APEX v.1501

[Jindřich Zelinka]

Hello Luca, 

I'd like to ask you a question. I'm modelling data for 45 years, but it only calculates soil moisture for the first year, and for the remaining years I have 0. This means that all the yields for the crops are not coming out. So could you please advise me where could be my mistake ?

Thank you for your help,
Jindřich Zelinka

[Luca Doro]

It is difficult to understand what is happening in your simulation. Is the model completing the simulation and running for 45 years as intended?

Luca

[Jindřich Zelinka]

The model completes a simulation of 45 years. But I have a problem in that from 2 years to 45 years I have 0 soil moisture. But in the results water is present in the soil profile. The rainfall comes out as it should and the results show that water is moving through the soil profile, but I don't know why the soil moisture comes out 0 from year 2. 

Thank you for your help,

Jindřich  

Dne neděle 5. ledna 2025 v 21:08:07 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

Can you share with me the .OUT output file? Hopefully, it will help me to understand what is happening with your simulation. Also, can you share the output file where you have soil moisture equal zero starting from the second year?

Thanks,

Luca

[Jindřich Zelinka]

Here I send the .OUT file and MS5. In the MS5 file I have WP 0 from the second year.  

Thank you for your help.

 Jindřich

Dne úterý 7. ledna 2025 v 9:02:28 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

I don't see anything that is obviously wrong in the OUT file, but I can tell that the simulation was not correctly terminated because the program stopped before printing the final time stamp at the end of the OUT file.

I don't have a solution for now. Can you share all the input files of your project? Also, are you trying to simulate two subareas one next to the other and both of them converging to the same outlet?

Luca

[Jindřich Zelinka]

I sent you the data via uschovna.cz to your email. I am trying to model one area, but I must have misunderstood the subareas function. I am learning with the model in the result myself, my boss knows how to use EPIC.

Jindřich  

Dne čtvrtek 9. ledna 2025 v 8:27:38 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

Dear Jindrich,

I was able to run your input files on my computer and everything looks fine (the model runs without errors, but for some reason it is not printing the final time stamp. Since you want to simulate just one field/subarea, you can sue the included subarea file, where I have removed the second subarea.

Now, we can begin to work on the problem related to the very low soil water content. First, I want to ask you the location of your simulation. Then, do you have information on the observed potential evapotranspiration for your simulated area? Is it close to the simulated value?

Also, do you have information about the average annual precipitation in the area?

These are the first things we can look at if something in the water cycle is off.

Luca

[Jindřich Zelinka]

Hello, 

Thank you for fixing my subarea file. The location is in the Czech Republic in the city of Trutnov. I don't have the evapotranspiration data, so I can't judge if they are close to the simulated results. Regarding precipitation, I have daily precipitation (from the meteorological station at the site) for 45 years, so that I can calculate the annual average.

 
Thank you for your help.

Jindřich 

Dne pátek 10. ledna 2025 v 16:57:57 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

Please try to gather information about the annual potential evapotranspiration (PET) and actual evapotranspiration (ET) for a system similar to the one you want to simulate. This information will be useful during the calibration process.

For instance, if the model is simulating a too high potential evapotranspiration, it might be the cause of the low soil water content. By knowing the PET, we can adjust some parameters if adjustments are required, or we can search elsewhere for the cause of the problem.

Luca

[Jindřich Zelinka]

Potential evapotranspiration (PET) per year should range from 550-600 mm per year, in the model it is slightly over 800 mm per year (this value corresponds to another location in the Czech Republic). My boss noticed that after the end of the first year the soil properties (POR, FC and CURNT) changed rapidly, what could be the cause ? At the same time, if you look at the turn of 1965/1966 (31/12/65 vs 1/1/66), the RSSF, WYLD, QRB, and RZSW and GWST parameters change in leaps and bounds without any logical relationship with the end of the calendar year, and in the groudwater component the water storage in the root zone of the RZSW moves to groundwater (GWST) and to surface and lateral runoff (WYLD, RSSF) in one day.  

Jindřich

Dne pondělí 13. ledna 2025 v 8:45:27 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

The OUT file you shared with me earlier, indicates that you are using the Hargreaves approach to simulate the potential evapotranspiration (PET). Because the model is simulating a high PET for your region, you can change PARM(23) and PARM(34) to adjust the simulated PET to fall in the range you provided me.

I think that adjusting the PET will not solve the problem with the soil water content. Thanks to the information you provided in the last message, I was able to pinpoint the problem (soil characteristics suddenly changing from year one to year two). I did a test changing the method used to estimate the wilting point and field capacity (from Rawls dynamic to Rawls static) and, from a quick (and superficial) observation of the results, it seems that it solved the problem. So, there must be something that is affecting the estimation of the hydraulic characteristics.

Over the weekend, I will try to find some time to work on this. I will let you know as soon as possible.

Luca

[Luca Doro]

Dear Jindřich,

I was able to find and fix a bug in the code of the APEX model that was causing the abrupt change in the soil characteristics at the beginning of the second year of simulation.

I am trying to make the new version available on the Blackland Research Center website. While I am working on that, you can download the modified version from this link: APEX1501.exe

Please, let me know when you have downloaded the executable and how your simulation goes.

Best,

Luca

[Jindřich Zelinka]

Hello Doro,

Thank you very much for your help. The simulation is running and the problem has disappeared. And I've adjusted the PET as you wrote and it fits the modeled region better now.

Best,
JIndřich

Dne středa 29. ledna 2025 v 8:19:41 UTC+1 uživatel lucad...@gmail.com napsal:

[Luca Doro]

Thank you for letting us know.

Best,

Luca