We here investigate the statistical characteristics of ocean waves and surges over Taiwan waters by cluster and return period analysis. The results of cluster analysis show the data records from neighbor stations would not be always fall into the same cluster due to the influence of topography around wave stations. In addition to, the wave height of the specified return period is significant related to the probability distribution of the data set. Because of the extreme wave record in Gueishandao sea area around Taiwan, the suitable probability distribution can not be well determined in this area. It would affect the wave height analysis of the return period. We would continue the study to discuss the suitable probability distribution, so as to determine the accurate wave height in different situations of the specified return period.
Ocean waves and surges have attracted considerable attentions throughout history. The mechanism of wave and surge is still not fully understood in the present day. Statistical analysis of the wave and surge data is a quite popular and practical way to understand the characteristics of these ocean phenomena. It is also possible to predict the sea states by some statistical methods (Liu et al., 1998; Tsai et al., 2002; Lee et al., 2007). Taiwan, located between the tropics and the subtropics, lies on the border between the largest land mass and the largest ocean in the world. The marine and atmospheric environments here are complex and sensitive. Due to the island of Taiwan is situated on typhoon tracks; typhoons always bring severe sea states. Thus Taiwanese coasts receive direct impacts by huge waves especially in the eastern part of the island as well as large surge heights in the western part. In the past 100 years, there were average 3.5 typhoons attacking Taiwan per year.
