Volume gets you a number, rate tells you whether it works
Sizing a rain garden looks like a volume problem and it is presented as one everywhere. Take the area draining in, multiply by a runoff coefficient and a storm depth, and you have a quantity of water. Divide by how much water a square foot of garden can hold and you have a footprint. That arithmetic is on this page and it is correct.
It is also only half the job, and it is the half that does not decide anything. A garden that holds 80 cubic feet is a functioning stormwater practice if the subsoil takes that away in a day and a liability if it takes five. The gallons are identical. What changed is a soil property you cannot see and cannot look up, and the only way to know it is to dig a hole and time the water going down.
This is why so many rain gardens built to a rule of thumb disappoint. The rule — around ten percent of the contributing area — encodes an assumption about soil that may or may not be true under your particular lawn, and the failure is not dramatic. The garden just stays wet, the plants that were chosen for periodic inundation sit in permanent inundation and die, and within two seasons it is a bare depression with cattails in it.
What a square foot of rain garden actually holds
Three layers store water, and they are not equal.
The ponding depth is open water standing on the surface, so it stores at 100 percent — six inches of ponding is six inches of water. It is the cheapest storage there is, which is why it is tempting to go deep, and the reason not to is that ponding depth translates directly into how long the garden looks like a puddle. Deep ponding on slow soil is what produces the mosquito complaint.
The engineered soil media, typically a sand-heavy mix with some compost and fines, stores water in its pore space. Only part of the total porosity is available for storm water, because some of the pore space stays occupied by water the soil holds against gravity, so 18 inches of media at 30 percent available porosity contributes about 5.4 inches. That percentage is a property of the specific mix and should come from whoever is blending it rather than from a general figure.
An optional gravel layer beneath, in open-graded clean stone, stores at whatever the void ratio of that gradation is — commonly somewhere around a third, again from the supplier. It buys storage cheaply but adds depth to the excavation.
| Layer | Depth | Storage factor | Water held |
|---|---|---|---|
| Surface ponding | 6" | 100% | 6.00" |
| Soil media | 18" | 30% | 5.40" |
| Gravel layer | 0" | 35% | 0 |
| Equivalent stored depth | 11.40" | ||
That 11.4 inches is 0.95 feet, or about 7.1 gallons per square foot. A thousand square feet of roof shedding an inch of rain at C = 0.95 produces about 79 cubic feet, which needs roughly 83 square feet of garden. Just over eight percent of the contributing area — which is where the rule of thumb came from, for this combination of assumptions and no others.
Drawdown, and the two different clocks
There are two drawdown times worth knowing and they answer different questions. The surface drawdown is how long standing water is visible: ponding depth divided by the design infiltration rate. At six inches of ponding and a design rate of half an inch per hour, that is twelve hours. This is the number neighbours and mosquitoes care about, and it is the one most programmes set a limit on.
The full profile drawdown is how long until the whole thing is empty and ready for the next storm: total equivalent stored depth divided by the same rate. At 11.4 inches and half an inch per hour, nearly 23 hours. This is the number that determines whether back-to-back storms are handled or whether the second one goes straight over the overflow.
Both use the design rate, which is your measured rate divided by a safety factor. The safety factor is not padding. A test in dry soil overreads, fines wash into the bottom over the years, and construction equipment on the excavation floor can cut the real rate by more than half on its own.
Getting the rate
Dig or bore a hole at the depth the garden bottom will sit — not at the surface, because the layer that governs is the one at the bottom of the excavation, and it is often a completely different material from the topsoil. Presoak it so you are measuring the wetted rate rather than the first gulp of dry ground. Then time the drop. The arithmetic that turns a drop and a stopwatch into inches per hour, and into the minutes-per-inch figure the same test is often reported in, is on the percolation test calculator.
If the rate comes back slow, the correct response is usually to move rather than to enlarge. Infiltration practices are commonly excluded below some rate threshold, that threshold is set locally, and a very large garden on ground that will not drain is a very large wet spot.
Where a rain garden is the wrong answer
Slow subsoil, a high water table, shallow bedrock, a steep slope above or below, and proximity to a foundation, a septic drainfield or a well all count against it, and the separation distances involved are set locally rather than by any general rule. Where infiltration is off the table, the alternatives are different in kind: an underdrained bioretention cell that filters and discharges instead of infiltrating, storage in a detention basin released slowly through an outlet, a dry well if the problem is depth to good material rather than the material itself, or simply carrying the water further away in a french drain to somewhere it can be released. If the water is arriving off a roof, the downspout discharge calculator gives the volume at each spout and the rainwater harvesting calculator covers storing it instead.
Questions people ask
Is a rain garden really about ten percent of the area draining into it?
That ratio falls out of a specific set of assumptions — around six inches of ponding, a foot and a half of media at roughly 30 percent available porosity, an inch of rain, and a runoff coefficient near one. Change any of them and the ratio moves. Halve the ponding depth and you need substantially more area; size for a two-inch storm instead of one and you need roughly double. More importantly, the ratio says nothing about whether the soil beneath will drain, which is the thing that actually determines whether the garden works. Use it as a sanity check on a calculated number, not as the number.
How deep should the ponding be?
Deep enough to store useful volume, shallow enough to disappear within whatever drawdown time your soil and your local programme allow. In practice the drawdown limit usually governs: at a design rate of half an inch per hour, twelve inches of ponding stands for a full day, which is past what a lot of plants and most neighbours tolerate. If the volume will not fit in an acceptable ponding depth, the answer is a wider garden or more storage in the media, not a deeper puddle.
Can I dig a rain garden next to my house?
This is the question to take to the building department rather than to a web page, and it is worth taking seriously. Deliberately concentrating and infiltrating water near a foundation can put water into a basement or a crawl space, can affect the soil supporting a footing, and separation distances from foundations, septic systems, drainfields, well heads and property lines are set locally and vary widely. This page states no distance for any of them. The general shape of the answer is that rain gardens belong well away from the building, downslope, with the roof water carried to them rather than released at the wall — but the specifics are a local determination, not a rule of thumb.
What goes in the soil mix?
Bioretention media is typically dominated by a washed coarse sand, with a modest fraction of compost for plant establishment and a small fines content, and specifications are usually published by the local stormwater programme with gradation and organic content limits. What matters for this calculator is the available porosity, which the supplier blending the mix should be able to state. Do not substitute topsoil or a general garden mix: they hold water rather than pass it, they compact, and the garden then behaves like a bathtub with plants in it.
What happens in a storm bigger than the one I sized for?
It overflows, and that is meant to happen. The design question is not how to prevent it but where the overflow goes, at what level it starts, and whether it leaves in a controlled path or by cutting through the low point in the berm. Set the overflow deliberately, armour it so it does not erode, and know where the water goes from there. Whether that discharge point is permitted, particularly if it heads toward a property line, a street or a watercourse, is a local question and one worth answering before construction rather than after the first big storm.