Point‐scale energy and mass balance snowpack simulations in the upper Karasu basin, Turkey

Abstract Since snowmelt runoff is important in the mountainous parts of the world, substantial efforts have been made to develop snowmelt models with many different levels of complexity to simulate the processes at the ground (soil–vegetation), within the snow, and at the interface with the atmosphere. Snow modifies the exchange of energy between the land surface and atmosphere and significantly affects the distribution of heating in the atmosphere by changing the surface albedo and regulating turbulent heat and momentum fluxes at the surface. Thus, for computing the amount of melt, the only strictly correct way is using an energy budget. A two‐layer point model (SNOBAL) was applied to calculate the energy and mass balance of snowmelt in the upper Karasu basin, in eastern Turkey, during the 2002–04 snow seasons. The data on snow and climate were provided from automated snow and meteorological stations installed and upgraded to collect high‐quality time series data of snow and meteorological variables, such as snow water equivalent, snow depth, precipitation and radiation, with automated data transfer. A number of analyses of snowpack energy and mass balance were carried out to understand the key processes that have major impacts on the snow simulation. Each form of energy transfer was evaluated during snow accumulation and ablation periods using a 2 h computational time step. The model results are appraised with respect both to temporal distribution (the model application for three consecutive snow seasons at one site) and to areal evaluation (the model application to three different sites for one season). The model performance is evaluated by comparing the results with observed snow water equivalent, snow depth and lysimeter yield. Copyright © 2006 John Wiley & Sons, Ltd.