82 



THEORY OF SEAKEEPING 



all of this work should now be repeated with proper re- 

 gard for the new theoretical requirements. As a mini- 

 mum recjuirement, the obser\'ed waves must be pre- 

 sented in spectral form. Consideration must be given 

 to short-crestedness; i.e., to the directional spectrum of 

 wa\-es. There are indications that this may strongly 

 differ in a wind flume from that of the larger natural 

 waves; a too high U/c ratio and wall reflections appear to 

 exaggerate the short-crestedness of the flume waves. 

 The surface-drag characteristics may be strongly af- 

 fected by the degree of short-crestedness. 



Velocity-gradient measurements in the air flow should 

 be made simultaneously with drag mea.'^uremcnts by 

 means of water-surface inclination. Although the use 

 of such measurements as an alternati\'e method of wa\'e- 

 drag determination has been known for long, the recent 

 work of Miles and Phillips has intensified interest in 

 them for a more complete in\'estigation of air-flow prop- 

 erties, including the evaluation of air turbulence. 



Corresponding measurements of water tiu'bulence also 

 may be useful in evaluating the transfer of energy from 

 wind and its dissipation in wa\'es. Presumably, Miles' 

 theory pertaining to the air flow can be applied also to 

 the conditions existing in the upper layers of water. A 

 distinction must be made between the potential air-flow 

 energy transmitted to the potential wa\'e energy and the 

 turbulent energies of aii' and watei- which are cau.sed by 

 viscosities. 



Comprehensive measurements in large sea wa\'es are 

 difficult to make and are expensive. Wind-flume ex- 

 periments have the wall effects on water wa\'es and on 

 air flow which are tlifl^cult to evaluate with certainty. 

 While both of the.se acti\ities should be continued, it ap- 

 pears to the author that the most rapid progress can be 

 made by ob.ser\-ations in moderately large but limited 

 water areas and in pred(.)minating wa\'es of, say, 10 to 30 

 ft long. The geography of the United States offers many 

 opportunities of locating and utilizing such areas. The 

 water mu,st be sufficiently deep to avoid wave distortion, 

 yet shallow enough to permit installations of poles or 

 towers carrying measuring instruments. Measurements 

 .should be made simultaneously at several stations along 

 the wind direction, and the simultaneous recording tech- 

 nic|ue must permit correlation analy.ses. This is particu- 

 larly important for the e\'entual separation of the energy- 

 transfer problems from those of energy dissipation. 

 Care should be taken in choosing a site so a.s to assure 

 freedom from swells arriving from elsewhere and from 

 wa\-e reflections. The results of observations should be 

 used in de\-elopment of semi-empirical and rational rela- 

 tionships, but it should not be assumed that small-scale 

 seas are directly equwalenl to ocean waves. The U/c 

 ratio appears here to be the prime controlling factor, 

 together with the fact that surface tension and viscosity 

 makes it impractical to conduct model observations at 

 sufficiently low U/c ratios. 



Condensed List of Suggested Research Topics^ 



1 Wind-Flume Experiments, as describetl in Section 

 2.3 .shoulil be repeated and more comprehensi\-e measure- 

 ments taken as outlined in Section 9.2. As a minimum, 

 measurements at three stations along the flume are 

 needed to pro\'ide a possible means of separating analyti- 

 cally the sinuUtaneous transmission of energy from the 

 wind and dissipation of energy in waves. Particular 

 care is needed to prevent wave contamination by waves 

 reflected from the leeward beach. If an artificial wave 

 generator is used to increase the effective fetch (as by 

 Francis, 1951), the generator should be adjusted to pro- 

 vide initial waves identical with the wind waves pre- 

 viously found at the leeward beach without generator ac- 

 tion. SuflScient flume height is needed to assure a nor- 

 mal wind-\-elocity gradient near the water surface. 



2 Wind-Tunnel Experiments sliould be made on wave 

 surface drag, similar to Motzfeld's, Section 2.3, but with 

 short-crested waves and ol)li(iue long-crested waves. 



3 Co-ordinated Wind-Tunnel and Wind-Flume Ex- 

 periments. It would lie desirable to have correlated 

 wind-tunnel and wind-flume experiments, where the 

 wind-tunnel model corresponds to the wave structure 

 instantaneously recorded in the flume (refer to Sections 

 2.3 and 2.4). Particular care must be taken to repro- 



" Approximately in the order of presentation in the body of the 

 monograph. 



duce sharp-crested wavelets. The measurements made 

 for model reproduction will also serve to estimate the 

 percentage of the water surface covered by such wavelets. 

 Proper attention to short-crestedness is important be- 

 cause the drag of a sharp-crested wavelet can be ex- 

 pected to diminish rapidly with obliciueness of the crest 

 to the normal to wind direction. 



4 Observations on Limited Natural Bodies of Water 

 (such as Van Dorn's, Section 2.5), l:)ut with a complete 

 record of wave and air conditions. Section 9.2, are 

 recommended. The.se can also, with advantage, be 

 made on larger water areas than were used by Van Dorn 

 and in larger waves. The limit here is set only by the 

 feasibilit.v and coiu'enience of instrument installation. 

 Larger waves are suggested in the hope of obtaining 

 c/U values more nearly corresponding to ocean-wave 

 conditions than is practicable with small waves. It is 

 suggested that two intermediate observation stations be 

 u.sed in addition to the two at the beginning and the end 

 of a fetch. The end stations must be located at rea.son- 

 able distances from the shore in order to record suf- 

 ficientl.v developed wa\-es at the beginning,-'- and to avoid 



" The author can see Httle practical usefulness in the study of 

 wave initiation. In his opinion, Roll's (1951) observations of 

 waves, which were irregular from the beginning and traveled in 

 oblique directions to the wind, exhaust the \isefulness of initial 

 wave study. The author realizes, however, that many oceanog- 



