The urbanization changes a watershed's response to precipitation. The most common effects include the reduced infiltration and the decreased travel time, which significantly increase runoff and peak discharges. This study attempts to analyze the impact of land use on runoff in the Toronto Region. In this report, the focus is on two aspects: (1) generating watershed boundaries using digital elevation model (DEM) data with the help of HEC-HMS, and (2) calculating runoff in the study area using the United States Soil Conservation Service (SCS) curve number method for the early 1990s and 2003. The study is based on the watershed boundaries generated from DEM data with 10 m resolution. Because of the flat surface in the south of the Toronto Region, the areas of the watersheds generated in this study are slightly less than the real ones, but the difference is within acceptable range. As a crucial parameter in the SCS method for runoff calculation, curve number is difficult to obtain. In this project, curve numbers for each watershed are calculated by using the land cover and soil data of the early 1990s and 2003 respectively. According to the theory, the higher the curve number is, the higher the potential of runoff generation in the area is. Unlike what is expected, the curve numbers have changed little from the early 1990s to 2003, although the impervious surface has increased. This is because the variation of the land cover is too little to increase the curve numbers. The curve number for each watershed is a weighted one. If the area of a specific lot which has changed from pervious to impervious surface is small, the weight variation of such area is also small. The other reason for the little change of curve numbers is that the land cover data sets of the early 1990s and 2003 used different classification systems. To eliminate the discrepancy resulting from those land cover classification systems, the curve numbers in 2003 were calculated by referring both classification schemes of the early 1990s and 2003. Because the land cover classification in this study is reasonable, the curve number of the Toronto Region in 2003, 80.4 can be used in the future research.