Mteroscopte Structures of Wind Waves 
high wind velocity. 
IV.2 Inception of high-frequency components of wind waves 
The standard deviations of the slope-distribution curves, 
shown in figure 7, are equivalent to rms water-surface slopes and 
are shown in figure 8. The trend of the slope variations shows that 
the rms water-surface slope is only a fraction of a degree at very 
low wind velocities, U_ < 1.9 m/sec. A sudden rise of the water- 
surface slope is observed in the wind-velocity range between 2 and 
3.5 m/sec. Beyond this range, the slope shows a steady but gradual 
increase with wind velocity. Figure 8 shows that the slope seems to 
reach a saturated value of 17.3°, when the wave growth with wind 
velocity ceases, and whitecaps appear at virtually every wave crest. 
As is shown in figure 2, the slope resolution of the present 
optical instrument is 1°, It is conceivable that increasing the slope 
resolution of the instrument (narrowing the slope-response curve 
shown in figure 2), would shift the lower end of the sudden slope rise 
(with slopes less than or near 1°) toward lower wind velocities, but 
the major rising portion (with slopes greater than 1°) will stay in the 
same wind-velocity range, shown in figure 8. 
The boundary-layer regimes of the wind are superimposed 
on the velocity scale in figure 8. It is very interesting to see that as 
soon as the effective transition of air flow from laminar to turbulent 
occurs, the surface slope rises suddenly. This rapid change indica- 
tes, therefore, very efficient inception of waves as soon as the wind 
boundary layer becomes turbulent, as suggested earlier by Phillips 
(1958b). These rapid changes soon slow down but the waves have al- 
ready been created and have grown to a considerable size. 
A similar sudden rise of the water-surface slope was repor- 
ted by Cox (1958); see the comparison between the present and Cox's 
results in figure 9. It is seen that these two sets of data behave simi- 
larly with regard to the sudden slope rise corresponding to the incep- 
tion of wind waves, as demonstrated in figure 8. The sudden slope 
rise obtained in the present study has been related to the initiation of 
the turbulent boundary layer of the wind. No wind-structure survey 
was provided by Cox; the close agreement between Cox's and the pre- 
sent data at the wave-inception stage suggests, however, that the sud- 
den rise of the water-surface slope in his experiment is also related 
to the transition of the airflow boundary layer, as first suggested by 
Phillips (1958b), 
1443 
