Enhanced snow cover mapping using combined optical and passive microwave data

Abstract

Snow cover is an important variable for climate and hydrologic models due to its effects on energy and moisture budgets. Over the past several decades both optical and passive microwave satellite data have been utilized for snow mapping at the regional to global scale. Snow mapping using optical data is based primarily on the magnitude of the surface reflectance, and in some cases on specific spectral signatures, while microwave data can be used to identify snow cover because the microwave energy emitted by the underlying soil is scattered by the snow grains resulting in a sharp decrease in brightness temperature and a characteristic negative spectral gradient. We present results which describe the respective advantages and disadvantages of these two types of satellite data for snow cover mapping and demonstrate how a multi-sensor product is optimal. Passive microwave and optical data sets for the Northern Hemisphere indicate similar patterns of inter-annual variability, although annual maximum snow extents and monthly variability derived from the optical data consistently exceed those provided by the microwave data. Because of the significant differences between snow products derived from passive microwave and optical data, especially during early winter, our current efforts to produce climate data records for snow cover focus on methods which combine the advantages of both data types. For the period 1978 to 2004 we combine data from the manually generated NOAA weekly snow charts, produced primarily from AVHRR and GOES data, with passive microwave data from the SMMR and SSM/I brightness temperature record. For the current and future time period we blend NASA EOS MODIS and AMSR-E data sets. Prototype snow cover maps from AMSR-E compare well with maps derived from SSM/I. Our current EASE-Grid blended product incorporates MODIS data from the Climate Modelers Grid (CMG) at approximately 5 km (0.05 deg.) with microwave-derived SWE at 25 km. resulting in a blended product that includes percent snow cover in the larger grid cell whenever the microwave snow water equivalent (SWE) signal is absent