open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I am new to the forum and also to the Telemac, I am modeling a river estuary where part of the sea is included.

So I want to specify two values for the salinity or the density.

Can any of you tell me how can I do that

Hello,

You can use a tracer (see the related section in the TELEMAC-3D manual) and call it SALINITY. Then you can choose add the keyword DENSITY LAW and choose you dependency law between SALINITY and DENSITY to account for variation in SALINITY in your domain.

I you feel advanturous, you can even implement your own DENSITY LAW by modifying the subroutine DRSURR (in TELEMAC-3D).

In TELEMAC-2D you can also include horizontal DENSITY EFFECTS (obviously there is no vertical calculation). Please read through the manual.

Hope this helps.

Sébastien.

Actually I am working in the 2D model. My problem is how to let Telemac knows that there are two parts each with different density or salinity. I have already specified two zones in matisse, but I am not sure which subroutine can help me in defining salinity or density for each zone

I think the first solution is good and more realistic where the interface between the sea and the river is not exactly known.

I will use this solution, thanks for your help

If you are looking at salt intrusion in an estuary where the vertical is well-mixed, then yes: TELEMAC-2D should be sufficient.

There are many ways you can initialise your salinity field:

- the first solution would be to start your simulation with no salt and add salt at your open/offshore boundary. After a few cycles (I would think up to 10 days or so) the salt should naturally have entered your estuary and stabilised to its "natural" state. You can then use the results as a hot-start-file, or PREVIOUS COMPUTATION FILE.
- the second solution, if you already know roughly where the salt is, is to create your hot-start-file with a graphical package, say Blue Kenue. You would have to set WATER DEPTH (or FREE SURFACE), which you can choose constant, VELOCITY U and V and the SALINITY. In Blue Kenue, it is easy to map values to part of your mesh.

Hope this helps,

Sébastien.

Sébastien,

I want to try the second solution. Can I map the salinity in Blue Kenue on the mesh and add this to the selafin file? I'm already doing this with the bottom friction. I mapped several areas with different values and called this mesh 'BOTTOM FRICTION' and added this to the selfin file. This works really nice. Can I do the same with the salinity? And how do I have to call the mesh for TELEMAC to recognize it? 'T1' or the name of the tracer?


Sven

Hi Sven
Of course you could map the salinity on the msh and generate a selafin file in order to generate an initial state.
The main question is about the variable name in order to well recognize it.
I think the name of tracer should work (take care to have the same name and unit). A short test with a small case should give you the right solution if it doesn't work with the name.

Good luck

Christophe,

I tried several names but none seemed to be recognized by TELEMAC. I tried: SALINITY (the same name as a gave the keyword 'names of tracer'),T1,T(1), T,TR1,TRACER1,TRACEUR1

If I give a value at the boundary the tracer takes too much time to get upstream my estuary. So I wanted to add the different values I got from another model and use them as initial state.

We want to estimate ecological value of locations in the estuary (i.e. benthic organisms) based on hydrodynamics, salinity and sediment structure. That's why I want to add salinity to my 2D model.

Any suggestions?

Sven

Hi Sven

In the listing, you have a list af variables read in the previous computational file, maybe this could give you an idea.
First, you try with a real telemac result (which contains the tracer). In such case, the computation continuation should work (I Hope). Then you use the same name for your own previous computational file.

Another option is to use fortran file to understand where is the problem.
For the computation continuation, the subroutines called are bief_suite, suite_serafin and litenr (all in bief). The first tracer should have number 32 and the ALIRE(32) should be equal to 1 if telemac try to read it. After that there is a comparison of the variable name but according to nomvar_telemac2d, the name should be exactly the name of tracer in the sterring file.

Hope this helps

Your welcomed !

Sébastien.