Monday, 10 February 2003
Impact of Regime Changes on Climate Rainfall Estimates from Passive Microwave Sensors
Reliable rainfall estimates from passive microwave sensors extend back to July of 1987 and provide the most physically direct measurements of global rainfall for many climate applications. While there is reasonably good agreement between rainfall retrievals from existing passive microwave sensors and algorithms, differences in seasonal and interannual variability in tropical mean rainfall still exist when compared to radar-derived estimates from the precipitation radar (PR) on board the Tropical Rainfall Measuring Mission (TRMM). Although there are issues related to the assumption of a constant drop size distribution in the PR retrievals, the TRMM radar provides globally consistent high resolution observations of the horizontal and vertical structure of precipitating storm systems. Therefore, we have investigated how regional, seasonal, and interannual changes in rainfall systems impact satellite retrievals from passive microwave observations using data from five years of available TRMM observations.
All satellite passive microwave techniques for estimating rainfall require assumptions regarding the atmospheric background state and the distribution of liquid and frozen hydrometeors within the atmospheric column. For example, retrievals are affected by changes in the freezing level, the vertical distribution of liquid and frozen hydrometeors, and horizontal inhomogeneity within the satellite field-of-view. Based on comparisons with the TRMM PR we will show how variations in precipitation systems due to differences in meteorological regimes produce systematic biases in many of the current retrievals. We will also show how several of the regime dependent biases we have identified to date impact variability in tropical mean rainfall.