The central theme of this study revolves around the impact of increased rainfall due to climate change on runoff in small urban watersheds. Is there a connection between current rainfall and anticipated runoff? Will peak flows and runoff volumes respond uniformly to this increase? Will runoff vary based on region, frequency, or duration of rainfall? Which watershed parameters will be more sensitive to these climate changes? We are attempting to address all these questions.

To achieve this, we developed an innovative methodology. It relies on modeling, using the SWMM software, of synthetic urban watersheds subjected to precipitation data from various meteorological stations, with different durations and frequencies. In total, 4,860 watersheds with diverse characteristics were assessed using 30 different time series of rainfall, durations, and frequencies, across 12 cities considering both current and future climates. This resulted in a total of 3,499,200 SWMM simulations that were generated and analyzed.

The analysis of precipitation parameters revealed a peak flow increase surpassing the rainfall scaling factor, while the increase in runoff volume exhibited a median variation equivalent to the rainfall scaling factor of 18%. Meteorological stations in southern Quebec (orange dot in figure 1), with long return periods and durations, displayed greater variability in results concerning runoff volume.

Figure 1 Median variation in peak flow (P) and runoff volume (V) between historical and increased rainfall to account for climate change, according to the meteorological station.

A sensitivity analysis (ANOVA) emphasized the significant influence of impermeability, area, and slope on peak flow. Moreover, it indicated that impermeability was the most crucial parameter for runoff volume across all stations, with some exceptions where soil type played a more predominant role.