Transpiration in HYDRUS-1D

A discussion forum for Hydrus-1D users.
Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Transpiration in HYDRUS-1D

Post by Goali » Tue Dec 01, 2020 4:31 pm

Hello,
I have a question about the simulation of evapotranspiration of plants. My goal is to simulate the daily evapotranspiration for one year. My model is already running quite well. I have all meteorological data and also the daily root depth and plant height as well as daily values for the LAI. Unfortunately, in the output file "T-Level", the transpiration of the plant during the growing season falls to zero. From one day to another.
Does anyone have any idea what could be causing this?
I would be grateful for any help, as this is a thesis.

Best regards Goali

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Wed Dec 02, 2020 3:34 pm

You likely have a problem with the Atmosph.in file. You must have made some formatting error there. J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Wed Dec 02, 2020 4:16 pm

Thank you very much for your answer!

It's a general attitude, but maybe it's my mistake.
Do I have to use "comma" or "dot" as decimal separator by the atmospheric parameters?
Does HYDRUS also accept negative values as Tmin?
In winter the air temperature is below 0 °C.
Otherwise I have the following atmospheric data [unit]:
Precipitation [mm]
Global radiation [J/cm²] --> converted to [MJ/m²]
Tmax,Tmin [°C] Humidity [%]
Wind [m/s] --> Why does HYDRUS calculate with km/d?
Sunshine duration [h]
Albedo is given and constant daily

Can modelling with HYDRUS cover several years? Like 60 years or more?

Thank you for help!

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Wed Dec 02, 2020 7:30 pm

You need to keep the same formatting as generated by GUI and as shows in the examples posted on the Hydrus website (including the same units for meteorological variables; as described in the manual). You need to simply extend these files (as described in the help). Many people have done that successfully, running simulations tens and hundreds years long. There is no limit on the length of time. J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Thu Dec 03, 2020 1:55 pm

Okay. Thanks a lot! I am still searching and so far I have not been able to find the mistake.
I have one more question:
I plotted the daily actual Transpiration and the daily amount of precipitation. It is noticeable that the actual Transpiration is zero or decreases when there is precipitation. I think it has to do with that and on days with precipitatoin the infiltration is very high. Would it be possible that in HYDRUS no transpiration takes place when precipitation falls?
Attachments
actual Transpiration and Precipitation.jpg
actual Transpiration and Precipitation.jpg (88.26 KiB) Viewed 1913 times

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Thu Dec 03, 2020 5:17 pm

There are multiple reasons why transpiration may go down during rainy days. For example:
1. Potential transpiration is likely smaller than on sunny hot days since it is likely a cloudy cold day.
2. If the rain is large enough, the pressure head in the root zone can go above P0 or POpt. See the stress response function.
3. If you consider interception. Transpiration is first taken from intercepted water before root water uptake is initiated.
And so on.

J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Tue Dec 22, 2020 4:24 pm

Dear Mr Simunek,

Thanks to your help! My model runs very well for a 1-year simulation. What value can I specify for the interception? I have specified 1mm for winter wheat and in the output file the interception was then 8mm. That’s way too high. In addition, I have a problem with the year-over-year modeling. Here I always get the following error message:

Numerical solution has been stopped since it has not converged in 10 consecutive time steps !

What is the cause of this? Which settings do I need to change? I want to model evapotranspiration every day for 26 years. That is a total of 9861 days.

Thank you for your help!

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Tue Dec 22, 2020 4:52 pm

The maximum interception is equal to a_intercept * LAI, where a_intercept is an input value (approximately 0.25 mm) and LAI is the leaf area index. The actual interception is given by equation 2.78 in the User Manual. You should not get more than that.

If you have a problem with convergence between different years, you likely are entering something wrong with respect to time. Note that time in both atmosph.in and meteo.in files need to be increasing (cannot go back to zero at the beginning of the new year).

Jirka

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Tue Dec 22, 2020 5:43 pm

Okay. Thanks!

Regarding the time, I have days as the time step and the days take starting from day 1 to day 9861 (26 years).
In the Time Information field I have the following information.
Initial Time (day): 1
Final Time (day): 9861
Minimum time step (day): 1e-005
Maximum time step (day) 0. 5

In the AtMOSPH.IN and METEO.IN file the time is also correct and I have no gaps. The number of days is increasing until 9861.

Do you have another idea?

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Tue Dec 22, 2020 7:04 pm

Then you do have some convergence problem. You need to see what’s happening at the time when you get nonconvergence and figure out why it is happening. The entire profile gets saturated, etc? J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Fri Jan 22, 2021 7:57 pm

Hi, 
I have made good progress in modeling transpiration  with HYDRUS. As a quick reminder, I am modeling the  current ET of winter wheat with HYDRUS for my master thesis. Now at 2 locations this has worked out without a problem with your help. Thanks a lot! I am now having  problems at the third location, and that is modeling the  AET of winter wheat on a loamy sand (lS). HYDRUS  breaks off here immediately. If I change the soil type to a silt, then the modeling works. What can it be? I'll give you all the meteorological data. I model the PET after  Penman-Montheith. Many thanks for your help!

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Fri Jan 22, 2021 9:16 pm

If the program works for other textures but not this one, then it is only a numerical issue and you need to adjust various parameters. You may try smaller time steps, lowering hCritA and wilting points, etc. J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Mon Jan 25, 2021 7:49 pm

Thank you for your tips! It works on a modified hCritA. What is meant by hCritA? I have to change this every time I do new year-over-year modeling for this location. Is that right?

Jirka
Posts: 5216
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Transpiration in HYDRUS-1D

Post by Jirka » Mon Jan 25, 2021 9:45 pm

hCritA – see the manual and help. J.

Goali
Posts: 9
Joined: Tue Dec 01, 2020 4:18 pm
Location: Germany

Re: Transpiration in HYDRUS-1D

Post by Goali » Sat Jan 30, 2021 10:23 am

Hello Jirka,

my modelling of the actual ET with a loamy sand is stable. Only now I have the problem that vRoot during the growing season on some days is just over zero mm. This is, of course, much too low, as shown by the comparison with another soil water content model and lysimeter data. My guess is it’s the saturated conductivity Ks. If I change this value, I get better results. E. g. for another soil type the saturated conductivity is lower . . .
Is there a span for the soil hydraulic parameter for each soil type? Or how can I adjust these parameters for a loamy sand? Thank you for your help!

Post Reply