The surface layers of the ocean form a part of the ocean--atmosphere boundary layer system, and as such are dominated by turbulent mixing processes and the air--sea heat and momentum fluxes. These layers are referred to collectively as the ``mixed-layer'' because there is almost always present at least one homogeneous layer near the surface in which the temperature and salinity profiles are constant with depth (the result of mixing), and the sound velocity increases with depth due to pressure effects. The bottom of the mixed-layer then most often represents the surface maximum of the sound velocity profile.
Many times, however, the situation is more complicated, in that the ``mixed-layer'' is actually composed of a sandwich of several homogeneous layers, each the result of different mixing, cooling and heating epochs on the diurnal, synoptic (i.e. weather) and seasonal scales, or any combination of these. Due to the limited scope of this chapter we will present only three of four of these configurations, selected on the basis of frequency of occurrence and their effect on acoustic propagation.
The boundary layer nature of the near-surface layers has strong implications for the models designed to simulate them. The principal requirement for these models is the correct specification of the fluxes of momentum and heat across the air--sea interface, and the correct simulation of the turbulent mixing due to surface wave breaking and shear of the wind-driven currents.
Due to the large discrepancy in spatial scales between the horizontal extent of atmospheric cyclones and the depth of the mixed layer, mixed-layer models have been generally constructed as 1-D models, with pressure gradients and all horizontal gradients neglected (see Kantha and Clayson , Martin ). These assumptions are very similar to the usual boundary layer approximations in hydrodynamics.