Wind fields are of key importance to predict loadings and power outputs on offshore wind turbines. However, they are complex to simulate because of global and regional meteorology, long- and short-term climate change, and air-sea interactions. In this research, wind speed time series at various locations and elevations are obtained from the Weather Research and Forecasting (WRF) model for North Atlantic offshore waters. Extreme statistics are derived and verified with floating light detection and ranging (LIDAR) measurements collected in the same time window. Histograms and probability density functions are further analyzed and compared with the standard Gaussian (Normal) distribution. Results show that WRF simulations are in good agreement with the LIDAR measurements and the empirical models. The implemented simulation approach can be generalized to worldwide coastal and offshore regions for wind energy development and improving the reliability of wind power supply.