Comparison of satellite snow cover estimates from NOAA snow charts and a SMMR show cover algorithm

This study compares two satellite-derived snow cover estimates with ground-based monthly snow cover measurements from the former Soviet Union. The satellite data are from NOAA/NESDIS snow charts, computed from reflected visible radiation, and an algorithm developed by Chang et al. (1987) which determines snow cover as a function of two Scanning Multichannel Microwave Radiometer (SMMR) microwave frequencies (18 and 37 Ghz).Results show the data agree relatively well during winter and summer months, but less well in spring and fall. It is argued that the SMMR algorithm underestimates snow cover conditions during spring and overestimates in fall as brightness temperatures are strongly affected by the presence of liquid water and frozen ground respectively. In addition, the visible data composites show disparities caused by extensive cloud cover, a rapidly moving snow boundary, and the use of monthly summary snow charts.