open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,


I am sorry to reply so late but i was rectify the bord.f file with the help from pprodano.

You are right my previous fortran file don't run.

So now for prescribing tide from the large model to the small model it's okay, but now in reality i have two liquid boundary, first one for the tide the second for the flow of the estuary.

I thought that it just sufficient to use the keyword Prescribed flowrate as long as the flow is constant.

So i tried the simulation with this keyword, but with checkeing the result i notice that in my boundary of tide the velocity don't taked from the liquid file and it's assigned with 0 value, the same for the flow boundary it's assigned as 0 m3/s.

If you have any suggestion to solve my problem it will be great.


Thank you
YASMINA.

Hi Yasmina,

You took my modified bord.f subroutine to accomplish nesting of global and local models. I said in post 30834 that specifications of liquid boundaries do not work as a consequence of the modification to the bord.f. In other words, I broke the specifications of liquid boundaries through my modification.

The way around it (again, not pretty, but works) is use the same bord.f from the above post and specify river flow with uvel and vvel, which you have to work our separately (perhaps by running a local model separately). Look at the bord.f and the ascii input file from post 30834 to see an example.

Pat

Hello Pat,

Yes Pat i had in mind your fortran file, but i thinked that it's just while i want to impose a velocity and a water depth to a river boundary.

In my project i have only one value of the discharge in m3/s, wich is constant in domain and time. This specification blocked me to use your fortran.

So if i understand what you said, even if i have only one value of debit, i must try to convert them to a velocity U and V to impose them in the fortran file?

Tell me if i miss some thing, in what i understood.

Thanks a lot for your help again.
Regards.

YASMINA.

Hi
If you have only one liquid boundary with imposed discharge constant in time, you just have to give the value in the steering file!
The other liquid boundary with imposed velocity and water level will have another code so there is no confusion.
In bord.f the things are separated.

Hope this helps

Dear Pat and Coulet,

Thank you for your reply.

Coulet it's different because i used the fortran file of Pat.I attached the fortran file to understand what i mean.

But pat, in the mean while i have countourned a second problem, i modified my geometrie file and now it's seems that the Boundary_Colour don't take it from the 13 colum of boundary file. So in my simulation the information taked only from 2 nodes.

I am confused because in the begain all things was gone well at this step

REGARDS.
YASMINA.

Hi Yasmina,

Yes, if you want to use my modified bord.f subroutine, you must provide uvel and vvel in the ascii input file (even when the the river discharge is constant in time and space). Look at the example from the previously referenced post.

Pat

Dear Pat and Coulet,

Thank you for your reply.

Coulet it's different because i used the fortran file of Pat.I attached the fortran file to understand what i mean.

But pat, in the mean while i have countourned a second problem, i modified my geometrie file and now it's seems that the Boundary_Colour don't take it from the 13 colum of boundary file. So in my simulation the information taked only from 2 nodes.

I am confused because in the begain all things was gone well at this step

REGARDS.
YASMINA.

Hi Yasmina,

If you change your geometry file, you must also change the ascii text file that is read by the modified bord.f subroutine. Recall that the begining of this file the subroutine reads the local nodes at the offshore and river boundaries. If your geometry file changes, so does the node numbering at the boundaries.

I hope this helps,

Pat

Hi Pat,

1-Yes i changed the ascii text file, when i posted this topic, the simulation didn't take into acount the text file in sequential mode neither in parrallel. So i thinked that is related to the position of the node in the texte file. I was correct them and make sur that the first value is in the right of the mesh, and the other take in the trigonometric direction.

Now the simulation in sequential is perfect, but in parrallel it is don't run, and i know that is related to the boundary_colour wich is not tooked from the 13 colum.

I realy don't know what i can add or modifie in the subroutine.

I attached the texte file, steering file and the boundary file.



2- I have another question about sediment plume dredging, i want to use two method to study the dispersion of the sediments

First with source tracer, wich make to see only the diffusion, but Pat for the tracer if i understand, with the use of your subroutine bord.f. To make a node as a source of sediment i can't use the keyword in the steering file? but i must use the bord.f?

Second, i want to use the module sisyphe. I want to be sure that with sisyphe we can study the difusion of plume sediment dredger in any node of the mesh. Hope that is possible, if yes, can i have an orientation to begain.


Sorry for my long message.
REGARDS
YASMINA.