Probabilistic Flood Hazard Analysis using RiverWare
Noah Friesen, Shaun Carney, Zelalem Mekonnen—RTI International
ABSTRACT
RTI International has used RiverWare to examine hydrologic hazards caused by extreme storms at many dams for TVA and New Brunswick Power. RiverWare is run in an overall framework including the Stochastic Event Flood Modeling (SEFM) tool that samples storm frequencies and storm templates and runs the generated meteorology through hydrologic models. The python framework routes the flows to each dam in the RiverWare model and works with RiverWare to model the storm and store the output. This modeling has generated some unique challenges and solutions. Running flows many times larger than historical observations (greater than 1 in 1 billion year events in some cases) requires modifications to the model to ensure the run can complete and still represents reasonable operations. The presentation will touch on a couple of these modifications and challenges. Additionally, we were interested in analyzing the effects of gate malfunctions and failures at a dam in Canada. SEFM was modified to sample the chance of a malfunction (stuck gate) or failure (broken gate) and provide information about the failure to RiverWare. RiverWare simulates the reduction in output capacity and the resulting increase in maximum pool elevation. The presentation will also go over some of the lesser challenges and results from that effort.
Click HERE for a PDF version of the presentation slides.
