Simulation is very slow or i got NaN in the simulation run

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!
Post Reply
ahmed_almayahi
Posts: 19
Joined: Tue Jun 12, 2018 12:49 pm
Location: Oman

Simulation is very slow or i got NaN in the simulation run

Post by ahmed_almayahi » Sat Aug 11, 2018 5:12 pm

Dear Hydrus users,
I am running a simulation of root water uptake in a box of silt loam (length=40 cm, width=40 and height=28) for 30 days. I set a mesh size of 2 cm. In the time variable BC, i set evaporation= 1.5 cm/day, transpiration=0.8 cm/day for sweet corn and area associated with transpiration=1,600 cm2 (the entire surface area of the box =40cm*40cm).
However, when i run the simulation, it was either very slow or sometimes i got NaN in the calculation. NaN (as you can see in the attachment) appears after 2 days of the simulation in the cumulative atmospheric BC, hAtm and hroot.

I red many comments in the PC-discussion forum and i tried to reduce the mesh size, or refined it, but that still did not work.

Can anyone advice me on how to solve the problem of slow simulation?
Also the problem of getting NaN?

Appreciating your helps
Attachments
Slide9.PNG
Slide9.PNG (567.67 KiB) Viewed 343 times

ahmed_almayahi
Posts: 19
Joined: Tue Jun 12, 2018 12:49 pm
Location: Oman

Simulation is very slow, take long time and i got NaN in the simulation run

Post by ahmed_almayahi » Mon Aug 13, 2018 11:13 am

Dear Hydrus experts,
I am still unable to solve my problem with simulation of root water uptake in a box of silt loam (length=40 cm, width=40 and height=28) for 30 days. I am using HYDRUS 2D/3D professional version 2.
I specify 2 cm mesh size, no flow BC from all side of the box except the surface. I assume full area associated with transpiration (1600 cm2) and no evaporation, i used sweet corn and i calculated the transpiration rate based on weather data using cropWAT. The transpiration which i used is 0.8 cm/day. The initial condition was full saturation of the soil which was imported from previous simulation for the same model. The parameters for root water uptake were
maximum rooting depth=26 cm
depth of maximum intensity= 13 cm
parameter pz=1
maximum rooting radius in x= 35 cm
radius of maximum intensity in x= 0 cm
parameter px=1
maximum rooting radius in y= 35 cm
radius of maximum intensity in y= 0 cm
parameter py=1
The problem is when i run the simulation it was slow or take long time, it then give me NaN in the cumulative atmospheric BC, hAtm and hroot as shown in the attachment of the previous post.
I tried the following to solve the problem but still i have the same problem:
1. reduce mesh size from 2 cm to 1.5 cm
2. reduce the transpiration rate from 0.8 to 0.4 cm/day
3. reduce the area associated with transpiration from full to 50 %
4. change the values of root uptake parameters
5. in the hydraulic model, i used VG-M with air entry value of -2 cm
6. in the time BC i changed hcritA from 10,000 to 1,000
7. reduce the simulation time from 28 to 25 days
8. change maximum time step from 5 to 0.5
I still can not understand why this problem happened and how to solve it. I need your advices
I attached the project, to inspect the problem and suggest.

Appreciating your helps

Ahmed
Attachments
st3_hw_dry_2.5atm_ruptake.rar
(1.95 MiB) Downloaded 30 times

Post Reply