Urban winter hydrology has garnered very little attention due to the general notion that high intensity rainfalls are the major flood-generating events in urban areas. As a result, few efforts have been made to research urban snow and its melt characteristics. This study investigated the characteristics of urban snow that differentiates it from rural snow, and the impact of incorporating these characteristics in an urban snowmelt model. A field study was conducted from fall of 2001 to spring of 2002 at the University of Calgary campus. Data collected includes snow depth and density, soil moisture, soil temperature, snow albedo, net radiation, snow evaporation, and surface temperature. Snow cover was classified into several types;​ snow piles, snow on road shoulders, snow on sidewalk edges, and snow in open areas. This resulted in the development of four separate functions for the changing snow albedo values. Shortwave radiation was found to be the main source of energy for urban snow, and as a consequence, the albedo of urban snow is a very important factor in urban snowmelt modelling. In addition, urban elements such as vehicle traffic and buildings can influence the energy balance of the snowpack. A study of the frozen ground conditions reveals that antecedent soil moisture conditions had very little impact on frozen ground, and thus frozen ground acts as a near impervious area. In the modelling component of this study, urban snowmelt was modeled using the energy balance method with hourly time steps and the incorporation of snow redistribution, and hence the simulation of snow piles. Three simulated tests of varying conditions revealed that peak volume, time to peak and runoff period differs for areas with snow piles versus a uniform urban snow cover. Simulation of rain-on-snow events revealed a sharp increase in runoff peak volume. Hence, under the adverse condition of intense snowmelt, frozen ground, and rainfall, flooding in urban areas can easily occur. Improved flood forecasting for urban catchments in cold regions can only be achieved with accurate modelling of urban winter runoff that involves the energy balance method, incorporating snow redistribution and urban snow cover characteristics, and using small time steps.