Statistical analysis of the snow cover variability in a subalpine watershed:: Assessing the role of topography and forest, interactions

This paper analyses the way in which the complex relationships among topography, climate, and forest canopy influence the spatiotemporal variability of accumulation and melting of snowpack in a subalpine valley. We used weekly to monthly snow course data and a numerical model (COUP) to estimate the snow water equivalent (SWE) at 16 sites distributed in the Alptal valley (Central Switzerland) from 1984 to 2004. From these SWE series, we derived indicators that represent the annual accumulation, the most intense annual melting events and the length of the period with snow cover. Generalized mixed linear models (GMLMs) were used to relate these indices to attitude and exposure to solar radiation, taking into account the interactions of both variables with the presence or absence of a forest cover. Regression analyses were carried out for the 20 analysed years and for specific years with particular climatic conditions during the accumulation or melting periods. On the basis of these results, we conclude that a small number of predictors are able to explain a large proportion of the observed SWE and snowmelt variability. The statistical analysis confirmed the role of a forest canopy in reducing snow accumulation. The reduction of the magnitude of snowpack below forest canopy will impact also in a lower amount of snow melted during the most intense thawing events. The role of forest in terms of the length of the snow cover season changes noticeably with attitude and exposure. (c) 2007 Elsevier B.V. All rights reserved.