open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear all,

I have a project of suspended sediments transport, in open water. I have a stable hydrodynamic (telemac3D) model coupled with tomawac. I used sisyphe for a time and chose to turn to Gaia but since I changed I have an error I had never seen before.

I started with a very simple simulation and I inspired mostly for hippodrome 3D example. It is just to test the coupling of Gaia T3D and Tomawac.

Simulation starts well but stops at the end after END OF TIME LOOP. I have an error I do not understand (cf file attached). Has someone already encountered this kind of problem ?

I must precise that since I changed to Gaia I had to modify the source file telemac3D.F line 302 and comment Allocate(num_isusp_icla(1)) (because apparently this table was allocated before with the coupling), hope it did not mess things up.

Hope someone can help me,
Léa

Hello Lea,

Which release do you use?
Can you change the comma between TOMAWAC and GAIA for the keyword 'COUPLING WITH' to a semicolon as suggested by the TELEMAC-3D user manual? COUPLING WITH = 'TOMAWAC;GAIA'. And rerun your computation?

Anyway, changing some parts of the source code may sometimes (often) be bold when you do not know so much about what you are doing: in your case, you comment an allocation command for a variable, which, at the end is to be deallocated (the error message says that a variable you try to deallocate is unknown), but cannot as it was not allocated. This case happens when you do not couple with GAIA (that is what is written as if condition just before line 302). And I think, as you use a comma rather than a semicolon, TELEMAC-3D only couples with TOMAWAC and not both TOMAWAC and GAIA. You can have a look at the listing file and tell me if you can read a call to GAIA.

Hope this helps,

Chi-Tuan

You were right, I changed for the semi-colon and fixed back my changes in the source to the original. I also changed " to ' and Gaia is finally called and seems to work !

Thank you so much for quick answer,
Léa

PS : I work on V8P1R0

Dear Léa,
May I suggest something? Your application is extremely ambitious and difficult. This is why I'd suggest you to proceed step-by-step, starting first from a good hydrodynamics solution (T3D), then move to only waves and currents (T3D+TOM), then start some simple tests only currents and sediment transport (T3D + Gaia), then include waves, and at the very end of the chain, couple everything with Nestor.
I know that we always want to attack the whole problem from the start, but this kind of multi-physics/multi-scale approach only succeed with you'are advancing 'in blocks', if not, most of cases is failure guaranteed.
All the success with your nice application,
Best wishes,
Pablo

I will do this way, you are right it is better to check every step before bringing every thing together to prevent mistakes.

Thank you,
Léa