Comparison of the Theoretical Clausius-Clapeyron Scaling and IDF_CC Tool for Updating Intensity-Duration-Frequency Curves under Changing Climatic Conditions in Canada

摘要

Changes in climatic conditions are expected to affect the hydrological cycle with intensification of extreme rainfall events caused by the disturbance in temperature and other atmospheric variables linked to precipitation. Extreme rainfall change will affect the intensity-duration-frequency (IDF) relationship, used in the design, maintenance, and operation of water infrastructure in Canada. This study presents a comparative analysis of the results from two IDF updating methods: (1)the IDF_CC tool, which applies an equidistance quantile-matching precipitation downscaling algorithm, and (2)the Clausius-Clapeyron (C-C) precipitation-temperature relationship, used with a proposed constant temperature scaling rate. The analyses were conducted using 358 selected Environment Canada hydro-meteorological stations from the IDF_CC tool database with record length longer than 20years. Results for the future period (2061-2100), are based on the multimodel ensemble of 24 global climate models (GCMs). The difference in (1)projected precipitation and (2)uncertainty range for both IDF updating methods are presented and analyzed. The uncertainty range is defined in this work as the difference between IDF relationships obtained using various GCMs. The C-C temperature scaling method resulted, overall, in higher extreme precipitation projections than the IDF_CC tool for the stations located in the Canadian Prairies (i.e.,the provinces of Alberta, Saskatchewan, and Manitoba). Stations located at the east and west coasts of Canada show smaller difference in the projected extremes. A similar pattern is observed for the multimodel ensemble median and the all individual GCMs. The difference in projected uncertainty range for both methods was analyzed for the multimodel ensemble and for representative concentration pathway (RCP) 2.6, RCP 4.5, and RCP 8.5 emission scenarios. The C-C scaling shows a smaller uncertainty range for RCP 2.6 and RCP 4.5, and the IDF_CC tool shows a smaller uncertainty range for the RCP 8.5 scenario (especially for stations located in the Canadian Prairies). The difference in percent uncertainty ranges from -75% to about 100%, considering all stations across Canada. Both methods show comparable uncertainty range in the future. One significant conclusion is that the high level of uncertainty cannot be avoided, regardless of the method selected for updating IDF curves for future conditions. Use of the precipitation-based IDF_CC tool is recommended because of serious issues in using a constant scaling rate with C-C temperature scaling.