open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,
I am able to get my model to work in barotropic mode (top panel). When I active SALINITY (DENSITY LAW = 2 and use field data to initialize salinity field in the water column), it appears that the model is out of balance, and that the flow had too much random energy which organized itself into spurious non-transient eddies that persists even during flood tide (see bottom panel).

Some of the relevant keywords that I use are:
DENSITY LAW : 2
NUMBER OF TRACERS : 1
NAMES OF TRACERS : 'SALINITY'
INITIAL VALUES OF TRACERS = 33.
AVERAGE WATER DENSITY = 1025.
STANDARD VALUES FOR TRACERS = 33.
PRESCRIBED TRACERS VALUES = 0;0;0
SOLVER FOR DIFFUSION OF TRACERS = 7
SCHEME FOR ADVECTION OF TRACERS : 13
TREATMENT OF FLUXES AT THE BOUNDARIES : 2;2;2

Has anyone run into similar problem in the past or has any idea on what I can try to eliminate these stationary, anomalous eddies?

Thanks.

Edwin

Hello,

I assume you solved your problem by now. Just in case, I note your boundary is set to 0. Maybe it should be 33. ?

Good luck.

Sébastien.

Hi Sébastien,

The boundary condition is being read in from the liquid boundary file. So that wasn't causing the problem.

I believe the problem is that the element sizes are too large in regions with rapid changes in bathymetry resulting in poor horizontal density gradient. The model works better after I further refined the mesh in those regions to better define the shelf.

Edwin

Hello Edwin,

oh yes -- I recommend 8% growth, no more. You can set this when you create your mesh with Blue Kenue (factor of 1.08).

Good kob.

Sébastien.