Y2ii Ci iiJ

We pay attention to the decrease in ujuh and the increase in c0/u,, and go^/ul over time. The last circumstance is obvious, but the first needs comment.

We have discussed above the limiting situation when the momentum and energy fluxes coming from the atmosphere are completely spent on wave generation (wave breakdown and drift current formation do not occur). In reality, everything is much more complicated. Momentum and energy of the wind are transferred simultaneously to waves and the drift current, and in doing so the momentum and energy transferred to waves is only partly spent on their development: at wave breakdown a part of the wave momentum is transformed into the momentum of the drift current, and a part of the wave energy into the energy of small-scale turbulence in the surface ocean layer. A solution of the wind-wave interaction problem, as applied to this more realistic situation, was obtained by Benilov (personal communication). As a result it was possible to clarify that establishing the dynamic equilibrium between waves and wind at the stage of wave breakdown occurs much more slowly than at the wave growth stage, and that at the intermediate stage when wave phase velocities are still increasing but wave breakdown already comes to manifest itself, the wave momentum and energy fluxes turn out to be much lower than the momentum and energy fluxes caused by wave breakdown. This conclusion is confirmed by the estimate presented by Kitaigorodskii (1970), according to which the wave momentum flux is of the order of 10% of the total momentum flux, while its remaining part is redistributed by some (still not established) means between two other components controlling the processes of wave breakdown and drift current maintenance.

3.6 Vertical distribution of the temperature and passive admixture over an immovable surface

In a similar way, as was demonstrated in Section 3.1, it can be shown that in a near-wall layer the condition of approximate constancy of the vertical flux F of any passive admixture has to be fulfilled, that is,

dz where c is the mean concentration of the passive admixture, c' is the departure from its mean value, x is the molecular diffusion coefficient; other symbols are the same.

From general considerations, it is clear that the profile of c should depend on the admixture transfer intensity for which the flux F serves as its measure, on statistical characteristics of the velocity field determined by parameters u% and v, and on the molecular diffusion coefficient x■ Then, on the basis of the 7r-theorem of dimensional analysis

where f^zu^/v, Sc) is a new non-dimensional universal function satisfying the condition /c(0, Sc) = 0; Sc = v/% is the molecular Schmidt number; subscript 0 indicates belonging to the underlying surface (z = 0); the von Karman constant k is included in the multiplier before the function fjzu^/v, cc) for convenience.

Let us consider the case of a smooth surface and z « v/u^. At values of the number Sc not very different from one the condition z « v/m# provides the existence not only of the viscous sublayer but of a molecular diffusion sublayer. In the latter, on the basis of (3.6.1), c-c0 = ™z (3.6.3)

0 0

Post a comment