open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

I want to setup a nested telemac2d fine-grid model based on a big scale estuary model with a coarser grid. Therefore I extracted waterlevel time series at specific nodes of the big-model which also conform to the same locations of my finer model, and are supposed to be my future liquid boundary conditions.

I was now looking into assigning these waterlevel time-series to the nodes of my fine-grid model as waterlevel boundaries (which vary in space).

Previous forum entries suggest the "bord.f" subroutine. If I would use this subroutine do I also need to use the "read_fic_frliq.f" to read the values(from a liquid-bnd file and interpolate them) or is there an easier way to assign the time-series to the nodes? And if yes, can I just assign the time-series to the nodes since according to the manual there has to be a "closed wall section" between different bnd-conditions ?

I also found the option of using a PUNCTUAL RESULTS FILE which was read as a BINARY file, when using TOMAWAC(www.opentelemac.org/index.php/kunena/19-...her-simulation#10656), is there a similar option for telemac2d?

I hope my description was not to misleading.
Thank you in advance for your time.

Best Regards,
Christian

Hello Christian,

I just completed something similar. The offshore boundary for my nearshore domain corresponded exactly to the nodes of my global model (that is how the boundary conditions were obtained).

I used BK to extract depths and velocities from the global model. Then, I modified bord.f to read the extracted depths and velocities (that vary with space and time) and force my nearshore model. My modified bord.f is in post #13844.

Please recognise this is only one way of doing this. There are others.

I hope it helps,

Pat

Hello Christian,

I had also come across a similar case, you may check the solution which I got in the post
www.opentelemac.org/index.php/kunena/16-...m-a-global-model-run
You may download the case file and the program, from the post.

The depths and velocities are extracted by using Bluekenue as suggested by Pat.
You may also check another case from www.opentelemac.org/media/kunena/attachm.../2535/intakecase.rar, where the currents are also imposed.

Withe Best regards,
Gourish

Hello Gourish and Pat,

Thank you both for your quick reply and the useful advice.

I continued to work on the file of Gourish(specicfally "ril.f" from the posthttp://www.opentelemac.org/index.php/kunena/16-telemac-2d/5757-how-to-impose-boundary-conditions-derived-from-a-global-model-run) because there it was easier to deduce the format of the boundary file supplied at fortran port 26.

I formatted my bnd-file in the same manner, see attachement, and took a look at the code which is why I thought it should work. I don't get any errors during the simulation, but the boundary nodes seem not to get the correct time-series allocated.

Is there something I am missing??

Best Regards,
Christian

Hi Christian,

There was a mistake in the numbering of the nodes in the data file which I have provided with reference to the boundary condition file. The node numbers in the data file at the third line should correspond to the node numbers of the mesh which are at the tide forcing boundaries, they are given in the 12th column in the boundary condition file (.cli file). The file which I have provided has the mistake of listing the boundary point number, which are given in the 13th column of the boundary condition file.
Please make the corrections by entering the correct node numbers for the tidal boundary, as per the 12th column of boundary condition file.

Probably this may be the reason as why the boundary nodes are not getting the correct time-series.

I have also updated the program where I have included the following statement
I am providing a test case for illustration.





Hope this helps

With Best regards,
Gourish

Dear Gourish,

Thank you for your help and the provided updated files, I really appreciate them.

I used the right nodes of the *.cli and after defining the Boundaries as Thomson boundaries it worked, thank you for the adapted fortran file.

I was only wondering what the difference is between using a "LIQUIDS BOUNDARY CONDTIONS FILE" with defined waterlevels (where a "classic" boundary setting delivers the expected results) and defining the waterlevels using the fortran file (where in my case only the Thompson boundary delivered the expected behavior). Do you maybe know, how the waterlevels in these two cases are treated differently??

Thank you for your help and Best Regards,
Christian