Characterizing Extreme Events Relevant to Hydrologic Design Using Next-Generation Intensity-Duration-Frequency Curves Over the Conterminous United States

Snowmelt and particularly rain-on-snow (ROS) are often associated with extreme flood events in snow-dominated regions but have been neglected or under-represented in standard precipitation-based Intensity-Duration-Frequency (PREC-IDF) curves. The recently developed Next-Generation IDF Curves (NG-IDF), as an alternative to PREC-IDF, estimates (10-, 50-, 100-year) extreme events based on the actual water reaching the land surface available for runoff (AWR) through explicit accounting of rain, snow melt (without concurrent rain), and ROS. This study describes the NG-IDF based assessment of extreme events including their magnitude, trend and dominant source in the Conterminous United States (CONUS). The well-validated Distributed Hydrology Soil Vegetation Model (DHSVM) was forced by 1/16th-degree Livneh near-surface meteorological data to produce the CONUS-wide gridded subdaily AWR datasets for the period 1950-2013. These data were used to construct NG-IDF at over 200,000 locations with recurrence intervals of 10, 50, 100 years and durations of 24 h, 48 h, and 72 h. Results suggest that the NG-IDF extreme AWR events (with varying return intervals and durations) are rain-dominated in over 80% of the Southeast, Great Plains, Southwest and Midwest regions, and are ROS-dominated in over 60% of the Northwest and Northeast regions. The extreme AWR events are typically higher in magnitude in the eastern U.S. (especially the rain-dominated southeastern U.S) and the western mountain ranges (particularly the ROS-dominated Cascade-Sierra and the Pacific Coast ranges). In comparison to PREC-IDF extreme events, extreme AWR events are significantly higher in magnitude in ROS- and melt-dominated regions (e.g. the mountain ranges in the western and northeastern U.S), and snowbelts near the Great Lakes strongly affected by lake-effect snowstorms, indicating substantial underestimation of design basis extreme events in these regions with PREC-IDF.