The Global Precipitation Mission (GPM), aside from a powerful core satellite, contains a conceptual element that combines many independent radiometers into a coherent framework dedicated to providing a global, 3-hourly rainfall product. To that end, the previously funded Precipitation Measurement Missions proposal was aimed at not only (a) delivering V6 of the GPROF algorithm, which is now operational, but also (b) developing the framework for a parametric algorithm for use with the GPM constellation satellites. This parametric approach uses TRMM PR and TMI to build significantly improved and transportable databases for the Bayesian scheme used in GPROF. These improved databases were constructed on a physical foundation that makes it relatively straightforward to adapt the results to different sensors such as SSMI, AMSR-E, SSMIS and WindSat. This new proposal is intended to expand upon the foundation laid in the previous research cycle by (a) delivering the TMI version (2A12 V7) of the new code to the PPS in the December 2007 timeframe, (b) implementing the new algorithm for the additional radiometers now flying, (c) explicitly addressing the distinct sources of errors in the new algorithm and quantifying the resulting net uncertainty, and (d) laying the foundation for the retrieval structure in the GPM timeframe when a dual-frequency radar with more detailed physical information about drop size distributions and High Frequency (HF) microwave channels can be used to construct the a-priori databases.
