Winds cosinus

[Felixstd]

To try to break the sea ice in the middle of the domain and to get rid of the instabilities coming from the corners, we added a cosinus part to the wind which is given by $-0.1cos(2\pi*x/(Nx))+10$. The slope is then 0 in both extremities of the domain.

Even with that, the ice seems to break in the corners for low $\mu_b$ but not for $\mu_b > 1$ as opposed to without the cosinus and only constant winds everywhere.

For $\mu_b = 1$ after 3 days:

For $\mu_b = 0.5 after 19 hours: