Where to start with swale digging?

Hi Everyone,

I've read in a few places that when you want to make a series of swales on a hillside, you should start at the top of the hill.  I am wondering why.  I want to start at the bottom of the hill - because it is near the house and it will be the first to start producing (zone 1).  If there is a good reason for starting at the top, I'd love to know before I start digging!

A bit of background - the hillside is about 3 acres with a 7% slope.  It gets about 1,000mm of rain a year.  The downhill side of each swale will be heavily planted.

Thanks,
Richard

I "think" the main reason to recommend starting at the top of the hill is because of the plume of water being created at the top of the property.  So with time, the ground below your top swale becomes more hydrated, and as you add swales below the top one in your system, the ground will already be a little more hydrated.

There are also instances where people have dug their top swale/swales in their systems and then added a lower swale years later, and the lower swale filled with water....because of how hydrated the ground above has become.  I think these are areas with lots of rain.

I don't think there's any specific reason that would hinder you in a large way by putting in your lower swale first...??  But I don't have enough personal experience to really know.  The main things I can think of is to consider if you'll need sediment traps at the top of your system, and the size of your runoff collection area uphill of your property if you have any.  Sediment traps where needed are easy enough to understand.  Calculating your runoff collection area may need some considering, because that is going to determine what size to dig your swales and how long to make your overflows.

If you start at the bottom, I think you are going to greatly increase your chance of a blowout.  All the water from the entire slope will accumulate in that one ditch.  If you start at the top, much less water will be captured in each swale.

Geoff Lawton talks about a drainage pipe he noticed when he first came on the property, and he knew someday there would be a swale there, but he had more important things to do.

If it's just a quick small garden bed then that's what it is. If it's part of a larger designed system then having that design at least on paper will help you fit each element together in the right place.

Keyline points in gullies are often origination points for swales, so locating all those points and their contour lines can help harmonize everything else you do. Often the highest keyline point is the place where  a design might begin to evolve, but even that contour does not preclude additional swales higher up

Also you might look at the lowest point on the highest boundary - this is usually the longest contour line and a frequent starting point for swale design, but even then it may not be the absolute highest swale.

It is true that ground hydration moves down from a swale and can impact swales below it, and it may be advantageous to start at the top, but if you have more pressing needs lower down, then  that would be the choice you make.
The best thing for the sake of maximizing effect with least work, is to know as much as possible what the main frame design will be and then fit the pieces together as you go keeping that final design in mind.

Thanks guys,

I did think about the plume and hopefully we'll end up will swales all the way to the top (maybe the second swale could be at the top).  I've done some calculations to work out the size and number of swales but it's not an exact science!  I read somewhere about the chance of landslides.  Some of the soil has a high clay content.  Maybe we'll have to re-think the idea and get in a digger and do all the swales at once.

I'd add a few thoughts, but start with these two provisos: all my comments below are based on the assumption that you do, in fact, need swales; and, I'm not an expert on design

There are general principles and there are specific plan needs. There is a general principle that says keep water as high on the landscape as possible. There's a host of reasons for this, but among them is that you have the aid of gravity to move stored or accumulated water to where you need or want it. If you have water at the bottom of the slope, you have to have a mechanical process of some form to get it back up slope to where you may need it at some point. This is usually thought of for ponds (dams) but depending on your rainfall and swale size, it will accumulate water that you can use by moving to lower areas where you need it.

Joshua is right about the swale plume affect. The size of the plume is dependent on many factors, but among those are the soil composition below the swale, size of the swale, rainfall amount (event and seasonal), density and type of berm cover, relative humidity, etc. Three acres of draw can accumulate a lot of water in some climates. Todd correctly cautions that a small swale at the bottom of a large draw runs the risk of overtopping or blowing out. Construction must be done really carefully and well in order to withstand a lot of pressure and volume.

Without being able to observe your place first hand, making recommendations is sketchy at best. But, for a 3-acre draw, I would guess you're going to need more than one swale. But, locating them is dependent on what you're overall, integrated plan is. Again, assuming your plan makes swales necessary, how you are planning the systems to work together will help you determine where, and how far apart. This really depends on why your plan is calling for swales. Often swales are seen as a water management tool and a tree-growing system. If your plan is for trees (fruit and nut trees, for example) I wouldn't feel that Zone 1 was necessary because you're not tending orchards on a daily basis.

Any chance you can give us a few more details about your plan and your landscape? Maybe we can all help a bit more with some more info.

Any time I see questions about people starting swales on a slope, I get worried.

My usual suggestion, especially if you have time to let hydrology do its thing, is to lay deadfalls and fallen tree limbs into rows above ground on contour where you want your swales. They will at least act as sediment traps, and will slow the water enough to infiltrate instead of washing right downhill. The sediment will gather behind your contour swale barrier, and you will end up with a terraced effect. At that point, you have the option of enhancing each individual contour line swale.

I am always concerned with digging into a slope located above someone's dwelling. It is so easy, especially with a clay content that will retain more water, to disrupt the slope in such a way that, with the right amount of saturation, could bury you. The above-ground method has the advantage of being less energy and time-intensive, and you can still physically dig the bottom one out. And if your slope is extreme enough to require it, you can use lengths of hardwood branch to peg the wood in place.

I am assuming that you are going to overseed where necessary to ensure you have a root mat to retain the soil in place as you increase the average amount of retained moisture. You could afford to wait and see if your house wasn't immediately downslope, but I would suggest playing it safe.

And yeah, more information about your specific situation would help tailor advice to your scenario.

-CK

Another reason to start at the keyline. The slope is much more stable from there on down.

Chris Kott wrote:Any time I see questions about people starting swales on a slope, I get worried.

Where else would you put a swale :)

Chris Kott wrote:My usual suggestion, especially if you have time to let hydrology do its thing, is to lay deadfalls and fallen tree limbs into rows above ground on contour where you want your swales. They will at least act as sediment traps, and will slow the water enough to infiltrate instead of washing right downhill. The sediment will gather behind your contour swale barrier, and you will end up with a terraced effect. At that point, you have the option of enhancing each individual contour line swale.
-CK

I'm doing more of this on my land on gentle slopes and I really like it.  I can work on it a little at a time and without equipment, and I find it really enjoyable work, like damming the little runoffs along the street and racing "boats" (really just sticks) like we did as kids.

I should, of course, have specified that I get worried when people start swales by digging into a slope. Good catch.

-CK

Thanks everyone for your help and interest,

Here are a few more details plus a site map.

We are in the process of buying a 20 acre property in central France.  This will be our 2nd Permaculture project.  We will be following the design practises described in Dave Jacke's 'Edible Forest Gardens' and also using ideas from http://MakingPermacultureStronger.net.

There is no water on the site.  The land is fairly degraded and has been overgrazed for some years.  We are initially thinking about Landform and Water.  Swales, dams, ponds, rainwater and grey water are all under discussion - hence the question about swales.  The slope behind the house rises to the north west.  The plot is naturally divided in half by the north/south track (driveway).  The section of land on the right of the driveway has a similar slope but to the north east, therefore we have a driveway gulley.  The entire length of the southern part of the land is a small valley extending far outside of the property.  After heavy rains, a small pond forms in the south west corner and drains within a few days.

Let me know if you want any more info.

I built my first swale near the cabin at the bottom of the property. Because it was an easy place to start. Made it easier to observe how water and sediment flows during rainstorms.

Chris Kott wrote:I should, of course, have specified that I get worried when people start swales by digging into a slope. Good catch.

-CK

Now I actually am confused.  I apologize if this is an ignorant question, but aren't swales always made by digging into a slope?  I do as you said earlier and pile things on top of the ground (I've taken to calling them fertility traps), but my understanding is that "proper" swales are always ditches dug into a slope on contour and with the dirt that is removed piled on the downhill side.  Is that not correct?

Todd, you are exactly correct. Swales are trenches containing biomass sponges. On really gentle slopes, I think there are no issues with ground disturbance. But I think that on steeper slopes, the contour sediment traps I described might be a really good preparation for a season or more before the swales are dug.

I would use them to make sure there are good systems of root mats holding the slope together, and with perennials and tree seedlings perhaps start forming the bones of terraces.

Doing it this way serves the short-term purpose of ensuring that the slope doesn't start encroaching on the house because of acute disruption and freak weather events, and also addresses long-term hydrology issues in a way that isn't cost, labour, and time-intensive, so it can be done relatively quickly, allowing hydrology to do the work of land-shaping for you. It may be unnecessary to dig swales at all, considering the local soil-loosening effect that nurse logs have.

What happens if your contour nurse logs, in their decomposition, nurse nitrogen-fixing bacteria host seedlings and a native pioneer succession guild, or even just a green manure guild that drops some deep taproots and various other types of root and encourages macrobiota to thrive in the soil? Wouldn't, for instance, the worms take decomposing biomass from the surface and drag it down into the soil, sort of exactly where you would have wanted to dig your swales?

So I guess I might be describing a type of no-dig, nurse log or hugelkultur-inspired surface system for getting soil life to dig swales for you. Sort of like a system that might be, wait for it, way better than swales.

But to answer the OP, that's where I'd start.

-CK

Joseph Lofthouse wrote:
I built my first swale near the cabin at the bottom of the property. Because it was an easy place to start. Made it easier to observe how water and sediment flows during rainstorms.

Thanks Joseph,

Did you observe anything interesting or have any problems?  Did you decide to dig more swales further up the slope?

Richard Cleaver wrote:Did you observe anything interesting or have any problems?  Did you decide to dig more swales further up the slope?

The most interesting thing I observed, out in the desert, is that runoff events are as much about sediment moving as they are about water moving. So I make a scratch on the surface where I want the swale, and it fills with loose soil during the first runoff event. It's simple to move that soil to the bund of the swale. Hard to dig a swale in bed-rocky ground, but easy to move loose dirt. Even now, I'm amazed at the amount of sediment that moves across the landscape. Problems? Oh yes, mostly regarding blow outs. Sediment swamping. etc... These days I plan for the swales to be inundated and overtopped with every runoff event. So spillways are created over bedrock. That way, the whole hillside doesn't wash away when they fail.

Other things I observed:

Water doesn't move at all like I thought that it did...

Twigs, branches, grasses, rocks laid on contour do a wonderful job of catching vegetation and sediment that would otherwise be washed away.

Packed down roadways shed a lot of water!

I'm working on sites of 20 and 90 acres. By hand. I expect to be building swales for the rest of my life. Every time I visit, I work on swales, check-dams, and contouring with vegetation.

Thanks Joseph - great info.  It looks like I'll be making swales for the rest of my life too.

I think you will make mistakes by starting at the bottom. I sure did. It was the area I had cleared to work, and there were unknown obstructions buried above. I have since cleared above and now the water flows have changed dramatically.

However, and I think this is key, is that a swale that needs modification is better than one that never was started. I have had to modify several water features because I didn't want to wait for four years to get a torrential rain. We still haven't, and we may not for ten years!  

I love Joseph's idea of making small scratches and watching them. I did the same thing on the hill above, in slightly larger scale. I basically had a clay lump that I could move more or less as a unit with the blade or bucket of a (small) tractor. This was my test swale. I left it in place for a year, and then decided to move it, and put in another above it and below it. After 18 months I had a pretty good idea of what the new water flow would be. This is not a static picture, the soil had very low infiltration, and now that it is hydrated, that will increase. The swale plume will change. It may require further modification, but it seems safe at the least.  And I have a new rice paddy as of last week.

I am an enemy of the circles on paper approach to design, unless you are doing a new paper every three months. The idea is that you are creating a more dynamic system, so don't be flustered when you get it! If you can work top down, that does limit the accrued changes, but starting an imperfect plan is better than not starting a really cool plan.  If you have the benefit of a larger property and some heavy equipment, and enough slope to identify keylines, then that seems like a really good jumping off point. I have a master plan in development for my evil long term plan, but more and more I am just working on pieces and modifying. The most important thing in my opinion is to follow a plan that reflects your resources and not try to be Geoff Lawton (unless you are Geoff Lawton). This should be a fun and inspiring adventure that stretches your ability to listen to your land.

I can understand not having access to heavy machinery to do everything at once.  I also have used branches and logs and rocks and anything else on contour to slow runoff, and start nutrient accumulation and water soaking. Spring before last I spent two months trying to dig a sediment pond above my small goldfish pond, then I brought in a backhoe, dug the sediment pond in ten minutes and put in a temporary gully dam above that.

This time last year I revised the temporary dam and where that small temporary pond was is now covered by the footprint of a gully dam 8 feet taller and holding back much more water. I dug another dam higher in the gully, as well as a couple hundred feet of a swale wide enough to drive a tractor through and another contour pond. I have learned lots (that means I made lots of mistakes), but overall I'm way past where I would've been if I had never tried.

I had submitted a mainframe water design as part of my course project, I thought about it for a year or more, and took time before I started the execution of it. I took time to look at the action of the temporary dam and study the water cycles. I never thought I knew everything just because I took a course, but I was never afraid to take the next step and practice what I had learned.

I may never be geoff lawton, but why would I take his course if I didn't intend to follow his footsteps?

Todd Parr wrote:If you start at the bottom, I think you are going to greatly increase your chance of a blowout.  All the water from the entire slope will accumulate in that one ditch.  If you start at the top, much less water will be captured in each swale.

Richard

I felt that this was quite important for considering, and wanted to express my opinion as such.  I didn't put enough time into my first post and only hinted at this, so I wanted to give it another go to ensure that it's been considered.

Looking over your property from the image, you may possibly get away with digging the lower swale first.?  But consider that the professionals with real world experience installing systems around the globe, "Geoff Lawton is the person that I tend to think of",  recommend starting at the top of the hill.

I'll give you a couple of my personal examples with some numbers, maybe it'll help.?

A system I installed above my home site that totals 2.4 acres.  I calculated and sized the swales for a 1.3" event, yet I calculated the overflows considering that the swales were already full and then another 1.3" of rain fell.  So the system will handle a 2.6" event within 60 minutes, and still allow a gentle overflow of roughly 3 cups of water per linear foot per second.  1.2 acres is the steep collection area above the swales, and these areas are, "guesstimating", roughly 85-95% runoff.  If I started at the bottom of that system then I would of needed to make the lowermost swales overflow roughly half the length of the entire swale, just to ensure the swale would stay intact while I began to dig the rest of the swales above.  But instead I only needed to make the lowermost swales overflow roughly 20 feet or so in length, and that's bigger than it needs to be.  This swale system holds roughly 65,000 gallons total, and any overflow is calculated to be gentle and slow, but able to handle to larger events.

Another example...same outcome.  Also calculated as a 1.3" rain event.  This area has 20 acres of catchment above the system with the upper half being 90-95% runoff which runs down to the lower half which is around 50-60% runoff....until the ground is saturated, then the runoff percentage increases.  If you crunch the numbers, it is a considerable amount of water all at once. Depending on the runoff, that can be up to 550,000 gallons in a 1.3" rain event.  If I started at the bottom then the total catchment area would increase by roughly 5 or so acres, which adds another 100,000 gallons of runoff.  I have begun at the top of the hill for this system as well.  The upper swale is around 523' in length with a 3' depth at its deepest point and it's 15' or so wide.  It's shaped like a right triangle sloping to the berm so that the trees planted on the berm benefit even more.  It holds roughly 65,000 gallons, and the overflow for this uppermost swale needs to be a minimum of 200' in length to satisfy me.  There will be five swales in this area, all of them roughly the same size.  With this amount of catchment and the percentage of runoff, it makes the most sense to start at the top of the system.  By the time all of that water runs through this system it will be adequately pacified and the water can slowly work it's way down through the rest of the property.  If I started the bottom swale of this system first then the overflow would increase in size...and it needs to be a small overflow area moving gently into the next set of swales that wraps around to a different side of the property.  So again for this system it makes the most sense to start at the top of the hill, and pacify the water as much as possible.

In my instance, for both of these examples starting at the bottom of the hill would result in too many failures.

I'm unsure if that will be of any help, but hopefully it makes sense.  The common theme for water is....slow, spread, sink.  Sometimes you can start at the bottom of a system and be okay, I've had a few acres in the past with no catchment area and a very low runoff percentage, that it would of been possible to start at the bottom of the slope.  So you may be able to get away with starting the lowermost swale first...but you'll need to do some number crunching if you want to verify that you experience success.  findlotsize.com is a great resource for calculating acreage using satellite images.

Here's a simple way I use to figure the size of the swale I'm digging or planning.  Swales will typically be shaped like a shallow "U" or like a right triangle.  For "U" shaped swales, find the area of the swale by calculating the cross sectional area as an ellipse, then split it in half.  For right triangle swales, find the area of a right triangle.  If you google these it makes it quite simple without needing to know formulas.  Then when you have the cross sectional area of your swale, you can find the cubic area of the swale in total and convert it to water capacity.

Hi Joshua,

I really appreciate you taking the time to answer my question so thoughtfully and with so much detail.  I now understand why I should start at the top!  It turns out that the region where our site is, has a reputation for large rain events.  I hadn't really given this much thought but thanks to you I'll be re-doing my calculations.

Just as a kind of side note. People often ask the "spacing" question when discussing swales. I think there are a lot of factors that determine how far apart to space your swales. Certainly water volume calculations are critical to this determination. Another key factor is the mature height of trees associated with the swales and sun angle. After that, on larger properties, there are likely other factors. For me, once I understood the minimum number of swales and their volume potential (I was planting semi-dwarf trees, so solar arc was not an issue), the distance between swales was based on the size of the machinery I would run between them (sickle mower). I wanted to only make a defined number of passes with my sickle mower with full swaths. Running a tractor and sickle mower with only a half swath is not efficient.

What a plan calls for between swales (silvopasture, cash crops, etc.) can have an impact on the distance between swales. Just another thing to think about.

Dan Grubbs wrote:Just as a kind of side note. People often ask the "spacing" question when discussing swales. I think there are a lot of factors that determine how far apart to space your swales. Certainly water volume calculations are critical to this determination. Another key factor is the mature height of trees associated with the swales and sun angle. After that, on larger properties, there are likely other factors. For me, once I understood the minimum number of swales and their volume potential (I was planting semi-dwarf trees, so solar arc was not an issue), the distance between swales was based on the size of the machinery I would run between them (sickle mower). I wanted to only make a defined number of passes with my sickle mower with full swaths. Running a tractor and sickle mower with only a half swath is not efficient.

What a plan calls for between swales (silvopasture, cash crops, etc.) can have an impact on the distance between swales. Just another thing to think about.

Excellent point and one I hadn't considered when planting trees.

I have a much shorter explanation, and it comes from experience. Just put the swale(s) slightly off contour. And just DO IT. No matter what your plans include, nature will reshape, and let you know what it wants to do. Dig it, let nature take its course, and work from there.

I may never be geoff lawton, but why would I take his course if I didn't intend to follow his footsteps?

Bob, I think my response may have had the wrong tone. I have nothing against taking courses. I have gone through the entire PDC/ATC video series and learned quite a bit. There are lots of good things to learn in courses. But  (and this is not a knock on courses) I have learned way more by watching projects as they progress and modifying them, and asking people smarter than me what they see in the situation. I spend at least twice as much time observing systems than I do reading about them, easily.

I just wanted to affirm to people that making "mistakes" and changes is necessary. The krameterhof was not built all at once or top down, this should be an iterative process. There is a military axiom from some famous general that "no plan survives first contact." This is true more often than not. The keen observer is not born, but developed over time, and factually that does not happen in a course. It happens when you stand out in the rain and watch infiltration and water movement and try to figure out why this fig tree is doing well and this other one sucks. It will happen again and again in a good system, and people should be encouraged to learn to observe over copying and templates. I think that is following in the footsteps of the great artisans. I think Geoff would be ecstatic at your observations, testing and eventual completion.

Planning certainly has a place, don't get me wrong. Big integrated earthworks would be a prime example, as a washout can be devastating to try to fix. Keyline planning lends itself to good outcomes, since it is somewhat predictable in the right situation. A house or other structure needs to have a solid plan. The big three- roads, structures and major earthworks seem to me to be about the only things that won't guarantee to require frequent changes if done right. That being said, moving any of them can be done and sometimes must be. Travis has been moving barns around! The first step is the hardest.

chad Christopher- I have a much shorter explanation, and it comes from experience. Just put the swale(s) slightly off contour. And just DO IT. No matter what your plans include, nature will reshape, and let you know what it wants to do. Dig it, let nature take its course, and work from there.

I think unless your grade is pretty nasty or your rainfall is torrential, this is sage.

Thanks tj, being said..that if the grade is too steep, or you  have THAT much rain...then swales are not the  proper  approach  anyways. I would say that the only  thing that needs a little  engineering, is if you  have  intersecting  swales. If these same swales are rentation basins as well. Things get a little tricky when contours collide. It would  be  rare that this would be required , as creating a  spillway into the  lower basin would  be  a  much safer and  easier idea. And always remember ,it's  SLOW spread, and sink. Not STOP spread and sink.

Observation is always the primary starting point, and overall plans/patterns are really the essence of Permaculture Design. That is not to say that every plan is fully developed before it is implemented, or may not need some reworking, especially as Nature starts to add her suggestions.. Start simple and then let nature complicate things. Main frame water design though is not something that is all that complicated. We are stopping, spreading and soaking water. The principles of locating dams and building swales are well documented even if I can't observe them on the ground close by.


Like I said, I did lots of handwork and still do developing features, but more and more as I make a determination of where i want ponds and swales, I stop trying to implement them with personal labor and use my personal labor for things the machines can't do well. It takes seven years to hydrate a property fully, After a few plantings of misc trees and bushes I realized I was getting nowhere fast. Water makes everything work better.  A light switch went off and I knew this was a place in time that justified going into debt for machine rental. So while I don't have a fully functioning swale system, I do have some dams that are trapping water and soaking it in, instead of rivers running down the side of my driveway for a month or two and drought the rest of the year.

Geoff had never seen a swale when he dug his first one. Some of us will have to study and study and relate the learning to what we can observe, and then make occasional leaps of faith based on as much information as we have. I didn't know when I built these dams whether they might ever hold water for long, but I did know they would stop runoff and put water in the soil, and likely eventually hold some water.  

I didn't understand just how much they might dry out systems farther down temporarily, but I did know they might do that. With an ongoing drought this year this has become a bit problematic for an established pond, but i still wouldn't change a thing.  I'm not even sure it was a mistake, more like unavoidable collateral damage in a successful campaign (to pursue the military symbolism a bit further).

In a landscape with no available water, but heavy singular events, that just screams to me to implement water storage as high as possible as soon as possible. Of course if the only available possibility is to dig a single swale or small pond by hand lower down, then that is where to start. We have to work with what we have where we are.

I don't think we really have any differences as far as our understandings of how these things work Tj, I just wanted to make sure people are empowered to make those next steps when they feel ready. I think Both Bill and Geoff have expressed the idea that designers that follow them will be better teachers and better designers over time. (it likely won't be me). But I do see where some of the young people are being weaned on this stuff, and they will start to have basic instincts that far surpass us.

A huge thanks to everyone for your advice and comments.  Plenty of food for thought.

I’ve just finished reading 'The Permaculture Earthworks Handbook' by Douglas Barnes, which I highly recommend.  For anyone who is interested, here are his four reasons for starting earthworks as high as possible in the landscape:

1.  The potential for groundwater recharge is maximised.  Water will have a longer, slower path through the landscape.  This can create greater flows in downhill wells, springs, and streams.  Slopes contain more surface area and more soil volume than flatter areas (when measured as a two-dimensional area).

2.  There is greater water absorption higher in the slopes because the soil is often made of larger particles.  The smaller clay and silt particles tend to get washed downhill.

3.  The possibility of erosion is reduced as water is slowed early in the landscape.

4.  The scale of the earthworks is much smaller at the top because there is less run-off than at lower points.

Bob, I totally agree. I have a feeling that different people have different tolerances for the "leap".  

My issue has been that I am by nature a planner/tinkerer. I force myself to be a bull in a china shop. I can modify a plan better than I can plan it, because I have spent so many days planning things and had something unforseen just trash the plan, which demoralizes me. So I have had to get pretty uncomfortable and just start doing something, modify, observe and sometimes reset. An example is that I order plants way ahead of time, then I have all these plants coming and have to get busy to prepare the areas. It is a way of forcing my hand, to get off the preparation treadmill.

 A light switch went off and I knew this was a place in time that justified going into debt for machine rental.

Yeah, this is not me! Not yet. Now what I should do is put a deposit on a machine rental, and then I would have to commit to a plan!

I guess this is part of the observation cycle, knowing how you work, not just what work might be done.