(Perma)Culture and Sanity
A 'swale' is simply a long, shallow depression in the ground, designed to collect or redirect water. In general, permaculture swales are used to mimic the water-collecting and -holding abilities of a thick forest mulch.
Swales are most useful in reforestation of degraded, mostly-bare, arid or semi-arid hillsides, to direct water to trees (this water would otherwise run off the bare soil and be lost to the local landscape). In a healthy forest with a thick mulch of leaves or needles covering the ground, very little runoff occurs and swales would usually be unnecessary. A healthy forest is very good at managing its own water resources—if it ain't broke, spend your time fixing something that is.
Each of these types of swale behaves differently and is useful in different situations.
Water-Harvesting Swales are laid out and dug along perfectly level lines ('contours'). This creates long, skinny, cross-slope water reservoirs that interrupt and capture soil-eroding rainwater runoff as it flows downhill. After the swales stop the runoff water, they hold it in place and allow it to soak into the ground to form an underground reservoir of water. Trees planted along water-harvesting swales benefit from a much higher effective rainfall than surrounding, untreated areas (they also benefit from the collected eroded soil).
Properly-sized swales catch all rainwater runoff from any but the strongest rainfalls. Properly designed swales allow extra water from extreme rainfall episodes to exit the swales peacefully, without causing gully formation. Ideally, the overflow would then be caught by larger swales—or even ponds—further downslope.
On-Contour Water-Harvesting Permaculture Swales have numerous features useful to the permaculturist:
Water-harvesting swales must be planted with trees or they will cause soil waterlogging by infiltrating larger amounts of water than grasses and herbaceous plants can remove. Modern-style thinking (wherein everything is treated first-and-foremost as an engineering problem) caused the creation of Soil Conservation Swales, which do not hold water (see below). Soil Conservation Swales are designed to slow water movement—and thereby erosion—without allowing the water to collect to the point of causing waterlogging in tree-less pastures.
However, this type of thinking constitutes a Type 1 Error—in this case, by treating an accumulation of water as a problem instead of as an unused resource. Knowing that trees absorb enough water to prevent waterlogging, instead of refusing the water we can plant water-harvesting swales to productive trees with higher water needs than local averages provide. We gain both the water and the yields from the trees by using win-win thinking, and have effectively widened the variety of trees we are able to grow without excessive care.
Planting trees in pastures is almost always resisted by ranchers, who believe—incorrectly—that the trees will lower grass production. Ironically, in arid and semi-arid lands widely-spaced trees (at up to around 30% coverage) actually increase grass production by enriching soil and offering partial shade to otherwise over-baked, light-saturated grasses. This effect can be readily seen by the taller, deeper-green grass that grows under dryland trees that are allowed to remain in place in pastures.
Diversion ditches are designed to move water safely (i.e., without causing erosion) as opposed to infiltrating it. For this reason, diversion ditches—rather than staying precisely on-contour—drop in one direction at the rate of about 1 inch of drop per 8 feet of run (or about one foot per hundred feet, or 1%). A 1% drop is sufficient to allow water to flow efficiently toward a storage point or drainage—at the same time it is slow enough to allow sediment carried in the runoff to drop out of suspension and collect in the ditch bottom (in heavy rains a great deal of soil can be deposited).
Since water movement and not infiltration is the goal, soil conservation swales do not tend to support high-moisture-need plants effectively. Even in arid areas, plant communities in soil conservation swale bottoms tend to closely resemble the plant communities growing in open areas where water is not restrained from moving downhill (unlike permaculture swales, which quickly establish richer, more diverse plant communities in their immediate vicinities).
Diversion Ditches:
Again, rejecting a resource instead of constructively managing it constitues a Type 1 Error, which causes later remedial action to be required—in this case, watering trees with pumped water instead of with the water that falls on the trees naturally. Permacultural thinking would suggest keeping the water, and preventing waterlogging by planting trees rather than by rejecting the water itself as a way to prevent waterlogging;
The fact that 1%-slope diversion ditches allow eroded soil to fall out of rainwater runoff is both an advantage and a weakness—especially on steeper slopes. This is because runoff coming down a hill can be carrying quite a lot of silt—up to coarse sand—all of which will be deposited in any diversion ditch which cuts across the slope. In hard rains and erodable soils, small ditches can fill in very quickly—in one rainfall episode in extreme cases. Even where diversion ditches cut across gentle slopes where erosion appears to be minimal, it's best to make the ditches larger than you think would be needed.
The most commonly-seen swales in this country (at least in our arid American Southwest) are called soil conservation swales. Most were installed by workers in the Civilian Conservation Corp during the 1930's, and are actually a form of diversion ditch designed to stop soil erosion rather than simply move water from A to B. Open on one end and sloped the standard 1% for a diversion ditch, soil conservation swales:
Soil conservation swales slow water runoff enough to allow soil to fall out of suspension while releasing the water from one end of the swale to prevent soil waterlogging. A better solution than open ends to prevent waterlogging, however, would be to shallowly-dam the ditches periodically to hold some water (or use a level water-harvesting swale), and then plant trees along the swale to prevent waterlogging.
Elizabeth Barrette June 11th, 2010; 10:28:24 AM ysabetwordsmith.livejournal.com
My only quibble is the assumption that water not retained in swales is wasted. Suppose you wanted a wetland at the bottom of your slope: it would make more sense to create water-transporting swales than water-harvesting swales, channeling the water gently downhill into the wetland. Then you have a self-maintaining water feature for plants and wildlife.
Jack Rowe June 6th, 2010; 09:30:56 AM
Thanks George,
Driveways are a great place to collect extra water, which often otherwise runs along the driveway causing erosion.
Don't forget to plant trees to avoid waterlogging, especially in rainy areas like Northland. Not only is waterlogging bad for soils, but in steep areas it can loosen soils over bedrock and cause mudslides or waterlogging in flat areas below.
Jack Rowe
New Mexico, Southwest USA
george June 6th, 2010; 12:32:45 AM geojo53 at yahoo dot com dot nz
Interesting site,
I am presently constructing swales into a bank along my driveway in terraces to reduce slope drainage etc.
Northland N.Z
Reprinted from (Perma) Culture and Sanity Website
http://tinyorb.net/Jack/
Jack Rowe June 11th, 2010; 12:00:33 PM
Good point, Elizabeth,
The idea of 'wasting' water is purely local, i.e., the water is always somewhere, being used by something or somebody, so it's hard to actually 'waste' in real terms (though it may temporarily divert into destructive uses, such as streambed erosion)... and certainly if you want a particular water feature you'd take the water to where you want it to collect. You could even keep some of the water in swales to begin reforesting the slopes and lower or halt soil erosion, and take the rest to the wetland at the bottom, having it both ways. Furthermore wetlands are some of the very richest ecosystems, with enormous production potential.
Mollison suggests that, by capturing water on slopes, lower areas are automatically rehydrated, and springs and wetlands reappear—but this would be a delayed effect. In unmolested Nature, slopes with 12-15 inches or more of annual precipitation would be forested and well mulched, rain would be infiltrated on slopes to nourish the forest, and excess water would make its way below ground down to the valleys to establish natural wetlands. Wetlands do naturally fall at the bottom of slopes, so it would be as self-maintaining as it gets, as you say.
One danger with taking water directly to the bottom of slopes in arid/semi-arid lands—if trees are not present on slopes and near the bottom—is that the water table at the bottom of the slope can rise to near the surface, setting up constant evaporation of groundwater that in arid areas tends to lead quickly to harmful concentrations of salts in topsoil and surface water. This is a common sight in the US Southwest.
How we go about re-establishing 'natural' conditions at this point in the degradation of our landscapes will remain a judgment call, but to paraphrase Mollison again, you can hardly do worse than we've done in the past and improvement is likely with any well-intentioned inputs. Thanks for your input!