A Method to Predict Wave Conditions in Island Environment

In many ocean regions the archipelagoes or even isolated islands provide shelter and this effect has an influence on the wave climate for coasts or ocean areas located in the shadow zones. Incident waves from the deep ocean are blocked by the island boundaries and are refracted over the island shoals. The wave energy is partially dissipated in surf zones or reflected back to the deep ocean. However, many other mechanisms, such as wave diffraction, wave scattering and wave-current interactions spread wave energy into the island lee regions. That is why, the most appropriate way of modeling the wave propagation into such an environment is probably to consider the island matrix both from local and regional perspective. The availability of realistic data concerning the nearshore wave field is most important in coastal engineer applications, as well as, in many military landing operations. The numerical wave models are nowadays able to produce forecast products of the oceanographic parameters and make available in a useful time scale the environmental support necessary for various rescue and military applications conducting to a rapid environmental assessment of a tactically significant area. In this context the nowcast and forecast techniques based on the interactions between large-scale and high-resolution wave models became more and more common and effective. This is due to the relatively low cost in comparison with the more expensive task of maintaining permanent networks for ‘in situ’ oceanographic measurements. Moreover, the predictions provided by these models may be significantly improved by an infusion of field collected or remotely sensed data. As a consequence, the directional boys, as well as, some other devices (ADCPs, pressure sensors, etc) are very useful in the process of data assimilation or when making the calibration of the numerical models on specific sites. The ocean-scale models, WAM (WAve Modeling) and WW3 (Wave Watch III), are based on a detailed physical description of the air/sea interactions and give a statistical description of the time evolution of the sea waves using the spectral action balance equation, Komen et al. (1994). The large-scale wave models have been coupled to the operational atmospheric forecast models and have been made some global and many regional implementations. Predictions of the wave climate near the European coasts of the Atlantic Ocean are available on various Internet sites. As concerns the high-resolution models, the SWAN spectral model seems to be the most effective and that is why it was denoted as the community wave model. SWAN (acronym for Simulation WAves Nearshore) is a phase averaging wave model designed to obtain realistic estimates of wave parameters in coastal areas, lakes and estuaries from given wind, bottom, and current conditions, Holthuijsen et al. (2001). In SWAN the following wave propagation processes are implemented: propagation through geographic space, refraction due to bottom and current variations, shoaling due to bottom and current variations, blocking and reflections by opposing currents, transmission through or blockage by obstacles. The model also accounts for the dissipation effects due to whitecapping, bottom friction and wave breaking. In its last version the SWAN model overpasses the condition of high-resolution model being increased its applicability from a scale of 25 km to almost any scale. However, SWAN does not support oceanic scales being less efficient in this area than WW3 and WAM. Consequently, from this point of view, SWAN can be considered still a high-resolution model but with more extended capacities. As a limitation, SWAN does not account for diffraction. For this reason, in the areas where this phenomenon is relevant might be used also the REF/DIF model, which is a phase-resolving, weakly nonlinear, combined refraction-diffraction model that incorporates the effects of shoaling, refraction, energy dissipation and diffraction, Kirby and Dalrymple (1994).