These two methods therefore result in the production of six synthesized data sets for each of the four sites, two at 80 m, 100 m, and 120 m, respectively, one produced at each height using an FWSC, and the second produced using a VWSC where the variable coefficient distribution has been calculated every hour and is displayed in Figure 3.
The root mean square error, (6), was used to compare the accuracy of the synthesized time series wind data produced at each of the three heights for each of the four sites using an FWSC and a VWSC with the actual time series wind data obtained at 80 m, 100 m, and 120 m using SoDAR.
These results present the difference in RMSE between the synthesized wind speeds produced using the two methods, FWSC and VWSC, (6).
Figures 4(a), 4(b), and 4(c) show that the RMSE obtained at 80 m, 100 m, and 120 m are consistently better using the VWSC than that obtained using the FWSC. The improvement ranges from 41% to 4% reduction in root mean squared error (RMSE) between predicted and actual wind speeds when using a variable wind shear coefficient at heights ranging from 33% to 100% above the highest actual wind measurement.