open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello there,

Familiar with the topic already discussed (~already 10years ago , see link below) I wanted to know how to set initial stratification with different grain sizes in each layer. The keyword adjustments in the .cas file (gaia) are known to me and are well documented in gaia user manual and also here in the forum.

But since it will be necessary to adapt a fortran ("init_compo.f")as well to enter the desired initial grain sizes, I´m a bit lost because I have no experience in writign and changing a fortran file.
I would be very grateful if there is someone who can briefly explain the variables to be adjusted or possibly has an example file ("init_compo.f") in which similar adjustments have been made.
I want 2 layers, each 1.5m thick with different grain sizes.

Many thanks in advance,
Matthias


www.opentelemac.org/index.php/kunena/17-...ith-mixed-sizes#4741

Did you have a look at the example in
\examples\gaia\yen-t2d\user_fortran_multi\USER_BED_INIT.f
That does define 2 layers with 2 different compositions and tow layer thicknesses.

Hi,

I have a related question to do with initial stratification. I'd like to copy the ratios from a prior results file into the current simulation. The fact of setting the ratios is relatively easy, as mafknaapen has established, but the part about reading the ratios from another file in is what I'm not so sure about.

Ideally, I would resume from prior computation, but the trouble is that the bathymetry has changed between the simulations, while the sediment distribution is essentially the same as the prior simulation (there are some specific changes, but I can handle those in the code further down). It's not clear to me what takes precedence on what between the sediment geometry file and the previous sedimentological computation file.

That said, my question is twofold:

1. Is there a way to "resume from prior sedimentological computation" in such a way that the stratification is preserved, but not the bathymetry?
2. If not, what would be the best strategy to read stratification ratios from a prior selafin and impute that into the current model, considering I'm running this as a parallel computation?

There may well be prior posts about this, but I'm not great at searching these forums, so apologies if I've missed anything – sometimes, the best search function is someone's memory of a prior post

Thanks!
André Renault

Oh, I should mention (though it was probably assumed) that I want the data from the final timestep of the results file. Unfortunately, I'm too late to edit my post, hence the double reply

If you add the relevant result variables for the sediment fractions to the geometry file in BlueKenue, the model should use these instead of the defined fractions in the steering file.

Hi Michiel,

Thanks for this idea. However, I am looking at having different sedimentological files as input, as I'm doing an informal sensitivity analysis, and the prospect of copying multiple variables several times over in Blue Kenue is not an enjoyable one (as BK does not lend itself well to batch jobs). Rather, I would like to be able to specify a file at runtime and use its sediment data, while retaining its own bathymetry.

Matthias has an interesting idea in doing the opposite: using a previous sedimentological computation file for the stratification, and changing only the bathymetry with the new geometry. However, the issue remains the same where I essentially need to merge data from two selafins, because I'm running tests with multiple inputs.

Thank you very much for your reply @mafknaapen!
I checked the fortran file "user_bed_init.f" [\examples\gaia\yen-t2d\user_fortran_multi\USER_BED_INIT.f] and with this file it should be possible to modify the layers in the way I need it.

Especially with the example mentioned in this specific example, it is easy to modify the file. One little question arised nevertheless: While I defined also my variables for the sediment (class) diameters within the fortran file, I think the classes / diameters are taken from the defined classes in the (gaia).cas file (because for the code only the variable "NSICLA"= number of sediment classes is needed) ?

@Renault:
For my purpose i casually checked the other provided fortran file: examples\gaia\yen-t2d\user_fortran_multi
and I think this is something that can help you with modifying the bottom topography (without changing your stratification?)
by changing the variable "ZF%R(I) = [fixed bottom_height]"

In that case, I think the answer would be in your initialisation run.

You could use SISYPHE for that with the setting STATIONARY MODE = YES (unfortunately, this is removed from GAIA).

The alternative might be (I haven't tested it yet), to run the initialisation run with an edited bed1_update.f. In this file comment line 231 (v8p4r0) out:
DO IPOIN = 1,NPOIN
ZF%R(IPOIN) = ZR%R(IPOIN)
DO ILAYER = 1,NOMBLAY
! ZF%R(IPOIN) = ZF%R(IPOIN) + ES(IPOIN,ILAYER)
ENDDO
ENDDO

If I read the code correctly, that will avoid the bed levels being updated. Bur again I haven't checked it.