Estimation of the snowpack water equivalent via remote sensing in the Alps

A large mass of water is stored in snowpack on mountains during winter and it is very important to estimate its volume before the melting season begins. The authors describe a way to estimate snowpack water equivalent combining satellite images and point observations. Snow covered area extension is obtained by remote sensing techniques; snow gauges give measurements of snow thickness and density at monitoring stations. Soft classifications are performed in order to have careful infonnation about snow presence on the ground. In fact soft classification of satellite images supplies pixels membership degrees to snow class that are good evaluations of the snow presence inside pixels. Simultaneous local measures of snow thickness are then used to calibrate a statistical model describing snow thickness as a function of terrain altitude and aspect. A calibrated model allows estimation of the snow thickness pixel by pixel and so estimation of pixel by pixel snow volume too. Snow density information is added to transform snow volume into water volume. The described method is applied to estimate water equivalent on the Adda river basin for April 1996