open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi all,
salut à tous,

I'm currently asking myself what to use? Telemac or Delft 3D. Could someone help me out?

I want to calculate overtopping (inundation) at first in 2D and later in 3D. Anyone that is expierienced with both software packages?

Is the 3D version freeware by now?

Thanks,

Yours sincerely,

Koen Wildemeersch

Hello,

I will not answer on Delft-3D. At EDF we use Telemac-3D for hydrodynamics and Delwaq for water quality, the two understand each other very well regarding mass conservation.
Telemac-3D will be released as open source next July, and universities can have it right now by signing an agreement (given the time it would take, better wait July...). You can also have Telemac-3D right now by paying an assistance contract.

On the technical part: Telemac-2D was from the origin specifically designed for dam breaks and performs very well on dry zones, with a number of specific algorithms (see test cases on malpasset in folder test.fr). Then it is easy to have the same result with Telemac-3D (same mesh, layers of 2D meshes superimposed to do prisms...) and the robustness is the same (see also malpasset test case in 3D), it can even work with 1 layer of elements on the vertical (factor 3 to 4 in computer time compared to 2D!). It is a full non hydrostatic Navier-Stokes, and work also in domain decomposition, several 10s of thousands of processors already tested.
A typical test case of overtopping with Telemac-2D is the one called "digue" in the test.fr folder.

I hope this helps,

With best regards,

Jean-Michel Hervouet

Jean-Michel Hervouet,

thank you very much for your answer. I'm very pleased with the fact that is goes open source next July. But, by next July do you mean 2011 or 2012?

Yours sincerely,

Koen Wildemeersch

ps. Anyone who can make a comparison between telemac and delft3d, especially when computing overtopping and inundation.

Ohh... And as well, I noticed that FlOW, MOR and WAVE are already open source. By next July, will all the modules be open source, or only some of them?

Yours sincerely,

Koen

Hello,

All the modules will be open source at the end of July 2011, including a 1D and 3D, except groundwater flows.

JMH

Dear Koen,

Even though it has been a couple of weeks since you posted this question, let me briefly comment on Delft3D.

Dam breaks were not the first application that my colleagues had in mind when developing Delft3D, but drying and flooding is also important in tidal areas for which it was developed. Later, the robust flooding scheme of Stelling & Duinmeijer (2003) was added. Our Delft3D validation document includes successful simulations of the typical dry and wet bed dam break cases described in that paper.

So, I think that Telemac and Delft3D are equally well suited for this kind of application. It all depends on the other features that you need; Telemac being unstructured and Delft3D still structured is only one of the many differences.

Best regards,

Bert

Indeed we, at HR Wallingford, have used (and developed and distributed) TELEMAC-3D since it was created, and the TELEMAC system in general for the past 20 years, principally because TELEMAC works with unstructured grids.

Contrary to solvers used by others (such as MIKEx or DELFTx), the TELEMAC system was designed from the outset, more than 20 years ago, to use the mathematically more advanced finite element techniques so that very flexible unstructured triangular grids can be used. This is superior to using an orthogonal and/or curvilinear grid as the engineer gains more control over grid refinement particularly in cases of detached coastlines, underwater features including navigation channels and manmade structures. In addition, currents that are tangential to solid or coastal boundary conditions (slip or nonslip conditions) are correct by design.

Horizontally, the model grid can therefore be setup over the entire model domain with a variable density mesh, which focuses on the details required while expanding to larger sizes towards offshore boundaries.


Hope this helps.

Sébastien