open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I am using TELEMAC-3D coupled with TELEMAC-2D to model the Gironde estuary. In the steering file, I specify the ORGINIAL DATE OF TIME to consider the tide in my model. At the end of the simulation, I obtain two result files, one 2D and one 3D.

I use Bluekenue to review the results obtained in these files. I can see the start date that I set by making an animation with the 2D results file, but I am unable to do the same with the 3D results file. Indeed, when I want to animate one of the variables included in the 3D results file, there is no information on the start date.

Do I need to specify somewhere so that the start date is also included in the 3D results file? Can you help me, please?

Thank you,

Mathilde

Hi
What do you mean by T3D coupled with T2D?
As far as I know, there is no internal coupling available for those modules in the Telemac system...
In Telemac-3D steering file you could specify the date and time, as in Telemac-2D...
Regards

Sorry, I'm new to Opentelemac and I may have expressed myself poorly. What I meant to say is that when I run a simulation on my 3D model, the 2D model also runs, which is why I have both 2D and 3D result files at the end of a simulation.

That's what I do, I specify the date in the Telemac3D steering file. However, for post-processing, the date doesn't seem to be present in the 3D result file. I hope I've explained it better this time.

Thank you.

Hello Mathilde,

Which release do you use?
Can you provide your steering file?

Looking at the tide example, running t3d_tide-jmj_real.cas where the keyword ORIGINAL DATE OF TIME is used, I can see that this date is written in both 3D RESULT FILE and 2D RESULT FILE, see the beginning of the listing when running run_telfile.py scan r3d_tide-jmj_real_gen.slf :

Interpreting command line options
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

~> Generic info

- Title: TELEMAC 3D: tide
- Date: 2011-10-22 00:00:00
- Format: SERAFIN
- Endianess: BIG_ENDIAN
- Precision: Single precision

~> Mesh info

- Number of dimensions: 3
- Element type: prism
- Number of points: 26246
- Number of elements: 43850
- Number of points per element: 6
- Number of planes: 11
- Number of points: 2386
- Number of elements: 4385
- Number of points per element: 3

+> Coordinates

- X offset, Y offset = 0, 0
- X range [184123.40625, 206226.0]
- Y range [136035.5, 160011.0]
- Z range [-67.40676879882812, 10.710958480834961]

~> Parallel info

- No parallel information

~> Boundary info

- No boundary file given

~> Data info

- Number of records: 51
- Time range: [0.0, 45000.0]
- Number of variables: 4
- Name: ELEVATION Z Unit: M
- Name: VELOCITY U Unit: M/S
- Name: VELOCITY V Unit: M/S
- Name: VELOCITY W Unit: M/S


My work is done

---

I cannot use Blue Kenue, I am not able to say if it may come from this tool or not. Please use another post-processing tool like Paraview to see if you get the same behaviour.

Chi-Tuan

Hello,

I am using v8p2r1, and here is my parameter file. The issue doesn't just come from Bluekenue. In fact, I also tested it with a Python code using a function I found in the Opentelemac documentation. With the following code:

from os import environ, path, getcwd
from data_manip.extraction.telemac_file import TelemacFile
import matplotlib.pyplot as plt
from postel.plot2d import plot2d_scalar_filled_contour, plot2d_triangle_mesh, plot2d_vectors
import numpy as np

# Fichier de résultats 3D Sélafin (à modifier)
file_name = path.join(getcwd(),
'data',
'RESULT-results_2d_18000s_22_02_2024_450s.slf')

# Ouverture fichier
res = TelemacFile(file_name)

date = res.get_mesh_date()

print(date)

I obtain : [2024 2 22 0 0 0]

But when I run the same code with the 3D results file I obtain this : [0 0 0 0 0 0].

I don't know if I am very clear.

Thanks for help

Mathilde

Hello Mathilde,

Next time, be careful with the extension of the file you upload. All are not allowed (e.g. .exe are forbidden, but .txt, .slf, .cli and .cas should work). I was not able to read your steering file.

And as written in the rules of this forum, please include as much information as you can, otherwise it may be difficult to understand or reproduce your issue. The key issue seems to be 3D results file in parallel (this last information was crucial for this bug).

Chi-Tuan

Hi, hopping in!

I can confirm that when running any of my Telemac3D models (parallel in version v8p4r0), that my res2d file has a correct original date, while the original date in the res3d file is always empty. See also attached. I haven't been able to figure out what causes this.

Interestingly enough when I run t3d_tide-jmj_real_gen.cas (sequential) this issue does not pop up, and my res3d file has a correct original date.

Hello Jurjen,

Thanks for having given every detail needed to reproduce this issue. I think I have found the bug, it is due to gretel (which merges the split results file into 1 single one).
If you have a look at $HOMETEL/sources/utils/gretel/gretel_autop.f program, you can find 2 lines:

Can you try to delete or comment them + compile the code and tell me if the issue still occurs or not in parallel please?

Chi-Tuan

Hello Mathilde and Jurjen,

Can you tell me if the suggested fix solved your issue of date, so that it can be pushed for next release please (minor release v8p5r1 which is to be created soon)?

Thanks,

Chi-Tuan

Hi!
I missed you earlier response, so thanks for the reminder.

I tried testing the fix in the user_fortran folder that does not seem to be possible. Testing your suggested recompilation now.