Posted Thu, 06 Sep 2018 06:55:33 GMT by LIONEL SCHAPER

I'm working on a sand column, which is unsaturated (partially saturated) and I want to simulate infiltration in the column in order to fit observed data.
The observed data show a strong hysteresis effect.

To do so, I have asked Feflow to consider hysteresis effect in the Problem settings and defined:
- the Van Genuchten parameters for the main drying curve and the main wetting curve
- the saturation limits for the main drying and wetting curve
- initial conditions, close to residual saturation

After injecting water at the top of the column (fluid flux), the water front mirgates well in the column and is welle drained at the bottom (fluid flux gradient).

No oscillations problems or other problem, as I limited the time steps and refined the mesh.

The problem is the following :
when I plot at observation points the curve "pressure = f(soil moisture)" or "pressure = f(saturation)", the curve stricly follows the wetting curve, even after the water front has passed, where I would have think that the curve would switch to a drying scanning curve ...
In fact, the hystérésis effect seems to be not considered ... and I don't uderstand why.

Another question is the threshold mentionned in the Problem setting. By default, it is 0,01 m. When I change it (1e6 m in order to switch from wetting to drying theoretically faster), it doesn't change.

Finally, how Feflow understand that the initial conditions correspond to drying or wetting conditions ?

Thank you for your help.
Posted Wed, 19 Sep 2018 13:59:12 GMT by Björn Kaiser
Your question is difficult to answer here in the forum. Could you please contact the FEFLOW support?

You must be signed in to post in this forum.