open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello everybody,
I am going to run several models in year 2020 (e.g. 14 days in June; 7 days in Nov.,...). Normally It is required to make the time series of boundary conditions (discharge, meteo, liquid) separately for each simulations.

But I am looking for an option to make time series of the boundary condition(s) for the whole year, only once. And then in the CAS file, by specifying the 'ORIGINAL DATE OF TIME' and 'ORIGINAL HOUR OF TIME', I want to read the corresponding boundary condition values at that timestamp (e.g. at 15 Jun 2020 00:00). Does the Telemac model has this possibility?

P.S. I added the keyword #REFDATE 2020-01-01 00:00:00 at the header of boundary condition text files. But, the model still reads the values of 1 Jan 2020, not 15 June 2020.

Hello,

#REFDATE in files (e.g. LIQUID BOUNDARIES FILE, SOURCES FILE or ASCII ATMOSPHERIC DATA FILE) is a date from which a shift is done with time of computation.
If you have a file starting on January 1st 2000 at 00:00 and you want to start the computation e.g. on February 1st 2000, you have to change initial time of your computation.

I can suggest you one way: to change AT in CONDIM subroutine (by default, time starts at 0 s or at a time from the previous computation file):

to e.g.

and add this subroutine in your Fortran file.

The 2 keywords ORIGINAL DATE OF TIME and ORIGINAL HOUR OF TIME are not dedicated to define the initial time of the simulation but a reference for a few features (tidal boundary conditions, tide generator force, heat exchange with atmosphere, one way coupling with DELWAQ) and are added to current time of the computation to use them.

Hope this helps,

Chi-Tuan

Dear Chi-Tuan,


Thanks for your advice. I modified the CONDIM subroutine and it works fine (tested for both the coldstart and hotstart models).

When changing the value of 'AT' parameter in 'CONDIM' subroutine, the only tip is that I MUST remove #REFDATE line from the boundary condition files. Otherwise, it starts the model from timestep number AT*2 (considers 'AT' twice: one from the 'CONDIM' subroutine and the other one from the CAS file).


Cheers,

Mostafa