Ponding over fully saturated soil column = GW above ground

A discussion forum for Hydrus-1D users.
Post Reply
mfd
Posts: 3
Joined: Wed Oct 21, 2020 8:32 am
Location: Germany

Ponding over fully saturated soil column = GW above ground

Post by mfd » Tue Jun 08, 2021 4:25 pm

Dear Jirka,

Í have seen you mention several times that if a soil column is fully saturated as an initial condition, the top centimetre should be set to slightly unsaturated (for example less than -2cm if the air entry pressure is -2cm for VGM). Does this mean that the initial condition should never be gw table above ground, such as a saturated profile plus additional ponding, for example +3cm on top, which I would simulate with h=3cm at the top node?
The lower BC in this case being for example variable flux. (As you say in the Hydrus 2D forum - "don't try to infiltrate into a saturated profile")

And does this also apply to the whole time series? Meaning a simulated groundwater table should not go over the top soil layer, and the ponding calculation is only there for irrigation and similar applications where some part of the profile is unsaturated, or at least there is free drainage?

Thank you and kind regards
Mariel

Jirka
Posts: 5265
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Ponding over fully saturated soil column = GW above ground

Post by Jirka » Tue Jun 08, 2021 4:44 pm

If you want to use an atmospheric BC, then you cannot have a fully saturated domain. That’s simply incompatible. This also means that you cannot have a GWT above the soil surface. You would need to use a different combination of BCs. The new Reservoir BC allows you to accumulate water at the surface (in the surface reservoir). You can try that if such a combination would work. However, I cannot guarantee that since I have never tested such conditions. J.

mfd
Posts: 3
Joined: Wed Oct 21, 2020 8:32 am
Location: Germany

Re: Ponding over fully saturated soil column = GW above ground

Post by mfd » Thu Jun 17, 2021 12:38 pm

Hello Jirka, thanks for the quick reply.

1. I have not set "atmospheric BC", I have set "atmospheric BC with surface layer"(h=7) to enable this GW of 3cm above soil surface. Are you saying that even with this BC, I cannot have a GW above ground = fully saturated profile with surface layer at any point in the time series? If so, what part of the code/formulas/theoretical background is not compatible? I would be very interested.

2. Because meanwhile I found this forum post where GW above surface seems not to be a problem in Hydrus-1D: viewtopic.php?f=4&t=2810&hilit=wetland

3. Also what is the theoretical background for the top node always having to be slightly unsaturated in an initial condition, if the rest of the profile is saturated? (which you mention several times in the forum). Does this only apply when there is an atmospheric BC with no specified surface layer?

4. Thanks for the tip with the reservoir BC, but as far as I can see this is a HYDRUS2D/3D function (and I am using Hydrus 1D), with the closest equivalent in 1D being the atmospheric BC+surface layer, correct?
EDIT: Ah yes sorry now I see in the manual that you meant formula 2.79 for reservoir BC [Mls, 1982]

Jirka
Posts: 5265
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Ponding over fully saturated soil column = GW above ground

Post by Jirka » Thu Jun 17, 2021 6:33 pm

You can try any combination of initial and boundary conditions. However, you may encounter numerical problems. For example, for many combinations of initial and boundary conditions, the solution may not be unique. For example, for a fully saturated column with an atmospheric BC and the top and a free drainage at the bottom, there is an infinite number of solutions, and the program will not be able to resolve it. I do not think that there are any theoretical explanations for numerical non-convergence of highly nonlinear problems (e.g., for the Richards equation), and even if there were, they would not be much useful. If the program does not converge, you need to find some pragmatic solutions to overcome that problem. J.

Post Reply