Radiative Transfer to Space through a Precipitating Cloud at Multiple Microwave Frequencies. Part II: Results and Analysis

Abstract

The time-dependent role of cloud liquid water in conjunction with its vertical heterogeneities on top-of-atmosphere (TOA) passive microwave brightness temperatures is investigated. A cloud simulation is used to specify the microphysical structure of an evolving cumulus cloud growing toward the rain stage. A one-dimensional multistream solution to the radiative transfer equation is used to study the upwelling radiation at the top of the atmosphere arising from the combined effect of cloud, rain, and ice hydrometeors. Calculations are provided at six window frequencies and one H2O resonance band within the EHF/SHF microwave spectrum. Vertically detailed transmission functions are used to help delineate the principal radiative interactions that control TOA brightness temperatures. Brightness temperatures are then associated with a selection of microphysical situations that reveal how an evolving cloud medium attenuates rainfall and surface radiation. The investigation is primarily designed to study the impact of cloud microphysics on space-based measurements of passive microwave signals, specifically as they pertain to the retrieval of precipitation over water and land backgrounds. Results demonstrate the large degree to which the relationship between microwave brightness temperature (BT) and rainrate (RR) can be altered purely by cloud water processes. The relative roles of the cloud and rain drop spectra in emissive contributions to the upwelling radiation are assessed with a normalized absorption index, which removes effects due purely to differences in the magnitudes of the cloud and rain liquid water contents. This index is used to help explain why the amplitudes of the BT-RR functions decrease with respect to cloud evolution time and why below-cloud precipitation is virtually masked from detection at the TOA. Although cloud water tends to obscure BT-RR relationships, it does so in a differential manner with respect to frequency, suggesting that the overall impact of cloud water is not necessarily debilitating to precipitation retrieval schemes. Furthermore, it is shown how a ?surface? of ?probability? can be defined, which contains an optimal time-dependent BT-RR function associated with an evolving cloud at a given frequency and removes ambiguities within the BT-RR functions at the critical retrieval frequencies. The influence of a land surface having varying emissivity characteristics is also examined in the context of an evolving cloud to show how the time-dependent cloud microphysics modulates the sign and magnitude of brightness temperature differences between various frequencies. Model results are assessed in conjunction with a Nimbus-7 SMMR case study of precipitation within an intense tropical Pacific storm. It is concluded that in order to obtain a realistic estimation and distribution of rainrates, the effects of cloud liquid water content must be considered.