Charles Hannah, Jennifer Shore, and John Loder (BIO)

Modelling studies have led to three basic ideas about the nature of Scotian Shelf cross-overs:
  1.   The seasonal mean circulation for winter and spring contain cross-over pathways from the offshore edge of Browns Bank to the outer edge of Georges Bank and from the western flank of Browns Bank to Georges Bank via a long pathway around Georges Basin.
  2.   Cross-overs do not require a major reorganization of the circulation. What is required is modest perturbations to get Scotian Shelf water onto flow pathways that lead onto the Northeast Peak.
  3.   The simplest way to get surface water and drifters onto the northeast peak of Georges Bank, in the winter and spring, is from the north via the near-surface wind-driven flow (Hannah et al. 1998). Therefore it seems likely that cross-over events (drifters or surface water properties) that get inside the 100 m isobath, do so by crossing onto the bank along the northern flank. This is supported by the drifter observations from the 1999 field year.

The discussion of circulation and cross-overs will make use of the seasonal-mean flow fields of Hannah et al. (2000) and the Lagrangian analysis of Shore et al. (2000). Figure B1 shows that for drifters released on the offshore side of Browns Bank, the near-surface drift pathways based on the seasonal-mean winter flow field are broadly consistent with those observed in the winter and spring of 1999. In particular the primary pathways are northwest into the Gulf of Maine and southwest along the southern flank of Georges Bank. The observations also show particles crossing the Northeast Channel and moving onto the eastern end of Georges Bank, whereas the simulated drifters that cross Northeast Channel tend to move along the southern flank between the 100 and 200-m isobaths. The observations also show drifters leaving the shelf; this is not seen on the simulated drifters.

Shore et al. (2000) used an approximate velocity streamfunction to characterize the flow patterns. In winter (Fig. B2), there is a coastal current along the coast of Nova Scotia and around the Gulf of Maine, a small closed area on Browns Bank, and a large closed area in Georges Basin. The closed area on the crest of Georges Bank does not show at the surface but it is evident at deeper levels. Notice the flow from the Scotian Shelf around the outside of Browns Bank and back into the Gulf of Maine, the apparent recirculation (closed streamline) around Georges Basin, the flow from the Gulf of Maine onto the Northeast Peak, and the saddle point near the mouth of Northeast Channel. The saddle point is the point where the streamline representing flow along the outer shelf touches the streamline representing flow around Georges Basin.

Particle trajectories are very sensitive to the details of the flow near saddle points and small perturbations can lead to large changes in particle trajectories. Consider a particle moving along the outer shelf and approaching the mouth of Northeast Channel. Relatively small perturbations in the cross-shelf direction will determine whether the particle moves into the Gulf of Maine or proceeds directly across the Channel. This is illustrated by the simulated trajectories in Figure B1. The saddle point in Northeast Channel is required by a combination of the closed area in Georges Basin, the flows around the Gulf of Maine and Georges Bank and the shelf-break flow across the mouth of the Channel. The exact location of the saddle point will depend on the details of the local processes and circulation.

What does this have to do with cross-overs? Scotian Shelf water flowing north along the eastern side of Northeast Channel (western flank of Browns Bank) can get onto eastern Georges Bank if at some point it is pushed a little to the west and onto the flow lines that cut across the eastern end of Georges Bank. For example, consider particles released on a line along the eastern side of Northeast Channel (Fig. B3). In the no wind solution (upper panels), the particles follow the circulation and travel north into the Gulf of Maine. There is very little shear in the upper water column and the 5 m and 20 m particles follow very similar trajectories. When the seasonal mean wind is added (lower panels), the 5 m drifters move south across Georges Basin and onto the eastern end of Georges Bank.

This does not mean that the wind is fully responsible for the cross-over events observed during the winter of 1998/99, although it may be a critical factor in the process(es). It appears that small perturbations of the mean flow in Northeast Channel and Georges Basin are required to provide a mechanism for Scotian Shelf water to cross directly onto the eastern end of Georges Bank.

While not directly related to cross-overs, we briefly discuss cross-shelf exchange via Northeast Channel (see Shore et al. (2000) for further details) to place the cross-overs in context. Passive particles were released on a vertical cross-section extending from surface to bottom at the mouth of the Channel and tracked for 60 days. The left column of Figure B4 shows the trajectories of particles that subsequently entered the Gulf of Maine. The right column shows the initial vertical positions of these particles.

In winter, drifters enter the Gulf on the eastern side of the Northeast Channel if they are in the lower half to two-thirds of the water column. The trajectories indicate that, over 2-months, some particles travel as far north as Jordan Basin, while others begin to curl (counter-clockwise) around Georges Basin. Drifters at initial depths above 150 m in winter move closest to the Bay of Fundy. These results are similar to the observed movement of cold Labrador Slope Water into the Gulf in the winter of 1998 (Drinkwater et al. 2000). Cold slope water observed at the mouth of the Northeast Channel in January penetrated into the Gulf and occupied most of Georges Basin by April and the rest of the Gulf by August.

In spring the entrance window is narrower and the particles do not travel as far as in other seasons. This is in contrast to the result for particles released inside the 100-m isobath on Browns Bank where drifters travel further towards the Bay of Fundy in spring than in winter. In summer the entrance window is largest and the particles travel the furthest. One group of particles moves north around Jordan Basin (shown as triangles in Figure B4), a second group circulates around Georges Basin and a third, smaller group, moves west into Franklin Basin north of Georges Bank. Fall (not shown) shows an intermediate state between summer and winter. Over all seasons, particles reaching Jordan Basin tend to originate in the upper eastern part of the mouth, and particles circulating in Georges or Franklin Basins tend to originate from deeper and farther west.

References

Drinkwater, K. F., D. B. Mountain, and A. Herman. 2000. The southward excursion of Labrador slope water off eastern North America during 1997-1998 and its effect on the adjacent shelves. Submitted to J. Geophys. Res.

Hannah, C.G., C. E. Naimie, J. W. Loder and F. E Werner. 1998. Upper-Ocean Transport Mechanisms from the Gulf of Maine to Georges Bank with implications for Calanus supply. Continental Shelf Research, 17, 1887-1991.

Hannah, C. G., J. Shore, J. W. Loder, and C. E. Naimie. 2000. Seasonal circulation on the western and central Scotian Shelf. J. Physical Oceanography. 2000. In press.

Shore, J. A., C. G. Hannah and J. W. Loder. 2000. Drift pathways on the western Scotian shelf and its environs. Can. J. Fish. Aquat. Sci. In press.