Subtidal Response of Scotian Shelf Circulation to Local and Remote Forcing. Part I: Observations

Abstract

Data collected during the Canadian Atlantic Storms Program (CASP) suggest that two primary current regimes exist on the Scotian Shelf during winter. Moorings inside the 100-m isobath feature currents that are parallel to the general bathymetry and have small eastward mean velocities. The time-varying part of the flow is nearly rectilinear alongshelf at typical maximum velocities of 0.2?0.3 m s?1 and oscillates on 2?5-day periods in response to meteorological forcing. The current at locations beyond the 100-m isobath exhibits a persistent westward flow of about 0.3 m s?1, denoting the dominance of the Nova Scotian Current, although smaller magnitude along- and cross-shelf fluctuations occur at synoptic frequencies. Flow in both regions is largely uniform with depth at this time of year. The velocity response to local and remote sources of forcing was examined using a frequency (?)-dependent multiple regression model. Flow is dominated by nonlocal forcing at ? < 0.2 cpd; alongshelf wind stress is the principal local influence at low frequencies. Remote forcing is less important at higher frequencies, while the role of local cross-shelf wind stress is greater. The effect of local wind forcing is generally small and varies substantially on horizontal scales of order 10 km. Current typically lags wind stress by about 30°?45° at ? < 0.5 cpd; the response to alongshelf wind stress also suggests a phase propagation to the west. The typical response to alongshelf wind forcing is consistent with previous quasi-steady observations and model results from the Scotian Shelf. The contribution of cross-shelf wind stress is limited to a near-surface flux in the synoptic band. The remote response decreases offshore and to the west and propagates westward, in the direction that coastally trapped waves propagate. The current response to local and remote forcing is uniform with depth at nearly all locations. Local and remote contributions to alongshelf velocity, integrated across the mooring array, are in geostrophic balance with the cross-shelf subsurface pressure (SSP) gradient. Together, the two wind stress components and a SSP record representing nonlocal forcing explain 40%?80% of alongshelf flow, compared to 95% of the SSP signal. Some of the large-scale flow features are consistent with variations seen in SSP data. The most obvious similarity is the remotely forced response. However, these results also reveal substantial variations in the current on scales of 10?100 km that are not reflected in the SSP field. Irregular shelf bathymetry is thought to induce the observed variations in circulation. This study also demonstrates the need to include remote forcing effects, in the form of a properly prescribed backward boundary condition, in shelf circulation models.