Because residential water usage is not metered by appliance, it's quite complicated to trace back usage to individual appliances, toilets, etc. In the mid-90's the American Water Works Association sponsored a detailed study by Aquacraft, Inc. of residential water usage in 12 study sites across the country. Using "data loggers" attached to the main water service meter with proprietary software used to decompose the data generated into discrete water-using events (e.g., toilets flushing, showering, outdoor sprinklers) Aquacraft developed the estimates recorded below. While the software used appears to have some difficulties interpreting simultaneous events the data provide a useful starting point.
Residential Water Use by Study Site
||Walnut Valley, CA
||Las Virgenes, CA
||San Diego, CA
||Waterloo, Cambridge, Ontario
The most important conclusion from a water supply planning standpoint is that indoor water usage reductions are achievable through the introduction of new fixtures and other water conservation measures.Outdoor usage, like indoor usage, can typically be reduced by 30-35% using available water conservation landscaping practices.
The 12 study sites showed remarkable similarities in indoor water use while outdoor water use varied with annual weather patterns. Across all 12 sites, outdoor water use averaged 58%, ranging from 59-67% in Scottsdale to 22-38% in Seattle, Tampa and Waterloo, Ontario. A good indicator of outdoor water requirements is annual evapotranspiration (ET) for turfgrass, which ranged from 15.65” in Waterloo to 73.40” in Phoenix, Tempe and Scottsdale. Evaporatranspiration is an indicator of how dry the air generally is at a location, and the data confirms our general understanding that in dryer locations, more water is used outdoors.
Indoor water use averaged 69.3 gallons per capita per day (gpcd) across all 12 study sites, including leakage, ranging from 57.1 gpcd in Seattle to 83.5 gpcd in Eugene, Oregon. The households studied in Eugene took more showers and flushed the toilet more often than those in the 11 other study sites and they used more water each time they flushed the toilet and took a shower. Study participants in Eugene used 36% more water than the city with the lowest mean toilet flush volume (San Diego) and 82% more water each time they showered than the city with lowest mean shower volume (Waterloo and Cambridge). A very small number of the homes in each study site were responsible for the majority of the leakage. Ten percent of the homes were responsible for 58% of the leaks found.
Residential water use summary
Toilets and clothes washing are the two largest indoor water uses. The study recorded over 289,000 toilet flushes over two years, with over 14.5% less than 2 gallons/flush and 50.8% greater than 4 gallons/flush. Net potential savings when comparing ULF (ultra-low flow toilets) only homes in the study to non-ULF homes is 10.5 gallons/person/day.