How the two differ in kind
Both systems replace mains water for uses that do not need to be potable — flushing, laundry, irrigation, washing down. They are not alternatives in the ordinary sense, because they have opposite availability profiles and opposite problems.
Rainwater is collected from roofs, which makes it relatively clean and entirely dependent on the weather. Its yield is not annual rainfall multiplied by roof area: it is the LESSER of supply and demand at each moment, accumulated. When the tank is full, extra rain overflows and is lost; when it is empty, demand goes unmet however much fell last month. That makes it a simulation over a time series rather than a multiplication, and it is why tank size shows steep diminishing returns — a small tank captures the frequent small events that make up most of the annual total, and each further increment only catches rarer ones.
Greywater comes from baths, showers and basins, so it arrives whenever the building is occupied — reliably, in proportion to the people in it, and in a drought as readily as in a wet week. The price of that reliability is that it is warm, full of organic matter and biologically active. Left standing it goes anaerobic within about a day, and an anaerobic greywater tank is an odour complaint waiting to be made. So greywater storage is measured in hours, and the design problem is treating and using it fast rather than keeping it.
The factors that actually differ
Show
| Rainwater harvesting | Greywater reuse | |
|---|---|---|
| When it is available | When it rains, which is not when you need it. Irrigation demand peaks precisely in the dry spell that empties the tank. | Whenever the house is occupied. Independent of weather, which makes it the drought-resilient one. |
| What limits the yield | Storage against a demand that does not coincide with supply. Doubling the tank rarely doubles what it delivers. | The occupants. You cannot collect more than the showers produce, and it is a fairly steady number per person per day. |
| How clean it arrives | Roof-clean: dust, leaves, bird droppings, and whatever the first minutes of a storm wash off. Filtration and a first-flush diverter deal with it. | Soap, skin, hair, lint, fats and warmth. Treatment proportional to the end use, and that treatment is a system with a maintenance regime rather than a filter. |
| How long it can be held | Months, in a dark cool tank. It is stable. | About a day untreated, before it turns and smells. Storage is hours, not weeks, and the design assumes it is used promptly. |
| What it can be used for | Irrigation, flushing, laundry, washing down — the full non-potable range, with modest treatment. | Depends on treatment level. Subsurface irrigation with little; flushing and laundry with a good deal more, including disinfection in most regimes. |
| Regulatory load | Backflow prevention and clear labelling of non-potable outlets; generally straightforward. | Heavier almost everywhere — treatment standards, permitting, and restrictions on surface application. Some jurisdictions permit only subsurface irrigation. |
| The serious failure mode | Cross-connection to the potable system. The reason mains top-up must be through an air gap rather than a valve. | The same cross-connection risk, plus a treatment system that is not maintained and quietly stops treating while still delivering. |
| Shape of the cost | Dominated by the tank and its installation — largely a one-off, with filters and a pump as the ongoing items. | Dominated by the treatment plant and its servicing. A continuing cost, and a system that stops working the year nobody services it. |
Which one, and when
Choose rainwater harvesting when…
- The demand is seasonal and outdoors — irrigation, washing down, a garden that drinks in summer.
- There is roof area and somewhere to put a tank, above or below ground.
- The building is intermittently occupied, so there is little greywater to collect anyway.
- You want a system that needs attention twice a year rather than a plant with a service contract.
Choose greywater reuse when…
- The demand is constant and indoors — WC flushing in an occupied building, which is a steady load rainfall does not match.
- The building is well occupied, so the supply is substantial and predictable.
- Drought resilience matters: the supply that does not fail in a hosepipe ban is the one that comes from the shower.
- Roof area is small or unavailable — a flat in a block, a building with a green roof already committed.
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 water will a rainwater tank actually give me?
- Considerably less than annual rainfall times roof area, and the gap is the whole point. That figure is what the roof COLLECTS. What a tank supplies is the lesser of supply and demand at each moment, accumulated — so every full-tank overflow and every empty-tank shortfall is a loss, and they are both common because rain arrives in events and demand happens continuously. The standards for these systems specify a daily or finer time-step calculation against a real rainfall record for exactly this reason. An annual-average figure is a useful first screen and a poor design, and a tank sized from one is usually larger than it needed to be in a wet climate and smaller than it needed to be in a dry one.
- Why can't I just store greywater until I need it?
- Because it is warm and full of organic matter, which makes it an excellent culture medium. Within roughly a day the dissolved oxygen is gone, the bacteria present switch to anaerobic metabolism, and the products of that are the sulphurous compounds that produce the smell people associate with failed greywater systems. That is why untreated greywater storage is designed in hours and why systems are built around prompt use — direct to subsurface irrigation, or through treatment to a small balancing tank. If a design calls for days of greywater storage, it is really calling for treatment first and storing treated water, which is a different and more expensive system.
- What is a first-flush diverter and do I need one?
- A device that discards the first few minutes of runoff from a storm before the rest reaches the tank, on the basis that the first water off a roof carries most of the accumulated dust, droppings and debris. It is cheap, it is mechanical, and it makes a real difference to how often the tank needs cleaning and how hard the filters work. Sizing it is a matter of a volume per unit of roof area — enough to take the dirty portion, not so much that a light shower produces nothing. The maintenance detail that catches people is that the diverter has to drain down between storms; one that stays full does not divert the next event, and it becomes a small stagnant reservoir plumbed into the top of the system.
- Can either of these supply drinking water?
- Not as described here, and treating them to potable standard is a different undertaking from either system on this page. It requires treatment designed against the actual contamination risk, verification that it is working, and in most jurisdictions regulatory approval, ongoing sampling and a duty holder. The reason these systems are specified as non-potable is not caution for its own sake: it is that the failure is invisible and delayed, and the person who drinks the water is not the person who last serviced the filter. Non-potable outlets are labelled, coloured and kept physically incapable of being confused with potable ones for the same reason.
- What is the payback?
- On water price alone, usually poor in most of the places this site is read, and the honest presentation of that is to say so rather than to produce a number from an assumed tariff. Mains water is cheap relative to the capital cost of a tank, a pump, filtration and the pipework to keep the non-potable system separate; the arithmetic works where water is metered and expensive, or where the volumes are large. The cases that genuinely justify these systems are usually different: drought resilience where supply is restricted, a planning or building-regulation requirement that has to be met somehow, or a credit in an assessment scheme. Those are real reasons and they do not need a payback figure to be defended.
- How is the mains top-up supposed to work?
- Through an AIR GAP, not a valve — and this is the detail that regulators care about most. The mains feed discharges into the tank from above the water level with a physical gap, so there is no continuous path from the non-potable side back into the mains under any pressure condition. A non-return valve is not accepted as equivalent in most regimes, because valves fail and the consequence of this one failing is contamination of a public supply rather than of a building. Everything downstream — the pipework, the outlets, the taps — is then marked and coloured as non-potable, and that labelling is not decoration: it is what protects the next person who works on the building and does not know what you installed.