An Assessment of the First- and Second-Generation Navy Operational Precipitation Retrieval Algorithms

Wesley Berg

Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder, Colorado

William Olson

Caelum Research Corporation, Silver Spring, Maryland

Ralph Ferraro

NOAA/Satellite Research Laboratory, Camp Springs, Maryland

Steven J. Goodman

NASA/Marshall Space Flight Center, Huntsville, Alabama

Frank J. LaFontaine

Hughes STX, Huntsville, Alabama


Rainfall estimates produced from the Special Sensor Microwave/Imager (SSM/I) data have been utilized operationally by the United States Navy since the launch of the first SSM/I sensor in June of 1987. The navy initially contracted Hughes Aircraft Company to develop a rainfall-retrieval algorithm prior to the launch of SSM/I. This first-generation operational navy rainfall retrieval algorithm, referred to as the D-Matrix algorithm, was used until the development of the second-generation algorithm by the SSM/I Calibration/Validation team, which has subsequently been replaced by a third-generation algorithm developed by the National Oceanic and Atmospheric Administration/National Environmental Satellite, Data and Information System. Results from both the D-Matrix and Cal/Val algorithms have been included in a total of five algorithm intercomparison projects conducted through the Global Precipitation Climatology Project and WetNet. A comprehensive summary of both quantitative and qualitative results from these intercomparisons is given detailing many of the strengths and weaknesses of the algorithms. Based on these results, the D-Matrix algorithm was found to produce excessively large estimates over land and to poorly represent the spatial structure of rainfall systems, especially at higher latitudes. The Cal/Val algorithm produces more realistic structure within storm systems but appears to overestimate the region of precipitation for many systems and significantly underestimates regions of intense rainfall. While the Cal/Val algorithm appears to provide better instantaneous rainfall estimates in the Tropics, the D-Matrix algorithm provides reasonable time-averaged results for monthly or longer periods.