open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I am investigating a particular case of TOMAWAC-ARTEMIS chaining in mode 2. The issue occurs with Westerly waves, i.e. ARTEMIS wave direction varies about 0.

The problem arises when wave direction at a particular element crosses 0 and the resultant wave height drops significantly, like the spectrum gets truncated.

I attach some results to illustrate the problem:
TOMAWAC OUTER results:

ARTEMIS result.

The project setup files:

Dropbox link: www.dropbox.com/t/wsdhSUWA9WnpjInX

Any advice would be greatly appreciated!

Best Regards,
Costas

Hello,

My colleague looked at your case and she found out that putting
minimum angle of propagation = 0.1
and maximum to 360,
the coupling works (even if it is not optimised anymore).

hope it helps

Thierry

Hi Costas

Following from Thierry's reply, have you checked whether you had the same problem without the coupling? i.e. when imposing a boundary wave centred on 0, and limiting the direction sector like you have done (from what I understand - I have not looked at your files). I looks like it may be that the spectrum is not read in properly in those cases, but only the 0+ part of it. Can you confirm?

Out of interest, why are you limiting the directions in ARTEMIS? If there is no energy in a sector, ARTEMIS will not run a simulation for that sector...

With best wishes - Noémie

Hello Noémie,

Thank you for your insight. Having revisited my files for an older project using CHAINING mode 1, I discovered that I have explicitly noted that MINIMUM/MAXIMUM ANGLE OF PROPAGATION is not used with TOMAWAC chaining. However I haven't noted that it creates problems and should be avoided.
I can also confirm that NO CHAINING works fine with MINIMUM/MAXIMUM ANGLE OF PROPAGATION.Yep, I got it now. Looking into the code it is clearly stated that the spectrum is discretized in equal energy directional bands...

Best Regards,
Costas

Hello Thierry,

Thank you for looking into it. Indeed it solves the issue. Why would you say it is not optimised anymore? Is it because you can no longer confine the directional spectrum?
In my test run with just 4 directions I got the following discretization:The energy spectrum seems efficiently resolved.
Best Regards,
Costas

I said it wouldn't be optimized because i don't know artemis very well, i thought it was a way to limit calculation to some angles but seeing what Noemie said i was wrong.

kind regards