Leaf Radiative Transfer Model Research Articles

The PROLIB leaf radiative transfer model: Simulation of the dorsiventrality of leaves from visible to mid-wave infrared

Many plant species have dorsiventral leaves that have significant differences in optical properties from one side to the other. Several studies have revealed that ignoring this asymmetry induces significant errors in plant canopy reflectance, and current leaf models simulating leaf dorsiventrality are limited to the 0.4–2.5 μm wavelength range. This article, partly based on two recently collected datasets in the 2.5–14 μm wavelength range, demonstrates that ignoring leaf dorsiventrality induces significant errors in brightness temperature and effective emissivity at the canopy scale. The PROLIB model, which inherits from the PROSPECT-VISIR and LIBERTY models, is the first radiative transfer model to simulate the reflectance and transmittance of both leaf sides from 0.4 to 5.7 μm. The palisade and spongy mesophylls are represented as plate and sphere layers, respectively, to account for the structural asymmetry of leaf cells. The sieve effect that explains the differences in transmittance between the adaxial and abaxial sides of the leaf is successfully incorporated into PROLIB. Evaluation of the model on several leaf datasets shows that: (1) It reproduces well the adaxial and abaxial optical properties of the leaves, with a root mean square error (RMSE) of 0.0109 for reflectance and transmittance. (2) It can be inversed to retrieve leaf traits, with RMSE values for leaf chlorophyll, carotenoid, anthocyanin, water, and dry matter content of 5.519 μg/cm2, 2.344 μg/cm2, 4.219 μg/cm2, 0.0022 g/cm2, and 0.0017 g/cm2, respectively (corresponding normalized RMSE values of 22.0%, 34.0%, 49.4%, 19.6%, and 24.7%). However, better and more complete leaf datasets are needed for leaf dorsiventrality analysis and model calibration.

Estimation of Oak Leaf Functional Traits for California Woodland Savannas and Mixed Forests: Comparison between Statistical, Physical, and Hybrid Methods Using Spectroscopy

Key leaf functional traits, such as chlorophyll and carotenoids content (Cab and Cxc), equivalent water thickness (EWT), and leaf mass per area (LMA), are essential to the characterization and monitoring of ecosystem function. Spectroscopy provides access to these four leaf traits by relying on their specific spectral absorptions over the 0.4–2.5 µm domain. In this study, we compare the performance of three categories of estimation methods to retrieve these four leaf traits from laboratory directional-hemispherical leaf reflectance and transmittance measurements: statistical, physical, and hybrid methods. To this aim, a dataset pooling samples from 114 deciduous and evergreen oak trees was collected on four sites in California (woodland savannas and mixed forests) over three seasons (spring, summer and fall) and was used to assess the performance of each method. Physical and hybrid methods were based on the PROSPECT leaf radiative transfer model. Physical methods included inversion of PROSPECT from iterative algorithms and look-up table (LUT)-based inversion. For LUT-based methods, two distance functions and two sampling schemes were tested. For statistical and hybrid methods, four distinct machine learning regression algorithms were compared: ridge, partial least squares regression (PLSR), Gaussian process regression (GPR), and random forest regression (RFR). In addition, we evaluated the transferability of statistical methods using an independent dataset (ANGERS Leaf optical properties database) to train the regression algorithms. Thus, a total of 17 estimations were compared. Firstly, we studied the PROSPECT leaf structural parameter N retrieved by iterative inversions and its distribution over our oak-specific dataset. N showed a more pronounced seasonal dependency for the deciduous species than for the evergreen species. For the four traits, the statistical methods trained on our dataset outperformed the PROSPECT-based methods. More particularly, statistical methods using GPR yielded the most accurate estimates (RMSE = 5.0 µg·cm−2; 1.3 µg·cm−2; 0.0009 cm; and 0.0009 g·cm−2 for Cab, Cxc, EWT, and LMA, respectively). Among the PROSPECT-based methods, the iterative inversion of this model led to the most accurate results for Cab, Cxc, and EWT (RMSE = 7.8 µg·cm−2; 2.0 µg·cm−2; and 0.0035 cm, respectively), while for LMA, a hybrid method with RFR (RMSE = 0.0030 g·cm−2) was the most accurate. These results showed that estimation accuracy is independent of the season. Considering the transferability of statistical methods, for the four leaf traits, estimation performance was inferior for estimators built on the ANGERS database compared to estimators built exclusively on our dataset. However, for EWT and LMA, we demonstrated that these types of statistical methods lead to better estimation accuracy than PROSPECT-based methods (RMSE = 0.0016 cm and 0.0013 g·cm−2 respectively). Finally, our results showed that more differences were observed between plant functional types than between species or seasons.

Physically based illumination correction for sub-centimeter spatial resolution hyperspectral data

Vegetation biophysical- and chemical traits, defined on the basis of leaf area, can be retrieved from their spectral reflectance. Ultra high resolution hyperspectral images, such as ones collected from drones, allows measuring the spectra of individual leaves. The reflectance signal of such data is calibrated with respect to the top-of-canopy (TOC) irradiance, as the local illumination conditions on leaf surfaces are largely unknown and can vary significantly from the TOC conditions. We developed an inversion algorithm that uses the PROSPECT leaf radiative transfer model and the theory of spectral invariants to retrieve the actual leaf reflectance from TOC-calibrated hyperspectral images. Compared with more traditional canopy reflectance models, this approach accounts for the spatial variation in leaf-level irradiance visible in sub-centimeter-resolution images and is computationally more efficient. We used simulated and measured leaf and canopy reflectance data to validate the approach and found the retrieved leaf reflectances to match closely the actual reflectances (relative RMSD was 12% for simulated data on the average and below 10% for measured data). The proposed method provides an efficient approach for illumination correction, enabling reliable, physically based applications for monitoring vegetation biochemical and biophysical properties from ultra-high-resolution spectral imagery.

Generality of leaf spectroscopic models for predicting key foliar functional traits across continents: A comparison between physically- and empirically-based approaches

Leaf spectroscopy provides an efficient way of predicting foliar functional traits, commonly using physically- and empirically-based models. However, the generality of both models has not been fully investigated, and it is not clear if inversion strategies of physically-based models can be transferred across datasets. In this study, we evaluated the generality of leaf spectroscopic models for predicting key foliar functional traits and compared the performance of physically- and empirically-based approaches. Two extensive datasets compiling a total of 3861 foliar samples were collected from 24 field sites in eastern United States and south China. The leaf radiative transfer model PROSPECT was coupled with COSINE (PROCOSINE) to retrieve foliar traits from leaf bidirectional reflectance factor (BRF). A commonly used empirically-based model, partial least squares regression (PLSR) was performed as a comparison. Results showed that both PROSPECT and PROCOSINE can accurately estimate leaf mass per area (LMA) and equivalent water thickness (EWT). Inversion strategies including the selection of optimal spectral domains and the use of prior information (IS3) greatly improved the estimation accuracy of leaf nitrogen, leaf chlorophyll a + b and carotenoids. The estimation accuracies were similar when transferring inversion strategies across datasets, indicating a high level of transferability of physically-based models. PLSR and interval PLSR (iPLSR, via feature selection) could predict foliar traits with high accuracies when cross-validation was performed, and iPLSR achieved higher accuracies. But both the empirical approaches demonstrated low transferability when applied to an independent dataset. Our findings highlight the importance of generalized traits models with respect to development and calibration of leaf radiative transfer model, as well as incorporating representative samples in training empirical models. This study can help us to better understand the variation of foliar traits among and within species, their response to environmental change, as well as plant biodiversity.

Estimating leaf mass per area with leaf radiative transfer model

Leaf mass per area (LMA) is an important leaf trait but challenging to be accurately estimated. This article proposes a simple leaf radiative transfer model called ISPECT. It explains the difference in optical properties observed on the adaxial (upper) and abaxial (lower) sides of leaves, i.e., their dorsiventrality, with a limited number of structural parameters. The performance of ISPECT in estimating LMA is compared to that of five other leaf radiative transfer models (FASPECT, DLM, PROSPECT-D, PROSPECT-5B, and Leaf-SIP). We tested six experimental datasets with 962 leaf samples and two spectral ranges: the solar domain (0.4–2.5 µm) and the shortwave infrared (1.7–2.4 µm). Results show that PROSPECT-D and PROSPECT-5B accurately estimate LMA using the shortwave infrared spectra, while ISPECT and FASPECT perform well in both spectral ranges. Further analysis demonstrates that leaf dorsiventrality is likely to be an influential factor for LMA estimation: thus ISPECT can accurately estimate LMA in the solar and shortwave infrared domains, with NRMSE of 26.0% and 28.8%, respectively. This motivates further studies on LMA mapping from spaceborne imaging spectrometers (e.g., PRISMA, GaoFen-5, EnMAP) by coupling ISPECT with canopy radiative transfer models.

Evaluation of the Methods for Estimating Leaf Chlorophyll Content with SPAD Chlorophyll Meters

Leaf chlorophyll content (LCC) is an indicator of leaf photosynthetic capacity. It is crucial for improving the understanding of plant physiological status. SPAD meters are routinely used to provide an instantaneous estimation of in situ LCC. However, the calibration of meter readings into absolute measures of LCC is difficult, and a generic approach for this conversion remains elusive. This study presents an evaluation of the approaches that are commonly used in converting SPAD readings into absolute LCC values. We compared these approaches using three field datasets and one synthetic dataset. The field datasets consist of LCC measured using a destructive method in the laboratory, as well as the SPAD readings measured in the field for various vegetation types. The synthetic dataset was generated with the leaf radiative transfer model PROSPECT-5 across different leaf structures. LCC covers a wide range from 1.40 μg cm−2 to 86.34 μg cm−2 in the field datasets, and it ranges from 5 μg cm−2 to 80 μg cm−2 in the synthetic dataset. The relationships between LCC and SPAD readings were examined using linear, polynomial, exponential, and homographic functions for the field and synthetic datasets. For the field datasets, the assessments of these approaches were conducted for (i) all three datasets together, (ii) individual datasets, and (iii) individual vegetation species. For the synthetic dataset, leaves with different leaf structures (which mimic different vegetation species) were grouped for the evaluation of the approaches. The results demonstrate that the linear function is the most accurate one for the simulated dataset, in which leaf structure is relatively simple due to the turbid medium assumption of the PROSPECT-5 model. The assumption of leaves in the PROSPECT-5 model complies with the assumption made in the designed algorithm of the SPAD meter. As a result, the linear relationship between LCC and SPAD values was found for the modeled dataset in which the leaf structure is simple. For the field dataset, the functions do not perform well for all datasets together, while they improve significantly for individual datasets or species. The overall performance of the linear (LCC=a∗SPAD+b), polynomial (LCC=a∗SPAD2+b∗SPAD+c), and exponential functions (LCC=0.0893∗10SPADα) is promising for various datasets and species with the R2 > 0.8 and RMSE <10 μg cm−2. However, the accuracy of the homographic functions (LCC=a∗SPAD/b−SPAD) changes significantly among different datasets and species with R2 from 0.02 of wheat to 0.92 of linseed (RMSE from 642.50 μg cm−2 to 5.74 μg cm−2). Other than species- and dataset-dependence, the homographic functions are more likely to produce a numerical singularity due to the characteristics of the function per se. Compared with the linear and exponential functions, the polynomial functions have a higher degree of freedom due to one extra fitting parameter. For a smaller size of data, the linear and exponential functions are more suitable than the polynomial functions due to the less fitting parameters. This study compares different approaches and addresses the uncertainty in the conversion from SPAD readings into absolute LCC, which facilitates more accurate measurements of absolute LCC in the field.

A convolution neural network for forest leaf chlorophyll and carotenoid estimation using hyperspectral reflectance

Forest leaf chlorophyll (Cab) and carotenoid (Cxc) are key functional indicators for the state of the forest ecosystem. Current machine learning models based on hyperspectral reflectance are widely applied to estimate leaf Cab and Cxc contents at leaf scale. However, these models have certain accuracy for non-independent datasets but have poor generalization for independent datasets when they are used to estimate leaf Cab and Cxc contents. This fact limits that hyperspectral remote sensing completely replaces destructive measurements for leaf Cab and Cxc contents. Thus, the development of an estimation model with high accuracy and satisfactory generalization is necessary. Convolutional neural networks (CNNs) have certain accuracy and generalization in many domains, and have the potential to solve above-mentioned problem. Therefore, this study developed a CNN using one-dimensional hyperspectral reflectance, which aimed to improve the model's accuracy and generalization in leaf Cab and Cxc content estimation at leaf scale. The proposed CNN was developed by three steps. First, in consideration of the correlation between leaf Cab and Cxc contents in natural leaves, 2500 physical data with leaf reflectance and corresponding Cab and Cxc contents were generated by leaf radiative transfer model and multivariable gaussian distribution function. Then, the proposed CNN was built by five strategies based on the architecture of the AlexNet. Finally, five-fold cross validation was performed with 70% of the physical data to determine the best strategy to develop the proposed CNN. These were executed to ensure the proposed CNN with the maximum accuracy and generalization. In addition, the accuracy and generalization of the proposed CNN were tested using a non-independent dataset and an independent dataset, respectively. The proposed CNN was also compared with back propagation neural network (BPNN), support vector regression (SVR) and gaussian process regression (GPR). Results showed that the best CNN could be developed with one input, five convolutional, three max-pooling and three fully-connected layers. Comprehensively considering the model's accuracy and generalization, the proposed CNN was the best model for leaf Cab and Cxc content estimation compared with BPNN, SVR and GPR. This study provides a development strategy of CNN estimation model using one-dimensional hyperspectral reflectance at leaf scale. The proposed CNN could further promote the practical application of hyperspectral remote sensing in leaf Cab and Cxc content estimation.

Spectral Retrieval of Eucalypt Leaf Biochemical Traits by Inversion of the Fluspect-Cx Model

Leaf biochemical traits indicating early symptoms of plant stress can be assessed using imaging spectroscopy combined with radiative transfer modelling (RTM). In this study, we assessed the potential applicability of the leaf radiative transfer model Fluspect-Cx to simulate optical properties and estimate leaf biochemical traits through inversion of two native Australian eucalypt species: Eucalyptus dalrympleana and E. delegetensis. The comparison of measured and simulated optical properties revealed the necessity to recalibrate the refractive index and specific absorption coefficients of the eucalypt leaves’ biochemical constituents. Subsequent validation of the modified Fluspect-Cx showed a closer agreement with the spectral measurements. The average root mean square error (RMSE) of reflectance, transmittance and absorptance values within the wavelength interval of 450–1600 nm was smaller than 1%. We compared the performance of both the original and recalibrated Fluspect-Cx versions through inversions aiming to simultaneously retrieve all model inputs from leaf optical properties with and without prior information. The inversion of recalibrated Fluspect-Cx constrained by laboratory-based measurements produced a superior accuracy in estimations of leaf water content (RMSE = 0.0013 cm, NRMSE = 6.55%) and dry matter content (RMSE = 0.0036 g·cm−2, NRMSE = 21.28%). The estimation accuracies of chlorophyll content (RMSE = 8.46 µg·cm−2, NRMSE = 24.73%), carotenoid content (RMSE = 3.83 µg·cm−2, NRMSE = 30.82%) and anthocyanin content (RMSE = 1.69 µg·cm−2, NRMSE = 37.12%) were only marginally better than for the inversion without any constraints. Additionally, we investigated the possibility to substitute the prior information derived in the laboratory by non-destructive reflectance-based spectral indices sensitive to the retrieved biochemical traits, resulting in the most accurate estimation of carotenoid content (RMSE = 3.65 µg·cm−2, NRMSE = 29%). Future coupling of the recalibrated Fluspect with a forest canopy RTM is expected to facilitate retrieval of biophysical traits from spectral air/space-borne image data, allowing for assessing the actual physiological status and health of eucalypt forest canopies.

PROSDM: Applicability of PROSPECT model coupled with spectral derivatives and similarity metrics to retrieve leaf biochemical traits from bidirectional reflectance

The PROSPECT model, one of the most popular leaf radiative transfer models, has long been used for retrieving leaf biochemical traits from leaf directional-hemispherical reflectance factor (DHRF) spectra, while remains underexplored for applications to the leaf bidirectional reflectance factor (BRF) spectra to retrieve leaf biochemical traits. Existence of leaf surface reflectance and anisotropic property could be the main issues that constrain the applicability of the PROSPECT inversion to assess leaf biochemical traits from leaf BRF spectra. This study presents an inversion method by integrating the PROSPECT model with spectral derivatives and similarity metrics (SDM), called PROSDM, to remove the difference between leaf BRF and DHRF spectra and retrieve leaf biochemical traits from leaf BRF spectra. Leaf BRF spectra and the corresponding contents of chlorophyll (Cab), carotenoid (Cxc), water (Cw), and dry matter (Cm) were obtained from ten datasets across a wide variety of plant species with varied growth stages, nutrition status, and planting regions. The datasets showed that the significant difference existed between leaf BRF and DHRF spectra, which varied with spectral wavelengths and plant species, and affected the accuracies of retrieving Cab, Cxc, Cw, and Cm. To eliminate the difference between leaf BRF and DHRF spectra independent on wavelengths, spectral derivatives were applied. When the difference between BRF and DHRF spectra varied with wavelengths, spectral derivatives only removed part of the difference, and implementation of the Manhattan distance compensated the limitation of the spectral derivatives to further reduce the difference. As a result, the PROSDM outperformed the PROSPECT and PROCOSINE inversions as well as PROCWT to retrieve Cab, Cxc, Cw, and Cm from leaf BRF spectra with the root mean square errors (RMSEs) of 7.64 μg/cm2, 2.77 μg/cm2, 0.0041 g/cm2, and 0.0024 g/cm2, respectively. Significant improvements in the retrievals of Cab, Cxc, Cw, and Cm were obtained with the RMSEs reduced by 20.33%, 29.34%, 25.45%, and 44.19% compared to the PROSPECT inversion, respectively. It was further found that the model inversions were affected by the spectral saturation, PROSPECT versions, spectral subdomains, and the range of model parameters. The retrieval results by different inversion methods may be improved with the suitable spectral subdomains and range of model parameters. Importantly, the proposed PROSDM showed the great potential to alleviate these negative effects on the retrievals of Cab, Cxc, Cw, and Cm. The performance assessment of PROSDM demonstrates a promising potential for its applications in remote sensing aiming at retrievals of leaf biochemical traits from leaf BRF spectra.

Cutting out the middleman: calibrating and validating a dynamic vegetation model (ED2-PROSPECT5) using remotely sensed surface reflectance

Abstract. Canopy radiative transfer is the primary mechanism by which models relate vegetation composition and state to the surface energy balance, which is important to light- and temperature-sensitive plant processes as well as understanding land–atmosphere feedbacks. In addition, certain parameters (e.g., specific leaf area, SLA) that have an outsized influence on vegetation model behavior can be constrained by observations of shortwave reflectance, thus reducing model predictive uncertainty. Importantly, calibrating against radiative transfer outputs allows models to directly use remote sensing reflectance products without relying on highly derived products (such as MODIS leaf area index) whose assumptions may be incompatible with the target vegetation model and whose uncertainties are usually not well quantified. Here, we created the EDR model by coupling the two-stream representation of canopy radiative transfer in the Ecosystem Demography model version 2 (ED2) with a leaf radiative transfer model (PROSPECT-5) and a simple soil reflectance model to predict full-range, high-spectral-resolution surface reflectance that is dependent on the underlying ED2 model state. We then calibrated this model against estimates of hemispherical reflectance (corrected for directional effects) from the NASA Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and survey data from 54 temperate forest plots in the northeastern United States. The calibration significantly reduced uncertainty in model parameters related to leaf biochemistry and morphology and canopy structure for five plant functional types. Using a single common set of parameters across all sites, the calibrated model was able to accurately reproduce surface reflectance for sites with highly varied forest composition and structure. However, the calibrated model's predictions of leaf area index (LAI) were less robust, capturing only 46 % of the variability in the observations. Comparing the ED2 radiative transfer model with another two-stream soil–leaf–canopy radiative transfer model commonly used in remote sensing studies (PRO4SAIL) illustrated structural errors in the ED2 representation of direct radiation backscatter that resulted in systematic underestimation of reflectance. In addition, we also highlight that, to directly compare with a two-stream radiative transfer model like EDR, we had to perform an additional processing step to convert the directional reflectance estimates of AVIRIS to hemispherical reflectance (also known as “albedo”). In future work, we recommend that vegetation models add the capability to predict directional reflectance, to allow them to more directly assimilate a wide range of airborne and satellite reflectance products. We ultimately conclude that despite these challenges, using dynamic vegetation models to predict surface reflectance is a promising avenue for model calibration and validation using remote sensing data.

Applications of radiative transfer models to greenhouse vegetation

In greenhouse horticulture, efficiency of climate control and plant protection can be improved by having an accurate impression of plant status, such as photosynthesis or chemical composition. Recent advances in remote sensing technologies have brought about a range of innovations in precision agriculture, with the potential for adaptation to greenhouses. Simple, traditionally used indices employ only one or two spectral bands, in which the contributions of various pigments and leaf or canopy structure can highly overlap. Consequently, such indices may be insufficient for applications. State-of-the-art models have been developed that can better interpret hyper- and multispectral leaf and canopy imagery by employing the biochemical and radiative transfer properties of vegetation. An example is the soil-canopy observation of photosynthesis and energy balance (SCOPE) model, which was developed specifically for crop canopies. Here we present one of the pillars of SCOPE, the leaf radiative transfer (RT) model Fluspect. Fluspect simulates leaf chlorophyll fluorescence, reflectance and transmittance spectra. The model can be inverted to obtain estimates of leaf chlorophylls, carotenoids, anthocyanins, xanthophyll epoxidation, water and dry matter content. Moreover, it can be linked to a model for leaf photosynthesis and when inverted, provide a method to estimate photosynthesis directly from leaf spectral information. We test the model against a tomato data set, with measured hyperspectral images, chlorophyll, sugar, acid, starch, dry matter content and nutrients. The first study of the data set, using partial least square regression, showed that hyperspectral images have a high correlation with important fruit and leaf compounds. We compared these results to Fluspect retrievals and conventional vegetation indices. In the paper, we discuss the potential added value of using RT models in greenhouse horticulture.

Unique contributions of chlorophyll and nitrogen to predict crop photosynthetic capacity from leaf spectroscopy.

The photosynthetic capacity or the CO2-saturated photosynthetic rate (Vmax), chlorophyll, and nitrogen are closely linked leaf traits that determine C4 crop photosynthesis and yield. Accurate, timely, rapid, and non-destructive approaches to predict leaf photosynthetic traits from hyperspectral reflectance are urgently needed for high-throughput crop monitoring to ensure food and bioenergy security. Therefore, this study thoroughly evaluated the state-of-the-art physically based radiative transfer models (RTMs), data-driven partial least squares regression (PLSR), and generalized PLSR (gPLSR) models to estimate leaf traits from leaf-clip hyperspectral reflectance, which was collected from maize (Zea mays L.) bioenergy plots with diverse genotypes, growth stages, treatments with nitrogen fertilizers, and ozone stresses in three growing seasons. The results show that leaf RTMs considering bidirectional effects can give accurate estimates of chlorophyll content (Pearson correlation r=0.95), while gPLSR enabled retrieval of leaf nitrogen concentration (r=0.85). Using PLSR with field measurements for training, the cross-validation indicates that Vmax can be well predicted from spectra (r=0.81). The integration of chlorophyll content (strongly related to visible spectra) and nitrogen concentration (linked to shortwave infrared signals) can provide better predictions of Vmax (r=0.71) than only using either chlorophyll or nitrogen individually. This study highlights that leaf chlorophyll content and nitrogen concentration have key and unique contributions to Vmax prediction.

Mapping within-field leaf chlorophyll content in agricultural crops for nitrogen management using Landsat-8 imagery

Spatial information on crop nutrient status is central for monitoring vegetation health, plant productivity and managing nutrient optimization programs in agricultural systems. This study maps the spatial variability of leaf chlorophyll content within fields with differing quantities of nitrogen fertilizer application, using multispectral Landsat-8 OLI data (30 m). Leaf chlorophyll content and leaf area index measurements were collected at 15 wheat (Triticum aestivum) sites and 13 corn (Zea mays) sites approximately every 10 days during the growing season between May and September 2013 near Stratford, Ontario. Of the 28 sites, 9 sites were within controlled areas of zero nitrogen fertilizer application. Hyperspectral leaf reflectance measurements were also sampled using an Analytical Spectral Devices FieldSpecPro spectroradiometer (400–2500 nm). A two-step inversion process was developed to estimate leaf chlorophyll content from Landsat-8 satellite data at the sub-field scale, using linked canopy and leaf radiative transfer models. Firstly, at the leaf-level, leaf chlorophyll content was modelled using the PROSPECT model, using both hyperspectral and simulated mulitspectral Landsat-8 bands from the same leaf sample. Hyperspectral and multispectral validation results were both strong (R2 = 0.79, RMSE = 13.62 μg/cm2 and R2 = 0.81, RMSE = 9.45 μg/cm2, respectively). Secondly, leaf chlorophyll content was estimated from Landsat-8 satellite imagery for 7 dates within the growing season, using PROSPECT linked to the 4-Scale canopy model. The Landsat-8 derived estimates of leaf chlorophyll content demonstrated a strong relationship with measured leaf chlorophyll values (R2 = 0.64, RMSE = 16.18 μg/cm2), and compared favourably to correlations between leaf chlorophyll and the best performing tested spectral vegetation index (Green Normalised Difference Vegetation Index, GNDVI; R2 = 0.59). This research provides an operational basis for modelling within-field variations in leaf chlorophyll content as an indicator of plant nitrogen stress, using a physically-based modelling approach, and opens up the possibility of exploiting a wealth of multispectral satellite data and UAV-mounted multispectral imaging systems.

A novel combined spectral index for estimating the ratio of carotenoid to chlorophyll content to monitor crop physiological and phenological status

Accurate estimation of the ratio of carotenoid (Car) to chlorophyll (Chl) content is crucial to provide valuable insight into diagnoses of plant physiological and phenological status in crop fields. Studies for assessing the ratio of Car to Chl content have been extensively conducted with semi-empirical approaches using spectral indices. However, spectral indices established in previous studies generally relied on site- or species-specific measured data and these indices typically lacked sufficient estimation accuracy for the ratio of Car to Chl content to be used across various species and under different physiological conditions. In this study, we propose a novel combined carotenoid/chlorophyll ratio index (CCRI) in the form of the carotenoid index (CARI) divided by the red-edge chlorophyll index (CIred-edge): The value of the index is illustrated using synthetic data simulated from the leaf radiative transfer model PROSPECT-5 and with extensive measured datasets at both the leaf and canopy level from the ANGERS dataset and winter wheat and maize field experiments. Results show that CCRI was the index with the highest correlation with the ratio of Car to Chl content in PROSPECT-5 simulations (R2 = 0.99, RRMSE = 8.65%) compared to other spectral indices. Calibration and validation results using the ANGERS and winter wheat leaf level data showed that CCRI achieved accurate estimation of the ratio of Car to Chl content (R2 = 0.52, RRMSE = 14.10%). CCRI also showed a good performance (R2 = 0.54, RRMSE = 17.08%) for estimation of the ratio of Car to Chl content in both calibration and validation with the winter wheat and maize canopy spectra measured in field experiments. Further investigation of the effect of the correlation between leaf Chl and Car content on the performance of CCRI indicated that variation of the correlation affected the retrieval accuracy of CCRI, and CCRI might not be very sensitive to changes of the ratio of Car to Chl content with low values (<0.10).

Extending Fluspect to simulate xanthophyll driven leaf reflectance dynamics

The xanthophyll cycle regulates the energy flow to photosynthetic reaction centres of plant leaves. Changes in the de-epoxidation state (DEPS) of xanthophyll cycle pigments can be observed as changes in the leaf absorption of light with wavelengths between 500 to 570 nm. These spectral changes can be a good remote sensing indicator of the photosynthetic efficiency, and are traditionally quantified with a two-band physiologically based optical index, the Photochemical Reflectance Index (PRI). In this paper, we present an extension of the plant leaf radiative transfer model Fluspect (Fluspect-CX) that reproduces the spectral changes in a wide band of green reflectance: a radiative transfer analogy to the PRI. The idea of Fluspect-CX is to use in vivo specific absorption coefficients for two extreme states of carotenoids, representing the two extremes of the xanthophyll de-epoxidation, and to describe the intermediate states as a linear mixture of these two states. The ‘photochemical reflectance parameter’ (Cx) quantifies the relative proportion of the two states. Fluspect-CX simulates leaf chlorophyll fluorescence (ChlF) excitation-emission matrices, as well as reflectance (R) and transmittance (T) spectra as a function of leaf structure, pigment contents and Cx. We describe the calibration of the model and test its performance using various experimental datasets. Furthermore, we retrieved Cx from optical measurements of various datasets. The retrieved Cx correlates well with xanthophyll DEPS (R2 = 0.57), as well with non-photochemical quenching (NPQ) of fluorescence (R2 = 0.78). The correlation with NPQ enabled us to incorporate Fluspect-CX in the model SCOPE to scale the processes to the canopy level. Introducing the dynamic green reflectance into a radiative transfer model provides new means to study chlorophyll fluorescence and PRI dynamics on leaf and canopy scales, which is crucial for the remote sensing.

Mapping forest canopy nitrogen content by inversion of coupled leaf-canopy radiative transfer models from airborne hyperspectral imagery

Foliar nitrogen is a critical factor in leaf physiological processes, plant growth, and ecosystem functioning, which has been proposed as one of the essential biodiversity variables. Nitrogen has been quantified by a number of empirical approaches using hyperspectral data, but the retrieval of nitrogen through a physically based approach remains a challenge. A recent study by Wang et al. (2015a) has revealed that leaf protein can be successfully estimated from fresh leaf spectra using a revised leaf radiative transfer model PROPECT-5 which incorporated the effects of leaf protein and cellulose + lignin on leaf reflectance and transmittance. This provides a potential approach of estimating nitrogen using radiative transfer models given the correlation between protein and nitrogen. However, such a revised leaf model has not been tested for the estimation of leaf nitrogen at the canopy level. In this study, a canopy reflectance model INFORM, coupled with the revised PROSPECT-5 model, was used to retrieve leaf and canopy nitrogen content in a mixed temperate forest using the wavelengths of 800–2500 nm from airborne hyperspectral imagery. Ecological criteria were applied to the parameterization of the model to reduce unrealistic combinations of input parameters. Global sensitivity analysis showed that leaf protein played a small but distinct role in driving the variation of canopy reflectance in the INFORM model. More accurate estimation was obtained for canopy nitrogen content (R2 = 0.64, RMSE = 1.90, NRMSE = 0.18) than leaf nitrogen content (R2 = 0.46, RMSE = 3.79e-05, NRMSE = 0.19). Moreover, inversion techniques, particularly regularized look-up tables, further improved the estimation accuracies compared to the original tables. Our results indicate that leaf and canopy nitrogen content can be retrieved successfully at the canopy level by inversion of INFORM. Both the direct and indirect effects of nitrogen on canopy reflectance are important for nitrogen estimation. The maps of leaf and canopy nitrogen content are the first to be generated using inversion of coupled leaf-canopy models, and the spatial variation of foliar nitrogen appears to be reasonable and consistent with ecological knowledge.

The mSCOPE model: A simple adaptation to the SCOPE model to describe reflectance, fluorescence and photosynthesis of vertically heterogeneous canopies

The vertical heterogeneity of leaf biophysical and biochemical properties may have a large effect on the bidirectional reflectance and fluorescence of vegetation canopies. This has implications for the interpretation of remote sensing data. We developed a model for light interaction and energy balance in vegetation canopies in which leaf biophysical and biochemical properties vary in the vertical. The model mSCOPE is an extension of the Soil-Canopy Observation of Photosynthesis and Energy fluxes (SCOPE) model, which simulates spectral and bidirectional reflectance, fluorescence, and photosynthesis of vertically heterogeneous vegetation canopies. The modelling of radiative transfer in mSCOPE is based on the classical SAIL theory. A solution to the radiative transfer equation for multi-layer canopies is given, which allows calculating top-of-canopy (TOC) reflectance and the flux profile. The latter is used for the simulation of fluorescence emission and photosynthesis of every leaf through the leaf radiative transfer model Fluspect and a biochemical model. The radiative transfer of fluorescence in multi-layer canopies is solved numerically in mSCOPE to obtain TOC bidirectional fluorescence. The significant effect of vertical heterogeneity of leaf properties on TOC reflectance, fluorescence and photosynthesis is demonstrated by different scenarios with customized vertical profiles of leaf chlorophyll content and leaf water content, and also with measured vertical profiles of leaf chlorophyll content in corn canopies. A preliminary validation of the reflectance calculating routine of mSCOPE is conducted by comparing measured and simulated TOC reflectance spectra of the corn canopies. We conclude that it is important to consider the vertical heterogeneity of leaf properties for the prediction of reflectance, fluorescence and photosynthesis. The model mSCOPE could serve as a tool to better understand vertically heterogeneous vegetation canopies.

Assessing the consistency of optical properties measured in four integrating spheres

ABSTRACTLeaf optical properties (LOPs) determine the radiation regime of vegetation, thus being key input parameters in leaf and canopy radiative transfer models. It is of great importance to know the uncertainties originating from the LOP measurements. The most common approach to measure LOP uses integrating spheres. They allow measurements of both directional-hemispherical reflectance factor (R) and transmittance factor (T). However, sphere’s design, inner surface coating and measurement protocol differ among the spheres’ models and manufacturers. Our main goal was to evaluate the consistency of optical properties measured in four integrating spheres (Dualsphere, Labspehre, ASD, and Li-cor). Our test samples were three Spectralon® panels, four artificial materials and leaves from six common broadleaf tree species. Results showed that spectra measured in the four integrating spheres were generally similar in the spectral region between 400 and 1600 nm. The average standard deviation computed among the spectra of all samples measured in all spheres was around 0.023 (and varied between 0.005 and 0.044). Statistically significant differences were detected mainly between Dualsphere and Li-cor spheres.

Assessment of leaf carotenoids content with a new carotenoid index: Development and validation on experimental and model data

Leaf carotenoids content (LCar) is an important indicator of plant physiological status. Accurate estimation of LCar provides valuable insight into early detection of stress in vegetation. With spectroscopy techniques, a semi-empirical approach based on spectral indices was extensively used for carotenoids content estimation. However, established spectral indices for carotenoids that generally rely on limited measured data, might lack predictive accuracy for carotenoids estimation in various species and at different growth stages. In this study, we propose a new carotenoid index (CARI) for LCar assessment based on a large synthetic dataset simulated from the leaf radiative transfer model PROSPECT-5, and evaluate its capability with both simulated data from PROSPECT-5 and 4SAIL and extensive experimental datasets: the ANGERS dataset and experimental data acquired in field experiments in China in 2004. Results show that CARI was the index most linearly correlated with carotenoids content at the leaf level using a synthetic dataset (R2=0.943, RMSE=1.196μg/cm2), compared with published spectral indices. Cross-validation results with CARI using ANGERS data achieved quite an accurate estimation (R2=0.545, RMSE=3.413μg/cm2), though the RBRI performed as the best index (R2=0.727, RMSE=2.640μg/cm2). CARI also showed good accuracy (R2=0.639, RMSE=1.520μg/cm2) for LCar assessment with leaf level field survey data, though PRI performed better (R2=0.710, RMSE=1.369μg/cm2). Whereas RBRI, PRI and other assessed spectral indices showed a good performance for a given dataset, overall their estimation accuracy was not consistent across all datasets used in this study. Conversely CARI was more robust showing good results in all datasets. Further assessment of LCar with simulated and measured canopy reflectance data indicated that CARI might not be very sensitive to LCar changes at low leaf area index (LAI) value, and in these conditions soil moisture influenced the LCar retrieval accuracy.

Quantifying the influences of spectral resolution on uncertainty in leaf trait estimates through a Bayesian approach to RTM inversion

Abstract The remote monitoring of plant canopies is critically needed for understanding of terrestrial ecosystem mechanics and biodiversity as well as capturing the short- to long-term responses of vegetation to disturbance and climate change. A variety of orbital, sub-orbital, and field instruments have been used to retrieve optical spectral signals and to study different vegetation properties such as plant biochemistry, nutrient cycling, physiology, water status, and stress. Radiative transfer models (RTMs) provide a mechanistic link between vegetation properties and observed spectral features, and RTM spectral inversion is a useful framework for estimating these properties from spectral data. However, existing approaches to RTM spectral inversion are typically limited by the inability to characterize uncertainty in parameter estimates. Here, we introduce a Bayesian algorithm for the spectral inversion of the PROSPECT 5 leaf RTM that is distinct from past approaches in two important ways: First, the algorithm only uses reflectance and does not require transmittance observations, which have been plagued by a variety of measurement and equipment challenges. Second, the output is not a point estimate for each parameter but rather the joint probability distribution that includes estimates of parameter uncertainties and covariance structure. We validated our inversion approach using a database of leaf spectra together with measurements of equivalent water thickness (EWT) and leaf dry mass per unit area (LMA). The parameters estimated by our inversion were able to accurately reproduce the observed reflectance (RMSE VIS = 0.0063, RMSE NIR-SWIR = 0.0098) and transmittance (RMSE VIS = 0.0404, RMSE NIR-SWIR = 0.0551) for both broadleaved and conifer species. Inversion estimates of EWT and LMA for broadleaved species agreed well with direct measurements (CV EWT = 18.8%, CV LMA = 24.5%), while estimates for conifer species were less accurate (CV EWT = 53.2%, CV LMA = 63.3%). To examine the influence of spectral resolution on parameter uncertainty, we simulated leaf reflectance as observed by ten common remote sensing platforms with varying spectral configurations and performed a Bayesian inversion on the resulting spectra. We found that full-range hyperspectral platforms were able to retrieve all parameters accurately and precisely, while the parameter estimates of multispectral platforms were much less precise and prone to bias at high and low values. We also observed that variations in the width and location of spectral bands influenced the shape of the covariance structure of parameter estimates. Our Bayesian spectral inversion provides a powerful and versatile framework for future RTM development and single- and multi-instrumental remote sensing of vegetation.