open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I would like to simulate a tracer (sediment) release in 3D with TELEMAC3D using a stationary velocity field in constant depth, and no introduction of mass of water. So basically what I want is using the T3D hydro results as a contant convective field only (like in SUBIEF in the past).

First I need to reproduce a 3D stationary current field. To this end I have set:
SCHEME FOR ADVECTION OF VELOCITIES = 0
SCHEME FOR DIFFUSION OF VELOCITIES = 0
with constant depth and velocity boundary/initial prescriptions (in CONDI3DUVW and BORD3D).
Friction law is Manning with m = 0 in order to remove bottom friction.

The velocity field is then ok. However, as I still had free surface variations I have commented the call to WAVE_EQUATION in the main routine TELEMAC3D (I was not too sure about that).

For the sediment release I have set:
VELOCITIES OF THE SOURCES ALONG X=0
VELOCITIES OF THE SOURCES ALONG Y=0
VELOCITIES OF THE SOURCES ALONG Z=0
in order to prevent any introduction of water into the domain.

The tracer release seems to be ok but I still obtain some strange free surface variations, I don't know why. I am probably mistaken somewhere in the method.

Could you please tell me how I could manage to achieve this goal?

Thank you very much!
JBS.

Hi JBS,

I may be off base here, but it seems to me for friction law, you would use LAW OF BOTTOM FRICTION = 0 to remove bottom friction (section 5.8.1 of T3D user manual)? A Manning's n of 0 would effectively mean division by 0, if I'm not mistaken.

Hope this helps,
André

Thank you André.
Yes, the friction may be deactivated this way. However, the Manning friction term is precisely proportional to m^2. So m = 0 should be ok (you were probably meaning Strickler).
JBS.

Hi!

Do you use T3D+GAIA or T3D/SEDI-3D?
How do you insert your sediment? Using a source with water discharge? When using:

VELOCITIES OF THE SOURCES ALONG X=0
VELOCITIES OF THE SOURCES ALONG Y=0
VELOCITIES OF THE SOURCES ALONG Z=0

you just insert a source without momentum if I am right.

Another explanation could be density effects due to sediment concentration?

Kind regards
PL

Dear Pilou,

Thank you for your reply.
I am using T3D-SEDI3D (v7p3 in this case).
That is true, there is zero momentum introduced, however there is still a water volume coming in. By reducing the flow rate, the free surface instabilities decrease, which is a good thing. However, I would like this source to be zero numerically whereas the tracer is still released. In that way, the free surface should not be affected and should remain still. That is why I commented the call to the WAVE_EQUATION routine in TELEMAC3D.f, but it is apparently not the right thing to do. Water is introduced in the domain somewhere else...
If you have any clue...

Thanks again!
JBS.

Hello,

If you don't want to introduce any additional water volume then avoid sources. A possibility would be to use source_trac.f (or user_source_trac.f depending on the version) in order to insert tracer quantities as a momentum source term. I don't know if it exists examples to do just that, but you could otherwise look at what is done for tracer decay for example and adapt it so that you can instead add tracer quantity per unit time at given nodes.

There might be more experienced users who also have nice tips & tricks to share :)

Good luck!
PL