Validation of rain/no-rain discrimination in the standard TRMM data products 1B21 and 1C21

Satellite observation of rain is attractive because it provides information about rainfall rates on a global basis that is unavailable with conventional ground-based instruments. The Tropical Rainfall Measuring Mission (TRMM) is a United States-Japan joint project for rain measurement from space. The first spaceborne rain radar has been aboard the TRMM satellite. Meteorological radar commonly uses averaging to reduce the effects of signal fluctuation. Mean values of the TRMM Precipitation Radar (TRMM/PR) signal are extracted from the fluctuating radar signals by averaging. Rain/no-rain discrimination is done basically to classify mean values of radar signals. Since experimentally obtained values always have some errors, one can not simply classify mean values of radar signal as either rain or no-rain. To meet the requirements for the various applications of the rain/no-rain discrimination, the rain status is classified into three classes as rain-certain, rain-possible, or no-rain by using two threshold values. This classification is applied to each angle bin of the TRMM/PR. In this paper, rain/no-rain discrimination for TRMM/PR signals has been validated by the simulation, and by a comparison with the ground-based radar data