Passive microwave radiometer channel selection based on cloud and precipitation information content

The information content of microwave frequencies between 5 and 200 GHz for rain, snow and cloud water retrievals over ocean and land surfaces was evaluated using optimal estimation theory. The study was based on large datasets representative of summer and winter meteorological conditions over North America, Europe, Central Africa, South America and the Atlantic obtained from short-range forecasts with the operational ECMWF model. The information content of rain, snow and cloud water was traded off against the uncertainties due to the natural variability of other variables that microwave observations are sensitive to. These are surface emissivity, land surface skin temperature, atmospheric temperature and moisture. The estimation of the underlying error statistics was based on ECMWF model forecast error statistics. The results suggest that a number of frequency bands are most suited for the retrieval of (i) rain over oceans: 15-18, 35-40, 80, 145, 118.75 10-14 GHz; rain over land: 85-100, 135-140 GHz, (ii) snow over land and oceans: 95-100, 140-150, 187 GHz, (iii) clouds over oceans: 40, 80-85 GHz; clouds over land: 90-100, 135-140 GHz. For radiometers designed for global and multi-season applications, several channels in all of the above frequency ranges would be desirable for optimizing channel usage in hydrometeor retrievals depending on the observed situation.