open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello Arman,

I can suggest you a few things to try:
- decrease your time step, try e.g. 2 s, 1 s, perhaps less,
- try to model with 8 passive tracers (use the same TELEMAC-2D steering file with COUPLING WITH = '' or comment the line of this keyword). Does the problem still occur?
- for such issues (nan), you should use a debug configuration with debug options to investigate.
See e.g. the S9.gfortran.debug configuration in the $HOMETEL/configs/systel.edf.cfg configuration file, in particular the flag fflags_debug_gfo for gfortran compiler.
fflags_debug_gfo: -g -Wall -fcheck=all -fbacktrace -fbounds-check -finit-integer=-1 -finit-real=nan -ffpe-trap=invalid,zero,overflowconfiguration as you have NaN in your listing (Not a Number).

Hope this helps,

Chi-Tuan

Dear Chi Tuan
Thanks for your reply.
I tried to decrease the time step and and I have still the same error. I am using water process=3 and I have no problem. Even when I have 8 tracers I have no errors.The error is just happens when I am using o2 (water process =5 or 2)
as a tracer.

Dear Chi Tuan
I am using Eutro coupling with Telemac2d. I think the simulation is highly sensitive to initial 8 tracers in Eutro. Can I use the initial values of tracers in my own case or I should just use the initial values of the Waqtel-Eutro example?
Regards

Hello,

Initial conditions may be dependent on the configuration (geometry, options etc.) you try to model. The most adapted to your configuration, the better results you will get, in particular at the beginning of your computation.

Hope this helps,

Chi-Tuan

Dear Chi Tuaan
When I am using my own initial condition I receive Error NAN in my results but when I am using the initial condition in the example if works good. Maybe it is due to the relative content of the different pollutants that can not happen for a starting point of simulation. Is it possible?
Regards

If you get NaN, as already told, you should use a debug configuration (see one of my previous post in this topic). It will tell you exactly in which subroutine the program crashes and may enable you to understand what is going wrong with your first choice of initial conditions.
Perhaps, there is a bug, but it is a mean to find it.

It is very efficient and will tell you more than I can in your case.

Hope this helps,

Chi-Tuan

I am using HPC and I cant update systel.edf to debug configuration.

Hello Arman,

You mean you run your computation on clusters?
If you look at gaia.intel.debug/gaia.gnu.debug or cronos.intel.debug/cronos.gnu.debug you have some examples of debug configurations for intel or gnu Fortran compilers.

Hope this helps,

Chi-Tuan