open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi

Is it possible to set the boundary as a solid boundary which reflects a tidal wave in telemac2d? If so what values do you assign to LIHBOR, LIUBOR & LIVBOR and which parameter controls the boundary damping?

Thanks

Fay

Hello,

A solid boundary with 2 2 2 for LIHBOR LIUBOR LIVBOR will reflect tidal waves but there will be no damping, we do not have such damping boundary conditions, unless we put a turbulence model with constant but variable in space diffusion, with a lot of diffusion at the boundary (this is just an idea, we never tried this).

Regards,

Jean-Michel Hervouet

Thanks Jean-Michel

So if I use 2 2 2 without damping, say the tidal wave approaches the boundary from the normal direction with a velocity of 2 m/s should the wave bounce off the wall and return in the direction from which it came at 2 m/s?

Regards

Fay

Hi

In an attempt to better understand how tidal waves in TELEMAC2D are reflected at the boundary I have tried to model a seiche in a swimming pool (attached are my files). The results of the first 18 seconds are as I expected, however after this the results are not what I would expect from theory; the pattern is different and the free surface elevations are very high (for example 2 m in a water depth of 2m, the boundary wave is 0.2m ). I will try and attach the results file but it might be too big. Can anyone tell me what is wrong with my set up, or if you have a successful seiche set up, would you mind sharing it please?

Thanks

Fay

Hello Fay,

I had a quick look to your files. You can use a SUPG option of 0;0 because the FREE SURFACE GRADIENT COMPATIBILITY will do the job of preventing inf-sup conditions, also the time step seems very small. The most important remark is that you use the incident wave feature. I was just thinking that this was obsolete and about to propose to remove it in the next versions... As a matter of fact it should be superseded by the Thompson boundary conditions, in which case you would have to give in subroutine bord.f the values of depths and velocities at the entrance, according to your incident wave, as we do for tidal conditions. The theory in both cases is about the same and based on characteristics, but Thompson happened to be better on cases like the one called "gouttedo".

Otherwise I have no experience with seiches caused by waves, I knew only seiches caused by wind or atmospheric pressure.

With best regards,

Jean-Michel Hervouet