IMPROVED ESTIMATES OF TROPICAL AND SUBTROPICAL PRECIPITATION USING THE GOES PRECIPITATION INDEX

Nine years (1986-94) of tropical and subtropical precipitation estimat es based on the GOES precipitation index (GPI) are examined. The GPI, based on the results of studies relating fractional coverage of cold c loud to convective rainfall, uses IR observations gathered by geostati onary and polar-ol biting satellites. Longitudinal discontinuities in mean GPI coincident with the boundaries of satellite coverage led to a comparison of GPI derived from each geostationary satellite in overla p regions. This study revealed both intersatellite calibration differe nces and satellite zenith angle dependence. Its goals are to remove th ese sources of systematic error within the GPI, investigate the climat ology of the corrected GPI, and compare against other estimated rainfa ll datasets. To correct calibration differences. Global Precipitation Climatology Project geostationary satellite IR data are standardized t o one satellite by temperature adjustments deduced by the Internationa l Satellite Cloud Climatology Project. The resulting GPI values are co rrected for zenith angle dependence based on a comparison between GOES -7 and Metosat-3 that found a systematic increase in GPI of 9% for eve ry 10 degrees of zenith angle beyond 25 degrees. The corrections remov e noticeable discontinuities in time-averaged GPI and are largest (>2 mm day(-1)) over the eastern Indian Ocean, the equatorial Pacific near the date line, and South America. The spatial correlation between cor rected GPI and rainfall derived from rain gauges is greater than 0.8 i n tropical regions with adequate gauge density. Empirical orthogonal f unctions of monthly anomalies of corrected GPI show the expected El Ni no-Southern Oscillation spatial pattern.