Satellite derived snow cover area (SCA) is a critical parameter in snowmelt modelling, and is used in numerous hydrological and climatological studies. A major limitation of current SCA modelling when using optical satellite sensors is mapping snow in forested areas. The most ideal case for mapping snow in dense forested areas is to have landcover data indicating the location of forested regions and then use separate classification criteria for forested and non-forested areas. This study investigates the MODIS (Moderate Resolution Imaging Spectroradiometer) snow-mapping algorithm "Snowmap" and its ability to map snow in the Northern Boreal Forest of Manitoba. Landcover data enhanced snow-mapping algorithms were developed and compared with MODIS snow products during the snowmelt period of 2001 and 2002. The use of Normalized Difference Snow Index (ND SI) and Normalized Difference Vegetation Index (NDVI) values in the MODIS Snowmap algorithm to detect snow in forested areas was proven successful in this study. Landcover based algorithms and the MODIS algorithm both mapped similar amounts of SCA during the melt period in each study year. The algorithm-derived SCA data for both years showed an exponential correlation with accumulated degree-days. In addition, because the variation in SCA provides an indication of the amount and rate of runoff produced by snowmelt in a watershed, snowmelt runoff computations using the derived SCA data at a daily time step were conducted in two watersheds located in the study area. A quick verification with streamfiow data indicated the inclusion of SCA data in runoff computation could yield better runoff estimates than the exclusion of SCA data. Snow depletion behavior at different elevation ranges in the study area was also investigated. Results show that snow located at higher elevation ranges melt faster compared to snow located at lower elevation ranges. This study also provides a relationship between SCA and accumulated ranges, degree-days at different elevation.