open TELEMAC-MASCARET The mathematically superior suite of solvers

Dear friends,
I am running a meandering river case using telemac2d coupling with sisyphe. I want to set tu different sediment size class distribution for different river section rather than only give one group sediment size for the whole river.
Could you give me some suggestions about this problem.
Your reply will be appreciated very much.
Best regards,
Fakai Lei

1) Run the model with multiple fractions
Than there are two ways it
1a) Add the bed composition to the geometry file (using BLueKenue or whatever tool you use to create a geometry file). You can then define the fractions in the bed of the sediment classes for each section.
1b) update the init_compo.f subroutine (see attached), making sure that the line WRITE(VAR_NAME,'(a11,I1,a4)')'FRACLAY 1 CL ',I,' '
produces the variable name used in the geometry file.

Alternatively
2a) Hard code the fractions in init_compo.f (if there are some easy rules you can use).

Dear Makfnappen,
I think the first way is easier for me to operation. However, now I have no ideal to add the composition to the geometry file. To my best knowledge, I should add bed compositions as a new variable to the selafin as the bottom elevations. But I don't know how to write the bed composition to the blukenue. What's more,after that, I have no ideal to define the fractions of the sediment classes for each section in the steering fiels.
Could you give me some operation details. And if you can give a similar case of yours, it will be very helpful to me.
Thank you for your very nice reply and looking forward your next reply.
Best wishes,
Fakai LEI

Hi Fakai Lei

In
examples\sisyphe\yen_multigrain
you can see how to define multiple sediment classes in the steering file.
As for the geometry file:
Adding the bed layer information to the geometry file works the same as defining the bed levels, you just need to give it another name. The Fortran file I attached previously uses the same names as SISYPHE outputs, but you can change it to whatever you like as long as the names match up in the file and the Fortran code. You will need to add an additional variable for each sediment class and each bed layer in your model (see fis_yen-exp_multi.slf, which has 10 variables
FRACLAY 1 CL 1 ........ FRACLAY 2 CL 5
to define fractions of 5 classes in 2 bed layers

Sorry that I can't make it easier, but that is because it isn't easy to set up (in any model).


Michiel

Dear Mafknaapen,
I can't express my appreciate any more.The help you give me is enough, I will try to figure out it.
Thank you again.
Best wishes,
Fakai Lei

Hi !
Personally for the initial distribution:
- firstly, I get the field's data of sediment grain size distribution (it's important to precise this step because too few modeller are doing that)
- then I incorporate the data in the serafin file in coding it in fortran
Best wishes !