Estimated Soil Loss in a Swale

[murdock]

first: sorry if this comes up twice, I think the post was eaten the first time around.

I have been tasked with estimating the amount of yearly soil loss in a roadside swale. The swale is approximately 1,100 feet long, and drains about 12.5 acres in a low-density residential area. The watershed has a steep (10-14%) section towards the top, and flattens out to about 4% for the rest of the area.

I tried several techniques, including the simple method, RUSLE equation, and measuring the depth of rills and calculating soil mass lost based on the volume removed. The results range from about 1,600 lbs./yr using the simple method to 20,000 lbs./yr using RUSLE to 160,000 lbs./yr. using rill measurement. I have confidence in my watershed calculations, but the direct measurement of rills were approximated as best as I could (as you can imagine, the channel was of varying geometries). The 160k lb/yr number seems way out of range to me, but this calculation is new to me and I don't trust my gut feelings yet.

Which one of these techniques seems best suited to estimate the soil loss in my swale? Can you identify where I may have made some incorrect assumptions? Much thanks.

--
Shane Mullen, Staff Engineer
Llewellyn-Howley Incorporated

http://www.lhinc.net

 

[cvg]

i don't know what the "simple" method is

Rusle will calculate the loss from the entire 12.5 acres watershed - not just from the swale - that doesn't sound like what you were asked to do.

measuring the rills would be difficult since you would need to measure on an annual basis for several years and come up with a statistical average rate per year. It would also be highly variable depending upon the amount of rain/runoff and on the amount of vegetation from year to year.

you might want to talk to somebody in your local NRCS office to get information on typical values for your region, as well as advice on a good methodology to use.

http://soils.usda.gov/

 

[murdock]

Thanks for the quick reply.

The The Simple Method to Calculate Urban Stormwater Loads is summarized here.

As for RUSLE, I only used the swale area, not the entire watershed to limit the scope of the calc.

My thought process in measuring the rills and channels was to make an assumption on what the original geometry was, find the volume eroded and divide that total soil loss over the lifetime of the swale (approximately 35 years, according to those with knowledge in the area). I called the DPW to see if there was any maintenance of the swale, and besides some work on culverts and headwalls, there was no maintenance of the ditch since construction. A good theory, but it seemed to me (as to you) that there was too much room for error in this methodology.

I'll check out the NRCS, thanks for the tip.

--
Shane Mullen, Staff Engineer
Llewellyn-Howley Incorporated

http://www.lhinc.net

 

[cvg]

the problem with the simple method as I see it is that it only calculates for one storm, not an annual average. how do you take the results and transform into an annual average?

 

[bltseattle]

My interpretation and past experience is that the Simple Method is used for annual loads.

I do not think it is useful here because it doesn't have an export/loading coefficient for what you are looking at - a roadside swale. You are looking for an erosion rate, which is a level of resolution in the analyses beyond the Simple Method: you would have to figure out if the concentrations used for the Simple Method reflect basins with swales, or not.

I think you are into the sediment transport equations used by geomorphologists. There is a very useful program called Win-XSPro that analyzes channels for sediment erosion/deposition. It has a decent manual, offers several methods, and relies on you to provide flows and soil gradation info. This makes sense, since the erosion rate will depend very much on the soil material (how is underlying soil reflected in the Simple Method? {it's not!}).

Here's the link for XSPro:
[blue]

http://www.stream.fs.fed.us/publications/winxspro.html

[/blue]