Comparison of Continuous and Quantile-Based Downscaling Approaches to Evaluate the Climate Change Impacts on Characteristics of Extreme Rainfall

Extreme precipitation events can cause flooding leading to severe socioeconomic and environmental impacts. Drawing on General Climate Model (GCMs) projections, an assessment of the potential impacts of climate change on the characteristics of extreme precipitation events was undertaken. Statistical downscaling approaches, along with temporal disaggregation techniques, were implemented to estimate continuous precipitation series at fine spatiotemporal scales, allowing extreme precipitation events to be extracted. Quantile-based downscaling approaches have recently been applied to estimate future extreme precipitation quantiles through empirical relationships between GCM and observed precipitation quantiles. It remains unclear how these two approaches differ from one another in terms of their projections. In this study, two quantile- and one continuous-based approaches were employed to estimate the future extreme precipitation quantiles in Montreal, Canada. These downscaling approaches provided an acceptable performance during the historical period; however, their estimates of the sign and magnitude of future extreme precipitation quantile changes differed considerably.