Impact of Air Temperature Distributed Calculation in Glacier Mass Balance Modeling

Thursday, 18 December 2014
Giancarlo Dalla Fontana1, Luca Carturan1 and Federico Cazorzi2, (1)TESAF, Department of Land, Environment, Agriculture and Forestry, University of Padova, Legnaro (PD), Italy, (2)University of Udine, Department of Agriculture and Environmental Sciences, Udine, Italy
Distributed models of snow and ice mass balance enable a better understanding of processes involved in glacier hydrology and the prediction of glacier runoff under possible future climatic scenarios. The so-called ‘Enhanced Temperature-Index’ (ETI) melt models are a good compromise between model simplicity, parsimony of input data, and the capability to account for dominant processes in snow and ice mass balance. Accurate spatial calculation of temperature input data is crucial, given the key role of air temperature in modeling ablation and accumulation processes, further emphasized in ETI models. Compared to ambient conditions, lower temperatures (the so-called glacier cooling effect), and temperature variability (the so-called glacier damping effect) generally occur over glaciers, complicating the extrapolation from off-glacier weather stations.

A comprehensive dataset of mass balance measurements and high-altitude meteorological observations was collected on La Mare and Careser glaciers (Ortles-Cevedale, Italian Alps) in 2010 and 2011. This dataset was used to analyze the air temperature distribution and wind regime over the glaciers, and to evaluate the impact of different calculation methods proposed in the literature for calculating on-glacier temperatures from off-glacier data. A general-purpose ETI model (EISModel - Energy Index Snow-and-ice Model) was used for simulating snow and ice accumulation and melt processes.

Results indicate that i) none of the existing methods fully accounts for the actual temperature distribution over glaciers, ii) even small deviations in air temperature calculations strongly impact the simulations, and iii) there is an important positive feedback related to glacier shrinking and disintegration. Among the tested methods, the more physically-based procedure of Greuell and Bohm (1998) provided the best overall results. Therefore, it was implemented in EISModel for distributed air temperature calculations over glaciers.