Conference Paper

Development and assessment of leaf area index algorithms for the Sentinel-2 multispectral imager

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Leaf area index (LAI) is identified as a Level 2b product to be derived from the Sentinel-2 (S2) Multispectral Imager (MSI) in support of user services [1]. The Validation of Sentinel 2 (VALSE2) project conducted a review, implementation and validation of LAI algorithms suitable for the MSI. Validation was performed using simulated MSI imagery co-located with in-situ LAI over 7 ESA Campaigns. Here we describe two implemented algorithms, the INRA Neural Network algorithm (NNET) and the CCRS Red-Edge algorithm (CCRS), and report on their verification using the PROSAILH radiative transfer model as well as validation both over the BARRAX ESA Campaign as well as prior campaigns. Results indicate both algorithms can provide reasonably unbiased LAI estimates with acceptable error (<1 unit) over prior validation sites but with larger (>1 unit) error over BARRAX. The larger error may be due to a combination of noisy input image data as well as the combination of sparse canopies and bright soils at that site.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... The physical model method, which is suitable for a variety of vegetation types, is also used to extract LAI from moderate-to high-resolution imagery [30][31][32][33][34][35][36][37]. Canopy reflectance models simulate the physical relationship between the canopy reflectance and the LAI in the forward direction. ...
... The hybrid methods include decision tree learning, artificial neural networks, kernel methods, and Bayesian networks [39]. Additionally, the currently used indirect methods of the radiative transfer model (RTM) for Landsat ETM+ and Sentinel-2 MSI data are the LUT and neural networks [31,32,35,36]. The accuracy of satellite LAI inversions is better than SVI-LAI empirical relationships, with R 2 values from 0.54 to 0.82 and RMSE values from 0.17 m 2 /m 2 to 0.71 m 2 /m 2 for crops (e.g., maize and soybean), shrubs, and planted forests [30,31,33,34]. ...
Article
Full-text available
In recent years, China has developed and launched several satellites with high spatial resolutions, such as the resources satellite No. 3 (ZY-3) with a multi-spectral camera (MUX) and 5.8 m spatial resolution, the satellite GaoFen No. 1 (GF-1) with a wide field of view (WFV) camera and 16 m spatial resolution, and the environment satellite (HJ-1A/B) with a charge-coupled device (CCD) sensor and 30 m spatial resolution. First, to analyze the potential application of ZY-3 MUX, GF-1 WFV, and HJ-1 CCD to extract the leaf area index (LAI) at the regional scale, this study estimated LAI from the relationships between physical model-based spectral vegetation indices (SVIs) and LAI values that were generated from look-up tables (LUTs), simulated from the combination of the PROSPECT-5B leaf model and the scattering by arbitrarily inclined leaves with the hot-spot effect (SAILH) canopy reflectance model. Second, to assess the surface reflectance quality of these sensors after data preprocessing, the well-processed surface reflectance products of the Landsat-8 operational land imager (OLI) sensor with a convincing data quality were used to compare the performances of ZY-3 MUX, GF-1 WFV, and HJ-1 CCD sensors both in theory and reality. Apart from several reflectance fluctuations, the reflectance trends were coincident, and the reflectance values of the red and near-infrared (NIR) bands were comparable among these sensors. Finally, to analyze the accuracy of the LAI estimated from ZY-3 MUX, GF-1 WFV, and HJ-1 CCD, the LAI estimations from these sensors were validated based on LAI field measurements in Huailai, Hebei Province, China. The results showed that the performance of the LAI that was inversed from ZY-3 MUX was better than that from GF-1 WFV, and HJ-1 CCD, both of which tended to be systematically underestimated. In addition, the value ranges and accuracies of the LAI inversions both decreased with decreasing spatial resolution.
... The coupling of RTM-MLRA seems to be a promising solution for operationalisation, despite the ill-posed nature of physicallybased approaches. Unfortunately, such efforts have focused mainly on NN [77,83], which has been proven inconsistent in semi-arid African environments [37]. Moreover, variable importance used to explain the model performance here is inadequate as it provides global interpretations (i.e., based on the entire model architecture and dataset) and explains overall relationships between the explanatory and response variables. ...
... Nevertheless, all MLRAs resulted in acceptable accuracy by GCOS/GMES, i.e., RRMSE of ≤10%, for all the biophysical parameters. This result is comparable (in other cases better when compared) to studies using simulated and hyperspectral data, RTMs, and advanced MLRAs such as NN and GPR [17,20,77,83]. Based on sPLS' better predictive performance using only seven variables over GBM in estimating LC ab , it is recommended that fewer, i.e., 4-8, Sentinel-2 spectral subsets should be evaluated in the future. ...
Article
Full-text available
Global food security is critical to eliminating hunger and malnutrition. In the changing climate, farmers in developing countries must adopt technologies and farming practices such as precision agriculture (PA). PA-based approaches enable farmers to cope with frequent and intensified droughts and heatwaves, optimising yields, increasing efficiencies, and reducing operational costs. Biophysical parameters such as Leaf Area Index (LAI), Leaf Chlorophyll Content (LCab), and Canopy Chlorophyll Content (CCC) are essential for characterising field-level spatial variability and thus are necessary for enabling variable rate application technologies, precision irrigation, and crop monitoring. Moreover, robust machine learning algorithms offer prospects for improving the estimation of biophysical parameters due to their capability to deal with non-linear data, small samples, and noisy variables. This study compared the predictive performance of sparse Partial Least Squares (sPLS), Random Forest (RF), and Gradient Boosting Machines (GBM) for estimating LAI, LCab, and CCC with Sentinel-2 imagery in Bothaville, South Africa and identified, using variable importance measures, the most influential bands for estimating crop biophysical parameters. The results showed that RF was superior in estimating all three biophysical parameters, followed by GBM which was better in estimating LAI and CCC, but not LCab, where sPLS was relatively better. Since all biophysical parameters could be achieved with RF, it can be considered a good contender for operationalisation. Overall, the findings in this study are significant for future biophysical product development using RF to reduce reliance on many algorithms for specific parameters, thus facilitating the rapid extraction of actionable information to support PA and crop monitoring activities.
... The SL2P algorithm (method 3 in Table 3) estimates LAI from Sentinel-2 top-of-canopy reflectance L2A data using neural networks trained with radiative transfer simulations from the PROSPECT (Jacquemoud and Baret 1990) and SAIL model (Verhoef 1984;Fernandes et al. 2014b;Weiss and Baret 2016). Validation results show that SL2P estimation is closer to effective LAI and might underestimate LAI in clumped canopies (Djamai et al. 2019;Brown et al. 2021). ...
Article
Full-text available
Remote sensing estimation of evapotranspiration (ET) directly quantifies plant water consumption and provides essential information for irrigation scheduling, which is a pressing need for California vineyards as extreme droughts become more frequent. Many ET models take satellite-derived Leaf Area Index (LAI) as a major input, but how uncertainties of LAI estimations propagate to ET and the partitioning between evaporation and transpiration is poorly understood. Here we assessed six satellite-based LAI estimation approaches using Landsat and Sentinel-2 images against ground measurements from four vineyards in California and evaluated ET sensitivity to LAI in the thermal-based two-source energy balance (TSEB) model. We found that radiative transfer modeling-based approaches predicted low to medium LAI well, but they significantly underestimated high LAI in highly clumped vine canopies (RMSE ~ 0.97 to 1.27). Cubist regression models trained with ground LAI measurements from all vineyards achieved high accuracy (RMSE ~ 0.3 to 0.48), but these empirical models did not generalize well between sites. Red edge bands and the related vegetation index (VI) from the Sentinel-2 satellite contain complementary information of LAI to VIs based on near-infrared and red bands. TSEB ET was more sensitive to positive LAI biases than negative ones. Positive LAI errors of 50% resulted in up to 50% changes in ET, while negative biases of 50% in LAI caused less than 10% deviations in ET. However, even when ET changes were minimal, negative LAI errors of 50% led to up to a 40% reduction in modeled transpiration, as soil evaporation and plant transpiration responded to LAI change divergently. These findings call for careful consideration of satellite LAI uncertainties for ET modeling, especially for the partitioning of water loss between vine and soil or cover crop for effective vineyard irrigation management.
... The development of methodologies that address forward and inverse problems has been essential in research in the fields of geomatics, physics, and remote sensing [3,4]. These problems have been approached with empirical, statistical and machine learning models and complex physical-based models [5,6], where the development of models that estimate accurately biophysical parameters from data coming from sensors remote areas has been a fundamental element in addressing these problems. ...
Chapter
Full-text available
In this paper, a chain of satellite image processing using free software libraries is proposed, to estimate biophysical parameters using data from the Sentinel-2 satellite. In particular, the processing chain proposed allows atmospheric correction, resampling and spatial cropping of satellite images. To evaluate the functionality of the developed processing chain, the sugarcane cultivation of the Mexican region of Jalisco is introduced as a case study; from the selected scene, the leaf area index (LAI) is estimated using a model based on the Gaussian Process Regression technique, which is trained employing synthetic reflectance data created utilizing the PROSAIL radiative transfer model.
... They concluded the red edge bands are the most important bands. Fernandes et al. [20] developed a physically based empirical relationship between LAI and the normalized difference of two Sentinel-2 red edge bands. ...
Article
Full-text available
Vegetation biophysical parameter retrieval is an important earth remote sensing system application. In this paper, we studied the potential impact of the addition of new spectral bands in the red edge region in future Landsat satellites on agroecosystem canopy green leaf area index (LAI) retrieval. The test data were simulated from SPARC ‘03 field campaign HyMap hyperspectral data. Three retrieval approaches were tested: empirical regression based on vegetation index, physical model-based look-up-table (LUT) inversion, and machine learning. The results of all three approaches showed that a potential new spectral band located between the Landsat-8 Operational Land Imager (OLI) red and NIR bands slightly improved the agroecosystem green LAI retrieval accuracy (R2 of 0.787 vs. 0.810 for vegetation index approach, 0.806 vs. 0.828 for LUT inversion approach, and 0.925 vs. 0.933 for machine learning approach). The results of this work are consistent with the conclusions from previous research on the value of Sentinel-2 red edge bands for agricultural green LAI retrieval.
... It has the combination of high spatial resolution and revisits frequently, novel spectral capabilities, and wide coverage, which offer more advantages over Landsat series in regional LULC classification. However, the existing researches mainly focus on parameter estimation (e.g., vegetation biophysical and water quality [4][5][6]) and specific target detection (e.g., water body, greenhouses, and built-up areas [7][8][9][10]). Vegetation species classification has also attracted great attention with the aim of assessing the potential of its three unique red-edge bands [11,12]. ...
Article
Full-text available
Land Use and Land Cover (LULC) classification is vital for environmental and ecological applications. Sentinel-2 is a new generation land monitoring satellite with the advantages of novel spectral capabilities, wide coverage and fine spatial and temporal resolutions. The effects of different spatial resolution unification schemes and methods on LULC classification have been scarcely investigated for Sentinel-2. This paper bridged this gap by comparing the differences between upscaling and downscaling as well as different downscaling algorithms from the point of view of LULC classification accuracy. The studied downscaling algorithms include nearest neighbor resampling and five popular pansharpening methods, namely, Gram-Schmidt (GS), nearest neighbor diffusion (NNDiffusion), PANSHARP algorithm proposed by Y. Zhang, wavelet transformation fusion (WTF) and high-pass filter fusion (HPF). Two spatial features, textural metrics derived from Grey-Level-Co-occurrence Matrix (GLCM) and extended attribute profiles (EAPs), are investigated to make up for the shortcoming of pixel-based spectral classification. Random forest (RF) is adopted as the classifier. The experiment was conducted in Xitiaoxi watershed, China. The results demonstrated that downscaling obviously outperforms upscaling in terms of classification accuracy. For downscaling, image sharpening has no obvious advantages than spatial interpolation. Different image sharpening algorithms have distinct effects. Two multiresolution analysis (MRA)-based methods, i.e., WTF and HFP, achieve the best performance. GS achieved a similar accuracy with NNDiffusion and PANSHARP. Compared to image sharpening, the introduction of spatial features, both GLCM and EAPs can greatly improve the classification accuracy for Sentinel-2 imagery. Their effects on overall accuracy are similar but differ significantly to specific classes. In general, using the spectral bands downscaled by nearest neighbor interpolation can meet the requirements of regional LULC applications, and the GLCM and EAPs spatial features can be used to obtain more precise classification maps.
... The forthcoming availability of Sentinel-2 data could allow a routine application of the algorithm based on the application of ANN inversion of the PROSAIL model at the International Journal of Remote Sensing 2455 regional scale (Fernandes et al. 2014), obtaining maps of biophysical variables at short time intervals and during the whole growing season of crops. ...
Article
Full-text available
Detailed information about the prediction of within-field potential in terms of yield at the field scale is an attractive goal that would allow useful applications in precision agriculture. Biophysical variables characterizing crop canopies, such as the leaf area index (LAI), fraction of absorbed photosynthetically active radiation (FAPAR), fractional ground cover (Fcover) and the concentration of chlorophyll-a and -b (Cab), can be estimated from satellite remote-sensing data through the application of a neural network inversion of a radiative transfer model, such as PROSAIL. The knowledge of the temporal and spatial variability of these variables can enhance the possibilities of estimating yield at the field scale. The aim of this study is to investigate the influence of acquisition time and spatial resolution of biophysical variables estimated from satellite data on the grain yield estimation of wheat crops. We used SPOT 4 (spatial resolution: 20 m) and SPOT 5 (spatial resolution: 10 m) images, acquired at six different dates during the wheat growing season in 2012, to obtain LAI, Fcover, FAPAR, and Cab on five fields in Maccarese (Central Italy). A preliminary survey was carried out to correlate spatially biophysical variables with the final grain yield at each acquisition date. Biophysical variables estimated at a spatial resolution of 10 m during the stem elongation stage showed the best simple and spatial correlation with yield. At this stage, all the biophysical variables showed the highest correlation values as compared to the other dates. Subsequently, we used the variables estimated from SPOT data at each growth stage to calibrate multiple linear regression (MLR) and cubist regression (CR) models for two fields, which were then validated on five independent fields. Although the CR calibration models provided better accuracy than MLR, the best validation statistics were gained from MLR models, obtaining a root mean square error (RMSE) of about 1 t ha−1 for three of these fields, using remote data having a spatial resolution of 10 metres and acquired between steam elongation and booting stage. The optimal acquisition time is affected, ceteris paribus, by the agricultural management and in particular by the variety that can influence the trend of crop growth. However, the optimal growth stage for yield estimation seems to be quite similar over the study area during a growth season. The validation of models on field data collected in another growing season is mainly affected by the climate conditions. These results highlight the importance of spatial resolution and the influence of acquisition time of satellite images on the estimation of yield at the field scale by remote-sensing data.
Article
Leaf Area Index (LAI) is a fundamental vegetation biophysical variable serving as an essential input to many land surface and atmospheric models. Long-term LAI maps are typically generated with satellite images at moderate spatial resolution (0.25 to 1 km), such as those from the Moderate Resolution Imaging Spectroradiometer (MODIS). While useful for regional-scale land surface modeling, these moderate resolution products often cannot resolve spatial heterogeneity important for many agricultural and hydrological applications. This paper proposes an approach to map LAI at 30-m resolution based on Landsat images for the Contiguous US (CONUS) consistent with the MODIS product, aimed at multi-scale modeling applications. The algorithm was driven by 1.6 million spatially homogeneous samples derived from MODIS LAI and Landsat surface reflectance products from 2006 to 2018. Based on these samples, we trained separate random forest models to estimate LAI from Landsat surface reflectance for eight biomes of the National Land Cover Database (NLCD). A balanced sample design regarding the saturation status of MODIS LAI and a machine-learning-based noise detection technique were introduced to mitigate the trade-off in estimation accuracy between medium LAI (e.g., 3 to 4, unsaturated) and high LAI (e.g., 4–6, saturated). This approach was evaluated using ground measurements from 19 National Ecological Observatory Network (NEON) sites and eight independent sites from other sources. These sites comprise a representative sample of forests, grasslands, shrublands, and croplands across the US. For NEON sites, the LAI estimates show an overall Root Mean Squared Error (RMSE) of 0.8 with r² of 0.88. For the eight independent sites, the Landsat LAI algorithm achieves RMSE between 0.52 and 0.91. The uncertainty in Landsat estimated LAI varies across biomes and locations. The proposed algorithm was implemented on the Google Earth Engine platform, allowing for the rapid generation of long-term high-resolution LAI records from the 1980s using Landsat images (code is available at https://github.com/yanghuikang/Landsat-LAI). Our findings also highlight the importance of sample balance on regression-based modeling in remote sensing applications.
Article
Leaf area index (LAI) and canopy water content (CWC) are important variables for monitoring crop growth and drought, which can be estimated from remotely sensed data. The goal of this study was to evaluate the suitability of the Sentinel-2 multispectral instrument (S2 MSI) data for winter wheat LAI and CWC estimation with three different inversion approaches in the main farming region in North China. During the winter wheat key growth stages in 2017, 22 fields, each with five independent samples, the total number of sample plot is 110, were designed for experimental measurements. In this study, the LAI and CWC were retrieved separately using empirical models through different spectral indices, neural network (NN) algorithms, and lookup table (LUT) methods based on the PROSAIL model. The accuracies of the estimated LAI and CWC were assessed through in situ measurements. The results show that the LUT inversion approach was more suitable for LAI and CWC estimation than the spectral index-based empirical model or the NN algorithm. With the LUT approach, LAI was obtained with a root mean square error (RMSE) of $\text{0.43}\,{\text{m}}^{\text{2}} \!\cdot\! {\text{m}}^{-\text{2}}$ and a relative RMSE (RRMSE) of 11% using seven S2 MSI bands, and CWC was obtained with an RMSE of $\text{0.41}\,{\text{kg}} \!\cdot\! {\text{m}}^{-\text{2}}$ , and an RRMSE of 32% using five S2 MSI bands. In all the three methods, S2 MSI was sensitive to LAI variation and able to reach higher accuracies when red edge bands were used. However, CWC inversion was still a challenge using S2 MSI data.
Article
Full-text available
Soil erosion is an important phenomenon that contributes to the degradation of agricultural land. Even though it is a natural process, human activities can significantly increase its impact on land degradation and present serious limitation on sustainable agricultural land use. Nowadays, the risk of soil erosion is assessed either qualitatively by expert assessment or quantitatively using model-based approach. One of the primary factors affecting the soil erosion assessment is a cover-management factor, C-factor. In the Czech Republic, several models are used to assess the C-factor on a long-term basis based on data collected using traditional tabular methods. This paper presents work to investigate the estimation of both long-term and short-term cover-management factors using remote sensing data. The results demonstrate a successful development of C-factor maps for each month of 2014, growing season average, and annual average for the Czech Republic. C-factor values calculated from remote sensing data confirmed expected trend in their temporal variability for selected crops. The results presented in this paper can be used for enhancing existing methods for estimating C-factor, planning future agricultural activities, and designing technical remediations and improvement activities of land use in the Czech Republic, which are also financially supported by the European Union funds.
Article
Spectral invariants can be considered fundamental descriptors of the impact of canopy structure on canopy scattering where the scattering objects are large compared with the wavelength of radiation. The paper uses the concept of canopy spectral invariants to explore scaling relationships within canopy scattering. A new approximation to the leaf-level PROSPECT scattering model of Jacquemoud and Baret [Jacquemoud, S., & Baret, F. (1990). PROSPECT: A model of leaf optical properties spectra. Remote Sensing of Environment, 34, 75–91.] is developed by applying the spectral invariant approach to leaf internal scattering, in a similar manner to that which has been used to describe canopy-level scattering. We show that it is possible to express both the canopy- and leaf-level single scattering albedo as a function of canopy spectral invariants. This approach provides a framework through which structural information can be maintained in a self-consistent manner across multiple scales from leaf- to canopy-level scattering, at least for the simple canopy architectures considered. It is demonstrated that the nesting of scales described in these relationships implies limits to the retrieval of absolute concentrations of any biochemical constituents or absolute quantities of the amount of scattering material from hyperspectral observations of total scattering. The implication is that in general it may not be possible to separate the components of structural and biochemical influences on measured total scattering signals.
Article
The concept of recollision probability originates from the theory of canopy spectral invariants (‘p-theory’) but is a simplification that involves several heuristic assumptions. Nonetheless, the concept has been shown to be a useful tool for incorporating the effects of 3D structure on canopy absorptive and reflective properties in forest reflectance models. A method is presented by which an average value of the canopy recollision probability (pˆ) can be calculated from canopy gap fraction data, which are provided for example by the LAI-2000 plant canopy analyzer or can be extracted from fisheye photographs. The new method was used to calculate the average recollision probabilities (pˆ values) of uniform leaf and shoot canopies, showing good agreement with results from Monte Carlo simulations. Strengths of the method presented here are the explicitly formulated relationship between recollision probability and canopy structure, and its direct applicability in canopy RT studies.
Article
A neural network is developed to operationally estimate biophysical variables over land surfaces from the observations of the ENVISAT-MERIS instrument: the leaf area index (LAI), the fraction of absorbed photosynthetically active radiation (fAPAR), the fraction of vegetation cover (fCover), and the canopy chlorophyll content (LAI×Cab). The neural network requires as input the geometry of observation and the top of canopy reflectances, corrected from the atmospheric effects, in eleven spectral bands. It is trained on a reflectance database made of radiative transfer model simulations. The principles underlying the generation of the database and the design of the network are first presented. The estimated variables are then compared to other existing products, LAI- and fAPAR-MODIS and MGVI-MERIS, and validated against ground measurements performed in the framework of the VALERI project. Results show remarkable consistency of the temporal dynamics between the several products with however some differences in the range of variation. When compared to actual VALERI ground measurements, the proposed algorithm shows the best performances for LAI (RMSE = 0.47) and fAPAR (RMSE = 0.09).
Article
The combined PROSPECT leaf optical properties model and SAIL canopy bidirectional reflectance model, also referred to as PROSAIL, has been used for about sixteen years to study plant canopy spectral and directional reflectance in the solar domain. PROSAIL has also been used to develop new methods for retrieval of vegetation biophysical properties. It links the spectral variation of canopy reflectance, which is mainly related to leaf biochemical contents, with its directional variation, which is primarily related to canopy architecture and soil/vegetation contrast. This link is key to simultaneous estimation of canopy biophysical/structural variables for applications in agriculture, plant physiology, or ecology, at different scales. PROSAIL has become one of the most popular radiative transfer tools due to its ease of use, general robustness, and consistent validation by lab/field/space experiments over the years. However, PROSPECT and SAIL are still evolving: they have undergone recent improvements both at the leaf and the plant levels. This paper provides an extensive review of the PROSAIL developments in the context of canopy biophysics and radiative transfer modeling
S2PAD -Sentinel-2 MSI -Level 2B Products Algorithm
  • F Baret
  • M Weiss
F. Baret. and M. Weiss, "S2PAD -Sentinel-2 MSI -Level 2B Products Algorithm,", S2PAD-VEGA-ATBD-0003, INRA, 2012.
Product User Manual, VEGETATION PARAMETERS ( FVC, LAI, FAPAR )
  • F Camacho
F. Camacho, "Product User Manual, VEGETATION PARAMETERS ( FVC, LAI, FAPAR ),"
Sentinel 2 Mission Requirements Document
ESA,"Sentinel 2 Mission Requirements Document," EOP-SM/1163/MR-dr, ESA, 2009.
  • F Baret
  • M Weiss
  • F Camacho
  • B Berthelot
  • R Fernandes
  • R Duca
F. Baret, M. Weiss, F. Camacho, B. Berthelot, R. Fernandes, and R. Duca, "VALSE2. Validation of Sentinel 2 Final Report", ESA, 2014.
CCRS Red-Edge LAI ATBD
  • R Fernandes
R. Fernandes, "CCRS Red-Edge LAI ATBD," VALSE2-TN-012-CCRS,ESA, 2014..