open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I created a small Tomawac domain, in which I use a constant in space but time-varying wind and wave forcing.
The model runs ok with a time-step of 20s and Courant condition is fulfilled, but I keep having some instabilities in the wave outputs, when wind gets stronger/wave higher (see figure).
I have similar instabilities if the model is offline-coupled with tidal current and elevation (read from a Telemac slf file).

I tried to improve it by:
*reducing the time-step
*using other wind generation formulations
*using more iteration for the source terms

But this did not improve the model.

Would you have any other recommendation please?
Regards,
Claire

Hello Claire

I can't run your case i got an erroneous arithmetic operation and a sigsev on another compiler. It seems we are not using the same version of telemac.

But i do not understand what you do in your limwac subroutine you read a H Fpic Teta but do not use them you have to convert them in a spectrum by calling a speini because when you set F(NBOR(IPTFR),IPLAN,IFF)=FBOR(IPTFR,IPLAN,IFF) line
you have nothing you can rely on in FBOR(IPTFR,IPLAN,IFF)

Hope it helps

Thierry

Hi Thierry,

And thank you for this quick reply.
Yes, I made a mistake in the name of the wavewind files when I copied it.
I have amended it, and hopefully, the attached case will work.

I believe I do call SPEINI function (line 202) in the fortran file. I have checked, we do go into this loop. Do I have to put this call after having read the wave variable (H, freq, etc)? I tried both, and I believe both work ok.

And the variable F is not empty. But not sure how to check whether it is sensible? Would you have any recommendation to check that?

Thank you,
Claire

Hello Claire

I had seen the problem of wave wind files before and had corrected that. The problem is that since you do not have the same number of arguments to limwac I know we are not using the same version of tomawac, which one are you using ? It is always better to use the last version as some bugs may have been corrected.

Yes you have to call speini AFTER you read your variables otherwise the spectrum does not take your H Tp and teta into account.

I am not talking about F but FBOR the value of F on the boundary that you construct in speini.

another thing i think that you meant if teta1l.le.0 instead of le 180

Thierry

Hi Thierry,

I am using Tomawac v7p3r0. Do I need to install a newer version?
Also, I am now using a newer (v7p2) version of limwac subroutine. The link on the telemac website to download limwac.f is not updated, that's why I was using limwac v6p3.

And no, I meant teta1l.le.180, as here I want to change the "direction from" to the "direction to". And as I want to get an angle modulo 2pi, I did it that way. Not very neat but I believe it works.

Claire

Hello Claire

If you are using V7P3R0 version you have to use the limwac that corresponds because variables are not the same. There are not the same number of variables.
Normally if you are using V7p3r0 you should have the corresponding limwac in the sources of your computer but anyway i attach the limwac for version v7p3r0.
As for direction i understand. I didn't think you wanted that.

regards

Thierry