ponded water with freezing temperatures

A discussion forum for users of the new HYDRUS 2D/3D. HYDRUS is a software package for simulating water, heat and solute movement in two- and three-dimensional variably saturated media. Happy Posting!
mgoddard
Posts: 13
Joined: Wed Dec 23, 2015 11:00 pm
Location: Canada
Location: Netherlands

ponded water with freezing temperatures

Post by mgoddard » Thu Apr 21, 2016 7:29 pm

Hi all,

Questions regarding freezing temperatures and how Hydrus deals with them.

I have ponded water/lake of 1 m depth (constant head) in my simulation, and several other small water filled ditches (0.05 m water depth, constant heads) along the top boundary surface of a large dam. When I apply heat transport in the form of daily average air temperatures (8 years), what does Hydrus do with the ponded water when the temperature drops below freezing? Does the ponded water freeze? or does it remain as a liquid? Also, what does Hydrus do with the water with the near surface unsaturated soils? Does it freeze?

The location of the dam is in northern Canada where the air temperature is below zero for 5 months of the year - and the ground surface is frozen solid during this time to ~3 m depth. It is important for me to understand how the model handles this when I am calibrating the simulated pressure heads to measured groundwater elevations.

Thanks
M

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Fri Apr 22, 2016 3:07 am

HYDRUS (2D/3D) does not consider freezing/thawing processes and thus it should not be used for temperatures below zero (Celcius). J.

mgoddard
Posts: 13
Joined: Wed Dec 23, 2015 11:00 pm
Location: Canada
Location: Netherlands

Re: ponded water with freezing temperatures

Post by mgoddard » Fri Apr 22, 2016 4:00 am

So could I approximate soil freezing by using an HcritA of -125 m during those months? This value was reported by K. Wanatabe in Water and heat flow in a directionally frozen soil, Hydrus Workshop, 2008, Japan.

Do you have any suggestions on what I could do to simulate soil freezing?

I have already removed any precipitation for these months to simulate snow fall.

Thanks,
M

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Fri Apr 22, 2016 2:48 pm

hCritA is rather unrelated to temperature. It simply affects the evaporation rate (a switch from the potential to actual values).

As I have written, HYDRUS does not consider freezing/thawing processes and thus cannot be easily used for such applications. I could make some quick modifications, such as reducing conductivity by several orders of magnitude, when the soil/air temperature drops below zero, and you would need to modify surface BCs to account for snow manually. Note that HYDRUS-1D has a simple "snow" routine (it snows when temperatures are below zero, snow stays at the soil surface, and then once temperature increases, there is a simple melting routine (based on degree day concept). I could perhaps add something as that to HYDRUS-2D. Let me know what you think.

Jirka

mgoddard
Posts: 13
Joined: Wed Dec 23, 2015 11:00 pm
Location: Canada
Location: Netherlands

Re: ponded water with freezing temperatures

Post by mgoddard » Sat Apr 23, 2016 6:26 pm

Thank you for the information regarding HcritA.

I would prefer to control the snow melt myself, as we have Site specific knowledge about sublimation, snow density, timing, etc. But I think that reducing the conductivity by several orders of magnitude is a great idea! This would simulate frozen soils and groundwater nicely. I'm looking for the pressure head to "flat line" during the winter months.

With regards to my original frozen pond question, I have been told that the lake never freezes due to man made influences. For the small ditches, I have successfully used a variable head BC to simulate winter and summer conditions.

Just a little more detail on our project: It is a semi-arid climate with around 350-500 mm ppt per year; we have yearly snow survey reports which are used to calculate the snow water equivalent; all the snow melts over a period of ~10 days in April which provides the majority of the deep infiltration; trees leaf out in early May; most of the summer rain is lost to the atmosphere or to vegetation (~85% ET); only large rain events (thunderstorms) produce sufficient ppt for deep infiltration; groundwater is near surface with 12 out of 14 monitoring wells screened within the frost line (< 3 m depth).

M

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Sun Apr 24, 2016 7:19 pm

I can send you an exe (if you send me a direct email), with K reduced when temperature drops below freezing. J.

tmalmir
Posts: 7
Joined: Tue Apr 10, 2018 6:41 pm
Location: Canada

Re: ponded water with freezing temperatures

Post by tmalmir » Thu Apr 12, 2018 5:25 pm

Dear Jirka,

I have the same problem. I am simulating a landfill biofilter in Canada, Quebec, by Hydrus 2D in order to analyse infiltration rate and so on. I used precipitation data of 2017. The results show that water content infiltrating to the biofilter increases highly, starting in February. However the fact is that soil is frozen approximately from December to March which means no infiltration. Here you have mentioned that you could add something or be any kind of help. Your advise will be so useful for me as well.

Moreover, on the discussion platform, we read something about decreasing the hydraulic conductivity by several order of magnitude when the temperature is below a certain threshold (not necessarily zero Celsius). How could this be implemented?

Thank you
TM

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Wed Apr 18, 2018 9:28 pm

HYDRUS-2D does not handle processes of freezing and thawing. Although you can specify negative temperatures when modeling heat transport, the code does not account for the heat of fusion when crossing the zero temperature, and thus its temperature prediction are not accurate for such conditions. In the past I was reducing conductivity when the air temperature was below zero, but this usually led to numerical instabilities, and thus I abandon this approach).

If you want to use HYDRUS for your purpose, I suggest that you simply preprocess your meteo data and do not apply any surface flux when the soil is frozen. You can then use some simple external melting model (e.g., something similar to the degree day melting model used in Hydrus-1D) to release surface water once air temperatures are above zero.

I hope this helps.

Jirka

Otthman
Posts: 18
Joined: Thu Nov 07, 2019 8:04 pm
Location: Canada

Re: ponded water with freezing temperatures

Post by Otthman » Tue Jan 19, 2021 7:37 pm

Hi M. Jirka

You suggest that we release the water slide by using an external melting model. I intend to do so in my case study.
You also say that we can decrease the hydraulic conductivity of the soil. How I can do this?
I would like to reconcile the two because I find it relevant in that melting snow does not automatically lead to infiltration. In reality, the ground can always be frozen and prevent infiltration. What do you think about it ?

Thank you.

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Tue Jan 19, 2021 8:33 pm

The melting model in H1D: The existing snow layer (if it exists) melts proportionally to the air temperature. The proportionality constant is the Snow Melting Constant. This is the amount of
snow (given in length units, such as cm of water; e.g., 0.43 cm) that will melt during one day for each oC.

Reduction of K as a function of temperature: There is only an option to adjust K as a result of temperature dependence of viscosity. However, this is only for positive temperatures (>0C).

J.

Otthman
Posts: 18
Joined: Thu Nov 07, 2019 8:04 pm
Location: Canada

Re: ponded water with freezing temperatures

Post by Otthman » Tue Jan 19, 2021 10:05 pm

Thank you for your response. In previous conversations you said that hydrus 1D has a simple routine based on the degree-day concept for melting snow. I would like to know more about the different variables such as melting temperature, snow density and melting factor. Can the user calibrate them?

Thank you

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Tue Jan 19, 2021 10:29 pm

This is only in the HYDRUS-1D package, not in HYDRUS (2D/3D). The detailed description is in the HYDRUS-1D manual. No, these constants cannot be calibrated directly using the internal H1D inverse procedure (but certainly can using external optimization). J.

Otthman
Posts: 18
Joined: Thu Nov 07, 2019 8:04 pm
Location: Canada

Re: ponded water with freezing temperatures

Post by Otthman » Wed Mar 17, 2021 5:32 pm

Good evening Mr. Jirka

In the past you have used the reduction of soil conductivity to imitate frozen soil in modelling. You abandoned it because it created numerical instabilities. I would like to know your assessment of this approach. Is it because it creates numerical instabilities and also does not take into account the freezing and thawing of snow?

Jirka
Posts: 5268
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: ponded water with freezing temperatures

Post by Jirka » Thu Mar 18, 2021 4:03 am

The Hydrus code has never been intended to support conditions with subzero temperatures. Initially, I thought that I can partially resolve this constrain by reducing the conductivity for subzero temperature, but once I saw that it did not help much, I abandoned that attempt. Certainly, this simple approach could not account for the energy loss/gain due to freezing/thawing, and thus it has always been a gross oversimplification. J.

Otthman
Posts: 18
Joined: Thu Nov 07, 2019 8:04 pm
Location: Canada

Re: ponded water with freezing temperatures

Post by Otthman » Thu Mar 18, 2021 10:44 pm

Thank you for your reply. I am asking these questions to see if I can combine the two methods together:

- Reducing conductivity
-Release the water equivalent of the snow with an external melting model.


Other things that interest me are the following: <<HYDRUS-1D has a simple "snow" routine (it snows when temperatures are below zero, the snow stays on the ground surface, then when the temperature rises, there is a simple melting routine (based on the degree-day concept). Perhaps I could add something like this to HYDRUS-2D>> (one of your answers to a student).

I am interested in the fact that one can add the snow routine from Hydrus -1D into Hydrus-2D. How do you do it?

Thanks,
O.

Post Reply