Passive microwave remote sensing from satellites has become the primary basis for global monitoring of precipitation. Cold-cloud precipitation processes dominate in midlatitude precipitation systems. Accurate models of the microwave optical properties of ice and mixed phase hydrometeors is critical to the success of physically based precipitation retrieval algorithms. To date, however, microwave radiative transfer calculations have generally relied on ad hoc assumptions about ice and mixed-phased particle properties. The validity of these assumptions, and the sensitivity of radiative transfer results and satellite rainfall retrievals to the associated uncertainties, have never been adequately documented.

We have undertaken a systematic analysis of the microwave brightness temperature variations associated with reasonable variations in assumed hydrometeor properties. In particular, a polarized plane parallel one dimensional radiative transfer model has been utilized to simulate brightness temperatures from idealized rain cloud structures. Changes in assumptions regarding the effective physical properties of frozen or semi-frozen hydrometeors, such as volume fraction of liquid water, ice, and air, as well as distribution of liquid-equivalent particle diameters, lead to significant (order 10 K or more) changes in brightness temperatures obtained from the radiative transfer model.

The ultimate objective of this work is to establish empirical constraints on hydrometeor properties assumed in radiative transfer models, based in part on comparisons between observed and simulated multichannel microwave radiances from current sensors, such as the Tropical Rainfall Measurement Mission (TRMM) Microwave Imager (TMI), the Advanced Microwave Scanning Radiometer (AMSR), and the Advanced Microwave Sounding Unit (AMSU).