Monitoring Polar Sea Ice Extent Using Ten Years of QuikSCAT Scatterometer Measurements

Polar sea ice extent is an important input to global climate models and is considered to be a sensitive indicator of climate change. Spaceborne scatterometers such as the SeaWinds scatterometer on the QuikSCAT satellite have long been used to monitor sea ice extent due to their relatively low sensitivity to atmospheric effects. The QuikSCAT satellite was launched June 1999 and continued collecting normalized radar cross section (NRCS) measurements of polar sea ice until end of mission in November 2009. Scatterometer vertical and horizontal polarization NRCS measurements, polarization ratio, and temporal variation statistics have proven to be sensitive to the presence of polar sea ice. An algorithm that exploits the information contained within these measured parameters to map sea ice extent is described. The algorithm is comprised of an iterative maximum likelihood classifier that segments sea ice and open ocean pixels. Residual classification errors are reduced through binary image processing techniques. The algorithm supplied operational sea ice extent during the ten year QuikSCAT mission. The resultant sea ice extent dataset is summarized showing Arctic and Antarctic sea ice extent seasonal variability and trends. The sea ice maps are compared with passive microwave radiometer-based sea ice concentration maps as measured by the Special Sensor Microwave/Imager (SSM/I) instruments. The scatterometer sea ice edges correlate with differing passive radiometer ice concentration levels during the seasonal ice advance and retreat phases. The sea ice edge results also show good correlation with sample Radarsat synthetic aperture radar (SAR) imagery.