Systems and Computing, pp. 369–381. Springer, Singapore.
Yang, C., Goodchild, M., Huang, Q., Nebert, D., Raskin, R., Xu, Y., Bambacus, M. and Fay, D. (2011). Spatial cloud
computing: How can the geospatial sciences use and help shape cloud computing? International Journal of Digital
Earth, vol. 4, no. 4, pp. 305–329.
Yang, P., Chua, L.H.C., Irvine, K.N. and Imberger, J. (2021). Radiation and energy budget dynamics associated with
a floating photovoltaic system. Water Research, vol. 206, no. 117745, pp. 1–17.
Yilmaz, O.S., Gulgen, F. and Ates, A.M. (2022). Determination of the appropriate zone on dam surface for Floating
Photovoltaic System Installation using Remote Sensing and GISc technologies. International Journal of Engineer-
ing and Geosciences, vol. 8, no. 1, pp. 63–75.
Zarei, S., Bozorg-Haddad, O., Singh, V.P. and Loáiciga, H.A. (2021). Developing water, energy, and food sustainability
performance indicators for agricultural systems. Scientific Reports, vol. 11, no. 1, p. 22831.
Zey, M. (2001). Rational Choice and Organization Theory. International Encyclopedia of the Social and Behavioral
Sciences, vol. 1, no. 2, pp. 12751–12755.
Ziar, H. (2021). Floating Solar Stations. Tech. Rep., Delft University of Technology, Delft, Netherlands.
348
