Don’t Try Building Hugel Swales – This is a Very and I Mean Very Bad Idea

Let’s start out with a basic understanding of what full scale Hugelkultur and full scale swale based systems are. Then we can jump into what they are not and why combining them can be a very bad idea.

First, as most permies know, a swale is a ditch on contour. They are also tree growing systems, and are not designed to grow annual gardens. They do a great many things well. Swales can fill ponds, prevent erosion, create pattern framework and the big one everyone focus on, they infiltrate water. In the end though, swales are simply non-compacted mounds on the down grade side of a ditch on contour for the purpose of establishing tree based systems.

Don’t get me wrong you can get a lot out of small scale swales and swale like paths for gardening, etc. But such is not what the term “hugel swale” means in the countless minds of those who keep recommending this practice. So far we have been lucky and most of the people who want to do this seem to have not acquired the resources to do it on large scale.

Second what actually is a Hugelkultur? This is mainly an Austrian/German technique made famous, but certainly not invented by Sepp Holzer. Holzer is considered the master of Hugelkultur with good reason by the way. He says in his book, “Sepp Holzer’s Permaculture”, “do not put hugel beds on contour”. There is a very good reason for this message. Understand for a moment, that a man who has built more hugels than likely anyone else on the planet says not to build them on contour, and that swales must be on contour by their very definition.

Now let’s discuss what Hugelkultur is. Loosely translated it means, “hill culture”. In other words, it is not burying wood deeply below grade. The proper execution of a hugelkultur would be piling it up at or perhaps just slightly below grade, then covering it with soil. A true hugel is approximately 30% wood core and 70% soil. It is 1.5 to 2 meters tall; the angles of the sides are at 70 degrees.

Image from
Image from

The purpose of this mound is twofold.

1. Break down organic matter and build soil

2. Grow annual production while number one happens and/or growing short term perennials and nurse trees for later planting in other locations.

How do I know this? When I met Sepp in Montana in 2012 and watched him build about 4 linear kilometers of hugels, he told me so and I believed him.

I hear cries of heresy and blasphemy, but I am just telling you the way this system is actually used successfully. So strap in if you are upset now, indeed it gets worse from here. The big shocker is what happens next. One of them you may have a hard time believing…

1. The mounds over time sort of flatten and are left. At this point, the succession proceeds into long term perennial production. The key though is that a few seasons of annual cropping and short term perennials are used first. Generally in this case they remain bush, shrub, small tree, herbaceous and annual crop producers.

2. More often than above, gasp! The mound is at some point spread out and full on perennial systems are established or even grazing systems are boosted. This can produce astounding amounts of soil, the value of which if trucked in would be measured in 10’s of thousands of dollars per acre.

The primary purpose of hugels is building soil, production is of secondary concern. Getting production out of hugels makes the method practical but none the less, still secondary to the original intent. Very few edicts to the concept are even aware of why Sepp Holzer did hugels in the first place. Quite simply it was done because he had a ton of low value trees around and removing them was more costly than their value.

Something had to be done, so Sepp buried them. That is it really. I have seen Holzer build berms with no wood in them for many other reasons; some of these go straight to full tree production because they will not slump in time like a hugel will.

So why not combine the two?

First look at the goals,

• A hugel is about annual and short term perennial production in a decaying mound designed to collapse on itself over time. A mound that may in fact be spread out in the future after having built amazing soil for almost free.

Image from Jacks' Flickr, reproduced with permission
Image from Jacks’ Flickr, reproduced with permission

• A swale is designed to be a non-compacted mound immediately planted to long term perennial tree based systems.

As permaculturists we are trained to combine systems, to function stack and at times we take this way too far. When you force two very different techniques into one, the intended benefit is lost. These two methods are just not compatible as a single technique. They can work beautifully together, but not as a single technique. I will conclude with how to combine them so please read on and hold judgement until then.

What I want to cover next is the bigger reason swales and hugels do not go together. As permaculture practitioners and designers we should not be creating potential natural disasters in our work. In order to convey my point, lets do a bit of math.

Say we build a full on swale 9 feet wide, average depth 10 inches, length say 350 feet long. Once enough rain infiltrates or fills ponds and this swale begins to hold water, what is its holding capacity? It scares me that people are putting in these structures without doing that math.

The answer is a bit over 19,000 gallons.

And remember, about 19,000 gallons were already infiltrated into the soil below grade and quite a bit wicked up into the mound. Oh that’s right, the un-compacted mound allows for wicking, an added reason to why we don’t compact it. So the swale will wick up water, and it is very good at doing that.

Now most of you are thinking but we don’t build a 2 meter high hugel when we do a swale. I agree, but if we create the thing we call a hugel swale we put quite a bit of wood in it. Say we stack a few layers of woody material along the contour, enough to be say 2 feet high and a few feet wide, then burry it with the dirt from the swale. We now have loosely compacted soil, holding down a massive buoyant core of organic matter that is in no way locked into the soil. Just in front (downhill) of a system with 19,000 gallons of water in it when full. Can you see what comes next?

Now great joy comes and our first major rain event occurs. Thousands upon thousands of gallons of water go into the swale, much into the subsoil; we super hydrate the land and the mound too. Now the excess water sits above ground and applies pressure on the uphill side of our hugel swale. The wood core begins to float as large amounts of water seep through our un-compacted dirt and next thing you know, there she blows! You have just sent a massive pile weighing many tons flowing down the landscape. Think this can’t happen? Well I had a conversation with Mark Shepard and he told me of an instance where it did happen.

The basics were, some permies got permission to build a permaculture project at a church. They got excited about the hugel swale concept. They brought in truckloads of woody material and piled it all up on contour. They hand dug a great swale, nice sill for overflow and all included. The rain came, everyone was excited.

Image from
Image from

Next thing you know, multiple metric tons go floating down at a relatively slow speed to another building. Slow speed of movement is deceptive though as mass time acceleration equals force.

Even slow velocity makes a lot of force with a lot of mass and that was a problem here. The wall was breached, mud and water poured in. Building owner is mad, church is mad, permaculture looks bad. Not a great story huh?

Guys and gals. For all the wonderful things we can do with little risk in permaculture when you step up to earth works you have to do some math and get it checked by engineers when it is beyond your current level of experience. Let me bring some of it home to you. Say you have a catchment area of 10,000 effective square feet, some is hard, some is soft but when you do all the math the effective catchment for 100% run off is just a measly 10,000 square feet. With that you collect about 6,200 gallons of water per inch of rainfall.

Many times we are working with dozens of acres of catchment. A single acre is 43,560 square feet. When you get one inch of rain on an acre you have 24,000 gallons of water and some change. If you get a big rain event, say 6 inches, that is 144,000 gallons of water moving through your system.

Image from Jacks' Flickr, reproduced with permission
Image from Jacks’ Flickr, reproduced with permission

I am not saying you can’t bury any wood in a swale mound. If you put down some scrub or say a single row or two of logs, the end system will be quite stable. But that example isn’t a hugel swale, it is a swale mound with bit of stray material buried in it and that material will do little or nothing to support the system beyond may be kick starting some fungal activity and sequestering a bit of carbon. It is simply a convenient way of getting rid of some material you have no other use for.

Can you make small hugel like beds on contour safely? Yep I have done so on my own property, the majority of the wood is below grade, the mounds on contour only account for about 2,000 square feet and the hills are only 36 inches wide and were built to about 24 inches high. They were annually cropped for a season then successed to become a small orchard. The total catchment they take water from is only about 5,000 square feet and there are no “ditches” just contour paths between their four rows.

Can you make a garden with a small hand dug swale like path and use wood cores? Yep, works fine. You can do a lot with wood cores but what you can’t do is put them into contour based systems on large scale without creating potential disasters.

So how might we effectively combine them? Well the way Sepp does on his farm, he installs terraces, on the terraces he has hugels that run vertically across the short dimension of the terrace. Well a terrace is basically a big flat swale like feature. So if we are not in the Austrian Alps and don’t have such large elevations to deal with. We can use actual swales instead of terraces.

To use this pattern you would develop your swales as per normal. Use them to establish ponds, infiltrate water, establish tree systems, etc. Then in the inner swale you build hugels and use them for whatever purpose you see fit.

Now this is proper function and technique stacking. Hugels wick up moisture a lot like swale berms. By infiltrating water with the swales you begin to raise the water table in the inner swale, this gives your wood core more to wick up. Because your swales take the momentum out of the surface water the hugels are less susceptible to erosion.

In time depending on the goals and scale of the system you either use the hugels to make an amazing shrub and herbaceous layer in your system or bring in equipment at some point and spread the thousands of cubic yards of soil worth many thousands of dollars on your grazing space for almost no money.

There are indeed countless ways these two design elements can work together. However, they are and should be treated as separate techniques for very good reasons.

I will conclude with this, the master of the swale in my opinion is Geoff Lawton. I have seen him do some amazing things with swales and I have even seen him make “organic matter swale berms” but these were above grade and let water move though them. They were deposition traps, designed to break down. Never have I witnessed Geoff Lawton make a hugel swale.

The master of Hugelkultur is Sepp Holzer, have you ever seen Holzer put a hugel into a swale mound? No and you won’t either, the reason is, Sepp can do math quite well and understands the forces he is playing around with.

I know we all want to be clever and come up with “original ideas” but trust me when I tell you this is one to avoid. There will likely be a lot of objections to what I have presented here, but none will come from engineers with hydrological and civil engineering backgrounds. Please consider that before you ignore this advice.

I wrote this article because I see the concept of hugel swales tossed around all the time. And I fear in time someone is going to get their hands on a 16 ton excavator and really try to prove it. Well if it was a great idea, someone with the resources would have done it by now. Such a project will eventually harm the earth, likely harm people and it will let a lot of surplus be lost.

Folks, that ain’t permaculture.


Jack Spirko clarifies his article, courtesy of his Youtube Channel

Jack Spirko also hosts The Survival Podcast and his own Youtube Channel which has been showing season two of his duck chronicles.

Jack on Facebook

You can find other posts by Jack Spirko here.

Feature Image by



7 thoughts on “Don’t Try Building Hugel Swales – This is a Very and I Mean Very Bad Idea

  1. Our cityscape is relatively flat. I’m working in areas with just a 7′ rise above the effective water table, plus aging drainage gullies bringing the total to around 22′. City storm drains feed a network of COE canals, engineered to recharge draw-down in the period following the Great Depression. Theoretically, (without expansive permaculture analysis,) my proposed “free local food forest” hugel-mounds will be placed in proximity to, but not within 30 t. or so of the likely flood-plane near the fragile, sandy-loamy edges of this network of urban canals. Increasing the fertility and depth of the soil profile throughout this town is critical because chemical contamination is evidenced by taste throughout the broader lowland regions.

    I am seeking suggestions and advice. The basis on my pitch weighs upon several goals. Downslope, the regenerative organic system-filter should help the environment greatly over time. Socially, it enables every grade of students and conscientious member of this destitute community to participate in a very low maintenance/no till, chop-n-drop community enriching process. This matches the city’s formally stated objectives and primary fiscal needs. Plus, almost all of the materials are already free for the asking, on site.

  2. If your beds are above the level sill spillways line and enough water can flow through the spillways I don’t see how this could be a problem.

    1. Because hydrology is the short answer.

      What does the mound do during rain events?

      In the end this whole thing comes down to people wanting to be clever and do something like take two elements that don’t belong together and mash them together.

      To put organic matter into soil in a swale system we grow trees they do the work with roots which make the system more stable, cutting up a bunch of slash and burying it in a swale mound is a bad idea. You want to do it, go ahead.

  3. Hello!

    The article was very interesting and well written. Your follow up video was satisfying as well.

    Soon im finishing my B.Sc in “sience of soil and water” in the faculty of agriculture – Hebrew university – Israel.

    In this current time I would like to do my M.S. on hugulkultur.
    There is one thing you said that cached my eye: “More often than above, gasp! The mound is at some point spread out and full on perennial systems are established or even grazing systems are boosted. This can produce astounding amounts of soil, the value of which if trucked in would be measured in 10’s of thousands of dollars per acre. ”
    I am trying to find more information about this use of hugelkultur: making high quality soil. Can you help me find some information on this subject?

    Thank you very much,

  4. Thanks for the article. I learned the term “swale” from LID (Low Impact Development) training when I worked as an Environmental Scientist. We used the term swale to describe ephemeral drainages that did not have scour. These “swales”, installed for treatment of stormwater, often crossed contours at about by 90 degrees and typically had a slope of about 2 degrees.

    The Mojave Desert, where I live and worked often has natural “swales” that drain sheet flow between hills or mounds. When flows concentrate in response to development, the “swales” often become washes or arroyos which have scoured beds.

    Following the definition used by my previous employers, I have a roadside swale. Given the permaculture definition, I do not have a swale. I guess this is an example of what happens when different groups assign different definitions to the same words – miscommunication.

    I was planning on writing a blog about an infiltration trench that I built a year ago and backfilled with logs, compost and mulch similar to hugelkulture piles . So what do I call it?

Leave a Reply

Your email address will not be published. Required fields are marked *