open TELEMAC-MASCARET The mathematically superior suite of solvers

I am trying to model gravel beach evolution under wave action. Do I need to import tomawac outputs into Telemac 2d and then couple Telemac2D with Sisyphe? Or can I connect Tomawac and Sisyphe directly?

Also when I attempt to view the flow velocity in the tomawac results file using Fudaa-Preparo I get an error message about bugs in the programme and reporting it to improve the system. Is there a later version of Fudaa since 11RC7?

Hi
About your first question, i'm not sure of you really want to do. Did you plan to run sisyphe directly coupled with telemac? If so, you should declare the tomawac result in telemac.
If you wants to run sisyphe uncoupled, i think you could declare tomawac result directly in sisyphe.

For Fudaa, the latest version (beta?) is 1.1RC25 but i'm quiet sure there was an official version higher than yours 1.1RC13 at least.
But for the tomawac result i didn't tried with those version.

regards

The overall goal here is to model beach erosion as caused by cross-shore wave action. (normal 90 degree incident waves no long shore drift).I have switched from Fudaa to Blue Kenue

Now about coupling the modules. I think I now understand it correctly. Tomawac models the wave state, then telemac2d takes the wave forces and wave heights and uses them to generate wave induced currents. Then sisyphe takes the wave induced currents and uses them to drive sediment transport.

I have got the tomawac part to run and I have also set up the telemac file. However when I run the programme it goes into some really long calculations at iteration zero. I have attached a print screen of my command prompt log during the calculation. My calculation is either stuck or taking very long and I am not sure why.

Hello,

You are using the solver "minimum error" in Telemac-2d, I find it rather strange, it could be that you have also the primitive equations option.

Try the wave equation option with conjugate gradient:

TREATMENT OF THE LINEAR SYSTEM : 2
SOLVER : 1


If it does not work better, then check that it works without wave driven currents, and then look if there are tidal flats in your domain. It could be that Tomawac gives you wave induced currents on dry zones. For wind there is a minimum depth to apply wind, but nothing is programmed for wave induced currents, this would have to be implemented in subroutine prosou.f in library telemac-2D.

I hope this helps,

Jean-Michel Hervouet

I have been avoiding this issue by using TREATMENT OF LINEAR SYSTEM: 1

I do not understand exactly what the solver is or what the seven choices in solver options described in the telemac2d user manual 6.0. In the telemac2d formula document I see that it references a publication of yours which discusses the iterative solver methods for linear systems.

The document is: J.-M. HERVOUET Iterative methods for solving matrix systems. Report HE-43/93.49 EDF.

I have been unable to find this document online. Do you know where I can access this and any other documents that would help to understand the "Solving of the Linear System"?

Hello,

The iterative methods are also explained in my book "hydrodynamics of free surface flows" (the EDF report is now out of print but just explained the different iterative methods).

The only thing to know is actually that :

TREATMENT OF THE LINEAR SYSTEM : 1 solves a linear system coupling depth and velocities, and this system is not symmetric, so the solver must be adapted to this : e.g. normal equation, minimum error or GMRES.


TREATMENT OF THE LINEAR SYSTEM : 2 simplifies the previous system by solving first the equation of depth (which is the continuity equation), and it gives a symmetric system, hence all methods, e.g. the conjugate gradient which is the simplest, will work.

We find that the second solution is much faster and stable than the first one.

With best regards,

Jean-Michel Hervouet

I see. I am now in the process of getting your book and the journal article.

I am using telemac2D to model set of gravel beach GWK experiments. I need to evaluate telemac's ability to model the erosion experienced on the strip of beach in the wave flume. For this reason I have been reading the journals that detail the formula that (tomawac telemac2d and sisyphe) use and also experimenting with different computational parameters to see which gives the best result.

When I get more questions I will post them here
Thank you very much for your help.