Improving multiple satellite precipitation estimation for hydrologic applications and uncertainty analysis

Recent development of satellite-based retrieval techniques has made it possible for monitoring precipitation events with improved quality for hydrologic applications. In this presentation, our recent progress of multiple satellite precipitation monitoring systems at UC Irvine will be presented. These multiple satellite algorithms using both Low Earth Orbital (LEO) and Geostationary Earth Orbiting (GEO) satellite information to estimate near real-time estimation of global precipitation. Retrospective estimation of precipitation climatology is also made available using historical LEO and GEO satellite imagery. This precipitation climate dataset is made available at daily and 0.25ox0.25o scale from 1983 to near current time. Uncertainty of precipitation estimation is presented using a generalized normal distribution. The parameters of the distribution is further extended to estimate uncertainty of precipitation products are various spatial and temporal scale. Experiments of using satellite-based precipitation products and uncertainty estimation for catchment scale streamflow simulation and analyzing the impact of precipitation uncertainty on the hydrological responses will be discussed.