Do household surveys estimate tap water use accurately?

Evidence from pressure sensor based estimates in Coimbatore, India


In developing countries, quantifying domestic water use at the household scale is crucial for any water policy intervention. But this is often challenging as piped water supply is one among many other water sources which households use in developing countries and piped water supply is often unmetered.

In the context of the recent policy shifts towards 24x7 piped water supply for cities in these countries, identification of a reliable approach to quantify household water use is paramount. The most common approach has been household surveys. But there is no evidence that household surveys are accurate in estimating water use. It is surprisingly difficult for people to quantify how much water they have used, when the water comes out of a tap.

The study was conducted on 82 households in Coimbatore, Tamil Nadu. It used pressure sensor data to estimate water consumption and supply at the household level. This data served as a reference to evaluate the effectiveness of conventional household survey methods. The pressure sensors produced detailed, continuous and accurate information on all sources of water accessed through the household, but they were expensive and intrusive.

While household surveys on their own do not give reliable estimates of household water use, the results from our study shows that that a combination of household surveys with a well-designed water diaries coupled with simple one-time field measurements produced water use estimates comparable to the high-quality pressure sensor derived estimates of tap water use.

ATREE researchers setting up Water Flow Meters in households in Coimbatore