Roof Material, Slope, and Runoff: How Much Do They Really Change Your Harvest?
Every rainwater-harvest estimate on this site starts from the same fixed constant: 0.623 gallons for every square foot of roof for every inch of rain. That part of the formula never changes, because it's a unit conversion, not an assumption. The runoff coefficient sitting right next to it in the same formula is a different story — it's the one variable in the whole calculation that's genuinely about your specific roof, and it moves the final number by a wider margin than most people expect.
What the coefficient is actually standing in for
Not every drop that lands on a roof reaches a downspout. Some evaporates before it can run off, especially in warm, dry conditions or a light drizzle that barely wets the surface. Some clings to a rough or absorbent surface and never drains at all. Some splashes past the gutter entirely during an intense downpour. The runoff coefficient is a single number, between 0 and 1, that bundles all of that loss into one multiplier — a coefficient of 0.85 means the model assumes 85% of the rain that hit the roof actually reached storage, and 15% was lost to evaporation, absorption, or splash.
General rainwater-harvesting guidance commonly cites a range of roughly 0.75 to 0.9 for typical residential roofing, with smoother, harder, less absorbent surfaces landing toward the higher end and rougher or more porous surfaces landing lower. That's a wide enough range that it's worth seeing what it actually does to a harvest estimate, rather than treating the calculator's 0.85 default as the only number that matters.
The same roof, five coefficients
Running a fixed 1,500-square-foot roof section through one inch of rain, changing nothing but the runoff coefficient:
| Runoff coefficient | Gallons harvested | Overflow past 1 barrel | 55-gal barrels needed |
|---|---|---|---|
| 0.65 (rougher, more absorbent, or aging surface) | 607.4 gal | 552.4 gal | 12 |
| 0.75 | 700.9 gal | 645.9 gal | 13 |
| 0.80 | 747.6 gal | 692.6 gal | 14 |
| 0.85 (this site's default) | 794.3 gal | 739.3 gal | 15 |
| 0.90 (smooth, hard, well-maintained metal) | 841.1 gal | 786.1 gal | 16 |
The spread from the lowest to the highest coefficient in that table is about 234 gallons, or roughly 38% more water harvested at 0.90 than at 0.65, from the exact same roof and the exact same storm. That's a bigger swing than most people assume a single input could cause, and it's the reason a generic "just use 0.85" answer is a reasonable starting point but not the end of the analysis for a roof you're actually planning storage around.
Roof type, in general terms
Rainwater-harvesting guidance generally associates the higher end of the coefficient range with smooth, non-porous, well-sealed surfaces — standing-seam metal roofing is the most commonly cited example, since it sheds water quickly with very little surface absorption. Asphalt shingle, the most common residential roofing material in much of North America, is generally cited in the middle of the range, since the granular surface absorbs and retains a bit more moisture than bare metal, particularly when new. Wood shake or shingle roofing tends to be cited lower still, since wood absorbs noticeably more water than either metal or asphalt before it starts running off. A living or green roof, by design, is built to retain and slowly release water rather than shed it quickly, and sits far below the range this site's calculators are built around — it's a different category of roof for harvesting purposes entirely, not a variation on the same math.
Treat all of this as general, commonly cited guidance rather than a lab-measured spec for your specific roof. The condition of a given roof — its age, how it's been maintained, and what's built up on its surface — moves the realistic coefficient more than the nominal material category alone.
Condition matters as much as material
A roof's coefficient isn't fixed for its whole service life. Moss or algae growth, common on shaded or humid roofs, holds moisture against the surface and reduces how efficiently it sheds. Heavy leaf litter or pine needles do the same thing physically, and also clog the gutters that are supposed to carry the water away in the first place. Aging asphalt shingles that have lost surface granules absorb more than the same shingles did when new. None of this shows up as a single clean number anywhere — it's a reason to treat the top of the commonly cited range as optimistic for anything but a newer, well-maintained, actively cleared roof, and to lean toward the lower end for an older roof or one under a lot of tree cover.
It's worth seeing what that gap is worth in absolute terms on a bigger roof and a bigger storm. A 2,000-square-foot roof at a 2-inch rain event harvests about 1,744.4 gallons at a rougher, less-maintained 0.7 coefficient, versus about 2,118.2 gallons at a well-maintained 0.85 — a difference of roughly 373.8 gallons from maintenance and condition alone, more water than an entire 55-gallon barrel holds, from the identical roof and the identical storm.
What the coefficient doesn't capture
Two related factors sit outside the runoff coefficient entirely and are worth knowing about separately. The first is gutter capacity: even a roof with an excellent coefficient can lose water to overflow at the gutter itself during a very intense, short-duration downpour, if the gutters and downspouts can't physically carry the peak flow rate fast enough — water that overshoots the gutter never reaches the coefficient calculation at all, since it never entered the drainage system. A steeper roof pitch generally sheds water faster than a shallow one for the same rainfall, which can push a marginal gutter system past its capacity sooner in a heavy storm, independent of the roof material's coefficient.
The second is how many separate sections and downspouts a roof actually has. A large roof isn't necessarily one footprint feeding one downspout — it's often several distinct sections, each with its own coefficient-adjusted harvest, only some of which may be plumbed toward storage at all. The roof-area-times-rainfall math in the rain barrel calculator is exactly right for each section considered on its own; adding sections together only makes sense if their water actually converges on the same storage.
Valleys and multiple downspouts, worked out
A roof with a simple gable shape and a single downspout is the easy case — one footprint, one coefficient, one number. Most real roofs are more complicated: an L-shaped or T-shaped house, a roof with dormers, or a garage roof tied into the main house all create valleys, where two roof planes meet and combine their runoff, and typically feed two, three, or more separate downspouts. Treating that whole roof as a single footprint overstates what any one downspout, and any one barrel connected to it, will actually receive.
The fix is the same principle already covered for measuring footprint in general: work out which downspout drains which section, and run each section through the calculator on its own. A 2,400-square-foot roof split evenly across three downspouts isn't one 2,400-square-foot harvest opportunity for a single barrel — it's three separate 800-square-foot opportunities, each capable of filling its own storage independently, and each worth evaluating on its own rather than assuming one barrel could somehow catch the whole roof's total. This matters most for anyone comparing "how many barrels do I need" against a whole-roof total instead of a per-downspout one; the per-downspout number is almost always the more useful one to actually plan storage around.
Material and water quality — not just quantity
It's worth repeating a point that matters more here than almost anywhere else on this site: whatever a roof is made of, and however efficiently it sheds water, the water it sheds is not drinking water and shouldn't be treated at home to try to make it so. Different materials shed different things into that runoff along with the rain itself — granule grit from asphalt shingles, tannins and preservative residue from treated wood, and, on some older homes, lead from aging flashing or old lead-based paint on surrounding trim. None of that changes the volume math above, but it's a direct reason harvested rainwater from any roof stays an outdoor, non-potable resource regardless of how well-maintained or expensive the roofing material is.
Using a realistic number for your own roof
None of this is meant to replace the 0.85 default with a more complicated guess. It's meant to explain what that default is standing in for, so you can adjust it deliberately — toward 0.75-0.8 for an older, shaded, or debris-prone roof, toward 0.9 for a newer, smooth, well-maintained one — rather than treating either extreme as automatically correct. Run your own roof's numbers through the rain barrel calculator at a couple of different coefficients to see the realistic range for your situation, the same way this article did for the two example roofs above, and the catchment reference table shows the same math applied consistently across several common roof sizes. If your roof has more than one downspout, run each section separately rather than the whole footprint at once — it's a five-minute exercise that turns a rough whole-house guess into a realistic, per-barrel planning number.