Runoff modelling of the glacierized Alpine Upper Salzach basin (Austria): multi-criteria result validation

Abstract

Snow cover and glaciers are the most important long-term forms of water storage and, hence, the main sources of runoff during the ablation period for many alpine headwater basins. We therefore investigated the application of the conceptual, distributed hydrological precipitation runoff evapotranspiration hydrological response unit model (PREVAH) to the alpine glacierized headwater basin of the Upper Salzach (593 km², 5% glacierized) river in Austria. Hourly meteorological data from 17 stations for a 6-year period were available for the calibration and validation of the hourly runoff simulations. Multi-criteria validation included hourly discharge, snow covered area (SCA), and glacier mass balances. SCA maps were generated from optical satellite images for six dates. These maps were compared to simulated maps of SCA to (1) calculate differences in SCA, (2) calculate altitudinal differences, and (3) show the ability to accurately model snow cover on different aspects. The differences between observed and simulated SCA for glacierized areas were between 1 and 9% during June and July, and between 10 and 36% during August and September observations. In general, the model overestimated SCA, which is the result of PREVAH not including redistribution of snow by wind or avalanches. The temporal variability of the simulated mass balance agreed well with observations from surrounding glaciers. Nash-Sutcliffe Efficiency Criteria (R²) of the hourly discharge simulations were between 0·83 and 0·89 with the exception of the extreme summer of 2003 which had an R² of 0·74. Contributions of glacier melt (firn/ice melt) to annual total runoff were between 1 and 4%. Again, the exception was 2003, when glaciers contributed 15% of the annual runoff and 60% to the August runoff alone. Copyright © 2008 John Wiley & Sons, Ltd.