Figure 1 - uploaded by Jonas Köhler
Content may be subject to copyright.
The relevance of Earth Observation data for the forecasting of land surface dynamics. Global change impacts the Earth’s surface in many ways. These dynamics directly affect the livelihoods of people and can be monitored with remote sensing satellites which generate time series of geospatial EO datasets. These can yield valuable insight into processes on the Earth’s surface and enable forecasting. Reliable forecasts can inform policy and decision makers, planners, as well as society as a whole to take action to mitigate global change itself and its impacts on the land surface. Several symbols modified courtesy of the Integration and Application Network, University of Maryland Center for Environmental Science (ian.umces.edu/symbols/).
Source publication
Reliable forecasts on the impacts of global change on the land surface are vital to inform the actions of policy and decision makers to mitigate consequences and secure livelihoods. Geospatial Earth Observation (EO) data from remote sensing satellites has been collected continuously for 40 years and has the potential to facilitate the spatio-tempor...
Citations
... In RS research, considerable efforts concentrate on DL models for supervised classification tasks, thus spanning land cover classification [10], building detection [11], and scene classification [12], among others. The interest in spatiotemporal data analysis is escalating, thus prompted by its extensive application potential [13]. Urgent global challenges, like uncontrolled deforestation [14,15] and growing environmental pollution [16,17], accentuate the necessity for pioneering research and solutions, thus attracting academia, policymakers, and decision makers. ...
Earth observation imaging technologies, particularly multispectral sensors, produce extensive high-dimensional data over time, thus offering a wealth of information on global dynamics. These data encapsulate crucial information in essential climate variables, such as varying levels of soil moisture and temperature. However, current cutting-edge machine learning models, including deep learning ones, often overlook the treasure trove of multidimensional data, thus analyzing each variable in isolation and losing critical interconnected information. In our study, we enhance conventional convolutional neural network models, specifically those based on the embedded temporal convolutional network framework, thus transforming them into models that inherently understand and interpret multidimensional correlations and dependencies. This transformation involves recasting the existing problem as a generalized case of N-dimensional observation analysis, which is followed by deriving essential forward and backward pass equations through tensor decompositions and compounded convolutions. Consequently, we adapt integral components of established embedded temporal convolutional network models, like encoder and decoder networks, thus enabling them to process 4D spatial time series data that encompass all essential climate variables concurrently. Through the rigorous exploration of diverse model architectures and an extensive evaluation of their forecasting prowess against top-tier methods, we utilize two new, long-term essential climate variables datasets with monthly intervals extending over four decades. Our empirical scrutiny, particularly focusing on soil temperature data, unveils that the innovative high-dimensional embedded temporal convolutional network model-centric approaches markedly excel in forecasting, thus surpassing their low-dimensional counterparts, even under the most challenging conditions characterized by a notable paucity of training data.
... Machine Learning (ML) models play a critical role in analyzing time series data and making robust predictions required for immediate and informed action (Shao et al., 2022). Recently, the increase in data collected via satellites and stations increases the importance of these algorithms in understanding climatic and environmental changes and predicting future situations (Amato et al., 2020;Koehler and Kuenzer, 2020). Sea Surface Temperature (SST) is one of these critical parameters, and it is important for understanding the interaction between the Ocean and Earth's atmosphere. ...
Spatiotemporal time series prediction plays a crucial role in a wide range of applications. However, in most of the studies, spatial information was ignored and predictions were carried out either on a few points or on average values. In this study, 37 different configurations of 4 traditional ML models and 3 Neural Network (NN) based models were utilized to provide a comprehensive comparison and evaluate the spatiotemporal data prediction capabilities of the ML models. Additionally, to reveal the importance of spatial data for the time series prediction process, the best configuration of each ML model was evaluated with and without using spatial information. The utilized models were: (i) Linear Regression (LR), (ii) K-Nearest Neighbors (KNN), (iii) Decision-Trees (DT), (iv) Support Vector Machine (SVM), (v) Multi-Layer Perceptron (MLP), (vi) Long Short-Term Memory (LSTM), and (vii) Gated Recurrent Unit (GRU). The study was performed on the Sea Surface Temperature (SST) data collected by satellite radiometers via infrared measurements. The models were evaluated according to their one-month ahead spatiotemporal SST prediction performance over the southern coasts of Turkey, and the effects of spatial information on model performance were presented. Results reveal that the spatial information increased the prediction performance by approximately 25%, in terms of RMSE. Additionally, acquired results show that the LSTM model outperforms all other ML models and gives the smallest prediction errors in all metrics.
... Related work At low spatial resolutions, machine learning models have been proposed for both vegetation modeling [11,[16][17][18][19][20][21][22] and crop forecasting [23][24][25]. Requena-Mesa et al. [26] introduced Earth surface forecasting as modeling the future spectral reflectance of the Earth surface and provided the first dataset in this respect, EarthNet2021. ...
Forecasting the state of vegetation in response to climate and weather events is a major challenge. Its implementation will prove crucial in predicting crop yield, forest damage, or more generally the impact on ecosystems services relevant for socio-economic functioning, which if absent can lead to humanitarian disasters. Vegetation status depends on weather and environmental conditions that modulate complex ecological processes taking place at several timescales. Interactions between vegetation and different environmental drivers express responses at instantaneous but also time-lagged effects, often showing an emerging spatial context at landscape and regional scales. We formulate the land surface forecasting task as a strongly guided video prediction task where the objective is to forecast the vegetation developing at very fine resolution using topography and weather variables to guide the prediction. We use a Convolutional LSTM (ConvLSTM) architecture to address this task and predict changes in the vegetation state in Africa using Sentinel-2 satellite NDVI, having ERA5 weather reanalysis, SMAP satellite measurements, and topography (DEM of SRTMv4.1) as variables to guide the prediction. Ours results highlight how ConvLSTM models can not only forecast the seasonal evolution of NDVI at high resolution, but also the differential impacts of weather anomalies over the baselines. The model is able to predict different vegetation types, even those with very high NDVI variability during target length, which is promising to support anticipatory actions in the context of drought-related disasters.
... For six catchments in the Alps, long-term linear trends of SLE during the ablation season were also calculated and a significant retreat of the SLE to higher elevations was detected in five cases [12]. Even though this example demonstrates that the detection of long-term trends of land surface dynamics from EO data is possible, the potential of long EO time series for forecasting future processes has not yet been fully utilized [16]. EO time series from missions such as Landsat encompass hundreds of approximately bi-weekly observations over several decades. ...
... The generation of long EO time series has only been possible for a few years since Landsat entered its fourth decade of operation. As a result, the potential of these time series to forecast land surface parameters such as snow cover dynamics has not yet been fully utilized [16]. In this study, we modeled future SLE dynamics in the Alps by training and evaluating well-established univariate forecasting models on long time series of freely available Landsat data. ...
Snow is a vital environmental parameter and dynamically responsive to climate change, particularly in mountainous regions. Snow cover can be monitored at variable spatial scales using Earth Observation (EO) data. Long-lasting remote sensing missions enable the generation of multi-decadal time series and thus the detection of long-term trends. However, there have been few attempts to use these to model future snow cover dynamics. In this study, we, therefore, explore the potential of such time series to forecast the Snow Line Elevation (SLE) in the European Alps. We generate monthly SLE time series from the entire Landsat archive (1985–2021) in 43 Alpine catchments. Positive long-term SLE change rates are detected, with the highest rates (5–8 m/y) in the Western and Central Alps. We utilize this SLE dataset to implement and evaluate seven uni-variate time series modeling and forecasting approaches. The best results were achieved by Random Forests, with a Nash–Sutcliffe efficiency (NSE) of 0.79 and a Mean Absolute Error (MAE) of 258 m, Telescope (0.76, 268 m), and seasonal ARIMA (0.75, 270 m). Since the model performance varies strongly with the input data, we developed a combined forecast based on the best-performing methods in each catchment. This approach was then used to forecast the SLE for the years 2022–2029. In the majority of the catchments, the shift of the forecast median SLE level retained the sign of the long-term trend. In cases where a deviating SLE dynamic is forecast, a discussion based on the unique properties of the catchment and past SLE dynamics is required. In the future, we expect major improvements in our SLE forecasting efforts by including external predictor variables in a multi-variate modeling approach.
... In most cases, the transition probabilities of CA-based models are computed based on the differences between the discrete observations. Koehler and Kuenzer [10] found that long satellite time-series data are seldom used in urban/settlement growth modelling and that the temporal information that is stored in urban time-series products, such as the WSF-evolution [8], is yet unexploited. ...
In the past decades, various Earth observation-based time series products have emerged, which have enabled studies and analysis of global change processes. Besides their contribution to understanding past processes, time series datasets hold enormous potential for predictive modeling and thereby meet the demands of decision makers on future scenarios. In order to further exploit these data, a novel pixel-based approach has been introduced, which is the spatio-temporal matrix (STM). The approach integrates the historical characteristics of a specific land cover at a high temporal frequency in order to interpret the spatial and temporal information for the neighborhood of a given target pixel. The provided information can be exploited with common predictive models and algorithms. In this study, this approach was utilized and evaluated for the prediction of future urban/built-settlement growth. Random forest and multi-layer perceptron were employed for the prediction. The tests have been carried out with training strategies based on a one-year and a ten-year time span for the urban agglomerations of Surat (India), Ho-Chi-Minh City (Vietnam), and Abidjan (Ivory Coast). The slope, land use, exclusion, urban, transportation, hillshade (SLEUTH) model was selected as a baseline indicator for the performance evaluation. The statistical results from the receiver operating characteristic curve (ROC) demonstrate a good ability of the STM to facilitate the prediction of future settlement growth and its transferability to different cities, with area under the curve (AUC) values greater than 0.85. Compared with SLEUTH, the STM-based model achieved higher AUC in all of the test cases, while being independent of the additional datasets for the restricted and the preferential development areas.
... For example, Remote Sensing (RS) missions such as Landsat have continuously been acquiring global imagery of the land surface for almost 40 years [15]. Since the inception of the first EO satellites in the 1970s, the archives of RS data have increased dramatically for a number of reasons: a multitude of sensors with different spatial, temporal, and spectral properties has been launched during the last decade that produced a continuous stream of global data; ongoing EO missions have been expanded to enable continuous observations over decades; an increasing variety of analysis-ready (ARD) products has been generated that facilitate scientific analysis of derived geophysical parameters, indices, and thematic information [16]. These developments have been accompanied by rapid advances in computer and data science in the past years. ...
... For six catchments in the Alps, long-term linear trends of SLE during the ablation season were also calculated and a significant retreat of the SLE to higher elevations was detected in five cases [12]. Even though this example demonstrates that the detection of long-term trends of land surface dynamics from EO data is possible, the potential of long EO time series for forecasting future processes has not yet been fully utilized [16]. EO time series from missions such as Landsat encompass hundreds of approximately bi-weekly observations over several decades. ...
... The generation of long EO time series has only been possible for a few years since Landsat entered its fourth decade of operation. As a result, the potential of these time series to forecast land surface parameters such as snow cover dynamics has not yet been fully utilized [16]. In this study, we modeled future SLE dynamics in the Alps by training and evaluating well-established univariate forecasting models on long time series of freely available Landsat data. ...
The inter and intra-annual dynamics of seasonal snow are of key interest in the tourism-based economies of many Alpine regions as well as for millions of people in the adjacent European lowlands when it comes to freshwater supply and electricity generation. However, accurate snow observations over long periods of time and at large spatial scales are especially challenging in inaccessible mountainous areas. This can be overcome by using data from Earth Observation satellites, which have been constantly monitoring the Earth’s surface for almost 40 years. On a catchment basis, we derive the Snow Line Elevation (SLE) from Landsat data for the entire Alpine region and model the spatio-temporal dynamics in monthly time-series ranging from 1984 to today. Based on the historical observations we model future SLE dynamics comparing different uni-variate and multi-variate approaches and assess them for their ability to generate multi-year forecasts from EO-derived time series data. These forecasts can enable local and regional stakeholders to adapt to a potentially changing snow regime under climate change.
... With the MOLUSE Plugin in QGIS 2.18, the LST is commonly forecasted using an artificial neural network (ANN) until 2030 ( Alam et al., 2021 ;Nugroho et al., 2018 ). ANN is a useful method for forecasting future time series LST and LULC using data from past years ( Faisal et al., 2021 ;Fattah et al., 2021 ;Imran and Mehmood, 2020 ;Koehler and Kuenzer, 2020 ; I. I. Maduako et al., 2016 ;I. D. I.D. Maduako et al., 2016 ;Mustafa et al., 2020 ;Nanjing et al., 2020 ). ...
This study, for the first time, assesses the impact of critical environmental factors on groundwater using Bayesian Network (BN) integrated with Analytical Hierarchy Process (AHP) and develop groundwater vulnerability map. The considered environmental factors are divided into: physical (rainfall, temperature (Tmax/Tmin), relative humidity (RHmax/RHmin)), water use and demand (surface water availability (SW), and number of tube wells (NTW)), agriculture and land use (total irrigated area (TIA), total cropped area (TCA), and total area sown (TAS)), and population. Results showed negative relationship of rainfall, RHmax/RHmin, and SW, while positive relationship of remaining variables with groundwater. Elasticities demonstrated that 1% change in SW (rainfall), major contributors, resulted in 0.64% (0.55%) decrease in groundwater in Bahawalpur (Multan). A 1% change in population (NTW), major consumers, resulted in 0.74% (0.70%) increase in DWT across Jhang (Khanewal). Vulnerability map depicted that high and very high vulnerability classes accounted for more than 50% of total Punjab.
... Although the field measurement method is convenient and easy to operate, it is limited by a small observation area, which is not conducive to comprehensive analysis. The remote sensing observation data are more intuitive, becoming an important method for obtaining and analyzing UHI information [12][13][14]. In remote sensing methods, the low-altitude remote sensing platform compensates for the low-resolution and real-time defects of the satellite remote sensing platform [15][16][17]. ...
Water and greenery space is an essential part of the urban ecosystem and plays a vital role in alleviating the urban heat island effect. Water and greenery differently affect the urban microclimate. After coupling the two factors, the quantitative human thermal environmental experience analysis needs to be revealed. This study investigated the thermal environment of the waterfront space with the greenery of Tianjin in the cold regions of China. Different types of microscale water were analyzed to evaluate their thermal environment in summer using subjective and objective analysis and model simulation. We concluded the coupling mechanism of waterfront greenery and thermal evaluation standard. The results show that water and greenery's coupling effect peaked at 14:00 and waterfront at 4–8 m. Conversely, the cooling effect was more significant at 3–6 m from the water's edge, where the location was suitable for construction as verified by the optimization model. The prerequisite of coupling mechanism was suitable green coverage rate and vegetation type, based on the comprehensive analysis of four cases. The objective effects of different waterfront distances on the thermal comfort of Heat Index (HI) and the subjective evaluation of Universal Thermal Climate Index (UTCI) for people were quantified. This research provides a theoretical basis and data support for urban waterfront renewal in cold regions.
... Only about 6% of studies were conducted at a global scale, 5% at a national scale, and 2% at a continental scale (Figure 4). Data processing and storage power has increased, but it seems that working with EO data at continental and global scale remain a challenge [99]. Another reason could also be because savannas are so variable at large scales, and, hence, methods must be developed and calibrated at smaller scales. ...
... Savanna mixed-pixel analysis of vegetation components often requires multi-temporal images spanning over several seasons to capture phenological differences in order to increase accuracy [5]. These multitemporal images are required to go through pre-processing, which increases the computational demand [99]. At a global scale, generating and collecting reliable and geographically representative validation data tend to be a challenge. ...
Reliable estimates of savanna vegetation constituents (i.e., woody and herbaceous vegetation) are essential as they are both responders and drivers of global change. The savanna is a highly heterogenous biome with high variability in land cover types while also being very dynamic at both temporal and spatial scales. To understand the spatial-temporal dynamics of savannas, using Earth Observation (EO) data for mixed-pixel analysis is crucial. Mixed pixel analysis provides detailed land cover data at a sub-pixel level which are essential for conservation purposes, understanding food supply for herbivores, quantifying environmental change, such as bush encroachment, and fuel availability essential for understanding fire dynamics, and for accurate estimation of savanna biomass. This review paper consulted 197 studies employing mixed-pixel analysis in savanna ecosystems. The review indicates that studies have so far attempted to resolve the savanna mixed-pixel issues by using mainly coarse resolution data, such as Terra-Aqua MODIS and AVHRR and medium resolution Landsat, to provide fractional cover data. Hence, there is a lack of spatio-temporal mixed-pixel analysis for savannas at high spatial resolutions. Methods used for mixed-pixel analysis include parametric and non-parametric methods which range from pixel-unmixing models, such as linear spectral mixture analysis (SMA), time series decomposition, empirical methods to link the green vegetation parameters with Vegetation Indices (VIs), and machine learning methods, such as regression trees (RT) and random forests (RF). Most studies were undertaken at local and regional scale, highlighting a research gap for savanna mixed pixel studies at national, continental, and global level. Parametric methods for modeling spatio-temporal mixed pixel analysis were preferred for coarse to medium resolution remote sensing data, while non-parametric methods were preferred for very high to high spatial resolution data. The review indicates a gap for long time series spatio-temporal mixed-pixel analysis of savannas using high resolution data at various scales. There is potential to harmonize the available low resolution EO data with new high-resolution sensors to provide long time series of the savanna mixed pixel, which, according to this review, is missing. Citation: Nghiyalwa, H.S.; Urban, M.; Baade, J.; Smit, I.P.J.; Ramoelo, A.; Mogonong, B.; Schmullius, C. Spatio-Temporal Mixed Pixel
... Earth observation sensors have been widely used in the last two decades to observe, survey, and monitor the built heritage environment [1][2][3]. The increased capabilities of space programs initiated and operated by several national agencies and the private sector facilitated research and application around the study, modelling, and predicting of various natural and anthropogenic phenomena affecting built heritage [4,5]. ...
This study combines satellite observation, cloud platforms, and geographical information systems (GIS) to investigate at a macro-scale level of observation the thermal conditions of two historic clusters in Cyprus, namely in Limassol and Strovolos municipalities. The two case studies share different environmental and climatic conditions. The former site is coastal, the last a hinterland, and they both contain historic buildings with similar building materials and techniques. For the needs of the study, more than 140 Landsat 7 ETM+ and 8 LDCM images were processed at the Google Earth Engine big data cloud platform to investigate the thermal conditions of the two historic clusters over the period 2013–2020. The multi-temporal thermal analysis included the calibration of all images to provide land surface temperature (LST) products at a 100 m spatial resolution. Moreover, to investigate anomalies related to possible land cover changes of the area, two indices were extracted from the satellite images, the normalised difference vegetation index (NDVI) and the normalised difference build index (NDBI). Anticipated results include the macro-scale identification of multi-temporal changes, diachronic changes, the establishment of change patterns based on seasonality and location, occurring in large clusters of historic buildings.
