How should the local infiltration conditions be set
Zhongkang Yan
I'm currently reproducing the examples in someone else's article. However, after setting the infiltration conditions for the top part, the result was not reflected. The result was still that there was infiltration throughout the entire top surface.
The scenario description, results and input files of the calculation example are as follows.
I don't think this calculation example is complicated. It's just that the local infiltration conditions at the top are somewhat rare. I don't know where the problem lies. Please give me your advice!
We validate our numerical implementation by comparing it to a widely-used open source simulator PFLOTRAN (Hammond et al., 2014). In the numerical experiment, at t = 0, a non-reactive tracer at a concentration of C0 = 0.5 mg/L is uniformly distributed in a 1 m × 1m × 1m 3D domain composed of Accusand. The domain is discretized using uniform cells ∆x = ∆y = ∆z = 1 cm. The water table depth is 90 cm and is kept constant during the simulation. The initial water pressure head is hydrostatic. The four side boundaries are all set as no-flux boundaries for water flow and solute transport. At the center of the top boundary, starting at t = 0, constant infiltration of clean water at 1 cm/d is applied to a 10 cm × 10 cm square area and lasts for 48 hours. The rest of the top boundary is set to zero flux. The 3D spatial distributions of water pressure head (h), water saturation (Sw), and aqueous concentration (C) are computed. The cross-sections of these distributions at y = 49.5 cm at the end of simulation using results computed from our code and PFLOTRAN are presented in Fig. B.1. Excellent agreement between the results from the two codes is observed.
Thank you very much for your time and support!
Sincerely,
Hammond, Glenn E
Zhongkang,
Define the domain
GRID
TYPE STRUCTURED
NXYZ 100 100 100
BOUNDS
0. 0. 0.
1. 1. 1.
/
END
To apply a boundary condition to the center 10x10cm portion of the domain, create a region:
REGION top_infiltrometer
FACE TOP
COORDINATES
0.45 0.45 1.
0.55 0.55 1.
/
END
and flow condition:
FLOW_CONDITION infiltration
TYPE
LIQUID_FLUX NEUMANN
/
LIQUID_FLUX LIST
TIME_UNITS h
DATA_UNITS cm/d
0. 1.
48. 0.
/
END
and couple in the boundary condition coupler:
BOUNDARY_CONDITION
REGION top_infiltrometer
FLOW_CONDITION infiltration
TRANSPORT_CONDITION <string>
END
I would initially use “NXYZ 10 10 10” to test and then migrate to the 1M cells.
Glenn
From:
pflotra...@googlegroups.com <pflotra...@googlegroups.com> on behalf of Zhongkang Yan <yzk...@163.com>
Date: Monday, June 16, 2025 at 7:33 AM
To: pflotran-users <pflotra...@googlegroups.com>
Subject: [pflotran-users: 8437] How should the local infiltration conditions be set
Check twice before you click! This email originated from outside PNNL.
Zhongkang Yan
--
You received this message because you are subscribed to the Google Groups "pflotran-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
pflotran-user...@googlegroups.com.
To view this discussion visit
https://groups.google.com/d/msgid/pflotran-users/8c6e62d6-ad6a-4c12-ba31-ce20a754200bn%40googlegroups.com.
Zhongkang
Glenn，
Thank you for your timely reply. It's very helpful to me.
I reconfigured the domain, top infiltrometer conditions and flow conditions as you said in the previous email and coupled them together.
However, the result I ran still showed that the entire top surfaceof the journal was infiltrating. This doesn't match the setting I made where only a part of the top area has seepage.
I also tried to manually set the area at the top, except for the seepage area, to a no-flow boundary, but it still didn't work.
I would like to ask you what other possible reasons there might be.My input file and result graph are enclosed with this letter.
You received this message because you are subscribed to a topic in the Google Groups "pflotran-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/pflotran-users/YqYNH5L61Nc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/PH0PR09MB78365EED42B162F7D71F41519A73A%40PH0PR09MB7836.namprd09.prod.outlook.com.
Hammond, Glenn E
Zhongkang,
Your permeabilities were huge (~10^-5 m^2). The water flows quickly into the domain, and you cannot see the wetting front. Run the attached input decks with lower permeability, increased infiltration rate, longer runtime, and an initial condition that has a lower water table.
Glenn
--
You received this message because you are subscribed to the Google Groups "pflotran-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to
pflotran-user...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/7d682931.a7a5.1978d73b7ed.Coremail.yzk1116%40163.com.
Zhongkang
Glenn，
Thank you for your timely reply.
I adjusted the input file according to your suggestion and the problem has been successfully solved now.
Sincerely,
Zhongkang
To view this discussion visit https://groups.google.com/d/msgid/pflotran-users/PH0PR09MB7836460B13CCE0A6FF68680A9A78A%40PH0PR09MB7836.namprd09.prod.outlook.com.
