Estimating Snow and Glacier Melt in a Himalayan Watershed Using an Energy Balance Snow and Glacier Melt Model

Abstract:

This study enhances an energy balance snowmelt model (Utah Energy Balance, UEB) to include the capability to quantify glacier melt. To account for clean and debris covered glaciers, substrate albedo and glacier outlines determined from remote sensing, are taken as inputs. The model uses the surface energy balance to compute the melting of seasonal snow and glacier substrate once the seasonal snow has melted. In this application the model was run over a 360 km² glacierized watershed, Langtang Khola, in the Nepal Himalaya for a 10-year simulation period starting in water year 2003. The model was run on a distributed mesh of grid cells providing the capability to quantify both timing and spatial variability in snow and glacier melt. The distributed UEB melt model has a relatively high data demand, while the Hindu-Kush Himalayan region is a data-scarce region, a limitation that affects most water resources impact studies in this region. In this study, we determined model inputs from the Modern Era Retrospective-Analysis for Research and Applications (MERRA) and Southern Asia Daily Rainfall Estimate (RFE2) data products. The model estimates that roughly 57% of total surface water input is generated from glacier melt, while snowmelt and rain contribute 34% and 9%, respectively over the simulation period. The melt model provided input to the USGS Geospatial Stream Flow Model (GeoSFM) for the computation of streamflow and produced reasonable streamflow simulations at daily scale with some discrepancies, while monthly and annual scale comparisons resulted in better agreement. The result suggests that this approach is of interest for water resources applications where monthly or longer scale streamflow estimates are needed. Mean annual streamflow was positively correlated with the total annual surface water input. However, mean annual streamflow was not correlated with total annual precipitation, highlighting the importance of energy balance melt calculation, in comparison to just using precipitation when considering streamflow availability. Overall, for a 10-year model run, the water equivalent of snow accumulation is 2.46 m compared to 7.13 m of glacier melt over the basin, suggesting a net loss in glacier mass.