Estimating coastal design extremes for given risks: A new approach

For many reasons, e.g., port operations, coastal construction planning, undersea structure survival, and underwater transport, man wishes to know the extreme values that are likely to occur in coastal oceanographic variables. This paper presents a hybrid statistics/ computer simulation method that uses archived oceanographic observations to estimate confidence levels on the most extreme values likely to occur over a given period in the future. The difference from previously developed methods is the ability to estimate the most extreme value over a time period for a given probability (as opposed to estimating the probability of exceeding a given value) and the ability to obtain results from empirical data without a great deal of theoretical oceanography. The method is applied to the California coast for a period of 100 years on the following variables: bottom surge particle velocity by water depth, wave height by water depth, wavelength by water depth, wave period, current velocity, regions of high density, regions of low density, and earthquake magnitude. Values are given for the 99- and 99.9-percent probability levels.