ESTIMATING SPATIAL AND TEMPORAL SNOW DISTRIBUTION USING NUMERICAL MODEL AND SATELLITE REMOTE SENSING IN THE WESTERN HINDUKUSH-HIMALAYA REGION

抄録

 In the Hindukush-Himalaya (HKH) region, most of the population are largely dependent on snowmelt and glacier meltwaters. Snowmelt and glacier meltwaters are considered to be a key source of water resources. We evaluated the characteristics of snowfall and snowmelt in the HKH region, which elevation ranges from 1593 m to 5694 m. Moreover, we estimated the spatial and temporal Snow Water Equivalent (SWE) distribution combining remote sensing and snow model. For this purpose, we optimized the precipitation gradient (PG) for snowfall estimation and degree-day factor (DDF) for snowmelt distribution by the combination between MODIS and simulated snow cover area (SCA). Temperature lapse rate was calculated using the observation data, recorded at different altitudes and value of -9.0 ℃ Km^-1 was obtained. As a result, we found the DDF value is 3 (mm ℃^-1 day^-1) for the low elevations (below 3000 m) and 3 to 5 (mm ℃^-1 day^-1) for the high elevations (above 3000 m). In addition, the precipitation distribution change with elevations the PG is 0 (m^-1) for high altitudes (above 4500 m) and 0.001 (m^-1) for low altitudes (below 4500 m). We estimated SCA and SWE through snow model using the two optimized parameters. MODIS and simulated SCA were examined and resulted in a determination coefficient of 0.98 and 0.96 for the years 2012-2013, 2013-2014 and 2014-2015, respectively. SWE accumulated from late October and continues until March. The SWE peaks in March and later on starts decreasing. The results justify that the snow model is applicable for water resources management under data-scarce and complex watersheds such as the HKH region.