Guidance on use of rainwater tanks

8. Tank rainwater to sustain development

Page last updated: March 2011

There are two common justifications for installing rainwater tanks. The first and most important is one of necessity in areas that are not served by mains water and where safe supplies of water are not readily or reliably available. The second is to provide a source of water that can be used as an alternative to mains water.

In areas without mains water

Rainwater tanks can provide a valuable source of drinking water in areas not supplied by mains water. Even in arid regions with low rainfall, tanks can provide a valuable resource. Conway et al. (1999) examined rainwater harvesting in an arid location (Giles) in central Australia where the median rainfall is only 119 mm per year. In an average year, a house with a 266 m2 roof area could collect 61.25 kL of water in a 27 kL tank and provide 168 L of water per day. In the worst rainfall year in 40 years, 9.5 kL could have been collected to provide 26 L per day. While these volumes would not be sufficient as a total resource, they could represent a substantial source of drinking water to augment a secondary source, such as groundwater used for other domestic purposes. This could be particularly important in areas where groundwater is too saline to drink.

At Mutitjulu (in the Northern Territory), with an average yearly rainfall of 300 mm, rainwater tanks have been installed to provide a minimum of 100 L of rainwater per day to seven houses (Grey-Gardner 2002). The community initiated the project to supply a better-tasting alternative to the local groundwater supply. The system was designed to provide ease of management in a remote area with limited access to water. To provide greater assurance of water quality each house was fitted with a point-of- use filter. However water quality tests have showed a very low prevalence rate of E.coli in the pre-filtered rain water (Chapman et al 2008, Grey-Gardner et al 2005).

In areas with mains water

Mains water is used for purposes ranging from drinking and food preparation, to toilet flushing and garden watering. The use of rainwater tanks as an alternative source of water for any of these purposes has the potential to reduce pressure on the limited surface and groundwater resources used to supply mains water to urban and rural communities. Reduced pressure on the mains supply provided by rainwater tanks could alleviate the need for additional dams in growth areas and the costs associated with producing water for all uses to drinking water standard.

One constraint that has been raised is that the lack of space in large urban centres limits the size of the tanks that can be installed. This problem is being exacerbated by the trend to higher density living. In these circumstances, at best, only small-capacity tanks can be installed but recent changes in tank design and shape provide a greater selection of choices.

Although small tanks will overflow during wet seasons, modest 1-2 kL tanks, which require little room, can capture a significant proportion of roof run-off. The proportion able to be collected is largely a function of volume and frequency of use. A household using rainwater for all domestic purposes will empty a small tank more often and hence increase the available storage when a rain event occurs. On the other hand, less storage will be available in households using rainwater just for drinking and food preparation. Garden watering is a high volume use, but it mostly occurs in drier months when replenishing rain is relatively infrequent, leading to long periods when small tanks may contain no rainwater.

Across Australia, average rainfall, patterns of rainfall, and residential water usage vary. In the capital cities rainfall ranges from 550-1670 mm per year (Bureau of Meteorology1), while indoor water usage varies from an estimated 150 kL to 350 kL. Hot water represents about 30%-35% of indoor use. These data have been used to calculate volumes that could be collected in 1 kL and 10 kL tanks and where the water is either used to supplement mains water for all indoor uses or just for hot water uses (see Table 4).

In Adelaide, for example, which has an average rainfall of just above 500 mm, about 57 kL of water would flow from a medium-sized roof of 150 m2 each year (see Table 4). Using data on rainfall patterns over the past 10 years, a 10 kL tank would be of sufficient size to ensure that most of the 57 kL was captured and available for use. A volume of 57 kL represents about 36% of annual use (160 kL). A small tank of 1 kL could still collect and contribute 39 kL (24%) to total household use.

If rainwater was used just to supply hot water, a 10 kL tank could contribute 37 kL, and a 1 kL tank 25 kL per year, or 65% and 44%, respectively, of the available roof run-off. Potential volumes of water that could be collected in 1 kL and 10 kL tanks in all Australian capital cities are shown in Table 4.

Other combinations of use would provide different proportions of collection. Mitchell et al. (1997) determined that for a house with a roof area of 203 m2 and with three occupants, use of rainwater from a 13 kL tank for laundry, toilet and outdoor use could achieve use of 49-56% of available roof run-off and a 30-40% reduction in mains water use.

Top of Page
Table 4: Indicative volumes of water collected in rainwater tanks in Australian capital cities
Potential volume of water collected per year (kL)*
1 kL tank
10 kL tank
for hot water supply
total indoor supply
for hot water supply
total indoor supply

* from a 150 m2 roof.
Source of data: M Allen, Department of Water, Land and Biodiversity Conservation (SA). Figures were derived from a daily water balance model using long-term rainfall data for each capital city and estimates of water use based on consumption for each city by a household with three occupants.
1. Data accessed on May 2010 from

In practical examples, the ‘Healthy House’ in Queensland used a 22 kL tank to collect rainwater from a 120 m2 roof. In 2000-2001 the tank supplied 165 kL (36%) of 458 kL used for total indoor use by a family of five (Gardner et al. 2002). The ‘Sustainable House’ in Sydney uses a 10 kL tank to collect rainwater from a small roof of 70 m2 to supply 230 L of water per day for all internal uses, except toilet flushing (Mobbs 1998). Coombes et al. (2000) showed that use of rainwater for hot water and toilet flushing from shared storages in the ‘Figtree Place’ development could result in up to a 45% reduction in mains water use. In a subsequent study at Maryville, Newcastle, where rainwater was collected in a 9 kL tank and used for hot water, toilet flushing and limited outdoor use, 28 kL of rainwater was used from the tank over 24 weeks (which represents 39% of available roof run-off) (Coombes et al. 2002c). During this period, only 25 kL of mains water was used in the house, representing a 52% reduction.

In addition to reducing consumption of mains water, household use of rainwater also reduces flows of stormwater into street gutters and then to receiving waters. Although house roofs only provide a fraction of the total urban surface run-off, even small tanks reduce these flows (for example, see Mitchell et al. 1997; Coombes et al. 2002a). A detailed discussion of this issue is beyond the scope of this guidance document.

Costs and benefits of rainwater tanks

There are several issues to consider when installing a rainwater tank. As well as the cost of the tank itself these can include:
  • transportation
  • installation
  • alterations to gutters and downpipes
  • a tank stand or foundation
  • additional plumbing
  • a first flush device
  • insect-proof screening and gutter guards
  • a pump (if necessary)
  • maintenance
  • development approval (if required).
Estimates of the cost of rainwater from domestic tanks have varied from $0.30 to $12.30 per kL (Van der Wel 2000; Coombes et al. 2002c; National Water Commission 2007). In 2006-2007 the average cost of mains water was about $1 per kL (National Water Commission 2007). However, since then costs of mains water have risen sharply with the result that the cost differential between mains water and rainwater is decreasing.

Top of Page

To a certain extent, a degree of variation in rainwater costs is to be expected, as a range of factors will influence costs. These factors include:
  • whether the rainwater tank is to be the sole source of supply or a supplementary supply (the former requires a larger storage capacity)
  • the range of uses, for example, using rainwater to supply hot water in a house will require less additional plumbing compared to substituting rainwater for parts of the cold water supply in an existing house
  • whether installation is part of new construction or is a retrofit to an existing dwelling
  • amount and seasonality of rainfall.
Most estimates have been limited in scope to determining the cost to the householder or installer of the tank. This approach reflects the general situation where the choice or decision to install a rainwater tank has largely been an independent and individual process. However, this excludes the potential benefits to the community of installing rainwater tanks, particularly if it was done in a coordinated way (for example, Coombes et al. 2002a). It has been suggested that installation of rainwater tanks in all new and redeveloped dwellings could provide community savings through reducing demands on mains water supplies, delaying the need for new water supply infrastructure and reducing the need for spending on stormwater infrastructure (Coombes et al. 2002a; Pezzaniti 2003).

There is evidence of increased recognition of potential community-wide benefits of installing rainwater tanks. In 2002-2003, when water restrictions were widely imposed in many parts of Australia, the potential benefits of rainwater tanks received greater attention. Offers of financial support, in the form of rebates, have been provided by the Australian Government, some state and territory governments and local councils.

In a few areas, local councils require installation of rainwater tanks with new dwellings and some developments are being designed with rainwater tanks included for all dwellings. Most of these programs have promoted outdoor use, together with indoor use for toilet flushing, laundry and, in some cases, hot water supply.

One note of caution is that householders have a poor record of maintaining rainwater tanks. In developments where rainwater tanks are included for all dwellings, depending on the nature of use of the water, it may be necessary to institute a centralised management system. This would increase costs.

In addition, in areas where rainwater tanks potentially provide breeding sites for mosquito vectors or viruses, such as dengue fever, the potential for inadequate maintenance has cost implications. Queensland and the Northern Territory have specific legislation relating to mosquito control, and monitoring and enforcement of compliance imposes costs. In the worst case scenario, illness and intervention programs associated with rainwater tanks would also have cost implications.