Author(s): Lijian Shi; Peng Lu; Bin Cheng; Zhijun Li; Qimao Wang; Yunfei Lu; Xiaomin Ye
Keywords: No Keywords
Abstract: Sea ice concentration (SIC) is one of the most important parameters describing the sea ice characteristics over polar region and a crucial input parameter for various regional numerical weather prediction (NWP) and climate models. Satellite-borne microwave radiometer is the most effective and consistent tool to measure the sea ice concentration over Polar Regions. SIC retrieval method with the brightness temperatures (Tb) from the Chinese marine satellite Hai Yang 2 (HY-2) scanning radiometer has been established. The HY-2 retrieval SIC in the Arctic is in good agreement with two operational products from NSIDC and University of Bremen. In this study, the validation of above three SIC products are presented. Aerial photography conducted during CHINARE-2012 was also employed to produce Arctic sea ice concentration and then compared with the retrieval of satellite remote sensing after spatial interpolation. HY2/NT is 16% higher than the results of aerial photography at the compact ice regions in the central Arctic where the helicopter flights during CHINARE-2012 was carried out. In contrast, the SSMIS/NT data underestimate the in-situ ice concentrations by 17%. While the SSMIS/ASI data show a nice agreement with the result of in-situ photography, possibly because of its higher spatial resolution of 6.25 km than other satellite data. Six SAR images were finally used for the validation of SIC retrievals HY2/NT, SSMIS/NT and SSMIS/ASI. The RMS errors between HY2/NT and SAR, ranging from 22.42% to 33.15%, are larger than that of SSMIS/NT and SSMIS/ASI which are in the range of 13-34% with an average of 22%. The RMS errors in ice edge areas are higher than that of the interior areas. If pixels at ice edges are removed, the RMS between SAR derived ice concentration and three microwave retrievals falls below 10%.