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.

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/Cab, 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.

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.

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.

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.

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.

Prototyping Crop Traits Retrieval Models for CHIME: Dimensionality Reduction Strategies Applied to PRISMA Data.

In preparation for new-generation imaging spectrometer missions and the accompanying unprecedented inflow of hyperspectral data, optimized models are needed to generate vegetation traits routinely. Hybrid models, combining radiative transfer models with machine learning algorithms, are preferred, however, dealing with spectral collinearity imposes an additional challenge. In this study, we analyzed two spectral dimensionality reduction methods: principal component analysis (PCA) and band ranking (BR), embedded in a hybrid workflow for the retrieval of specific leaf area (SLA), leaf area index (LAI), canopy water content (CWC), canopy chlorophyll content (CCC), the fraction of absorbed photosynthetic active radiation (FAPAR), and fractional vegetation cover (FVC). The SCOPE model was used to simulate training data sets, which were optimized with active learning. Gaussian process regression (GPR) algorithms were trained over the simulations to obtain trait-specific models. The inclusion of PCA and BR with 20 features led to the so-called GPR-20PCA and GPR-20BR models. The 20PCA models encompassed over 99.95% cumulative variance of the full spectral data, while the GPR-20BR models were based on the 20 most sensitive bands. Validation against in situ data obtained moderate to optimal results with normalized root mean squared error (NRMSE) from 13.9% (CWC) to 22.3% (CCC) for GPR-20PCA models, and NRMSE from 19.6% (CWC) to 29.1% (SLA) for GPR-20BR models. Overall, the GPR-20PCA slightly outperformed the GPR-20BR models for all six variables. To demonstrate mapping capabilities, both models were tested on a PRecursore IperSpettrale della Missione Applicativa (PRISMA) scene, spectrally resampled to Copernicus Hyperspectral Imaging Mission for the Environment (CHIME), over an agricultural test site (Jolanda di Savoia, Italy). The two strategies obtained plausible spatial patterns, and consistency between the two models was highest for FVC and LAI (R 2 = 0.91, R 2 = 0.86) and lowest for SLA mapping (R 2 = 0.53). From these findings, we recommend implementing GPR-20PCA models as the most efficient strategy for the retrieval of multiple crop traits from hyperspectral data streams. Hence, this workflow will support and facilitate the preparations of traits retrieval models from the next-generation operational CHIME.

An approach to fill in missing data from satellite imagery using data-intensive computing and DINEOF.

This paper proposes an approach to fill in missing data from satellite images using data-intensive computing platforms. The proposed approach merges satellite imagery from diverse sources to reduce the impact of the holes in images that result from acquisition conditions: occlusion, the satellite trajectory, sunlight, among others. The amount of computation effort derived from the use of large high-resolution images is addressed by data-intensive computing techniques that assume an underlying cluster architecture. As a start, satellite data from the region of study are automatically downloaded; then, data from different sensors are corrected and merged to obtain an orthomosaic; finally, the orthomosaic is split into user-defined segments to fill in missing data, and filled segments are assembled to produce an orthomosaic with a reduced amount of missing data. As a proof of concept, the proposed data-intensive approach was implemented to study the concentration of chlorophyll at the Mexican oceans by merging data from MODIS-TERRA, MODIS-AQUA, VIIRS-SNPP, and VIIRS-JPSS-1 sensors. The results revealed that the proposed approach produces results that are similar to state-of-the-art approaches to estimate chlorophyll concentration but avoid memory overflow with large images. Visual and statistical comparison of the resulting images revealed that the proposed approach provides a more accurate estimation of chlorophyll concentration when compared to the mean of pixels method alone.

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.

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 R 2 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 R 2 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.

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.

Systematic Assessment of MODTRAN Emulators for Atmospheric Correction.

Atmospheric radiative transfer models (RTMs) simulate the light propagation in the Earth's atmosphere. With the evolution of RTMs, their increase in complexity makes them impractical in routine processing such as atmospheric correction. To overcome their computational burden, standard practice is to interpolate a multidimensional lookup table (LUT) of prestored simulations. However, accurate interpolation relies on large LUTs, which still implies large computation times for their generation and interpolation. In recent years, emulation has been proposed as an alternative to LUT interpolation. Emulation approximates the RTM outputs by a statistical regression model trained with a low number of RTM runs. However, a concern is whether the emulator reaches sufficient accuracy for atmospheric correction. Therefore, we have performed a systematic assessment of key aspects that impact the precision of emulating MODTRAN: 1) regression algorithm; 2) training database size; 3) dimensionality reduction (DR) method and a number of components; and 4) spectral resolution. The Gaussian processes regression (GPR) was found the most accurate emulator. The principal component analysis remains a robust DR method and nearly 20 components reach sufficient precision. Based on a database of 1000 samples covering a broad range of atmospheric conditions, GPR emulators can reconstruct the simulated spectral data with relative errors below 1% for the 95th percentile. These emulators reduce the processing time from days to minutes, preserving sufficient accuracy for atmospheric correction and providing model uncertainties and derivatives. We provide a set of guidelines and tools to design and generate accurate emulators for satellite data processing applications.

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 (Cab ), leaf water content (Cw ), fractional vegetation coverage (FVC), leaf area index (LAI), and upscaled leaf biochemical compounds, i.e., LAI * Cab (laiCab) and LAI * Cw (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 Cab . 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).

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.

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.

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.

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.

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.

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.