TY - DATA AB - **Overview** The purpose of this work is to assess the sensitivity of the forecast for turbine height wind speed to initial condition (IC) uncertainties over the Columbia River Gorge and Columbia River Basin for two typical weather phenomena: a local thermal gradient induced by a marine air intrusion and passage of a cold front. The Weather Research and Forecasting (WRF) model data assimilation system (WRFDA) was used to generate ensemble ICs from the North American Regional Analysis (NARR) for the WRF model initialization. The simulated turbine-height wind speeds were categorized into four types using the self-organizing map (SOM) technique. This work advances understanding of IC uncertainties impacts on wind speed forecasts and locates the high-impact regions. **Data Details** This dataset contains 100 ensemble simulation outputs over the WFIP2 study region (outer domain, d01) for the cold-front case. Variables including surface pressure (PSFC), geopotential height (PH), pressure (P), 2-m temperature (T2), and wind speed (U & V) are compressed in a single zip file for each case. The files are named in the format of [var].[timestamp].nc, which contain 100 ensembles in one file. WRF namelists are also included. The cold-front case is for the time period 2016-08-21 to 2016-08-22. AU - Liu, Ye DB - Open Energy Data Initiative (OEDI) DP - Open EI | National Renewable Energy Laboratory DO - 10.21947/1819817 KW - a2e KW - atmosphere to electrons KW - wind KW - weto KW - eere KW - wfip2 KW - Wind Forecast Improvement Project 2 (model) KW - Simulated wind speed and initial conditions over the WFIP2 region: Cold-front (D01) KW - casestudy KW - wrf KW - 01 KW - coldfront LA - English DA - 2016/08/20 PY - 2016 PB - Wind Energy Technologies Office (WETO) T1 - Simulated wind speed and initial conditions over the WFIP2 region: Cold-front (D01) UR - https://doi.org/10.21947/1819817 ER -