All techniques › Water harvesting & retention
Swales, Keyline & Contour Earthworks
Contour earthworks that slow and infiltrate runoff. Far more popular than the evidence supports, permanent once dug, and the only technique here that can fail dangerously.
The short answer
We could not find a single peer-reviewed European farm-scale trial of swales. They are among the most enthusiastically promoted practices in regenerative farming, and the practical safety guidance — slope limits, clay behaviour — exists only in practitioner forums rather than in the literature. That is not a reason to dismiss them. It is a reason to know you would be going beyond the evidence.
If you farm sloping Mediterranean land, restore the terraces you already have instead. Terrace restoration is European and fixes a documented problem: over half of abandoned terraced fields show moderate to severe erosion. It is a smaller permitting question than new works, not an exempt one — check with your water authority before you dig.
This is the one technique here that can hurt someone. Swales wet the ground below them, and saturated clay slides. Do not build above ~15° slopes, on clay, or uphill of buildings or roads without geotechnical advice. And get a permit first — Slovak farmers had retention structures declared illegal and were fined after building them.
Read this before anything else
Two warnings belong at the top of this page rather than buried at the bottom.
Safety
This is the only technique on this site that can injure people or destroy property. A swale deliberately holds water in the ground, which means the ground immediately downslope becomes wetter — and saturated clay is prone to sliding.
Practitioner-reported limits, cited as such because that is the only place they exist:
- Not advised on slopes above about 15°, and especially hazardous in clay.
- Recommended maximum gradients vary wildly between experienced practitioners — some say 25%, some 15%, some 5%. That disagreement is itself a signal that this is not a well-characterised practice.
- Terraces without retaining walls are suitable only at roughly 3:1 (18.4°, ~33%) or less.
- Infiltration uphill of a building or road on a clay slope can increase landslide risk.
Sources: Permaculture Reflections, "When Swales Can Kill"; permies.com earthworks discussions.
If there is anything downslope that matters — a house, a road, a neighbour — get geotechnical advice rather than taking a forum's word for it, or ours.
Permission
Earthworks that intercept water are not purely a private matter. The EU Water Framework Directive governs abstraction, ecological flow standards and river-basin management, and explicitly protects downstream users and groundwater-dependent ecosystems (European Commission). Water you retain is water that does not reach them.
Implementation is national, and we have not been able to pin down farm-earthworks permitting rules for most member states. What we do know is that the risk is real and not hypothetical: in Slovakia, officials ruled small water-retention structures illegal constructions and fined the people who built them.
Check with your national water authority and your river basin management plan before you dig. An unpermitted structure may have to be removed at your own cost. The rules for England, Spain, Poland and Portugal are set out on the permits page: in Spain, anything within 100 m of a watercourse, dry channels included, needs a six-month authorisation; in Poland a non-permanent barrier on your own ditch needs nothing.
What it is
A family of contour-following earthworks that slow, spread and infiltrate runoff instead of letting it leave the field.
Swales are level ditches dug on contour with the spoil mounded downhill, holding runoff until it soaks in. From the permaculture tradition.
Keyline design is P.A. Yeomans' 1950s Australian system: read the landscape's ridges and valleys, then subsoil-rip on a slight off-contour gradient to move water from wet valleys out toward dry ridges. Uses a chisel or Yeomans plough rather than open ditches.
Contour bunds and ridges are low walls or ridges on contour forming micro-catchments.
Terraces are level benches cut into a slope, often with dry-stone retaining walls — Europe's own two-thousand-year-old version of this idea, and by far the best documented.
How it is meant to help
On crusted or sloping Mediterranean soil, a summer downpour largely runs off. These structures are intended to hold it in place long enough to infiltrate, converting surface flow into stored soil moisture and groundwater recharge, and extending the period roots have access to water after rain. They also cut erosion by shortening slope length and slowing flow velocity — which is usually the largest measurable benefit. That premise, scarce rain arriving fast on a slope, is why this page does not list the Atlantic zone: where rainfall is high and the land is mostly flat there is far less runoff to harvest, and the soil-cover techniques address the drought problem more directly.
Keyline specifically aims to redistribute water laterally from valleys toward ridges, and to deepen topsoil by ripping compacted subsoil without inverting it.
How good is the evidence? It varies enormously
| Sub-technique | Evidence | Basis |
|---|---|---|
| Contour bunds, ridges, stone terraces | Good — but non-European sites | Replicated field trials in arid and semi-arid conditions |
| Terraces in Europe | Good — but mostly about failure | Extensive European geomorphology research |
| Keyline design | Weak | One modelling study, one comparative study, 1950s founder reports |
| Swales at European farm scale | Effectively absent | No peer-reviewed European farm trials found |
Water-harvesting structures: the solid part
A replicated study across arid and semi-humid sites found (Environmental Management):
| Structure | Runoff reduction | Sedimentation reduction |
|---|---|---|
| Stone terraces and semi-circular bunds, semi-humid site | 65–85% | 58–69% |
| Stone terraces, arid site | 80% | — |
| Contour ridges, arid site | 73% | — |
Soil moisture increased significantly under every water-harvesting treatment at all three sites. The mechanism works — that much is clear and directly drought-relevant.
The limitation is transferability. These are dryland sites outside Europe. The physics should carry, but do not expect these specific magnitudes on a European farm.
Keyline: promising, not established
Keyline was developed by P.A. Yeomans in 1940s–50s New South Wales, using chisel ploughs to break compaction and increase infiltration.
The famous claim needs flagging. Yeomans reported subsoil converting to topsoil within two to three years, deepening the profile by 10–15 cm. Soil formation at that rate would be remarkable — and the report comes from the inventor of the method, not an independent measurement. It is the single most-quoted keyline figure online and we are not going to repeat it as fact.
Independent work is thin but not absent: a modelling study of keyline effects on erosion control (ResearchGate), and a comparative analysis finding positive effects on both erodibility and runoff infiltration, with control plots losing an average of 75 t/ha more soil than keyline treatments.
Worth separating out: keyline's core physical act — non-inversion subsoil ripping to break compaction — is well-supported agronomy independent of the keyline framework, and connects directly to no-till, where compaction must be fixed before transitioning. The defensible part of keyline may be the ploughing pattern rather than the philosophy, and ripping is reversible and far cheaper than earthworks.
Swales: the gap worth naming
No peer-reviewed European farm-scale trials of swales were found.
Swales are widely promoted in regenerative circles on the strength of physical plausibility and practitioner testimony. That is not nothing — the mechanism is sound and many people report success — but it is not the same as the meta-analyses behind mulching or no-till, and you deserve to know the difference before moving earth permanently.
Terraces in Europe: the cautionary tale, and our best evidence
Europe already did this, at scale, for centuries. The dominant modern research finding is about what happens when maintenance stops.
Dry-stone walls degrade, and their maintenance is complex and expensive — identified as one of the most relevant drivers of agricultural land abandonment across Europe (Land Degradation & Development). Abandonment causes wall collapse and slides in risers, accelerating erosion (Water 10(10):1400). More than half of abandoned fields showed moderate to severe erosion, with terraced land the most damaged (Iberian Range case study). In lowland settings, abandonment leads to landsliding, terrace destruction and increased catchment connectivity — meaning worse downstream flooding than before.
The literature's own framing is the warning: "terraced landscapes: from an old best practice to a potential hazard for soil degradation due to land abandonment" (Springer chapter).
Well-maintained stone terraces retain water, increase vegetation cover and reduce erosion. Unmaintained ones are worse than never having been built. That is the most important sentence on this page.
Where it fits
| Context | Fit | Why |
|---|---|---|
| Sloping Mediterranean land with existing terraces | Restore, don't rebuild | Cheaper, a smaller permitting question than new works, fixes a documented problem |
| Arid and semi-arid rangeland and tree crops | Reasonable | Where the water-harvesting trial evidence applies most directly |
| Gentle slopes under 5–10%, free-draining soil | Lowest-risk case | If you are going to build new, this is where |
| Compacted soils, any zone | Keyline ripping worth considering | Reversible, cheap, and supported independently of keyline theory |
| Clay soils | Avoid or engineer properly | Water sits rather than infiltrating; saturation triggers slides |
| Slopes above 15° | Do not | Practitioner consensus, and the terrace-failure literature supports caution |
| Above buildings, roads or neighbours | Not without engineering advice | You are deliberately wetting the ground above them |
| Flat arable | Not applicable | No runoff to intercept — use the soil-cover techniques instead |
What it costs
This is the most capital-intensive technique on this site, and the only one that is effectively irreversible.
The one European figure we have comes from Slovakia's landscape restoration programme: €5/m³ of retention capacity for countryside projects, against €28/m³ for urban construction, on a 2019 cost basis (EEA Climate-ADAPT). So roughly €5,000 for 1,000 m³ of dispersed retention capacity.
Treat that as order-of-magnitude only. It is a programme-level cost for many small dispersed structures in a relatively low-construction-cost member state — not a farm-pond or terrace quote. We have found no European excavator hire rates or terrace restoration costs, and would rather say so than guess.
What the literature does establish is that maintenance cost, not construction cost, is the binding constraint. Terrace upkeep is expensive enough to be a documented cause of land abandonment across Mediterranean Europe. If you build, you must cost the next thirty years, not one summer.
Erosion and runoff benefits arrive immediately on completion. The obligation is permanent.
Commitment
Permanent — the highest on this site, with the harshest failure mode.
You cannot un-dig a swale. Unlike cover crops or mulching, which you simply stop doing, a swale or terrace is a lasting change to the shape of your land that then carries a lasting maintenance obligation.
Three things make this worse than a long commitment:
- Abandonment is actively harmful. Collapsed terraces cause accelerated erosion, landsliding and increased catchment connectivity. Walking away leaves you worse off than never starting.
- Failure can be dangerous, not merely disappointing.
- The regulatory status can change retroactively, as the Slovak fines show.
Contrast this with agroforestry, which is a longer commitment in years but fails gently — an abandoned plot simply becomes woodland. Length and severity are different things, and this page is severe.
Two reversible alternatives worth exhausting first: keyline-style subsoil ripping, and surface management through mulching and residue retention. Much of the runoff these earthworks are meant to catch is caused by surface crusting that mulch alone can fix.
How to start
- Look for existing terraces or contour features first. Across Mediterranean Europe, restoring an abandoned terrace is cheaper, better evidenced, lower risk and permit-free compared with building something new.
- Fix the cheap things first. Mulching and residue retention deliver infiltration gains for a fraction of the cost and with none of the permanence.
- Survey slope and soil honestly. Above 15°, or on clay, the practitioner consensus is don't. Get geotechnical advice near anything downslope.
- Check the regulatory position with your national water authority and river basin plan.
- Consider keyline subsoil ripping separately from keyline philosophy. Breaking compaction on a slight off-contour gradient is defensible, reversible and far cheaper.
- Start small and watch a full wet season before committing to a whole-farm layout.
- Budget maintenance for decades, not construction for one summer.
Risks and failure modes
Landslide and slope failure — covered at the top, and the reason that section is at the top.
Abandonment makes things worse than never building. Over half of abandoned terraced fields showed moderate to severe erosion.
Irreversibility. A mistake here is permanent and expensive, unlike every other technique on this site.
Waterlogging and crop damage where clay prevents infiltration and water simply sits.
Downstream and legal harm. Water you intercept does not reach neighbours, rivers, or the groundwater-dependent ecosystems the Water Framework Directive protects.
Going beyond the evidence. Recommending swales at European farm scale means acting on plausibility and testimony rather than trials. We think you should know that rather than discover it later.
Founder-claim inflation. The most widely circulated keyline figures come from the method's inventor — the same caution we apply to industry-funded material on no-till.
Claims checked. Two circulating claims about this technique are run down to their sources: keyline turns subsoil into topsoil in two to three years and swales are a proven drought technique.