How the two differ in kind
Both carry surface water from where it falls to where it is allowed to go, and both are open-channel or part-full pipe problems governed by Manning's relationship rather than by pressure. The difference that drives everything else is ROUGHNESS. A smooth plastic or concrete pipe offers very little resistance; grass offers a great deal. For the same flow and gradient, the swale needs a substantially larger cross-section — several times the area, not a few per cent.
That is the cost, and the slowness it buys is the product. Water moving slowly through vegetation drops its sediment, and sediment is what carries most of the pollutant load off a road or a yard. Water standing briefly on a permeable base infiltrates instead of leaving. A pipe does neither: whatever enters it arrives at the outfall, at the same rate and with the same contents, a few minutes later.
The two are also constrained from opposite directions by the same variable. A pipe has a MINIMUM gradient, because below a self-cleansing velocity solids settle out and it silts up — and, less well known, a maximum, because too steep leaves the liquid outrunning the solids. A swale has a MAXIMUM velocity, because above it the flow scours the grass off and the channel starts eroding. The same fall can be too flat for the pipe and too steep for the swale, which is why a site with a strong gradient tends toward pipes and a flat one toward swales with check dams.
The factors that actually differ
Show
| Grassed swale | Piped storm drain | |
|---|---|---|
| Cross-section for the same flow | Several times larger, because vegetation's resistance is an order above a pipe's — and it changes through the season as the grass grows. | Small. Roughness is low, constant, and known. |
| What it does besides convey | Settles sediment, infiltrates where the base allows, and reduces the peak by simply being slow. Three functions from one feature. | Conveys. Whatever enters arrives, at the same rate and with the same load. |
| Land take | A strip — bed width plus side slopes flat enough to mow and to be safe to walk on, conventionally one in three or gentler. | A trench during construction and nothing afterwards. The surface above it is usable. |
| Governing velocity | A ceiling. Too fast and the channel scours; check dams are the standard way to hold the gradient down on a sloping site. | A floor. Too slow and it silts; the minimum falls quoted as one in forty or one in eighty come from that, not from capacity. |
| Capacity through the year | Varies. Long grass is rougher and slower than mown, so the same channel carries different flows in July and February — which is why the design roughness is stated with a mowing regime attached. | Constant, until it silts. |
| How it fails | Compaction and scalping. Vehicles crossing it, or mowing to bare soil, destroy the infiltration and the vegetation that made it work. Sediment build-up at the inlet is the other one. | Blockage, at a point, invisibly, until it surcharges. Roots, silt, fat, and whatever was dropped down a gully. |
| Maintenance | Landscape maintenance — mowing at the right height, arisings removed, and periodic sediment removal at the inlet. Cheap, frequent, and skipped often enough to be the usual reason one fails. | Jetting and camera survey. Infrequent, specialist, and only triggered once something has gone wrong. |
| Where guidance puts it | High. Modern surface water hierarchies prefer infiltration and surface conveyance over discharge to a sewer, so a swale is often the option a drainage strategy has to justify NOT using. | Last. A piped discharge to a sewer is the fallback where the hierarchy above it has been shown to be impracticable. |
Which one, and when
Choose grassed swale when…
- There is width available along the route — a verge, a boundary strip, a margin that is not doing anything else.
- The runoff carries sediment or road pollutants and something has to treat it before discharge.
- The ground will infiltrate, and reducing the volume leaving the site is worth something in the drainage strategy.
- The gradient is gentle, so the flow stays below the velocity that would erode it.
Choose piped storm drain when…
- There is no land: the route crosses a car park, a yard in use, or a building line.
- The gradient is steep enough that an open channel would scour without a great many check dams.
- The flow has to cross beneath something — an access, a building, a road — where an open channel is not an option.
- The flow is large enough that the swale required would be a watercourse rather than a landscape feature.
Now run your own numbers
This page holds no prices on purpose — a national average is wrong for almost every real project. Quantify both options with your dimensions and your local quotes.
Frequently asked questions
- How much bigger does the swale actually have to be?
- Enough that it is a land-take decision rather than a detailing one. Manning's relationship puts roughness in the denominator, and typical values for mown grass are roughly an order of magnitude above those for a smooth pipe — so at the same gradient and the same flow the swale needs a cross-section several times larger, and more than that if the grass is long. The calculator here returns the capacity for a stated section, roughness and gradient, which is the right way round: choose the width you can actually have, then see what it carries, rather than deriving a width and discovering the boundary is in the way. And design it at the roughness it will have in summer, not the one it has the week it is built.
- Why is my swale permanently wet?
- It may be doing exactly what it was designed to do. A DRY swale is designed to drain down between storms and be usable grass in between; a WET swale is designed to hold water permanently, because a permanent pool treats better. Which one you have was decided by the ground beneath it and by the designer's intent, and a wet swale on a site that was sold a dry one usually means the underlying soil does not infiltrate the way the design assumed. The other possibility is that it is a dry swale that has stopped draining — compacted by vehicles or blinded by sediment at the inlet — and the tell is that it used to dry out and no longer does.
- Can vehicles drive across a swale?
- At designed crossing points, yes; anywhere else, it destroys it. A swale's infiltration depends on the soil beneath staying uncompacted, and a vehicle compacts it in a single pass when it is wet. Crossings are built deliberately — reinforced grass, a culvert, or a hardened section — and the rest is kept off, which is a management problem rather than a construction one. The same applies during construction: a swale formed early and then used as a haul route is a drainage feature in shape only, and re-forming it afterwards means excavating the compacted layer out, not just re-seeding it.
- Does a swale count as attenuation?
- Only partly, and the distinction matters when a discharge rate has to be demonstrated. A plain swale CONVEYS: it slows the flow and reduces the peak somewhat by doing so, but it does not store a volume in the way a basin or a tank does. Adding check dams turns sections of it into storage between them, and that storage can be counted. If the strategy needs a volume held back, the swale is usually one component alongside a basin or a tank rather than the whole answer. And, as the stormwater methodology sets out, detention reduces a peak without reducing a volume — only infiltration and reuse remove water from the system.
- Which is cheaper?
- The swale is normally cheaper to build and dearer to neglect; the pipe is dearer to build and tolerates neglect for longer before it fails. A swale is earthworks and seeding rather than excavation, bedding, pipe and chambers, so the construction comparison usually favours it where the land exists — and the land is the hidden cost, because a strip of ground given to drainage is a strip not given to anything else. Over a life, the swale needs frequent cheap attention and the pipe needs rare expensive attention, and the honest comparison is an annualised one of the kind the life-cycle methodology describes rather than a construction figure for either.
- What actually kills a swale?
- Three things, in order of how often they happen. Mowing too short, which removes the vegetation that provides the roughness and the filtration and exposes bare soil to scour. Sediment accumulating at the inlet until flow bypasses or ponds behind it, which is a maintenance task nobody was told about. And compaction, from vehicles, storage or construction traffic, which ends the infiltration without changing the appearance at all. None of the three is visible as a failure until a storm arrives, which is why swales are specified with a maintenance regime written into the drainage strategy rather than left to whoever cuts the grass.