Optimal precipitation estimation using multisensor microwave datasets

The quantification and variability of water cycle and its mechanisms are key scientific issues. Systematic errors and biases in precipitation measurements can be eliminated using merging techniques from multisatellite data and data assimilation techniques. Microwave measurement data sources are used in this investigation to merge precipitation estimates from several satellites to produce optimal precipitation datasets. The emphasis of the research is on several case studies with different synoptic conditions and backgrounds to enhance the global precipitation measurement. Analysis and detection of severe snowfall in high latitudes and light rain over desert using AMSU-B and AMSR-E data are discussed. Space-based precipitation estimation is validated using ground-based radar (WSR88D) and rain gauge data for different cases representing different synoptic conditions and surface types. Three hourly global precipitation assimilation schemes for a land-atmosphere couple model system using merged, satellite-based precipitation is explained and the sensitivity of the couple system to the precipitation assimilation is explored. The high frequency passive microwave, such as 150 GHz, is useful particularly for ice and mixed precipitation. Results of the comparative study of precipitation totals based on multisensor estimate, Doppler radar estimate and gauge data are presented.