Wave records of the Central Weather Bureau (CWB) of the Republic of China have been analyzed. Our primary objective for the study is to find the most suitable statistical model(s) for the wave climate around Taiwan. The other objective is to try to detect possible trends for the extremes. Due to some unexpected difficulties, these have not been achieved. Only some results of fitting statistical models can be presented here. Fourteen statistical models, most of them are commonly found in frequency analysis, were tested. These include the Gaussian, the Rayleigh, the Weibull, log-normal, Gumbel, generalized Pareto, as well as the generalized extreme value (GEV) distributions. For the present study, no independency and homogeneity checks were made. Wave heights of two-hourly data were used directly to fit the model. Chisquare (χ2) and Kolmogorov-Smirnov goodness-of-fit tests were then used to select the most suitable model(s). To test the possible differences of dimensionalization, both dimensional and nondimensional wave heights were used for the analyses. Our present results show, that the Gaussian, the log-normal, the gamma, Gumbel and GEV distributions can all be used as models for the wave climate around Taiwan. The values of the shape parameter of the GEV model are found to scattered around zero. This seems to indicate that the Weibull and/or Fréchet distributions are more suitable than the Gumbel distribution for wave climate around Taiwan.
Nowadays, the majority of the scientific world believes that global warming is inevitable. Since water can absorb more heat than air, it seems natural to assume that the ocean will be warmer (Schubert et al., 2006). There will probably also result in more tropical cyclone generations, as well as more heavy rain in some regions around the world, although admittedly, there is no consent among scientists.
