Composition and accuracy assessment of snow cover product based on MODIS and AMSR-E data

Abstract: The daily snow cover products of Terra/MOD10A1 and Aqua/MYD10A1 in Qinghai Province from October 1, 2002 to March 31, 2008 were used to composite the fiveday snow products (MOYD_5D), and then combing the MOYD_5D and the fivedays snow water equivalent product of AMSRE/Aqua further composited the new fiveday snow products (MDAE_5D) by userdefined algorithm. Based on snow depth data from climate stations, the accuracies of MOYD_5D, AE_5D and AEMD_5D were validated assessed in this study. The results of this study showed that the snow classification accuracies of MOYD_5D, AE_5D and AEMD_5D were 17.5%, 49.8%, and 23.2%, respectively when the snow depth (SD) was below 3 cm, and they were 46.2%, 55.1% and 56.9% when SD was between 4 cm and 6 cm; and the snow classification accuracies of MOYD_5D, AE_5D and AEMD_5D were 58.6%, 78.5% and 78.6%, respectively when SD was between 7 cm and 9 cm, and they were 66.7%, 82.2% and 84.1% when SD was over 10 cm. This study suggested that the MDAE_5D improved snow classification accuracies, which would be provide useful information for monitoring and evaluating the snow disaster in the pastoral areas.