Hyperspectral remote sensing of foliar nitrogen content

Joint optimization of land carbon uptake and albedo can help achieve moderate instantaneous and long-term cooling effects

Both carbon dioxide uptake and albedo of the land surface affect global climate. However, climate change mitigation by increasing carbon uptake can cause a warming trade-off by decreasing albedo, with most research focusing on afforestation and its interaction with snow. Here, we present carbon uptake and albedo observations from 176 globally distributed flux stations. We demonstrate a gradual decline in maximum achievable annual albedo as carbon uptake increases, even within subgroups of non-forest and snow-free ecosystems. Based on a paired-site permutation approach, we quantify the likely impact of land use on carbon uptake and albedo. Shifting to the maximum attainable carbon uptake at each site would likely cause moderate net global warming for the first approximately 20 years, followed by a strong cooling effect. A balanced policy co-optimizing carbon uptake and albedo is possible that avoids warming on any timescale, but results in a weaker long-term cooling effect.

Leaf photosynthetic pigment as a predictor of leaf maximum carboxylation rate in a farmland ecosystem

The leaf maximum rate of carboxylation (V_cmax) is a key parameter of plant photosynthetic capacity. The accurate estimation of V_cmax is crucial for correctly predicting the carbon flux in the terrestrial carbon cycle. V_cmax is correlated with plant traits including leaf nitrogen (N_area) and leaf photosynthetic pigments. Proxies for leaf chlorophyll (Chl_area) and carotenoid contents (Car_area) need to be explored in different ecosystems. In this study, we evaluated the relationship between leaf maximum rate of carboxylation (scaled to 25°C; V_cmax25) and both leaf N_area and photosynthetic pigments (Chl_area and Car_area) in winter wheat in a farmland ecosystem. Our results showed that V_cmax25 followed the same trends as leaf Chl_area. However, leaf N_area showed smaller dynamic changes before the flowering stage, and there were smaller seasonal variations in leaf Car_area. The correlation between leaf V_cmax25 and leaf Chl_area was the strongest, followed by leaf Car_area and leaf N_area (R² = 0.69, R² = 0.47 and R² = 0.36, respectively). The random forest regression analysis also showed that leaf Chl_area and leaf Car_area were more important than leaf N_area for V_cmax25. The correlation between leaf V_cmax25 and N_area can be weaker since nitrogen allocation is dynamic. The estimation accuracy of the V_cmax25 model based on N_area, Chl_area, and Car_area (R² = 0.75) was only 0.05 higher than that of the V_cmax25 model based on Chl_area and Car_area (R² = 0.70). However, the estimation accuracy of the V_cmax25 model based on Chl_area and Car_area (R² = 0.70) was 0.34 higher than that of the V_cmax25 model based on N_area (R² = 0.36). These results highlight that leaf photosynthetic pigments can be a predictor for estimating V_cmax25, expanding a new way to estimate spatially continuous V_cmax25 on a regional scale, and to improve model simulation accuracy.

Challenges and opportunities in remote sensing-based crop monitoring: a review

Building a more resilient food system for sustainable development and reducing uncertainty in global food markets both require concurrent and near-real-time and reliable crop information for decision making. Satellite-driven crop monitoring has become a main method to derive crop information at local, regional, and global scales by revealing the spatial and temporal dimensions of crop growth status and production. However, there is a lack of quantitative, objective, and robust methods to ensure the reliability of crop information, which reduces the applicability of crop monitoring and leads to uncertain and undesirable consequences. In this paper, we review recent progress in crop monitoring and identify the challenges and opportunities in future efforts. We find that satellite-derived metrics do not fully capture determinants of crop production and do not quantitatively interpret crop growth status; the latter can be advanced by integrating effective satellite-derived metrics and new onboard sensors. We have identified that ground data accessibility and the negative effects of knowledge-based analyses are two essential issues in crop monitoring that reduce the applicability of crop monitoring for decisions on food security. Crowdsourcing is one solution to overcome the restrictions of ground-truth data accessibility. We argue that user participation in the complete process of crop monitoring could improve the reliability of crop information. Encouraging users to obtain crop information from multiple sources could prevent unconscious biases. Finally, there is a need to avoid conflicts of interest in publishing publicly available crop information.

An autofocus algorithm considering wavelength changes for large scale microscopic hyperspectral pathological imaging system

Microscopic hyperspectral imaging technology has been widely used to acquire pathological information of tissue sections. Autofocus is one of the most important steps in microscopic hyperspectral imaging systems to capture large scale or even whole slide images of pathological slides with high quality and high speed. However, there are quite few autofocus algorithm put forward for the microscopic hyperspectral imaging system. Therefore, this article proposes a Laplace operator based autofocus algorithm for microscopic hyperspectral imaging system which takes the influence of wavelength changes into consideration. Through the proposed algorithm, the focal length for each wavelength can be adjusted automatically to ensure that each single band image can be autofocused precisely with adaptive image sharpness evaluation method. In addition, to increase the capture speed, the relationship of wavelength and focal length is derived and the focal offsets among different single band images are calculated for pre-focusing. We have employed the proposed method on our own datasets and the experimental results show that it can capture large-scale microscopic hyperspectral pathology images with high precise.

Practical spectral photography II: snapshot spectral imaging using linear retarders and microgrid polarization cameras

Despite recent advances, customized multispectral cameras can be challenging or costly to deploy in some use cases. Complexities span electronic synchronization, multi-camera calibration, parallax and spatial co-registration, and data acquisition from multiple cameras, all of which can hamper their ease of use. This paper discusses a generalized procedure for multispectral sensing using a pixelated polarization camera and anisotropic polymer film retarders to create multivariate optical filters. We then describe the calibration procedure, which leverages neural networks to convert measured data into calibrated spectra (intensity versus wavelength). Experimental results are presented for a multivariate and channeled optical filter. Finally, imaging results taken using a red, green, and blue microgrid polarization camera and the channeled optical filter are presented. Imaging experiments indicated that the calculated spectra's root mean square error is highest in the region where the camera's red, green, and blue filter responses overlap. The average error of the spectral reflectance, measured of our spectralon tiles, was 6.5% for wavelengths spanning 425-675 nm. This technique demonstrates that 12 spectral channels can be obtained with a relatively simple and robust optical setup, and at minimal cost beyond the purchase of the camera.

Detection of Apple Valsa Canker Based on Hyperspectral Imaging

Approximately half of the world’s apple production occurs in East Asia, where apple Valsa canker (AVC) is a prominent disease. This disease affects the bark of the tree, ultimately killing it and resulting in significant economic loss. Visual identification of the diseased area of the bark, particularly in the early stages, is extremely difficult. In this study, we conducted hyperspectral imaging of the trunks and branches of AVC-infected apple trees and revealed that the diseased area can be identified in the near-infrared reflectance, even when it is difficult to distinguish visually. A discriminant analysis using the Mahalanobis distance was performed on the normalized difference spectral index (NDSI) obtained from the measured spectral reflectance. A diagnostic model for discriminating between the healthy and diseased areas was created using the threshold value of NDSI. An accuracy assessment of the diagnostic model presented the overall accuracy as >0.94 for the combinations of spectral bands at 660–690 nm and 720–760 nm. This simple diagnostic model can be applied to other tree bark canker diseases.

Remote Sensing Monitoring of Winter Wheat Stripe Rust Based on mRMR-XGBoost Algorithm

For the problem of multi-dimensional feature redundancy in remote sensing detection of wheat stripe rust using reflectance spectrum and solar-induced chlorophyll fluorescence (SIF), a feature selection and disease index (DI) monitoring model combining mRMR and XGBoost algorithm was proposed in this study. Firstly, characteristic wavelengths selected by successive projections algorithm (SPA) were combined with the vegetation indices, trilateral parameters, and canopy SIF parameters to constitute the initial feature set. Then, the max-relevance and min-redundancy (mRMR) algorithm and correlation coefficient (CC) analysis were used to reduce the dimensionality of the initial feature set, respectively. Features selected by mRMR and CC were input as independent variables into the extreme gradient boosting regression (XGBoost) and gradient boosting regression tree (GBRT) to monitor the severity of stripe rust. The experimental results show that, compared with CC analysis, the monitoring accuracy of the features selected by mRMR in the XGBoost and GBRT models increased by 12% and 17% on average, respectively. Meanwhile, the mRMR-XGBoost model achieved the best monitoring accuracy (R2 = 0.8894, RMSE = 0.1135). The R2 between the measured DI and predicted DI of mRMR-XGBoost was improved by an average of 5%, 12%, and 22% compared with mRMR-GBRT, CC-XGBoost, and CC-GBRT models. These results suggested that XGBoost is more suitable for the remote sensing monitoring of wheat stripe rust, and mRMR has more advantages than the commonly used CC analysis in feature selection. Field survey data validation results also confirm that the mRMR-XGBoost algorithm has excellent monitoring applicability and scalability. The proposed model could provide a reference for data dimensionality reduction and crop disease index monitoring based on hyperspectral data.

Nitrogen use efficiency (NUE): elucidated mechanisms, mapped genes and gene networks in maize (Zea mays L.)

Nitrogen, the vital primary plant growth nutrient at deficit soil conditions, drastically affects the growth and yield of a crop. Over the years, excess use of inorganic nitrogenous fertilizers resulted in pollution, eutrophication and thereby demanding the reduction in the use of chemical fertilizers. Being a C₄ plant with fibrous root system and high NUE, maize can be deployed to be the best candidate for better N uptake and utilization in nitrogen deficient soils. The maize germplasm sources has enormous genetic variation for better nitrogen uptake contributing traits. Adoption of single cross maize hybrids as well as inherent property of high NUE has helped maize cultivars to achieve the highest growth rate among the cereals during last decade. Further, considering the high cost of nitrogenous fertilizers, adverse effects on soil health and environmental impact, maize improvement demands better utilization of existing genetic variation for NUE via introgression of novel allelic combinations in existing cultivars. Marker assisted breeding efforts need to be supplemented with introgression of genes/QTLs related to NUE in ruling varieties and thereby enhancing the overall productivity of maize in a sustainable manner. To achieve this, we need mapped genes and network of interacting genes and proteins to be elucidated. Identified genes may be used in screening ideal maize genotypes in terms of better physiological functionality exhibiting high NUE. Future genome editing may help in developing lines with increased productivity under low N conditions in an environment of optimum agronomic practices.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s12298-021-01113-z.

Poor relationships between NEON Airborne Observation Platform data and field-based vegetation traits at a mesic grassland

Understanding spatial and temporal variation in plant traits is needed to accurately predict how communities and ecosystems will respond to global change. The National Observatory Ecological Network (NEON) Airborne Observation Platform (AOP) provides hyperspectral images and associated data products at numerous field sites at 1 m spatial resolution, potentially allowing high-resolution trait mapping. We tested the accuracy of NEON's readily available AOP derived data products - Leaf Area Index, Total biomass, Ecosystem structure (Canopy height model; CHM), and Canopy Nitrogen by comparing them to spatially extensive field measurements from a mesic tallgrass prairie. Correlations with AOP data products exhibited generally weak or no relationships with corresponding field measurements. The strongest relationships were between AOP LAI and ground-measured LAI (r = 0.32) and AOP Total biomass and ground-measured biomass (r = 0.23). We also examined how well the full reflectance spectra (380-2500 nm), as opposed to derived products, could predict vegetation traits using partial least-squares regression models. Only one of the eight traits examined, Nitrogen, had a validation R² of more than 0.25. For all vegetation traits, validation R² ranged from 0.08-0.29 and the root mean square error of prediction ranged from 14-64%. Our results suggest that currently available AOP derived data products should not be used without extensive ground-based validation. Relationships using the full reflectance spectra may be more promising, although careful consideration of field and AOP data mismatches in space and/or time, biases in field-based measurements or AOP algorithms, and model uncertainty are needed. Finally, grassland sites may be especially challenging for airborne spectroscopy because of their high species diversity within a small area, mixed functional types of plant communities, and heterogenous mosaics of disturbance and resource availability. Remote sensing observations are one of the most promising approaches to understanding ecological patterns across space and time, yet the opportunity to engage a diverse community of NEON data users will depend on establishing rigorous links with in-situ field measurements across a diversity of sites.

Hyperspectral reflectance-based phenotyping for quantitative genetics in crops: Progress and challenges

Many biochemical and physiological properties of plants that are of interest to breeders and geneticists have extremely low throughput and/or can only be measured destructively. This has limited the use of information on natural variation in nutrient and metabolite abundance, as well as photosynthetic capacity in quantitative genetic contexts where it is necessary to collect data from hundreds or thousands of plants. A number of recent studies have demonstrated the potential to estimate many of these traits from hyperspectral reflectance data, primarily in ecophysiological contexts. Here, we summarize recent advances in the use of hyperspectral reflectance data for plant phenotyping, and discuss both the potential benefits and remaining challenges to its application in plant genetics contexts. The performances of previously published models in estimating six traits from hyperspectral reflectance data in maize were evaluated on new sample datasets, and the resulting predicted trait values shown to be heritable (e.g., explained by genetic factors) were estimated. The adoption of hyperspectral reflectance-based phenotyping beyond its current uses may accelerate the study of genes controlling natural variation in biochemical and physiological traits.

A methodological roadmap to quantify animal-vectored spatial ecosystem subsidies

Energy, nutrients and organisms move over landscapes, connecting ecosystems across space and time. Meta-ecosystem theory investigates the emerging properties of local ecosystems coupled spatially by these movements of organisms and matter, by explicitly tracking exchanges of multiple substances across ecosystem borders. To date, meta-ecosystem research has focused mostly on abiotic flows-neglecting biotic nutrient flows. However, recent work has indicated animals act as spatial nutrient vectors when they transport nutrients across landscapes in the form of excreta, egesta and their own bodies. Partly due to its high level of abstraction, there are few empirical tests of meta-ecosystem theory. Furthermore, while animals may be viewed as important mediators of ecosystem functions, better integration of tools is needed to develop predictive insights of their relative roles and impacts on diverse ecosystems. We present a methodological roadmap that explains how to do such integration by discussing how to combine insights from movement, foraging and ecosystem ecology to develop a coherent understanding of animal-vectored nutrient transport on meta-ecosystems processes. We discuss how the slate of newly developed technologies and methods-tracking devices, mechanistic movement models, diet reconstruction techniques and remote sensing-that when integrated have the potential to advance the quantification of animal-vectored nutrient flows and increase the predictive power of meta-ecosystem theory. We demonstrate that by integrating novel and established tools of animal ecology, ecosystem ecology and remote sensing, we can begin to identify and quantify animal-mediated nutrient translocation by large animals. We also provide conceptual examples that show how our proposed integration of methodologies can help investigate ecosystem impacts of large animal movement. We conclude by describing practical advancements to understanding cross-ecosystem contributions of animals on the move. Understanding the mechanisms by which animals shape ecosystem dynamics is important for ongoing conservation, rewilding and restoration initiatives around the world, and for developing more accurate models of ecosystem nutrient budgets. Our roadmap will enable ecologists to better qualify and quantify animal-mediated nutrient translocation for animals on the move.

Evaluation of the FluorWPS Model and Study of the Parameter Sensitivity for Simulating Solar-Induced Chlorophyll Fluorescence

Solar-induced chlorophyll fluorescence (SIF) has been used as an indicator for the photosynthetic activity of vegetation at regional and global scales. Canopy structure affects the radiative transfer process of SIF within canopy and causes the angular-dependencies of SIF. A common solution for interpreting these effects is the use of physically-based radiative transfer models. As a first step, a comprehensive evaluation of the three-dimensional (3D) radiative transfers is needed using ground truth biological and hyperspectral remote sensing measurements. Due to the complexity of forest modeling, few studies have systematically investigated the effect of canopy structural factors and sun-target-viewing geometry on SIF. In this study, we evaluated the capability of the Fluorescence model with the Weighted Photon Spread method (FluorWPS) to simulate at-sensor radiance and SIF at the top of canopy, and identified the influence of the canopy structural factors and sun-target-viewing geometry on the magnitude and directional response of SIF in deciduous forests. To evaluate the model, a 3D forest scene was first constructed from Goddard’s LiDAR Hyperspectral and Thermal (G-LiHT) LiDAR data. The reliability of the reconstructed scene was confirmed by comparing the calculated leaf area index with the measured ones from the scene, which resulted in a relative error of 3.5%. Then, the performance of FluorWPS was evaluated by comparing the simulated at-sensor radiance spectra with the spectra measured from the DUAL and FLUO spectrometer of HyPlant. The radiance spectra simulated by FluorWPS agreed well with the measured spectra by the two high-performance imaging spectrometers, with a coefficient of determination (R2) of 0.998 and 0.926, respectively. SIF simulated by the FluorWPS model agreed well with the values of the DART model. Furthermore, a sensitivity analysis was conducted to assess the effect of the canopy structural parameters and sun-target-viewing geometry on SIF. The maximum difference of the total SIF can be as large as 45% and 47% at the wavelengths of 685 nm and 740 nm for different foliage area volume densities (FAVDs), and 48% and 46% for fractional vegetation covers (FVCs), respectively. Leaf angle distribution has a markedly influence on the magnitude of SIF, with a ratio of emission part to SIF range from 0.48 to 0.72. SIF from the grass layer under the tree contributed 10%+ more to the top of canopy SIF even for a dense forest canopy (FAVD = 3.5 m−1, FVC = 76%). The red SIF at the wavelength of 685 nm had a similar shape to the far-red SIF at a wavelength of 740 nm but with higher variability in varying illumination conditions. The integration of the FluorWPS model and LiDAR modeling can greatly improve the interpretation of SIF at different scales and angular configurations.

High nitrogen addition decreases the ozone flux by reducing the maximum stomatal conductance in poplar saplings

Ground-level ozone (O₃) and nitrogen (N) deposition are major environmental pollutants, often occurring concurrently. Ozone exposure- and flux-response relationships for tree biomass are used for regional O₃ risk assessment. In order to investigate whether soil N addition affects stomatal O₃ uptake of poplar, poplar saplings were exposed to treatment combinations of five O₃ levels and four N addition levels. High N addition treatment reduced the accumulated stomatal O₃ uptake in the leaf due to reduced maximum stomatal conductance (g_s). Nitrogen addition also significantly reduced the steady-state light-saturated g_s in August and September. Elevated O₃ significantly reduced and N addition increased total plant biomass; however, there were no significant O₃ × N interactions. The slopes of biomass-based O₃ exposure- and flux-response relationships did not differ significantly among N treatments. The critical levels for a 5% biomass reduction were estimated at 15.4 ppm h and 17.1 mmol O₃ m^-2 projected leaf area (PLA) for Accumulated O₃ exposure Over an hourly Threshold of 40 ppb (AOT40) and Phytotoxic Ozone Dose above a threshold 1 nmol O₃ m^-2 PLA s^-1 (POD₁). These results can facilitate the evaluations of O₃ effect on the carbon-sink capacity and productivity of forest.

Spectrum- and RGB-D-Based Image Fusion for the Prediction of Nitrogen Accumulation in Wheat

The accurate estimation of nitrogen accumulation is of great significance to nitrogen fertilizer management in wheat production. To overcome the shortcomings of spectral technology, which ignores the anisotropy of canopy structure when predicting the nitrogen accumulation in wheat, resulting in low accuracy and unstable prediction results, we propose a method for predicting wheat nitrogen accumulation based on the fusion of spectral and canopy structure features. After depth images are repaired using a hole-filling algorithm, RGB images and depth images are fused through IHS transformation, and textural features of the fused images are then extracted in order to express the three-dimensional structural information of the canopy. The fused images contain depth information of the canopy, which breaks through the limitation of extracting canopy structure features from a two-dimensional image. By comparing the experimental results of multiple regression analyses and BP neural networks, we found that the characteristics of the canopy structure effectively compensated for the model prediction of nitrogen accumulation based only on spectral characteristics. Our prediction model displayed better accuracy and stability, with prediction accuracy values (R2) based on BP neural network for the leaf layer nitrogen accumulation (LNA) and shoot nitrogen accumulation (SNA) during a full growth period of 0.74 and 0.73, respectively, and corresponding relative root mean square errors (RRMSEs) of 40.13% and 35.73%.

Simulation-Based Evaluation of the Estimation Methods of Far-Red Solar-Induced Chlorophyll Fluorescence Escape Probability in Discontinuous Forest Canopies

The escape probability of Solar-induced chlorophyll fluorescence (SIF) can be remotely estimated using reflectance measurements based on spectral invariants theory. This can then be used to correct the effects of canopy structure on canopy-leaving SIF. However, the feasibility of these estimation methods is untested in heterogeneous vegetation such as the discontinuous forest canopy layer under evaluation here. In this study, the Discrete Anisotropic Radiative Transfer (DART) model is used to simulate canopy-leaving SIF, canopy total emitted SIF, canopy interceptance, and the fraction of absorbed photosynthetically active radiation (fAPAR) in order to evaluate the estimation methods of SIF escape probability in discontinuous forest canopies. Our simulation results show that the normalized difference vegetation index (NDVI) can be used to partly eliminate the effects of background reflectance on the estimation of SIF escape probability in most cases, but fails to produce accurate estimations if the background is partly or totally covered by vegetation. We also found that SIF escape probabilities estimated at a high solar zenith angle have better estimation accuracy than those estimated at a lower solar zenith angle. Our results show that additional errors will be introduced to the estimation of SIF escape probability with the use of satellite products, especially when the product of leaf area index (LAI) and clumping index (CI) was underestimated. In other results, fAPAR has comparable estimation accuracy of SIF escape probability when compared to canopy interceptance. Additionally, fAPAR for the entire canopy has better estimation accuracy of SIF escape probability than fPAR for leaf only in sparse forest canopies. These results help us to better understand the current estimation results of SIF escape probability based on spectral invariants theory, and to improve its estimation accuracy in discontinuous forest canopies.

Empirical validation of photon recollision probability in single crowns of tree seedlings

Physically-based methods in remote sensing provide benefits over statistical approaches in monitoring biophysical characteristics of vegetation. However, physically-based models still demand large computational resources and often require rather detailed informative priors on various aspects of vegetation and atmospheric status. Spectral invariants and photon recollision probability theories provide a solid theoretical framework for developing relatively simple models of forest canopy reflectance. Empirical validation of these theories is, however, scarce. Here we present results of a first empirical validation of a model based on photon recollision probability at the level of individual trees. Multiangular spectra of pine, spruce, and oak tree seedlings (height = 0.38-0.7 m) were measured using a goniometer, and tree hemispherical reflectance was derived from those measurements. We evaluated the agreement between modeled and measured tree reflectance. The model predicted the spectral signatures of the tree seedlings in the wavelength range between 400 and 2300 nm well, with wavelength-specific bias between -0.048 and 0.034 in reflectance units. In relative terms, the model errors were the smallest in the near-infrared (relative RMSE up to 4%, 7%, and 4% for pine, spruce, and oak seedlings, respectively) and the largest in the visible wavelength region (relative RMSE up to 34%, 20%, and 60%). The errors in the visible region could be partly attributed to wavelength-dependent directional scattering properties of the leaves. Including woody parts of tree seedlings in the model improved the results by reducing the relative RMSE by up to 10% depending on species and wavelength. Spectrally invariant model parameters, i.e. total and directional escape probabilities, depended on spherically averaged silhouette to total area ratio (STAR) of the tree seedlings. Overall, the modeled and measured tree reflectance mainly agreed within measurement uncertainties, but the results indicate that the assumption of isotropic scattering by the leaves can result in large errors in the visible wavelength region for some tree species. Our results help increasing the confidence when using photon recollision probability and spectral invariants -based models to interpret satellite images, but they also lead to an improved understanding of the assumptions and limitations of these theories.

Foliar functional traits from imaging spectroscopy across biomes in eastern North America

Foliar functional traits are widely used to characterize leaf and canopy properties that drive ecosystem processes and to infer physiological processes in Earth system models. Imaging spectroscopy provides great potential to map foliar traits to characterize continuous functional variation and diversity, but few studies have demonstrated consistent methods for mapping multiple traits across biomes. With airborne imaging spectroscopy data and field data from 19 sites, we developed trait models using partial least squares regression, and mapped 26 foliar traits in seven NEON (National Ecological Observatory Network) ecoregions (domains) including temperate and subtropical forests and grasslands of eastern North America. Model validation accuracy varied among traits (normalized root mean squared error, 9.1-19.4%; coefficient of determination, 0.28-0.82), with phenolic concentration, leaf mass per area and equivalent water thickness performing best across domains. Across all trait maps, 90% of vegetated pixels had reasonable values for one trait, and 28-81% provided high confidence for multiple traits concurrently. Maps of 26 traits and their uncertainties for eastern US NEON sites are available for download, and are being expanded to the western United States and tundra/boreal zone. These data enable better understanding of trait variations and relationships over large areas, calibration of ecosystem models, and assessment of continental-scale functional diversity.

Data-Driven Methods for the Estimation of Leaf Water and Dry Matter Content: Performances, Potential and Limitations

Leaf equivalent water thickness (EWT) and dry matter content (expressed as leaf mass per area (LMA)) are two critical traits for vegetation function monitoring, crop yield estimation, and precise agriculture management. Data-driven methods are widely used for remote sensing of leaf EWT and LMA because of their simplicity, satisfactory accuracy, and computation efficiency, such as the vegetation indices (VI)-based and machine learning (ML)-based methods. However, most of the data-driven methods are utilized at the canopy level, comparison of the performances of the data-driven methods at the leaf level has not been well documented. Moreover, the ML-based data-driven methods generally adopt leaf optical properties directly as their inputs, which may subsequently decrease their ability in remote sensing of leaf biochemical constituents. Performances of the ML-based methods cooperating with VI are rarely evaluated. Using the independent LOPEX and ANGERS datasets, we compared the performances of three data-driven methods: VI-based, ML-reflectance-based, and ML-VI-based methods, for the estimation of leaf EWT and LMA. Three sampling strategies were also utilized for evaluation of the generalization of these data-driven methods. Our results evidenced that ML-VI-based methods were the most accurate among these data-driven methods. Compared to the ML-reflectance-based and VI-based methods, the ML-VI-based model with support vector regression overall reduced errors by 5.7% (41.5%) and 1.8% (12.4%) for the estimation of leaf EWT (LMA), respectively. The ML-VI-based model inherits advantages of vegetation indices and ML techniques, which made it sensitive to changes of leaf biochemical constituents and capable of solving nonlinear tasks. It is thus recommended for the estimation of EWT and LMA at the leaf level. Moreover, its performance can further be enhanced by improving its generalization ability, such as adopting techniques on the selection of better wavelengths and definition of new vegetation indices. These results thus provided a prior knowledge of the data-driven methods and can be helpful for future studies on the remote sensing of leaf biochemical constituents.

Ensemble projections elucidate effects of uncertainty in terrestrial nitrogen limitation on future carbon uptake

The magnitude of the nitrogen (N) limitation of terrestrial carbon (C) storage over the 21st century is highly uncertain because of the complex interactions between the terrestrial C and N cycles. We use an ensemble approach to quantify and attribute process-level uncertainty in C-cycle projections by analysing a 30-member ensemble representing published alternative representations of key N cycle processes (stoichiometry, biological nitrogen fixation (BNF) and ecosystem N losses) within the framework of one terrestrial biosphere model. Despite large differences in the simulated present-day N cycle, primarily affecting simulated productivity north of 40°N, ensemble members generally conform with global C-cycle benchmarks for present-day conditions. Ensemble projections for two representative concentration pathways (RCP 2.6 and RCP 8.5) show that the increase in land C storage due to CO₂ fertilization is reduced by 24 ± 15% due to N constraints, whereas terrestrial C losses associated with climate change are attenuated by 19 ± 20%. As a result, N cycling reduces projected land C uptake for the years 2006-2099 by 19% (37% decrease to 3% increase) for RCP 2.6, and by 21% (40% decrease to 9% increase) for RCP 8.5. Most of the ensemble spread results from uncertainty in temperate and boreal forests, and is dominated by uncertainty in BNF (10% decrease to 50% increase for RCP 2.6, 5% decrease to 100% increase for RCP 8.5). However, choices about the flexibility of ecosystem C:N ratios and processes controlling ecosystem N losses regionally also play important roles. The findings of this study demonstrate clearly the need for an ensemble approach to quantify likely future terrestrial C-N cycle trajectories. Present-day C-cycle observations only weakly constrain the future ensemble spread, highlighting the need for better observational constraints on large-scale N cycling, and N cycle process responses to global change.

Estimation of Gross Primary Productivity (GPP) Phenology of a Short-Rotation Plantation Using Remotely Sensed Indices Derived from Sentinel-2 Images

This study aimed to understand which vegetation indices (VIs) are an ideal proxy for describing phenology and interannual variability of Gross Primary Productivity (GPP) in short-rotation coppice (SRC) plantations. Canopy structure- and chlorophyll-sensitive VIs derived from Sentinel-2 images were used to estimate the start and end of the growing season (SOS and EOS, respectively) during the period 2016–2018, for an SRC poplar (Populus spp.) plantation in Lochristi (Belgium). Three different filtering methods (Savitzky–Golay (SavGol), polynomial (Polyfit) and Harmonic Analysis of Time Series (HANTS)) and five SOS- and EOS threshold methods (first derivative function, 10% and 20% percentages and 10% and 20% percentiles) were applied to identify the optimal methods for the determination of phenophases. Our results showed that the MEdium Resolution Imaging Spectrometer (MERIS) Terrestrial Chlorophyll Index (MTCI) had the best fit with GPP phenology, as derived from eddy covariance measurements, in identifying SOS- and EOS-dates. For SOS, the performance was only slightly better than for several other indices, whereas for EOS, MTCI performed markedly better. The relationship between SOS/EOS derived from GPP and VIs varied interannually. MTCI described best the seasonal pattern of the SRC plantation’s GPP (R2 = 0.52 when combining all three years). However, during the extreme dry year 2018, the Chlorophyll Red Edge Index performed slightly better in reproducing growing season GPP variability than MTCI (R2 = 0.59; R2 = 0.49, respectively). Regarding smoothing functions, Polyfit and HANTS methods showed the best (and very similar) performances. We further found that defining SOS as the date at which the 10% or 20% percentile occurred, yielded the best agreement between the VIs and the GPP; while for EOS the dates of the 10% percentile threshold came out as the best.

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, Narea) 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 Narea 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 Narea. 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.

Application of organic carbon affects mineral nitrogen uptake by winter wheat and leaching in subsoil: Proximal sensing as a tool for agronomic practice

We tested the hypothesis that application of stable forms of organic carbon (C) into the soil reduces leaching of nitrogen (N). We also examined the potential to estimate N leaching employing N-sensitive spectral reflectance indices. During three growing seasons 2013-2015, field experiment at two experimental sites combining application of distinct N doses (0 (N₀), 35 (N₃₅), 70 (N₇₀), and 140 (N₁₄₀) kg N ha^-1) and two stable forms of organic C (lignohumate and compost) was established to measure N uptake by winter wheat and its leaching to subsoil layers. The spectral reflectance at canopy level was measured simultaneously with N content in leaf dry matter at the beginning of the grain filling phase. At full maturity, the above-ground biomass, grain yield, and grain protein content were evaluated. That data was used to calculate N uptake in grain. The N₁₄₀ dose led to increased N uptake by grain of 64% and 73% in the wetter years 2013 and 2014, respectively, and even by 118% in the drier year 2015 in comparison with the N₀ treatment. N leaching to subsoil increased substantially with higher N dose, but only in wetter years 2013 (by 74%) and 2014 (by 87%). By contrast, no effect of N dose on leached N was found in the dry year 2015. The application of organic C along with the N₁₄₀ dose substantially reduced N leaching by 26% and 29% in 2014 and 2015, respectively. Moreover, we demonstrated that normalized red-edge spectral reflectance index (NRERI) is able to predict N uptake by wheat and it can serve as an indicator of N leaching in heavy-rainfall years. Our results thus point towards possible agronomic practices and use of remote-sensing techniques to reduce groundwater contamination by N-based fertilizers.

Leaf Photosynthetic Capacity of Sunlit and Shaded Mature Leaves in a Deciduous Forest

A clear understanding of the dynamics of photosynthetic capacity is crucial for accurate modeling of ecosystem carbon uptake. However, such dynamical information is hardly available and has dramatically impeded our understanding of carbon cycles. Although tremendous efforts have been made in coupling the dynamic information of photosynthetic capacity into models, using “proxies” rooted from the close relationships between photosynthetic capacity and other available leaf parameters remains the popular selection. Unfortunately, no consensus has yet been reached on such “proxies”, leading them only applicable to limited cases. In this study, we aim to identify if there are close relationships between the photosynthetic capacity (represented by the maximum carboxylation rate, Vcmax) and leaf traits for mature broadleaves within a cold temperature deciduous forest. This is based on a long-term in situ dataset including leaf chlorophyll content (Chl), leaf nitrogen concentration (Narea, Nmass), leaf carbon concentration (Carea, Cmass), equivalent water thickness (EWT), leaf mass per area (LMA), and leaf gas exchange measurements from which Vcmax was derived, for both sunlit and shaded leaves during leaf mature periods from 2014 to 2019. The results show that the Vcmax values of sunlit and shaded leaves were relatively stable during these periods, and no statistically significant interannual variations occurred (p > 0.05). However, this is not applicable to specific species. Path analysis revealed that Narea was the major contributor to Vcmax for sunlit leaves (0.502), while LMA had the greatest direct relationship with Vcmax for shaded leaves (0.625). The LMA has further been confirmed as a primary proxy if no leaf type information is available. These findings provide a promising way to better understand photosynthesis and to predict carbon and water cycles in temperate deciduous forests.

UNVI-Based Time Series for Vegetation Discrimination Using Separability Analysis and Random Forest Classification

Land cover data is crucial for earth system modelling, natural resources management, and conservation planning. Remotely sensed time-series data capture dynamic behavior of vegetation, and have been widely used for land cover mapping. Temporal profiles of vegetation index (VI), especially normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI), are the most used features derived from time-series spectral data. Whether NDVI or EVI is optimal to generate temporal profiles has not been evaluated. The universal normalized vegetation index (UNVI), a relatively new index with all spectral bands incorporated, has been proved to be more effective than several commonly used satellite-derived VIs in some application scenarios. In this study, we explored the ability of UNVI time series for discriminating different vegetation types in Chaoyang prefecture, northeast China, in comparison with normalized NDVI, EVI, triangle vegetation index (TVI), and tasseled cap transformation greenness (TCG). These five indices were calculated using Landsat 8 surface reflectance data, and two comparative experiments were conducted. The first experiment analyzed class separabilities using pairwise JM (Jeffries–Matusita) distance as indicator, and the results showed that UNVI was superior to EVI, TVI, and TCG, and almost equivalent to NDVI, especially during the peak of vegetation growing season and for the most indistinguishable vegetation pair broadleaf and shrubs. The second experiment compared the vegetation classification accuracies using the features of these VI temporal profiles and the corresponding phenological parameters, and the results showed that UNVI can better classify the five major vegetation in Chaoyang prefecture than other four indices. Therefore, we conclude that UNVI time series has considerable potential for regional land cover mapping, and we recommend that the use of the UNVI is considered in the future time series related studies.

Towards comparable assessment of the soil nutrient status across scales-Review and development of nutrient metrics

Nutrient availability influences virtually every aspect of an ecosystem, and is a critical modifier of ecosystem responses to global change. Although this crucial role of nutrient availability in regulating ecosystem structure and functioning has been widely acknowledged, nutrients are still often neglected in observational and experimental synthesis studies due to difficulties in comparing the nutrient status across sites. In the current study, we explain different nutrient-related concepts and discuss the potential of soil-, plant- and remote sensing-based metrics to compare the nutrient status across space. Based on our review and additional analyses on a dataset of European, managed temperate and boreal forests (ICP [International Co-operative Programme on Assessment and Monitoring of Air Pollution Effects on Forests] Forests dataset), we conclude that the use of plant- and remote sensing-based metrics that rely on tissue stoichiometry is limited due to their strong dependence on species identity. The potential use of other plant-based metrics such as Ellenberg indicator values and plant-functional traits is also discussed. We conclude from our analyses and review that soil-based metrics have the highest potential for successful intersite comparison of the nutrient status. As an example, we used and adjusted a soil-based metric, previously developed for conifer forests across Sweden, against the same ICP Forests data. We suggest that this adjusted and further adaptable metric, which included the organic carbon concentration in the upper 20 cm of the soil (including the organic fermentation-humus [FH] layer), the C:N ratio and pH CaCl 2 of the FH layer, can be used as a complementary tool along with other indicators of nutrient availability, to compare the background nutrient status across temperate and boreal forests dominated by spruce, pine or beech. Future collection and provision of harmonized soil data from observational and experimental sites is crucial for further testing and adjusting the metric.

Novel Combined Spectral Indices Derived from Hyperspectral and Laser-Induced Fluorescence LiDAR Spectra for Leaf Nitrogen Contents Estimation of Rice

Spectra of reflectance (Sr) and fluorescence (Sf) are significant for crop monitoring and ecological environment research, and can be used to indicate the leaf nitrogen content (LNC) of crops indirectly. The aim of this work is to use the Sr-Sf features obtained with hyperspectral and laser-induced fluorescence LiDAR (HSL, LIFL) systems to construct novel combined spectral indices (NCIH-F) for multi-year rice LNC estimation. The NCIH-F is in a form of FWs* Φ + GSIs* Φ , where Φ is the Sr-Sf features, and FWs and GSIs are the feature weights and global sensitive indices for each characteristic band. In this study, the characteristic bands were chosen in different ways. Firstly, the Sr-Sf characteristics which can be the intensity or derivative variables of spectra in 685 and 740 nm, have been assigned as the Φ value in NCIH-F formula. Simultaneously, the photochemical reflectance index (PRI) formed with 531 and 570 nm was modified based on a variant spectral index, called PRIfraction, with the Sf intensity in 740 nm, and then compared its potential with NCIH-F on LNC estimation. During the above analysis, both NCIH-F and PRIfraction values were utilized to model rice LNC based on the artificial neural networks (ANNs) method. Subsequently, four prior bands were selected, respectively, with high FW and GSI values as the ANNs inputs for rice LNC estimation. Results show that FW- and GSI-based NCIH-F are closely related to rice LNC, and the performance of previous spectral indices used for LNC estimation can be greatly improved by multiplying their FWs and GSIs. Thus, it can be included that the FW- and GSI-based NCIH-F constitutes an efficient and reliable constructed form combining HSL (Sr) and LIFL (Sf) data together for rice LNC estimation.

Watson on the Farm: Using Cloud-Based Artificial Intelligence to Identify Early Indicators of Water Stress

As demand for freshwater increases while supply remains stagnant, the critical need for sustainable water use in agriculture has led the EPA Strategic Plan to call for new technologies that can optimize water allocation in real-time. This work assesses the use of cloud-based artificial intelligence to detect early indicators of water stress across six container-grown ornamental shrub species. Near-infrared images were previously collected with modified Canon and MAPIR Survey II cameras deployed via a small unmanned aircraft system (sUAS) at an altitude of 30 meters. Cropped images of plants in no, low-, and high-water stress conditions were split into four-fold cross-validation sets and used to train models through IBM Watson’s Visual Recognition service. Despite constraints such as small sample size (36 plants, 150 images) and low image resolution (150 pixels by 150 pixels per plant), Watson generated models were able to detect indicators of stress after 48 hours of water deprivation with a significant to marginally significant degree of separation in four out of five species tested (p < 0.10). Two models were also able to detect indicators of water stress after only 24 hours, with models trained on images of as few as eight water-stressed Buddleia plants achieving an average area under the curve (AUC) of 0.9884 across four folds. Ease of pre-processing, minimal amount of training data required, and outsourced computation make cloud-based artificial intelligence services such as IBM Watson Visual Recognition an attractive tool for agriculture analytics. Cloud-based artificial intelligence can be combined with technologies such as sUAS and spectral imaging to help crop producers identify deficient irrigation strategies and intervene before crop value is diminished. When brought to scale, frameworks such as these can drive responsive irrigation systems that monitor crop status in real-time and maximize sustainable water use.

Nitrogen and Phosphorus effect on Sun-Induced Fluorescence and Gross Primary Productivity in Mediterranean Grassland

Sun-Induced fluorescence at 760 nm (F760) is increasingly being used to predict gross primary production (GPP) through light use efficiency (LUE) modeling, even though the mechanistic processes that link the two are not well understood. We analyzed the effect of nitrogen (N) and phosphorous (P) availability on the processes that link GPP and F760 in a Mediterranean grassland manipulated with nutrient addition. To do so, we used a combination of process-based modeling with Soil-Canopy Observation of Photosynthesis and Energy (SCOPE), and statistical analyses such as path modeling. With this study, we uncover the mechanisms that link the fertilization-driven changes in canopy nitrogen concentration (N%) to the observed changes in F760 and GPP. N addition changed plant community structure and increased canopy chlorophyll content, which jointly led to changes in photosynthetic active radiation (APAR), ultimately affecting both GPP and F760. Changes in the abundance of graminoids, (%graminoids) driven by N addition led to changes in structural properties of the canopy such as leaf angle distribution, that ultimately influenced observed F760 by controlling the escape probability of F760 (Fesc). In particular, we found a change in GPP–F760 relationship between the first and the second year of the experiment that was largely driven by the effect of plant type composition on Fesc, whose best predictor is %graminoids. The P addition led to a statistically significant increase on light use efficiency of fluorescence emission (LUEf), in particular in plots also with N addition, consistent with leaf level studies. The N addition induced changes in the biophysical properties of the canopy that led to a trade-off between surface temperature (Ts), which decreased, and F760 at leaf scale (F760leaf,fw), which increased. We found that Ts is an important predictor of the light use efficiency of photosynthesis, indicating the importance of Ts in LUE modeling approaches to predict GPP.

A Global Sensitivity Analysis of Commonly Used Satellite-Derived Vegetation Indices for Homogeneous Canopies Based on Model Simulation and Random Forest Learning

Remote sensing (RS) provides operational monitoring of terrestrial vegetation. For optical RS, vegetation information is generally derived from surface reflectance (ρ). More generally, vegetation indices (VIs) are built on the basis of ρ as proxies for vegetation traits. At canopy level, ρ can be affected by a variety of factors, including leaf constituents, canopy structure, background reflectivity, and sun-sensor geometry. Consequently, VIs are mixtures of different information. In this study, a global sensitivity analysis (GSA) is made for several commonly used satellite-derived VIs in order to better understand the application of these VIs at large scales. The sensitivities of VIs to different parameters are analyzed on the basis of PROSPECT-SAIL (PROSAIL) radiative transfer model simulations, which apply for homogeneous canopies, and random forest (RF) learning. Specifically, combined factors such as canopy chlorophyll content (CCC) and canopy water content (CWC) are introduced in the RF-based GSA. We find that for most VIs, the leaf area index is the most influential factor, while the broad-band sensor-derived enhanced VI (EVI) exhibits a strong sensitivity to CCC, and the universal normalized VI (UNVI) is sensitive to CWC. The potential and uncertainty for the application of all the considered VIs are analyzed according to the GSA results. The results can help to improve the use of VIs in different contexts, and the RF-based GSA method can be further applied in more sophisticated situations.

Regional Scale Dryland Vegetation Classification with an Integrated Lidar-Hyperspectral Approach

The sparse canopy cover and large contribution of bright background soil, along with the heterogeneous vegetation types in close proximity, are common challenges for mapping dryland vegetation with remote sensing. Consequently, the results of a single classification algorithm or one type of sensor to characterize dryland vegetation typically show low accuracy and lack robustness. In our study, we improved classification accuracy in a semi-arid ecosystem based on the use of vegetation optical (hyperspectral) and structural (lidar) information combined with the environmental characteristics of the landscape. To accomplish this goal, we used both spectral angle mapper (SAM) and multiple endmember spectral mixture analysis (MESMA) for optical vegetation classification. Lidar-derived maximum vegetation height and delineated riparian zones were then used to modify the optical classification. Incorporating the lidar information into the classification scheme increased the overall accuracy from 60% to 89%. Canopy structure can have a strong influence on spectral variability and the lidar provided complementary information for SAM’s sensitivity to shape but not magnitude of the spectra. Similar approaches to map large regions of drylands with low uncertainty may be readily implemented with unmixing algorithms applied to upcoming space-based imaging spectroscopy and lidar. This study advances our understanding of the nuances associated with mapping xeric and mesic regions, and highlights the importance of incorporating complementary algorithms and sensors to accurately characterize the heterogeneity of dryland ecosystems.

Exploring the spatial relationship between airborne-derived red and far-red sun-induced fluorescence and process-based GPP estimates in a forest ecosystem

Terrestrial gross primary productivity (GPP) plays an essential role in the global carbon cycle, but the quantification of the spatial and temporal variations in photosynthesis is still largely uncertain. Our work aimed to investigate the potential of remote sensing to provide new insights into plant photosynthesis at a fine spatial resolution. This goal was achieved by exploiting high-resolution images acquired with the FLuorescence EXplorer (FLEX) airborne demonstrator HyPlant. The sensor was flown over a mixed forest, and the images collected were elaborated to obtain two independent indicators of plant photosynthesis. First, maps of sun-induced chlorophyll fluorescence (F), a novel indicator of plant photosynthetic activity, were successfully obtained at both the red and far-red peaks (r2 = 0.89 and p < 0.01, r2 = 0.77 and p < 0.01, respectively, compared to top-of-canopy ground-based measurements acquired synchronously with the overflight) over the forested study area. Second, maps of GPP and absorbed photosynthetically active radiation (APAR) were derived using a customised version of the coupled biophysical model Breathing Earth System Simulator (BESS). The model was driven with airborne-derived maps of key forest traits (i.e., leaf chlorophyll content (LCC) and leaf area index (LAI)) and meteorological data providing a high-resolution snapshot of the variables of interest across the study site. The LCC and LAI were accurately estimated (RMSE = 5.66 μg cm-2 and RMSE = 0.51 m2m-2, respectively) through an optimised Look-Up-Table-based inversion of the PROSPECT-4-INFORM radiative transfer model, ensuring the accurate representation of the spatial variation of these determinants of the ecosystem's functionality. The spatial relationships between the measured F and modelled BESS outputs were then analysed to interpret the variability of ecosystem functioning at a regional scale. The results showed that far-red F is significantly correlated with the GPP (r2 = 0.46, p < 0.001) and APAR (r2 = 0.43, p < 0.001) in the spatial domain and that this relationship is nonlinear. Conversely, no statistically significant relationships were found between the red F and the GPP or APAR (p > 0.05). The spatial relationships found at high resolution provide valuable insight into the critical role of spatial heterogeneity in controlling the relationship between the far-red F and the GPP, indicating the need to consider this heterogeneity at a coarser resolution.

Characterizing Boreal Peatland Plant Composition and Species Diversity with Hyperspectral Remote Sensing

Peatlands, which account for approximately 15% of land surface across the arctic and boreal regions of the globe, are experiencing a range of ecological impacts as a result of climate change. Factors that include altered hydrology resulting from drought and permafrost thaw, rising temperatures, and elevated levels of atmospheric carbon dioxide have been shown to cause plant community compositional changes. Shifts in plant composition affect the productivity, species diversity, and carbon cycling of peatlands. We used hyperspectral remote sensing to characterize the response of boreal peatland plant composition and species diversity to warming, hydrologic change, and elevated CO2. Hyperspectral remote sensing techniques offer the ability to complete landscape-scale analyses of ecological responses to climate disturbance when paired with plot-level measurements that link ecosystem biophysical properties with spectral reflectance signatures. Working within two large ecosystem manipulation experiments, we examined climate controls on composition and diversity in two types of common boreal peatlands: a nutrient rich fen located at the Alaska Peatland Experiment (APEX) in central Alaska, and an ombrotrophic bog located in northern Minnesota at the Spruce and Peatland Responses Under Changing Environments (SPRUCE) experiment. We found a strong effect of plant functional cover on spectral reflectance characteristics. We also found a positive relationship between species diversity and spectral variation at the APEX field site, which is consistent with other recently published findings. Based on the results of our field study, we performed a supervised land cover classification analysis on an aerial hyperspectral dataset to map peatland plant functional types (PFTs) across an area encompassing a range of different plant communities. Our results underscore recent advances in the application of remote sensing measurements to ecological research, particularly in far northern ecosystems.

Hyperspectral Leaf Reflectance as Proxy for Photosynthetic Capacities: An Ensemble Approach Based on Multiple Machine Learning Algorithms

Global agriculture production is challenged by increasing demands from rising population and a changing climate, which may be alleviated through development of genetically improved crop cultivars. Research into increasing photosynthetic energy conversion efficiency has proposed many strategies to improve production but have yet to yield real-world solutions, largely because of a phenotyping bottleneck. Partial least squares regression (PLSR) is a statistical technique that is increasingly used to relate hyperspectral reflectance to key photosynthetic capacities associated with carbon uptake (maximum carboxylation rate of Rubisco, Vc,max ) and conversion of light energy (maximum electron transport rate supporting RuBP regeneration, Jmax ) to alleviate this bottleneck. However, its performance varies significantly across different plant species, regions, and growth environments. Thus, to cope with the heterogeneous performances of PLSR, this study aims to develop a new approach to estimate photosynthetic capacities. A framework was developed that combines six machine learning algorithms, including artificial neural network (ANN), support vector machine (SVM), least absolute shrinkage and selection operator (LASSO), random forest (RF), Gaussian process (GP), and PLSR to optimize high-throughput analysis of the two photosynthetic variables. Six tobacco genotypes, including both transgenic and wild-type lines, with a range of photosynthetic capacities were used to test the framework. Leaf reflectance spectra were measured from 400 to 2500 nm using a high-spectral-resolution spectroradiometer. Corresponding photosynthesis vs. intercellular CO2 concentration response curves were measured for each leaf using a leaf gas-exchange system. Results suggested that the mean R 2 value of the six regression techniques for predicting Vc,max (Jmax ) ranged from 0.60 (0.45) to 0.65 (0.56) with the mean RMSE value varying from 47.1 (40.1) to 54.0 (44.7) μmol m-2 s-1. Regression stacking for Vc,max (Jmax ) performed better than the individual regression techniques with increases in R 2 of 0.1 (0.08) and decreases in RMSE by 4.1 (6.6) μmol m-2 s-1, equal to 8% (15%) reduction in RMSE. Better predictive performance of the regression stacking is likely attributed to the varying coefficients (or weights) in the level-2 model (the LASSO model) and the diverse ability of each individual regression technique to utilize spectral information for the best modeling performance. Further refinements can be made to apply this stacked regression technique to other plant phenotypic traits.

Exploiting genetic variation in nitrogen use efficiency for cereal crop improvement

Cereals are the most important sources of calories and nutrition for the human population, and are an essential animal feed. Food security depends on adequate production and demands are predicted to rise as the global population rises. The need for increased yields will have to be coupled to the efficient use of resources including fertilisers such as nitrogen to underpin the sustainability of food production. Although optimally performing crops with high yields require a balanced mineral nutrition, nitrogen fundamentally drives growth and yield as well as requirements for other nutrients. It is estimated that globally only 33% of applied nitrogen fertiliser is recovered in the harvested grain, indicative of a huge waste of resource and potential major pollutant and is thus a major target for crop improvement. Both agronomy and breeding will contribute to improved nitrogen use efficiency (NUE) and an important component of the latter is harnessing germplasm variation. This review will consider the key traits involved in NUE, the potential to exploit genetic variation for these specific traits, and the approaches to be utilised.

Variability and Uncertainty Challenges in Scaling Imaging Spectroscopy Retrievals and Validations from Leaves Up to Vegetation Canopies

Imaging spectroscopy of vegetation requires methods for scaling and generalizing optical signals that are reflected, transmitted and emitted in the solar wavelength domain from single leaves and observed at the level of canopies by proximal sensing, airborne and satellite spectroradiometers. The upscaling embedded in imaging spectroscopy retrievals and validations of plant biochemical and structural traits is challenged by natural variability and measurement uncertainties. Sources of the leaf-to-canopy upscaling variability and uncertainties are reviewed with respect to: (1) implementation of retrieval algorithms and (2) their parameterization and validation of quantitative products through in situ field measurements. The challenges are outlined and discussed for empirical and physical leaf and canopy radiative transfer modelling components, considering both forward and inverse modes. Discussion on optical remote sensing validation schemes includes also description of a multiscale validation concept and its advantages. Impacts of intraspecific and interspecific variability on collected field and laboratory measurements of leaf biochemical traits and optical properties are demonstrated for selected plant species, and field measurement uncertainty sources are listed and discussed specifically for foliar pigments and canopy leaf area index. The review concludes with the main findings and suggestions as how to reduce uncertainties and include variability in scaling vegetation imaging spectroscopy signals and functional traits of single leaves up to observations of whole canopies.

Spatial and Temporal Monitoring of Pasture Ecological Quality: Sentinel-2-Based Estimation of Crude Protein and Neutral Detergent Fiber Contents

Frequent, region-wide monitoring of changes in pasture quality due to human disturbances or climatic conditions is impossible by field measurements or traditional ecological surveying methods. Remote sensing imagery offers distinctive advantages for monitoring spatial and temporal patterns. The chemical parameters that are widely used as indicators of ecological quality are crude protein (CP) content and neutral detergent fiber (NDF) content. In this study, we investigated the relationship between CP, NDF, and reflectance in the visible–near-infrared–shortwave infrared (VIS–NIR–SWIR) spectral range, using field, laboratory measurements, and satellite imagery (Sentinel-2). Statistical models were developed using different calibration and validation data sample sets: (1) a mix of laboratory and field measurements (e.g., fresh and dry vegetation) and (2) random selection. In addition, we used three vegetation indices (Normalized Difference Vegetative Index (NDVI), Soil-adjusted Vegetation Index (SAVI) and Wide Dynamic Range Vegetation Index (WDRVI)) as proxies to CP and NDF estimation. The best models found for predicting CP and NDF contents were based on reflectance measurements (R2 = 0.71, RMSEP = 2.1% for CP; and R2 = 0.78, RMSEP = 5.5% for NDF). These models contained fresh and dry vegetation samples in calibration and validation data sets. Random sample selection in a model generated similar accuracy estimations. Our results also indicate that vegetation indices provide poor accuracy. Eight Sentinel-2 images (December 2015–April 2017) were examined in order to better understand the variability of vegetation quality over spatial and temporal scales. The spatial and temporal patterns of CP and NDF contents exhibit strong seasonal dependence, influenced by climatological (precipitation) and topographical (northern vs. southern hillslopes) conditions. The total CP/NDF content increases/decrease (respectively) from December to March, when the concentrations reach their maximum/minimum values, followed by a decline/incline that begins in April, reaching minimum values in July.

Assessing Across-Scale Optical Diversity and Productivity Relationships in Grasslands of the Italian Alps

The linearity and scale-dependency of ecosystem biodiversity and productivity relationships (BPRs) have been under intense debate. In a changing climate, monitoring BPRs within and across different ecosystem types is crucial, and novel remote sensing tools such as the Sentinel-2 (S2) may be adopted to retrieve ecosystem diversity information and to investigate optical diversity and productivity patterns. But are the S2 spectral and spatial resolutions suitable to detect relationships between optical diversity and productivity? In this study, we implemented an integrated analysis of spatial patterns of grassland productivity and optical diversity using optical remote sensing and Eddy Covariance data. Across-scale optical diversity and ecosystem productivity patterns were analyzed for different grassland associations with a wide range of productivity. Using airborne optical data to simulate S2, we provided empirical evidence that the best optical proxies of ecosystem productivity were linearly correlated with optical diversity. Correlation analysis at increasing pixel sizes proved an evident scale-dependency of the relationships between optical diversity and productivity. The results indicate the strong potential of S2 for future large-scale assessment of across-ecosystem dynamics at upper levels of observation.

Prediction of cadmium concentration in brown rice before harvest by hyperspectral remote sensing

Cadmium (Cd) contaminated rice has become a global food security issue. Hyperspectral remote sensing can do rapid and nondestructive monitoring of environmental stress in plant. To realize the nondestructive detection of Cd in brown rice before harvest, the leaf spectral reflectance of rice exposed to six different levels of Cd stress was measured during the whole life stages. In addition, the dry weight of rice grain and Cd concentrations in brown rice were measured after harvest. The impact of Cd stress on the quantity and the quality of rice grain and on the leaf reflectance of rice was analyzed, and hyperspectral estimation models for predicting the Cd content in brown rice during three growth stages were established. The results showed that rice plants can impact the quality of the brown rice seriously, even if the impact on the quantity was not significant. All the established models had the capability to estimate Cd concentrations in brown rice (R² > 0.598), and the best performance model, with the R² value of 0.873, was use first derivative spectrum of booting stage as variable. It was concluded that the hyperspectral of rice leaves provides a new insight to predict Cd concentration in brown rice before harvest.

Decoupling Canopy Structure and Leaf Biochemistry: Testing the Utility of Directional Area Scattering Factor (DASF)

Biochemical properties retrieved from remote sensing data are crucial sources of information for many applications. However, leaf and canopy scattering processes must be accounted for to reliably estimate information on canopy biochemistry, carbon-cycle processes and energy exchange. A coupled leaf-canopy model based on spectral invariants theory has been proposed, that uses the so-called Directional Area Scattering Factor (DASF) to correct hyperspectral remote sensing data for canopy structural effects. In this study, the reliability of DASF to decouple canopy structure and biochemistry was empirically tested using simulated reflectance spectra modelled using a Monte Carlo Ray Tracing (MCRT) radiative transfer model. This approach allows all canopy and radiative properties to be specified a priori. Simulations were performed under idealised conditions of directional-hemispherical reflectance, isotropic Lambertian leaf reflectance and transmittance and sufficiently dense (high LAI) canopies with black soil where the impact of canopy background is negligible, and also departures from these conditions. It was shown that both DASF and total canopy scattering could be accurately extracted under idealised conditions using information from both the full 400–2500 nm spectral interval and the 710–790 nm interval alone, even given no prior knowledge of leaf optical properties. Departures from these idealised conditions: varying view geometry, bi-directional reflectance, LAI and soil effects, were tested. We demonstrate that total canopy scattering could be retrieved under conditions of varying view geometry and bi-directional reflectance, but LAI and soil effects were shown to reduce the accuracy with which the scattering can be modelled using the DASF approach. We show that canopy architecture, either homogeneous or heterogeneous 3D arrangements of canopy scattering elements, has important influences over DASF and consequently the accuracy of retrieval of total canopy scattering. Finally, although DASF and total canopy scattering could be retrieved to within 2.4% of the modelled total canopy scattering signal given no prior knowledge of leaf optical properties, spectral invariant parameters were not accurately retrieved from the simulated signal. This has important consequences since these parameters are quite widely used in canopy reflectance modelling and have the potential to help derive new, more accurate canopy biophysical information. Understanding and quantifying the limitations of the DASF approach as we have done here, is an important step in allowing the wider use of these methods for decoupling canopy structure and biochemistry.

Improving Land Cover Classifications with Multiangular Data: MISR Data in Mainland Spain

In this study, we deal with the application of multiangular data from the Multiangle Imaging Spectroradiometer (MISR) sensor for studying the effect of surface anisotropy and directional information on the classification accuracy for different land covers with different rate of disaggregation classes (from four to 35 different classes) from a Mediterranean bioregion in Iberian, Spain. We used various MISR band groups from nadir to blue, green, red, and NIR channels at nadir and off-nadir. The MISR data utilized here were provided by the L1B2T product (275 m spatial resolution) and belonged to two different orbits. We performed 23 classifications with the k-means algorithm to test multiangular data, number of clusters, and iteration effects. Our findings confirmed that the multiangular information, in addition to the multispectral information used as the input of the k-means algorithm, improves the land cover classification accuracy, and this improvement increased with the level of disaggregation. A very large number of clusters produced even better improvements than multiangular data.

Influence of Leaf Specular Reflection on Canopy Radiative Regime Using an Improved Version of the Stochastic Radiative Transfer Model

Interpreting remotely-sensed data requires realistic, but simple, models of radiative transfer that occurs within a vegetation canopy. In this paper, an improved version of the stochastic radiative transfer model (SRTM) is proposed by assuming that all photons that have not been specularly reflected enter the leaf interior. The contribution of leaf specular reflection is considered by modifying leaf scattering phase function using Fresnel reflectance. The canopy bidirectional reflectance factor (BRF) estimated from this model is evaluated through comparisons with field-measured maize BRF. The result shows that accounting for leaf specular reflection can provide better performance than that when leaf specular reflection is neglected over a wide range of view zenith angles. The improved version of the SRTM is further adopted to investigate the influence of leaf specular reflection on the canopy radiative regime, with emphases on vertical profiles of mean radiation flux density, canopy absorptance, BRF, and normalized difference vegetation index (NDVI). It is demonstrated that accounting for leaf specular reflection can increase leaf albedo, which consequently increases canopy mean upward/downward mean radiation flux density and canopy nadir BRF and decreases canopy absorptance and canopy nadir NDVI when leaf angles are spherically distributed. The influence is greater for downward/upward radiation flux densities and canopy nadir BRF than that for canopy absorptance and NDVI. The results provide knowledge of leaf specular reflection and canopy radiative regime, and are helpful for forward reflectance simulations and backward inversions. Moreover, polarization measurements are suggested for studies of leaf specular reflection, as leaf specular reflection is closely related to the canopy polarization.

Imaging Spectroscopic Analysis of Biochemical Traits for Shrub Species in Great Basin, USA

The biochemical traits of plant canopies are important predictors of photosynthetic capacity and nutrient cycling. However, remote sensing of biochemical traits in shrub species in dryland ecosystems has been limited mainly due to the sparse vegetation cover, manifold shrub structures, and complex light interaction between the land surface and canopy. In order to examine the performance of airborne imaging spectroscopy for retrieving biochemical traits in shrub species, we collected Airborne Visible Infrared Imaging Spectrometer—Next Generation (AVIRIS-NG) images and surveyed four foliar biochemical traits (leaf mass per area, water content, nitrogen content and carbon) of sagebrush (Artemesia tridentata) and bitterbrush (Purshia tridentata) in the Great Basin semi-desert ecoregion, USA, in October 2014 and May 2015. We examined the correlations between biochemical traits and developed partial least square regression (PLSR) models to compare spectral correlations with biochemical traits at canopy and plot levels. PLSR models for sagebrush showed comparable performance between calibration (R2: LMA = 0.66, water = 0.7, nitrogen = 0.42, carbon = 0.6) and validation (R2: LMA = 0.52, water = 0.41, nitrogen = 0.23, carbon = 0.57), while prediction for bitterbrush remained a challenge. Our results demonstrate the potential for airborne imaging spectroscopy to measure shrub biochemical traits over large shrubland regions. We also highlight challenges when estimating biochemical traits with airborne imaging spectroscopy data.

Testing of Automated Photochemical Reflectance Index Sensors as Proxy Measurements of Light Use Efficiency in an Aspen Forest

Commercially available autonomous photochemical reflectance index (PRI) sensors are a new development in the remote sensing field that offer novel opportunities for a deeper exploration of vegetation physiology dynamics. In this study, we evaluated the reliability of autonomous PRI sensors (SRS-PRI) developed by METER Group Inc. as proxies of light use efficiency (LUE) in an aspen (Populus tremuloides) forest stand. Before comparisons between PRI and LUE measurements were made, the optical SRS-PRI sensor pairs required calibrations to resolve diurnal and seasonal patterns properly. An offline diurnal calibration procedure was shown to account for variable sky conditions and diurnal illumination changes affecting sensor response. Eddy covariance measurements provided seasonal gross primary productivity (GPP) measures as well as apparent canopy quantum yield dynamics (α). LUE was derived from the ratio of GPP to absorbed photosynthetically active radiation (APAR). Corrected PRI values were derived after diurnal and midday cross-calibration of the sensor’s 532 nm and 570 nm fore-optics, and closely related to both LUE (R² = 0.62, p < 0.05) and α (R² = 0.72, p < 0.05). A LUE model derived from corrected PRI values showed good correlation to measured GPP (R² = 0.77, p < 0.05), with an accuracy comparable to results obtained from an α driven LUE model (R² = 0.79, p < 0.05). The automated PRI sensors proved to be suitable proxies of light use efficiency. The onset of continuous PRI sensors signifies new opportunities for explicitly examining the cause of changing PRI, LUE, and productivity over time and space. As such, this technology represents great value for the flux, remote sensing and modeling community.

Spectral Invariant Provides a Practical Modeling Approach for Future Biophysical Variable Estimations

This paper presents a simple radiative transfer model based on spectral invariant properties (SIP). The canopy structure parameters, including the leaf angle distribution and multi-angular clumping index, are explicitly described in the SIP model. The SIP model has been evaluated on its bidirectional reflectance factor (BRF) in the angular space at the radiation transfer model intercomparison platform, and in the spectrum space by the PROSPECT+SAIL (PROSAIL) model. The simulations of BRF by SIP agreed well with the reference values in both the angular space and spectrum space, with a root-mean-square-error (RMSE) of 0.006. When compared with the widely-used Soil-Canopy Observation of Photochemistry and Energy fluxes (SCOPE) model on fPAR, the RMSE was 0.006 and the R2 was 0.99, which shows a high accuracy. This study also suggests the newly proposed vegetation index, the near-infrared (NIR) reflectance of vegetation (NIRv), was a good linear approximation of the canopy structure parameter, the directional area scattering factor (DASF), with an R2 of 0.99. NIRv was not influenced much by the soil background contribution, but was sensitive to the leaf inclination angle. The sensitivity of NIRv to canopy structure and the robustness of NIRv to the soil background suggest NIRv is a promising index in future biophysical variable estimations with the support of the SIP model, especially for the Deep Space Climate Observatory (DSCOVR) Earth Polychromatic Imaging Camera (EPIC) observations near the hot spot directions.

Earth Observations from DSCOVR/EPIC Instrument

The NOAA Deep Space Climate Observatory (DSCOVR) spacecraft was launched on February 11, 2015, and in June 2015 achieved its orbit at the first Lagrange point or L1, 1.5 million km from Earth towards the Sun. There are two NASA Earth observing instruments onboard: the Earth Polychromatic Imaging Camera (EPIC) and the National Institute of Standards and Technology Advanced Radiometer (NISTAR). The purpose of this paper is to describe various capabilities of the DSCOVR/EPIC instrument. EPIC views the entire sunlit Earth from sunrise to sunset at the backscattering direction (scattering angles between 168.5° and 175.5°) with 10 narrowband filters: 317, 325, 340, 388, 443, 552, 680, 688, 764 and 779 nm. We discuss a number of pre-processingsteps necessary for EPIC calibration including the geolocation algorithm and the radiometric calibration for each wavelength channel in terms of EPIC counts/second for conversion to reflectance units. The principal EPIC products are total ozone O3amount, scene reflectivity, erythemal irradiance, UV aerosol properties, sulfur dioxide SO2 for volcanic eruptions, surface spectral reflectance, vegetation properties, and cloud products including cloud height. Finally, we describe the observation of horizontally oriented ice crystals in clouds and the unexpected use of the O2 B-band absorption for vegetation properties.

Machine Learning Using Hyperspectral Data Inaccurately Predicts Plant Traits Under Spatial Dependency

Spectral, temporal and spatial dimensions are difficult to model together when predicting in situ plant traits from remote sensing data. Therefore, machine learning algorithms solely based on spectral dimensions are often used as predictors, even when there is a strong effect of spatial or temporal autocorrelation in the data. A significant reduction in prediction accuracy is expected when algorithms are trained using a sequence in space or time that is unlikely to be observed again. The ensuing inability to generalise creates a necessity for ground-truth data for every new area or period, provoking the propagation of “single-use” models. This study assesses the impact of spatial autocorrelation on the generalisation of plant trait models predicted with hyperspectral data. Leaf Area Index (LAI) data generated at increasing levels of spatial dependency are used to simulate hyperspectral data using Radiative Transfer Models. Machine learning regressions to predict LAI at different levels of spatial dependency are then tuned (determining the optimum model complexity) using cross-validation as well as the NOIS method. The results show that cross-validated prediction accuracy tends to be overestimated when spatial structures present in the training data are fitted (or learned) by the model.