Comparison of SSMIS and AMSR2 sea ice concentration in the Amundsen Sea, West Antarctica, with KOMPSAT-5 SAR Images
- Comparison of SSMIS and AMSR2 sea ice concentration in the Amundsen Sea, West Antarctica, with KOMPSAT-5 SAR Images
- Han, Hyangsun
- Issue Date
- Han, Hyangsun and Hyun-cheol Kim. 2016. Comparison of SSMIS and AMSR2 sea ice concentration in the Amundsen Sea, West Antarctica, with KOMPSAT-5 SAR Images. XXXIV SCAR Biennial Meetings including 2016 Open Science Conference, Kuala Lumpur. 2016.08.20-10.30.
- Sea ice in the Arctic and Antarctic Ocean plays a role of an indicator of climate changes. Since 1970s, passive microwave sensors have been widely used to study the Arctic and Antarctic sea ice because they have observed the whole polar oceans at least once in every two days, regardless of weather conditions and sun altitudes. The Special Sensor Microwave Imager/Sounder (SSMIS) and the Advanced Microwave Scanning Radiometer 2 (AMSR2), the successor to the Special Sensor Microwave Imager (SSM/I) and the Advanced Microwave Scanning Radiometer - Earth Observing System (AMSR-E), respectively, are the representative passive microwave sensors, producing sea ice concentration since 2008 and 2012, respectively. As sea ice concentration products derived from SSMIS and AMSR2 are universally used in various research fields, it is very important to evaluate accuracies of the products, especially in sea ice melting season. Synthetic Aperture Radar (SAR) is very useful tool for observing sea ice because it is all-weather and day-and-night imaging system. In this study, SSMIS and AMSR2 sea ice concentration was evaluated by using Korea Multi-Purpose SATellte-5 (KOMPSAT-5) SAR, South Korea’s first satellite SAR, images obtained in the Amundsen Sea in the West Antarctica during January-February 2016. All KOMPSAT-5 SAR images were obtained in HH-polarization at Enhanced Wide (EW) swath mode with spatial resolution of 6.25 m. Sea ice, icebergs and open water were classified from the SAR images based on Random Forest, a rule-based machine learning approach, in which texture features derived from gray level co-occurrence matrix were used as input variables. Then sea ice concentration was computed from the SAR images and compared with the passive microwave sea ice concentration. The comparisons showed that both SSMIS and AMSR2 sea ice concentrations were underestimated. This would be probably caused by melting of ice surface which is regarded as open water in sea ice algorithms of the passive microwave sensors. Particularly, SSMIS and AMSR2 sea ice concentration were very low over icebergs due to different microwave radiation characteristics between sea ice and glacial ice.
- Conference Name
- XXXIV SCAR Biennial Meetings including 2016 Open Science Conference
- Conference Place
- Kuala Lumpur
- Conference Date
- Files in This Item
- Can archive pre-print and post-print or publisher's version/PDF
Can archive post-print (ie final draft post-refereeing) or publisher's version/PDF
Can archive pre-print (ie pre-refereeing)
Archiving not formally supported
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.