There are a considerable number of wave records around the coastline of the United Kingdom and elsewhere, but all too often the engineer is faced with a short data set which is situated too far from his particular area of interest to be of use. With this in mind, a wave predication model has recently been developed at Hydraulics Research. It has wide applicability, is easy to use and can produce a long data sequence without the need for large computer storage.
The techniques used in the model are not sophisticated but they do allow the effective use of any wave data which may be available. The model, called HINDWAVE, can estimate the mean annual wave climate at a site or produce a time-history of wave height, period and direction. It can be used in the absence of measured wave data or, equally well, it can extend an existing short wave record to a longer, more representative period. The model does not require a complex description of the wind field; instead it uses hourly averaged wind speeds and directions from the nearest coastal anemometer station.
The results may be produced hour by hour, month by month, or as seasonally averaged values. A certain amount of calibration can be done by adjusting the land-based wind speeds for use over the sea. With quite minor changes, i.e. an increase in land-based data of say 10%, very good agreement with measured wave data is usually obtained.
The HINDWAVE model was developed to meet the needs of coastal engineers requiring large wave data sets at specific sites, quickly, and at low cost. The input to the model consists of two sets of variables, one representing the shape of the wave generation area and the other the hourly averaged wind velocities in that area. The programs are written in an efficient way enabling long sequences of wind data to be processed rapidly.
The model may be used on its own in order to estimate a directionally dependent wave climate distribution, or it may be used in conjunction with measured wave data. Alternatively, it can be calibrated against wave recordings, and then used at other nearby locations and with other or longer periods of wind data. For example, this procedure can be used to put a single year of wave records into a longer-term perspective, or to extend the effective length of sequences of wave records, or to predict wave conditions at a point nearby which has a different exposure to wave action. In addition, the model adds the important wave direction parameter which is not usually available with wave measurements. This is particularly valuable for coastal or harbour engineering projects.
The model has been used successfully on about 20 recent studies at various sites around the British coast, and in several cases has been calibrated and proved against measured wave data. It has been used in conjunction with a refraction model, a beach plan shape model, and a method for extremes analysis.
