Summary
Fan-Beam and Pencil-Beam Radar Scaterrometers on many satellites in polar orbits, e.g. ERS-1/2, ADEOS-1/2 and QuikSCAT, have been designed primarily for estimation of vector winds over the global oceans. However, these scatterometers, due to their reasonably high resolutions and precise backscatter measurements in different polarizations, have emerged as powerful tools to map and study the ice in the polar region. Dual polarization 0'0( HH and VV) measurements provided by these scatterometers have been exploited to successfully delineate the presence of sea ice. Algorithms and procedures have been developed to locate the boundary between open sea and sea ice regions i.e. the sea ice edge, for flagging the ice corrupted oceanic wind retrievals as well as for providing sea ice advisories for navigation. In the past, India launched its first passive microwave radiometer in polar orbit i.e. MSMR onboard Oceansat-l during 1999. MSMR data have been used extensively by us for the study of spatio-temporal variability of sea ice over the Antarctic region and to examine its role in climate change. Currently, India is planning to launcha Ku-band Pencil-BeamS catterometeor nboardO ceansat-2 in near future, coinciding with the IPY (2007-2008) period. To give boost to applications of Oceansat-2S catterometeirn polar ice studies,w e have undertakenth e developmento f algorithms for delineation and study of sea ice based on QuikSCA T data. In this paper,w e havem adeu se of the dual polarizationm easurementsfr om QuikSCAT by using the polarizationd ifference( O'OVY-O '~H) and the polarizationr atio (0'0V V / 0'0H H) in a synergistic manner to develop a new physically based algorithm for unambiguous sea ice detection over the Southern Ocean around Antarctica. The resulting sea ice maps have been used to locate the seai ce edgew ith high accuracy,a s evidencedb y comparisonw ith simultaneouss eai ce concentrationim agesf rom SSM/! NASA team Algorithm. The seai ce edgel ocated by the present algorithm is found to correspond well to the 10% sea ice concentration contour in the SSM/! images, during both the summer and winter conditions. The PR+PD based sea ice edge algorithm developed by us is found to be robust and, unlike the other QuikSCA T based algorithms published in literature, does not require any a priori information and statistics related to history of sea ice presence in the region. The paper presents the details of the new algorithm and discusses the comparison and validation of sea ice maps generated using QuikSCAT data with those from SSM/I and other QuikSCAT based algorithms. An attempt is also being made to develop techniques to categorize sea ice regions into different classes( FY/ MY, thick/thin, low/high concentrations) based on the use of PR and PD as the discriminating parameters.