Abstract: The relationship between snow melting rate (SMR) and temperature is analyzed by using ground observation data during the winter half year from current November to the next April in years from 1980 to 2007, at 13 weather stations over the Northern Tibetan Plateau. The daily products of snow cover days, snow depth and snowstorms intensity level are offered based on the linear regression model between SMR and the maximum temperature, minimum temperature. This method may be applied to assessing the snowstorm severity on time, and providing quantitative products in meteorology for the grassland snowstorm. In assessments for three winter half years of 2008-2009, 2009-2010 and 2010-2011, the averaged absolute forecast-error of daily snow depth is less than 2cm for 1-15 days; of snow cover days is less than 9d, of which 78.5% errors are not more than 1 day; of the accuracy of snowstorms intensity level is about 84%. So, it has a better forecast function for the snow disaster assessment in the Northern Tibet. The defect is that no consideration of the new snow cover for the later snowfall, because there is no accurately quantitative snowfall forecasting products.