Snow cover mapping and snowmelt runoff modelling studies can be carried out using satellite data input. Modelling snowmelt runoff with remote sensing inputs is particularly useful in Himalayan basins, which are inaccessible for detailed ground surveys and lack climatological and hydrological data networks. The Snowmelt Runoff Model (SRM) is one such model in which variables like temperature, precipitation and snow covered area are considered along with some externally derived parameters like temperature lapse rate, degree-day factor etc. The SRM was taken-up for implementation in the Beas basin, in the western Himalayas, under the Indo-German Technical Cooperation Programme. The Beas basin up to Thalot (5144 km2) and Parbati River up to Pulga dam site (1154km2) were chosen for the studies. Landsat-MSS data for the runoff seasons of 1986 and 1987 were digitally analysed using sophisticated interpretation techniques. Tne areal extent of the snow cover was evaluate for each elevation zone. Tne corresponding depletion curves could be derived and the values for the daily snow coverage extracted. This information along with temperature, precipitation, degree-day factor, temperature lapse rate and runoff coefficients was entered into the model, which runs on a personal computer. Simulation studies were carried out to get a good fit between the simulated discharges at Thalot and Pulga dam site, and the actual discharges as measured by user departments. An experimental forecast of the snowmelt runoff every 15 days from March to July was given in 1991. This was the first time such real-time forests were made. In 1992, as an improvement over 1991, 7-day predictions were made for Beas- and Parbati. While the ground data transfer from the sites to NRSA needs to be improved, this will help in improving the accuracy of the forecasts. The model parameters also need to be adjusted based on experience every year. The experience gained over these years leads us to extend this study to other basins.