Description of Variability

Time series of current and temperature sampled near the bottom at DS are dominated by fluctuations at the semidiurnal frequency (Figure 3.1). Temperature changes $\geq0.3^{\circ}C$ occur during spring tides, and $\leq0.1^{\circ}$ C during neap tides. The horizontal tidal current is stronger in the across-slope direction than the alongslope, with peak speeds of $0.2 ms^{-1}$ during springs and $<0.1 ms^{-1}$ during neaps. The alongslope flow is typically $<0.1 ms^{-1}$ , although the ratio of across-slope to alongslope tidal flow varies over the course of the experiment, due to changes in the baroclinic tidal component. The general variability at DN is similar to DS (Figure 3.2).

The standard deviation of the current velocity is similar at the two moorings ( $0.04-0.08 ms^{-1}$ ), with an increase in amplitude away from the boundary (Figure 3.3). Over the entire experiment the overall ratio of across to along slope flow is 1.2 at DS (Figure 3.3). The standard deviations of the temperature oscillations and the corresponding vertical displacements are larger at DS than DN ($120 m$ compared to $60 m$ ), consistent with the advection of the background stratification by similar currents up and down a steeper slope at DS than at DN. Besides the strength of the vertical displacements, the main difference between the two moorings is the partition of the variability between the tidal (detailed in the next section) and non-tidal contribution, namely the near-inertial to diurnal variability described in chapter 4. At DS, most of the variability can be explained by the tides, while a significant fraction of the variability at DN is related to near-inertial motions.

At both moorings, the long-term averaged current is weak, $<0.01 ms^{-1}$ at all depths (Figure 3.3). At the lowest ADCP bin at DS, the long-term averaged current is anomalously large ( $\sim 0.03 ms^{-1}$ ) and directed downslope. We checked the amplitude and correlation of the acoustic return signal for this bin, and could not dismiss it based on sensor noise or mooring contamination. Nevertheless, with only one bin showing the anomalous mean current, we do not consider it further.