In this paper, the artificial neural network is applied to predict the short-term typhoon surge in order to overcome the problem of exclusive and the nonlinear relationships. The observations obtained during three typhoons of Jiangjyun station in Taiwan will be verify the present model. The results indicate that the short-term storm surge can be efficiently forecasted one to six hours ahead by artificial neural network.
A storm surge is an onshore rush of water associated with a low pressure weather system, typically a tropical typhoon. Storm surge is caused primarily by high winds pushing on the ocean's surface. The wind causes the water to pile up higher than the ordinary sea level. Low pressure at the center of a weather system also has a small secondary effect, as can the bathymetry of the body of water. It is this combined effect of low pressure and persistent wind over a shallow water body which is the most common cause of storm surge, which can increase the mean water level 5 meters. Such a fast sea level raised in coastal areas can cause severe flooding and cost people their lives, particularly when the storm surge coincides with high tides.
Numerous storm surge investigations have been carried out since the 1950's. For instance, Hansen (1956) presented a numerical fluid dynamic model to predict storm surges for the North Sea. Jelesnianski (1972) used a coarse and fine mesh system to list the amplitudes of surges from hurricanes. Kawahara et al. (1982) applied a two-step explicit finite element method for storm surge propagation analysis. Jelesnianski et al. (1992) proposed the SLOSH model of Sea, Lake, and Overland Surge from Hurricane for surge prediction, which has been commonly used by emergency managers to determine which areas must be evacuated.
