SECT. 5] 



TIDES 



797 



Fig. 28 indicates the accuracy which could have been possible in forecasting 

 the storm-surge of December, 1954. In this example the computation has been 

 done without taking account of the tides ; only the wind stress and air pressure 

 have been used. Full lines indicate the residuals given by Rossiter (1958) ; the 

 broken lines have been computed. 



In this context another well-known phenomenon is of interest. Under 

 particular conditions, so-called external surges enter the North Sea from the 

 Norwegian Sea ; these move to the south, and in British coastal areas sometimes 



RJy 



Fig. 28. Storm-surge in the North Sea, December, 1954. Full lines and open circles, residuals 

 determined from observations of sea-level by Rossiter for River Tyne R. Ty, 

 Lowestoft L, Ostend O, Ijmuiden I, Terschelling T, Norderney N, Biisum Bii, 

 Thyboran Ty, Bergen B ; broken lines and full circles, computation. 



increase in amplitude. In eastern coastal areas these increases in amplitude are 

 unimportant ; indeed they are almost unknown. Fig. 29 shows the result of 

 the computation of an external surge which happened in December, 1954. The 

 lines marked with figures 10, 20 or 30 cm designate the areas where the maxi- 

 mum amplitudes of the surge were never greater than these values. The pattern 

 of these lines demonstrates the decrease in amplitude of this surge as it ap- 

 proached the south-eastern part of the North Sea. 



These examples indicate that the field of application of these numerical 

 processes is much wider than had previously been supposed. 



