Deriving Future Rainfall Depth-Duration-Frequency Curves from Hourly Regional Climate Projections and Simple Scaling in Sicily

Abstract Climate change is increasingly recognized as a major driver of extreme rainfall intensification worldwide. Quantifying its effects on extreme rainfall, expressed through depth-duration-frequency curves (DDFs), is crucial for designing flood protection infrastructures. In this study, we propose a methodology for deriving future DDFs under RCP 4.5 and RCP 8.5 scenarios for the near (2021–2050), middle (2041–2070), and far future (2071–2100). Hourly climate projections at the regional scale from three EURO-CORDEX CMIP5 models are used to derive annual maxima series (AMS) at durations of 1, 3, 6, 12, and 24 h for historical and future periods, enabling the computation of change factors by comparing the two. A simple scaling assumption is used for the derivation of DDFs. These change factors are applied to observed precipitation averages and the moments of the observed dimensionless rainfall depth series, which are then used for parameter estimation of AMS probability distributions according to the method of moments. The approach is tested at four locations in Sicily spanning distinct climatic subregions. Results indicate an overall increase in future DDFs, particularly under RCP 4.5 and in the far future, with changes ranging from approximately 30–70% relative to current conditions, despite significant climate model uncertainty implying variations of up to 80 mm across models. These increases imply that a 50-year DDF in the future climate corresponds to a return period exceeding 100 years in the control period, reaching up to 300 years in some scenarios. Despite the uncertainty, these findings further underscore the importance of integrating future climate scenarios into flood protection design.