SummaryA laboratory wind tunnel has been used to study the effect of wind on a water surface. The surface shearing stress 7'0 and the slope of the surface induced by wind have been measured. Values of the surface stress, in good agreement with each other, have been obtained from: (a) the velocity profile of the wind above the water surface, (b) the measured values of surface slope or set-up, and (c) the spreading characteristics of surface films. The drag coefficient, C h = TO/ pU~, was found to be constant for wind speeds up to about 412 cm/s and then to rise gradually for higher wind speeds.When the surface waves are damped out by the addition of small amounts of surface-active material to the water, the shearing effect of the wind on the surface is somewhat modified. If detergent solution is used for this purpose, there is a critical concentration at which conditions in the boundary layer near the water surface appear to undergo a marked change in character.

SummaryA laboratory wind tunnel has been used to study the movement of the surface of a body of water over which a wind is blowing. The ratio of the surface velocity Us to the wind velocity V has been measured for both smooth and wavy surfaces at wind speeds between 350 and 750 cm/s. It has been found that this ratio is markedly affected by the damping-out of surface waves. For a wavy surface, as obtained with clean water, us/V has a constant value of about 0·03. The damping of the surface waves is achieved by the addition of detergent solution to the water. There is a particular concentration of detergent, and a corresponding surface pressure, at which u s/ V ceases to have this constant value and begins to rise to' values around 0·045. For a fully damped surface us/V rises linearly with V for low wind speeds and tends to a constant value of 0·045 for wind speeds greater than 550 cm/s. These findings are, to some extent, in conflict with the observations and theories of previous workers.

SummaryThe surface velocity, surface slope, and velocity profile produced by the application of a wind stress to the smooth surface of a closed channel have been determined by adapting the empirical laws of flow in smooth tubes. The estimated responses agree well with the available experimental data.It is shown that the response of a channel with a rough water surface may usefully be predicted by separating the wind stress into two parts, one producing liquid flow and the other, surface waves. This formal division was suggested by Keulegan (1951).