Snowmelt detection in Greenland ice sheet based on AMSR2 89GHz

Abstract

The snowmelt on the surface of the Greenland ice sheet can show the climate change in the Arctic. In order to obtain high spatial resolution and high precision results for snowmelt detection on the polar ice sheet surface, a method of snowmelt detection based on an Advanced Microwave Scanning Radiometer-2 (AMSR-2) 89GHz channel is presented. Firstly, the stable relationship between the polarization ratio of 89GHz data and 36GHz data under clear and cloudless weather conditions is used to screen out the 89GHz data that is not affected by external factors such as clouds and water vapor. Then, the Nobuyuki Otsu (Otsu) method is used to obtain the freeze-thaw threshold, and the snowmelt detection results of the ice sheet surface are obtained from the unaffected data by cloud conditions. The areas affected by cloud cover are replaced by the snowmelt detection results of the ice sheet surface from the gradient ratio (GR) model based on 19GHz and 37GHz horizontal polarization data. Finally, we get the snowmelt detection results of the Greenland ice sheet surface. To verify the feasibility and rationality of the proposed method, the results are compared with those of the cross-polarized gradient ratio algorithm (XPGR) and the simple physical model algorithm. The snowmelt area obtained by our method is between the XPGR and the simple physical model, and the snowmelt trend of our method and the simple physical model algorithm is closer. Then the results of the three algorithms are compared and analyzed with the temperature data from the automatic weather stations showing that the results obtained by our method are closer to the measurement data.