open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I'm trying to run 3D hydrodynamics (on v8p5) simulations with 2nd order predictor-corrector advection scheme (MURD), and I found that turning on tidal flats somehow stabilizes the solution. However, Table 4.1 of the theory guide (attached) shows that this 2nd order scheme cannot be used for tidal flats. Then I just wondered what scheme Telemac3d actually applies when I specify 2nd order pred/cor scheme with tidal flats=YES. Does it automatically switch to LIPS scheme?

Also, does anyone know 2nd order pred/cor scheme or LIPS scheme generally has lower numerical diffusion?

I'd appreciate any insight!

Thank you,
Jason

Hello Jason,

I am not a specialist of these advection schemes.
Anyway, what I can say is PC and LIPS are based on N and PSI MURD schemes. If there is a volume around a node which is lower than a threshold, there is no specific treatment for those advection schemes (in particular if no water). I would expect no better results with PC schemes on tidal flats or dry areas.
Your description seems strange, perhaps you are lucky. :)

I was told that PC is theoretically of order 2 (in practice around 1.7 during convergence tests). LIPS is not of order 2, but is quite less diffusive than many advection schemes of TELEMAC (2D or 3D).

Hope this helps,

Chi-Tuan