17. SURGES 



P. Groen and G. W. Groves^ 



1. Introduction 



A. Definition of Surges 



From the point of view of the spectrum of sea-level, surges may be charac- 

 terized as being sea-surface disturbances with dominant periods ranging 

 roughly from 10^ to 10^ sec, or from 1 to 10^ h, thus falling between the tsunamis 

 and the lower frequency astronomical tides. 



From the causal point of view, surges are, in this treatment, understood to 

 be phenomena of atmospheric origin, with exclusion of disturbances of crustal 

 and of astronomical origin (tsunamis, lunar and solar tides). It is, of course, also 

 possible that a surge in a certain restricted sea area or bay originates from a 

 travelling disturbance of the sea elsewhere (in the same way as tides in marginal 

 seas are induced there by the oceanic tides), so that it is, in a way, of marine 

 origin, but its first cause will still be atmospheric if the primary disturbance is 

 of atmospheric origin. 



Only surface surges will be dealt with here. 



B. Separation of Surges from Astronomical Tides 



Separating the "surge" or the "atmospheric effect" in a sea-level record from 

 the astronomical tide is made fundamentally difficult by the fact that these 

 phenomena are dynamically non-linear : even the definition of what is tide and 

 what is atmospheric effect in the variation of sea-surface height offers a prob- 

 lem. In many cases a simple subtraction of the predicted astronomical tide 

 from the recorded heights will not work because of coupling effects, which 

 make the difference obtained in that way show more or less pronounced 

 secondary oscillations with tidal periods. A formal analysis of these coupling 

 effects is briefly explained later (page 625). One of them is particularly simple, 

 physically, viz. a time shift of the astronomical tide caused by an increased 

 depth of the water through which the tide travels. This effect is particularly 

 important in very shallow sea areas ; it may be eliminated by giving the pre- 

 dicted astronomical tide curve the proper time shift before subtracting it from 

 the tide record. Various practical ways of extracting the true "surge" from a 

 tide record have been described by (among others) W. F. Schalkwijk (1947), 

 R. H. Corkan (1950), G. W. Groves (1955), A. R. Miller (1958) and J. R. Rossiter 

 (1959). 



C. Classification 



From the point of view of local time sequence of sea-surface heights, a 

 classification of surges may be found in the extent to which the dominant 

 period of a surge is determined by its cause, the extremes being, on the one 



1 Of this chapter, sections 1, 3A and 3B have been written by P. Groen, sections 2 and 

 3C by G. W. Groves. 



[MS received June, 1960] 611 



