Abstract:
Land surface processes significantly influence weather thereby affecting terrestrial water balance. Even though land surface models (lsms) offer a complementary means of examining fluxes, their simulations are often subjected to uncertainties from model parameters, model structure and forcing data. Of these, precipitation forcing uncertainty is known to significantly contribute to lsm output. This study examines the impact of different satellite-based precipitation products on lsm simulated soil moisture over india based on a data period of 8 years (2008 to 2015). The precipitation products used are namely, the global precipitation measurement mission (gpm) integrated multi-satellite retrievals (imerg) late run, sm2rain-climate change initiative (sm2rain-cci) and sm2rain-advanced scaterometer (sm2rain-ascat). The first product is a multi-satellite precipitation product (top-down approach) with the remaining two products derived from surface satellite soil moisture (bottom up approach). The uncertainty in these three precipitation products are evaluated against simulations resulting from india meteorological department (imd) precipitation dataset. The imerg precipitation indicates a bias (mm) of 1.3014 and rmse (mm) of 1.8024 during monsoon. Whereas for sm2rain-ascat it shows a bias of 0.0206 and rmse of 0.9368 during monsoon and sm2rain-cci which showed a bias of −0.3711 and rmse of 0.9345 during monsoon. Vic simulated soil moisture using imerg indicate a bias (m3/m3) of 0.0356 and rmse (m3/m3) of 0.0402 during monsoon. This is in comparison to sm2rain-ascat which showed a bias of 0.0211 and rmse of 0.0286 during monsoon and sm2rain-cci which showed a bias of −0.0178 and rmse of 0.0327 during monsoon. Our study has evaluated precipitation products derived from two different algorithms i.e top down and bottom up approach, it will provide insights on which algorithm is performing better in a country like india on different topographic condition. © 2021