SECT. 5] 



SURGES 



617 



terms of forced planetary waves of the ty])e discussed by Veronis and Stommel 

 (1956). In the equatorial region of the Pacific there is, at times, a regular 

 oscillation of sea-level associated with waves in the trade-wind regime (Groves, 

 1956). 



It should be mentioned that phenomena other than atmospheric ones influence 

 the day-to-day variation of sea-level : thermal expansion of the water, long- 

 period tides, etc. Variation in temperature can give rise to sea-level changes of 

 the order of centimeters. Long-period tides, especially the lunar fortnightly, 

 account for much of the coherence between stations in Fig. 4. 



The largest anomalies in the day-to-day variation seldom exceed a meter. 

 The surges which do the most damage by inundating coastal areas are associated 

 with the transient response of the sea surface. 



B. Transient Surges 



We shall now consider the high-frequency end of the surge spectrum, in which 

 the sea surface is not in equilibrium with the disturbing forces. The usual 

 procedure is to take a tide record, subtract the predicted tide, and to consider 



Allonlic City N-J. 



iTroced from onginol tide gage records 

 [predicted tide curve is superficial J 



Fig. 5. Observed and predicted tide at Atlantic City, Sept. 14-15, 1944. (After Harris, 

 1959. By courtesy of the U.S. Department of Commerce Weatlier Bureau.) 



the remainder as the surge. Essentially one neglects the difficulties of non- 

 linearity discussed in section 1-B of this Chapter and assumes that an accurate 

 prediction is available. Even if a prediction is available for the very locality 

 under study, there are errors, and so the "surge" will always contain tidal 

 periodicities to some extent. 



The surges caused by tropical-type hurricanes are among the most intense 

 observed. Regions where a wide, shallow shelf is combined with the frequent 

 occurrence of hurricanes are particularly susceptible to damaging surges. Such 

 conditions exist along the Atlantic and Gulf coasts of North America, where 

 surges have been extensively studied by Harris (1956, 1957a, 1958, 1959), 

 Kajiura (1958, 1959), and Reid (1956, 1958). Fig. 5 (taken from Harris, 1959) 

 shows a predicted and observed tide curve for Atlantic City during the hurricane 



