PNNL

Abstract

Prior research on climate change impacts on flooding has primarily focused on changes in extreme precipitation magnitudes, often neglecting snow processes and spatiotemporal storm patterns, such as hyetograph shapes and areal reduction factors (ARF). This study examines projected changes in extreme water available for runoff (W) events within two snow-dominated basins in the western U.S.: the Yakima River Basin (YRB) in Washington State and the Walker River Basin (WRB) spanning the California-Nevada border. We analyze changes in W magnitudes, mechanisms, hyetograph shapes, and ARFs, considering their compounded impacts on flood risk. Our findings suggest that future climate scenarios predict increased extreme W magnitudes in both basins, steeper hyetographs at higher elevations (despite larger uncertainties), and higher ARF values. These changes are linked to a shift from snowmelt to more rain-on-snow events at higher elevations and more rainfall at lower elevations. Using a single event-based rainfall-runoff model, we estimate flood risk changes based on extreme W magnitudes, hyetograph shapes, ARFs, and their compounded impacts. Our analysis reveals that focusing solely on magnitude changes of extreme W can significantly underestimate future flood risks and uncertainties. Ignoring spatiotemporal patterns can underestimate future flood risk by 63.3% and future flood uncertainty by 18.1% in the WRB. These results underscore the necessity of incorporating spatiotemporal dynamics into future flood risk assessments to provide a more accurate evaluation of potential impacts.