It is of interest to compare the KAMM and MesoMap model performances as used in both modeling studies. Validation of the KAMM model against tower data give an RMSE of 9.3% but no appropriate comparison with MesoMap is possible as model validation with Thai data was not given in the report (TrueWind Solution, 2001).
It is likely that differences between our results and those of TrueWind Solution are due to several factors. Firstly, fifteen years of hourly wind data were used in our study to capture the wind climatology as opposed to the use of representative sample data employed in the earlier study. Secondly, measurements at 100 m were used for display in our study while the earlier study provides a wind assessment at 30 m and 65 m. However at lower elevations wind systems from localized sources such as katabatic flows, valley flows and localized wind channeling are more likely to be operating and would be less representative of a regional wind. Lastly, the treatments of turbulent flow in complex terrains used by both models are probably different.
The percentage of the area which has various mean yearly wind speeds at the height of 100 m above the ground level with respect to the total area of the country was also investigated and results are shown in Fig. 6.