Bias Adjustment of Dynamically Downscaled Climate Simulations for Hydrologic Modeling of the Rifle River Watershed

The 1025 km² Rifle River Watershed inputs into the Saginaw Bay, a location that provides key ecosystem services to the western Lake Huron region. The NCAR Weather Research and Forecast model (WRF) is used to dynamically downscale climate simulations from CMIP5 to a nested domain of 3km grids centered over the Great Lakes region. Two 15-year simulations are completed, one at the end of the 20^th Century for validation purposes, and one at the end of the 21^st Century with an RCP 8.5 scenario. Quantile mapping is applied to daily temperature, precipitation, and relative humidity to achieve bias correction by transforming the model data to fit with the empirical cumulative distribution function of the observed data. These corrections result in model distributions that are statistically similar to observed values for almost all variables at all stations. WRF output is coupled to the Soil and Water Assessment Tool (SWAT) hydrologic model to assess the impact of bias-correction and projected climate conditions on streamflow into the Saginaw Bay.