open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi

I'm trying to run a 3D simulation (v7p1r1) and I'm getting instabilities in the free surface. The model has a fresh water input with salinty as a tracer, to investigate stratification.

I've run the model in 2D and it works fine. I've run in 3D without the tracer and I still get the instabilities. I'm guessing its a numerical scheme problem?

The model is forced with time series of elevation and velocity at the two open boundaries at either end with a flow rate for the river source.

Any suggestions?

Kind regards

David

Sorry find attached the steering file and liquid boundary file as well.

Hello,

You can try the following things (separately) :

IMPLICITATION FOR DEPTH = 1.
IMPLICITATION FOR VELOCITY = 1.

COEFFICIENT FOR HORIZONTAL DIFFUSION = 1. (think that to be numerically effective this diffusion must be of the order of 0.5 * velocity * mesh size).

FREE SURFACE GRADIENT COMPATIBILITY : 0.

I would also investigate the effect of the Nikuradse law, first by setting :

VERTICAL VELOCITY DERIVATIVES = 1 (we do not have much feedback on option 2 though it seems a good idea)

and then by trying another law (all other laws do not depend on the velocity profile at the bottom).

With best regards,

Jean-Michel Hervouet