A land surface emissivity database for conically scanning microwave sensors

Accurate knowledge of local surface emissivity is required for lower tropospheric microwave remote sensing over land. Ideally, for a stand alone microwave system, accuracies of 0.01 or less are needed to minimize the impact of cloud liquid water on temperature and water vapor retrievals and for improving surface temperature retrievals. Because surface properties may change rapidly, the emissivity database must be frequently updated. Surface emissivity may be well characterized in clear areas using collocated microwave and infrared observations although, in certain areas, terrain and surface type inhomogeneities may be a limiting factor. In cloudy (non-precipitating) conditions one must rely on temporal persistence. The use of such an approach is necessarily limited to areas for which frequency of occurrence of relatively clear measurements is higher than the rate at which surface properties change.

In this work land surface emissivity in the AMSR-E channels is retrieved from combined observations from the AMSR-E, AIRS and MODIS instruments on the EOS/Aqua platform in relatively clear conditions. We examine the temporal variability of retrieved local surface emissivity over selected regions of the globe and provide a preliminary assessment of the usefulness of the product for cloudy microwave retrievals. Since the emissivity retrieval is sensitive to the specification of land surface temperature (LST) we have also investigated the use of different sources for LST (such as those from physical land surface models and analyses), and attempt to characterize these differences both in terms of the respective LSTs, as well as the retrieved emissivities.