Thicker surface cells might not solve that well the Ekman Spiral

[willaguiar]

I was talking to Andrew (Kiss) and he was telling me that the slower surface cell in the 5m case is caused by the fact that this thicker cell takes a longer slice of the Ekman Spiral. That means that thicker cells would have not only weaker velocities, but the angle of the velocities at the surface cell might be shifted too.

So if (over)simplify the effect of the easterlies and apply that logic.... than the Easterlies-driven Ekman transport by a thicker top cell might be slightly more poleward (> $45^o$) than the one from a thinner cell (~ $45^o$).

Maybe such a poleward shift in surface cell speed pushes light surface offshore waters to the shelf, making them more resistant to DSW by buoyancy loss.

[willaguiar]

If we plot the surface cell velocity streamlines (below) we can see that indeed it seems that the surface velocities in the 5m case ( orange) are slightly more poleward directed.

A simple calculation of the Ekman layer depth using $0.1\ m^2\ s^{-1}$ for eddy viscosity, predicts an average Ekman layer depth of ~30m in the shelf latitudes. so perhaps thickening the cell by a 5m levels would be enough to cause this shifting?

Edit 1: Winds around Antarctica are more complex than the easterlies alone, so I likely need to produce similar plots as above, but with winds on top so we can see how a angle shift would alter the surface cell Ekman transport

Edit 2: Other suggestion is to remove the geostrophic speeds from the plot above, and (assuming what is left is mostly Ekman) we can have a more clear view on how the surface Ekman speed direction changed