Polarimetric Remote Sensing of Land and Snow/Ice Covers with the SpaceborneMicrowave Radiometer WindSat

Abstract

Measurements from spaceborne microwave radiometers, such as the Scanning Multichannel Microwave Radiometer (SMMR), the Special Sensor Microwave/Imager (SSM/I) and the Advanced Microwave Scanning Radiometer (AMSR), are found to be useful in estimating various earth surface geophysical quantities, e.g. soil moisture and vegetation characteristics over land, snow water equivalent for snow covers and sea ice concentration. All these instruments have measured onlythe vertical and horizontal polarization component of the brightness temperature (Tb ). WindSat is the first spaceborne radiometer to provide fully polarimetric measurements of the earthà ¯Ã ¿Ã ½s emission. It waslaunched by US Navy in February 2003. It determines the polarization state of the emission in the form of Stokes vector consisting of four components. The first two components are the typically measured vertical and horizontal TB. WindSat additionally determines the difference between à ?à ±45 Deg linearly polarized (3rd Stokes component) and left and right hand circular polarized radiation (4th Stokes component). The polarimetric radiometry is primarily used to estimate the sea surface wind speed and direction. So far little was known aboutthe information content of the Stokes vector over vegetation, bare soil, snow and sea ice. This thesis explores the polarimetric signal observed by WindSat over land, the Antarctic ice sheet and Arctic sea ice. Over land, it is shown that the polarimetric signal depends on theorientation of surface features such as sand dunes in deserts, and extended structures in agricultural fields. This dependency is validated over the test sites of the Taklamakan desert and the Helongjiang agriculture fields in China using correlative data collected by the Advanced Spaceborne Thermal Emission and Refection Radiometer(ASTER). The ASTER images from thermal infrared sensor of 90 meter resolution are used to identify the vegetation and desert surface structures. Over the Antarctic ice sheet, the findings show that the polarimetric signatures at higher frequencies (37 GHz) depend on snow surface features, such as sastrugi orientation and surface topography, whereas at the lower frequency (10.7 GHz) the signal additionallydepend on snow properties such as grain size and density.This dependency is validated using the results from previously made scatterometer studies with European Space agency SCATterometer (ESCAT) and NASA SCATterometer (NSCAT), demonstrating the potential of polarimetric radiometers to estimate scattering and emission properties from the single instrument. Finally, the analysis of one year of the weekly averaged 3rd and 4th Stokes components of Arctic sea ice shows the highest anisotropic signal during a week of early summer (June 21-27). This was interpreted as a result of melting of the overlying snow due to which the penetration depth decreases making higher frequencies (37 GHz) sensitive to surface structure.