Use of TRMM and GPM Multi-Satellite Precipitation Data for Hydrological Modeling

Abstract:

Satellite data has proven to be a credible alternative to gauge precipitation for use in hydrologic models as gauge data are sparse and sometimes not available. In the case of large watersheds, the spatial distribution of precipitation is the key for representing hydrologic variability. Spatially distributed gauge data is usually available for experimental watersheds at small scales. Satellite precipitation data such as the Integrated MultisatellitE Retrievals for GPM (IMERG) and the Tropical Rainfall Measuring Mission (TRMM) Multi-satellite Precipitation Analysis (TMPA) are the most extensively available global data. TMPA data is available at 0.25° spatial and 3 hour temporal resolution for the latitude band 50°S-50°N . TMPA data has been used in hydrologic models. With the end of the TRMM mission, more-detailed IMERG data is replacing TMPA. IMERG data is available at 0.1° spatial and 0.5 hour temporal resolution for the latitude band 60°S-60°N. IMERG data is in its early stages and its application in hydrologic models has not been extensively tested. The objective of this study is to evaluate IMERG precipitation data in a hydrologic model against gauge, radar and TMPA precipitation. We created a model of the Upper Mississippi River Basin (UMRB) using the Soil and Water Assessment Tool (SWAT). We have used areal averaging for IMERG, TMPA and radar precipitation data, and Ordinary Krigging (OK) for gauge precipitations to interpolate precipitation data to SWAT model sub-watersheds. Finally, SWAT model simulation results are compared for each precipitation data input. Preliminary results indicated that IMERG data produced promising model performance compare to other precipitation inputs. The calibrated SWAT model results for each precipitation data will be presented.