An investigation of inversion methodologies to retrieve the leaf area index of corn from C-band SAR data

Mandal, Dipankar; Hosseini, Mehdi; McNairn, Heather; Kumar, Vineet; Bhattacharya, Avik; Rao, Y. S.; Mitchell, Scott; Robertson, Laura Dingle; Davidson, Andrew; Dąbrowska-Zielińska, K.

doi:10.1016/j.jag.2019.06.003

Cited by 32 publications

(20 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While for various types of crops, many studies have shown a positive relationship between VH backscatter coefficient in C-band and LAI or NDVI (Mandal et al, 2019; Wang et al, 2019), our results show that, in deciduous forests, these relationships are inverted. Temporal dynamics of LAI and NDVI is accompanied by a decrease in VH and an increase of VV/VH ratio.…”

Section: Discussioncontrasting

confidence: 65%

Potential of Synthetic Aperture Radar Sentinel-1 time series for the monitoring of phenological cycles in a deciduous forest

Soudani

Delpierre

Berveiller

et al. 2021

Preprint

View full text Add to dashboard Cite

Annual time-series of the two satellites C-band SAR (Synthetic Aperture Radar) Sentinel-1 A and B data over five years were used to characterize the phenological cycle of a temperate deciduous forest. Six phenological markers of the start, middle and end of budburst and leaf expansion stage in spring and the leaf senescence in autumn were extracted from time-series of the ratio (VV/VH) of backscattering at co-polarization VV (vertical-vertical) and at cross polarization VH (vertical-horizontal). These markers were compared to field phenological observations, and to phenological dates derived from various proxies (Normalized Difference Vegetation Index NDVI time-series from Sentinel-2 A and B images, in situ NDVI measurements, Leaf Area Index LAI and litterfall temporal dynamics). We observe a decrease in the backscattering coefficient (σ0) at VH cross polarization during the leaf development and expansion phase in spring and an increase during the senescence phase, contrary to what is usually observed on various types of crops. In vertical polarization, σ0VV shows very little variation throughout the year. S-1 time series of VV/VH ratio provides a good description of the seasonal vegetation cycle allowing the estimation of spring and autumn phenological markers. Estimates provided by VV/VH of budburst dates differ by approximately 8 days on average from phenological observations. During senescence phase, estimates are positively shifted (later) and deviate by about 20 days from phenological observations of leaf senescence while the differences are of the order of 2 to 4 days between the phenological observations and estimates based on in situ NDVI and LAI time-series, respectively. A deviation of about 7 days, comparable to that observed during budburst, is obtained between the estimates of senescence from S-1 and those determined from the in situ monitoring of litterfall. While in spring, leaf emergence and expansion described by LAI or NDVI explains the increase of VV/VH (or the decrease of σ0VH), during senescence, S-1 VV/VH is decorrelated from LAI or NDVI and is better explained by litterfall temporal dynamics. This behavior resulted in a hysteresis phenomenon observed on the relationships between VV/VH and NDVI or LAI. For the same LAI or NDVI, the response of VV/VH is different depending on the phenological phase considered. This study shows the high potential offered by Sentinel-1 SAR C-band time series for the detection of forest phenology for the first time, thus overcoming the limitations caused by cloud cover in optical remote sensing of vegetation phenology.

show abstract

Section: Discussioncontrasting

confidence: 65%

Potential of Synthetic Aperture Radar Sentinel-1 time series for the monitoring of phenological cycles in a deciduous forest

Soudani

Delpierre

Berveiller

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Several studies have used semi-empirical modeling to characterize the contribution of these soil and vegetation parameters to SAR backscatter, with promising results [8][9][10][11][12]. The water cloud model (WCM) [13] is a semi-empirical radiative transfer model which has been used successfully to estimate crop biophysical parameters such as LAI and biomass [14][15][16][17][18][19]. For example, Bériaux et al (2011) used the WCM with C-Band satellites (ERS-1/2, ENVISAT/ASAR and RADARSAT-2) to estimate wheat LAI, later applying the WCM to corn LAI [11,16].…”

Section: Introductionmentioning

confidence: 99%

A Comparison between Support Vector Machine and Water Cloud Model for Estimating Crop Leaf Area Index

et al. 2021

Self Cite

View full text Add to dashboard Cite

The water cloud model (WCM) can be inverted to estimate leaf area index (LAI) using the intensity of backscatter from synthetic aperture radar (SAR) sensors. Published studies have demonstrated that the WCM can accurately estimate LAI if the model is effectively calibrated. However, calibration of this model requires access to field measures of LAI as well as soil moisture. In contrast, machine learning (ML) algorithms can be trained to estimate LAI from satellite data, even if field moisture measures are not available. In this study, a support vector machine (SVM) was trained to estimate the LAI for corn, soybeans, rice, and wheat crops. These results were compared to LAI estimates from the WCM. To complete this comparison, in situ and satellite data were collected from seven Joint Experiment for Crop Assessment and Monitoring (JECAM) sites located in Argentina, Canada, Germany, India, Poland, Ukraine and the United States of America (U.S.A.). The models used C-Band backscatter intensity for two polarizations (like-polarization (VV) and cross-polarization (VH)) acquired by the RADARSAT-2 and Sentinel-1 SAR satellites. Both the WCM and SVM models performed well in estimating the LAI of corn. For the SVM, the correlation (R) between estimated LAI for corn and LAI measured in situ was reported as 0.93, with a root mean square error (RMSE) of 0.64 m2 m−2 and mean absolute error (MAE) of 0.51 m2 m−2. The WCM produced an R-value of 0.89, with only slightly higher errors (RMSE of 0.75 m2 m−2 and MAE of 0.61 m2 m−2) when estimating corn LAI. For rice, only the SVM model was tested, given the lack of soil moisture measures for this crop. In this case, both high correlations and low errors were observed in estimating the LAI of rice using SVM (R of 0.96, RMSE of 0.41 m2 m−2 and MAE of 0.30 m2 m−2). However, the results demonstrated that when the calibration points were limited (in this case for soybeans), the WCM outperformed the SVM model. This study demonstrates the importance of testing different modeling approaches over diverse agro-ecosystems to increase confidence in model performance.

show abstract

“…However, the iterative method produce accurate estimates at the expense of high computational resources when optimizing such inversion problems. Mandal et al (2019a) indicated the highest computation-intensive nature of IO approach in memory-time performance analysis compared with other inversion approaches. Besides, the intrinsic problem associated with such an optimization of non-linear multi-variate merit function is the possibility to get confined into a local minimum instead of reaching the global minimum (Perez, Jansen, and Martins 2012).…”

Section: Introductionmentioning

confidence: 99%

“…Recognizing the potential issues of the traditional approaches (IO and LUT search techniques) for applications to larger areas, ill-posed inversion problems in remote sensing are often solved by data-driven nonparametric models (Durbha, King, and Younan 2007;Bériaux, Lambot, and Defourny 2011;Verrelst et al 2012;Mandal et al 2019a), which provides a stable and optimum solution. Also, machine learning regression models can provide an optimum solution with a lower computational cost (Mandal et al 2019a). Machine learning regression approaches such as the support vector regression (SVR) can cope with the nonlinearity of the functional dependence between crop descriptors (in WCM) and the SAR backscatter intensities.…”

Section: Introductionmentioning

confidence: 99%

“…Machine learning regression approaches such as the support vector regression (SVR) can cope with the nonlinearity of the functional dependence between crop descriptors (in WCM) and the SAR backscatter intensities. Hence, SVR can serve as a suitable technique for model inversion in operational scale applications (Caicedo et al 2014;Mandal et al 2019a). However, the formulation of standard SVR involves a single target output.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Crop biophysical parameter retrieval from Sentinel-1 SAR data with a multi-target inversion of Water Cloud Model

Mandal

Kumar

López-Sánchez

et al. 2020

International Journal of Remote Sensing

Self Cite

View full text Add to dashboard Cite

Estimation of bio-and geophysical parameters from Earth observation (EO) data is essential for developing applications on crop growth monitoring. High spatiotemporal resolution and wide spatial coverage provided by EO satellite data are key inputs for operational crop monitoring. In Synthetic Aperture Radar (SAR) applications, a semi-empirical model (viz., Water Cloud Model (WCM)) is often used to estimate vegetation descriptors individually. However, a simultaneous estimation of these vegetation descriptors would be logical given their inherent correlation, which is seldom preserved in the estimation of individual descriptors by separate inversion models. This functional relationship between biophysical parameters is essential for crop yield models, given that their variations often follow different distribution throughout crop development stages. However, estimating individual parameters with independent inversion models presume a simple relationship (potentially linear) between the biophysical parameters. Alternatively, a multi-target inversion approach would be more effective for this aspect of model inversion compared to an individual estimation approach. In the present research, the multi-output support vector regression (MSVR) technique is used for inversion of the WCM from C-band dual-pol Sentinel-1 SAR data. Plant Area Index (PAI, m 2 m −2 ) and wet biomass (W, kg m −2 ) are used as the vegetation descriptors in the WCM. The performance of the inversion approach is evaluated with in-situ measurements collected over the test site in Manitoba (Canada), which is a super-site in the Joint Experiment for Crop Assessment and Monitoring (JECAM) SAR inter-comparison experiment network. The validation results indicate a good correlation with acceptable error estimates (normalized root mean square error-nRMSE and mean absolute error-MAE) for both PAI and wet biomass for the MSVR approach and a better estimation with MSVR than single-target models (support vector regression-SVR). Furthermore, the correlation between PAI and wet biomass is assessed using the MSVR and SVR model. Contrary to the single output SVR, the correlation between biophysical parameters is adequately taken into account in MSVR based simultaneous inversion technique. Finally, the spatio-temporal maps for PAI and W at different growth stages indicate their variability with crop development over the test site. KEYWORDS model inversion; ill-posed problem; biomass; PAI; JECAM.

show abstract

An investigation of inversion methodologies to retrieve the leaf area index of corn from C-band SAR data

Cited by 32 publications

References 58 publications

Potential of Synthetic Aperture Radar Sentinel-1 time series for the monitoring of phenological cycles in a deciduous forest

Potential of Synthetic Aperture Radar Sentinel-1 time series for the monitoring of phenological cycles in a deciduous forest

A Comparison between Support Vector Machine and Water Cloud Model for Estimating Crop Leaf Area Index

Crop biophysical parameter retrieval from Sentinel-1 SAR data with a multi-target inversion of Water Cloud Model

Contact Info

Product

Resources

About