Permeability Testing

[geotechguy1]

Hi All,

I have a few queries regarding permeability testing and applicability.

Geology for my site consists of ~6m of clayey silt, overlying a saturated sand and gravel deposit. The sand and gravel deposit is the source of the towns groundwater so I assume that the recharge rate must be fairly high. I have a couple of wells installed, (50mm diameter, with slotted lower casings). One well is installed with the slotted pipe in the gravel layer, sand around the screen, and then is sealed from the top of the screen to the surface with bentonite. The other well is installed in the clayey silt, similar installation details.

It seems to me that there are two ways I could test the permeability: I could add water to the pipe, and then measure how long it takes for the head to drop; or, I could pump/bail the water out and measure how fast it takes to recharge.

Would both methods give the same permeability?

 

[geomane]

Permeability/hydraulic conductivity is based on Darcy's Law v=ki. Do you know the boundary conditions?

This is similar to discussions we have in our company. We perform DRI testing and some engineers will report an unsaturated vertical hydraulic conductivity. The DRI estimates infiltration not permeability in my opinion. The boundary conditions are unknown in a DRI.

I'm interested to hear feedback from some of the more experienced engineers on here.

 

[GeoPaveTraffic]

In theory you would get the same result by adding water, falling head test, and by pumping water, rising head test. Note, you normally would not add water since you could contaminate the groundwater. In stead you would use slug instead to create an artificial rise in the groundwater in the the well.

Given your description, a slug test is likely the best option in the clay and an pumping test in the sand/gravel.

Mike Lambert

 

[fattdad]

You will only get the same results if the screened interval is entirely below the phreatic surface during the rising head test. If the drawdown from slug removal exposes any part of the well interval below the bentonite plug you have different boundary conditions for the rising head test.

I would also encourage you to do grain-size distribution and correlate to permeability as a reference point to your test.

I would also encourage you to look at the various test conditions of Hvorslev.

f-d

ípapß gordo ainÆt no madre flaca!

 

[Ron]

If you force water into the well, you will increase the interstitial head loss, thus giving you a lower permeability. If you pump from the well and allow a slow recharge of the well, you will get a more representative permeability of the soil (IMHO )

 

[geotechguy1]

Ron,

My thinking was that since the purpose of the test is to provide the town with a site-specific permeability they can use to determine stormwater infilitration, the 'adding water' test would be more applicable since it seems directly analogous to what they are trying to do, even if the permeability is lower

 

[fattdad]

using horizontal flow boundary conditions to forecast vertical infiltration is a bit simplistic. Anisotropy in soils often returns a 2 to 4x factor between horizontal and vertical permeability. This is almost codified in the Federal guidance on PVD design, where Ch/Cv is shown as a range from 2 to 4. Many specialty contractors actually use 3.5, based on their experiences in ground improvement and what's considered, "Allowed" by the FHWA.

This is one reason that the Johnson or Guelph permeameters are used in infiltration design.

f-d

ípapß gordo ainÆt no madre flaca!