open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

does Telemac distribute the rain over the nodes or over the area?

I have two surfaces with the same area. But if i change the number of the interior nodes i got different results from a section control.

Best regards

Michael

Hello,

The rain is modelled as a source term on each node (rain intensity in m/s). The added volume during one time step by rainfall is equal to intensity * area around node * time step.

Could it be that the differences you have in control sections are related to mesh discretizations?

Best regards
PL

Thanks for the answer.

No i dont think so but i could be possible.

I created two test areas with the same amount of nodes.

One of them with more nodes in the upper half and the other is mirrored.
The first one creates more flow in the control section then the other.

Iadded the files from the test area. The difference here is really tiny but in my real area the difference is extreme.

Best regards
Michael

Hello,

I tried your case and the mass balance says that the final volume in the domain is 210 m3 which corresponds exactely to your rainfall definition. So there is no problem related to the way rainfall is introduced into you domain.

The problem is certainly your geometry: you have a very steep slope (10%) and large mesh size in the flow direction (3.45 m). In finite elements you don't satisfy the criteria for how tidal flats are handled since you have very small water depths. Runoff is actually a tough thing to model in 2D and the best is finite volumes for that. I tried it but was not better, I suspect due to the geometry / mesh.

Try to refine the mesh / lower the slope and see how things change.

Good luck!

Best regards
PL