Content uploaded by Pablo Martinez
Author content
All content in this area was uploaded by Pablo Martinez on Dec 09, 2021
Content may be subject to copyright.
1
Preprint submitted to Journal of Cleaner Production December 12, 2019
Towards automated aquaponics: a review on monitoring, IoT, and
smart systems
A. Reyes Yanesa, P. Martinezb, R. Ahmada*
a Department of Mechanical Engineering, Laboratory of Intelligent Manufacturing, Design and
Automation (LIMDA), University of Alberta, Edmonton, AB T6G 2G8
5
b Department of Civil and Environmental Engineering, University of Alberta, Edmonton, AB T6G 2W2
* Corresponding author. E-mail address: rafiq.ahmad@ualberta.ca
Abstract
Aquaponics is a farming method that promises to be a good alternative against the food and environmental problem
10
the world is facing. It is a combination between aquaculture (farming of fish) and hydroponics (growing plants without
soil), being a technique to grow plants with the aquaculture effluent. This technique claims to have high water use
efficiency, does not use pesticides and reduce the use of fertilizers, which make this technology green and sustainable.
Since the interest in aquaponics is increasing, the major challenge is to do it feasible and reliable at commercial scale.
The concept of precision farming usually applied in the traditional farming sense is now being introduced, leading to
15
the need to adopt sensing, smart and IoT systems for monitoring and control of its automated processes. Lately,
valuable contributions have been made towards the introduction of fully and semi-automated systems in small-scale
aquaponics systems by automation and manufacturing experts. This paper aims to support research towards a viable
commercial aquaponics solution by identifying, listing, and providing an in-depth explanation of each of the
parameters sensed in aquaponics, and the smart systems and IoT technologies in the reviewed literature. Further, the
20
proposed review highlights potential gaps in the research literature and future contributions to be made in regards of
automated aquaponics solutions.
Keywords: aquaponics ; farming 4.0 ; IoT ; precision farming ; monitoring.
List of Acronyms and Notations
TAN
Total ammonia nitrogen
AOB
Ammonia oxidizing bacteria
NOB
Nitrite oxidizing bacteria
DO
Dissolved oxygen
EC
Electro conductivity
TDS
Total dissolved solids
SL
Salinity
RH
Relative humidity
PPF
Photosynthetic photon flux
YPF
Yield photon flux
PPFD
Photosynthetic photon flux density
PAR
Photosynthetically active radiation
AI
Artificial intelligence
PPT
Parts per trillion
PPM
Parts per million
ISFET
Ion-sensitive field-effect transistor
TAN
Total ammonia nitrogen
1. Introduction
More than 113 million people across 53 countries
25
experienced acute hunger requiring urgent food,
nutrition and livelihoods assistance (IPC/CH Phase 3
or above) in 2018 (Food Security Information
Network (FSIN), 2019). In front of this reality, it is
necessary to look for alternatives to solve this
30
problem. The food demand the world is facing cannot
be maintained by additional natural resources or land
exploitation (König et al., 2018). Looking for
sustainable solutions that contribute to the food
production and consumption, some of the alternatives
35
that can be implemented are: (1) reduce actual meat
consumption, (2) minimize food waste, and (3) modify
the current food production processes (van der Goot et
al., 2016).
In this scenario of food and environmental crisis,
40
the aquaponics farming method come into as a
solution to improve farming productivity. Aquaponics
is defined as the process of growing aquatic organisms
and plants symbiotically (Yep and Zheng, 2019).
Aquaponics have attracted increasing attention
45
because the ability to save resources, high efficiency
and low consumption. Aquaponics have become the
trend of the agricultural development nowadays
(Mchunu et al., 2018).Any aquaponics system is
defined by a recirculating aquaculture system (RAS)
50
and a hydroponic system working together. As a brief
summary, aquatic animals excrete waste, then bacteria
convert the animal waste into nutrients, and plants
2
Preprint submitted to Journal of Cleaner Production December 12, 2019
make use of the nutrients to grow, thus improving the
water quality for the aquatic animals (Love et al.,
55
2014).
Nowadays, the agriculture industry is the world’s
largest user of water, around the 70% of the total
consumption, of which 70% of the water is wasted
throughout their different processes (Kloas et al.,
60
2015), (Murad et al., 2017). One person drinks around
two and five liters of water daily, but it requires nearly
5000 liters of water to produce the daily diet for a
single person (Manju et al., 2017). The sustainable
development strategies have become a global trend,
65
and a circular economy is the general trend of
sustainable development and the best mode of
economic development (Wei et al., 2019). Aquaponics
is known as a form of sustainable agriculture because
it imitates natural systems, where the efficiency of the
70
water is dramatically increased, and has fewer
environmental impacts (Blidaria and Grozea, 2011).
As a sustainable, circular, efficient and intensive low-
carbon production mode in the future, the aquaponics
system has realized the transformation from waste to
75
nutrients and has effectively solve the problem of
environmental pollution (Nichols and Savidov, 2012).
As a modern approach, the earliest application of
aquaponics, as a research area, was in the 70’s and
80’s. With the purpose of improving the quality of
80
water by removing the excess of ammonia in RAS
aquaculture systems, plants were used as bio filters
(Lewis et al., 1978). Aquaponics research started to
grow after 2010 (Junge et al., 2017). Nowadays,
aquaponics is being practiced in at least 43 countries
85
around the world, but 84% of the practitioners use this
technology as a hobby (Love et al., 2014). The
successful development of aquaponics could
guarantee a major part of a more sustainable world
food supply (Junge et al., 2017). The global
90
application of aquaponics will succeed helping the
food crisis and world sustainability as long as it
becomes widely spread as a commercial alternative.
Only 31% of the commercial aquaponics facilities
reported to be profitable and 47% rely in others
95
products or services for additional income (Love et al.,
2014). As such, research challenges still exist to
procure viable commercial aquaponics facilities.
Yep and Zheng wrote a review named Aquaponics
trends and challenges where after conducting a
100
literature analysis were addressed major trends : types
of aquaponic systems, hydroponics components, plant
species, fish species, nitrifying bacteria, microflora
and additional species (Yep and Zheng, 2019). This
contribution helps in the understanding of Aquaponics
105
science and to direct efforts into future developments,
nevertheless the automation and control systems are
not included as a factor that impact the developing of
the addressed parameters.
A proper design and correct management of the
110
system are the key points towards the economic
feasibility and the global success of aquaponics itself
(Tyson et al., 2011). The design and management of
an aquaponic system is a difficult challenge when
trying to achieve high yields and quality. Being a
115
greenhouse and a symbiotic environment, the
parameters and factors (light, temperature, pH,
moisture, etc.) that need to be controlled are diverse.
This makes the task of manually analyzing and
managing such systems exponentially hard when
120
scaling the system up to commercial levels.
With the introduction of automation, smart
strategies, and connectivity in the farming industry, a
new door was opened for the improvement of these
aquaponics systems. The expected benefits of smart
125
automation are a significant reduction of manual labor,
a more robust control of the process by increasing the
accessibility and connectivity of the parameters, and
using computer capabilities to make data-driven
decisions (Martinez et al., 2019a). One of the recurrent
130
researches in the last three years is the implementation
of sensing, smart, or IoT systems in aquaponics with
the aim of solving some of the proposed challenges by
the previous authors. By analyzing past publications,
contributions make notorious too many differences
135
between their proposed systems. The learning curve
for automation experts interested to contribute in
aquaponics, and vice versa, is steep and does not
encourage cross contributions, limiting the progress
and feasibility of aquaponics at a commercial scale.
140
Questions arise that, if answered, will help the
future academic and commercial contributors in the
research area. Which parameters are involved, and
which ones can control, monitor and/or predict
aquaponics system behavior? Which monitoring,
145
smart and IoT technologies are currently being
researched towards commercialization of aquaponics
systems? The information provided in this paper may
be used then as a bridge between these two academics
areas and serve as a guide for those interested to
150
initiate in the enhancement of aquaponics through
precision farming.
2. Research methodology
The analysis is based in a comprehensive literature
review of monitoring, smart and IoT systems in
155
aquaponics. The objective of this research is to
synthesize the current knowledge and approaches on
monitoring systems for aquaponics. Towards this
objective, a quantitative review method, e.g.
3
Preprint submitted to Journal of Cleaner Production December 12, 2019
systematic approach, is employed in this study. Since
160
there are scarce publications in literature about
aquaponics systems, some hydroponic systems were
included in the analysis to complement the
information.
The systematic analysis presented in this paper is
165
based on a qualitative analysis of carefully selected
journal and conference papers. The task was
performed manually by the authors; as a result, the
identified and presented articles are reviewed one by
one and categorized based on their research focus and
170
results. Publications written only in English were
considered and no review articles were included in this
research. The systematic review is carried out to
provide a comprehensive view of existing research
with the purpose of identifying gaps in the body of
175
knowledge and provide deep understanding of the
current status of the studied research area (Martinez et
al., 2019b; Zhang et al., 2018). Figure 1 displays the
overview of the research methodology adopted
Table 1 lists all the selected publications and their
180
integration of sensors based on the researched
parameters, smart approaches, and IoT technology. In
summary, 21 aquaponic (A) or hydroponic (H)
systems are selected from the literature and their
proposed monitoring systems are evaluated in depth.
185
Overall, 19 physical parameters are identified as being
actively studied and are considered by researchers as
critical for aquaponic systems. ‘Smart’ includes the
use of machine learning, deep learning, prediction
algorithms and decision-making. IoT involves, for
190
example, remote control through web based or mobile
applications. The remaining parts of the paper focus
on the following topics:
The monitoring parameters found in the analysis
are then defined, classified by water and
195
environmental factors. Then, the importance of each
of them in an aquaponics system is presented and how
to measure and control them in an automated system
is discussed. The admissible high and low levels for
each parameter in the symbiotic (aquaculture,
200
nitrification, hydroponic) environment are finally
summarized. Then, a suggested location for each of
the mentioned sensors for both coupled and decoupled
aquaponics system is proposed based on literature
analysis. As such, a summary of all the parameters that
205
influence the behavior and final growth results in
aquaponics systems is given. Each parameter is listed
with the proposed adequate ranges for each
component, namely the aquaculture, the nitrification
process, and the hydroponic component. Further, the
210
potential side effects on the system when outside of
the proposed range are provided for each parameter.
Smart and IoT frameworks and techniques are
presented. Some future considerations and possible
applications in the areas are then proposed. A
215
discussion part is presented where high and low levels
for a sustainable equilibrium in the three components
is proposed for an aquaponics environment. The
authors hope that illustrating the relationship between
sensors and each parameter serves as a good start to
220
introduce automation in aquaponics at a commercial
level.
3. Monitoring parameters
The approach to measure the recurring parameters
can vary from system to system. As an example, the
225
water pH sensing system can vary from pH test strips,
some standalone sensors with LCD screen attached, or
analog sensors capable of transmitting the
information, wireless or not, to some controller (PLC,
micro-controller, etc.). With the aim of designing a
230
reliable, sustainable and economic feasible system,
automated sensing techniques need to be evaluated.
The monitoring and control of environment and
equipment through intelligent technology is the
premise and foundation to ensure the stable operation
235
of aquaponics system (Wei et al., 2019). Along the
years, different types of aquaponics systems have been
proposed, adopted and explained (Goddek et al.,
2015), mainly categorized as coupled and decoupled
systems. Figure 2 summarizes the proposed location of
240
the sensors for the aforementioned aquaponics
systems. In the following subsections, each parameter
is introduced and its role in aquaponic systems is
explained. Then, an explanation on how authors
measured, used, and controlled them is given.
245
Figure 1. Overview of research methodology.
5
Preprint submitted to Journal of Cleaner Production December 12, 2019
Table 1. List of sensed parameters, smart systems and IoT systems in aquaponic or hydroponic selected publications.
Author
Type
Sensing
Smart
IoT
Water
Air
Light
Moisture
NO3-
NO2-
Height
pH
EC
T
Level
DO
NH3
TDS
SL
Flow
T
RH
CO2
(Wang et al., 2015)
A
✓
✓
✓
✓
✓
✓
(Kumar et al., 2016)
A
✓
✓
✓
✓
✓
(Kyaw and Ng, 2017)
A
✓
✓
✓
✓
✓
(Murad et al., 2017)
A
✓
✓
✓
✓
(Nagayo et al., 2017)
A
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
(Mamatha and Namratha, 2017)
A
✓
✓
✓
✓
✓
✓
(Pitakphongmetha et al., 2016)
H
✓
✓
✓
✓
(Palande et al., 2018)
H
✓
✓
✓
(Mehra et al., 2018)
H
✓
✓
✓
✓
✓
✓
(Aishwarya et al., 2018)
A
✓
✓
✓
✓
(Vernandhes et al., 2017)
A
✓
✓
✓
✓
(Manju et al., 2017)
A
✓
✓
✓
✓
✓
✓
(Sreelekshmi and Madhusoodanan, 2018)
A
✓
✓
✓
✓
✓
✓
✓
(Jacob, 2017)
A
✓
✓
✓
✓
✓
✓
(Dutta et al., 2018)
A
✓
✓
✓
✓
✓
(Zamora-Izquierdo et al., 2019)
H
✓
✓
✓
✓
(Odema et al., 2018)
A
✓
✓
✓
✓
✓
✓
✓
(Elsokah and Sakah, 2019)
A
✓
✓
✓
✓
(Haryanto et al., 2019)
A
✓
✓
✓
✓
(Mandap et al., 2018)
A
✓
✓
✓
(Naser et al., 2019)
A
✓
✓
✓
✓
6
Preprint submitted to Journal of Cleaner Production December 12, 2019
3.1 Water
The quality of the water is commonly presented as
250
the most important factor in aquaponics processes
(Odema et al., 2018). This can be easily understood
since water is used as the medium to provide nutrients
to the plants. In addition, water is the most complex
factor in the automation point of view because implies
255
synchronous control of several parameters that are
dependent of each other. As a combination of
aquaculture and hydroponics, aquaponics is the result
of mixing two well-known techniques that, nowadays,
are still being widely adopted and developed
260
individually. The RAS design in aquaculture came up
for water efficiency and sustainability. However, the
ammonia in the water started to accumulate at deadly
levels for the fish. As such, bio filters started to be used
for recirculating the water. In the other hand, the plants
265
in the hydroponic approach need nutrients and
elements that the water cannot provide without the use
of fertilizers. Using fertilizers in hydroponics leads to
water disposal and replacement. The dependency
between these two techniques in aquaponics lies in the
270
transformation that the water goes through between
them. Technically, the plants grow with the effluent of
the fish tank. However, this process is not
straightforward and occurs thanks to a process called
nitrification, explained below.
275
To ensure ideal quality standards in the water
solution to favor the nitrification process and plant
growth, while keeping the fishes healthy, it is
theoretically necessary to maintain the correct amount
of nutrients, pH, temperature, dissolved oxygen and
280
salt throughout the whole process. The methods to
measure or sense each of the parameters are varied. In
the next sections, each parameter will be explained in
depth.
3.1.1 Nitrification
285
Nitrogen is the most important inorganic nutrient
for the plants. The nitrification process base is the
ammonia, which is obtained from fish waste. It can be
found in the form of ammonia (NH3) and ammonium
(NH4+), where the concentration of both in the water
290
solution is a function of the pH, the temperature and
the salinity (Anthonisen et al., 1976; Ebeling et al.,
2006). The sum of both is known as total ammonia-
nitrogen concentration TAN (NH4+ + NH3)
(Wongkiew et al., 2017). The process that transforms
295
the TAN into nitrates NO3-, which is a form of nitrogen
that the plants can uptake, is called nitrification (Hu et
al., 2015). First, TAN is oxidized intro nitrite NO2- by
ammonia oxidizing bacteria (AOB). This nitrite NO2-
is then broken down by nitrite oxidizing bacteria
300
(NOB) into Nitrates NO3- (Ebeling et al., 2006). A
typical aquaponic system consists of an aquaculture
component, a biofilter for the nitrification and a
hydroponic component, as illustrated in Figure 3
(Love et al., 2015).
305
When designing the hydroponic component three
different choices can be made for the grow bed: (1)
Figure 2. Location of sensors in a Coupled and Decoupled Aquaponics Systems, after (Yep and Zheng, 2019).
Figure 3. Overview of the nitrification process in a CAS
system with nutrient film technique grow bed, after
(Love et al., 2015).
7
Preprint submitted to Journal of Cleaner Production December 12, 2019
nutrient film technique (NFT); (2) deep water culture
(DWC); and (3) media-based (Goddek et al., 2015).
3.1.1.1 TAN
310
Ammonia is a dissolved gas present naturally in
surface and wastewaters (Stone and Thomforde,
2004). It is a form of nitrogen found in organic
materials and many fertilizers. Ten percent of the
protein in fish feed will be converted into ammonia in
315
the system water (Tyson et al., 2008).Ammonia is
produced by the waste excreted by fish and plays an
important role in the aquaponics system, since it will
serve as initial component for plants nutrients. The
total ammonia-nitrogen generated in the aquaculture
320
component can be calculated by the equation given
below:
𝑃𝑇𝐴𝑁 = 𝐹 ∗ 𝑃𝐶 ∗ 𝑐
where 𝑃𝑇𝐴𝑁 is the production rate of TAN (kg/day), 𝐹
represents the feeding rate (kg/day), 𝑃𝐶 is the protein
content (fractional), and c is the constant amount of
325
excreted TAN per protein input based upon the
feeding rate. The constant is empirically obtained and,
for aquaponics systems, 𝑐 = 0.092 (Ebeling et al.,
2006). The feeding process can be then adjusted by
controlling the production rate of TAN.
330
Ammonia is highly toxic for fish in small amounts,
and it is predominant (relative proportion increases) in
the water solution when it becomes strongly acidic or
alkaline. Stone and Thomforde mentioned that the
desirable range for fishes of TAN is from 0-2 mg/L
335
(Stone and Thomforde, 2004). Somerville et al. make
a difference between warm water fish and cold water
fish (Somerville et al., 2014). In the first case, the
optimum TAN range is <3 mg/L, and in the second
one, it is <1 mg/L. For bacterial activity, namely AOB
340
and NOB, the optimum value is <3 mg/L and <30
mg/L for plants (Somerville et al., 2014).
Ammonia is colorless and odorless in small
amounts, so sensing it is the best form of knowing the
presence of the parameter. Usually these sensors
345
consist of a wire electrode in a custom filling solution.
The internal solution is segregated from the sample
medium by a ion selective membrane, which interacts
with ammonium ions (YSI, n.d.). To increase the
ammonia measurement accuracy, it is necessary to
350
know the pH and temperature of the water. As such,
quantifying the amount of ammonia in water solutions
becomes a data fusion problem between pH sensors,
temperature sensors, and ammonia concentration
sensors. As shown in Figure 2, those sensors need to
355
be included in the water tank, as the concentration of
ammonia past the biofilter can be considered
negligible.
3.1.1.2 Nitrite
Nitrite is obtained from ammonia by the AOB.
360
Nitrite is another form of nitrogenous waste that is
deadly for aquatic life (Bhatnagar and Devi, 2013).
The desirable range of nitrite dissolute in water to
allow fish, plants and bacteria’s life is 0-1 mg/l (Stone
and Thomforde, 2004). A similar range of nitrite is
365
required for bacterial activity and plant growth. The
presence of nitrite should not be a problem if
maintained in the optimal range. Nitrite concentration
sensors are usually a combination of nitrite-ionized
electrodes and a sensing element made of a plastic
370
(PVC) membrane, working as an ion exchanger and
reference electrode. The sensor develops an electrical
potential proportional to the concentration of nitrite
ions in the solution, thus providing the concentration
of nitrite in the water.
375
3.1.1.3 Nitrates
Nitrate is the result of the nitrification process by NOB
and is a form of nitrogen component that plants can
uptake. Nitrate is relatively non-toxic to fish. Stone
and Thomforde mentioned that nitrate should not
380
cause health issues if maintained below 90 mg/L
(Stone and Thomforde, 2004). Bhatnagar and Devi,
recommend a normally stabilized range of 50-100
ppm. This factor is important when designing the
biofilter. High amounts of nitrates could mean an
385
under-sized bio filtration and be toxic for fishes
(Bhatnagar and Devi, 2013). The nitrate measurement
is usually done with the same sensor that estimates the
concentration of ammonia, which are commonly
prepared to measure both signals.
390
3.1.2 pH
pH is a measure of hydrogen ion concentration,
usually known as a measure of acidity of alkalinity of
a solution. The water pH affects plant nutrient
availability and the nitrification rate (Kuhn et al.,
395
2010; Tyson et al., 2011). Currently, pH
measurements are obtained using three different
approaches: (1) test strips (2) manual electronic probes
and (3) automatic probes in controllers. Due to the aim
of this paper, only the third method will be covered. A
400
pH meter is an electric device that measures the
hydrogen-ion activity (acidity or alkalinity) in some
solution (Gregersen, 2009).
In the aquaculture component, the desirable range
for the water pH goes from 6.5-9.5 and the acceptable
405
range from 5.5-10, however this range can slightly
vary with fish species (Stone and Thomforde, 2004).
Outside that pH range, water can alter the equilibrium
of the aquaponic system. For example, fish
reproduction rate may be diminished in slightly acidic
410
environments (Stone and Thomforde, 2004). In the
hydroponics component, the optimum pH is around
6.0. A higher pH than 7.0 will cause precipitation of
8
Preprint submitted to Journal of Cleaner Production December 12, 2019
Fe or Mn, and a lower pH than 4.5 can cause root
injury (Wada, 2019) and plants experience nutrient
415
deficiencies (Somerville et al., 2014).
In general, aquaponics systems are sensible to
changes in the water pH. For the nitrification process,
the required pH range goes from 7.0 to 9.0. An
efficiency increase in the range of 8.4-8.8 was reported
420
(Rakocy, 2012). To adjust the pH in the aquaponics
system, bases like potassium and calcium should be
prioritized as they serve as base for nutrients (Rakocy,
2012; Somerville et al., 2014). Smalls changes (< 0.3)
in short periods of time (18-24 h) can highly affect the
425
health of the fishes (Somerville et al., 2014).
In an automated system, the pH meter is connected
to a controller where the controller receives the change
in the pH meter output (mV, mA). For the controller is
necessary to calibrate the sensor. The pH meter is
430
connected to the controller and then tested in a solution
with known ph. The gotten output is then correlated to
the pH units into the controller programing interface.
Zamora-Izquierdo et al., developed a smart farming
IoT platform that focused on pH equilibrium as an
435
agent to ensure high yield. They used a pH meter
(B&C Electronics - SZ 1093) with a range of 0-13,
maximum temperature of 80°C and maximum
pressure of 7 bars (Zamora-Izquierdo et al., 2019).
Mandap et al., 2018 used an ISFET (ion-sensitive
440
field-effect transistor) in their system claiming nearly
the same performance as a 0.01 resolution digital pH
meter (Dr. Meter pH pen tester). In this study, Mandap
et al. ended up suggesting the Atlas EZO pH Sensor,
having this one significantly less percentage error than
445
other possible tested options (Mandap et al., 2018).
Nonetheless, other options still exist. Manju et al.,
used, in their aquaponics system, an OMEGA PHE-
45P pH sensor with lower maximum temperature
resistance (60°C) (Manju et al., 2017), and Kuhn et al.,
450
used an Orion 3 Star meter from Thermo Fisher
Scientific to determine the pH (Kuhn et al., 2010).
3.1.3 Temperature
The temperature in the water is linked to most of
the other water-related parameters in the aquaponics
455
system. For the nitrification process, the optimal
temperature is around 17-34°C (Somerville et al.,
2014), if the water temperature goes below this range,
the productivity of the bacteria will tend to decrease
and the nitrification process will not be successful. For
460
the hydroponics component, the suitable temperature
range is 18-30 °C (Somerville et al., 2014). For the
fish, maintaining a correct temperature diminishes the
risk of diseases. The appropriate temperature varies
depending on the species: for tropical fish, the
465
optimum temperature is 22-32 °C; while for cold-
water fish species, the required temperature is 10-18
°C. Some other fishes have a wider range of suitable
temperature, i.e. 5-30 °C (Somerville et al., 2014).
High water temperatures can restrict plants’
470
absorption of calcium.
The methods to measure the temperature in the
water are varied. The most common practice to
measure water temperature is to check the temperature
range, resolution and the tolerance to the salinity.
475
Since most of water temperature sensors cover the
desired range, the sensor’s resolution is the most
important factor in the selection. Note that the sensor
must be made to be submerged long periods and not
just ‘‘waterproof’’. Thermistors are the most common
480
and widely used to measure water temperature.
Mandap et al., Sreelekshmi and Madhusoodanan, and
Mamatha and Namratha, used a DS1820 temperature
sensor for Arduino controller. The temperature range
in this sensor goes from - 55 °C to 125 °C and the
485
resolution is ± 0.5 °C. Manju et al., 2017 used a LM35
IC National semiconductor temperature sensor
(Mamatha and Namratha, 2017; Mandap et al., 2018;
Sreelekshmi and Madhusoodanan, 2018).
490
3.1.4 Level
The amount of water needed in the system is
determined by the size of the components, especially
the aquaculture ones (fish tanks) (Somerville et al.,
2014). The stocking density in the aquaculture tank
495
highly affects the fish’s growth and health and is one
of the most common root causes for fish stress. The
recommended amount of stocking is 20 kg of fish per
1000 liters of water (Somerville et al., 2014). It is not
recommended to grow fishes for consume in tanks
500
with less than 500 liters. In bigger systems, to mitigate
changes in the water quality parameters has been
reported to be easier than in smaller units.
Every aquaponics system has natural water losses
for mainly four reasons: 1) fish sludge removal; 2)
505
evaporation; 3) plant evapotranspiration; and 4) fish
splashing during feeding (Maucieri et al., 2017). Also,
the hydroponic system consumes a daily amount of
water that typically goes from 0.1% to 3%, depending
on the hydroponic/fish tank ratio, water temperature,
510
flow, plant and fish species, and the hydroponic type
of system used (Maucieri et al., 2017).
The water level in tanks can be manually measured
with sight glass or floating devices, but the automated
measuring systems available are numerous. One can
515
simply sense when the water gets to a desired level
(binary output) or decide to measure the total amount
of water in the tank/system (range output, usually 4-20
mA). The most advanced sensors for measuring fluid
level are ultrasonic, radar or laser-based sensors.
520
9
Preprint submitted to Journal of Cleaner Production December 12, 2019
Water level is the third most researched parameter
in existing literature. Zamora-Izquierdo et al., used a
Omron K8AK-LS1 water level controller with a
maximum temperature tolerance of 50 °C (Zamora-
Izquierdo et al., 2019). Mehra et al., used an analog
525
output water level sensor connected to an Arduino
controller (Mehra et al., 2018). Wang et al., followed
the same approach (Wang et al., 2015). Mamatha and
Namratha, used an array of sensors to determine the
water level in tanks(Mamatha and Namratha, 2017).
530
They used three probes at different levels to know
when actions were needed to keep the water level as
required, e.g. start pumping water when water level
was low. Jacob, used a BC546 NPN transistor circuit
to make an overflow level sensor in the water tank
535
(Jacob, 2017). Kyaw and Ng, and Sreelekshmi and
Madhusoodanan, used an ultrasonic level sensor to
control the water tank levels (Kyaw and Ng, 2017;
Sreelekshmi and Madhusoodanan, 2018).
3.1.5 Dissolved Oxygen
540
Dissolved oxygen (DO) is a measure of how much
oxygen is dissolve in the water available for the
aquatic living organisms. The amount of Do in the
water is an important parameter for the three
organisms (fish, bacteria and plants) that share the
545
aquaponics environment. Alongside the water level,
the amount of oxygen in water determines the ability
to support aquatic life (Sallenave, 2012). Oxygen is
dissolved in the water at very low concentrations (in
parts per million or ppm) and has been reported to be
550
the parameter that has most immediate and drastic
effects on aquaponics (Somerville et al., 2014).
In natural environments, oxygen is produced by
photosynthesis in aquatic green plants and algae. It is
of high importance to monitor the dissolved oxygen in
555
any aquaponics system because its level varies
dramatically in short periods of time (24 hours)
(Sallenave, 2012).
There is a strong relationship between temperature
and the DO, as warmer water can hold less oxygen.
560
When fishes are eating, the consumption of DO
increases and it needs to be compensated in some
cases. Further, the nitrification process is an oxidative
reaction; thus, it depends on the existing dissolved
oxygen to happen. When the levels of oxygen are low,
565
the bacteria will stop to break down the ammonia and
nitrite, increasing potential health’s risks for fishes and
plants. Optimum levels for nitrifying bacteria’s go
from 4 to 8 mg/L (Somerville et al., 2014).
Plants use their leaves to absorb oxygen during
570
respiration, but still need to absorb oxygen through
their roots (Somerville et al., 2014). For the
hydroponics component, plants need high levels of
DO, typically >3 mg/L(Somerville et al., 2014). When
oxygen is low, plants’ roots start to die, and some
575
fungus may appear. In the aquaculture component,
most of the fish species require a DO concentration >5
mg/L (Bhatnagar and Singh, 2010). In cases of low
concentration, fish production of TAN will diminish
(Bhatnagar and Devi, 2013).
580
In small sized systems, dissolved oxygen is
expensive to measure since there are no low-cost
methods available. In this case, manual visual
inspection of the fish is the most common approach,
i.e. red zones around the eyes or fish swimming close
585
to the surface are indicators that DO levels are low.
When using DO sensors, it is necessary to be aware
that DO measurements are affected by temperature,
pressure, salinity, and some compensations may be
needed. The two methods available to measure DO
590
concentration are optical and electromechanical, if
excluding laboratory-based methods such as
colorimetric approaches and Winkler titration.
The optical sensors measure the interaction
between the oxygen and certain luminescent dyes.
595
When DO is present, the returned wavelengths are
limited or altered due to oxygen molecules interacting
with the dye (Fondriest Environmental, 2014). The
existent electromechanical option to measure the
dissolved oxygen concentration can be galvanic and
600
polarographic. Inducing voltage to polarize (or not)
the system, the presence of DO is measured by the
change in the electrical signal.
DO concentration measurement systems are
expensive, and as such, few publications that target
605
such parameter exist in the literature. Odema et al.,
connected a DO sensor to a Modbus and used TCP/IP
technology to transfer the data (Odema et al., 2018).
Mandap et al., successfully used an Atlas DO probe
with a capacity range of 0-100 mg/L, a maximum
610
pressure of 3,447 kPa and maximum depth of 343
meters, in an aquaponics system (Mandap et al., 2018).
3.1.6 Electro-Conductivity
The electro-conductivity (EC) is a measurement of
the ability of a medium to conduct electric current, and
615
in the case of aquaponics systems, it is highly
correlated to salinity (amount of salt concentration in
the water) (Manju et al., 2017; Nagayo et al., 2017).
Therefore, fish population is the most sensible to
changes in the EC. It also relates to how fresh the water
620
is and low levels could indicate unbalanced systems
(Nagayo et al., 2017). High levels of EC indicate that
water is polluted, and it may cause death of the fish
population.
A minimum salt content is desirable though to help
625
fish maintain their osmotic balance (Stone and
Thomforde, 2004). The optimum range for fishes is
100-2000 µS/cm, but a wider range has been found
10
Preprint submitted to Journal of Cleaner Production December 12, 2019
acceptable (30-5000 µS/cm) (Stone and Thomforde,
2004). A measurement of the electro-conductivity in
630
the hydroponic component can be used as an estimator
of the water nutrient content if added to a pH
measurement. However, such measurement would not
be able to differentiate between all the different
nutrients.
635
Day by day, measurements of water EC may
provide insight on nutrient consumption. Thus, it may
help maintain consistency with each crop cycle, and
ensure maximization of nutrients use without over-
fertilizing (“Electrical Conductivity and hydroponic
640
gardening,” 2015). A method to control the nutrient
solution in the hydroponics system through EC
monitoring was proposed (Wada, 2019). For example,
Enshi-shoho nutrient solution was used to provide
control over EC, as it has a known EC of 2400 µS/cm.
645
The electro-conductivity meters usually employ a
potentiometric method and four platinum electrodes.
Some current is applied to the outer pair and the
potential between the inner pair is then measured.
Nagayo et al., suggested the use of this parameter in
650
their work (Nagayo et al., 2017).
3.1.7 Total Dissolved Solids
Dissolved solids are naturally present in water.
TDS levels represent the content of inorganic salts,
organic matter and other dissolved materials in water
655
(Weber-Scan and Duffy, 2007). Typically an optimum
amount of TDS in water for fish life is inferior to 1000
mg/L, although values below 2500 mg/L have been
found acceptable (Stone and Thomforde, 2004). High
amounts (>1000 mg/L) of TDS can cause a toxic
660
medium for most fish species.
As sensing units, TDS meters are commercially
available. Usually used to measure TDS in potable
water, TDS meters are similar to EC meters. In fact,
the same sensor used to measure EC can be used.
665
3.1.8 Salinity
The salinity (SL) indicates the amount of salt
concentration in the water (Nagayo et al., 2017) and is
a driving factor that affects the density and growth of
the fishes (Bhatnagar and Devi, 2013). Salinity is often
670
derived by the electro-conductivity measurement, just
like TDS. The desirable range of SL varies with each
fish species. The most common adoption was given by
(Garg and Bhatnagar, 1996 for the common carp and
its range goes from 0 to 2 ppt (part per trillion).
675
3.1.9 Water hardness
Water hardness is a measure of the concentration of
existing positively charged calcium and magnesium
salts in water solution. Calcium and magnesium are
essential to fish metabolic reaction, namely bone and
680
scale formation, thus, relevant to fish growth
(Bhatnagar and Devi, 2013). Whereas low levels of
water hardness only cause stress in fishes, high levels
could be lethal since it increases water pH, resulting in
a low nitrification and nutrients absorption rate for
685
plants. The desirable range for water hardness goes
from 50-150 mg/L, but >10 mg/L is acceptable for
most species (Stone and Thomforde, 2004).
Water test kits can be used to manually measuring
water hardness, relying in test tablets or paper test
690
strips. However, water hardness is usually determined
qualitatively by the TDS or EC measurements (Brand,
2011). Some colorimeters or spectrophotometer
sensors are used when lower measurements than
4mg/L CaCO3 are expected.
695
3.1.10 Alkalinity
The alkalinity is a measure of the concentration of
bases, typically carbonate and bicarbonate in
aquaponics systems. Water hardness and alkalinity are
often confused as alkalinity measures the negative ions
700
(carbonate and bicarbonate) and hardness the positive
ions (calcium and magnesium). Alkalinity is usually
referred as the water ability to resist changes in pH or
the capacity to neutralize acids. Low levels indicates
that even small amounts of acids can cause large
705
change in the pH (Bhatnagar and Devi, 2013), high
levels of alkalinity cause non-toxic ammonia to
become toxic. Desirable ranges of alkalinity go from
50 to 150 mg/L CaCO3 (Stone and Thomforde, 2004).
3.1.11 Flow
710
Water flow through the aquaponics system is
extremely important to estimate the capacity of
filtration (solids) and bio-filtration (nitrification), as
well as to determine the nutrients availability for the
plants. It is recommendable that the flow in the system
715
is maintained constant to avoid stress in the fishes and
to avoid nutrient deficiencies in the plants. Most
importantly, it is needed to measure the flow between
the filters and the grow bed.
The flow rate will vary depending in the
720
hydroponic system adopted. In NFT-based systems,
flowing water in the channels ensures that the roots
receive large amounts of oxygen and nutrition. The
recommended water flow for NFT should be lower
than 1-2 L/min (Somerville et al., 2014). In the media-
725
based technique, a siphon is used to filtrate the water
through the media. The recommendation is to set the
flow rate to be able to filter the entire water fish tanks
every hour through the grow beds. In the DWC-based
systems, the water flow is mostly due to gravity. As
730
such, water needs to flow during approximatively 1 to
4 hours through the channels to guarantee adequate
replenishment of nutrients. The growth of the plants in
DWC systems benefit from high flow rates and
turbulent water because plants’ roots absorb more
735
nutrients.
11
Preprint submitted to Journal of Cleaner Production December 12, 2019
Optimal water flow is calculated from channel size
and water capacity. Once the requirements are set, a
flowmeter will be useful to guarantee water flow
throughout the system and enable detection of major
740
problems, i.e. obstructions in the piping system. A
flowmeter is a device capable of measuring the
amount of water that is passing through a pipe. It exists
four types of different flowmeters: 1) mechanical; 2)
vortex; 3) ultrasonic; and 4) magnetic. Murad et al.,
745
use a water sensor to detect the water flow into the fish
tank through the siphon outlet in a media grow bed
(Murad et al., 2017), and Kyaw and Ng, put a
flowmeter between the fish tank and the grow beds
(Kyaw and Ng, 2017).
750
3.2 Environment
The parameter related to the air conditions in
contact with the plants are analyzed in this section. To
obtain and maintain balanced and safe optimal crops
in the system and to ensure the stable and healthy
755
growth of fish and vegetables it is necessary to monitor
and control some environmental parameters(Wei et
al., 2019). Namely, the air temperature, where the
ranges changes between the plant species; the light
intensity, which depends on the growth stage of the
760
plant; the air humidity, the air content of CO2 (carbon
dioxide), and the media moisture, in case of adopting
a media base as the hydroponic component.
3.2.1 Air Temperature
The air temperature influences the health of plants.
765
The suitable temperature for most of the vegetables
commonly grown in aquaponics systems is 18-30 °C.
At higher temperatures, leafy greens bolt and begin to
flower and seed (Somerville et al., 2014). Further, the
air temperature is responsible of a correct transpiration
770
of the crops.
The factors that have taken into consideration when
selecting air temperature sensors are: 1) temperature
range; 2) contact or contactless; 3) sensing element;
and 4) calibration method. For measuring the air
775
temperature in aquaponics systems, a thermistor has
been the recurrent option as the air temperature and
humidity are measured together. Sreelekshmi and
Madhusoodanan, used a DTH11 thermistor
(Sreelekshmi and Madhusoodanan, 2018).
780
Vernandhes et al., used a DHT22 thermistor, which is
a more accurate sensor and has a larger range of
temperature values than the DTH11 (Vernandhes et
al., 2017).
3.2.2 Relative Humidity
785
The relative humidity is an expression for moisture
in the air. Most of the grown crops in aquaponics
systems need humid air for thriving, thus the relative
humidity needs to be well managed. The humidity in
the air can be measured in different ways: 1) mass of
790
water in unit volume of air; 2) unit mass of air; or 3)
partial pressure of water vapor in the air (Bakker,
1991). Air humidity can be also expressed as a
proportion of the air-water saturation or relative
humidity (RH).
795
Warm air has a higher moisture-holding capacity
than cooler air. To maintain control of the level of
humidity, a good control over ventilation and heating
systems is necessary. The ventilation system allows
the exchange of the moisture inside the greenhouse
800
with drier air from outside the facilities. The heating
devices are necessary to warm up the outdoor air to the
optimum growing temperature and increase the
capacity of the air to hold moisture. To increase
moisture on indoor air, it is common to use
805
moisturizers.
The optimum level of RH varies depending in the
type of crop and the growth stage of the plants. The
most common considerations are 50%-80%, but those
depend on the indoor temperature. An excess of RH in
810
the air would interfere with the plants’ transpiration
and prevent roots and stems to supply an adequate
quantity water to the leaves.
As mentioned previously, air RH is usually
provided by the air temperature sensor. However,
815
individual sensors can be used: Wang et al., used a
DTH11 capacity humidity sensor to measure RH
values in an aquaponics systems (Wang et al., 2015).
3.2.3 CO2
The carbon dioxide is an essential component of the
820
photosynthesis, vital chemical reaction for plants
sustain. In mass production indoor systems, it is
possible that plants use all the CO2 available in the air.
As such, artificial addition of CO2 and control of its
levels is necessary.
825
The optimum range level for most of indoor crops
is within 340-1300 ppm (Blom et al., 2002). However,
a smaller range may be needed for some crops. For
example, 800-1000 ppm of CO2 in air is necessary to
grow tomatoes, cucumbers, peppers, or even lettuces.
830
Depending on the type of crop, the lighting conditions,
air temperature and RH, different CO2 air
concentrations are needed.
Elevated CO2 also leads to changes in the chemical
composition of plant tissue. The extra carbon
835
molecules may be dissolved in the systems’ water,
forming carbonic acid, HCO3-. This would provoke a
decrease in the water pH. Recommended levels of
carbonic acid must be less than 5 mg/L to enable
proper fish growth (Bhatnagar and Devi, 2013;
840
Santosh and Singh, 2007). More than the indicated
level is toxic for aquatic life.
12
Preprint submitted to Journal of Cleaner Production December 12, 2019
The commercially available sensors that measure
CO2 concentration in the air are infrared gas sensors
(NDIR) or chemical gas sensors. Nagayo et al., used a
845
MG811 sensor to measure the CO2 content in the air
of their proposed system (Nagayo et al., 2017).
3.2.4 Media Moisture
Media moisture is the soil water content in the
media base. This measurement is only necessary when
850
is used the media-base type in the hydroponic
component. For this type of hydroponic component, it
is good practice to implement a moisture-soil sensor to
guarantee that the media has correct amount of water
for the plants. More plants die due oversaturated root
855
than of drought. It is highly recommendable to check
for the water holding capacity of the soil selected in
order to set the parameters for the sensor system.
Werner did an analysis for traditional farming using
different types of sands, loams, clays and
860
combinations. After sensing using a tensiometer (0-
100 centibars), he found that the optimum ranges vary
from type to type of soil, going from 30-60 centibars
(Werner, 2002). Traditional meters are electrical
resistance blocks, tensiometer and time domain
865
reflectometry (TDR). Another soil-moisture sensor
works under capacitance and measure the dielectric
permittivity of the water in the medium and is the
sensor commonly adopted in Hydroponics due the
simplicity. In this case, Vernandhes et al., used a FC-
870
28 as a soil-moisture sensor. Since various materials
for the media base are used and their levels are not
fixed, no generic recommendation can be provided for
this parameter (Vernandhes et al., 2017).
Nevertheless, interested people can find the
875
appropriate levels after selecting the media base and
crops to farm.
3.2.5 Light Intensity
Sunlight is critical for plants but is unavailable or
limited in indoor facilities. Artificial lighting is placed
880
in aquaponics systems as a substitute to provide light
to the plants. Light is usually measured in terms of its
intensity (lux). However, plants use a limited part of
the light spectrum called photosynthetically active
radiation (PAR). It designates the spectral range of
885
solar radiation that photosynthetic organisms can
process.
Most of the plants do not require PAR regulations
and grow independently of lighting conditions, but
some light sensitive crops, i.e. lettuce, salad greens,
890
and cabbages, can bolt, seed and become bitter and
unpalatable with high levels of PAR (Somerville et al.,
2014). Additionally, with low light intensity, the
growing rate of plants is greatly diminished. Contrary
to plants, water does not need direct light radiation and
895
it is paramount to isolate any water system to help
maintain the water temperature and prevent algae
growth. Further, the nitrifying bacteria are
photosensitive organisms during the initial formation
of the bacteria colonies. For new aquaponics systems,
900
it is recommendable to cover the area from direct light
for the first 3-5 days.
Different equipment can be used to obtain PAR
measurements; however, research has shown that
using a photosynthetic photon flux (PPF) and yield
905
photon flux (YPF) specific PAR meter is the most
accurate measuring sensor for narrow spectrum
radiation sources, such as artificial lights (Barnes et
al., 1993). In efficient light systems, a balance between
correct PAR usually given in photosynthetic photon
910
flux density (PPFD) and the right light intensity (lux
or lumens) must be reached. In general, crops need
between 14 to18 hours of light per day. The amount of
PPFD that plants need vary from its growth stage and
type of plant, but an average optimal range of 600-900
915
PPFD is required (“Why is PAR Rating a Big Deal for
Indoor Grow Light Systems?,” n.d.). To measure the
radiation intensity of a lighting system, a light
dependent resistor (LDR) can be used. From such
measurement, PPFD can be estimated. Both, Mamatha
920
and Namratha, and Sreelekshmi and Madhusoodanan,
used an LDR to successfully measure the ambient light
intensity (Mamatha and Namratha, 2017; Sreelekshmi
and Madhusoodanan, 2018).
4. Smart Systems
925
Intelligent or smart systems is a broad concept in
the academic world with the purpose of optimizing
production by making use of cutting-edge information,
communication and computing technologies (Ahmad
et al., 2017). The opportunity to use smart systems in
930
industrial applications came from the research
developments in artificial intelligence (AI) in general.
Some misconceptions have been made in this research
area when naming proposed systems as “smart”.
Usually, researchers tend to label as smart a system
935
that is just automated or wireless. A machine that work
under input signals, comparison between signals and
ranges, triggers and output cannot be called smart and
is just an automated system. Smart is a concept more
related to the Industry 4.0 itself and involves complex
940
logical process, algorithms and it is not limited to basic
logical operators. The adoption of the smart systems in
the farming is going towards the concept of precision
farming, which looks to apply only the water and
nutrients that plants need (Pinstrup-Andersen, 2018).
945
This concept was analyzed and presents a dominance
of heuristic approaches over the quantitative working
13
Preprint submitted to Journal of Cleaner Production December 12, 2019
methods when applying tools from the Industry 4.0
(Braun et al., 2018).
König et al., presented a review of aquaponics as
950
an emerging technological innovation system where
changing the food production technologies themselves
was proposed as one way of creating more sustainable
food systems (König et al., 2018). The immersion of
smart techniques in aquaponics is helping to minimize
955
production times, reducing the need of labor, lowering
the expertise need it to regulate the systems and
enhance the quality of the products. When managing a
farming system, the adoption of Cyber-Physical
systems is increasing. These systems are presented as
960
collaboration levels between self-configuration
(machines), local (analysis of the production system)
and extended (collaboration between different actors
as clients, farmers, etc.). The usage of advanced
learning techniques (machine learning), can support
965
and expand this concept (Braun et al., 2018) .
4.1 Parameters prediction
In all the literature reviewed, only one smart
application was found for aquaponics systems. Kumar
et al., developed an autonomous wireless aquaponics
970
system. The smart component of the system relies in
the application of regression techniques to predict
future values for some of the sensed parameters
(nitrate and pH) and make smart decisions with the
outputs (Kumar et al., 2016).
975
Smart applications in hydroponics, in the other
hand, have more contributors and developments using
deep neural networks, predictions, decision making
have been made. Mehra et al., trained a deep neural
network to predict pH, humidity, light intensity,
980
temperature, and the water level in hydroponic tanks’
sensors outputs. Then, this trained neural network was
installed in a Raspberry Pi to control the outputs
depending on the sensed values (Mehra et al., 2018).
Pitakphongmetha et al., used an artificial neural
985
network with the pH, electro conductivity,
temperature, humidity, light intensity, and plant age,
as inputs to predict the pH and electrical conductivity.
Then, the error between predicted values and sensor
outputs was used to monitor and control the
990
parameters (Pitakphongmetha et al., 2016).
4.2 Quality and growth rate
The use of Convolutional Neural Networks is
commonly used in quality assessments of the crops. A
monitoring growth rate of lettuce using deep
995
convolutional neural networks was implemented in a
hydroponics system by Lu et al. (Lu et al., 2019).
Moving forward, this image prediction models can be
used to monitor some parameters in the aquaculture
component, e.g. the health of the fishes based in the
1000
known physical reactions of some parameters, (i.e. red
areas in the eyes when the level of ammonia is
dangerous), turbidity of the water for triggering the
filter or cleaning, etc. The image processing and
prediction techniques based on images is not found
1005
often in the literature of Aquaponics.
5. IoT systems
Internet of Things (IoT) is looking to dismiss the
bridge of connectivity issues between systems. One of
its main objectives is to make industrial machinery
1010
capable of communicating between each other and
provide a framework where data-driven decisions can
be taken without human intervention. The
enhancement of the network capacity, with the 4G and
5G technologies, increases the feasibility of IoT
1015
implementations and leads to the creation of new
communication hardware, protocols, and frameworks.
The communication between devices and interfaces is
currently less limited, increasing the flexibility,
interoperability, and integration of complex
1020
communication systems into complex industrial
scenarios such as aquaponics. In farming, IoT
technologies are being implemented with very
different objectives, including: improve GPS systems,
weather predictions, inventories, producer-consumer
1025
information, and so forth. In our review, 71% percent
of the publications used an IoT technique in their
proposed system. The findings are categorized in three
different sections: 1) monitoring interfaces, 2) remote
applications and 3) wireless technologies. In some
1030
cases, combinations of these categories were found.
5.1 Remote monitoring interfaces
Monitoring interfaces are commonly an
environment (interactive or not) that displays some of
the interested parameters in the process to the user or
1035
stakeholder. This visualization process is key to final
decision making. IoT technology enable these
monitoring interfaces to display values through
wireless networks, even in real-time. Manju et al.,
designed a web application that showcased a
1040
dashboard connected to a microcontroller to monitor
selected aquaponics’ parameters (Manju et al., 2017).
In the same year, Dutta et al., connected a Raspberry
Pi to all the system measurement units, then the
sensors’ data are sent to a web-based platform where
1045
it is stored and displayed (Dutta et al., 2018). A year
later, Elsokah and Sakah, programmed an iOS
application that allowed to monitor the system
environment continuously by obtaining data directly
from the systems’ microcontrollers (Elsokah and
1050
Sakah, 2019). The direction of these collaborations is
heading towards the real time reliability and mobility
14
Preprint submitted to Journal of Cleaner Production December 12, 2019
(not only web based but also application for mobile
devices)
5.2 Remote control applications
1055
Remote control applications are defined based on
their capability to signal system actuators to interact or
change certain parameter. Such applications are a step
forward from only monitoring the system, as presented
in the previous subsection. For example, with remote
1060
control applications (web-based or applications),
operators can turn on/off water pumps or lights when
necessary, change values of critical timers to modify
the plants’ growth process, and so forth.
From the reviewed papers, Nagayo et al.,
1065
implemented a GSM Arduino-based monitoring and
control system that can send alert messages to
operators when measurements are outside specific
ranges. A graphical user interfaces is designed to
display the information and data could be extracted
1070
from the system (Nagayo et al., 2017). The
collaboration of Pitakphongmetha et al., was using
Blynk, a multi-language platform that enables remote
control of different microcontrollers (i.e. Arduino,
Raspberry Pi) (Pitakphongmetha et al., 2016). The
1075
next year, Aishwarya et al., integrated a GSM receptor
with a microcontroller in an aquaponics system. As
such, operators can send messages to the receptor so
that real-time control over the water supply or
temperature is achieved(Aishwarya et al., 2018).
1080
Vernandhes et al., used an Arduino connected to a web
server through an Ethernet Shield. A user interface
was created to allow real-time monitoring and control
of the water-related sensor measurements, i.e. switch
on or off the exhaust, pumps, and mist makers
1085
(Vernandhes et al., 2017). Odema et al., created an
IoT-based aquaponics system that allows remote
monitoring and control of the system parameters. The
authors used a Modbus TCP standard protocol to pull
measurement data from the sensing nodes of a
1090
supervisory computer (Odema et al., 2018). Haryanto
et al., designed a system with a microcontroller
connected to an Ubuntu IoT Cloud. The system could
be accessed to monitor and control the parameters
automatically based on the sensed inputs(Haryanto et
1095
al., 2019). The authors in this section added the
controlling parameter into the scenario. At this time,
the visualization of the parameters in the system is not
enough and is necessary to control such parameters for
a better system.
1100
5.3 Wireless technologies
The wireless technologies are rarely presented
alone and are mostly linked to the two past sections
(remote monitoring or control interfaces).
Nevertheless, was found that some contributors were
1105
focus in develop/apply some wireless technologies
into Aquaponics to improve the connectivity. Wang et
al., designed an architecture to monitor and control an
aquaponics system with Arduino and sensors
information. Data is efficiently stored on WRT nodes
1110
and transmitted to an OpenWrt server using a Wi-Fi
data transmission module(Wang et al., 2015). Kumar
et al., designed an aquaponics system using the
6LOWPAN protocol and a wireless sensor network
(WSN)(Kumar et al., 2016). Murad et al., used GSM
1115
technology to send notifications if pH and temperature
values go out of range (Murad et al., 2017). Mamatha
and Namratha, used a data logging platform,
ThingSpeak, to store all the information from an
aquaponics(Mamatha and Namratha, 2017).
1120
Sreelekshmi and Madhusoodanan, developed a
web-based monitoring system using ThingSpeak IoT
platform with Arduino Uno and an ESP8266-01 Wi-Fi
transceiver (Sreelekshmi and Madhusoodanan, 2018).
Jacob, used a Raspberry Pi along with a Wi-Fi dongle
1125
to give internet connectivity to the system. The system
uses cloud-based platforms (Pubnub, Cloudinary, and
Dweet) to store and control the diverse parameters of
the aquaponics system, i.e. motors and lights, with an
IoT dashboard using Freeboard (Jacob, 2017). The use
1130
of wireless technologies in sensors or a transmission
of data opens the door to improvements in the
monitoring and control of parameters.
15
Preprint submitted to Journal of Cleaner Production December 12, 2019
Table 2. Parameter ranges and potential effects in
1135
aquaponics systems.
Parameter
Aquaculture
Range
Nitrification
Hydroponic
Range
Low Level Effect
High Level Effect
References
pH
6.5-9.5
7.0-9.0
4.5-7.0
Fish reproduction rate
is diminished. Root
injury and plants
experience nutrient
deficiencies.
Plants experience
nutrient deficiencies.
(Rakocy, 2012) (Wada,
2019) (Stone and
Thomforde, 2004)
Water T
5-32 °C *
17-34 °C
18-30 °C
Increase the risk of
diseases in fishes.
Increases the risk of
diseases in fishes.
(Somerville et al.,
2014)
Water Level
1000 L per 20 kg of
fish
-
-
Fish stress leading to
grow health issues.
Plants experience
nutrient deficiencies.
(Somerville et al.,
2014)
Dissolved
Oxygen
4-5 mg/L
4-8 mg/L
>3 mg/L
Plant roots may die,
and some fungus can
start to grow. Fish stop
to eat. Bacteria will
stop nitrification
process.
-
(Somerville et al.,
2014) (Bhatnagar and
Singh, 2010)
Electro-
Conductivity
100-2000 µS/cm
-
-
Loss of nutrients in the
water. Indicates
unbalanced systems.
High levels of EC
indicate that water is
pollute and may cause
death of the fish
population.
(Stone and Thomforde,
2004)
Total Dissolved
Solids
<1000 mg/L
-
-
-
Toxic for most aquatic
life, especially fish.
(Stone and Thomforde,
2004)
Salinity
0-2 ppt
-
-
-
Affects the density and
growth of the fishes.
(Garg and Bhatnagar,
1996)
Water Hardness
50-150 mg/l CaCO3
-
Fish stress.
Increase of pH , resulting
in a low nitrification and
nutrients absorption rate
for plants
(Stone and Thomforde,
2004)
Alkalinity
50-150 mg/L CaCO3
-
-
Poor status of water
body. Low ability to
neutralize acids, risk of
high ph.
Cause non-toxic
ammonia to become
toxic. Fish stop
breathing.
(Stone and Thomforde,
2004)
Total Ammonia-
Nitrogen
0-2 mg/ L
<3 mg/ L
<30 mg/ L
-
Highly toxic for fish.
(Somerville et al.,
2014) (Stone and
Thomforde, 2004)
Nitrites
0-1 mg/ L
0-1 mg/ L
0-1 mg/ L
-
Highly toxic for fish,
plants and bacterial
activity.
(Stone and Thomforde,
2004)
Nitrates
50-100 ppm
-
-
Nutrient deficiencies in
plants
Toxic for fishes.
(Bhatnagar and Devi,
2013)
Flow
-
-
1-2 L/min*
Low availability of
nutrients.
Low availability of
nutrients
(Somerville et al.,
2014)
Air T
-
-
18-30 °C
Incorrect transpiration
of the crops.
Leafy greens bolt and
begin to flower and seed.
Increases transpiration of
the crops. Reduces
efficiency of water
supply to the plants.
(Somerville et al.,
2014)
Relative
Humidity
-
-
50%-80 %
Curled leaves and dry
leaf.
Inadequate supply of
water to plants. Causes
mold and fungus growth.
N/A
CO2
-
-
340 ppm-1300
ppm
Decrease in plants
photosynthesis.
Changes in the chemical
composition of plant
tissue.
(Bhatnagar and Devi,
2013) (Blom et al.,
2002)(Santosh and
Singh, 2007)
Bed Moisture
-
-
-
Not enough nutrient
availability in plants.
Drought.
Plants will start to wilt,
and roots start to dye due
a lack of oxygen.
N/A
Light Intensity
-
-
600-900 PPFD
Decrease in plants’
photosynthesis.
Carbon limitation in
plants.
(“Why is PAR Rating a
Big Deal for Indoor
Grow Light Systems?,”
n.d.)
16
Preprint submitted to Journal of Cleaner Production December 12, 2019
6. Discussion
6.1 Overview
1140
The success of an aquaponics system relies on a
correct management and implementation of sensors,
IoT techniques and smart systems. This paper aims to
summarize through bibliometric analysis the
necessary and proposed solutions available in the
1145
literature in an effort to support commercial
availability. Moreover, this paper targets to ease the
introduction of automation, smart technologies and
IoT in aquaponics systems by simplifying the selection
of sensors based on biological needs.
1150
An automated system had proven in other mature
areas (i.e. automotive, manufacturing, construction
etc.) to increase the productivity, reduce the human
error and reduce time and amount of labor.
Extrapolating it to Aquaponics will lead to fulfill the
1155
concept of precision farming with the inherent
attributes such as improving resources utilization
(water, electricity, fertilizers) reducing human
intervention, reduce field expertise and even
accelerating the grow time of most of the crops since
1160
the ideal environment can be automatically
maintained.
Introducing sensors is a mandatory step in working
towards the fully or semi-automation of this systems.
Here the importance of having a guide for those
1165
planning to dive into the automation of aquaponics
systems. Table 3 summarizes the parameter ranges;
optimized to avoid potential threats to the aquaponic
system. Figure 4 shows the direct correspondence
found in the literature between sensors and the
1170
aquaponic parameters.
Table. 3. Optimal parameters range for aquaponics
systems.
Parameters
Aquaponic
pH
6.5-7.0
Water T
17°C -30°C
Water Level
.02 kg/L
Dissolved Oxygen
>4 mg/L
Electro-Conductivity
100-2000 µSiemens/cm
Total Dissolved Solids
<1000 mg/L
Salinity
0-2 ppt
Water Hardness
50-150 mg/L CaCO3
Alkalinity
50-150 mg/L CaCO3
Total Ammonia-Nitrogen
<2 mg/L
Nitrites
<1 mg/L
Nitrates
50ppm-100 ppm
Flow
1-2 liters/min *
Air T
18°C -30°C
Relative Humidity
60%-80%
CO2
340 ppm-1300 ppm
Light Intensity
600 PPFD -900 PPFD
6.2 Control Strategies
1175
The analysis of the mentioned literature showcased a
consistent use of micro-controllers, such as Raspberry
Pi and Arduino, in aquaponic systems. Overall, three
different levels of control strategies were noted.
The most basic control strategy can be understood as
1180
a local approach with external communication. Wang
et al. utilized an Arduino and a WRTnod to monitor
the data acquisition and manage the aquaponic system
built (Wang et al., 2015). The sensor acquisition
module built consisted in temperature, humidity, light,
1185
water level, and DO sensors. The data was sent
wirelessly to the control and management center,
whose function was to store the data, process it, and
send it to a remote server. Once the information was
stored in the server, users were able then to analyze the
1190
information and make decisions in regard to the state
of the air and water pumps, lights, and so forth.
A step further implies to gather and analyze
information wirelessly through cloud servers. Kumar
Figure 4. List of all sensors and parameters found in
literature and their correspondence.
17
Preprint submitted to Journal of Cleaner Production December 12, 2019
et al. included in their system wireless sensor network
1195
devices (temperature, pH and nitrate sensors) (Kumar
et al., 2016). The network had a 10-meter
communications range with a transfer rate of 250
Kbits/s. The authors used this time the run-time
platform IBM Mote Runner
1
as the sensor network. A
1200
cloud data storage system was used to store the data
from the sensors, then, trend analysis helped to predict
the next time series values of the variables. A
regression technique was implemented to make
predictions about nitrate and pH values, aiming to
1205
create an autonomous aquaponic system regarding the
control of these two parameters.
Finally complex control strategies found in literature
aimed to achieve autonomous systems through a
variety of smart techniques that go from linear
1210
regression to more complex prediction models, such as
neural networks. Murad et al. differed from this two
strategies and kept the automation and deployment
locally (Murad et al., 2017). Several sensors were used
and controlled with Arduino. The system was locally
1215
controlled and connected to a GSM (Global System
for Mobile) communication interface, capable to send
notifications or alarms as predefined actions based on
sensors levels.
However, no discrete or event-triggered control
1220
strategies were found during the analysis as authors
tend more to apply distributed control systems (DCS).
Those missing strategies are key to the implementation
of Programmable Logic Controllers (PLCs) in such
environments and would require further research.
1225
6.3 Current Limitations and Future Work
Some limitations were encountered in the current
body of knowledge. Based on Table 1, Figure 5
displays the concurrence of the parameters in the
reviewed literature.
1230
It is important to note that while some parameters
have been thoroughly researched, i.e. pH or water
temperature, some other parameters are being
neglected by the academic community. With varying
degrees of effect in the aquaponics system, research is
1235
needed to provide practitioners of a clear impact of
each of the listed parameters. An effort is then required
to draw precedent in certain parameters, while keeping
improving past contributions.
1
www.zurich.ibm.com/moterunner/
Regarding IoT systems, their implementation has
1240
influenced the success of other industries, i.e.
automotive or aerospace, thus working in the use of
those technologies in aquaponics to aim towards
precision farming seems to be an accurate solution for
the feasibility of such farming systems. Nonetheless,
1245
the available solutions are still primitive, with a widely
spread use of micro-controllers and commercially
available software that would limit its industrial
application, economically speaking.
When moving to commercial-industrial levels, the
1250
adoption of the adequate equipment becomes
mandatory. Migration towards controllers and sensing
equipment that are more robust, such as a
Programmable Logic Controller (PLC), is necessary to
promote automatic aquaponic solutions to commercial
1255
production. Nowdays, monitoring off-the-shelve
equipment can be purchased, capable of giving real-
time data, namely parameters such as temperature, pH,
electroconductivity (EC) and relative humidity (RH)
currently used to improve yield performance and
1260
quality of crops in greenhouses. Nevertheless, these
systems cannot deal with closed-loop feedback
systems that may, potentially, control external
variables and/or hardware, which are necessary to
push aquaponics towards smart technologies. On the
1265
same page, some monitoring systems are comercially
available for correspondent systems in aquaculture.
These are capable to measure parameters such as
temperature, dissolved oxygen (DO), and pH, with the
same limitations aforementioned. When planning
1270
aquaponics’ control systems, high flexibility is
needed. Since the complexity and interaction between
parameters, as well as the cross-incidence (i.e.
variations in temperature levels yield variations to pH
levels), is high and sometimes difficult to predict or
1275
still unknown, it is then necessary to have a control
system flexible enough to control and monitor the
variety of sensors and equipment shared (and not)
between this two technologies (aquaculture and
hydroponics).
1280
The introduction of advanced industrial control
systems, such as PLCs, together with wireless
networks and data servers in aquaponics can highly
impact the development of this industry. PLC offers
high flexibility when dealing with variety of sensors,
1285
motors, pumps, and other hardware needed in the
aquaponics industry.
At last, water is the most important and complex
parameter in an aquaponic system. Aquaculture alone
18
Preprint submitted to Journal of Cleaner Production December 12, 2019
have been one of the fastest growing food production
1290
industry, with an average rate of 8.5 % annually over
the last 30 years. Promoting the adoption of the
hydroponics technology as a business opportunity to
the aquaculture sector will exponentially expand the
aquaponics impact. To achieve this, it is necessary to
1295
enhance the research of the water treatment strategies
towards a feasible and easy to adopt business model.
Even though some research efforts have been made in
this area that can contribute to the popularization of
aquaponics, further research is required to provide a
1300
real stimulus towards the commercialization of such
systems. For example, Lie et al. reported the
performance of a immobilized biofilm treatment in
aquaponics pilot scale (Li et al., 2019) . Boxman et al.
recently evaluated the water treatment capacity,
1305
nutrient cycling, and biomass production of a marine
aquaponic system (Boxman et al., 2018) and Calone et
al. investigated the implications of water management
through a series of experimentations in aquaponics
systems (Calone et al., 2019) However, even in this
1310
research area, it is still necessary to move towards the
goal of adopting smart technologies. Currently, no
smart system has been designed towards the quality of
water in such conditions, and the development of a
smart system that involves water within the
1315
aquaculture, bio filtration, and hydroponic
components is becoming necessary.
7. Conclusions
Research contributions in the topics of aquaculture
and hydroponics are increasing and attracting attention
1320
from researchers and practitioners. A systematic
analysis was presented to explore the status and global
trends of aquaponics systems, focusing on their
relevant sensing parameters, smart and IoT
technologies. This paper presents a study of the field
1325
as a whole aiming to simplify the decision making
regarding the setup of sensors in aquaponic systems
and provide a clear image of the research trends in
smart aquaponics. The final purpose of this work is to
create a bridge between biological and electrical
1330
engineering knowledge to enable aquaponic
development as a sustainable source of food. This
paper contributes by giving aquaponics experts’
technical knowledge about automation, IoT and smart
systems; and automation expert’s knowledge
1335
regarding the biological processes happening in
aquaponic systems. Creating a bridge towards scaled
up aquaponics systems will accelerate contributions in
the area and enable viability in commercial solutions.
1340
Acknowledgements
The authors acknowledge the support of this work by
the Council on Science and Technology (CONACYT)
through funding 2018-000039-01EXTF-00050 and
Transportes Pitic Scholarship. The authors would like
1345
to acknowledge NSERC (Grant No. RGPIN-2017-
04516) for funding this project.
References
Ahmad, R., Tichadou, S., Hascoet, J.-Y., 2017. A knowledge-
based intelligent decision system for production
1350
planning. The International Journal of Advanced
Manufacturing Technology 89, 1717–1729.
https://doi.org/10.1007/s00170-016-9214-z
Aishwarya, K.S., Harish, M., Prathibhashree, S., Panimozhi, K.,
2018. Survey on Automated Aquponics Based
1355
Gardening Approaches, in: 2018 Second International
Conference on Inventive Communication and
Computational Technologies (ICICCT). IEEE, pp. 1377–
1381. https://doi.org/10.1109/ICICCT.2018.8473155
Anthonisen, A.C., Loehr, R.C., Prakasam, T.B., Srinath, E.G.,
1360
1976. Inhibition of Nitrification by Ammonia and
Nitrous Acid. Journal - Water Pollution Control
Federation 48, 835–852.
Bakker, J.C., 1991. Analysis of humidity effects on growth and
production of glasshouse fruit vegetables. Agricultural
1365
University, Wageningen, The Netherlands.
Barnes, C., Tibbitts, T., Sager, J., Deitzer, G., Bubenheim, D.,
Koerner, G., Bugbee, B., 1993. Accuracy of quantum
sensors measuring yield photon flux and photosynthetic
photon flux. HortScience 28, 1197–1200.
1370
Bhatnagar, A., Devi, P., 2013. Water quality guidelines for the
management of pond fish culture. International Journal
of Environmental Sciences 3, 1980–2009.
https://doi.org/10.6088/ijes.2013030600019
Bhatnagar, A., Singh, G., 2010. Culture fisheries in village
1375
ponds: a multi-location study in Haryana, India.
Agriculture and Biology Journal of North America 1,
961–968.
https://doi.org/10.5251/abjna.2010.1.5.961.968
Blidaria, F., Grozea, A., 2011. Increasing the Economical
1380
Efficiency and Sustainability of Indoor Fish Farming by
Means of Aquaponics - Review. Scientific Papers Animal
Science and Biotechnologies 44, 1–8.
Blom, T.., Straver, W.A., Ingratta, F.J., -OMAFRA, S.K.-O.W.B.,
2002. Carbon Dioxide In Greenhouses [WWW
1385
Document]. Ministry of Agriculture, Food and Rural
Affairs. URL
http://www.omafra.gov.on.ca/english/crops/facts/00-
077.htm (accessed 10.9.19).
Boxman, S.E., Nystrom, M., Ergas, S.J., Main, K.L., Trotz, M.A.,
1390
2018. Evaluation of water treatment capacity, nutrient
cycling, and biomass production in a marine aquaponic
system. Ecological Engineering 120, 299–310.
https://doi.org/10.1016/j.ecoleng.2018.06.003
Brand, G.W.A.X., 2011. Can you determine water hardness from
1395
conductivity or total dissolved solids measurements?
[WWW Document]. URL
http://www.globalw.com/support/hardness.html
(accessed 9.17.19).
Braun, A.-T., Colangelo, E., Steckel, T., 2018. Farming in the Era
1400
of Industrie 4.0. Procedia CIRP 72, 979–984.
https://doi.org/10.1016/j.procir.2018.03.176
Calone, R., Pennisi, G., Morgenstern, R., Sanyé-Mengual, E.,
Lorleberg, W., Dapprich, P., Winkler, P., Orsini, F.,
19
Preprint submitted to Journal of Cleaner Production December 12, 2019
Gianquinto, G., 2019. Improving water management in
1405
European catfish recirculating aquaculture systems
through catfish-lettuce aquaponics. Science of The Total
Environment 687, 759–767.
https://doi.org/10.1016/j.scitotenv.2019.06.167
Dutta, A., Dahal, P., Prajapati, R., Tamang, P., Saban Kumar K.C,
1410
E., 2018. IOT Based Aquaponics Monitoring System, in:
1st KEC Conference on Engineering and Technology. pp.
164–168.
Ebeling, J.M., Timmons, M.B., Bisogni, J.J., 2006. Engineering
analysis of the stoichiometry of photoautotrophic,
1415
autotrophic, and heterotrophic removal of ammonia–
nitrogen in aquaculture systems. Aquaculture 257, 346–
358.
https://doi.org/10.1016/j.aquaculture.2006.03.019
Electrical Conductivity and hydroponic gardening [WWW
1420
Document], 2015. URL
https://fifthseasongardening.com/electrical-
conductivity-and-hydroponic-gardening
Elsokah, M.M., Sakah, M., 2019. Next Generation of Smart
Aquaponics with Internet of Things Solutions, in: 2019
1425
19th International Conference on Sciences and
Techniques of Automatic Control and Computer
Engineering (STA). IEEE, pp. 106–111.
https://doi.org/10.1109/STA.2019.8717280
Fondriest Environmental, I., 2014. Measuring Dissolved
1430
Oxygen. [WWW Document]. Fundamentals of
Environmental Measurements. URL
https://www.fondriest.com/environmental-
measurements/measurements/measuring-water-
quality/dissolved-oxygen-sensors-and-methods/
1435
Food Security Information Network (FSIN), 2019. Global
Report on Food Crises 2019 - Joint Analysis for Better
Decisions 2019.
Garg, S.K., Bhatnagar, A., 1996. Effect of varying closes of
organic and inorganic fertilizers on plankton production
1440
and fish biomass in brackish water fish ponds.
Aquaculture Research 27, 157–166.
https://doi.org/10.1111/j.1365-2109.1996.tb00980.x
Goddek, S., Delaide, B., Mankasingh, U., Ragnarsdottir, K., Jijakli,
H., Thorarinsdottir, R., 2015. Challenges of Sustainable
1445
and Commercial Aquaponics. Sustainability 7, 4199–
4224. https://doi.org/10.3390/su7044199
Gregersen, E., 2009. pH Meter [WWW Document]. URL
https://www.britannica.com/technology/pH-meter
Haryanto, Ulum, M., Ibadillah, A.F., Alfita, R., Aji, K., Rizkyandi,
1450
R., 2019. Smart aquaponic system based Internet of
Things (IoT). Journal of Physics: Conference Series
1211, 012047. https://doi.org/10.1088/1742-
6596/1211/1/012047
Hu, Z., Lee, J.W., Chandran, K., Kim, S., Brotto, A.C., Khanal, S.K.,
1455
2015. Effect of plant species on nitrogen recovery in
aquaponics. Bioresource Technology 188, 92–98.
https://doi.org/10.1016/j.biortech.2015.01.013
Jacob, N.K., 2017. IoT powered portable aquaponics system, in:
Proceedings of the Second International Conference on
1460
Internet of Things, Data and Cloud Computing - ICC ’17.
ACM Press, New York, New York, USA, pp. 1–5.
https://doi.org/10.1145/3018896.3018965
Junge, R., König, B., Villarroel, M., Komives, T., Jijakli, M., 2017.
Strategic Points in Aquaponics. Water 9, 182.
1465
https://doi.org/10.3390/w9030182
Kloas, W., Groß, R., Baganz, D., Graupner, J., Monsees, H.,
Schmidt, U., Staaks, G., Suhl, J., Tschirner, M., Wittstock,
B., Wuertz, S., Zikova, A., Rennert, B., 2015. A new
concept for aquaponic systems to improve
1470
sustainability, increase productivity, and reduce
environmental impacts. Aquaculture Environment
Interactions 7, 179–192.
https://doi.org/10.3354/aei00146
König, B., Janker, J., Reinhardt, T., Villarroel, M., Junge, R., 2018.
1475
Analysis of aquaponics as an emerging technological
innovation system. Journal of Cleaner Production 180,
232–243.
https://doi.org/10.1016/j.jclepro.2018.01.037
Kuhn, D.D., Drahos, D.D., Marsh, L., Flick, G.J., 2010. Evaluation
1480
of nitrifying bacteria product to improve nitrification
efficacy in recirculating aquaculture systems.
Aquacultural Engineering 43, 78–82.
https://doi.org/10.1016/j.aquaeng.2010.07.001
Kumar, N.H., Baskaran, S., Hariraj, S., Krishnan, V., 2016. An
1485
Autonomous Aquaponics System Using 6LoWPAN
Based WSN, in: 2016 IEEE 4th International Conference
on Future Internet of Things and Cloud Workshops
(FiCloudW). IEEE, pp. 125–132.
https://doi.org/10.1109/W-FiCloud.2016.37
1490
Kyaw, T.Y., Ng, A.K., 2017. Smart Aquaponics System for Urban
Farming. Energy Procedia 143, 342–347.
https://doi.org/10.1016/j.egypro.2017.12.694
Lewis, W.M., Yopp, J.H., Schramm, H.L., Brandenburg, A.M.,
1978. Use of Hydroponics to Maintain Quality of
1495
Recirculated Water in a Fish Culture System.
Transactions of the American Fisheries Society 107, 92–
99. https://doi.org/10.1577/1548-
8659(1978)107<92:UOHTMQ>2.0.CO;2
Li, C., Zhang, B., Luo, P., Shi, H., Li, L., Gao, Y., Lee, C.T., Zhang, Z.,
1500
Wu, W.-M., 2019. Performance of a pilot-scale
aquaponics system using hydroponics and immobilized
biofilm treatment for water quality control. Journal of
Cleaner Production 208, 274–284.
https://doi.org/10.1016/j.jclepro.2018.10.170
1505
Love, D.C., Fry, J.P., Genello, L., Hill, E.S., Frederick, J.A., Li, X.,
Semmens, K., 2014. An International Survey of
Aquaponics Practitioners. PLoS ONE 9, 10.
https://doi.org/10.1371/journal.pone.0102662
Love, D.C., Uhl, M.S., Genello, L., 2015. Energy and water use of
1510
a small-scale raft aquaponics system in Baltimore,
Maryland, United States. Aquacultural Engineering 68,
19–27. https://doi.org/10.1016/j.aquaeng.2015.07.003
Lu, J.-Y., Chang, C.-L., Kuo, Y.-F., 2019. Monitoring Growth Rate
of Lettuce Using Deep Convolutional Neural Networks,
1515
in: 2019 Boston, Massachusetts July 7- July 10, 2019.
American Society of Agricultural and Biological
Engineers, St. Joseph, MI.
https://doi.org/10.13031/aim.201900341
Mamatha, M.N., Namratha, S.N., 2017. Design & implementation
1520
of indoor farming using automated aquaponics system,
in: 2017 IEEE International Conference on Smart
Technologies and Management for Computing,
Communication, Controls, Energy and Materials
(ICSTM). IEEE, pp. 396–401.
1525
https://doi.org/10.1109/ICSTM.2017.8089192
Mandap, J.P., Sze, D., Reyes, G.N., Matthew Dumlao, S., Reyes, R.,
Danny Chung, W.Y., 2018. Aquaponics pH Level,
Temperature, and Dissolved Oxygen Monitoring and
Control System Using Raspberry Pi as Network
1530
Backbone, in: TENCON 2018 - 2018 IEEE Region 10
Conference. IEEE, pp. 1381–1386.
https://doi.org/10.1109/TENCON.2018.8650469
Manju, M., Karthik, V., Hariharan, S., Sreekar, B., 2017. Real time
monitoring of the environmental parameters of an
1535
aquaponic system based on Internet of Things, in: 2017
Third International Conference on Science Technology
Engineering & Management (ICONSTEM). IEEE, pp.
943–948.
https://doi.org/10.1109/ICONSTEM.2017.8261342
1540
Martinez, P., Ahmad, R., Al-Hussein, M., 2019a. A vision-based
system for pre-inspection of steel frame manufacturing.
20
Preprint submitted to Journal of Cleaner Production December 12, 2019
Automation in Construction 97, 151–163.
https://doi.org/10.1016/j.autcon.2018.10.021
Martinez, P., Al-Hussein, M., Ahmad, R., 2019b. A scientometric
1545
analysis and critical review of computer vision
applications for construction. Automation in
Construction 107, 102947.
https://doi.org/10.1016/j.autcon.2019.102947
Maucieri, C., Nicoletto, C., Junge, R., Schmautz, Z., Sambo, P.,
1550
Borin, M., 2017. Hydroponic Systems and Water
Management in Aquaponics: A Review. Italian Journal of
Agronomy 11, 1–11.
https://doi.org/10.4081/ija.2017.1012
Mchunu, N., Lagerwall, G., Senzanje, A., 2018. Aquaponics in
1555
South Africa: Results of a national survey. Aquaculture
Reports 12, 12–19.
https://doi.org/10.1016/j.aqrep.2018.08.001
Mehra, M., Saxena, S., Sankaranarayanan, S., Tom, R.J.,
Veeramanikandan, M., 2018. IoT based hydroponics
1560
system using Deep Neural Networks. Computers and
Electronics in Agriculture 155, 473–486.
https://doi.org/10.1016/j.compag.2018.10.015
Murad, S.A.Z., Harun, A., Mohyar, S.N., Sapawi, R., Ten, S.Y.,
2017. Design of aquaponics water monitoring system
1565
using Arduino microcontroller, in: AIP Conference
Proceedings. p. 020248.
https://doi.org/10.1063/1.5002442
Nagayo, A.M., Mendoza, C., Vega, E., Izki, R.K.S. Al, Jamisola, R.S.,
2017. An automated solar-powered aquaponics system
1570
towards agricultural sustainability in the Sultanate of
Oman, in: 2017 IEEE International Conference on Smart
Grid and Smart Cities (ICSGSC). IEEE, pp. 42–49.
https://doi.org/10.1109/ICSGSC.2017.8038547
Naser, B.A.A.-Z., Saleem, A.L., Ali, A.H., Alabassi, S., Al-Baghdadi,
1575
M.A.R.S., 2019. Design and construction of smart IoT-
based aquaponics powered by PV cells. International
Journal of Energy and Environment 10, 87–96.
Nichols, M.A., Savidov, N.A., 2012. AQUAPONICS: A NUTRIENT
AND WATER EFFICIENT PRODUCTION SYSTEM. Acta
1580
Horticulturae 129–132.
https://doi.org/10.17660/ActaHortic.2012.947.14
Odema, M., Adly, I., Wahba, A., Ragai, H., 2018. Smart
Aquaponics System for Industrial Internet of Things
(IIoT), in: Hassanien, A.E., Tolba, M.F., Shaalan, K., Azar,
1585
A.T. (Eds.), Proceedings of the International Conference
on Advanced Intelligent Systems and Informatics,
Advances in Intelligent Systems and Computing.
Springer International Publishing, Cham, pp. 844–854.
https://doi.org/10.1007/978-3-319-64861-3_79
1590
Palande, V., Zaheer, A., George, K., 2018. Fully Automated
Hydroponic System for Indoor Plant Growth. Procedia
Computer Science 129, 482–488.
https://doi.org/10.1016/j.procs.2018.03.028
Pinstrup-Andersen, P., 2018. Is it time to take vertical indoor
1595
farming seriously? Global Food Security 17, 233–235.
https://doi.org/10.1016/j.gfs.2017.09.002
Pitakphongmetha, J., Boonnam, N., Wongkoon, S., Horanont, T.,
Somkiadcharoen, D., Prapakornpilai, J., 2016. Internet of
things for planting in smart farm hydroponics style, in:
1600
2016 International Computer Science and Engineering
Conference (ICSEC). IEEE, pp. 1–5.
https://doi.org/10.1109/ICSEC.2016.7859872
Rakocy, J.E., 2012. Aquaponics-Integrating Fish and Plant
Culture, in: Aquaculture Production Systems. Wiley-
1605
Blackwell, Oxford, UK, pp. 344–386.
https://doi.org/10.1002/9781118250105.ch14
Sallenave, R., 2012. Understanding Water Quality Parameters
to Better Manage Your Pond. College of Agricultural,
Consumer and Environmental Sciences, New Mexico
1610
State University Guide W-10, 4.
Santosh, B., Singh, N.P., 2007. Guidelines for water quality
management for Fish culture in tripura. ICAR Research
Complex for NEH Region, Tripura Centre, Lembucherra.
10.
1615
Somerville, C., Cohen, M., Pantanella, E., Stankus, A., Lovatelli,
A., 2014. Small-scale aquaponic food production.
Integrated fish and plant farming. FAO Fisheries and
Aquaculture Technical Paper No.589.
Sreelekshmi, B., Madhusoodanan, K.N., 2018. Automated
1620
aquaponics system. Emerging Trends in Engineering,
Science and Technology for Society, Energy and
Environment - Proceedings of the International
Conference in Emerging Trends in Engineering, Science
and Technology, ICETEST 2018 695–700.
1625
Stone, N.M., Thomforde, H.K., 2004. Understanding your fish
pond water analysis report, Cooperative Extension
Program, University of Arkansas at Pine Bluff, Pine
Bluff.
Tyson, R. V., Simonne, E.H., Treadwell, D.D., White, J.M.,
1630
Simonne, A., 2008. Reconciling pH for Ammonia
Biofiltration and Cucumber Yield in a Recirculating
Aquaponic System with Perlite Biofilters. HortScience
43, 719–724.
https://doi.org/10.21273/HORTSCI.43.3.719
1635
Tyson, R. V., Treadwell, D.D., Simonne, E.H., 2011.
Opportunities and Challenges to Sustainability in
Aquaponic Systems. HortTechnology 21, 6–13.
https://doi.org/10.21273/HORTTECH.21.1.6
van der Goot, A.J., Pelgrom, P.J.M., Berghout, J.A.M., Geerts,
1640
M.E.J., Jankowiak, L., Hardt, N.A., Keijer, J., Schutyser,
M.A.I., Nikiforidis, C. V., Boom, R.M., 2016. Concepts for
further sustainable production of foods. Journal of Food
Engineering 168, 42–51.
https://doi.org/10.1016/j.jfoodeng.2015.07.010
1645
Vernandhes, W., Salahuddin, N.., Kowanda, A., Sari, S.P., 2017.
Smart Aquaponic with Monitoring and Control System
based on IoT, in: 2017 Second International Conference
on Informatics and Computing (ICIC). IEEE, pp. 1–6.
https://doi.org/10.1109/IAC.2017.8280590
1650
Wada, T., 2019. Theory and Technology to Control the Nutrient
Solution of Hydroponics, in: Plant Factory Using
Artificial Light. Elsevier, pp. 5–14.
https://doi.org/10.1016/B978-0-12-813973-8.00001-4
Wang, D., Zhao, J., Huang, L., Xu, D., 2015. Design of A Smart
1655
Monitoring and Control System for Aquaponics Based
on OpenWrt, in: Proceedings of the 5th International
Conference on Information Engineering for Mechanics
and Materials. Atlantis Press, Paris, France, pp. 937–
942. https://doi.org/10.2991/icimm-15.2015.171
1660
Weber-Scan, P.K., Duffy, L.K., 2007. Effects of Total Dissolved
Solids on Aquatic Organisms: A Review of Literature
and Recommendation for Salmonid Species. American
Journal of Environmental Sciences 3, 1–6.
https://doi.org/10.3844/ajessp.2007.1.6
1665
Wei, Y., Li, W., An, D., Li, D., Jiao, Y., Wei, Q., 2019. Equipment
and Intelligent Control System in Aquaponics: A Review.
IEEE Access 7, 169306–169326.
https://doi.org/10.1109/ACCESS.2019.2953491
Werner, H., 2002. Measuring Soil Moisture for Irrigation Water
1670
Management, SDSU College of Agriculture and Biological
Sciences.
Why is PAR Rating a Big Deal for Indoor Grow Light Systems?
[WWW Document], n.d. URL
https://growace.com/blog/why-is-par-rating-a-big-
1675
deal-for-indoor-grow-light-systems/ (accessed
9.15.19a).
Why is PAR Rating a Big Deal for Indoor Grow Light Systems?
[WWW Document], n.d.
Wongkiew, S., Hu, Z., Chandran, K., Lee, J.W., Khanal, S.K., 2017.
1680
21
Preprint submitted to Journal of Cleaner Production December 12, 2019
Nitrogen transformations in aquaponic systems: A
review. Aquacultural Engineering 76, 9–19.
https://doi.org/10.1016/j.aquaeng.2017.01.004
Yep, B., Zheng, Y., 2019. Aquaponic trends and challenges – A
review. Journal of Cleaner Production 228, 1586–1599.
1685
https://doi.org/10.1016/j.jclepro.2019.04.290
YSI, n.d. Ammonia, Ammonium [WWW Document]. URL
https://www.ysi.com/parameters/ammonia
Zamora-Izquierdo, M.A., Santa, J., Martínez, J.A., Martínez, V.,
Skarmeta, A.F., 2019. Smart farming IoT platform based
1690
on edge and cloud computing. Biosystems Engineering
177, 4–17.
https://doi.org/10.1016/j.biosystemseng.2018.10.014
Zhang, H., Asutosh, A., Hu, W., 2018. Implementing Vertical
Farming at University Scale to Promote Sustainable
1695
Communities: A Feasibility Analysis. Sustainability 10,
4429. https://doi.org/10.3390/su10124429
