Hydrologist USGS Alaska Science Center, Alaska, United States
Abstract Submission: Characterization of flood-generating mechanisms for extreme floods measured at USGS streamgages informs flood frequency analyses that are critical for design of infrastructure, management of water resources, and ecosystem studies. Minimum extreme flood thresholds consisting of the magnitude of the 1-percent annual exceedance probability flood or the 95th percentile of a measure of relative magnitude known as the Creager coefficient identified 149 annual peak flows in Alaska as extreme floods for the period of record through water year 2022. Flood-generating mechanisms for the floods were identified from streamgaging and climate data, historical documents, and seasonal associations. Outburst floods, mostly from glacier dammed lakes, accounted for 25 percent of the extreme floods. Of the non-outburst extreme floods, atmospheric river data for 1980-2019 showed that 45 percent occurred concurrently with an atmospheric river, and historical documents helped show that other rainstorms, including typhoon remnants, convective storms, and undifferentiated rainstorms, generated 29 percent. Melt-based mechanisms, including spring snowmelt, rain-on-snow, and summer high-elevation melt, generated 22 percent. Contributing factors including wet antecedent soil conditions from recent rainfall, high antecedent streamflow from high-elevation melt, warm temperatures that altered seasonal rain-to-snow proportions, and the contributions of snowmelt to rainfall events indicate that intense precipitation is an incomplete predictor for extreme floods. Outburst flood magnitudes were often more related to basin conditions than to meteorologic conditions, making them generally unsuited to conventional flood frequency analysis. Peak flows generated by atmospheric rivers or snowmelt formed common peak-flow populations that could be addressed by advanced techniques such as multiple-population analysis.
