Storage coefficient [1/L] in saturated zone

A discussion forum for old Hydrus-2D & Meshgen. Happy Posting!
Download PDF document with all topics (February 8, 2008, 1.9MB):
http://www.pc-progress.com//Downloads/F ... on_H2D.pdf
Post Reply
Hector
Posts: 36
Joined: Sat Mar 16, 2002 4:02 pm
Location: Germany

Storage coefficient [1/L] in saturated zone

Post by Hector » Tue Jul 27, 2010 2:40 pm

Hi out there,

I wonder if water storage is considered in hydrus2D in a completely saturated domain (horizontal and vertical plane). The vG Model yields C=0 below the phreatic surface however one would expect some small value depending on the materials compressibility.

If there is no consideration, is there a possibility to assign some value to the storage (or capacity in the terminology of Richard's equation) via the soil parameters?

Regards

Héctor

Jirka
Posts: 4716
Joined: Sat Mar 16, 2002 3:47 pm
Location: USA
Location: Riverside, CA

Post by Jirka » Tue Jul 27, 2010 2:51 pm

Héctor

Neither the old HYDRUS-2D, nor the new HYDRUS (2D/3D) consider the storage coefficient in the saturated zone. However, I do have it implemented in the new code and only have it disabled. If you have a license for the new code, I could send you an executable module with this option. Sorry, not for the old one. While we still support that, we are not developing it further for some last 5 years.

Jirka

Hector
Posts: 36
Joined: Sat Mar 16, 2002 4:02 pm
Location: Germany

Post by Hector » Wed Jul 28, 2010 9:04 am

Jirka,

thank you for your hints. I decided to perform the modifications in the code (I guess the old one) since I am interested in the incorporation of a pressure dependent storage coefficient (or capacity term) to represent air entrapment below the phreatic surface. Even a very small volume of entrapped gas may influence water pressure propagation in the "saturated"-zone.

My Hydrus code now reads a table of hydraulic properties not only for suction but also for pressure (below the phreatic surface). I assigned C a constant value for pressure and checked that against a quasi 1-dimensional analytical solution (was a bit surprised that Excel has a complementary error function already built in ;-). The results showed quite a strong dependence on temporal discretization but yielded for small time steps almost perfect agreement.

Apparently a non-zero C-term at saturation yields much "smoother" runs (saturated/unsaturated) than the conventional approach with C(h=0)=0 - at least for certain cases and soil properties.

Héctor

Post Reply