open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I am trying to calibrate my TELEMAC3D model to observed sea level elevation data using the attached steering file. I am using variable river discharge (attached). The model will run successfully with a COEFFICIENT TO CALIBRATE TIDAL RANGE = 1.0, but will fail if I change this parameter to 1.1, which produces a better result compared to data in shorter model runs. I'm assuming the error results because the increase in tidal range causes the boundary to have too little water. The job file with the error is attached. Could someone please advise on the error and on how I could change the steering file to calibrate the model this way without it crashing when using this river discharge file? Could someone also please explain the function of the COEFFICIENT TO CALIBRATE TIDAL RANGE to ensure I have the correct understanding?

Thank you,
Taylor

Hello Taylor,

Which release do you use? You seem to use a release older than 8.1. You should use the latest release (currently v8p3r0) as there are some new features and bug fixes which may trouble your computation.

Anyway, you have some explanations of the use of the keyword COEFFICIENT TO CALIBRATE TIDAL RANGE (or SEA LEVEL) in the TELEMAC-3D user manual and the TELEMAC-3D reference manual available here:
wiki.opentelemac.org/doku.php?id=documentation_latest

I assume the title of this topic is wrong and should be COEFFICIENT TO CALIBRATE TIDAL RANGE?
This last keyword is a coefficient to multiply the tidal signal on water depth (this last one is the sum of tidal waves). It is a mean to calibrate tide amplitude if your computation does not match measurements e.g.

I cannot see your BOUNDARY CONDITIONS FILE, but I assume the code for your 1st open boundary is 5 4 4 o 5 6 6?
If you want to prescribe water depth, having wet nodes on this boundary is mandatory and using a too big COEFFICIENT TO CALIBRATE TIDAL RANGE may lead to dry nodes. In that case, you can lower the bottom elevation of some boundary nodes. You can have a look at the Fortran file of the TELEMAC-3D tide example (user_t3d_corfon) where the bottom elevation of open boundary nodes lower or equal 0 m for some boundary nodes.

Hope this helps,

Chi-Tuan

PS : I cannot see your job file with the error. You should use an allowed extension for the name of this file (e.g. .txt).

Thank you for your reply. I am using v7p2, the version available to me on my organization's supercomputer, so it is not an option to switch to a newer version unfortunately.

Yes, the topic should have been COEFFICIENT TO CALIBRATE TIDAL RANGE, and yes the code for the first open boundary is 566.

Since I am using v7p2, is it correct that I should instead use the Fortran file t3d_corfon from v6p2, then modify the first and last node values (below) to correspond to my open boundary? Am I correct in my understanding that this will eliminate tidal flats on the open boundary?

! SUPPRESS TIDAL FLATS ON MARITIME BOUNDARIES
! PARAMETERS FIRST AND LAST THAT MAY BE CHANGED!!!
!
FIRST = 250
LAST = 339
!
DO N=1,MESH2D%NPTFR
I=BOUNDARY_COLOUR%I(N)
IF(I.GE.FIRST.AND.I.LE.LAST) THEN
ZF(MESH2D%NBOR%I(N)) = MIN(0.D0,ZF(MESH2D%NBOR%I(N)))
ENDIF
ENDDO

This was very helpful, thank you.
Taylor

Hello Taylor,

When you run a version of TELEMAC, you should use modified subroutines from this release (e.g. v7p2 not another one).
To modify bottom elevation, you can indeed change t3d_corfon subroutine.

The example of modified t3d_corfon for tide example is very specific to this case (only one tidal boundary) and should be modified to fit to your example. If you read the BOUNDARY CONDITIONS FILE of the tide example, you can see that FIRST corresponds to the 1st boundary number of the 566 boundary and 339 to the last one.

I would strongly recommend you to upgrade the release of TELEMAC-3D you use as there have been several bugs fixed since v7p2 (released 5 years ago!) in particular for tide modelling with Thompson boundary conditions.

Hope this helps,

Chi-Tuan