Set initial condition spatially in FE mesh

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
s.kumar2
Posts: 3
Joined: Wed Dec 16, 2020 7:09 am
Location: India

Set initial condition spatially in FE mesh

Post by s.kumar2 » Wed Dec 16, 2020 12:11 pm

We are facing problem in setting initial condition spatially in FE mesh nod for simulation in Hydrus 2D. When we try to set initial condition by selecting nodes in FE mesh, it shows error "there are difference between the values of the Fe mesh and geometric object" (snapshot attached). Kindly help us in rectifying error
Attachments
Picture4.jpg
Picture4.jpg (127.28 KiB) Viewed 1027 times
Picture3.jpg
Picture3.jpg (169.57 KiB) Viewed 1027 times

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

Re: Set initial condition spatially in FE mesh

Post by Jirka » Wed Dec 16, 2020 5:36 pm

When you select in the “Domain Type and Units” dialog that you want to “Edit domain properties, initial and boundary conditions on geometric objects” (rather than directly in the FE mesh), the program will check and compare what is specified on “Geometric Objects” and “FE mesh”, and if there is a discrepancy, it will issue this warning (to let you know about this discrepancy). The program will then proceed with using the information that is specified at the FE mesh. Thus, you do not have a problem, this is just a notification. J.

s.kumar2
Posts: 3
Joined: Wed Dec 16, 2020 7:09 am
Location: India

Re: Set initial condition spatially in FE mesh

Post by s.kumar2 » Mon Dec 28, 2020 11:08 am

We are simulating sub surface drainage impact on water table fluctuation in field condition. The dimension (geometry) of project is 6000cm x 200 cm height (depth). Atmospheric boundary condition is assigned to upper surface. Water table depth below land surface is 46 cm i.e. pressure head at observation nod is 154 cm.
Bottom and side boundary is assigned as no flux and subsurface drain (tile) opening has assigned as seepage flux.
When an irrigation event of 6 cm is assigned “numerical solution is stopped as it has not converged in 10 consecutive time”
We want to know that does hydrus 2d not have option of “atmospheric pressure with surface layer” (ponding) as it is available in Hydus 1d.
In real field condition, when irrigation apply in waterlogged/saturated field, it remains on surface for few days.
May pl look into and suggest possible solution.
Attachments
Picture1.jpg
Picture1.jpg (139.71 KiB) Viewed 1005 times

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

Re: Set initial condition spatially in FE mesh

Post by Jirka » Mon Dec 28, 2020 4:28 pm

HYDRUS-2D does not have an option of “atmospheric pressure with surface layer”, it only considers “atmospheric pressure with surface runoff”. However, the code can handle only infiltration excess runoff, it cannot handle the saturation excess runoff.

From the figure, it seems that your vertical discretization is on the order of 50-100 cm. That’s clearly not sufficient, see FAQ 8:
http://www.pc-progress.com/Documents/No ... zation.pdf

s.kumar2
Posts: 3
Joined: Wed Dec 16, 2020 7:09 am
Location: India

setting of boundary condition in sub surface drainage project

Post by s.kumar2 » Tue Dec 29, 2020 11:54 am

With reduced vertical discretization, problem is not solved. What we understand that numerical solution performs till pressure head < 200 cm at the observation nod. This shows that on upper surface, flux is < 0 it works and once it increases under irrigation/rainfall event which is more than the amount needed for saturating upper layer/portion, convergence problem occurs with atmospheric boundary condition as top flux.
In field condition where sub surface drainage is installed, there is situation that some top portion (layer) of soil (above drain) is unsaturated and when irrigation applies, positive head (water depth) is created on surface temporarily for some time. With time, it recedes under atmospheric drying and subsoil movement through drainage network (seepage flux).
Now we are trying to simulate this situation in hydrus 2d. We think that problem is with boundary condition as we tried to run the similar situation with example file available for drainage(pc-progress), but encountered similar problem of convergence error.
Kindly help us in this regard.

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

Re: Set initial condition spatially in FE mesh

Post by Jirka » Wed Dec 30, 2020 12:22 am

I do not have anything to add to my previous answer. J.

Post Reply