An enhanced temperature-index glacier melt model including the shortwave radiation balance: development and testing for Haut Glacier d’Arolla, Switzerland

Abstract An enhanced temperature-index glacier melt model, incorporating incoming shortwave radiation and albedo, is presented. The model is an attempt to combine the high temporal resolution and accuracy of physically based melt models with the lower data requirements and computational simplicity of empirical melt models, represented by the ‘degree-day’ method and its variants. The model is run with both measured and modelled radiation data, to test its applicability to glaciers with differing data availability. Five automatic weather stations were established on Haut Glacier d’Arolla, Switzerland, between May and September 2001. Reference surface melt rates were calculated using a physically based energy-balance melt model. The performance of the enhanced temperature-index model was tested at each of the four validation stations by comparing predicted hourly melt rates with reference melt rates. Predictions made with three other temperature-index models were evaluated in the same way for comparison. The enhanced temperature-index model offers significant improvements over the other temperature-index models, and accounts for 90–95% of the variation in the reference melt rate. The improvement is lower, but still significant, when the model is forced by modelled shortwave radiation data, thus offering a better alternative to existing models that require only temperature data input.