Wet snow detection in the south of China by passive microwave remote sensing

Snow mapping is of great importance in meteorological, hydrology and global change researches. In the heavy snow event in 2008 in the south of China, the optic remote sensing fails to map snow cover due to the existence of thick clouds, and the traditional passive microwave snow detection algorithm could only be used for dry snow. Therefore, to map snowcover in the south of China where the snowpacks are mostly wet, shallow snow, AMSR-E brightness temperatures from Jan 1^st to Feb 20^th, 2008 is used to analyze the brightness temperature characteristics in the region of 23-43°N, 102-122°E. Eight land surface types are extracted based on the site-observed snow depth and ground temperature, IMS (Interactive Multi-sensor Snow and Ice Mapping System) and MODIS snowcover. Then, a snow detection decision tree algorithm is established. Comparison of the algorithm-detected snowcover with the IMS product in 2008 and 2011 shows that, the use of 89 GHz channel can improve the snow-detection ability in the southern part of the study region. The performance of the algorithm in the sparse-vegetated region is better than that in the forest-covered region. The total accuracy of the algorithm is about 94%.