The turbulent bottom boundary layer and its influence on local dynamics over the continental shelf

Abstract

We report on measurements of the structure of the bottom boundary layer on the continental shelf off Vancouver Island. A time series of vertical profiles obtained with the microstructure profiler FLY II revealed large temporal variations in the dissipation rate and in the density structure. The near-bottom current structure was simultaneously measured at fixed heights with conventional current meters. The data reveal the association between the predominantly diurnal tide and the variations in the structure of the bottom boundary layer. A clear distinction appears between the turbulent bottom boundary layer (8–40 m deep in a total water depth of 138 m) and the well-mixed layer (20–40 m deep). The two layers vary independently, with horizontal advection dominating the fluctuations in the thickness of the well-mixed layer while local dissipation is more closely related to the thickness of the turbulent layer. Variations in the density structure of the bottom layer are related to the strength and direction of the vertical shear and to the regional distribution of isopycnals. Current veering is commonly concentrated above the well-mixed layer. The evolution of the characteristics of the bottom layers is followed through a tidal cycle and related to local dissipation and other variables. Microstructure measurements from six locations over the southern portion of the Vancouver Island continental shelf are used to estimate the influence of turbulent energy dissipation on regional-scale flows. That fraction of the dissipation taking place in the bottom boundary layer is attributed to barotropic tidal flows, while that occurring above it is associated with nearly geostrophic baroclinic flows. Results give a lower limit of ∼ 1070 km for the length scale of shelf wave decay, in good agreement with current models; the estimates of tidal friction based on our dissipation measurements are however much lower than required by astronomical observations. An estimate of 230 h is obtained for the spin-down time of the local Tully eddy, rather longer than the decay time of 68 h obtained from observations.