open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I run a 3D Telemac model and I would like to calculate maps of the bed shear stress. What is the best way to do it?

I can use this formula: τ=0.5C_f ρV_mag^2 Should I use the depth average velocity magnitude or velocity in the layer near the bottom?

Best regards,
Tatiana

Hi,
there are more possibilities but probably the easiest way is to write out the friction velocity US via VARIABLES FOR 2D GRAPHIC PRINTOUTS = 'US' and to calculate the bed shear stress in a postprocessing step. The friction velocity is calculated from the depth-averaged values unless you use the Nikuradse friction law.

Hope this helps,
Clemens

Hi,

I get different flow-rates for the exact same simulation. The only difference is that in the red one, I included 'US' for 'VARIABLES FOR GRAPHIC PRINTOUTS' in the steering file.
Can someone comment about what can be the possible reason?

Thanks in advance!

Hello Nitesh,

Which release do you use?

Have you written US for VARIABLES FOR 2D GRAPHIC PRINTOUTS or for VARIABLES FOR 3D GRAPHIC PRINTOUTS?

With 2D, it should write the friction velocity whereas with 3D, it should
write Stokes velocity along the x axis.

Do you couple TELEMAC-3D with TOMAWACT3D?

Chi-Tuan

Hi,

I am using TELEMAC-2D. In the steering file, it is 'VARIABLES FOR GRAPHIC PRINTOUTS' only, not 3D or 2D.

Regards,
Nitesh

Hello Nitesh,

As you have posted in the TELEMAC-3D section, I thought you ran TELEMAC-3D.

Anyway, can you send all your files to reproduce your issue, in particular the steering file and potential user Fortran files + tell which release you use please? In addition tell exactly what you drew and at which location and time step as I cannot use Blue Kenue.

Chi-Tuan

Hi Ci-Tuan,

ohh.. I posted in telemac-3D by mistake.
I am using v8p2 of telemac2D.
Attached are the files which I used for the simulation (note that they are not the same as my previous post, so don't get confused). Only steering file is different for the two simulations, all the other input files used are same.
'problem.png' is the screenshot of the problem as in my previous post.

After the simulation is finished. I opened the result (.slf) files in bluekenue. Then I calculated the flow rate by integration along the cross-section (04.i2s file) for the entire period of the simulation ('flowrate_extraction.png').
I don't know how to do this without bluekenue.


Regards,
Nitesh

Hi,

Here is the link to remaining files (cross-section line, geometry file and rainfall file)
I don't know why the files don't get attached. It happened to me many times.

drive.google.com/drive/folders/1xHqPmyMy...lwGDcZAb?usp=sharing

Regards,
Nitesh

Hello Nitesh,

Your issue comes from your Fortran file.
In your modified runoff_scs_nf subroutine, you use the work array T7 with a cumulative effect. It is not calculated from scratch everytime this subroutine is called.

The problem is that the work array T7 is used, and filled in particular when friction velocity (US) is asked in the keyword VARIABLES FOR GRAPHIC PRINTOUTS: so this array contains something different from what you expect with the cumulative effect:

IF(RAINDEF.EQ.4) THEN
CALL OV('X=X+Y ', X=T7%R, Y=T11%R, DIM1=NPOIN)
ELSE
CALL OV('X=X+C ', X=T7%R, C=RFM, DIM1=NPOIN)
ENDIF

This Fortran file should be modified if you want to be more generic.

Anyway, for your specific use, I can suggest you a temporary fix, if you do not use Thompson boundary conditions (that is your case): you can replace T7 by T22 (see e.g. enclosed file).

I think it should better work.

Hope this helps,

Chi-Tuan

Thank you so much! It looks good now :)