open TELEMAC-MASCARET The mathematically superior suite of solvers

I am currently trying to simulate a case of an estuary. So, I considered that the boundary condition at the sea side as imposed head and imposed salinity (not standard by MATISSE, normally the imposed tracer is imposed in MATISSE when an imposed flow is given).

Do you think that this a correct boundary condition.

My opinion is that I have to make it this way because of the presence of sea water which suppose to intrude into the river estuary

Is there anyone made similar simulation or is there an example from TELEMAC about this, actually I want to compare the results of others with mine as I do not know whether what I got is correct or not.

Attached is the steering file, a fortran file for the initial condition, an image showing the initial condition, an image showing the result after 40 days simulation, the flow field, the geometry file and the boundary conditions file

sorry the geometry file and the conlim file are here

Hello,

It is logical to impose a free surface elevation and a constant tracer density as boundary conditions. In such kind of models, we proceed exactly by the same manner. However, there is 2 important remarks for your case:

1- To impose the water elevation and a constant tracer density, you have to verify that the boundary is located far enough of the embouchure of the river. This will ensure that the velocity field is dumped enough in the vicinity of the boundary compared to the one in the river/estuary. This condition is necessary; otherwise, you need to prescribe a Thompson boundary condition which is not recommanded in your case. Thus, the solution, in my opinion is to make an extension of your madel by adding a sea area with an increasing coarse mesh. This will permit to Telemac to impose smoothly the prescribed BC.
2- Based on my humble experience, it is important to include tide effects for this kind of application. Moreover, I don't know if you have some measurements, but, very often, the salinity is not constant and it decreases due to the big quantity of incoming fresh water.

I hope that this is helpful

Riadh

Many thanks for the your usual help.

Actually I am modeling an estuary at the Mediterranean sea, and the tides are very few therefore I ignored the impact of the tides.

In addition I do not have enough information about the sea bathymetry, therefore I have stopped just at the end of the river. And I am using Thompson boundary condition, because without it I usually got errors.

Do you think that this is the reason of getting very high velocity at the downstream boundary and non-normal flow field.

Also do you think that the results are logic (the images attached in my first contribution)

Hello again,

I can not have a straight out opinion whether the results are logic or not; however, it seems that the huge recirculation you got at the downstream is tightly linked to the boundary condition you are imposing.
I think that it preferable that you extend the model to include a small area in the sea direction, even with an approximate bathymetry. This will stabilize the model and dump the spurious effects of the boundary condition. You can find approximate description of the bathymetry of coastal area when you look at a classic map or surely with internet.

My best regards

Riadh

Hello,

I will try to extend the domain to include part of the sea. But, I think I have by any mean to set the salinity of the part of the sea to be constant during the calculations, otherwise most of it will be substituted by fresh water after short time. Do you know how can I do this

Hello,

The most important is that your constant salinity condition is satisfied rigorously exactly at the boundary. The area in the vicinity of the estuary mouth will undoubtedly be affected by the fresh water coming from the river. This is more than normal.
For a steady state simulation, you will see that the salinity increases smoothly until it reaches the prescribed value at the boundary.

Riadh

Hi,

Do you know if there is one of the examples in the validation document of Telemac or on the basic examples which related to salinity.

Hi

You can have a look on the test cases called "Point discharge without diffusion" or "Point discharge with diffusion". These test cases show how to include tracer into the simulation.
However, if you want a case with salinity, you can see the validation of Telemac3d and particularly the case called "stratification". The keywords used for Telemac3D and Telemac2D are almost the same.

With my best regards
Riadh ATA