Assessing Climatic Drivers of Spring Mean and Annual Maximum Flows in Western Canadian River Basins

Water

Paulin Coulibaly¹  Barrie Bonsal²  Colin Johnson³  Yonas B. Dibike³  Rajesh R. Shrestha³ 

[1] Civil Engineering Department and the School of Earth Environment and Society, McMaster University, Hamilton, ON L8S 4K1, Canada;Environment and Climate Change Canada, Watershed Hydrology and Ecology Research Division, National Hydrology Research Centre, Saskatoon, SK S7N 3H5, Canada;Environment and Climate Change Canada, Watershed Hydrology and Ecology Research Division, University of Victoria, 2472 Arbutus Rd., Victoria, BC V8N 1V8, Canada;

关键词: peak flows;    multiple linear regression;    predictor;    predictand;    snow water equivalent;    annual maximum flow;   

DOI  :  10.3390/w13121617

来源: DOAJ

【 摘 要 】

Flows originating from cold and mountainous watersheds are highly dependent on temperature and precipitation patterns, and the resulting snow accumulation and melt conditions, affecting the magnitude and timing of annual peak flows. This study applied a multiple linear regression (MLR) modelling framework to investigate spatial variations and relative importance of hydroclimatic drivers of annual maximum flows (AMF) and mean spring flows (MAMJflow) in 25 river basins across western Canada. The results show that basin average maximum snow water equivalent (SWEmax), April 1st SWE and spring precipitation (MAMJprc) are the most important predictors of both AMF and MAMJflow, with the proportion of explained variance averaging 51.7%, 44.0% and 33.5%, respectively. The MLR models’ abilities to project future changes in AMF and MAMJflow in response to changes to the hydroclimatic controls are also examined using the Canadian Regional Climate Model (CanRCM4) output for RCP 4.5 and RCP8.5 scenarios. The results show considerable spatial variations depending on individual watershed characteristics with projected changes in AMF ranging from −69% to +126% and those of MAMJflow ranging from −48% to +81% by the end of this century. In general, the study demonstrates that the MLR framework is a useful approach for assessing the spatial variation in hydroclimatic controls of annual maximum and mean spring flows in the western Canadian river basins. However, there is a need to exercise caution in applying MLR models for projecting changes in future flows, especially for regulated basins.

【 授权许可】

Unknown