Canada has experienced a warming trend since the industrial revolution, surpassing the global average by a factor of two, with northern regions witnessing an even more pronounced threefold increase in temperature. This significant and consistent shift in the climate has led to extensive alterations in weather patterns across the country, intensifying the frequency, severity, and duration of climate-related natural disasters including extreme rainfall, which significantly affects the structural integrity and performance of roofs. In this paper, we will present the approach adopted by the National Research Council Canada to obtain the magnitudes and distributions of extreme design rainfall, which will be used to evaluate the performance of climate resilient roofs in Canada. The approach involves an analysis of one minute frequency rainfall data to derive the temporal distribution of extreme 5-min duration rainfall events historically observed around cities in Canada. The magnitudes of future projected rainfall are obtained by temperature-based scaling of historically observed extreme rainfall in accordance with the future projected changes in temperature forecasted for these cities by the Canadian Regional Climate Model – version 4 (CanRCM4). The future projected magnitudes of extreme rainfall are then distributed across the duration of the event making use of the rainfall distribution functions previously obtained for the cities. The process is implemented for three Canadian cities of Montreal, St. Johns, and Calgary and peak rainfall rates of 13.7 mm/min, 8 mm/min, and 4.1 mm/min respectively are obtained for the design of roof systems for their hydraulic capacity. The outcomes of this work will contribute to enhancing the resilience of these roof assemblies in Canada in the face of a changing climate.