distance from a creek when silt fence is not needed 2

[fivestring]

I am working on an erosion control plan for construction within a national forest. It would seem that at some point silt would dissipate before reaching a creek or stream if it is a certain distance from the construction. Could there be a rule of thumb based on the distance, slope and condition of native ground, and maybe soil type in which silt fence would not be needed?

 

[beej67]

The most important contributing factors will be soil erosivity, total amount of disturbed area, and the rate of runoff from the disturbed area, but I don't know of any 'rules of thumb' that aren't state specific. Erosion control laws are typically in the hands of the state. I'm not sure whether the state is still the permitting entity for federal parks though.

Where's the project?

If you're clearing a lot of trees and/or brush with your construction, I've seen brush barriers made of onsite vegetation work wonders here in the southeast. Sometimes those can be more effective than silt fence.

Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[fivestring]

Thanks for replying. I am in GA and this project is atypical for the EPD. They are allowing me some freedom in this design; not just following a checklist. If I can justify that silt fence is not needed when minor disturbance is a particular distance from a stream, I think they will go along with it. I understand that a lot of factors would be involved.

 

[beej67]

Heh, well, if it's Georgia my opinion is follow all the rules and minutia and call it a day, for CYA reasons.

No matter what sort of back-up you find for not having silt fence now, you only have two defenses in court if someone comes after you:

1) Turbidity tests showing no increase, or
2) Proof that you followed the checklist.

If you're a Georgia PE then you know how out of whack those turbidity tests can get, due to no fault of your own or of the contractor. I know our natural tendency as engineers is to find the 'best' solution, but the best way to stay out of court in modern litigious society is often to find the 'safest' one.

When the Georgia NPDES construction permit first came out, there was a lot of language in it that made engineers criminally liable for erosion failures even if their plans met every code. Like, jail-liable. A lot of guys spent a long time campaigning for revisions so that as long as the plans were done to a particular standard, the engineer was covered. Your safest bet is to follow the standard.

Now, beyond that, your post does bring up some interesting academic / engineering questions. All erosion control design for construction I've seen personally followed sets of pretty basic design standards, such as the Georgia Green Book, which crafts its criteria in the style of "always have Type XYZ silt fence at ABC location" and "all basins bigger than PQR must have a sediment pond" etc, without actually trying to model sediment discharge itself.

South Carolina is one of the few states in the southeast I haven't done any work in, but I heard a rumor that they require a SEDCAD model for construction plans, which is the sort of thing that would calculate what you're talking about, because it's a treatment-train calculation software for sediment loading. I'm not familiar with SEDCAD personally, but if they've got a treatment efficiency listed for undisturbed buffer, and it varies with width, and you can compare the SEDCAD predicted efficiency of the buffer with the SEDCAD predicted efficiency of Type C silt fence, you're well on your way. Go through their documentation and find out where they got their numbers from, and cite that to the EPD. But you absolutely ABSOLUTELY must get a signed hardcopy letter from someone reasonably big at the EPD saying they buy your approach. Again I stress, the important thing here is having the ability to deflect liability if it comes up.

Then again, by the time you've done all that you could have done the plans to standard. I also think you'll find that regulators aren't in the business of assuming liability, so the moment you try and get them to hang their hat on a decision that departs from a design standard, they usually balk and refer you back to the standard. Regulators never get fired unless they screw up a hard decision, which means the smart ones can get infinite job security by never making any decisions. It's just an unfortunate artifact of the business model.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[fivestring]

I totally understand what you are saying about the EPD. Yes, normally I would just follow the checklist and be done with it. This erosion control design is for miles of 3' wide trails in which the disturbance for installing silt fence may be more than the trail itself. On this particular project the EPD is allowing me to calculate sediment storage using silt fence, allowing type A silt fence, and not requiring permanent grassing. These are trails in the woods so these allowances would make sense. The cost of the silt fence would be more than the trail if it was installed on every bit of it. So my first inclination was to install the silt fence far below the trails. Then I got to thinking that in thick wooded areas, the silt probably wouldn't even make it to the fence. This led to my question. I will look into SEDCAD; thanks for the reply. Since this project is unusual to the EPD, they are allowing me some freedom.

 

[beej67]

Ahh, gotcha. Exposure is pretty low on that anyway.

Cue engineers from other areas of the country replying with "what the hell do you need silt fence for on a trail anyway?" You and I will probably have to just grin and shake our heads.

Another approach you might be able to take, depending on the alignment of the trails, is to go with rock filter dams in draws downhill. That's a more natural BMP and could conceivably be left in place after construction, to some environmental benefit.


Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[fivestring]

Thanks for the assistance. You did give me an idea to maybe try to tie this in somehow with the water quality spreadsheet and TSS loading. The woods could be recognized as as undisturbed filter.

 

[fattdad]

diversion ditches are effective too.

f-d

¡papá gordo ain’t no madre flaca!

 

[beej67]

Fivestring -

EPD might accept that, and you might be able to lean on that legally as well, but you should be aware that the TSS spreadsheets released by northgeorgiastormwater.com, as well as the local ones for Gwinnett and Forsyth and etc/etc, aren't really based on anything I'd call "reality," and the undisturbed buffer portions are a "credit" that's been purely manufactured out of thin air as encouragement for low impact design. They're also intended to be for permanent BMPs, not temporary erosion control. They're basically the CYA document that local munis run up the flagpole to EPD, so EPD can claim compliance with the non point source provisions of the Clean Water Act.

You could easily use the things to show that you meet water quality criteria for new development, but I'm not sure you could reasonably claim to calculate sediment storage with them. Especially since the Blue Book doesn't publish TSS removal efficiencies for temporary erosion control measures.

Those things will be particularly helpful to you if you end up having to go through the stream buffer variance process for your trails. Hopefully you're exempt though, I don't have all the exemptions memorized.




Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[fivestring]

I wasn't looking to use the spreadsheet or try to use this as sediment storage. I just mentioned it from a theoretical point. I am also looking at soil loss and transport to technically show that the sediment, which would be topsoil, would become part of the native ground at a certain point.

 

[beej67]

Possibly look into the USLE too, or modern variants, if you're going to develop your own spreadsheet approach.

http://fargo.nserl.purdue.edu/rusle2_dataweb/

I'd love to see what you come up with.



Hydrology, Drainage Analysis, Flood Studies, and Complex Stormwater Litigation for Atlanta and the South East -

http://www.campbellcivil.com

 

[chicopee]

In response to the original post, is there a set back requirement from the creek? if so at the setback limit.

 

[francesca]

I worked on a site in Tennessee on the shores of Kentucky Lake. We were able to place sediment traps instead of sediment ponds even though the "rules" required sediment ponds because the ponds would've had to be sited in undisturbed, possibly virgin, woods. It would've been far more disruptive to put them in and disturbance was going to be minimal. It was all academic in the end, though. It was the summer of 2007 and a luxury "holiday" residential subdivision. The client went belly-up and we didn't get paid.

 

[TerryScan]

What is the actual trail material/construction? If they don't buy your argument for no control:

As an alternative to silt fence, many jurisdictions have details for perimeter dikes for sediment control. I would proposed a very small version consisting on the scraping of the trail bed on the downslope side of the trail. I would detail this miniature version.

 

[fivestring]

The trail will be left natural (no grass or additional surfacing). I don't think GA EPD will go for the perimeter dikes, although the trail builder will probably do that anyway. Thanks for the post.