This research examines how distinct climatic conditions affect the runoff reduction functions of green roofs by comparing performance in Vancouver, BC, Kelowna, BC and Shanghai, P.R. China. To quantify the reduction in runoff volume effectuated by green roofs, both the Soil Conservation Service Curve Number (SCS-CN), crop coefficient method and the Hargreaves-Samani method are applied in calculating the annual water gains and losses of green roofs during a year of average precipitation, using local climate data such as precipitation, evapotranspiration, and temperature. Using a soil water balance model, the research also analyzes the change in soil water content of a typical green roof with a soil depth of 150 mm, and compares the potential irrigation requirements of plants with low versus high water requirements in each of the three cities. The calculation results show that the typical green roof could reduce annual rooftop runoff by 29% in Vancouver, 55% in Shanghai, and 100% in Kelowna. Furthermore, these results illustrate the important role that soil properties, soil depth, and plant selection play in maintaining growth of plants and minimizing green roof irrigation requirements. Les résultats montrent qu'un toit vert typique pourrait réduire la quantité les eaux de ruissellement annuels de 29% à Vancouver, de 55% à Shanghai et de 100% à Kelowna. De plus, il s'avère que les spécificités du toit vert, en particulier, la qualité du sol, l'épaisseur du substrat de croissance et la séléction des plantes jouent un role important pour assurer la bonne croissance des plantes et amoindrir le besoin d'irrigation du toit vert.