Online water-use feedback: household user interest, savings and implications

Abstract

This paper reports on the short- and long-term impacts of online water-use feedback provided via a smart metering trial involving 120 households in New South Wales, Australia. Near-real time water consumption feedback was provided via an online portal to half of the sample. Water consumption was uniquely analysed one year pre- and post-intervention, and in conjunction with login data. During one year of available access, the intervention group saved an overall average of 24.1 litres per household per day (L/hh/d) (4.2%). Regression analysis showed the significant savings of active users related specifically to portal login activity. Significant short-term effects persisted for 42 days, averaging at 63.1 L/hh/d. The article discusses the implications for research and practice, including a consideration of how, in addition to providing ongoing access, online portals could be leveraged further by water authorities to help meet urgent short-term supply constraints such as in drought.

Keywords:

• Sustainable urban water management
• water demand management
• education
• innovation
• metering
• monitoring

Acknowledgements

This research was supported by the Australian Research Council’s Linkage Projects funding scheme (LP110200767). The authors thank Chenxi Zeng (MidCoast Water), and James Riddell and Robert Yin (Outpost Central) and the household study participants for their collaboration. Thanks are also due to Thomas Boyle and Candice Delaney (ISF) for contributions to the study and portal design and Ian Chong for design work.

Notes

1. The existing Elster V100 meters were 20 mm helical vane (mechanical device) meters with 0.5L/pulse, 4x4 number wheel counters, and a backflow check valve for the potable water supply; compliant with the metrological and technical requirements of NMI R49–1 and AS 3565.1. These meters typically last 15 years or 3500 kL. Since most were low age (either manufactured in 2009 or 2010) with few kilolitres on the dial, measurement error was unlikely to be an issue.

2. In retrospect, internet access could have been required for study participation.

3. One additional household responded to an offer to have alerts set up by calling MCW’s Customer Service, rather than by logging on to the portal, so this household’s registration for alerts is included in this analysis.

4. This data includes the household that registered for alerts by calling MCW.