Land Surface Microwave Emissivity: Modeling and Validation to Support Global Precipitation Mission

The land surface emissivity at microwave frequencies is a critical variable to remote sensing and data assimilation with passive microwave observations. Two radiative transfer models, the Community Radiative Transfer Model (CRTM) and the Community Microwave Emission Modeling Platform (CMEM) have been widely adopted for both retrieval and forward modeling applications. But they have not been extensively validated due to the lack of reliable reference data on the global scale. In this study, we devised a new approach to validate both models. The approach is based on the fact that at lower frequencies, the microwave polarization difference index (MPDI) computed from satellite's brightness temperature (Tb) measurements is nearly identical to the same index computed from true emissivity values. This enables us to quantitatively assess the models' simulation of microwave emissivity. With both CRTM and CMEM coupled to the Land Information System, global-scale land surface microwave emissivities were simulated for 4.5 years, and validated with the MPDI-based method. It is found that both models have promises and challenges over various types of land surfaces. Among them, the desert shows the largest errors (200-300%) due to the wrong physical models used, and requires an overhaul in both models. Other snow-free surface types exhibit quantitative differences and it is expected that parameter tuning can improve their performances.