It has been revealed that higher moments of parent distributions influence the tail characteristics of annual maximum wind speed distribution. Some studies have been conducted to show that once the standard deviations of skewness and kurtosis of yearly variation for the 10 minute mean wind speeds are properly estimated, a good probability model for the annual maxima can be obtained. However since the parameters for four moments have various effects on the location and scale parameters of estimated
Gumbel distribution for annual maxima, it is interesting to see more general relationship between those parameters of parent distributions and the location and scale parameters of Gumbel models. This paper aims at examining the characteristics of the moment parameters and their effects on the Gumbel models. Once the relation between the moment parameters and approximate annual maximum statistics is known,
even a short period statistics such as 5 years or 10 years may help to have an approximate estimation of annual maximum. It is expected that such examinations of probabilistic characteristics of 10 minute mean wind speeds may provide good information for wind hazard models for individual sites.
Proceedings of ASRANet Integrating Structural Analysis, Risk & Reliability 5th International Conference 2010, 9p.