Surface Soil Moisture Estimation during Dry Winter in the Terai Region of Eastern India: Use of Trapezoid Models based on Landsat 8 and Sentinel 2 Data

Abstract

In the eastern Terai ecoregion of India, the winter and pre-monsoon months experience a dry period with scanty rainfall. Surface soil moisture (SSM) status becomes a key state variable governing soil and crop health during this period. In this context, we evaluated the capability of two trapezoid models namely, Thermal-optical trapezoid model (TOTRAM) and Optical trapezoid model (OPTRAM) to estimate SSM using spatial data in the dry winter month of December, 2020 covering four Community Development Blocks of this region. Using optical and thermal satellite data (Landsat 8 and Sentinel 2), the physical foundation of these models was tested through pixel-cloud distribution within the trapezoid spaces. The OPTRAM model was found to be advantageous over TOTRAM for easier parameterization. Landsat 8 data based TOTRAM overestimated the SSM, while Sentinel 2 data based OPTRAM underestimated the same. Landsat 8 data based OPTRAM showed the best goodness of fit with ground measured SSM (R2 = 0.648, MAE = 0.081, RMSE = 0.096). The study inferred that soil type, land use and crops influenced the SSM. The overall SSM of the study area was found good even in the dry winter month of December. The high groundwater table and possible capillary rise of water was found to be the contributing factors behind this. The success of this pilot-scale research indicated the opportunity to measure the soil moisture of the entire Indian Terai ecoregion in dry months for site- and need-specific irrigation scheduling.