Investments in both on- and off-shore wind energy are being rapidly scaled up to transform the energy system and reduce greenhouse gas emissions. Our research addresses two themes; (1) How much electricity can be generated by existing technology (i.e. bottom-mounted wind turbines) along the US east coast? And what is the uncertainty? (2) Will global climate non-stationarity change the amount of electricity that can be generated from wind turbines? To address theme 1 we are performing detailed simulations with the Weather and Research Forecasting model and sampling uncertainty across; a range of wind turbine layouts, different wind turbine parameterizations and different versions of WRF - specifically atmosphere-only WRF and the Coupled Ocean Atmosphere Wave Sediment Transport (COAWST) model.

Currently, wind turbines supply approximately 10% of total U.S. electricity consumption. For the most likely wind turbine spacing of 1.85 km up to a further 5% of national supply could be generated from areas along the US east coast that have already by leased for development. However, the wind farm parameterization is a major source of uncertainty and preliminary results suggest there may be substantial benefits to fully coupled simulations in terms of accuracy of power generation predictions. For theme 2 we have performed projections of daily electricity generation from existing onshore wind farms in the contemporary and future climate using both machine learning and WRF. These indicate small declines for the Midwest, Northern Great Plains and the Northeast primarily driven by reduced summertime windiness. Electricity generation is projected to be stable in the wintertime and over the Southern Great Plains. Energy system electrification will preferentially enhance wintertime electricity demand over much of the US. Our work suggests evidence this may lead to greater synchronization between demand and times of highest wind energy production.