"Keeping in mind that this is a model" It would be more useful to me to have an accurate hindcast, but ...
At the moment I'm looking at the Atl. forcing. The tides/currents/winds and atmospherics[mslp] it seems can alter the amount of water passing through the 'Faroes gap' by almost an order of magnitude[x8??] . The greater the magnitude the easier it finds it to overcome the inertia both in the Greenland/Norwegian basin on the Barents shelf and in the Arctic proper. At it's center I'm guessing, from observation and retroduction, that the current can reach @20mph when peaking, compared to as little as 1mph when fully dampened. Though a more regular peak is @16mph, and I'm only speaking of the G.-N. basin. The further east a persistent low is then the more likely the flow will surge up the slope and into Barent, west means more likely to show up at Fram, or even be diverted back down the east coast of Greenland. To get near to peak flow the low[mslp] needs to hang around for at least three tidal cycles near the new/full moon very close to the ridges. The current [GS] seems to vary itself but after about 5 surges is usually exhausted so a period of quietitude follows, this is often signalled by circular features appearing off the Norwehian coast, these may be slow rising vortices, in which case I'd expect the local fishermen to know about them.
Then there's the inverse, the tidal harmonic is such that even as a surge passes north another is pulled south down the east coast of Greenland and has to make up for the deficit headed north.
Something I'm fairly confident of is that each time a low passes over the Faroes gap a little more momentum is added to the Atl./ Arctic exchange and the currents evolving in the Arctic get more entrenched.
