The Development of Microwave Vegetation Indices from WindSat Data

Abstract

The microwave vegetation indices (MVIs), including parameter A ( $\mathbf{MVIs}\_\mathbf{A}$ ) and parameter B ( $\mathbf{MVIs}\_\mathbf{B}$ ), have been recently developed based on advanced microwave scanning radiometer-earth observing system (AMSR-E) measurements. Coriolis/WindSat (WindSat) is a space-borne multifrequency polarimetric microwave radiometer with similar frequencies to the AMSR-E. Unlike the AMSR-E instrument configuration, the WindSat observation angles vary at the different frequencies and range from 49.9° to 55.3°. This variation results in significant uncertainty in deriving MVIs using WindSat data. In this study, we extended our algorithm for deriving MVIs from AMSR-E to that under WindSat sensor configuration by considering the measurements from different observation angles. We will present the theoretical basis for this new algorithm and then compare the two MVIs derived from these two sensors’ data at global and pixel scales, respectively. We found that the MVIs from WindSat data and AMSR-E data share similar global distribution patterns and temporal trends. The $\mathbf{MVIs}\_\mathbf{B}$ at the 6.8- and 10.7-GHz frequency pair [ $\mathbf{MVIs}\_\mathbf{B}$ (6.8,10.7)] from WindSat data is somewhat higher than that from AMSR-E data, whereas the $\mathbf{MVIs}\_\mathbf{B}$ at the 10.7- and 18.7-GHz frequency pair [ $\mathbf{MVIs}\_\mathbf{B}$ (10.7,18.7)] is higher for AMSR-E. The MVIs from both the WindSat data and the AMSR-E data can be affected by satellite overpass times. The extended MVIs are expected to provide possible complementary information and contribute to vegetation monitoring, vegetation water content, biomass and soil moisture retrieval, and global climate change research.