The relationship among precipitation, cloud-top temperature, and precipitable water over the tropics

The relationship of monthly precipitation P to precipitable water w and cloud-top temperature as represented by the Geostationary Operational Environmental Satellite (GOES) Precipitation Index (GPI) is obtained over tropical land, coast, and ocean: P = exp[a₁(w - a₂)] GPI, where coefficients a₁ and a₂ are determined using one year of the Global Precipitation Climatology Project (GPCP) monthly rain gauge data and then independently tested using four other years of gauge data. This algorithm, over land, gives more accurate precipitation estimates than are obtained using the cloud-top temperature alone (i.e., GPI) and is as accurate as the state-of-the-art multisatellite algorithm (MS) from GPCP. Over coastal and oceanic regions, this algorithm has a smaller bias in precipitation estimation than GPI but has the same correlation coefficient with gauge data as GPI. Compared with MS, it has a much smaller bias but larger mean absolute deviation. Evaluation using the Pacific atoll-island gauge data also shows that this algorithm can reproduce well the observed meridional distribution of precipitation across the ITCZ and SPCZ near the date line. This algorithm is then used to produce a five-year (January 1988-December 1992) 2.5°X 2.5°integrated dataset of precipitation and precipitable water between 40°N and 40°S for climate model evaluation. The small bias of this algorithm (particularly over ocean) also suggests that it would be a good data source for precipitation merging algorithms.