Evaluation of TRMM satellite rainfall in driving Variable Infiltration Capacity (VIC) model in Ganjiang Basin

摘要

Global rainfall data is very important for water resource research. Current rainfall data are mainly from the rain gauge on the ground, ground radar and space borne passive radiometer. This study assesses the successive Version-6 and Version-7 TRMM Multi-satellite Precipitation Analysis (TMPA) products (3B42V6, 3B42V7, 3B42RTV6 and 3B42RTV7) in Ganjiang River Basin, southeastern China. The results shows that compared with ground observed precipitation, the TMPA rainfall estimates are well captured by 3B42V6, 3B42V7 and 3B42RTV7. In terms of daily grid-based comparison, 3B42RTV7 has been improved over 3B42RTV6 by reducing relative bias from -24, 15% to 12.05%, and the root-mean-square error (RMSE) improves from 5.98mm to 5.69mm. Using these TMPA products as input to simulate daily hydrologic with VIC hydrologic model shows that: 1) 3B42V6 and 3B42V7 show very comparable hydrologic skills, with high Nash-Sutcliffe index (NSCE, 0.72 and 0.75, respectively) and strong correlation (CC=0.89); 2) 3B42RTV7 shows display a better hydrologic performance over 3B42RTV6 by increasing the NSCE from 0.55 to 0.58, improving CC from 0.81 to 0.86, and reducing relative bias from -34.28% to 15.71%.