open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everyone,

This is my first post to this forum, and I am a bit nervous

I am trying to run a 3D tide model, but have been unsuccesful so far. Here are a few things that I would like to ask for clarification

1. I created the mesh in Blue Kenue (BK). Knowing that Fudaa won't accept Geographic coordinate, I change the coordinate system projection in BK to UTM. However, when importing .slf and .cli files in Fudaa, the following warning appears : No errors found "Cartesian grid is not supported". Why is this warning still appearing eventhough I have changed the projection for all files in BK?

2. I have read a lot of posts in the forum regarding tidal modeling, and I have set my parameters in FUDAA according according to the discussions/steering files shared in this forum (many thanks to Gourish). When I started the simulation in the DOS windows, I get the following notes (Error I guess)
2.1 Maximum number of Elements around a point : 9 / maximum number of points around a point : 9 (Is this an error warning?)
2.2 EQUNOR <BIEF> : Exceeding maximum iterations (does it mean my boundary is wrong?)

3.I have quite a large domain for my simulation. For the open boundary, I used open boundary with prescribed H, and open boundary with free tracer. May I ask what is the use of tracer code? (The manual didn't mention much about this part)

Below are my steering file, geometry and BC file.









Thanks!

Anas

Hello,

A few answers:

Maximum number of points or elements = 9 is not an error, just a message.

maximum iterations in EQUNOR : you can try instead a conjugate gradient :

SOLVER FOR PROPAGATION : 1

but there is probably a problem at the beginning of the computation, try a smaller number of time steps to see where it is and when you get a problem with a maximum of iterations of the solver.

With OPTION FOR LIQUID BOUNDARIES : 2;2 you are using Thompson boundary conditions everywhere, is it what you want ?

The code for tracers is like velocity : 4 is free and 5 is prescribed, but you have no tracer, like salinity or temperature, in your model so no problem.

Regards,

Jean-Michel Hervouet

Dear jmhervoue,

Thank you for your reply and advice. I have tried to modify several setting / input parameters based on your suggestion and also from discussions in this forum. However, I am still having trouble to run my model.

1. at the beginning of the simulation, if I keep on getting GRACJG <BIEF> Exceeding Iteration or EQUNOR <BIEF> messages, then I know it is as warning, and I need to check my parameters.

2. SO I did managed to eliminate those messages. However, upon running the simulation in the DOS terminal, the simulation stopped after the line "END OF TPXO SIMULATION".

3. I am beginning to think that my geometry file is the source of the problem, but I did not get any warning regarding my .slf file.

I have changed the setting / type of open boundaries several times to see if it has an effect, but nothing happens. I have also tried several configurations in the FUDAA with no luck. I am stuck with this problem for quite some time now. I would very much appreciate any suggestions/advices on how to improve and run my tidal model.

I used Delft3D before this, and just moved to Telemac about a month ago for my research. From my brief experience using Telemac, I already like this software since it is so flexible in term of mesh generation and input parameters. Hopefully I could solve this issue and move on using this software.

Attached are the latest .slf , .cli, and .cas files that I created.

Thank you in advance for any help/suggestions!

Hello,

Are you sure that the computation stopped? We see no crash and as you have a period for listing printouts set to 600 maybe you'll have to wait some time before seing the following print on the listing.

To be sure you can either choose a period of 1, or if it crashes somewhere add:

DEBUGGER = 1

in your parameter file, so that we see the main steps of the time loop and where it stops.

With best regards,

Jean-Michel Hervouet