Abstract: Snow-melt-water is an important source of water supply for rivers in high altitude areas, but the lack of observed data brings great challenges to hydrological simulation in such areas．Precipitation and temperature data from China meteorological forcing dataset (CMFD) and MODIS snow-cover data were used to establish an SRM model for Nianchu River basin．The model simulations of snow-melt-runoff were compared to model driven by meteorological station data．The Nash-Sutcliffe efficiency coefficients of the model driven by measured meteorological data were found to be 0.75 and 0.68 in the calibration (2004-2009) and validation (2010-2015) periods respectively, but these coefficients of the model driven by CMFD data became 0.77 and 0.78, indicating improved runoff-simulation．It is concluded that CMFD could provide input data for hydrological simulation in un-gauged areas．This would be useful for snowmelt-runoff simulation in alpine areas．