open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi all,

I am trying to force Telemac2D with TPXO with an open boundary that surrounds some islands. However, in BlueKenue I cannot select the full boundary so I have to split into 2 boundaries. However, when I try to run Telemac2D I get the error:

OPTION FOR TIDAL BOUNDARY CONDITIONS:
GIVE THE SAME NUMBER OF VALUES AS THE NUMBER
OF LIQUID BOUNDARIES I.E., 0

PLANTE STOPPED AFTER AN ERROR
RETURNING EXIT CODE: 2

For reference I am using V7P2R1

Files are attached. I really don't seem to able to solve this so any help would be much appreciated!

John

Hi
This is linked to the fact you splitted your liquid boundary in 2 parts...
you should give 2 values for some keywords

Nevertheless, you could also manualy change the boundary condition file to merge the 2 parts in one and this will works...

Hope this helps

Thanks a lot for your reply. However, if you look in the .cli file the boundary is merged together i.e. it is not separated in the file. Does the ordering matter? I have mine in the order from bottom left going anti-clockwise. Why does it say I have 0 liquid boundaries?

Sorry, I didn't look your files.
Even you use tide on your boundaries, you should give the keywords prescribed elevations and prescribed velocities...
Hope this will solve your problem

Regards

Hi,

It still doesn't work. It must be something to do with the ordering of my boundary or the way the .cli file is setup. If I have a continuous boundary it doesn't recognize that there is any liquid boundaries. However, if I just add one boundary along the west for example it recognizes this boundary and the code runs.

Any ideas?

Update: I was able to get the model running by defining 4 boundaries each with two nodes between (as a closed boundary). I could then alter the .cli file to add the code 5 6 6 at the locations that had 2 2 2. I tried various combinations but only that one worked.

Hi,

This is a bug which has been corrected in latest releases to allow for a circular liquid boundary (in bief subroutine front2.f).
I tested your case, and it works wellwith v7p3r0 and v8p0r0 versions.

I hope it helps.

Regards,
Laurent

By the way, there a newer version TPXO database: v9 at 1/30°, 12 constituents (combining global/regional databases).

After getting my mesh to work with TPXO v9 as Leballeur suggested I developed another mesh with a higher resolution boundary and now I get a very unstable tidal signal at the boundary (see attached figure). Does anyone know what could be causing this?

I also want to make sure I have the projections correct as there seems to be a lot of confusion w.r.t TPXO about how to set it up. My mesh is projected in UTM29N or EPSG:32629). Therefore, in the cas file I have:

SPATIAL PROJECTION TYPE : 1
LATITUDE OF ORIGIN POINT = 61.1540
GEOGRAPHIC SYSTEM : 2
ZONE NUMBER IN GEOGRAPHIC SYSTEM : 29

Is this the correct setup to use TPXO or should I be defining the projections differently?

I have attached the setup so please take a look if there is anything to do with the mesh or boundary that is suddenly causing the TPXO boundary conditions to be ill-defined.

Many thanks,
John

Hi John,

Here are few hints to progress with your case:
- LATITUDE OF ORIGIN POINT is useless here as you're using UTM projection.
- If your mesh horizontal resolution is about 25m, a time step of 30 seconds seems to be a little too high to ensure CFL<1. Try with a time step of about 10s to be sure.
- You mentioned TPXO v9 in your post but in your steering file it's TPXO v7.2. I don't really know which version of the global tidal atlas you're using, but be aware that Telemac is not fully compatible with the latest v9 as there is 1 extra constituent (S1) which is not expected by TELEMAC yet. So you should keep v7.2 or v8 of TPXO global solution.
- Check the locations of the boundary points of your new mesh in comparison to tpxo grid resolution.
- Try to prescribe only water levels (not velocities)
- Try to turn off/on Thompson option

I hope it helps.

Regards,
Laurent