Abstract:
Geographically there is a possibility of having high wind speeds in Puttalam.
This can cause damages to man- made structures such as bridges, wind-
turbines, buildings, radio masts, etc. Therefore assessment of wind speed is utmost important to prevent from such disasters to some extent. Daily wind speed (kmh-1) data for Puttalam from January 2007 to April 2014 was used in
the analysis. “Extremes” package of R software was used for the data
analysis. In this study we present two different methods for extreme value analysis, Peak-Over-Threshold method, which is based on Generalized Pareto Distribution and Block Maxima approach, which is based on Generalized Extreme Value distribution. Parameter estimation was done using Maximum Likelihood estimation and L-moments method. Hypothesis testing on the shape parameter confirmed the type of the distribution to be used. According to Peak-Over–Threshold method, using the Mean residual life plot, threshold stability plot and diagnostic plots the best threshold value was found as 12.25 kmh-1. The shape parameter was negative and the null hypothesis of data fits exponential distribution was rejected (p-value<0.05). Hence the best distribution was identified as the Beta distribution. According to the Block Maxima method the null hypothesis that the data fits Gumbel distribution was rejected (p-value<0.05). Since the shape parameter was negative, data fitted well with the Weibull distribution. Using the identified
distributions return levels and their 95% confidence bands were obtained.
The 100 year return level using Peak-Over-Threshold method was 18.29 kmh-1 whereas according to the Block Maxima method it was 17.99 kmh-1.
