Cumulative water flux problem

A discussion forum for Hydrus-1D users.
Post Reply
Victoria
Posts: 2
Joined: Thu Nov 26, 2015 4:12 pm
Location: Chile

Cumulative water flux problem

Post by Victoria » Mon Nov 30, 2015 7:03 pm

Dear Jirka,

I'm using Hydrus 1D to model water transport (without vapor flow) in a vertical soil column of 5 cm. The experimental conditions are: No flow in the bottom and a variable evaporation flux at the top of the column (without precipitation). Also, I have pressure heads in two observation points (1.23 and 3.75 cm from the bottom) and the average water content in the entire soil column. I want to find the hydraulic parameters (water retention curve and hydraulic conductivity) with inverse simulation. The available data for the inverse simulation are the pressure heads, mean water content, and I want to compare the experimental cumulative water flux at the top of the column (calculated from the evaporation flux) with the modeled cumulative water flux at the top (which I see in the Actual Surface Flux tab of the H1D results).
The simulation time is 14.2 days, the time units are seconds and the length units are cm. The evaporation fluxes are measured every 2 h and the pressure heads and water content every 10 min. The water balance error it’s very low in the simulations (in the order of 10^-2 %), so that’s ok. The initial conditions are hydrostatic equilibrium with the soil completely saturated.
The problem that I have is that when I compare the simulated cumulative evaporation with the experimental cumulative evaporation, I don’t obtain the same surface flux, as shown in the following figure:
CASF.jpg
CASF.jpg (31.72 KiB) Viewed 3205 times

I hope you can help me. Thank you so much!

Victoria Sandoval
Alumna Magister en Ciencias de la Ingeniería.
Departamento de Ingeniería Hidráulica y Ambiental.
Pontificia Universidad Católica de Chile.

Jirka
Posts: 4726
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Cumulative water flux problem

Post by Jirka » Mon Nov 30, 2015 7:49 pm

Victoria,

I assume that sometimes about 700,000 s, the surface pressure head reaches the critical pressure head value hCritA and the calculated actual evaporation from the column start deviating from the specified values. You have couple of options that you may consider:
a. Lower the hCritA (this will likely not help much)
b. Consider also vapor flow (which may raise the calculated actual evaporation).
c. Modify the hydraulic conductivity function (to increase conductivity for lower water content). This may be accomplished by lowering the l parameter, even taking negative values.

We are currently implementing in HYDRUS the more complex (PDI) models for soil hydraulic properties (see http://onlinelibrary.wiley.com/doi/10.1 ... 3/abstract). Perhaps that may help, once it is implemented.

J.

swhitman
Posts: 9
Joined: Fri Apr 15, 2016 8:36 pm
Location: USA

Re: Cumulative water flux problem

Post by swhitman » Fri Jul 20, 2018 10:29 pm

I know this is an old thread, but I am wondering if there are still plans to implement the PDI models in HYDRUS (and if so, when they might be available).

Thanks,

Jirka
Posts: 4726
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Cumulative water flux problem

Post by Jirka » Fri Jul 20, 2018 11:32 pm

Although we have planned to do that, we have not done it. It became obvious that you can easily accomplish the same if you use the look-up tables, rather than the parametric form of the model. You can generate the look-up tables using any model you wish. J.

infoleon
Posts: 10
Joined: Tue Nov 07, 2017 6:05 pm
Location: Brazil

Re: Cumulative water flux problem

Post by infoleon » Thu Sep 06, 2018 12:30 pm

Jirka wrote:
Fri Jul 20, 2018 11:32 pm
Although we have planned to do that, we have not done it. It became obvious that you can easily accomplish the same if you use the look-up tables, rather than the parametric form of the model. You can generate the look-up tables using any model you wish. J.
Hi,

Nice! Is there some manual teaching how to do that? Is it possible to use inverse modeling to fit a PDI curve?


Thanks,

Jirka
Posts: 4726
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Cumulative water flux problem

Post by Jirka » Tue Sep 25, 2018 7:34 am

If you select the lookup table options, the GUI will give you quite clear indication what you need to do.
Tables.png
Tables.png (142.64 KiB) Viewed 1109 times
The exact formatting of the Mater.in file is also described in the manual.

The inverse module of HYDRUS support only the parametric models of soil hydraulic properties (e.g., VG, BC, K, D), not the tabular form.

J.

Post Reply