Incorporating well sorted gravel into model

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
Posts: 4
Joined: Wed May 30, 2018 5:51 pm
Location: Canada

Incorporating well sorted gravel into model

Post by ReubenD » Fri Jul 06, 2018 5:22 pm

Hello, I have read some other posts on the topic which suggested using the van Genuchten fitting parameters for a sand then increase the Ksat. I was wondering if there is possibly a better solution for my situation.

I would like to include a clean well sorted gravel in my model. I have estimated van Genuchten parameters that I believe to be roughly appropriate, but I am experiencing numerical instability and convergence problems when incorporating this material. I was wondering if there is any way to maybe assign it a maximum suction value or some other simplification which would make it easier numerically, but still let it behave as a clean gravel should.

Specifically I would like to show that there is essentially no flux into this unit in the unsaturated zone (because of having much lower suction values than the surrounding material and it lies at the base of a building foundation which shields it from any gravity driven infiltration). The gravel is an engineered material with grain sizes between 1/2" and 3/4" and essentially zero sand or fines. I don't need to use exactly this material's properties. Modeling it as for example a poorly sorted gravel with grain sizes between 1/8" and 6" would be fine, but treating it as a sand doesn't really allow me to illustrate what I am trying to show with the model.

Thanks for any help or advice you're able to offer!

Posts: 4688
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Incorporating well sorted gravel into model

Post by Jirka » Mon Jul 09, 2018 9:03 am

I’m currently traveling in Tibet and thus not much time. When you have soil hydraulic properties for gravel, you likely have a very quick desaturation at relatively small pressure heads and a dramatic decrease of the conductivity. To be able to describe flow in such material, you will need to use relatively find spatial and temporal (on the order of s or smaller) discretization and small water content tolerance (0.0001).

FAQ 8 describes some of the rules about spatial and temporal discretization behind these recommendations: ... zation.pdf

Post Reply