open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi all,
I need to simulate the flow with Telemac 2D, considering 5 pumps distributed near the boundary of my computational domain. I have the x and y coordinates and the discharges. I thought about using the SOURCE command, but with negative discharge, as the following:

ABSCISSAE OF SOURCES: 559972.95; 565668.0; 566850.0; 569146.0; 569667.0
ORDINATES OF SOURCES: 9019533.32; 9024490.0; 9026214.0; 9027894.0; 9028277.0
WATER DISCHARGE OF SOURCES: -0.89; -0.078; -0.09; -0.127; -0.164

The program gives me this error:

SPECTRUM OR SOURCE POINT 1 OUTSIDE DOMAIN
NOT A MISTAKE IN PARALLEL MODE

IN PARALLEL SOURCES OR SPECTRUM POINTS
MUST COINCIDE WITH
NODES IN THE MESH, FOR SOURCE 1 CHOOSE:
X= 559939.0625 Y= 9019513.

PLANTE: PROGRAM STOPPED AFTER AN ERROR
Application 1696614 exit codes: 2
Application 1696614 resources: utime ~1s, stime ~1s, Rss ~11036, inblocks ~24732, outblocks ~79843

Do you know what I made wrong? Does it mean that I have to give the program points already into the mesh?

Thanks a lot and best regards

Hello

When running in parallel, source coordinates have to be exactly on the nodes of the mesh, which is not your case.
In serial run you can introduce any source coordinates and Telemac will search for the nearest node.

with my best regards
Riadh ATA

Thanks, that is helpful, I will search with Janet the nearest nodes and change a bit the coordinates. Despite for that, the concept is correct? Can I use the sources in my case?

Thnaks in advance and best regards

I forgot... do I have to give also the z values to the points?

Hello

Yes the concept is correct, it is one way to modelize your problem. An other way to do is to add interior boundary (small islands, or one bigger boundary) in the location of the pumps and prescribe your flowrates there.
There is no need for z.
If your flowrates will change in time, you can use the source file with the key-word SOURCES FILE, see user manual for further details.

My best regards
Riadh ATA