open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi!

I have been trying to apply wind to a 3d model with version 7.0.0 without success - it seems to be different keywords for wind in 2d and 3d (the keywords for 2d are new compared to older versions but they are the same as the old in 3d). In 2d it works fine with the new keywords. I would be very happy if anyone can help me to solve this!

/Charlotta

Hi,

I am not familiar with wind computations, but did you have a look on the keywords listed in the .dico file?

Regards,
Gabi

Yes, and the keywords for wind are not the same for telemac2d and telemac3d, I think that could be a problem, since it seems that telemac3d also uses telemac2d for some computations. I tried to use the keyword TELEMAC-2D RELEASE ='V6P3'(there the keywords are the same as in 3d v7) but it didn't help.

Thanks,
Charlotta

It works well when I use variable wind with the meteo-fortran file. (It is only the constant wind that doesn't work.)

/Charlotta

Hello,

The person in charge of wind and exchanges with atmosphere will be back in January, but I have looked at the implementation and, yes, it seems that old keywords are obsolete. There is now a:

OPTION FOR WIND with default value 0 (no wind) and options 1, 2 and 3 (you can see them in the file telemac3d.dico in the telemac-3D sources. The components of the wind velocity are no longer used and you have now a keyword :

SPEED AND DIRECTION OF WIND : two numbers separated by a semi-colon, speed in m/s and angle in degrees from 0 to 360.

It is indeed difficult to know what to do without the new documentation but you can look at what is done in subroutine meteo.f (library telemac-2d). The SPEED AND DIRECTION OF WIND is here WIND_SPD(1) and WIND_SPD(2). You can see the formulas implemented depending on the OPTION FOR WIND.

I hope this helps, with best regards,

Jean-Michel Hervouet

Hello!
Anybody solve this problem?
I successfully implemented for Tel2d the subroutine meteo.f (option 99, calling IDWM_T2D for the spatial interpolation).
Now, I want to run the same simulation with Tel3d, I implemented similarly meteo.f but I have problems with the parallelism (i guess) in the subroutine IDWM_T2D.
Did you experience similar problem?
How can I manage this?

Thanks.

G

Hello,

Make sure that you have the latest version of this subroutine, in the enclosed file.

Regards,

JMH

Yes, that's the same subroutine!
Each time I launch the run, the index J of the do cycle (DO J = 1, M) in IDWM_T2D routine stops at a different step, also varying the processor number.
Advices?

Thanks in advance!

G

Hello,

What is exactly the problem ? This subroutine works in parallel because every processor has all the data on stations (NUMSTA is the number of stations). Then the loop on points in the sub-domain is the loop of J from 1 to NPOIN,NPOIN being the local number of points and so it varies depending on the sub-domain. As a consequence there is no surprise in what you say and it seems normal.

With best regards,

Jean-Michel Hervouet

The error is the following attached in the screenshot image.
I don't really have idea in how resolving this problem.
I also used the same routine in tel2d and worked also in parallel mode..
Any hints?
Thanks
G