Study of water cloud model vegetation descriptors in estimating soil moisture in Solani catchment

Abstract

AbstractWater cloud model (WCM) relates the backscatter coefficient (σo) with soil moisture. The backscatter coefficient includes the backscatter coefficient due to vegetation (σoveg), and the backscatter coefficient due to soil (σosoil). The σoveg of WCM depends upon vegetation characteristics. The present study is aimed to investigate the effect of different vegetation descriptors in estimating soil moisture from WCM. The study is carried out in Solani River catchment of India. Envisat Advanced Synthetic Aperture Radar (ASAR) images of three dates were acquired for the study. The field data, volumetric soil moisture from the upper 0–10 cm soil layer, soil texture, soil surface roughness, leaf area index (LAI), leaf water area index, normalized plant water content and average plant height corresponding to satellite pass dates were collected. Genetic algorithm optimization technique is used to estimate the WCM vegetation parameters. The use of LAI as vegetation descriptor results in minimum root mean square error (RMSE) of 1.77 dB between WCM computed backscatter and Envisat ASAR observed backscatter. Also, use of LAI in WCM as vegetation descriptor results in the least RMSE of 4.19%, between estimated and observed soil moisture for the first field campaign, whereas it was 5.64% for the last field campaign which was undertaken after 35 days of first campaign. It is concluded that LAI can be treated as the best vegetation descriptor in studies retrieving soil moisture and backscatter from microwave remote sensing data. Copyright © 2014 John Wiley & Sons, Ltd.