SeepageFace Boundary Condition

[Mirna Kassem]

Hello,

I have been trying to use the SeepageFace boundary condition at the bottom boundary (z-lower) when modeling groundwater flow within a watershed for rainfall-induced landslide applications.

The ParFlow manual does not appear to provide an example of how to implement this boundary condition. Is the following setup correct?

pfset Patch.z-lower.BCPressure.Type SeepageFace pfset Patch.z-lower.BCPressure.Cycle "constant"

Since changing the bottom boundary condition from FluxConst to SeepageFace, ParFlow has been crashing (see attached snapshot), so I am wondering whether I am missing an additional parameter or setting.

1. How can I fix this issue?
Do I need to specify a pressure value, for example:

pfset Patch.z-lower.BCPressure.alltime.Value 0.0

However, my understanding is that this would impose a zero-pressure condition at the bottom boundary, which seems inconsistent with the purpose of using a seepage-face boundary condition.

2. Is applying a SeepageFace boundary condition at the bottom boundary appropriate for modeling shallow rainfall-induced landslides within a watershed?
My goal is to represent groundwater flow conditions as realistically as possible while assessing slope instability triggered by rainfall.

Thank you in advance for your help.

Best,
Mirna

[Smith, Steve]

Hi,

Thanks for reporting this, the documentation is not clear and as you indicate we do not have an example using the SeepageFace BC, which is bad since it means we are not regularly testing this boundary condition.   The https://parflow.readthedocs.io/en/latest/keys.html#boundary-conditions-pressure manual section doesn't indicate it  but SeepageFace does require an input value.   I believe the Overland Flow comment:

The key OverlandFlow corresponds to a Value key with a positive or negative value, to indicate uniform fluxes (such as rainfall or evapotranspiration) over the entire domain 

Applies to the SeepageFace value as well.   I'm checking with team to verify the input value is a flux for SeepageFace.

The output of the parflow run is captured in the <runname>.out.txt file.   This file can sometimes provide additional error messages for what can causing execution failures.   I think you will see an error indicating a missing Value key for the SeepageFace alltime Value.

Steve

---

From: 'Mirna Kassem' via ParFlow <par...@googlegroups.com>
Sent: Monday, June 15, 2026 11:01 AM
To: ParFlow <par...@googlegroups.com>
Subject: SeepageFace Boundary Condition

 

--
You received this message because you are subscribed to the Google Groups "ParFlow" group.
To unsubscribe from this group and stop receiving emails from it, send an email to parflow+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/parflow/31f84f94-f4b1-499e-9779-20f146263a69n%40googlegroups.com.

[Mirna Kassem]

Hi Steve,

Thank you for your reply!

What is the advantage of using the SeepageFace boundary condition if I still need to specify a constant flux? In that case, wouldn't a FluxConst boundary condition achieve the same result?

The manual mentions that SeepageFace allows flow freely through the boundary.

Thank you for your help.

Best,
Mirna

--

Mirna Kassem, Ph.D.

University of California, Berkeley

Civil and Environmental Engineering Department

Email: kassem...@berkeley.edu

Website: Mirna Kassem