Analyzing Changing Flood Risk in the St. Joseph River Basin Using Observations and Modeling

Session: Applications of Simulation Models in Watershed Science and Lake Ecology (1)

Alan Hamlet, University of Notre Dame, [email protected]
Kyuhyun Byun, University of Notre Dame, [email protected]

Abstract

In Feb, 2018 the St. Joseph River experienced unprecedented flooding due to warm conditions that caused rapid melt of heavy snowpack, followed by more than five inches of rain over three days.  Peak stage measurements during the event almost exceeded the maximum capacity of the USGS gauge. Retrospective analysis of historical USGS peak flow records for the St. Joseph at Niles showed that the event had a return interval of about 2,500 years, and was an extreme outlier in comparison with historical peak flows.  Additional analysis revealed that there has been a statistically significant increase in mean annual peak flows pre- and post-1975.  Future projections using statistically downscaled climate change projections and the VIC hydrology model also showed substantial increases in flood risk over time, with the magnitude of the 100-yr flood increasing by about 45% relative to a late 20^th century baseline for the 2080s and RCP8.5 emissions scenario.  These three analyses all point to substantial increases in flood risk that should be carefully considered in long-term planning and management for the St. Joseph and Lake Michigan.