Van Genuchten: effective/total porosity & Se

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
tjoudega
Posts: 4
Joined: Mon May 22, 2017 3:41 pm
Location: Austria

Van Genuchten: effective/total porosity & Se

Post by tjoudega » Tue May 23, 2017 9:27 am

Hello, I am trying to calibrate a 3D model on the drawdown during a period of pumping. I am trying to get a K value, leaving porosity constant.

1. However, is the porosity in the Van Genuchten hydraulic model total porosity, or effective porosity? If it is total porosity, how is the effective porosity calculated?
2. Does this have something to do with the effective water content Se? (I couldn't find how this parameter is calculated.)

Thank you,
Thomas

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Van Genuchten: effective/total porosity & Se

Post by Jirka » Tue May 23, 2017 8:33 pm

The porosity in the van Genuchten hydraulic model is the total porosity (entire pore space). In the soil physics, we do not recognize the effective porosity in the same sense as in groundwater hydrology (where it is not well defined either; and depends not only on texture, but also the groundwater table depth, etc). In the soil physics we have effective saturation, which is defined as (theta-theta_r)/(theta_s-theta_r), and which varies between zero and one. I guess you can estimate the effective porosity based on the texture and depth of groundwater (L), i.e., equal to theta_s-theta(-L).

J.

tjoudega
Posts: 4
Joined: Mon May 22, 2017 3:41 pm
Location: Austria

Re: Van Genuchten: effective/total porosity & Se

Post by tjoudega » Thu Jun 01, 2017 8:00 am

Dear sir,

Thank you for your reply. Another quick question:

When simulating a well screen, while the vertical mesh is unequally distributed, should I calculate the nodal recharge for each node, depending on their spatial distribution? Or should you enter a total pumping value value for all the nodes inside the well screen?

Thank you,
Thomas

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Van Genuchten: effective/total porosity & Se

Post by Jirka » Thu Jun 01, 2017 10:06 pm

When you specify the flux BC, you need to enter the flux in [L/T] (e.g., cm/d) and the code will calculates nodal fluxes by multiplying this flux with the length/area associated with each node (in 2D/3D, respectively). J.

tjoudega
Posts: 4
Joined: Mon May 22, 2017 3:41 pm
Location: Austria

Re: Van Genuchten: effective/total porosity & Se

Post by tjoudega » Fri Jun 02, 2017 11:30 am

The well screen is in the center of the domain (in 3D layered), so I can only simulate this pumping with the function of nodal recharge. This cannot be entered in L/T, as it asks for L^3/T. I thought that I have to do it this way, because it's not possible to assign a BC to internal nodes, understandably.

Or should I, after adding these points in the section 'nodal recharge', edit them in 'initial conditions'? That's the only other place where I can see these nodes in the FE-Mesh.

Thomas

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Van Genuchten: effective/total porosity & Se

Post by Jirka » Fri Jun 02, 2017 3:48 pm

Indeed, boundary conditions can be assigned only to boundary nodes. When you use internal nodes with nodal recharge, then the units are L3/T. You should be able to see these nodes (and modify recharge values) under Domain Properties->Nodal Recharge. J.

tjoudega
Posts: 4
Joined: Mon May 22, 2017 3:41 pm
Location: Austria

Re: Van Genuchten: effective/total porosity & Se

Post by tjoudega » Wed Jun 07, 2017 8:42 am

I understand. This brings me back to the original question:

In the pumping well, the total pumping is 2000 m^3/day. Should I assign 2000 L^3/T to all the nodes inside the well screen, or should I divide this 2000 by the amount of nodes?

And if the latter is true, should I account for the spatial distribution of these nodes? I.e. the layers in my 3D layered domain are vertically unequally distributed (depending on the different geological layers), so simply dividing 2000 by the amount of nodes could lead to wrong estimates. Is this thought correct?

Thank you (and sorry for being unclear in my question before),
Thomas

Jirka
Posts: 4712
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Re: Van Genuchten: effective/total porosity & Se

Post by Jirka » Wed Jun 07, 2017 4:06 pm

You should divide your flux proportionally to each node, depending how large area/volume they represent. J.

Post Reply