In terms of loss of life and property damage, storm induced coastal flooding has become the world's foremost natural hazard. As atmospheric weather systems pass over ocean areas, water level oscillations are induced both by the wind stress and the horizontal gradients in atmospheric pressure associated with such systems. Storm surge is by definition restricted to the storm induced oscillations of water level with periods ranging from a few minutes to a few days. This definition excludes the wind waves which typically have periods on the order of a few to several seconds. These high frequency waves do play a significant role in the coastal flooding problem and can contribute directly to high coastal sea levels due to wave set-up, but these effects are not considered in this thesis. It is the combination of large amplitude storm surge, large amplitude wind waves and high tides which can cause enormous devastation to coastal areas.

Much of the work done on the study of storm surge has dealt with the large, highly destructive surges found in broad, shallow areas such as the continental shelf regions of the Bay of Bengal, the North Sea and the Gulf of Mexico. In those areas and many others storm surge can play a dominant role in coastal flooding. But often it is the coincidence of storm surge with high tides and other oceanographic factors which generates coastal flooding.

It was the occurrence of many oceanographic and meteorological factors which caused the extensive (damage estimates totaled over $100,000,000) coastal floods of the winter of 1982-1983 along the California coast. In addition to wave attack, four factors were;

1)High predicted astronomical tides;

2) Storm surge due to strong North Pacific storms;

3) High sea level associated with the EI Nino event of 1982-1983;

4) The cumulative effect of slow, secular rise in sea level.

Tides, which are the most predictable of all oceanic processes, contribute the largest amount of variance to the sea level signal. However, it has been suggested that more than half of the variance of sea level anomaly (observed level minus predicted tide) at short (hours to days) time scales can be related to atmospheric forcing by pressure and wind. Secular sea level rise, though important for its long term effects on coastal regions, is negligible for day to day or even year to year variations. Similarly, the increased ocean temperatures associated with El Ninos generally influence sea surface variability on seasonal time scales. Monthly mean sea levels, in 1983, were on the order of 10 cm above the long term monthly means. This sea level rise played a significant role in the coastal flooding seen that .year, but the influence of El Nino events on high frequency sea level anomalies is most likely due to enhanced storminess in the central North Pacific, which is attributed to El Nino episodes, not to above normal water temperatures.