open TELEMAC-MASCARET The mathematically superior suite of solvers

hello！

What is the reason for this problem? How to solve it?

I simulated the flow field of a watershed, and the result showed that the velocity on the boundary was much larger than that inside the region. The velocity on the boundary was about 20m/s, while the velocity inside the region was 0~2m/s. Please consult professionals with relevant experience on how to solve the problem.

Thank you very much!

Hello,

You may provide wrong information of the location where you are.
As you use SPHERICAL COORDINATES = YES and SPATIAL PROJECTION TYPE = 3 (longitude-latitude), you have to set the 2 keywords LATITUDE OF ORIGIN POINT and LONGITUDE OF ORIGIN POINT. Otherwise the default values (equal to 48 and 0) are used and you will have quite different values if you expect to model an area closed to China.
It is recommended to give values around the center of the domain. You can try e.g. 20.4 for LATITUDE and 110.3 for LONGITUDE.

Anyway, I would recommend you to read the TELEMAC-2D user manual available here:
wiki.opentelemac.org/doku.php?id=documentation_latest
in particular section 14.3 for the use of the keyword SPHERICAL COORDINATES + chapter 16 Recommendations, in particular what choice is recommended for TREATMENT OF NEGATIVE DEPTHS (= 2 is now recommended if you do not want to have negative depths in the domain). Using linear elements with FREE SURFACE GRADIENT COMPATIBILITY = 0.9 or less may be more efficient than using quasi-bubble elements (DISCRETIZATIONS IN SPACE = 12).

Hope this helps,

Chi-Tuan

Thank you for your reply!

First of all, thank you very much for your reply. The result has changed and seems to be quite ideal, but I still have some doubts.

According to what you said, I set these two keywords LATITUDE OF ORIGIN POINT and LONGITUDE OF ORIGIN POINT. However, I have not modified FREE SURFACE GRADIENT COMPATIBILITY = 0.9, DISCRETIZATIONS IN SPACE = 12, TREATMENT OF NEGATIVE DEPTHS (error after setting to 2).

After running, the results show that the maximum boundary value is 4.67m/s and the maximum internal point is 1.0m/s.

I think the result is still a little inappropriate, because I looked up the data and found that the maximum tidal current speed in Qiongzhou Strait should be about 2m/s.

I don't know what went wrong.

Hello,

Which release do you use?
If it is recent, TELEMAC automatically changes GEOGRAPHIC SYSTEM to 5 if using SPHERICAL COORDINATES = YES. If not, you have to change it in your steering file.

I suspect an error of location when interpolating tidal constituents.
Can you add tpxo.f file to your FORTRAN FILE, add the writing of the longitude and latitude of the maritime boundary conditions, e.g. adding a loop at line 3686 with something like:

Run a computation only with 1 core and 1 time step and read the output.
If the coordinates are not the one expected in longitude/latitude, that means that the conversion is not correct and could explain high velocities at the liquid boundary nodes.

What error did you get with TREATMENT OF NEGATIVE DEPTHS = 2? If it is written that you have to set other keywords to mandatory values with this choice, you should follow advice:
MASS-LUMPING ON H = 1.
CONTINUITY CORRECTION = YES
SUPG OPTION = 0;0 or 1;0 (the 2nd 0 is mandatory)

Moreover,
MINIMUM VALUE OF DEPTH is useless if H CLIPPING = NO.
THRESHOLD FOR NEGATIVE DEPTHS is useless if not using TREATMENT OF NEGATIVE DEPTHS = 1
Perhaps you would get better result by decreasing SOLVER ACCURACY

If you set SPHERICAL COORDINATES = YES, CORIOLIS COEFFICIENT is not read because it is automatically computed depending on latitude of nodes.

Hope this helps,

Chi-Tuan

Hello, thank you again!

Firstly, I'm using version v7p0.

Second, I'm not using a Fortranfile, and I don't know how to set up a Fortranfile to verify this error (and line 3686?). .

Third, I successfully set the TREATMENT OF NEGATIVE DEPTHS = 2, MASS-LUMPING ON H = 1.

CONTINUITY CORRECTION = YES

SUPG OPTION = 0; 0. The results are shown below.




Finally, I would like to send all the documents to you. Please help me debug them and find out the specific problem.

Thank you very much for your help.

Hello,

Please upgrade the version to the latest one: v8p4! There were 8 new releases since v7p0! Many bugs have been fixed for 8 years!
In addition, no assistance is given for old releases except the latest and main.

Anyway, you should carefully read what I have already written, in particular, at least, change GEOGRAPHIC SYSTEM = 5. If it is not enough, upgrade the release of TELEMAC and see if your issue still occurs.

Chi-Tuan

hello,
Thank you very much!
When I changed the GEOGRAPHIC SYSTEM = 5, the result became much better, in line with my expected idea.
How to upgrade telemac to release v8p4 , is there a tutorial?
If there are any more questions, I will continue to consult you.

Hello,

Good news if your results became better. For sure without GEOGRAPHIC SYSTEM = 5, they could not have been good with your old release.

If you use Linux, you can read wiki.opentelemac.org/doku.php?id=installation_on_linux

If you use Windows, you can read gitlab.pam-retd.fr/otm/wintel

Hope this helps,

Chi-Tuan

Thank you for your reply!

First of all, thank you very much for your reply. The result has changed and seems to be quite ideal, but I still have some doubts.

According to what you said, I set these two keywords LATITUDE OF ORIGIN POINT and LONGITUDE OF ORIGIN POINT. However, I have not modified FREE SURFACE GRADIENT COMPATIBILITY = 0.9, DISCRETIZATIONS IN SPACE = 12, TREATMENT OF NEGATIVE DEPTHS (error after setting to 2).

After running, the results show that the maximum boundary value is 4.67m/s and the maximum internal point is 1.0m/s.

I think the result is still a little inappropriate, because I looked up the data and found that the maximum tidal current speed in Qiongzhou Strait should be about 2m/s.

I don't know what went wrong.