With the launches of the Tropical Rainfall Measuring Mission (TRMM) Microwave Imager (TMI) in 1997 and the Advanced Microwave Sounding Unit (AMSU) in 1998, and the upcoming launch of the Advanced Microwave Sounding Radiometer (AMSR), there will be many sources of information for global, microwave rainfall estimates in addition to the Special Sensor Microwave/Imager (SSM/I). With the reduction of temporal sampling error due to the multiple sensors, algorithm error becomes the main concern. There are many sources of uncertainty in land rainfall estimation that can cause significant error, but in general attempts to quantify the relative contribution of the different sources to the overall error have been unsuccessful. In fact, the error variance of the estimates has not been well-quantified either.

We are attempting to quantify these errors from both physical and statistical approaches. We use observational studies of rainfall processes to understand the physical causes, and we use ground validation estimates whose errors are quantified themselves to quantify the retrieval error. Using these results, we form a models of the quantitative contribution of the individual error sources to the overall errors. These models are sensor-dependent, as each sensor has different frequencies and footprint resolutions.