hydraulic conductivity of crushed stone 4

[murdock]

This question is kind of a mash-up of a this forum and the hydrocad/storm modeling forum. I have been asked to model storm flow down a swale that has check dams installed. I am planning to use hydrocad's pond node and use the exfiltration device for an outlet. I need to spec the hydraulic conductivity (velocity in hydrocad-speak) of the crushed stone used for the dam (the regulator is concerned with ponding behindthe check dam and the creation of nuisance conditions).

I did a bit of hunting around and found a reference to a book (On-Site Stormwater Management: Applications for Landscape & Engineering by Thomas N. Debo and Bruce K. Ferguson, Feb 1990) which stated the rates for 2" crushed stone were 50,000 in/h.

If I plugged this figure into my model, there would not be any appreciable ponding behind my check dam, which does not sound right on the first pass. Is it? Am I missing a step in this process? Also, I am trying to square in my head this K-value, and the stone's effect of reducing the velocity of water in the swale. Is the K-value unrelated to velocity reduction?

I look forward to hearing your insights.

 

[RWF7437]

50,000 inches per hour !!?? Unheard of. See the link:

http://www.p2pays.org/ref/37/36132/manual/slides/chapter5a/img1

Reread your reference. Could this be inches per day or week or month ?

Generally, check dams are designed to reduce the velocity to near zero. You may be able to model this in either HydroCad or Hydrographs 2007 as a series of small detention basins. Try posting on the HydroCad forum for expert advice.

Don't know what you mean by "K value". Try spelling out any abbreviation, acronym or any similar term before using it. Works wonders for communication.

Loosely placed, 3" round rock typically will have about 30% voids. Why not take a smaple of your rock and measure the infiltration rate ?

good luck

 

[RWF7437]

http://www.intelisolve.com/home.htm

 

[francesca]

Heaven forbid there should be ponding behind a check dam... next thing you know it'll be trapping sediment!

 

[RWF7437]

http://en.wikipedia.org/wiki/Hydraulic_conductivity

 

[murdock]

Thanks for your replies. I had exclamation points and question marks coming from my head when I first read that citation, too. Here's where it came from (PDF). See the bottom of page 2. The wikipedia link (thanks, RWF7437) and my groundwater and wells book had numbers in the same order of magnitude.

RWF7437, by K-value I mean the hydraulic conductivity. Measuring it myself would be a fine check, but I wanted to see if there was any literature on it before I invested the time and effort. BTW, your p2pays.org link doesn't work, could you post it again?

I have a good idea on how to model this in hydrocad, so I decided to post it here.

 

[RWF7437]

Murdock,

Sorry about the dead link. The Wikipedia link tells you just about the same thing so I won't try to post it again. These things suffer "webrot" from time to time.

As a check, you can also model this in Hydraflow Hydrographs 2007. For some things it is easier to use than HydroCad and produces nicer looking, more succinct reports.

Meanwhile, use the best references you can find and verify them by your own testing. Shouldn't be that much time and trouble and you'll be following in the footsteps of Gallileo, of Papal fame.

As Franseca reminds us, next thing you know rainfall will be running off the land creating rivers and streams which can become contaminated with bugs and fish.

good luck

 

[ptmoss]

FYI, from the table in wikipedia: unconsolidated sand & gravel "K" = 100,000 ft/day (the upper limit of the range), which stated another way equals.....50,000 in/hr

 

[RWF7437]

ptmoss,

Would you use this ( upper limit ) to design a check dam ? If so, how would it be used ?

 

[RWF7437]

Murdock,

You had originally asked "Is the K-value unrelated to velocity reduction?".

I think the answer is probably "No". A check dam works by effectively decreasing the slope of the ditch. Of any water drains out of the ditch through the bottom it is likely it is a minimal amount and if you want to model it you would treat it as "infiltration". The "K-value " is NOT the infiltration rate which depends on the material of the ditch bottom; not the material of the check dam itself.

The amount of water flowing through the check dam is usually ignored. Even if you wanted to account for it you would need to use the infiltration rate ( also expressed in inches per hour ). For well graded, compacted crushed aggregate it seem very unlikely that it is as high as 50,000 in/hr. An infiltration rate ( not K) of 1-inch per hour would be considered very well drained. 50,000 in/hr corresponds to a velocity of about 1.15 feet per second. This seems very fast even for groundwater in a good, porous, aquifer. Unless you are worried about how long it may take the water behind the check dam to drain you probably don't need to calculate this. If it is necessary to drain the water behind the dam quickly you can always install a small pipe through the base of the dam.

If the concern is mosquito larvae, a drain time of a few hours to as many as 72 hours is widely used in southern California and Arizona, I believe.

good luck

 

[RWF7437]

Also, in the HydroCad program click on the Help button and search on "exfiltration". Note the probable exfiltration rates listed there.

In the end, I think you don't need to know this in order to design, or analyze, check dams.

good luck

 

[ptmoss]

RWF7437, No I can't see myself using numbers like that for anything. Although I have seen water moving through native sand and gravel like that. Just once, though. It was in a test hole for a septic system - we hit a vein of sandy gravel and cobbles about 5' deep and it was like a river flowing through it. Moved a bit more than 50' away laterally and there was just a little water seeping.

I've tryed modeling check dams in swales to see the impact and it never really worked for me. My areas were just too small. Just from observations though, if your swale is flat then you'll get some ponding/infiltration behind them, which would certainly be a water quality benefit for that first half inch of rain or so.

I also assume checkdams will get silted up to a certain extent, so just that reason alone would keep k-value slower than stone alone.

Right or wrong, I just ignore them in my calc's.

 

[psmart]

Although check-dams are definitely beneficial for low-flow water quality events, they are often overwhelmed by a 10-year storm or greater. Unless the available storage is a "significant" fraction of the total inflow volume, it won't have much effect on the hydrograph routing.

For larger events, it may work better to use a reach with a higher Manning's value, in order to represent the rougher surface created by (frequent) check-dams.

Check out a Manning's table. For example: "Natural Streams, Clean, winding, some pools and shoals" n=0.040 The value could increase to 0.050 or more.

 

[psmart]

As a side note, the sample exfiltration rates (previously) listed in the HydroCAD help system were admitedly on the low side, and were removed some time ago. Actual rates can vary greatly, as proven in this thread. HydroCAD will now accept any value from 0.001 to 999 in/hour, and larger values can be used by increasing the upper limit on the Settings|Units screen.

 

[bltseattle]

Going back to murdock's original post, if you use a high Ksat (permeability) and apply Darcy's Law, for typical check dam dimensions, the estimated ponding is negligible. To achieve the ponding behind a check dam (thus flattening the slope of the channel as RWF discusses above) either Ksat must be lower than reported in the standard tables for open gravels or Darcy's law does not apply.

In "Seepage, Drainage, and Flow Nets," Cedergren 1989, it states that "in the analysis of seepage in coarse sands and gravels Darcy's Law is not strictly applicable. Forchheimer (1902) found the frictional resistance of pervious gravel to be: (eqn 3.4)"

[Δ]H / [Δ]L = 1.77 / 1000 * V + 3.18 / 10000 * V[²]

where H is the head difference, L is the length, and V is velocity in meters/day.

When you apply this formula for the check dam scenario, ponding is predicted, as is observed in actual installations.

I am not suggesting anyone waste their time running calcs for check dam seepage and ponding - designers should use the reviewing agency's standard check dam design and move on to more important items.

 

[runoff]

Murdock;

Also, check the Virginia best management practices manual. They have a great section in the engineering references that outline design conditions for grassy swales. I just finished a project that included a grassy swale design and hydrocad exfiltration numbers ended up being a little negligible as PSMART stated above. The project was constructed based upon their design calculations and I visited the site after a very large event and it worked like a charm.

You might even consider a biofiltration design that acts very similar to a grassy swale but has the vegetation to retard heavy flows and allow for 72 hour exfiltration. Just a thought.