AWI Science Week poster
My newest plume video:
The goal is to recreate a plume that occurred over the 36 hours between noon on September 13 (2021) and midnight on September 15–16.
Our model setup uses a heavily-modified MITgcm tutorial_deep_convection test case. Some key changes:
- Resolution: $4$ m $\times$ $4$ m $\times$ variable in the vertical ($1$ m at the surface to $11$ m at depth)
- Domain: $594$ $\times$ $33$ $\times$ $99$ grid cells (9 cores)
- Initialisation: Temperature and salt profile from mooring time series (24 h mean starting from noon on September 12, 2021)
- Initial velocities: $0$ m s$^{-1}$ in all directions
- Timestep: $4$ s
- Equation of state: TEOS-10
- Coriolis: $f_0=-1.358 \times 10^{-4}$, $\beta=8.194 \times 10^{-12}$
- Boundary conditions:
no_slip_bottom=.FALSE.andno_slip_sides=.FALSE., with no tracer relaxation sponges - Surface forcing: Buoyancy forcing is applied for 24 hours through a 100 m-wide pseudo sea ice lead; the heat flux is constant at -$200$ W m$^{-2}$ and the salt flux follows Stefan’s law
- Surface temperature limitation (i.e.,
allowFreezing=.TRUE.):Tfreezing=-1.9(i.e., a bit of supercooling) - Non-hydrostatic solver:
cg3dMaxIters=40 - Advection scheme:
tempAdvScheme=33andsaltAdvScheme=33(and no prescribed diffusivities) - Viscosity: Three dimensional Smagorinsky scheme with the default coefficient
