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

Proximal remote sensing: an essential tool for bridging the gap between high-resolution ecosystem monitoring and global ecology

A new proliferation of optical instruments that can be attached to towers over or within ecosystems, or 'proximal' remote sensing, enables a comprehensive characterization of terrestrial ecosystem structure, function, and fluxes of energy, water, and carbon. Proximal remote sensing can bridge the gap between individual plants, site-level eddy-covariance fluxes, and airborne and spaceborne remote sensing by providing continuous data at a high-spatiotemporal resolution. Here, we review recent advances in proximal remote sensing for improving our mechanistic understanding of plant and ecosystem processes, model development, and validation of current and upcoming satellite missions. We provide current best practices for data availability and metadata for proximal remote sensing: spectral reflectance, solar-induced fluorescence, thermal infrared radiation, microwave backscatter, and LiDAR. Our paper outlines the steps necessary for making these data streams more widespread, accessible, interoperable, and information-rich, enabling us to address key ecological questions unanswerable from space-based observations alone and, ultimately, to demonstrate the feasibility of these technologies to address critical questions in local and global ecology.

Monitoring functional traits of complex temperate forests using Sentinel-2 data during a severe drought period

Forest ecosystems are crucial for preserving biodiversity and providing ecosystem services. The Ticino Park is a temperate mixed forest, one of the few natural ecosystems in northern Italy, which is facing increasing natural and anthropogenic disturbances exacerbated by climate change. Remote sensing is a cost-effective tool for the indirect estimation of forest status. However, it has typically relied on indirect proxies that often have coarse spatio-temporal resolution. This study investigates the possibility of deriving high temporal resolution time series of forest traits to identify drought-induced anomalies and relate them to differences in forest type and environmental factors. Sentinel-2 images from 2017 to 2022 were analysed, with 2022 being characterised by a severe drought. Leaf area index (LAI), canopy chlorophyll content (CCC), and canopy water content (CWC) were retrieved from Sentinel-2 time series through the S2 Toolbox and validated using measurements collected during an intensive field campaign in 2022. A positive and statistically significant correlation was found for all traits. The best results were obtained for LAI (R2 = 0.75, nRMSE = 11.49 %) and CCC (R2 = 0.82, nRMSE = 13.56 %), while slightly worse results were obtained for CWC (R2 = 0.64, nRMSE = 8.84 %). The accurate retrieval of LAI, CCC and CWC enabled the analysis of the temporal and spatial variations of the daily standardised anomalies (DSA). CCC reached the most negative DSA values, highlighting its higher sensitivity in detecting the effects of water shortage compared to CWC and LAI. The statistical analysis showed that DSA was linked to forest types. Pine and black cherry exhibited the highest stress response, while hygrophilic black alder and chestnut were the least impacted. These results highlight the species-specific responses to drought and the importance of incorporating species information in forest monitoring. The developed methodology provides a cost-effective approach for monitoring forest status and supporting effective management strategies.

Machine Learning and New-Generation Spaceborne Hyperspectral Data Advance Crop Type Mapping

Hyperspectral sensors provide near-continuous spectral data that can facilitate advancements in agricultural crop classification and characterization, which are important for addressing global food and water security issues. We investigated two new-generation hyperspectral sensors, Germany’s Deutsches Zentrum für Luft‐ und Raumfahrt Earth Sensing Imaging Spectrometer (DESIS) and Italy’s PRecursore IperSpettrale della Missione Applicativa (PRISMA), within California???s Central Valley in August 2021 focusing on five irrigated agricultural crops (alfalfa, almonds, corn, grapes, and pistachios). With reference data from the U.S. Department of Agriculture Cropland Data Layer, we developed a spectral library of the crops and classified them using three machine learning algorithms (support vector machines [SVM], random forest [RF], and spectral angle mapper [SAM]) and two philosophies: 1. Full spectral analysis (FSA) and 2. Optimal hyperspectral narrowband (OHNB) analysis. For FSA, we used 59 DESIS four-bin product bands and 207 of 238 PRISMA bands. For OHNB analysis, 9 DESIS and 16 PRISMA nonredundant OHNBs for studying crops were selected. FSA achieved only 1% to 3% higher accuracies relative to OHNB analysis in most cases. SVM provided the best results, closely followed by RF. Using both DESIS and PRISMA image OHNBs in SVM for classification led to higher accuracy than using either image alone, with an overall accuracy of 99%, producer’s accuracies of 94% to 100%, and user???s accuracies of 95% to 100%.

Upscaling vascular aboveground biomass and topsoil moisture of subarctic fens from Unoccupied Aerial Vehicles (UAVs) to satellite level

Arctic and subarctic ecosystems are experiencing rapid changes in vegetation composition and productivity due to global warming. Tundra wetlands are especially susceptible to these changes, which may trigger shifts in soil moisture dynamics. It is therefore essential to accurately map plant biomass and topsoil moisture. In this study, we mapped total, wood, and leaf above ground biomass and topsoil moisture in subarctic tundra wetlands located between Norway and Finland by linking models derived from Unoccupied Aerial Vehicles with multiple satellite data sources using the Extreme Gradient Boosting algorithm. The most accurate predictions for topsoil moisture (R2 = 0.73) used a set of red edge-based vegetation indices with a spatial resolution of 20 m per pixel. On the contrary, wood biomass showed the lowest accuracies across all tested models (R2 = 0.38). We found a trade-off between the spatial resolution and the performance of upscaling models, where smaller pixel sizes generally led to lower accuracies. However, we were able to compensate for reduced accuracy at smaller pixel sizes using Copernicus phenology metrics. A modelling uncertainty assessment revealed that the uncertainty of predictions increased with decreasing pixel sizes and increasing number of co-predictors. Our approach could improve efforts to map and monitor changes in vegetation at regional to pan-Arctic scales.

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.

Identification for the species of aquatic higher plants in the Taihu Lake basin based on hyperspectral remote sensing

Aquatic plants are crucial for aquatic ecosystems and their species and distribution reflect aquatic ecosystem health. Remote sensing technology has been used to monitor plant distributions over large scales. However, the fine identification of the species of aquatic higher plants is challenging due to large temporal-spatial changes in optical water body properties and small spectral differences among plant species. Here, an aquatic plant identification method was developed by constructing a decision tree file in the C4.5 algorithm based on the canopy spectra of eight plants in the Changguangxi Wetland water area from hyperspectral remote sensing technology. The method was used to monitor the distribution of different plants in the Changguangxi Wetland area and two other water areas. The results showed that the spectral characteristics of plants were enhanced by calculating their spectral index, thereby improving the comparability among different species. The total recognition accuracy of the constructed decision tree file for eight types of plants was 85.02%. Nymphaea tetragona, Pontederia cordata, and Nymphoides peltatum had the highest recognition accuracy and Eichhornia crassipes was the lowest. The specific species and distributions of aquatic plants were consistent with the water quality in the area. The results can provide a reference for the accurate identification of aquatic plants in the same type of water area.

Multi-Temporal Hyperspectral Classification of Grassland Using Transformer Network

In recent years, grassland monitoring has shifted from traditional field surveys to remote-sensing-based methods, but the desired level of accuracy has not yet been obtained. Multi-temporal hyperspectral data contain valuable information about species and growth season differences, making it a promising tool for grassland classification. Transformer networks can directly extract long-sequence features, which is superior to other commonly used analysis methods. This study aims to explore the transformer network's potential in the field of multi-temporal hyperspectral data by fine-tuning it and introducing it into high-powered grassland detection tasks. Subsequently, the multi-temporal hyperspectral classification of grassland samples using the transformer network (MHCgT) is proposed. To begin, a total of 16,800 multi-temporal hyperspectral data were collected from grassland samples at different growth stages over several years using a hyperspectral imager in the wavelength range of 400-1000 nm. Second, the MHCgT network was established, with a hierarchical architecture, which generates a multi-resolution representation that is beneficial for grass hyperspectral time series' classification. The MHCgT employs a multi-head self-attention mechanism to extract features, avoiding information loss. Finally, an ablation study of MHCgT and comparative experiments with state-of-the-art methods were conducted. The results showed that the proposed framework achieved a high accuracy rate of 98.51% in identifying grassland multi-temporal hyperspectral which outperformed CNN, LSTM-RNN, SVM, RF, and DT by 6.42-26.23%. Moreover, the average classification accuracy of each species was above 95%, and the August mature period was easier to identify than the June growth stage. Overall, the proposed MHCgT framework shows great potential for precisely identifying multi-temporal hyperspectral species and has significant applications in sustainable grassland management and species diversity assessment.

Low-Cost Hyperspectral Imaging to Detect Drought Stress in High-Throughput Phenotyping

Recent developments in low-cost imaging hyperspectral cameras have opened up new possibilities for high-throughput phenotyping (HTP), allowing for high-resolution spectral data to be obtained in the visible and near-infrared spectral range. This study presents, for the first time, the integration of a low-cost hyperspectral camera Senop HSC-2 into an HTP platform to evaluate the drought stress resistance and physiological response of four tomato genotypes (770P, 990P, Red Setter and Torremaggiore) during two cycles of well-watered and deficit irrigation. Over 120 gigabytes of hyperspectral data were collected, and an innovative segmentation method able to reduce the hyperspectral dataset by 85.5% was developed and applied. A hyperspectral index (H-index) based on the red-edge slope was selected, and its ability to discriminate stress conditions was compared with three optical indices (OIs) obtained by the HTP platform. The analysis of variance (ANOVA) applied to the OIs and H-index revealed the better capacity of the H-index to describe the dynamic of drought stress trend compared to OIs, especially in the first stress and recovery phases. Selected OIs were instead capable of describing structural changes during plant growth. Finally, the OIs and H-index results have revealed a higher susceptibility to drought stress in 770P and 990P than Red Setter and Torremaggiore genotypes.

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.

A Review of Hybrid Approaches for Quantitative Assessment of Crop Traits Using Optical Remote Sensing: Research Trends and Future Directions

Remote sensing technology allows to provide information about biochemical and biophysical crop traits and monitor their spatiotemporal dynamics of agriculture ecosystems. Among multiple retrieval techniques, hybrid approaches have been found to provide outstanding accuracy, for instance, for the inference of leaf area index (LAI), fractional vegetation cover (fCover), and leaf and canopy chlorophyll content (LCC and CCC). The combination of radiative transfer models (RTMs) and data-driven models creates an advantage in the use of hybrid methods. Through this review paper, we aim to provide state-of-the-art hybrid retrieval schemes and theoretical frameworks. To achieve this, we reviewed and systematically analyzed publications over the past 22 years. We identified two hybrid-based parametric and hybrid-based nonparametric regression models and evaluated their performance for each variable of interest. From the results of our extensive literature survey, most research directions are now moving towards combining RTM and machine learning (ML) methods in a symbiotic manner. In particular, the development of ML will open up new ways to integrate innovative approaches such as integrating shallow or deep neural networks with RTM using remote sensing data to reduce errors in crop trait estimations and improve control of crop growth conditions in very large areas serving precision agriculture applications.

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.

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.