open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

For the last few months I have been using TELEMAC-3D to work on various interesting problems. Of, late I have been trying to delve a bit deeper into the implementation of the bottom shear stress in TELEMAC-3D. I have been referring to the Finite Element book, by Dr. Hervouet, but things are not exactly clear. If I understand it correctly, the treatment of bottom shear stress is same for TELEMAC-2D and TELEMAC-3D. Please clarify ...

Hello,

The treatment is different, but it is implemented to give the same effect if Strickler or Chézy law is used. As a matter of fact the friction in 2D acts on all points of the domain (like a source term in the momentum equation), whereas in 3D it acts only on the bottom (and appears as a boundary condition of the diffusion step).

See subroutine tfond.f in telemac-3D for the treatment in 3D. As the friction in 2D is based on the depth-averaged velocity U2D and V2D, we also use it in 3D for friction laws such as Strickler or Chézy (which are basically depth-averaged or section-averaged laws). In this way a calibration done in 2D will somewhat remain valid in 3D (see page 52 of the book). If you use a Nikuradse law in 3D the treatment will be different, based on an assumption of logarithmic profile.

With best regards,

Jean-Michel Hervouet

Thanks a lot Dr. Hervouet,

The cases with Chezy/Strickler law is clear to me now, and for the Nikuradse law implementation I will look into the subroutine "tfond.f" to have a better understanding. Is there a document (apart from the sub-routine), which talks about the Nikuradse law implementation in some details ?


best regards

Som

Hello,

I'm afraid not, it is based on the logarithmic profile: the logarithmic profile in rough regime is assumed and the friction velocity is deduced from the value of velocity at the first plane above the bottom (it has always been a wonder to me that you tend to a logarithmic profile by just assuming that it is already there...).

Regards,

JMH