open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi ,

I am working with telemac2D in v6p2 version for hydrodynamic simulation
I built a mesh and boundary conditions with BK.
Tidal conditions are prescribed with TPXO model.

I used a similar case to fill my case file.
When i'm running the code (with debugger enabled) the simulation seems to be stuck after "CALLING CHARAC".
I tried to rebuild several times my mesh but i got always the same result.
I've attached my steering and input files.

If someone could help ...

Regards

Matt

Hello,

This is strange, unless you have an infinite velocity somewhere. It seems that the characteristics called by Thompson work, so you could try :

FORME DE LA CONVECTION : 3;5

and see what happens, if you see a stop with too many iterations, it is definitely a large velocity somewhere. If it works we'll have to do the test with your files.

You can also try to set :

OPTION POUR LA DIFFUSION DES VITESSES : 1

as 2 is dangerous on tidal flats.

Unfortunately I'll be back at work only on the 8th of April.

With best regards,

Jean-Michel Hervouet

Thank you for your quick answer
I've just run a test with your parameters (FORME DE LA CONVECTION and DIFFUSION DES VITESSES). The program stop with an error message concerning iterations of VF scheme. (see attached output log file)
So according to you , it's due to a large velocity somewhere ?
Do you know how i can fix this ? is there a problem with the mesh ?

Best Regards,

Matt

I checked the initial conditions (in the result file) and there are high velocities (around 2 m/s) in small depths (< 1m).
Is there a way to set a minimal depth for hydrodynamic computation ?

Regards

Matt

Hi Matt,

Prior to answer you, I have a few questions to ask:
- what is the reference of the sea level in your model? Chart Datum or Mean Sea Level? If it is CD, please read the following post (there was a bug in version 6.2):
www.opentelemac.org/index.php/kunena/16-...h-v6p2?start=20#6118
Anyway, can you just try to start your simulation with a "constant elevation" initial condition just to see if your problem comes from the initial conditions?

As told in the paper "Use of tidal harmonic constants databases to force open boundary conditions in TELEMAC" from the last TELEMAC-MASCARET User Conference available at the following link www.opentelemac.org/images/clubu/2012/TE...ceedings-2012-R1.pdf
at the end of section III or in the slides available in the post:
www.opentelemac.org/index.php/kunena/16-...h-v6p2?start=10#6080
when possible, you are advised to use a local or a regional solution from OSU rather than the TPXO global solution, in particular when tidal flats are present in the tidal solution.

If you want to implement some modifications, you have to copy the whole TPXO module available in the telemac2d directory (please see in particular the BORD_TIDE_TPXO subroutine for the boundary conditions and CONDI_TPXO subroutine for the initial conditions).

Hope this helps,

Chi-Tuan

Hi Chi-Tuan,

Thanks for your help.
The sea level reference is CD.
So i tried to use the TPXO and CORFON module you attached in the post (CD2MSL) but it doesn't change anything.
I also tried to start a simulation with a "constant elevation" but i got the same results. So it doesn't come from initial conditions ?
I'm gonna read the docs you supply.

Thanks

Matt

I will try to use regional OSU (Pacific Zone) instead of the global TPXO solution. Is this the same way to configure it in the steering file as TPXO ?

Thanks

Matt

Hi Matt,

You can use every regional or local solution from OSU as you do with the TPXO global solution. You only need to change the name of the two binary databases for tide files.

Have you tried a "relevant" initial elevation corresponding to an elevation in the domain at the beginning of your calculation, or not so far or a random elevation?

Regards,

Chi-Tuan

Hello Chi-Tuan,

I tried regional solution from OSU and i don't have problems anymore in shallow water near the coast ( velocities up to 2 m /s) . but my run still crash with the same error ( too many iterations with the VF scheme)
When i'am looking at the initial conditions (see attached result file) i have velocities in good range ( from 1e-6 to 0.24 m/s ) so i dont see where it's crashing (high velocities)

Regards,

Matt

i forgot the attached file. sorry