Page 1 of 1

Pressure data and model do not match

Posted: Thu Mar 25, 2021 9:35 am
by Mitra
Hello,

I have pressure (in the middle), moisture, and outflow data for a soil column. When I use HYDRUS to reproduce the data. I am successful in reproducing the flux, and soil moisture; however, when I compare the pressure values ( where negative values indicate absence of fully saturated condition and positive value indicates presence of a saturated colum several cm above) with that of HYDRUS. I see the same trend, but the values are different. All the pressure data I collected are positive (fluctuating in range of 2-0 kPa), while in HYDRUS it starts from the same point +2, it immediately drops to negative values and fluctuate with the same magnitude as my measured data (but in the range of 0 to -2kPa). I was thinking how can I overcome this issue, or what can be the possible reason for it?
Thanks,
Mitra

Re: Pressure data and model do not match

Posted: Thu Mar 25, 2021 4:15 pm
by Jirka
I think you need to first figure out whether the conditions in your soil column are saturated or unsaturated. We cannot really help you with that. J.

Re: Pressure data and model do not match

Posted: Tue Apr 06, 2021 11:36 am
by Mitra
Actually, what you mentioned is very correct.
I applied a constant suction at the bottom of my soil column (-300cm) and in my model, suction pressure of -300cm (I have also tried seepage face at this suction pressure) as the bottom boundary condition without considering evaporation (measured values were negligible). This gives me the minimum water content at the bottom and a maximum water content at the top while in my soil column experimental data, I have the most saturated condition at the bottom and most dry condition at the top.

So now, I am wondering if I am using the right bottom boundary condition?

Thanks,
Mitra

Re: Pressure data and model do not match

Posted: Tue Apr 06, 2021 4:38 pm
by Jirka
I think you need to first think about the physics of the entire problem. What different BCs represent physically? I have a feeling that you do not fully understand the physics. You can not use HYDRUS successfully without such understanding. There are extensive texts on the physics of different BCs in various textbooks, such as:

Radcliffe, D., and J. Šimůnek, Soil Physics with HYDRUS: Modeling and Applications, CRC Press, Taylor & Francis Group, Boca Raton, FL, ISBN: 978-1-4200-7380-5, pp. 373, 2010.

Rassam, D., J. Šimůnek, D. Mallants, and M. Th. van Genuchten, The HYDRUS-1D Software Package for Simulating the Movement of Water, Heat, and Multiple Solutes in Variably Saturated Media: Tutorial, Version 1.00, CSIRO Land and Water, Adelaide, Australia, 183 pp., ISBN 978-1-4863-1001-2, 2018.
https://www.pc-progress.com/Downloads/P ... 0_2018.pdf

J.

Re: Pressure data and model do not match

Posted: Wed Apr 07, 2021 5:20 pm
by Mitra
I know theoretically what does each condition represents, what I do not know is which condition my experimental setup represents at the bottom boundary. I have a pump that applies a high suction pressure to a bottle through a tube, and the bottle is completely closed and connected to the bottom of lysimeter with a tube (a single point). and water drips from soil passing through a highly conductive membrane to the bottle with the tube. Bottom of lysimeter and tip of the bottle are at approximately at the same height. In the reference you mention in working with lysimeters, usually seepage face boundary condition is used. I thought as pump is applying a suction pressure, maybe a seepage face at that suction pressure will be the bottom boundary condition. If you can provide me with some options or tests to identify the right bottom boundary condition, it will be precious for me.

Re: Pressure data and model do not match

Posted: Wed Apr 07, 2021 6:02 pm
by Jirka
It seems that you are on the right track. Why don’t you then use “Seepage Face” with a specified pressure head? J.

Re: Pressure data and model do not match

Posted: Thu Apr 08, 2021 12:05 pm
by Mitra
I thought as I am applying a suction of 30kPa in reality, so I can use a seepage face boundary condition of -300 cm (equivalent to 30kPa) at the bottom, does it seem sensible to you?
This boundary condition still gives me, the lowest soil moisture at the bottom of the soil column while in my setup and in reality, I am getting the highest soil moisture reading at the bottom. So, I am thinking as the pressure is not directly applied from the pump to the plate (it is a tube connected to a bottle, and the bottle is connected with a tube to the lysimeter (only to a single point of the whole cross-sectional area)), I might need to readjust this seepage face boundary condition to lower pressures than what is actually the pump applying. I don't know if this can be the case or not. and if it is the case, how can I estimate, by how much I should readjust the pressure.

Re: Pressure data and model do not match

Posted: Wed Apr 14, 2021 7:34 am
by Mitra
Actually, it seems the boundary condition at the bottom is alway seepage face (no matter which suction I apply); probably because of a membrane at the bottom. However, I am thinking how the suction applied should best be represented in the model (the model works with higher Ks at higher suctions) ?

Re: Pressure data and model do not match

Posted: Wed Apr 14, 2021 2:38 pm
by Jirka
I do not have much to add. As you know, HYDRUS-1D has a seepage face BC, where you can specify the boundary pressure head, h_seep. When the simulated pressure head is larger than h_seep, h_seep is imposed on the boundary. When the simulated pressure head is smaller than h_seep, zero flux is imposed on the boundary. This is obviously described in the manual and help. J.