Long-term joint distribution of wave height and period based on Copula functions

This paper deals with the long-term joint distribution of wave height and period based on the bivariate copula functions.A case is studied by using annual maximum wave heights and the relevant periods measured in Shanwei sea waters along eastern Guangdong.Some conclusions were reached,as follows:(1)Optimized marginal distributions of annual maximum wave heights and relevant wave periods can be represented by Pearson pattern three distribution and generalized extreme value distribution,respectively,and Gumbel-Hougaard Copula that belongs to Archimedean copula family was selected by using goodness-of-fit test;(2)Comparing the joint distribution,the design values of annual maximum wave heights and relevant periods were higher than those of single variable under the same frequency condition,in which difference of design values of wave heights was slightly larger than that of wave period;(3)The design values of joint return periods and co-occurrence periods of wave heights and the relevant periods could be used as upper and lower bound in the coastal ocean engineering construction design;(4)Conditional probability 1 indicated that the encounter probabilities of annual maximum waves with the relevant wave periods were very high,and the combination of probabilities can be as an engineering building damage risk rate.