684 BABBER AND TUCKER [CHAP. 19 



damping. Secondly, the hydrodynamic exciting force depends on the wave- 

 length of the wave, and for pitching in head seas is usually greatest when the 

 wavelength is about 1.25 the length of the ship. For a stationary ship, the 

 component wave whose frequency is that of the ship resonance usually has a 

 wavelength considerably below that giving maximum hydrodynamic force, 

 but when the ship steams into the waves, the "Doppler" increase in the fre- 

 quency of encounter means that the resonant wave has a lower frequency 

 relative to a fixed point, and hence a longer wavelength and greater exciting 

 force. In a storm, it will probably also lie in a region of the spectrum where the 

 energy is higher. Thus, a ship heading into a storm can experience dangerously 

 large motion, which can be reduced by slowing down. 



The phase of the motion relative to the waves is also important, and governs 

 whether the ship will "slam", for example. 



It is apparent that to calculate the total response of the ship, the wave 

 spectrum is required in both frequency and direction. 



The wave information required for this field of research may be summarized 

 as follows : 



(1) Measurement of the directional wave spectrum and of instantaneous 

 wave height during experiments on the correlation of ship motion and per- 

 formance with wave conditions. 



(2) Knowledge of wave spectra under various meteorological conditions, to 

 allow them to be reproduced in model experiments, and to allow calculation of 

 the behaviour of ships (when the theory has been adequately developed). 



(3) Knowledge of the statistics of wave conditions on the shipping routes of 

 the world, to allow calculation of the optimum compromise between the various 

 factors affecting the economic running of ships. 



9. Methods of Observation and Analysis — Methods Taking No Account of 



Direction of Travel 



Many laboratories have attempted to design instruments which will measure 

 and record the wave spectrum directly (for example. Barber, 1949 and 1954; 

 Valembois, 1955), but none of these has been used to any extent in practice. 

 This is because there is little incentive to overcome the considerable practical 

 difficulties involved. In the early days of spectrum analysis it was thought that 

 an immediate knowledge of the wave spectrum might be useful for storm 

 detection in areas where there is little meteorological coverage. Detection of 

 long-period swell shows the existence of a storm, and the rate at which its lower 

 period-limit changes gives the distance from the recording station (see section 5 

 of this Chapter). However, the information becomes available rather late and 

 measurement of the direction of the storm is difficult, so that this method has 

 not been used in practice. At present, there is no application which requires an 

 immediate knowledge of the wave spectrum. Thus, it has proved more con- 

 venient to record the waves and to perform any required analyses later. 



For research purposes, the recent trend has been towards instruments 



