A dynamic algorithm for mapping of snow cover using SSMI data

Ongoing information on snow distribution is a crucial issue for Hydro-Quebec in terms of reservoir management late in the winter or in the early months of spring. A dynamic approach that use passive microwave data have been evaluated over province of Quebec (Canada). The method combines a K-mean algorithm and ensemble mapping approach. The first results show that the methods can perform a realistic snow mapping, and unlike standard approaches, can quantify some part of the uncertainty which is an advantage in forecasting.