open TELEMAC-MASCARET The mathematically superior suite of solvers

Hello,

In term of velocity, the unstationnary open boundaries (forced by a tidal signal for instance) are treated with the Thompson method in Telemac.

Is there currently a way of dealing with this type of boundary condition for a tracer flux?
Indeed, an open boundary should allow the perturbations generated inside the computational domain to leave it without deterioration of the inner model solution (made possible in Telemac with the flag "4=FREE TRACER" in the .cli file). However, there is sometimes competing requirement that physically important information be conveyed inwards (if not my salinity plume dissipate with no limit).

Thank you for indicating me how I should deal with this BC issue.

Best regards,


Alexis

Hello,

I am not sure to understand. On an open boundary condition the velocity is either entering or leaving the domain. You can perhaps prescribe the value of the tracer, and Telemac will automatically relax it (putting free tracer = 4 instead of 5 when it exits, this is done by subroutine diffin in library bief).

Otherwise the Thompson boundary condition is also designed for tracers. There will be a new version that works in parallel in version 6.1.

With best regards,

Jean-Michel Hervouet

Okay, that is what I wanted to hear.
Thank you very much Jean-Michel.

Best regards,

Alexis