Application of sentinel-1 SAR-derived vegetation descriptors for soil moisture retrieval and plant height prediction during the wheat growth cycle

ABSTRACT

Soil moisture is the crucialparameter impacting the plant growth during its phenological cycle. The Water Cloud Model (WCM) is one of the most widely used semi-empirical models for retrieval of soil moisture of vegetated lands from synthetic aperture radar (SAR) images. The model considers the effect of soil as well as vegetation on the radar backscatter. The study assesses the applicability of Sentinel-1 SAR-derived vegetation descriptors in the WCM for soil moisture retrieval during the wheat growth cycle. Various combinations of vegetation descriptors (V₁and V₂), viz. VH polarized backscatter coefficient (σ⁰_VH), Radar Vegetation Index (RVI) and depolarization ratio (χ_v), were used in the model. The model performed better when different parameters are used as vegetation descriptors (V₁≠V₂) in the WCM rather than using the same parameter for both the vegetation descriptors (V₁= V₂). The best results were observed when σ⁰_VHwas considered as one of the vegetation descriptors (V₁) while either χ_vor RVI were utilized as the other vegetation descriptor (V₂) giving a Pearson correlation coefficient (R) of 0.959 and 0.958 and a root mean square error (RMSE) of 0.499 dB and 0.516 dB respectively. The validation of the model-retrieved soil moisture against the in-situ measured values gave an R value of 0.72 and a RMSE of 0.096m³/m³. The plant height was also predicted by the WCM in which the retrieved soil moisture from SAR data was used as a parameter. The predicted plant height was compared to in-situ measured plant height and an R value of 0.76 and RMSE of 0.214 was obtained as the best result. The study demonstrates the capability of SAR-derived parameters as vegetation descriptors in the WCM for soil moisture retrieval.

KEYWORDS:

• Soil moisture
• Water Cloud Model, Sentinel-1
• Vegetation descriptors
• Plant height

Acknowledgement

The authors acknowledge the research facility provided by the Director of Research, Anand Agricultural University to support the research work. Authors extend thanks to the Dy. Director and Head, MTDD/AMHTDG/EPSA, Space Applications Centre, ISRO, Ahmedabad for providing funds to conduct the research work under MAHTRAM, Soil Moisture collaborative study.

Disclosure statement

No potential conflict of interest was reported by the authors.