No convergence in near saturated water flow problem

A discussion forum for users of the new HYDRUS 2D/3D. HYDRUS is a software package for simulating water, heat and solute movement in two- and three-dimensional variably saturated media. Happy Posting!
Post Reply
ghassemi
Posts: 5
Joined: Fri May 17, 2019 3:21 pm
Location: Canada

No convergence in near saturated water flow problem

Post by ghassemi » Fri May 17, 2019 5:27 pm

Hi there,

I am modeling water flow through a silt embankment under variable intensity rainfall by HYDRUS 2D using soil-atmosphere boundary. The initial condition of the embankment which is near saturation has been obtained from another HYDRUS 2D analysis. During the run, mass balance error increases constantly until it stops before the peak of the rainfall event giving the following message in running folder:
“Numerical solution has been stopped since it has not converged in 10 consecutive time steps !”
However, the absolute error is not large relative to the total amount of water in the system. A simulation with hydraulic properties for a sand with everything else the same runs fine.
I appreciate if anyone could help me in this regard.

Thank you,
Ali

Jirka
Posts: 4726
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: No convergence in near saturated water flow problem

Post by Jirka » Fri May 17, 2019 10:06 pm

Different soil textures require different spatial discretization. Please see “rules about spatial and temporal discretization (FAQ: http://www.pc-progress.com/Documents/No ... zation.pdf).

You do not mention, which texture does not work. Note that for fine-textured materials you should use the VGM model with the air-entry value of -2cm.
J.

ghassemi
Posts: 5
Joined: Fri May 17, 2019 3:21 pm
Location: Canada

Re: No convergence in near saturated water flow problem

Post by ghassemi » Wed May 22, 2019 2:01 pm

Dear Jirka,

I checked all model inputs again considering "rules about spatial and temporal discretization". The initial time step is 1 sec and finer mesh in the order of cm has been defined at soil-atmosphere boundary. I am using VG model with reasonable values for hydraulic parameters of a silt laom. As I mentioned in my first post, the initial condition of the embankment which is near saturation has been obtained from another HYDRUS 2D analysis. Attached please find the initial pore pressure head distribution. In some areas close to the slope toe, the soil is fully saturated.
During running the flow simulation of this embankment under variable intensity rainfall, mass balance error increases constantly until it stops before the peak of the rainfall event.

Thank you,
Ali
Attachments
Initial condition.jpg
Initial condition.jpg (130.84 KiB) Viewed 187 times

Jirka
Posts: 4726
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: No convergence in near saturated water flow problem

Post by Jirka » Wed May 22, 2019 6:26 pm

The code will have numerical problems if you try to infiltrate water into a fully saturated profile. The program can handle infiltration excess, but not saturation excess conditions. J.

Post Reply