A new method was developed to estimate the mass balance in unsampled areas from existing
datasets. Three years of mass-balance data from two glaciers in the central Italian Alps were used to
develop and test a multiple-regression method based exclusively on a 10 m resolution digital terrain
model. The introduction of a relative elevation attribute, which expresses the degree of wind exposure of
the gridcells, notably increased the amount of explainable variance in winter balance with respect to
altitude itself. The summer balance is highly correlated with elevation, but, in order to obtain reliable
extrapolations, the clear-sky shortwave radiation and the diurnal cloud-cover cycle had to be taken into
account. The net annual mass balance on a glacier system comprising the two monitored glaciers was
calculated by applying both a single regression of winter and summer balance with altitude and the new
regression method. The consistency of results was assessed against measured net balances and snow-cover maps drawn in the ablation season. The results of the new method were in close agreement with
observations and proved to be less sensitive to the spatial representation of the sampled areas
