Bridging the NASA Global Precipitation Measurement and Soil Moisture Active Passive Missions: Variability of Microwave Surface Emissivity from both In-situ and Remote Sensing Perspectives

Abstract

The overland precipitation retrievals from satellite passive microwave (PMW) radiometer such as the Global Precipitation Mission (GPM) microwave imager (GMI) suffer from the contamination of observed brightness temperatures from the land surface. Microwave land surface emissivity is the quantitative indicator of the magnitude of the microwave radiance from the land surface. Research has shown that the estimation of PMW emissivity faces challenges because emissivity cannot be directly measured but it is highly heterogeneous and highly variable under the influence of various surface properties such as soil moisture, surface roughness and vegetation. Our research aims to achieve an improved quantitative understanding of the emissivity by studying the relationship between the emissivity and different surface parameters. Surface parameter information can be obtained through in-situ measurements (International Soil Moisture Network stations) and satellite measurements. In particular, the Soil Moisture Active and Passive mission (SMAP) provides global scale soil moisture data. The overarching goal of our work is to incorporate the SMAP soil moisture to improve the precipitation retrieval from GPM in order to bridge two satellite missions on the water cycle by linking remotely sensed precipitation with soil moisture. Our results quantitatively and holistically describe the variation of the emissivity with soil moisture, surface temperature and vegetation at various frequencies / polarization over different type of land surfaces. This sheds light into the processes governing the emission of the land and provides reliable support to the development of a physically based emissivity model. Moreover, while most of the emissivity retrieval approaches apply only to cloud-free scenes, the analyses established in this study can be used to reproduce and predict the emissivity also for the rainy/cloudy conditions when surface parameters such as soil moisture and vegetation are available. Furthermore, the analyses based on in-situ measurements serve as a benchmark for satellite-based models established using the satellite observations, which paves a way toward a global scale dataset of emissivity based on our models.