## Heterogeneous material property

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!
sughosh
Posts: 1
Joined: Fri Sep 29, 2017 8:51 pm
Location: India

### Heterogeneous material property

Dear Sir,
Iam trying to model leachate recirculation process in a typical landfill of 100m X 20m (X-Z axis) for 100days using Hydrus 2D. The landfill model is divided into 5 equal layers of 4m deep to incorporate the heterogeneous nature of waste. The permeability (decreases) and SWCC parameters change with depth and the material properties assumed for 5 layers are given in the figure1 below.
Recirculation is carried out using a vertical well placed at the left side of the landfill and is of 0.15m in radius & 17m deep. The leachate enters the landfill through a 3m deep screen built at the bottom of the 17m deep vertical well. The range of recirculation rate used is 5.5, 27.5 and 55m3/d.

We are facing the following problems while modeling the above problem in "2D Axisymetrical vertical flow domain" and would appreciate any help/advice in this regard.

1) The model fails to converge and stops at around 9 days for the following material properties(figure1). But when homogeneous material (K=1E-5m/s or 0.854m/d) property is assumed for all layers, model successfully runs for all 5.5, 27.5 and 55m3/d recirculation rates.
Figure1-Heterogenous hydrus model.png (25.16 KiB) Viewed 1178 times

2) Calculation of Variable Flux:
The Variable Flux corresponding to recirculation rate of 5.5m3/d is equal to -->[5.5/(3m screen depth +0.15 screen radius)]=-0.873m/d. similarly for 27.5 and 55m/d the flux calculated is -4.365 and -8.73m/d. Please correct me if the above calculation is wrong.

Warm Regards,
Sughosh

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

### Re: Heterogeneous material property

1. Once the model runs for one set of soil hydraulic properties (e.g., for the homogeneous case), it is usually only a question of discretization for it to run also for other sets (e.g., for the heterogeneous case). In your case, you have water infiltrating into an initially dry system and relatively fine discretization around the well, with increasing elements further away from it. It is thus likely that the model stops converging (in your case after day once the moisture front reaches the domain with coarser discretization.
2. I think that your calculations of flux should also include PI, since you are dealing with radially symmetric system ([5.5/area = 5.5/(2*3.14*0.15 screen radius*3 m screen depth + 3.14 * 0.15^2 screen radius)]

I would also use water content tolerance of 0.001 and pressure head tolerance of 0.05 m (max).