Knowing the wind properties is a key to the assessment of wind energy resources and for wind power forecasting. In the complex terrain and coastal regions, where a significant portion of wind energy arises from regional/local winds, it is beneficial to utilize a chain of numerical models to dynamically refine the associated wind fields. Obtained wind speed forecasts from ALADIN 8 km (hydrostatic) and ALADIN 2 km (non-hydrostatic) grid spacing model simulations and refined CFD- like model version at 2 km horizontal resolution (DADA ; hydrostatic) have been compared with measurements for the period of 2011-2012. Statistical verification procedure was performed considering wind speed as continuous and categorical variable. The continuous statistical scores, such as multiplicative mean systematic error (MBIAS), root-mean-square error (RMSE) and mean absolute error (MAE) were calculated and averaged over monthly periods to show their seasonal variability. For verification of modeled mean wind speed as categorical predictand 4 x 4 contingency tables were used (thresholds were 3.4, 5.5 and 8 m/s). Frequency bias (FBIAS ; unity implies no bias) and critical success index (CSI ; measure of relative accuracy) were considered for each category. Overall skill was estimated using Heidke skill score. As expected, both versions of the 2 km grid model give an overall improvement over the 8 km grid spacing model. Unlike ALADIN 8 km, both DADA and ALADIN 2 km more often predict category of strong winds, and less often category of weak winds. That leads to better relative accuracy in category of strong winds, especially at more complex terrain. |