EVALUATION OF POTENTIAL ERROR IN AREAL MEAN PRECIPITATION AND ITS IMPACT ON RAINFALL-RUNOFF SIMULATION USING SATELLITE PRECIPITATION PRODUCT

摘要

Areal Mean Precipitation (AMP) uncertainties have been a long concerning issue. It is necessary to quantify the AMPerror and assess its impacts on Rainfall Runoff (RR) simulation, which is one of the guidance for improving theaccuracy and robustness of RR models. In this paper, AMP uncertainties were quantified by both published andproposed indices in five variety sized catchments in Asia. Moreover, AMP errors were also computed using both in-situobservations and remote sensing data. Satellite based precipitation product (GsMAP-MVK) was used with the effort tomitigate the problem of sparse data, especially severe in poorly gauged river basins. The declining trends in the relationof model performances and AMP error have been observed. The AMP error estimation was improved when accountingfor the spatial variability of precipitation and the gauge locations along with the gauge density. The daily basisGsMAP-MVK grasps the precipitation variability in space better than the annual basis and it can reliably use to estimatethe AMP error based on “Sample Design method” (Esdd).