open TELEMAC-MASCARET The mathematically superior suite of solvers

Good evening,

I'm doing a simulation of a close trapezoidal basin with a costant velocity of the wind (I've used both k-epsilon, mixing length and smagorinsky to model turbulence).

I'm trying to realise a graphic with
- x axis: averaged velocity (result from computation) / velocity on the surface (datum)
- y axis: bottom stress (from 'US' output) / stress on the surface (datum).

I'm considering velocities in the middle of my basin, far from set-up or set-down, in order to compare to Delft/Roms results.

The velocity profiles are correct; for example, in the deepest part of the basin I can see an inversion of the velocity in the wind direction (near the bottom the velocity is in a certain direction, while near the surface it changes direction).

But if I put my results on the graphic, I cannot see this.

May anyone have an idea of what doesn't work?

I attached something that may be useful.

Thank you so much

Hi
Not fully sure but I think the bottom stress and surface stress are computed as a scalar that couldn't be negative because it use the velocity magnitude, not only the U component along the X axis...

Hope this helps