Abstract
Navigation in coastal areas requires accurate water levels modeling and prediction. Basically, tides are generated as a response to the attraction forces exerted by the moon and the sun. However, such attraction forces are not the only factors affecting water levels. The shape of bays, local wind and weather patterns also can affect tides. In this paper, the least-squares spectral analysis (LSSA) approach is used to analyze long series of tidal data, atmospheric pressure, and wind speed extended more than 9 years. The tide prediction model is developed by determining the harmonic constituents of the tidal data using LSSA approach. It is found that the resultant spectrum still contains different peaks after forcing all tidal constituents. The water level response to atmospheric pressure is also investigated. The amplitude and phase response of tidal data to atmospheric pressure are determined. It is shown that the response of water level to the atmospheric pressure has an average of about 4.5 mm/millibar. Moreover, the amplitude and phase response of tidal data to wind speed is also investigated. It is found that the power ratio of pressure effect to wind effect is about 1.64 x 10(6). That means the effect of wind is too small compared to the effect of atmospheric pressure, which can be considered a special case for this location as it is surrounded by mountains that affect the wind speed and its variation.