NH31D-02
Incorporating Climate Change Projections into a Hydrologic Hazard Analysis for Friant Dam

Wednesday, 16 December 2015: 08:15
309 (Moscone South)
Kathleen D Holman1, Nicole Novembre1, Victoria Sankovich-Bahls2 and John F. England3, (1)Bureau of Reclamation Denver, Denver, CO, United States, (2)Denver Boulder American Meteorological Society, Denver-Boulder, CO, United States, (3)US Army Corps of Engineers, Risk Managment Center, Denver, CO, United States
Abstract:
The Bureau of Reclamation’s Dam Safety Office has initiated a series of pilot studies focused on exploring potential impacts of climate change on hydrologic hazards at specific dam locations across the Western US. Friant Dam, located in Fresno, California, was chosen for study because the site had recently undergone a high-level hydrologic hazard analysis using the Stochastic Event Flood Model (SEFM). SEFM is a deterministic flood-event model that treats input parameters as variables, rather than fixed values. Monte Carlo sampling allows the hydrometeorological input parameters to vary according to observed relationships. In this study, we explore the potential impacts of climate change on the hydrologic hazard at Friant Dam using historical and climate-adjusted hydrometeorological inputs to the SEFM. Historical magnitude-frequency relationships of peak inflow and reservoir elevation were developed at Friant Dam for the baseline study using observed temperature and precipitation data between 1966 and 2011. Historical air temperatures, antecedent precipitation, mean annual precipitation, and the precipitation-frequency curve were adjusted for the climate change study using the delta method to create climate-adjusted hydrometeorological inputs. Historical and future climate projections are based on the Bias-Corrected Spatially-Disaggregated CMIP5 dataset (BCSD-CMIP5). The SEFM model was run thousands of times to produce magnitude-frequency relationships of peak reservoir inflow, inflow volume, and reservoir elevation, based on historical and climate-adjusted inputs. Results suggest that peak reservoir inflow and peak reservoir elevation increase (decrease) for all return periods under mean increases (decreases) in precipitation, independently of changes in surface air temperature.