Simulating Extreme Precipitation in the Lake Champlain Basin using a Regional Climate Model: Limitations and Uncertainties

Extreme precipitation, defined as the amount of precipitation falling during the heaviest 1% wet days, has increased significantly over the Northeastern United States, with an abrupt shift in 1996. The extreme precipitation events have caused damage to infrastructure and reduced water quality in the Lake Champlain Basin. Robust simulation of extreme precipitation events is thus critically needed to assess climate change risk in the basin. Here we utilize the Weather Research and Forecasting (WRF) model, a mesoscale numerical weather prediction system, to downscale ERA-Interim reanalysis data (~79 km spatial resolution) to a higher resolution (4 km) for the Lake Champlain Basin. We then evaluate the performance of the regional climate model in capturing historical climate from 1980 to 2014, with a focus on precipitation extremes in the pre- and post-1996 periods. Specifically, an ensemble of WRF experiments integrating multiple physics schemes using a one-way, three-domain nested model configuration (36, 12, and 4 km) are compared to gridded observational datasets. Simulated total and extreme precipitation were contrasted with observed precipitation across model configurations at both spatially averaged (244 km×268 km) and grid point (4 km×4 km) scales for a variety of metrics, including annual and seasonal precipitation, annual and seasonal extreme precipitation, and frequency and magnitude of daily extreme precipitation events. Regional biases in simulated precipitation were identified, which are partially attributed to limitations in representing terrain effects by WRF and indicated by overestimated precipitation in high-altitude areas and underestimated precipitation in lower altitudes. Our findings will facilitate hydrological and ecological modeling over the Lake Champlain Basin and support risk assessments under past and future climate.