The aim of this study is to simulate wind waves at the northern coast of the Persian Gulf and estimate extreme wave characteristics based on the modeled wave data. Hence, the SWAN model was employed for wave hindcasting forced by the 6-hourly ECMWF wind data with a resolution of 0.5°. After hindcasting 11 year wave data, extreme value analysis was conducted based on the numerically simulated wave data. The results of extreme value analysis showed that the Weibull distribution is better fitted to the wave data in the studied area compared to other distributions.


The first step in the design of marine structures is the estimation of long term data about environmental conditions such as waves. The design of coastal and offshore structures is based on the extreme wave height that would be exceeded once in a given period of years. Estimation of the design wave height requires existence of short-term data at the desired location. Instrumentally recorded, visually observed or numerically modeled wave data can be used for this objective. Due to the lack of records and observations, and recent developments of the numerical wave models, numerically modeled wave data are widely used as the data source for the estimation of extreme wave characteristics. Some researchers have studied the effect of wave modeling on the extreme wave analysis. As an example, Caires and Sterl (2005) estimated global return value for significant wave height based on a reanalysis meteorological parameters from 1957 to 2002. They found that estimated return values of Hs are underestimated and proposed a linear relationship to correct them based on the buoys return value estimates. Because of the lack of field data and low resolution, their findings cannot supply extensive information for countries around semi-enclosed basin such as the Persian Gulf.

This content is only available via PDF.
You can access this article if you purchase or spend a download.