The kinematics and dynamics of the New England continental shelf and shelf/slope front

Abstract

A 37 day long field program was carried out in March 1974 on the New England continental shelf break to study the current and hydrographic structure and variability on the shelf and in the shelf/slope front. A second experiment was conducted in the shelf break region for one week in January 1975 to study frontal exchange processes. The mean currents during the March 1974 experiment all had a westward alongshore component, increasing in magnitude progressing offshore from ~5 cm/sec to a maximum at the nearshore edge of the shelf/slope front of between 10 and 20 cm/ sec, and decreasing in magnitude with depth. The current structure was such that the velocity vector rotated clockwise with depth in the shelf waters inside the front. The mean alongshore transport of shelf water was on the order of 0.4 Sverdrups through a cross-shelf transect south of Block Island. About 30% of the transport occurred in the wedge-shaped region offshore of the 100 m isobath and inshore of the front. Comparison of the observed mean currents with those predicted by the steady frictional boundary layer model of Csanady (1976) indicates that the model captures most of the essential features of the shelf circulation. The low frequency currents contain approximately 30% of the total current variance. An empirical orthogonal modal analysis indicates that for low frequency alongshore motions the whole shelf together with the water above the front moves as a unit and that the on- offshore currents are characterized by opposing flows at surface and bottom. The alongshore wind stress component is the dominant forcing term for these low frequency motions and for the subsurface pressure field as well. For motion with periods longer than 33 hours, the time derivative term in the cross-shelf momentum balance is comparable with the Coriolis term while the advective terms are 2 to 10 times smaller, on the average. The semi-diurnal tide is barotropic over the shelf with current magnitudes that increase almost by a factor of two between the shelf break and the inshore mooring 70 km shoreward. At the shelf break one-dimensional continuity gives the correct relation between the surface tide and the semi-diurnal currents. The semi-diurnal tide is clockwise polarized. The diurnal tide is baroclinic, increasing somewhat toward the bottom, is less clockwise polarized than the semi-diurnal, and has tidal ellipses aligned with the isobaths. The diurnal tidal energy decreases toward shore. Inertial energy in the frontal zone is equal to the semi-diurnal tidal energy near the surface. The inertial energy decreases with depth and is an order of magnitude smaller further on the shelf. The inertial oscillations are shown to be highly correlated with the wind stress record, arising and decaying on a time scale of 3 to 4 days. The inertial oscillations are shown to be preferentially forced by wind stress events that have a large amount of clockwise energy at near inertial periods. The frontal zone is shown to be in near geostrophic balance with an anticipated vertical shear across the front of the order of 5 to 8 cm/sec. Thus, there is a wedge-shaped region of velocity deficit that is confined directly under the front and above ~200 m. Outside of this region the velocity is alongshore to the west. Low frequency motion of the front is shown to exist on time scales from 3 to 10 days although the complete nature of the motions is not known. An oscillation of the front about its mid-depth position at periods of 3 1/2 to 4 days was caused initially by an eastward wind stress event forcing the front offshore near surface and onshore along the bottom. This was accompanied by large temperature oscillations near the bottom at midshelf and current oscillations confined to those current meters near the front. The internal wave band is most energetic in the center of the front, is about half as energetic above the front where it is subject to variations associated with the wind stress, and is smaller and nearly constant below the front. The internal wave energy decreases shoreward reflecting the decreasing stratification shoreward of the wintertime hydrography. Linear internal wave theory seems to break down in the conditions of the frontal zone. A stability analysis of the front to small perturbations is carried out by extending the model of Margules frontal stability of Orlanski (1968) to include the steep bottom topography of the shelf break region. The study covers the parameter range pertinent to the New England continental shelf break region and indicates that the front is indeed unstable; however, the associated growth rates are so slow that baroclinic instability does not seem to be a viable explanation for the observed frontal motions. Application of the theory to the nearly flat topography of the shelf itself shows that the front would be at least 20 times more unstable there suggesting that the front would migrate offshore to the shelf break region until a stable equilibrium was established between frictional dissipation and the instabilities.