open TELEMAC-MASCARET The mathematically superior suite of solvers

Hi,

I have a problem setting the boundary conditions for an instationary flood wave simulation. I am modelling a river including a flood plain. When starting the flood wave, all discharge is in the river (100m wide), but at the 100-year flood peak the outlet section is approx. 3km wide.

My problem is that I have to vary the boundary condition to take this wetting (and the drying after the wave) into account. I have the free surface elevations in the river, but I do not know the water depth or the elevations in the flood plain, because they are filled from backward (overtopping of the river banks more than 4km upstream – part of the water is flowing through the one of the old river beds parallel to the river).

Does anybody has a suggestion?

For this kind of cases, it would be nice to have a downstream boundary condition like in HydroAS, where you can give an energy slope for the boundary and this is taken into account when the node is getting wet…..

Many thanks,
Gabi

Hello,

Here I give you hints but maybe not a full solution. The flow on previously dry land or with such a big flood is probably supercritical, so maybe a condition 4 4 4 at the exit would work. Another solution would be to try a Thompson boundary condition, setting the known elevation only in the main channel. You can try to mix conditions 4 4 4 and 5 4 4, I hope it will work (I am not entirely sure). In a similar situation we have noticed that conditions 4 4 4 on dry land may cause a crash before the water of the flood arrives (probably truncation errors coupled with a problem not well posed) whereas Thompson boundary conditions are stable.

With best regards,

Jean-Michel Hervouet