Prediction of Root Zone Soil Moisture using Remote Sensing Products and In-Situ Observation under Climate Change Scenario

Abstract:

Prediction of root zone soil moisture status at field level is vital for developing efficient agricultural water management schemes. In this study, root zone soil moisture was estimated across the Rana watershed in Eastern India, by assimilation of near-surface soil moisture estimate from SMOS satellite into a physically-based Soil-Water-Atmosphere-Plant (SWAP) model. An ensemble Kalman filter (EnKF) technique coupled with SWAP model was used for assimilating the satellite soil moisture observation at different spatial scales. The universal triangle concept and artificial intelligence techniques were applied to disaggregate the SMOS satellite monitored near-surface soil moisture at a 40 km resolution to finer scale (1 km resolution), using higher spatial resolution of MODIS derived vegetation indices (NDVI) and land surface temperature (T_s). The disaggregated surface soil moisture were compared to ground-based measurements in diverse landscape using portable impedance probe and gravimetric samples. Simulated root zone soil moisture were compared with continuous soil moisture profile measurements at three monitoring stations. In addition, the impact of projected climate change on root zone soil moisture were also evaluated. The climate change projections of rainfall were analyzed for the Rana watershed from statistically downscaled Global Circulation Models (GCMs). The long-term root zone soil moisture dynamics were estimated by including a rainfall generator of likely scenarios. The predicted long term root zone soil moisture status at finer scale can help in developing efficient agricultural water management schemes to increase crop production, which lead to enhance the water use efficiency.