Crop type discrimination using Geo-Stat Endmember Extraction and machine learning algorithms

The identification of crop diversity in today's world is very crucial to ensure adaptation of the crop with changing climate for better productivity as well as food security. Towards this, Hyperspectral Remote Sensing (HRS) is an efficient technique based on imaging spectroscopy that offers the opportunity to discriminate crop types based on morphological as well as physiological features due to availability of contiguous spectral bands. The current work utilized the benefits of Airborne Visible Infrared Imaging spectrometer- New Generation (AVIRIS-NG) data and explored the techniques for classification and identification of crop types. The endmembers were identified using the Geo-Stat Endmember Extraction (GSEE) algorithm for pure pixels identification and to generate the spectral library of the different crop types. Spectral feature comparison was done among AVIRIS-NG, Analytical Spectral Device (ASD)-Spectroradiometer and Continuum Removed (CR) spectra. The best-fit spectra obtained with the Reference ASD-Spectroradiometer and Pure Pixel spectral library were then used for crop discrimination using the ten supervised classifiers namely Spectral Angle Mapper (SAM), Spectral Information Divergence (SID), Support Vector Machine (SVM), Minimum Distance Classifier (MDC), Binary Encoding, deep learning-based Convolution Neural Network (CNN) and different algorithms of Ensemble learning such as Tree Bag, AdaBoost (Adaptive Boosting), Discriminant and RUSBoost (Random Under Sampling). In total, nine crop types were identified, namely, wheat, maize, tobacco, sorghum, linseed, castor, pigeon pea, fennel and chickpea. The performance evaluation of the classifiers was made using various metrics like Overall Accuracy, Kappa Coefficient, Precision, Recall and F1 score. The classifier 2D-CNN was found to be the best with Overall Accuracy, Kappa Coefficient, Precision, Recall and F1 score values of 89.065 %, 0.871,87.565%, 89.541% and 88.678% respectively. The output of this work can be utilized for large scale mapping of crop types at the species level in a short interval of time of a large area with high accuracy.

Assessment of maize nitrogen uptake from PRISMA hyperspectral data through hybrid modelling

The spaceborne imaging spectroscopy mission PRecursore IperSpettrale della Missione Applicativa (PRISMA), launched on 22 March 2019 by the Italian Space Agency, opens new opportunities in many scientific domains, including precision farming and sustainable agriculture. This new Earth Observation (EO) data stream requires new-generation approaches for the estimation of important biophysical crop variables (BVs). In this framework, this study evaluated a hybrid approach, combining the radiative transfer model PROSAIL-PRO and several machine learning (ML) regression algorithms, for the retrieval of canopy chlorophyll content (CCC) and canopy nitrogen content (CNC) from synthetic PRISMA data. PRISMA-like data were simulated from two images acquired by the airborne sensor HyPlant, during a campaign performed in Grosseto (Italy) in 2018. CCC and CNC estimations, assessed from the best performing ML algorithms, were used to define two relations with plant nitrogen uptake (PNU). CNC proved to be slightly more correlated to PNU than CCC (R2 = 0.82 and R2 = 0.80, respectively). The CNC-PNU model was then applied to actual PRISMA images acquired in 2020. The results showed that the estimated PNU values are within the expected ranges, and the temporal trends are compatible with plant phenology stages.

In-situ start and end of growing season dates of major European crop types from France and Bulgaria at a field level

Crop phenology data offer crucial information for crop yield estimation, agricultural management, and assessment of agroecosystems. Such information becomes more important in the context of increasing year-to-year climatic variability. The dataset provides in-situ crop phenology data (first leaves emergence and harvest date) of major European crops (wheat, corn, sunflower, rapeseed) from seventeen field study sites in Bulgaria and two in France. Additional information such as the sowing date, area of each site, coordinates, method and equipment used for phenophase data estimation, and photos of the France sites are also provided. The georeferenced ground-truth dataset provides a solid base for a better understanding of crop growth and can be used to validate the retrieval of phenological stages from remote sensing data.

Spatial and Temporal Analysis of Water Quality in High Andean Lakes with Sentinel-2 Satellite Automatic Water Products

The water of high Andean lakes is strongly affected by anthropic activities. However, due to its complexity this ecosystem is poorly researched. This study analyzes water quality using Sentinel-2 (S2) images in high Andean lakes with apparent different eutrophication states. Spatial and temporal patterns are assessed for biophysical water variables from automatic products as obtained from versions of C2RCC (Case 2 Regional Coast Color) processor (i.e., C2RCC, C2X, and C2X-COMPLEX) to observe water characteristics and eutrophication states in detail. These results were validated using in situ water sampling. C2X-COMPLEX appeared to be an appropriate option to study bodies of water with a complex dynamic of water composition. C2RCC was adequate for lakes with high transparency, typical for lakes of highlands with excellent water quality. The Yambo lake, with chlorophyll-a concentration (CHL) values of 79.6 ± 5 mg/m3, was in the eutrophic to hyper-eutrophic state. The Colta lake, with variable values of CHL, was between the oligotrophic to mesotrophic state, and the Atillo lakes, with values of 0.16 ± 0.1 mg/m3, were oligotrophic and even ultra-oligotrophic, which remained stable in the last few years. Automatic S2 water products give information about water quality, which in turn makes it possible to analyze its causes.

Simultaneous retrieval of sugarcane variables from Sentinel-2 data using Bayesian regularized neural network

Quantifying biophysical and biochemical vegetation variables is of great importance in precision agriculture. Here, the ability of artificial neural networks (ANNs) to generate multiple outputs is exploited to simultaneously retrieve Leaf area index (LAI), leaf sheath moisture (LSM), leaf chlorophyll content (LCC), and leaf nitrogen concentration (LNC) of sugarcane from Sentinel-2 spectra. We apply a type of ANNs, Bayesian Regularized ANN (BRANN), which incorporates the Bayes' theorem into a regularization scheme to tackle the overfitting problem of ANN and improve its generalizability. Quantitatively assessing the result accuracy indicated RMSE values of 0.48 (m²/m²) for LAI, 2.36 (% wb) for LSM, 5.85 (μg/cm²) for LCC, and 0.23 (%) for LNC, applying simultaneous retrieval. It was demonstrated that simultaneous retrievals of the variables outperformed the individual retrievals. The superiority of the proposed BRANN over a conventional ANN trained with the Levenberg-Marquardt algorithm was confirmed through statistical comparison of their results. The model was applied over the entire Sentinel-2 images to map the considered variables. The maps were probed to qualitatively evaluate the model performance. The results indicated that the retrievals reasonably represent spatial and temporal variations of the variables. Generally, this study demonstrated that the BRANN simultaneous retrieval model can provide faster and more accurate retrievals than those obtained from conventional ANNs and individual retrievals.

Neural Network Emulation of Synthetic Hyperspectral Sentinel-2-like Imagery with Uncertainty

Hyperspectral satellite imagery provides highly-resolved spectral information for large areas and can provide vital information. However, only a few imaging spectrometer missions are currently in operation. Aiming to generate synthetic satellite-based hyperspectral imagery potentially covering any region, we explored the possibility of applying statistical learning, i.e. emulation. Based on the relationship of a Sentinel-2 (S2) scene and a hyperspectral HyPlant airborne image, this work demonstrates the possibility to emulate a hyperspectral S2-like image. We tested the role of different machine learning regression algorithms (MLRA) and varied the image-extracted training dataset size. We found superior performance of Neural Network (NN) as opposed to the other algorithms when trained with large datasets (up to 100'000 samples). The developed emulator was then applied to the L2A (bottom-of-atmosphere reflectance) S2 subset, and the obtained S2-like hyperspectral reflectance scene was evaluated. The validation of emulated against reference spectra demonstrated the potential of the technique. R 2 values between 0.75-0.9 and NRMSE between 2-5% across the full 402-2356 nm range were obtained. Moreover, epistemic uncertainty is obtained using the dropout technique, revealing spatial fidelity of the emulated scene. We obtained highest SD values of 0.05 (CV of 8%) in clouds and values below 0.01 (CV of 7%) in vegetation land covers. Finally, the emulator was applied to an entire S2 tile (5490x5490 pixels) to generate a hyperspectral reflectance datacube with the texture of S2 (60Gb, at a speed of 0.14sec/10000pixels). As the emulator can convert any S2 tile into a hyperspectral image, such scenes give perspectives how future satellite imaging spectroscopy will look like.

Quantifying Irrigated Winter Wheat LAI in Argentina Using Multiple Sentinel-1 Incidence Angles

Synthetic aperture radar (SAR) data provides an appealing opportunity for all-weather day or night Earth surface monitoring. The European constellation Sentinel-1 (S1) consisting of S1-A and S1-B satellites offers a suitable revisit time and spatial resolution for the observation of croplands from space. The C-band radar backscatter is sensitive to vegetation structure changes and phenology as well as soil moisture and roughness. It also varies depending on the local incidence angle (LIA) of the SAR acquisition's geometry. The LIA backscatter dependency could therefore be exploited to improve the retrieval of the crop biophysical variables. The availability of S1 radar time-series data at distinct observation angles holds the feasibility to retrieve leaf area index (LAI) evolution considering spatiotemporal coverage of intensively cultivated areas. Accordingly, this research presents a workflow merging multi-date S1 smoothed data acquired at distinct LIA with a Gaussian processes regression (GPR) and a cross-validation (CV) strategy to estimate cropland LAI of irrigated winter wheat. The GPR-S1-LAI model was tested against in situ data of the 2020 winter wheat campaign in the irrigated valley of Colorador river, South of Buenos Aires Province, Argentina. We achieved adequate validation results for LAI with R C V 2 = 0.67 and RMSE _CV = 0.88 m² m^-2. The trained model was further applied to a series of S1 stacked images, generating temporal LAI maps that well reflect the crop growth cycle. The robustness of the retrieval workflow is supported by the associated uncertainties along with the obtained maps. We found that processing S1 smoothed imagery with distinct acquisition geometries permits accurate radar-based LAI modeling throughout large irrigated areas and in consequence can support agricultural management practices in cloudprone agri-environments.

Hyperspectral response of agronomic variables to background optical variability: Results of a numerical experiment

Understanding how biophysical and biochemical variables contribute to the spectral characteristics of vegetation canopies is critical for their monitoring. Quantifying these contributions, however, remains difficult due to extraneous factors such as the spectral variability of canopy background materials, including soil/crop-residue moisture, soil-type, and non-photosynthetic vegetation (NPV). This study focused on exploring the spectral response of two important agronomic variables (1) leaf chlorophyll content (Cab ) and (2) leaf area index (LAI) under various canopy backgrounds through a global sensitivity analysis of wheat-like canopy spectra simulated using the physically-based PROSAIL radiative transfer model. Our results reveal the following general findings: (1) the contribution of each agronomic variable to the simulated canopy spectral signature varies considerably with respect to the background optical properties; (2) the influence of the soil-type and NPV on the spectral response of canopy to Cab and LAI is more significant than that caused by soil/crop-residue moisture; (3) spectral bands at 560 and 704 nm remain sensitive to Cab while being least affected by the impacts of variations in the NPV, soil-type and moisture; (4) the near-infrared (NIR) spectral bands exhibit higher sensitivity to LAI and lower background effects only in the cases of soil/crop-residue moisture but are relatively strongly affected by soil-type and NPV. Comparative analysis of the correlations of twelve widely used vegetation indices with agronomic variables indicates that LICI (LAI-insensitive chlorophyll index) and Macc01 (Maccioni index) are more effective in estimating Cab , while OSAVI (optimized soil adjusted vegetation index) and MCARI2 (modified chlorophyll absorption ratio index 2) are better LAI predictors under the simulated background variability. Overall, our results highlight that background reflectance variability introduces considerable differences in the agronomic variables' spectral response, leading to inconsistencies in the VI- Cab /-LAI relationship. Further studies should integrate these results of spectral responsivity to develop trait-specific hyperspectral inversion models.

Retrieval of carbon content and biomass from hyperspectral imagery over cultivated areas

Spaceborne imaging spectroscopy is a highly promising data source for all agricultural management and research disciplines that require spatio-temporal information on crop properties. Recently launched science-driven missions, such as the Environmental Mapping and Analysis Program (EnMAP), deliver unprecedented data from the Earth's surface. This new kind of data should be explored to develop robust retrieval schemes for deriving crucial variables from future routine missions. Therefore, we present a workflow for inferring crop carbon content (Carea ), and aboveground dry and wet biomass (AGBdry , AGBfresh ) from EnMAP data. To achieve this, a hybrid workflow was generated, combining radiative transfer modeling (RTM) with machine learning regression algorithms. The key concept involves: (1) coupling the RTMs PROSPECT-PRO and 4SAIL for simulation of a wide range of vegetation states, (2) using dimensionality reduction to deal with collinearity, (3) applying a semi-supervised active learning technique against a 4-years campaign dataset, followed by (4) training of a Gaussian process regression (GPR) machine learning algorithm and (5) validation with an independent in situ dataset acquired during the ESA Hypersense experiment campaign at a German test site. Internal validation of the GPR-Carea and GPR-AGB models achieved coefficients of determination (R 2) of 0.80 for Carea and 0.80, 0.71 for AGBdry and AGBfresh , respectively. The mapping capability of these models was successfully demonstrated using airborne AVIRIS-NG hyperspectral imagery, which was spectrally resampled to EnMAP spectral properties. Plausible estimates were achieved over both bare and green fields after adding bare soil spectra to the training data. Validation over green winter wheat fields generated reliable estimates as suggested by low associated model uncertainties (< 40%). These results suggest a high degree of model reliability for cultivated areas during active growth phases at canopy closure. Overall, our proposed carbon and biomass models based on EnMAP spectral sampling demonstrate a promising path toward the inference of these crucial variables over cultivated areas from future spaceborne operational hyperspectral missions.

Multi-sensor spectral synergies for crop stress detection and monitoring in the optical domain: A review

Remote detection and monitoring of the vegetation responses to stress became relevant for sustainable agriculture. Ongoing developments in optical remote sensing technologies have provided tools to increase our understanding of stress-related physiological processes. Therefore, this study aimed to provide an overview of the main spectral technologies and retrieval approaches for detecting crop stress in agriculture. Firstly, we present integrated views on: i) biotic and abiotic stress factors, the phases of stress, and respective plant responses, and ii) the affected traits, appropriate spectral domains and corresponding methods for measuring traits remotely. Secondly, representative results of a systematic literature analysis are highlighted, identifying the current status and possible future trends in stress detection and monitoring. Distinct plant responses occurring under shortterm, medium-term or severe chronic stress exposure can be captured with remote sensing due to specific light interaction processes, such as absorption and scattering manifested in the reflected radiance, i.e. visible (VIS), near infrared (NIR), shortwave infrared, and emitted radiance, i.e. solar-induced fluorescence and thermal infrared (TIR). From the analysis of 96 research papers, the following trends can be observed: increasing usage of satellite and unmanned aerial vehicle data in parallel with a shift in methods from simpler parametric approaches towards more advanced physically-based and hybrid models. Most study designs were largely driven by sensor availability and practical economic reasons, leading to the common usage of VIS-NIR-TIR sensor combinations. The majority of reviewed studies compared stress proxies calculated from single-source sensor domains rather than using data in a synergistic way. We identified new ways forward as guidance for improved synergistic usage of spectral domains for stress detection: (1) combined acquisition of data from multiple sensors for analysing multiple stress responses simultaneously (holistic view); (2) simultaneous retrieval of plant traits combining multi-domain radiative transfer models and machine learning methods; (3) assimilation of estimated plant traits from distinct spectral domains into integrated crop growth models. As a future outlook, we recommend combining multiple remote sensing data streams into crop model assimilation schemes to build up Digital Twins of agroecosystems, which may provide the most efficient way to detect the diversity of environmental and biotic stresses and thus enable respective management decisions.

Seasonal Mapping of Irrigated Winter Wheat Traits in Argentina with a Hybrid Retrieval Workflow Using Sentinel-2 Imagery

Earth observation offers an unprecedented opportunity to monitor intensively cultivated areas providing key support to assess fertilizer needs and crop water uptake. Routinely, vegetation traits mapping can help farmers to monitor plant development along the crop's phenological cycle, which is particularly relevant for irrigated agricultural areas. The high spatial and temporal resolution of the Sentinel-2 (S2) multispectral instrument leverages the possibility to estimate leaf area index (LAI), canopy chlorophyll content (CCC), and vegetation water content (VWC) from space. Therefore, our study presents a hybrid retrieval workflow combining a physically-based strategy with a machine learning regression algorithm, i.e., Gaussian processes regression, and an active learning technique to estimate LAI, CCC and VWC of irrigated winter wheat. The established hybrid models of the three traits were validated against in-situ data of a wheat campaign in the Bonaerense valley, South of the Buenos Aires Province, Argentina, in the year 2020. We obtained good to highly accurate validation results with LAI: R² = 0.92, RMSE = 0.43 m² m^-2, CCC: R² = 0.80, RMSE = 0.27 g m^-2 and VWC: R² = 0.75, RMSE = 416 g m^-2. The retrieval models were also applied to a series of S2 images, producing time series along the seasonal cycle, which reflected the effects of fertilizer and irrigation on crop growth. The associated uncertainties along with the obtained maps underlined the robustness of the hybrid retrieval workflow. We conclude that processing S2 imagery with optimised hybrid models allows accurate space-based crop traits mapping over large irrigated areas and thus can support agricultural management decisions.

Introducing ARTMO's Machine-Learning Classification Algorithms Toolbox: Application to Plant-Type Detection in a Semi-Steppe Iranian Landscape

Accurate plant-type (PT) detection forms an important basis for sustainable land management maintaining biodiversity and ecosystem services. In this sense, Sentinel-2 satellite images of the Copernicus program offer spatial, spectral, temporal, and radiometric characteristics with great potential for mapping and monitoring PTs. In addition, the selection of a best-performing algorithm needs to be considered for obtaining PT classification as accurate as possible. To date, no freely downloadable toolbox exists that brings the diversity of the latest supervised machine-learning classification algorithms (MLCAs) together into a single intuitive user-friendly graphical user interface (GUI). To fill this gap and to facilitate and automate the usage of MLCAs, here we present a novel GUI software package that allows systematically training, validating, and applying pixel-based MLCA models to remote sensing imagery. The so-called MLCA toolbox has been integrated within ARTMO's software framework developed in Matlab which implements most of the state-of-the-art methods in the machine learning community. To demonstrate its utility, we chose a heterogeneous case study scene, a landscape in Southwest Iran to map PTs. In this area, four main PTs were identified, consisting of shrub land, grass land, semi-shrub land, and shrub land-grass land vegetation. Having developed 21 MLCAs using the same training and validation, datasets led to varying accuracy results. Gaussian process classifier (GPC) was validated as the top-performing classifier, with an overall accuracy (OA) of 90%. GPC follows a Laplace approximation to the Gaussian likelihood under the supervised classification framework, emerging as a very competitive alternative to common MLCAs. Random forests resulted in the second-best performance with an OA of 86%. Two other types of ensemble-learning algorithms, i.e., tree-ensemble learning (bagging) and decision tree (with error-correcting output codes), yielded an OA of 83% and 82%, respectively. Following, thirteen classifiers reported OA between 70% and 80%, and the remaining four classifiers reported an OA below 70%. We conclude that GPC substantially outperformed all classifiers, and thus, provides enormous potential for the classification of a diversity of land-cover types. In addition, its probabilistic formulation provides valuable band ranking information, as well as associated predictive variance at a pixel level. Nevertheless, as these are supervised (data-driven) classifiers, performances depend on the entered training data, meaning that an assessment of all MLCAs is crucial for any application. Our analysis demonstrated the efficacy of ARTMO's MLCA toolbox for an automated evaluation of the classifiers and subsequent thematic mapping.

Quantifying mangrove leaf area index from Sentinel-2 imagery using hybrid models and active learning

Mangrove forests provide vital ecosystem services. The increasing threats to mangrove forest extent and fragmentation can be monitored from space. Accurate spatially explicit quantification of key vegetation characteristics of mangroves, such as leaf area index (LAI), would further advance our monitoring efforts to assess ecosystem health and functioning. Here, we investigated the potential of radiative transfer models (RTM), combined with active learning (AL), to estimate LAI from Sentinel-2 spectral reflectance in the mangrove-dominated region of Ngoc Hien, Vietnam. We validated the retrieval of LAI estimates against in-situ measurements based on hemispherical photography and compared against red-edge NDVI and the Sentinel Application Platform (SNAP) biophysical processor. Our results highlight the performance of physics-based machine learning using Gaussian processes regression (GPR) in combination with AL for the estimation of mangrove LAI. Our AL-driven hybrid GPR model substantially outperformed SNAP (R2 = 0.77 and 0.44 respectively) as well as the red-edge NDVI approach. Comparing two canopy RTMs, the highest accuracy was achieved by PROSAIL (RMSE = 0.13 m2.m-2, NRMSE = 9.57%, MAE = 0.1 m2.m-2). The successful retrieval of mangrove LAI from Sentinel-2 can overcome extensive reliance on scarce in-situ measurements for training seen in other approaches and present a more scalable applicability by relying on the universal principles of physics in combination with uncertainty estimates. AL-based GPR models using RTM simulations allow us to adapt the genericity of RTMs to the peculiarities of distinct ecosystems such as mangrove forests with limited ancillary data. These findings bode potential for retrieving a wider range of vegetation variables to quantify large-scale mangrove ecosystem dynamics in space and time.

Analysis of Biophysical Variables in an Onion Crop (Allium cepa L.) with Nitrogen Fertilization by Sentinel-2 Observations

The production of onions bulbs (Allium cepa L.) requires a high amount of nitrogen. According to the demand of sustainable agriculture, the information-development and communication technologies allow for improving the efficiency of nitrogen fertilization. In the south of the province of Buenos Aires, Argentina, between 8000 and 10,000 hectares per year^-1 are cultivated in the districts of Villarino and Patagones. This work aimed to analyze the relationship of biophysical variables: leaf area index (LAI), canopy chlorophyll content (CCC), and canopy cover factor (fCOVER), with the nitrogen fertilization of an intermediate cycle onion crop and its effects on yield. A field trial study with different doses of granulated urea and granulated urea was carried out, where biophysical characteristics were evaluated in the field and in Sentinel-2 satellite observations. Field data correlated well with satellite data, with an R² of 0.91, 0.96, and 0.85 for LAI, fCOVER, and CCC, respectively. The application of nitrogen in all its doses produced significantly higher yields than the control. The LAI and CCC variables had a positive correlation with yield in the months of November and December. A significant difference was observed between U250 (62 Mg ha^-1) and the other treatments. The U500 dose led to a yield increase of 27% compared to U250, while the difference between U750 and U500 was 6%.

Improving the remote estimation of soil organic carbon in complex ecosystems with Sentinel-2 and GIS using Gaussian processes regression

BACKGROUND AND AIMS

The quantitative retrieval of soil organic carbon (SOC) storage, particularly for soils with a large potential for carbon sequestration, is of global interest due to its link with the carbon cycle and the mitigation of climate change. However, complex ecosystems with good soil qualities for SOC storage are poorly studied.

METHODS

The interrelation between SOC and various vegetation remote sensing drivers is understood to demonstrate the link between the carbon stored in the vegetation layer and SOC of the top soil layers. Based on the mapping of SOC in two horizons (0-30 cm and 30-60 cm) we predict SOC with high accuracy in the complex and mountainous heterogeneous páramo system in Ecuador. A large SOC database (in weight % and in Mg/ha) of 493 and 494 SOC sampling data points from 0-30 cm and 30-60 cm soil profiles, respectively, were used to calibrate GPR models using Sentinel-2 and GIS predictors (i.e., Temperature, Elevation, Soil Taxonomy, Geological Unit, Slope Length and Steepness (LS Factor), Orientation and Precipitation).

RESULTS

In the 0-30 cm soil profile, the models achieved a R² of 0.85 (SOC%) and a R² of 0.79 (SOC Mg/ha). In the 30-60 cm soil profile, models achieved a R² of 0.86 (SOC%), and a R² of 0.79 (SOC Mg/ha).

CONCLUSIONS

The used Sentinel-2 variables (FVC, CWC, LCC/C_ab, band 5 (705 nm) and SeLI index) were able to improve the estimation accuracy between 3-21% compared to previous results of the same study area. CWC emerged as the most relevant biophysical variable for SOC prediction.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s11104-022-05506-1.

Estimating soil moisture content under grassland with hyperspectral data using radiative transfer modelling and machine learning

The monitoring of soil moisture content (SMC) at very high spatial resolution (<10m) using unmanned aerial systems (UAS) is of high interest for precision agriculture and the validation of large scale SMC products. Data-driven approaches are the most common method to retrieve SMC with UAS-borne data at water limited sites over non-disturbed agricultural crops. A major disadvantage of data-driven algorithms is the limited transferability in space and time and the need of a high number of ground reference samples. Physically-based approaches are less dependent on the amount of samples and are transferable in space and time. This study explores the potential of (1) a hybrid method targeting the soil brightness factor of the PROSAIL model using a variational heteroscedastic Gaussian Processes regression (VHGPR) algorithm, and (2) a data-driven method employing VHGPR for the retrieval of SMC over three grassland sites based on UAS-borne VIS-NIR (399-1001 nm) hyperspectral data. The sites were managed by mowing (Fendt), grazing (Grosses Bruch) and irrigation (Marquardt). With these distinct local pre-conditions we aimed to identify factors that favor and limit the retrieval of SMC. The hybrid approach presented encouraging results in Marquardt (RMSE = 1.5 Vol_%, R² = 0.2). At the permanent grassland sites (Fendt, Grosses Bruch) the thatch layer jeopardized the application of the hybrid model. We identified the complex canopy structure of grassland as the main factor impacting the hybrid SMC retrieval. The data-driven approach showed high accuracy for Fendt (R² = 0.84, RMSE = 8.66) and Marquardt (R² = 0.4, RMSE = 10.52). All data-driven models build on the LAI-SMC relationship. However, this relationship was hampered by mowing (Fendt), leading to a lack of transferability in time. The alteration of plant traits by grazing prevents finding a relationship with SMC in Grosses Bruch. In Marquardt, we identified the timelag between changes in SMC and plant response as the main reason of decrease in model accuracy. Yet, the model performance is accurate in undisturbed and water-limited areas (Marquardt). The analysis points to challenges that need to be tackled in future research and opens the discussion for the development of robust models to retrieve high resolution SMC from UAS-borne remote sensing observations.

Gaussian processes retrieval of crop traits in Google Earth Engine based on Sentinel-2 top-of-atmosphere data

The unprecedented availability of optical satellite data in cloud-based computing platforms, such as Google Earth Engine (GEE), opens new possibilities to develop crop trait retrieval models from the local to the planetary scale. Hybrid retrieval models are of interest to run in these platforms as they combine the advantages of physically- based radiative transfer models (RTM) with the flexibility of machine learning regression algorithms. Previous research with GEE primarily relied on processing bottom-of-atmosphere (BOA) reflectance data, which requires atmospheric correction. In the present study, we implemented hybrid models directly into GEE for processing Sentinel-2 (S2) Level-1C (L1C) top-of-atmosphere (TOA) reflectance data into crop traits. To achieve this, a training dataset was generated using the leaf-canopy RTM PROSAIL in combination with the atmospheric model 6SV. Gaussian process regression (GPR) retrieval models were then established for eight essential crop traits namely leaf chlorophyll content, leaf water content, leaf dry matter content, fractional vegetation cover, leaf area index (LAI), and upscaled leaf variables (i.e., canopy chlorophyll content, canopy water content and canopy dry matter content). An important pre-requisite for implementation into GEE is that the models are sufficiently light in order to facilitate efficient and fast processing. Successful reduction of the training dataset by 78% was achieved using the active learning technique Euclidean distance-based diversity (EBD). With the EBD-GPR models, highly accurate validation results of LAI and upscaled leaf variables were obtained against in situ field data from the validation study site Munich-North-Isar (MNI), with normalized root mean square errors (NRMSE) from 6% to 13%. Using an independent validation dataset of similar crop types (Italian Grosseto test site), the retrieval models showed moderate to good performances for canopy-level variables, with NRMSE ranging from 14% to 50%, but failed for the leaf-level estimates. Obtained maps over the MNI site were further compared against Sentinel-2 Level 2 Prototype Processor (SL2P) vegetation estimates generated from the ESA Sentinels' Application Platform (SNAP) Biophysical Processor, proving high consistency of both retrievals (R ² from 0.80 to 0.94). Finally, thanks to the seamless GEE processing capability, the TOA-based mapping was applied over the entirety of Germany at 20 m spatial resolution including information about prediction uncertainty. The obtained maps provided confidence of the developed EBD-GPR retrieval models for integration in the GEE framework and national scale mapping from S2-L1C imagery. In summary, the proposed retrieval workflow demonstrates the possibility of routine processing of S2 TOA data into crop traits maps at any place on Earth as required for operational agricultural applications.

Hybrid retrieval of crop traits from multi-temporal PRISMA hyperspectral imagery

The recently launched and upcoming hyperspectral satellite missions, featuring contiguous visible-to-shortwave infrared spectral information, are opening unprecedented opportunities for the retrieval of a broad set of vegetation traits with enhanced accuracy through novel retrieval schemes. In this framework, we exploited hyperspectral data cubes collected by the new-generation PRecursore IperSpettrale della Missione Applicativa (PRISMA) satellite of the Italian Space Agency to develop and test a hybrid retrieval workflow for crop trait mapping. Crop traits were mapped over an agricultural area in north-east Italy (Jolanda di Savoia, FE) using PRISMA images collected during the 2020 and 2021 vegetative seasons. Leaf chlorophyll content, leaf nitrogen content, leaf water content and the corresponding canopy level traits scaled through leaf area index were estimated using a hybrid retrieval scheme based on PROSAIL-PRO radiative transfer simulations coupled with a Gaussian processes regression algorithm. Active learning algorithms were used to optimise the initial set of simulated data by extracting only the most informative samples. The accuracy of the proposed retrieval scheme was evaluated against a broad ground dataset collected in 2020 in correspondence of three PRISMA overpasses. The results obtained were positive for all the investigated variables. At the leaf level, the highest accuracy was obtained for leaf nitrogen content (LNC: r²=0.87, nRMSE=7.5%), while slightly worse results were achieved for leaf chlorophyll content (LCC: r²=0.67, nRMSE=11.7%) and leaf water content (LWC: r²=0.63, nRMSE=17.1%). At the canopy level, a significantly higher accuracy was observed for nitrogen content (CNC: r²=0.92, nRMSE=5.5%) and chlorophyll content (CCC: r²=0.82, nRMSE=10.2%), whereas comparable results were obtained for water content (CWC: r²=0.61, nRMSE=16%). The developed models were additionally tested against an independent dataset collected in 2021 to evaluate their robustness and exportability. The results obtained (i. e., LCC: r²=0.62, nRMSE=27.9%; LNC: r²=0.35, nRMSE=28.4%; LWC: r²=0.74, nRMSE=20.4%; LAI: r²=0.84, nRMSE=14.5%; CCC: r²=0.79, nRMSE=18.5%; CNC: r²=0.62, nRMSE=23.7%; CWC: r²=0.92, nRMSE=16.6%) evidence the transferability of the hybrid approach optimised through active learning for most of the investigated traits. The developed models were then used to map the spatial and temporal variability of the crop traits from the PRISMA images. The high accuracy and consistency of the results demonstrates the potential of spaceborne imaging spectroscopy for crop monitoring, paving the path towards routine retrievals of multiple crop traits over large areas that could drive more effective and sustainable agricultural practices worldwide.

Evaluation of Hybrid Models to Estimate Chlorophyll and Nitrogen Content of Maize Crops in the Framework of the Future CHIME Mission

In the next few years, the new Copernicus Hyperspectral Imaging Mission (CHIME) is foreseen to be launched by the European Space Agency (ESA). This missions will provide an unprecedented amount of hyperspectral data, enabling new research possibilities within several fields of natural resources, including the “agriculture and food security” domain. In order to efficiently exploit this upcoming hyperspectral data stream, new processing methods and techniques need to be studied and implemented. In this work, the hybrid approach (HYB) and its variant, featuring sampling dimensionality reduction through active learning heuristics (HAL), were applied to CHIME-like data to evaluate the retrieval of crop traits, such as chlorophyll and nitrogen content at both leaf (LCC and LNC) and canopy level (CCC and CNC). The results showed that HYB was able to provide reliable estimations at canopy level (R² = 0.79, RMSE = 0.38 g m⁻² for CCC and R² = 0.84, RMSE = 1.10 g m⁻² for CNC) but failed at leaf level. The HAL approach improved retrieval accuracy at canopy level (best metric: R² = 0.88 and RMSE = 0.21 g m⁻² for CCC; R² = 0.93 and RMSE = 0.71 g m⁻² for CNC), providing good results also at leaf level (best metrics: R² = 0.72 and RMSE = 3.31 μg cm⁻² for LCC; R² = 0.56 and RMSE = 0.02 mg cm⁻² for LNC). The promising results obtained through the hybrid approach support the feasibility of an operational retrieval of chlorophyll and nitrogen content, e.g., in the framework of the future CHIME mission. However, further efforts are required to investigate the approach across different years, sites and crop types in order to improve its transferability to other contexts.

Quantifying Fundamental Vegetation Traits over Europe Using the Sentinel-3 OLCI Catalogue in Google Earth Engine

Thanks to the emergence of cloud-computing platforms and the ability of machine learning methods to solve prediction problems efficiently, this work presents a workflow to automate spatiotemporal mapping of essential vegetation traits from Sentinel-3 (S3) imagery. The traits included leaf chlorophyll content (LCC), leaf area index (LAI), fraction of absorbed photosynthetically active radiation (FAPAR), and fractional vegetation cover (FVC), being fundamental for assessing photosynthetic activity on Earth. The workflow involved Gaussian process regression (GPR) algorithms trained on top-of-atmosphere (TOA) radiance simulations generated by the coupled canopy radiative transfer model (RTM) SCOPE and the atmospheric RTM 6SV. The retrieval models, named to S3-TOA-GPR-1.0, were directly implemented in Google Earth Engine (GEE) to enable the quantification of the traits from TOA data as acquired from the S3 Ocean and Land Colour Instrument (OLCI) sensor.Following good to high theoretical validation results with normalized root mean square error (NRMSE) ranging from 5% (FAPAR) to 19% (LAI), a three fold evaluation approach over diverse sites and land cover types was pursued: (1) temporal comparison against LAI and FAPAR products obtained from Moderate Resolution Imaging Spectroradiometer (MODIS) for the time window 2016-2020, (2) spatial difference mapping with Copernicus Global Land Service (CGLS) estimates, and (3) direct validation using interpolated in situ data from the VALERI network. For all three approaches, promising results were achieved. Selected sites demonstrated coherent seasonal patterns compared to LAI and FAPAR MODIS products, with differences between spatially averaged temporal patterns of only 6.59%. In respect of the spatial mapping comparison, estimates provided by the S3-TOA-GPR-1.0 models indicated highest consistency with FVC and FAPAR CGLS products. Moreover, the direct validation of our S3-TOA-GPR-1.0 models against VALERI estimates indicated with regard to jurisdictional claims in good retrieval performance for LAI, FAPAR and FVC. We conclude that our retrieval workflow of spatiotemporal S3 TOA data processing into GEE opens the path towards global monitoring of fundamental vegetation traits, accessible to the whole research community.

Intelligent Sampling for Vegetation Nitrogen Mapping Based on Hybrid Machine Learning Algorithms

Upcoming satellite imaging spectroscopy missions will deliver spatiotemporal explicit data streams to be exploited for mapping vegetation properties, such as nitrogen (N) content. Within retrieval workflows for real-time mapping over agricultural regions, such crop-specific information products need to be derived precisely and rapidly. To allow fast processing, intelligent sampling schemes for training databases should be incorporated to establish efficient machine learning (ML) models. In this study, we implemented active learning (AL) heuristics using kernel ridge regression (KRR) to minimize and optimize a training database for variational heteroscedastic Gaussian processes regression (VHGPR) to estimate aboveground N content. Several uncertainty and diversity criteria were applied on a lookup table (LUT) composed of aboveground N content and corresponding hyperspectral reflectance simulated by the PROSAIL-PRO model. The best-performing AL criteria were Euclidian distance-based diversity (EBD) resulting in a reduction of the LUT training data set by 81% (50 initial samples plus 141 samples selected from a pool of 1000 samples). This reduced LUT was used for training VHGPR, which is not only a competitive algorithm but also provides uncertainty estimates. Validation against in situ N reference data provided excellent results with a root-mean-square error (RMSE) of 1.84 g/m² and a coefficient of determination (R² ) of 0.92. Mapping aboveground N content over an agricultural region yielded reliable estimates and meaningful associated uncertainties. These promising results encourage the transfer of such hybrid workflows into space and time within the frame of future operational N monitoring from satellite imaging spectroscopy data.

Assessing Non-Photosynthetic Cropland Biomass from Spaceborne Hyperspectral Imagery

Non-photosynthetic vegetation (NPV) biomass has been identified as a priority variable for upcoming spaceborne imaging spectroscopy missions, calling for a quantitative estimation of lignocellulosic plant material as opposed to the sole indication of surface coverage. Therefore, we propose a hybrid model for the retrieval of non-photosynthetic cropland biomass. The workflow included coupling the leaf optical model PROSPECT-PRO with the canopy reflectance model 4SAIL, which allowed us to simulate NPV biomass from carbon-based constituents (CBC) and leaf area index (LAI). PROSAIL-PRO provided a training database for a Gaussian process regression (GPR) algorithm, simulating a wide range of non-photosynthetic vegetation states. Active learning was employed to reduce and optimize the training data set. In addition, we applied spectral dimensionality reduction to condense essential information of non-photosynthetic signals. The resulting NPV-GPR model was successfully validated against soybean field data with normalized root mean square error (nRMSE) of 13.4% and a coefficient of determination (R2) of 0.85. To demonstrate mapping capability, the NPV-GPR model was tested on a PRISMA hyperspectral image acquired over agricultural areas in the North of Munich, Germany. Reliable estimates were mainly achieved over senescent vegetation areas as suggested by model uncertainties. The proposed workflow is the first step towards the quantification of non-photosynthetic cropland biomass as a next-generation product from near-term operational missions, such as CHIME.

Vegetation Types Mapping Using Multi-Temporal Landsat Images in the Google Earth Engine Platform

Vegetation Types (VTs) are important managerial units, and their identification serves as essential tools for the conservation of land covers. Despite a long history of Earth observation applications to assess and monitor land covers, the quantitative detection of sparse VTs remains problematic, especially in arid and semiarid areas. This research aimed to identify appropriate multi-temporal datasets to improve the accuracy of VTs classification in a heterogeneous landscape in Central Zagros, Iran. To do so, first the Normalized Difference Vegetation Index (NDVI) temporal profile of each VT was identified in the study area for the period of 2018, 2019, and 2020. This data revealed strong seasonal phenological patterns and key periods of VTs separation. It led us to select the optimal time series images to be used in the VTs classification. We then compared single-date and multi-temporal datasets of Landsat 8 images within the Google Earth Engine (GEE) platform as the input to the Random Forest classifier for VTs detection. The single-date classification gave a median Overall Kappa (OK) and Overall Accuracy (OA) of 51% and 64%, respectively. Instead, using multi-temporal images led to an overall kappa accuracy of 74% and an overall accuracy of 81%. Thus, the exploitation of multi-temporal datasets favored accurate VTs classification. In addition, the presented results underline that available open access cloud-computing platforms such as the GEE facilitates identifying optimal periods and multitemporal imagery for VTs classification.

Urban morphology detection and it's linking with land surface temperature: A case study for Tehran Metropolis, Iran

Expansion of urban areas and alteration of natural land cover exacerbate the local climate change. To find out the effect of land cover changes on the local climate, in this study, the Local Climate Zone (LCZ) concept was utilized to detect urban morphology in Tehran Metropolis. LCZ and Land Surface Temperature (LST) can be identified and classified based on available remote sensing products. Firstly, LCZ maps of Tehran metropolis were extracted using Landsat imagery, and secondly, relationships between LCZ and LST were explored for three years (1990, 2004, and 2018). We found that Tehran urban structure has 13 LCZs based on imagery from Landsat 5 and 14 LCZs based on images for Landsat 7 and 8. Overall accuracy and kappa coefficient were estimated at 62% and 0.60, respectively. Results show that built-up classes including compact high-rise, compact mid-rise, and heavy industrial areas tended to increase the surface temperature, while except for bare land, all other land cover types tended to decrease the surface temperature. The findings also suggest that complementary optical and thermal remote sensing data, such as the combination of OLI with TIRS imageries, were sufficient for supervised LCZ and LST classification in a semi-arid region of Tehran metropolis.

Emulation of Sun-Induced Fluorescence from Radiance Data Recorded by the HyPlant Airborne Imaging Spectrometer

The retrieval of sun-induced fluorescence (SIF) from hyperspectral radiance data grew to maturity with research activities around the FLuorescence EXplorer satellite mission FLEX, yet full-spectrum estimation methods such as the spectral fitting method (SFM) are computationally expensive. To bypass this computational load, this work aims to approximate the SFM-based SIF retrieval by means of statistical learning, i.e., emulation. While emulators emerged as fast surrogate models of simulators, the accuracy-speedup trade-offs are still to be analyzed when the emulation concept is applied to experimental data. We evaluated the possibility of approximating the SFM-like SIF output directly based on radiance data while minimizing the loss in precision as opposed to SFM-based SIF. To do so, we implemented a double principal component analysis (PCA) dimensionality reduction, i.e., in both input and output, to achieve emulation of multispectral SIF output based on hyperspectral radiance data. We then evaluated systematically: (1) multiple machine learning regression algorithms, (2) number of principal components, (3) number of training samples, and (4) quality of training samples. The best performing SIF emulator was then applied to a HyPlant flight line containing at sensor radiance information, and the results were compared to the SFM SIF map of the same flight line. The emulated SIF map was quasi-instantaneously generated, and a good agreement against the reference SFM map was obtained with a R² of 0.88 and NRMSE of 3.77%. The SIF emulator was subsequently applied to 7 HyPlant flight lines to evaluate its robustness and portability, leading to a R² between 0.68 and 0.95, and a NRMSE between 6.42% and 4.13%. Emulated SIF maps proved to be consistent while processing time was in the order of 3 min. In comparison, the original SFM needed approximately 78 min to complete the SIF processing. Our results suggest that emulation can be used to efficiently reduce computational loads of SIF retrieval methods.

Estimating the phenological dynamics of irrigated rice leaf area index using the combination of PROSAIL and Gaussian Process Regression

The growth of rice is a sequence of three different phenological phases. This sequence of change in rice phenology implies that the condition of the plant during the vegetative phase relates directly to the health of leaves functioning during the reproductive and ripening phases. As such, accurate monitoring is important towards understanding rice growth dynamics. Leaf Area Index (LAI) is an important indicator of rice yields and the availability of this information during key phenological phases can support more informed farming decisions. Satellite remote sensing has been adopted as a proxy to field measurements of LAI and with the launch of freely available high resolution Satellite images such as Sentinel-2, it is imperative that accurate retrieval methods are adopted towards monitoring LAI at irrigated rice fields. Here, we evaluate the potential of a hybrid radiative transfer model (i.e., PROSAIL - Gaussian Process Regression (GPR), for estimating the phenological dynamics of irrigated rice LAI using imager derived from the Sentinel-2 multispectral instrument. LAI field measurements were obtained from an experimental rice field in Nasarawa state, Nigeria during the dry season. We used the PROSAIL radiative transfer model to create a look up table (LUT) that was subsequently used to train a GPR model. Afterwards, we evaluated the potential of the hybrid modelling approach by assessing the overall model accuracy and the extent to which LAI was able to accurately predict LAI during key rice phenological phases. We compared the predicted hybrid GPR LAI values with LAI values generated from the SNAP toolbox, based on a hybrid Artificial Neural Network (ANN) modelling approach. Our results show that the overall predictive accuracy of the hybrid GPR model (R2 = 0.82, RMSE = 1.65) was more accurate than that of the hybrid ANN model (R2 = 0.66, RMSE = 3.89) for retrieving LAI values from Sentinel-2 imagery. Both models underestimated LAI values during the reproductive and ripening phases . However, the accuracy during the phenological phases were more significant when using the hybrid GPR model (P < 0.05). During the different phenological phases, the hybrid GPR model predicted LAI more accurately during the reproductive (R² = 0.7) and ripening (R² = 0.59) phases compared to the hybrid ANN reproductive and ripening phases. When monitoring LAI phenological profiles of both hybrid models, the hybrid GPR and ANN models underestimated LAI during the reproductive and ripening phases. However, the ANN model underestimations were statistically significantly greater than those for the hybrid GPR model (P < 0.05). Our results highlight the potential of hybrid GPR models for estimating the phenological dynamics of irrigated rice LAI from Sentinel-2 data. They provided more accurate estimation of LAI patterns from varying nitrogen and water applications than hybrid ANN models.

Classification of Plant Ecological Units in Heterogeneous Semi-Steppe Rangelands: Performance Assessment of Four Classification Algorithms

Plant Ecological Unit’s (PEUs) are the abstraction of vegetation communities that occur on a site which similarly respond to management actions and natural disturbances. Identification and monitoring of PEUs in a heterogeneous landscape is the most difficult task in medium resolution satellite images datasets. The main objective of this study is to compare pixel-based classification versus object-based classification for accurately classifying PEUs with four selected different algorithms across heterogeneous rangelands in Central Zagros, Iran. We used images of Landsat-8 OLI that were pan-sharpened to 15 m to classify four PEU classes based on a random dataset collected in the field (40%). In the first stage, we applied the following classification algorithms to distinguish PEUs: Minimum Distance (MD), Maximum Likelihood Classification (MLC), Neural Network-Multi Layer Perceptron (NN-MLP) and Classification Tree Analysis (CTA) for pixel based method and object based method. Then, by using the most accurate classification approach, in the second stage auxiliary data (Principal Component Analysis (PCA)) was incorporated to improve the accuracy of the PEUs classification process. At the end, test data (60%) were used for accuracy assessment of the resulting maps. Object-based maps clearly outperformed pixel-based maps, especially with CTA, NN-MLP and MD algorithms with overall accuracies of 86%, 72% and 59%, respectively. The MLC algorithm did not reveal any significant difference between the object-based and pixel-based analyses. Finally, complementing PCA auxiliary bands to the CTA algorithms offered the most successful PEUs classification strategy, with the highest overall accuracy (89%). The results clearly underpin the importance of object-based classification with the CTA classifier together with PCA auxiliary data to optimize identification of PEU classes.

Mapping landscape canopy nitrogen content from space using PRISMA data

Satellite imaging spectroscopy for terrestrial applications is reaching maturity with recently launched and upcoming science-driven missions, e.g. PRecursore IperSpettrale della Missione Applicativa (PRISMA) and Environmental Mapping and Analysis Program (EnMAP), respectively. Moreover, the high-priority mission candidate Copernicus Hyperspectral Imaging Mission for the Environment (CHIME) is expected to globally provide routine hyperspectral observations to support new and enhanced services for, among others, sustainable agricultural and biodiversity management. Thanks to the provision of contiguous visible-to-shortwave infrared spectral data, hyperspectral missions open enhanced opportunities for the development of new-generation retrieval models of multiple vegetation traits. Among these, canopy nitrogen content (CNC) is one of the most promising variables given its importance for agricultural monitoring applications. This work presents the first hybrid CNC retrieval model for the operational delivery from spaceborne imaging spectroscopy data. To achieve this, physically-based models were combined with machine learning regression algorithms and active learning (AL). The key concepts involve: (1) coupling the radiative transfer models PROSPECT-PRO and SAIL for the generation of a wide range of vegetation states as training data, (2) using dimensionality reduction to deal with collinearity, (3) applying an AL technique in combination with Gaussian process regression (GPR) for fine-tuning the training dataset on in field collected data, and (4) adding non-vegetated spectra to enable the model to deal with spectral heterogeneity in the image. The final CNC model was successfully validated against field data achieving a low root mean square error (RMSE) of 3.4 g/m² and coefficient of determination (R ²) of 0.7. The model was applied to a PRISMA image acquired over agricultural areas in the North of Munich, Germany. Mapping aboveground CNC yielded reliable estimates over the whole landscape and meaningful associated uncertainties. These promising results demonstrate the feasibility of routinely quantifying CNC from space, such as in an operational context as part of the future CHIME mission.

Crop Nitrogen Retrieval Methods for Simulated Sentinel-2 Data Using In-Field Spectrometer Data

Nitrogen (N) is one of the key nutrients supplied in agricultural production worldwide. Over-fertilization can have negative influences on the field and the regional level (e.g., agro-ecosystems). Remote sensing of the plant N of field crops presents a valuable tool for the monitoring of N flows in agro-ecosystems. Available data for validation of satellite-based remote sensing of N is scarce. Therefore, in this study, field spectrometer measurements were used to simulate data of the Sentinel-2 (S2) satellites developed for vegetation monitoring by the ESA. The prediction performance of normalized ratio indices (NRIs), random forest regression (RFR) and Gaussian processes regression (GPR) for plant-N-related traits was assessed on a diverse real-world dataset including multiple crops, field sites and years. The plant N traits included the mass-based N measure, N concentration in the biomass (Nconc), and an area-based N measure approximating the plant N uptake (NUP). Spectral indices such as normalized ratio indices (NRIs) performed well, but the RFR and GPR methods outperformed the NRIs. Key spectral bands for each trait were identified using the RFR variable importance measure and the Gaussian processes regression band analysis tool (GPR-BAT), highlighting the importance of the short-wave infrared (SWIR) region for estimation of plant Nconc-and to a lesser extent the NUP. The red edge (RE) region was also important. The GPR-BAT showed that five bands were sufficient for plant N trait and leaf area index (LAI) estimation and that a surplus of bands effectively reduced prediction performance. A global sensitivity analysis (GSA) was performed on all traits simultaneously, showing the dominance of the LAI in the mixed remote sensing signal. To delineate the plant-N-related traits from this signal, regional and/or national data collection campaigns producing large crop spectral libraries (CSL) are needed. An improved database will likely enable the mapping of N at the agro-ecosystem level or for use in precision farming by farmers in the future.

Comparison of Crop Trait Retrieval Strategies Using UAV-Based VNIR Hyperspectral Imaging

Hyperspectral cameras onboard unmanned aerial vehicles (UAVs) have recently emerged for monitoring crop traits at the sub-field scale. Different physical, statistical, and hybrid methods for crop trait retrieval have been developed. However, spectra collected from UAVs can be confounded by various issues, including illumination variation throughout the crop growing season, the effect of which on the retrieval performance is not well understood at present. In this study, four retrieval methods are compared, in terms of retrieving the leaf area index (LAI), fractional vegetation cover (fCover), and canopy chlorophyll content (CCC) of potato plants over an agricultural field for six dates during the growing season. We analyzed: (1) The standard look-up table method (LUTstd), (2) an improved (regularized) LUT method that involves variable correlation (LUTreg), (3) hybrid methods, and (4) random forest regression without (RF) and with (RFexp) the exposure time as an additional explanatory variable. The Soil–Leaf–Canopy (SLC) model was used in association with the LUT-based inversion and hybrid methods, while the statistical modelling methods (RF and RFexp) relied entirely on in situ data. The results revealed that RFexp was the best-performing method, yielding the highest accuracies, in terms of the normalized root mean square error (NRMSE), for LAI (5.36%), fCover (5.87%), and CCC (15.01%). RFexp was able to reduce the effects of illumination variability and cloud shadows. LUTreg outperformed the other two retrieval methods (hybrid methods and LUTstd), with an NRMSE of 9.18% for LAI, 10.46% for fCover, and 12.16% for CCC. Conversely, LUTreg led to lower accuracies than those derived from RF for LAI (5.51%) and for fCover (6.23%), but not for CCC (16.21%). Therefore, the machine learning approaches—in particular, RF—appear to be the most promising retrieval methods for application to UAV-based hyperspectral data.

Top-of-Atmosphere Retrieval of Multiple Crop Traits Using Variational Heteroscedastic Gaussian Processes within a Hybrid Workflow

In support of cropland monitoring, operational Copernicus Sentinel-2 (S2) data became available globally and can be explored for the retrieval of important crop traits. Based on a hybrid workflow, retrieval models for six essential biochemical and biophysical crop traits were developed for both S2 bottom-of-atmosphere (BOA) L2A and S2 top-of-atmosphere (TOA) L1C data. A variational heteroscedastic Gaussian process regression (VHGPR) algorithm was trained with simulations generated by the combined leaf-canopy reflectance model PROSAILat the BOA scale and further combined with the Second Simulation of a Satellite Signal in the Solar Spectrum (6SV) atmosphere model at the TOA scale. Established VHGPR models were then applied to S2 L1C and L2A reflectance data for mapping: leaf chlorophyll content (C_ab), leaf water content (C_w), fractional vegetation coverage (FVC), leaf area index (LAI), and upscaled leaf biochemical compounds, i.e., LAI * C_ab (laiCab) and LAI * C_w (laiCw). Estimated variables were validated using in situ reference data collected during the Munich-North-Isar field campaigns within growing seasons of maize and winter wheat in the years 2017 and 2018. For leaf biochemicals, retrieval from BOA reflectance slightly outperformed results from TOA reflectance, e.g., obtaining a root mean squared error (RMSE) of 6.5 μg/cm² (BOA) vs. 8 μg/cm² (TOA) in the case of C_ab. For the majority of canopy-level variables, instead, estimation accuracy was higher when using TOA reflectance data, e.g., with an RMSE of 139 g/m² (BOA) vs. 113 g/m² (TOA) for laiCw. Derived maps were further compared against reference products obtained from the ESA Sentinel Application Platform (SNAP) Biophysical Processor. Altogether, the consistency between L1C and L2A retrievals confirmed that crop traits can potentially be estimated directly from TOA reflectance data. Successful mapping of canopy-level crop traits including information about prediction confidence suggests that the models can be transferred over spatial and temporal scales and, therefore, can contribute to decision-making processes for cropland management.

Retrieving and Validating Leaf and Canopy Chlorophyll Content at Moderate Resolution: A Multiscale Analysis with the Sentinel-3 OLCI Sensor

ESA’s Eighth Earth Explorer mission “FLuorescence EXplorer” (FLEX) will be dedicated to the global monitoring of the chlorophyll fluorescence emitted by vegetation. In order to properly interpret the measured fluorescence signal, essential vegetation variables need to be retrieved concomitantly. FLEX will fly in tandem formation with Sentinel-3 (S3), which conveys the Ocean and Land Color Instrument (OLCI) that is designed to characterize the atmosphere and the terrestrial vegetation at a spatial resolution of 300 m. In support of FLEX’s preparatory activities, this paper presents a first validation exercise of OLCI vegetation products against in situ data coming from the 2018 FLEXSense campaign. During this campaign, leaf chlorophyll content (LCC) and leaf area index (LAI) measurements were collected over croplands, while HyPlant DUAL images of the area were acquired at a 3 m spatial resolution. A multiscale validation strategy was pursued. First, estimates of these two variables, together with the combined canopy chlorophyll content (CCC = LCC × LAI), were obtained at the HyPlant spatial resolution and were compared against the in situ measurements. Second, the fine-scale retrieval maps from HyPlant were coarsened to the S3 spatial scale as a reference to assess the quality of the OLCI vegetation products. As an intermediary step, vegetation products extracted from Sentinel-2 data were used to compare retrievals at the in-between spatial resolution of 20 m. For all spatial scales, CCC delivered the most accurate estimates with the smallest prediction error obtained at the 300 m resolution (R2 of 0.74 and RMSE = 26.8 μg cm−2). Results of a scaling analysis suggest that CCC performs well at the different tested spatial resolutions since it presents a linear behavior across scales. LCC, on the other hand, was poorly retrieved at the 300 m scale, showing overestimated values over heterogeneous pixels. The introduction of a new LCC model integrating mixed reflectance spectra in its training enabled to improve by 16% the retrieval accuracy for this variable (RMSE = 10 μg cm−2 for the new model versus RMSE = 11.9 μg cm−2 for the former model).

Hybrid inversion of radiative transfer models based on high spatial resolution satellite reflectance data improves fractional vegetation cover retrieval in heterogeneous ecological systems after fire

In forest landscapes affected by fire, the estimation of fractional vegetation cover (FVC) from remote sensing data using radiative transfer models (RTMs) enables to evaluate the ecological impact of such disturbance across plant communities at different spatio-temporal scales. Even though, when landscapes are highly heterogeneous, the fine-scale ground spatial variation might not be properly captured if FVC products are provided at moderate or coarse spatial scales, as typical of most of operational Earth observing satellite missions. The objective of this study was to evaluate the potential of a RTM inversion approach for estimating FVC from satellite reflectance data at high spatial resolution as compared to the standard use of coarser imagery. The study was conducted both at landscape and plant community levels within the perimeter of a megafire that occurred in western Mediterranean Basin. We developed a hybrid retrieval scheme based on PROSAIL-D RTM simulations to create a training dataset of top-of-canopy spectral reflectance and the corresponding FVC for the dominant plant communities. The machine learning algorithm Gaussian Processes Regression (GPR) was learned on the training dataset to model the relationship between canopy reflectance and FVC. The GPR model was then applied to retrieve FVC from WorldView-3 (spatial resolution of 2 m) and Sentinel-2 (spatial resolution of 20 m) surface reflectance bands. A set of 75 plots of 2x2m and 45 plots of 20x20m was distributed under a stratified schema across the focal plant communities within the fire perimeter to validate FVC satellite derived retrieval. At landscape scale, the accuracy of the FVC retrieval was substantially higher from WorldView-3 (R² = 0.83; RMSE = 7.92%) than from Sentinel-2 (R² = 0.73; RMSE = 11.89%). At community level, FVC retrieval was more accurate for oak forests than for heathlands and broomlands. The retrieval from WorldView-3 minimized the over- and under-estimation effects at low and high field sampled vegetation cover, respectively. These findings emphasize the effectiveness of high spatial resolution satellite reflectance data to capture FVC ground spatial variability in heterogeneous burned areas using a hybrid RTM retrieval method.

Prototyping Sentinel-2 green LAI and brown LAI products for cropland monitoring

For agricultural applications, identification of non-photosynthetic above-ground vegetation is of great interest as it contributes to assess harvest practices, detecting crop residues or drought events, as well as to better predict the carbon, water and nutrients uptake. While the mapping of green Leaf Area Index (LAI) is well established, current operational retrieval models are not calibrated for LAI estimation over senescent, brown vegetation. This not only leads to an underestimation of LAI when crops are ripening, but is also a missed monitoring opportunity. The high spatial and temporal resolution of Sentinel-2 (S2) satellites constellation offers the possibility to estimate brown LAI (LAI _G ) next to green LAI (LAI _G ). By using LAI ground measurements from multiple campaigns associated with airborne or satellite spectra, Gaussian processes regression (GPR) models were developed for both LAI _G and LAI _B , providing alongside associated uncertainty estimates, which allows to mask out unreliable LAI retrievals with higher uncertainties. A processing chain was implemented to apply both models to S2 images, generating a multiband LAI product at 20 m spatial resolution. The models were adequately validated with in-situ data from various European study sites (LAI _G : R² = 0.7, RMSE = 0.67 m²/m²; LAI _B : R² = 0.62, RMSE = 0.43 m²/m²). Thanks to the S2 bands in the red edge (B5: 705 nm and B6: 740 nm) and in the shortwave infrared (B12: 2190 nm) a distinction between LAI _G and LAI _B can be achieved. To demonstrate the capability of LAI _B to identify when crops start senescing, S2 time series were processed over multiple European study sites and seasonal maps were produced, which show the onset of crop senescence after the green vegetation peak. Particularly, the LAI _B product permits the detection of harvest (i.e., sudden drop in LAI _B ) and the determination of crop residues (i.e., remaining LAI _B ), although a better spectral sampling in the shortwave infrared would have been desirable to disentangle brown LAI from soil variability and its perturbing effects. Finally, a single total LAI product was created by merging LAI _G and LAI _B estimates, and then compared to the LAI derived from S2 L2B biophysical processor integrated in SNAP. The spatiotemporal analysis results confirmed the improvement of the proposed descriptors with respect to the standard SNAP LAI product accounting only for photosynthetically active green vegetation.

Denoising AVIRIS-NG Data for Generation of New Chlorophyll Indices

The availability of Airborne Visible and Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) data has enormous possibilities for quantification of Leaf Chlorophyll Content (LCC). The present study used the AVIRIS-NG campaign site of Western India for generation and validation of new chlorophyll indices by denoising the AVIRIS-NG data. For validation, concurrent to AVIRIS-NG flight overpass, field samplings were performed. The acquired AVIRIS-NG was subjected to Spectral Angle Mapper (SAM) classifier for discriminating the crop types. Three smoothing techniques i.e., Fast-Fourier Transform (FFT), Mean and Savitzky-Golay filters were evaluated for their denoising capability. Raw and filtered data was used for developing new chlorophyll indices by optimizing AVIRIS-NG bands using VIs based on parametric regression algorithms. In total, 20 chlorophyll indices and corresponding 20 models were developed for mapping LCC in the area. SAM identified 17 crop types in the area, while FFT found to be the best for filtering. Performance of these models when checked based on Pearson correlation coefficient (r) and Centered Root Mean Square Difference (CRMSD), indicated that LCC-CCI10 based on normalized difference type index formed through Near Infrared band and blue band is the best estimator of LCC (r_cal = 0.73, r_val = 0.66, CRMSD = 4.97). The approach was also tested using AVIRIS-NG image of the year 2018, which also showed a promising correlation (r = 0.704, CRSMD = 8.98, Bias = -0.5) between modeled and field LCC.

Potential use of grapevine cv Askari for heavy metal phytoremediation purposes at greenhouse scale

Grapevine varieties possess desirable characteristics for phytoremediation purposes. We investigated the potential of grapevine cv Askari in phytoremediation of heavy metal (HM) pollutions. In total 80 grapevine seedlings were exposed to four levels of HM stress (mild, low, moderate, and severe) in greenhouse condition during two growing years (2018 and 2019). The HM concentrations (Zn, Cu, Cd, Cr, and Pb) were subsequently determined in the soils, roots, and grapevine aboveground parts (AGPs), and then phytoextraction and phytostabilization potential assessment indices, i.e., biological absorption coefficient (BAC), bioconcentration factor (BCF), and translocation factor (TF), were calculated. Results led to ranking of the cumulative concentration order of the HMs, i.e.: soils (3476.6) > AGPs (1418.8) > roots (562.2) mg/kg-DW. The mean concentrations ranking order of studied HMs were in soil: Cu (1184.8) > Pb (865.5) > Cd (803.2) > Cr (623.0) > Zn (277.9) mg/kg-DW; roots, Cu (242.0) > Cd (239.5) > Zn (188.8) > Pb (63.5) > Cr (17.2) mg/kg-DW; and AGP environments, Cu (910.2) > Cd (322.9) > Zn (160.3) > Pb (152.9) > Cr (25.3) mg/kg-DW. Principal component analysis results demonstrated the same distribution pattern for the studied HMs between soil, root, and AGP environments, and the highest correlation coefficients were found for Cu, Zn, and Cd. Based on the obtained results (Cu-BAC (> 1), Zn-BCF (> 1), Zn-TF (< 1), Cu-AGP mean concentrations (> 1000 mg/kg-DW), and Cd-AGP mean concentrations (> 100 mg/kg-DW)), it can be concluded that grapevine cv Askari possesses potentials for phytoremediation purposes of Cu, Zn, and Cd. These results were acquired in a greenhouse environment and under controlled conditions; we suggest that the phytoremediation potential of this grapevine variety be assessed in a contaminated vineyard environment as well.

Green LAI Mapping and Cloud Gap-Filling Using Gaussian Process Regression in Google Earth Engine

For the last decade, Gaussian process regression (GPR) proved to be a competitive machine learning regression algorithm for Earth observation applications, with attractive unique properties such as band relevance ranking and uncertainty estimates. More recently, GPR also proved to be a proficient time series processor to fill up gaps in optical imagery, typically due to cloud cover. This makes GPR perfectly suited for large-scale spatiotemporal processing of satellite imageries into cloud-free products of biophysical variables. With the advent of the Google Earth Engine (GEE) cloud platform, new opportunities emerged to process local-to-planetary scale satellite data using advanced machine learning techniques and convert them into gap-filled vegetation properties products. However, GPR is not yet part of the GEE ecosystem. To circumvent this limitation, this work proposes a general adaptation of GPR formulation to parallel processing framework and its integration into GEE. To demonstrate the functioning and utility of the developed workflow, a GPR model predicting green leaf area index (LAIG) from Sentinel-2 imagery was imported. Although by running this GPR model into GEE any corner of the world can be mapped into LAIG at a resolution of 20 m, here we show some demonstration cases over western Europe with zoom-ins over Spain. Thanks to the computational power of GEE, the mapping takes place on-the-fly. Additionally, a GPR-based gap filling strategy based on pre-optimized kernel hyperparameters is also put forward for the generation of multi-orbit cloud-free LAIG maps with an unprecedented level of detail, and the extraction of regularly-sampled LAIG time series at a pixel level. The ability to plugin a locally-trained GPR model into the GEE framework and its instant processing opens up a new paradigm of remote sensing image processing.

Quantifying vegetation biophysical variables from the Sentinel-3/FLEX tandem mission: Evaluation of the synergy of OLCI and FLORIS data sources

The ESA's forthcoming FLuorescence EXplorer (FLEX) mission is dedicated to the global monitoring of the vegetation's chlorophyll fluorescence by means of an imaging spectrometer, FLORIS. In order to properly interpret the fluorescence signal in relation to photosynthetic activity, essential vegetation variables need to be retrieved concomitantly. FLEX will fly in tandem with Sentinel-3 (S3), which conveys the Ocean and Land Colour Instrument (OLCI) that is designed to characterize the atmosphere and the terrestrial vegetation at a spatial resolution of 300 m. In this work we present the retrieval models of four essential biophysical variables: (1) Leaf Area Index (LAI), (2) leaf chlorophyll content (Cab), (3) fraction of absorbed photosynthetically active radiation (fAPAR), and (4) fractional vegetation cover (FCover). These variables can be operationally inferred by hybrid retrieval approaches, which combine the generalization capabilities offered by radiative transfer models (RTMs) with the flexibility and computational efficiency of machine learning methods. The RTM SCOPE (Soil Canopy Observation, Photochemistry and Energy fluxes) was used to generate a database of reflectance spectra corresponding to a large variety of canopy realizations, which served subsequently as input to train a Gaussian Process Regression (GPR) algorithm for each targeted variable. Three sets of GPR models were developed, based on different spectral band settings: (1) OLCI (21 bands between 400 and 1040 nm), (2) FLORIS (281 bands between 500 and 780 nm), and (3) their synergy. Their respective performances were assessed based on simulated reflectance scenes. Regarding the retrieval of Cab, the OLCI model gave good model performances (R²: 0.91; RMSE: 7.6 μg. cm ^-2), yet superior accuracies were achieved as a result of FLORIS' higher spectral resolution (R²: 0.96; RMSE: 4.8 μg. cm ^-2). The synergy of both datasets did not further enhance the variable retrieval. Regarding LAI, the improvement of the model performances by using only FLORIS spectra (R²: 0.87; RMSE: 1.05 m².m^-2) rather than only OLCI spectra (R²: 0.86; RMSE: 1.12 m².m^-2) was less evident but merging both data sets was more beneficial (R²: 0.88; RMSE: 1.01 m².m^-2). Finally, the three data sources gave good model performances for the retrieval of fAPAR and Fcover, with the best performing model being the Synergy model (fAPAR: R²: 0.99; RMSE: 0.02 and FCover: R²: 0.98; RMSE: 0.04). The ability of the models to process real data was subsequently demonstrated by applying the OLCI models to S3 surface reflectance products acquired over Western Europe and Argentina. Obtained maps showed consistent patterns and variable ranges, and comparison against corresponding Sentinel-2 products (coarsened to a 300 m spatial resolution) led to reasonable matches (R²: 0.5-0.7). Altogether, given the availability of the multiple data sources, the FLEX tandem mission will foster unique opportunities to quantify essential vegetation properties, and hence facilitate the interpretation of the measured fluorescence levels.

Retrieval of aboveground crop nitrogen content with a hybrid machine learning method

Hyperspectral acquisitions have proven to be the most informative Earth observation data source for the estimation of nitrogen (N) content, which is the main limiting nutrient for plant growth and thus agricultural production. In the past, empirical algorithms have been widely employed to retrieve information on this biochemical plant component from canopy reflectance. However, these approaches do not seek for a cause-effect relationship based on physical laws. Moreover, most studies solely relied on the correlation of chlorophyll content with nitrogen, and thus neglected the fact that most N is bound in proteins. Our study presents a hybrid retrieval method using a physically-based approach combined with machine learning regression to estimate crop N content. Within the workflow, the leaf optical properties model PROSPECT-PRO including the newly calibrated specific absorption coefficients (SAC) of proteins, was coupled with the canopy reflectance model 4SAIL to PROSAIL-PRO. The latter was then employed to generate a training database to be used for advanced probabilistic machine learning methods: a standard homoscedastic Gaussian process (GP) and a heteroscedastic GP regression that accounts for signal-to-noise relations. Both GP models have the property of providing confidence intervals for the estimates, which sets them apart from other machine learners. Moreover, a GP-based sequential backward band removal algorithm was employed to analyze the band-specific information content of PROSAIL-PRO simulated spectra for the estimation of aboveground N. Data from multiple hyperspectral field campaigns, carried out in the framework of the future satellite mission Environmental Mapping and Analysis Program (EnMAP), were exploited for validation. In these campaigns, corn and winter wheat spectra were acquired to simulate spectral EnMAP data. Moreover, destructive N measurements from leaves, stalks and fruits were collected separately to enable plant-organ-specific validation. The results showed that both GP models can provide accurate aboveground N simulations, with slightly better results of the heteroscedastic GP in terms of model testing and against in situ N measurements from leaves plus stalks, with root mean square error (RMSE) of 2.1 g/m². However, the inclusion of fruit N content for validation deteriorated the results, which can be explained by the inability of the radiation to penetrate the thick tissues of stalks, corn cobs and wheat ears. GP-based band analysis identified optimal spectral settings with ten bands mainly situated in the shortwave infrared (SWIR) spectral region. Use of well-known protein absorption bands from the literature showed comparative results. Finally, the heteroscedastic GP model was successfully applied on airborne hyperspectral data for N mapping. We conclude that GP algorithms, and in particular the heteroscedastic GP, should be implemented for global agricultural monitoring of aboveground N from future imaging spectroscopy data.

Retrieval of aboveground crop nitrogen content with a hybrid machine learning method

Hyperspectral acquisitions have proven to be the most informative Earth observation data source for the estimation of nitrogen (N) content, which is the main limiting nutrient for plant growth and thus agricultural production. In the past, empirical algorithms have been widely employed to retrieve information on this biochemical plant component from canopy reflectance. However, these approaches do not seek for a cause-effect relationship based on physical laws. Moreover, most studies solely relied on the correlation of chlorophyll content with nitrogen, and thus neglected the fact that most N is bound in proteins. Our study presents a hybrid retrieval method using a physically-based approach combined with machine learning regression to estimate crop N content. Within the workflow, the leaf optical properties model PROSPECT-PRO including the newly calibrated specific absorption coefficients (SAC) of proteins, was coupled with the canopy reflectance model 4SAIL to PROSAIL-PRO. The latter was then employed to generate a training database to be used for advanced probabilistic machine learning methods: a standard homoscedastic Gaussian process (GP) and a heteroscedastic GP regression that accounts for signal-to-noise relations. Both GP models have the property of providing confidence intervals for the estimates, which sets them apart from other machine learners. Moreover, a GP-based sequential backward band removal algorithm was employed to analyze the band-specific information content of PROSAIL-PRO simulated spectra for the estimation of aboveground N. Data from multiple hyperspectral field campaigns, carried out in the framework of the future satellite mission Environmental Mapping and Analysis Program (EnMAP), were exploited for validation. In these campaigns, corn and winter wheat spectra were acquired to simulate spectral EnMAP data. Moreover, destructive N measurements from leaves, stalks and fruits were collected separately to enable plant-organ-specific validation. The results showed that both GP models can provide accurate aboveground N simulations, with slightly better results of the heteroscedastic GP in terms of model testing and against in situ N measurements from leaves plus stalks, with root mean square error (RMSE) of 2.1 g/m². However, the inclusion of fruit N content for validation deteriorated the results, which can be explained by the inability of the radiation to penetrate the thick tissues of stalks, corn cobs and wheat ears. GP-based band analysis identified optimal spectral settings with ten bands mainly situated in the shortwave infrared (SWIR) spectral region. Use of well-known protein absorption bands from the literature showed comparative results. Finally, the heteroscedastic GP model was successfully applied on airborne hyperspectral data for N mapping. We conclude that GP algorithms, and in particular the heteroscedastic GP, should be implemented for global agricultural monitoring of aboveground N from future imaging spectroscopy data.

Retrieval of aboveground crop nitrogen content with a hybrid machine learning method

Hyperspectral acquisitions have proven to be the most informative Earth observation data source for the estimation of nitrogen (N) content, which is the main limiting nutrient for plant growth and thus agricultural production. In the past, empirical algorithms have been widely employed to retrieve information on this biochemical plant component from canopy reflectance. However, these approaches do not seek for a cause-effect relationship based on physical laws. Moreover, most studies solely relied on the correlation of chlorophyll content with nitrogen, and thus neglected the fact that most N is bound in proteins. Our study presents a hybrid retrieval method using a physically-based approach combined with machine learning regression to estimate crop N content. Within the workflow, the leaf optical properties model PROSPECT-PRO including the newly calibrated specific absorption coefficients (SAC) of proteins, was coupled with the canopy reflectance model 4SAIL to PROSAIL-PRO. The latter was then employed to generate a training database to be used for advanced probabilistic machine learning methods: a standard homoscedastic Gaussian process (GP) and a heteroscedastic GP regression that accounts for signal-to-noise relations. Both GP models have the property of providing confidence intervals for the estimates, which sets them apart from other machine learners. Moreover, a GP-based sequential backward band removal algorithm was employed to analyze the band-specific information content of PROSAIL-PRO simulated spectra for the estimation of aboveground N. Data from multiple hyperspectral field campaigns, carried out in the framework of the future satellite mission Environmental Mapping and Analysis Program (EnMAP), were exploited for validation. In these campaigns, corn and winter wheat spectra were acquired to simulate spectral EnMAP data. Moreover, destructive N measurements from leaves, stalks and fruits were collected separately to enable plant-organ-specific validation. The results showed that both GP models can provide accurate aboveground N simulations, with slightly better results of the heteroscedastic GP in terms of model testing and against in situ N measurements from leaves plus stalks, with root mean square error (RMSE) of 2.1 g/m². However, the inclusion of fruit N content for validation deteriorated the results, which can be explained by the inability of the radiation to penetrate the thick tissues of stalks, corn cobs and wheat ears. GP-based band analysis identified optimal spectral settings with ten bands mainly situated in the shortwave infrared (SWIR) spectral region. Use of well-known protein absorption bands from the literature showed comparative results. Finally, the heteroscedastic GP model was successfully applied on airborne hyperspectral data for N mapping. We conclude that GP algorithms, and in particular the heteroscedastic GP, should be implemented for global agricultural monitoring of aboveground N from future imaging spectroscopy data.

Gaussian processes retrieval of LAI from Sentinel-2 top-of-atmosphere radiance data

Retrieval of vegetation properties from satellite and airborne optical data usually takes place after atmospheric correction, yet it is also possible to develop retrieval algorithms directly from top-of-atmosphere (TOA) radiance data. One of the key vegetation variables that can be retrieved from at-sensor TOA radiance data is leaf area index (LAI) if algorithms account for variability in atmosphere. We demonstrate the feasibility of LAI retrieval from Sentinel-2 (S2) TOA radiance data (L1C product) in a hybrid machine learning framework. To achieve this, the coupled leaf-canopy-atmosphere radiative transfer models PROSAIL-6SV were used to simulate a look-up table (LUT) of TOA radiance data and associated input variables. This LUT was then used to train the Bayesian machine learning algorithms Gaussian processes regression (GPR) and variational heteroscedastic GPR (VHGPR). PROSAIL simulations were also used to train GPR and VHGPR models for LAI retrieval from S2 images at bottom-of-atmosphere (BOA) level (L2A product) for comparison purposes. The BOA and TOA LAI products were consistently validated against a field dataset with GPR (R² of 0.78) and with VHGPR (R ² of 0.80) and for both cases a slightly lower RMSE for the TOA LAI product (about 10% reduction). Because of delivering superior accuracies and lower uncertainties, the VHGPR models were further applied for LAI mapping using S2 acquisitions over the agricultural sites Marchfeld (Austria) and Barrax (Spain). The models led to consistent LAI maps at BOA and TOA scale. The LAI maps were also compared against LAI maps as generated by the SNAP toolbox, which is based on a neural network (NN). Maps were again consistent, however the SNAP NN model tends to overestimate over dense vegetation cover. Overall, this study demonstrated that hybrid LAI retrieval algorithms can be developed from TOA radiance data given a cloud-free sky, thus without the need of atmospheric correction. To the benefit of the community, the development of such hybrid models for the retrieval vegetation properties from BOA or TOA images has been streamlined in the freely downloadable ALG-ARTMO software framework.

Gaussian processes retrieval of LAI from Sentinel-2 top-of-atmosphere radiance data

Retrieval of vegetation properties from satellite and airborne optical data usually takes place after atmospheric correction, yet it is also possible to develop retrieval algorithms directly from top-of-atmosphere (TOA) radiance data. One of the key vegetation variables that can be retrieved from at-sensor TOA radiance data is leaf area index (LAI) if algorithms account for variability in atmosphere. We demonstrate the feasibility of LAI retrieval from Sentinel-2 (S2) TOA radiance data (L1C product) in a hybrid machine learning framework. To achieve this, the coupled leaf-canopy-atmosphere radiative transfer models PROSAIL-6SV were used to simulate a look-up table (LUT) of TOA radiance data and associated input variables. This LUT was then used to train the Bayesian machine learning algorithms Gaussian processes regression (GPR) and variational heteroscedastic GPR (VHGPR). PROSAIL simulations were also used to train GPR and VHGPR models for LAI retrieval from S2 images at bottom-of-atmosphere (BOA) level (L2A product) for comparison purposes. The BOA and TOA LAI products were consistently validated against a field dataset with GPR (R² of 0.78) and with VHGPR (R ² of 0.80) and for both cases a slightly lower RMSE for the TOA LAI product (about 10% reduction). Because of delivering superior accuracies and lower uncertainties, the VHGPR models were further applied for LAI mapping using S2 acquisitions over the agricultural sites Marchfeld (Austria) and Barrax (Spain). The models led to consistent LAI maps at BOA and TOA scale. The LAI maps were also compared against LAI maps as generated by the SNAP toolbox, which is based on a neural network (NN). Maps were again consistent, however the SNAP NN model tends to overestimate over dense vegetation cover. Overall, this study demonstrated that hybrid LAI retrieval algorithms can be developed from TOA radiance data given a cloud-free sky, thus without the need of atmospheric correction. To the benefit of the community, the development of such hybrid models for the retrieval vegetation properties from BOA or TOA images has been streamlined in the freely downloadable ALG-ARTMO software framework.

Partitioning net carbon dioxide fluxes into photosynthesis and respiration using neural networks

The eddy covariance (EC) technique is used to measure the net ecosystem exchange (NEE) of CO2 between ecosystems and the atmosphere, offering a unique opportunity to study ecosystem responses to climate change. NEE is the difference between the total CO2 release due to all respiration processes (RECO), and the gross carbon uptake by photosynthesis (GPP). These two gross CO2 fluxes are derived from EC measurements by applying partitioning methods that rely on physiologically based functional relationships with a limited number of environmental drivers. However, the partitioning methods applied in the global FLUXNET network of EC observations do not account for the multiple co-acting factors that modulate GPP and RECO flux dynamics. To overcome this limitation, we developed a hybrid data-driven approach based on combined neural networks (NNC-part ). NNC-part incorporates process knowledge by introducing a photosynthetic response based on the light-use efficiency (LUE) concept, and uses a comprehensive dataset of soil and micrometeorological variables as fluxes drivers. We applied the method to 36 sites from the FLUXNET2015 dataset and found a high consistency in the results with those derived from other standard partitioning methods for both GPP (R2  > .94) and RECO (R2  > .8). High consistency was also found for (a) the diurnal and seasonal patterns of fluxes and (b) the ecosystem functional responses. NNC-part performed more realistic than the traditional methods for predicting additional patterns of gross CO2 fluxes, such as: (a) the GPP response to VPD, (b) direct effects of air temperature on GPP dynamics, (c) hysteresis in the diel cycle of gross CO2 fluxes, (d) the sensitivity of LUE to the diffuse to direct radiation ratio, and (e) the post rain respiration pulse after a long dry period. In conclusion, NNC-part is a valid data-driven approach to provide GPP and RECO estimates and complementary to the existing partitioning methods.

Crop nitrogen monitoring: Recent progress and principal developments in the context of imaging spectroscopy missions

Nitrogen (N) is considered as one of the most important plant macronutrients and proper management of N therefore is a pre-requisite for modern agriculture. Continuous satellite-based monitoring of this key plant trait would help to understand individual crop N use efficiency and thus would enable site-specific N management. Since hyperspectral imaging sensors could provide detailed measurements of spectral signatures corresponding to the optical activity of chemical constituents, they have a theoretical advantage over multi-spectral sensing for the detection of crop N. The current study aims to provide a state-of-the-art overview of crop N retrieval methods from hyperspectral data in the agricultural sector and in the context of future satellite imaging spectroscopy missions. Over 400 studies were reviewed for this purpose, identifying those estimating mass-based N (N concentration, N%) and area-based N (N content, N_area) using hyperspectral remote sensing data. Retrieval methods of the 125 studies selected in this review can be grouped into: (1) parametric regression methods, (2) linear nonparametric regression methods or chemometrics, (3) nonlinear nonparametric regression methods or machine learning regression algorithms, (4) physically-based or radiative transfer models (RTM), (5) use of alternative data sources (sun-induced fluorescence, SIF) and (6) hybrid or combined techniques. Whereas in the last decades methods for estimation of N_area and N% from hyperspectral data have been mainly based on simple parametric regression algorithms, such as narrowband vegetation indices, there is an increasing trend of using machine learning, RTM and hybrid techniques. Within plants, N is invested in proteins and chlorophylls stored in the leaf cells, with the proteins being the major nitrogen-containing biochemical constituent. However, in most studies, the relationship between N and chlorophyll content was used to estimate crop N, focusing on the visible-near infrared (VNIR) spectral domains, and thus neglecting protein-related N and reallocation of nitrogen to non-photosynthetic compartments. Therefore, we recommend exploiting the estimation of nitrogen via the proxy of proteins using hyperspectral data and in particular the short-wave infrared (SWIR) spectral domain. We further strongly encourage a standardization of nitrogen terminology, distinguishing between N% and N_area. Moreover, the exploitation of physically-based approaches is highly recommended combined with machine learning regression algorithms, which represents an interesting perspective for future research in view of new spaceborne imaging spectroscopy sensors.

DATimeS: A machine learning time series GUI toolbox for gap-filling and vegetation phenology trends detection

Optical remotely sensed data are typically discontinuous, with missing values due to cloud cover. Consequently, gap-filling solutions are needed for accurate crop phenology characterization. The here presented Decomposition and Analysis of Time Series software (DATimeS) expands established time series interpolation methods with a diversity of advanced machine learning fitting algorithms (e.g., Gaussian Process Regression: GPR) particularly effective for the reconstruction of multiple-seasons vegetation temporal patterns. DATimeS is freely available as a powerful image time series software that generates cloud-free composite maps and captures seasonal vegetation dynamics from regular or irregular satellite time series. This work describes the main features of DATimeS, and provides a demonstration case using Sentinel-2 Leaf Area Index time series data over a Spanish site. GPR resulted as an optimum fitting algorithm with most accurate gap-filling performance and associated uncertainties. DATimeS further quantified LAI fluctuations among multiple crop seasons and provided phenological indicators for specific crop types.

Optimizing Gaussian Process Regression for Image Time Series Gap-Filling and Crop Monitoring

Image processing entered the era of artificial intelligence, and machine learning algorithms emerged as attractive alternatives for time series data processing. Satellite image time series processing enables crop phenology monitoring, such as the calculation of start and end of season. Among the promising algorithms, Gaussian process regression (GPR) proved to be a competitive time series gap-filling algorithm with the advantage of, as developed within a Bayesian framework, providing associated uncertainty estimates. Nevertheless, the processing of time series images becomes computationally inefficient in its standard per-pixel usage, mainly for GPR training rather than the fitting step. To mitigate this computational burden, we propose to substitute the per-pixel optimization step with the creation of a cropland-based precalculations for the GPR hyperparameters θ . To demonstrate our approach hardly affects the accuracy in fitting, we used Sentinel-2 LAI time series over an agricultural region in Castile and Leon, North-West Spain. The performance of image reconstructions were compared against the standard per-pixel GPR time series processing. Results showed that accuracies were on the same order (RMSE 0.1767 vs. 0.1564 [m²/m²], 12% RMSE degradation) whereas processing time accelerated about 90 times. We further evaluated the alternative option of using the same hyperparameters for all the pixels within the complete scene. It led to similar overall accuracies over crop areas and computational performance. Crop phenology indicators were also calculated for the three different approaches and compared. Results showed analogous crop temporal patterns, with differences in start and end of growing season of no more than five days. To the benefit of crop monitoring applications, all the gap-filling and phenology indicators retrieval techniques have been implemented into the freely downloadable GUI toolbox DATimeS.

Urban Heat Island Monitoring and Impacts on Citizen’s General Health Status in Isfahan Metropolis: A Remote Sensing and Field Survey Approach

Urban heat islands (UHIs) are one of the urban management challenges, especially in metropolises, which can affect citizens’ health and well-being. This study used a combination of remote sensing techniques with field survey to investigate systematically the effects of UHI on citizens’ health in Isfahan metropolis, Iran. For this purpose, the land surface temperature (LST) over a three-year period was monitored by Landsat-8 satellite imagery based on the split window algorithm. Then, the areas where UHI and urban cold island (UCI) phenomena occurred were identified and a general health questionnaire-28 (GHQ-28) was applied to evaluate the health status of 800 citizens in terms of physical health, anxiety and sleep, social function, and depression in UHI and UCI treatments. The average LST during the study period was 45.5 ± 2.3 °C and results showed that the Zayandeh-Rood river and the surrounding greenery had an important role in regulating the ambient temperature and promoting the citizens’ health. Citizens living in the suburban areas were more exposed to the UHIs phenomena, and statistical analysis of the GHQ-28 results indicated that they showed severe significant (P < 0.05) responses in terms of non-physical health sub-scales (i.e., anxiety and sleep, social functioning, and depression). Therefore, it can be concluded that not all citizens in the Isfahan metropolis are in the same environmental conditions and city managers and planners should pay more attention to the citizens living in the UHIs. The most important proceedings in this area would be the creation and development of parks and green belts, as well as the allocation of health-medical facilities and citizen education.

Comparative analysis of atmospheric radiative transfer models using the Atmospheric Look-up table Generator (ALG) toolbox (version 2.0)

Atmospheric radiative transfer models (RTMs) are software tools that help researchers in understanding the radiative processes occurring in the Earth's atmosphere. Given their importance in remote sensing applications, the intercomparison of atmospheric RTMs is therefore one of the main tasks used to evaluate model performance and identify the characteristics that differ between models. This can be a tedious tasks that requires good knowledge of the model inputs/outputs and the generation of large databases of consistent simulations. With the evolution of these software tools, their increase in complexity bears implications for their use in practical applications and model intercomparison. Existing RTM-specific graphical user interfaces are not optimized for performing intercomparison studies of a wide variety of atmospheric RTMs. In this paper, we present the Atmospheric Look-up table Generator (ALG) version 2.0, a new software tool that facilitates generating large databases for a variety of atmospheric RTMs. ALG facilitates consistent and intuitive user interaction to enable the running of model executions and storing of RTM data for any spectral configuration in the optical domain. We demonstrate the utility of ALG in performing intercomparison studies of radiance simulations from broadly used atmospheric RTMs (6SV, MODTRAN, and libRadtran) through global sensitivity analysis. We expect that providing ALG to the research community will facilitate the usage of atmospheric RTMs to a wide range of applications in Earth observation.

Estimation of leaf area index using PROSAIL based LUT inversion, MLRA-GPR and empirical models: Case study of tropical deciduous forest plantation, North India

Forests play a vital role in biological cycles and environmental regulation. To understand the key processes of forest canopies (e.g., photosynthesis, respiration and transpiration), reliable and accurate information on spatial variability of Leaf Area Index (LAI), and its seasonal dynamics is essential. In the present study, we assessed the performance of biophysical parameter (LAI) retrieval methods viz. Look-Up Table (LUT)-inversion, MLRA-GPR (Machine Learning Regression Algorithm-Gaussian Processes Regression) and empirical models, for estimating the LAI of tropical deciduous plantation using ARTMO (Automated Radiative Transfer Models Operator) tool and Sentinel-2 satellite images. The study was conducted in Central Tarai Forest Division, Haldwani, located in the Uttarakhand state, India. A total of 49 ESUs (Elementary Sampling Unit) of 30m×30m size were established based on variability in composition and age of plantation stands. In-situ LAI was recorded using plant canopy imager during the leaf growing, peak and senescence seasons. The PROSAIL model was calibrated with site-specific biophysical and biochemical parameters before used to the predicted LAI. The plantation LAI was also predicted by an empirical approach using optimally chosen Sentinel-2 vegetation indices. In addition, Sentinel-2 and MODIS LAI products were evaluated with respect to LAI measurements. MLRA-GPR offered best results for predicting LAI of leaf growing (R2 = 0.9, RMSE = 0.14), peak (R2 = 0.87, RMSE = 0.21) and senescence (R2 = 0.86, RMSE = 0.31) seasons while LUT inverted model outperformed VI's based parametric regression model. Vegetation indices (VIs) derived from 740 nm, 783 nm and 2190 nm band combinations of Sentinel-2 offered the best prediction of LAI.