Hydroinformatics Blog - Advancing Technology for Understanding Residential Water Use

Posted Nov 9, 2022

Hydroinformatics Blog Post

Organized by the CUAHSI Informatics Standing Committee. Contributions are welcome, please contact Veronica Sosa Gonzalez at email hidden; JavaScript is required.

By: Camilo J. Bastidas Pacheco, Jeffery S. Horsburgh, Nour A. Attallah

Achieving higher efficiency in urban water management and planning requires understanding of how water is used at the household level. Daily patterns in consumption, potential for water savings and distribution of water use across end uses are essential inputs to water demand estimation, leak identification, design of programs to manage water demand, and water planning to ensure adequate supply (Giurco et al., 2008; Willis et al., 2011). In the past, our ability to characterize water demand has been limited by the temporal resolution (typically monthly values) of residential water metering data. Higher resolution data can increase the accuracy of peak demand estimation and reduce leak volumes that can go undetected. Sub-minute resolution data can be used to record and quantify end uses of water (e.g., toilets, faucets, showers, etc.) that have short duration (Cominola et al., 2018; Nguyen et al., 2015) and is usually collected using so-called “smart” water meters. However, while commercially-available “smart” meters are capable in some instances of collecting high temporal resolution data, applications that use those data to provide actionable information are still emerging. Collecting, managing, and analyzing high temporal resolution residential water use data are all challenging due to cost and technical requirements associated with the volume and velocity of data collected. Thus, to realize the full potential of smart water metering in urban water management, robust hardware and software systems are needed to not only enable the data collection and management, but also to enable the “smart” analysis and applications of the data to aid in decision making - both by water consumers and by water providers.

Methods and Infrastructure

The Cyberinfrastructure for Intelligent Water Supply (CIWS) project at Utah State University produced multiple, open-source, low-cost data-logging devices to demonstrate new techniques for collecting smart metering data along with data management software and tools that can be used to provide insights into residential water usage. They conducted multiple residential water use studies in Utah to demonstrate the applicability of these tools.

The vast majority of water meters currently used by water supply utilities for quantifying residential water consumption are analog, magnetically-driven, positive-displacement meters. Replacing these meters is expensive and can disrupt utility operations. With this in mind, we designed low-cost data collection devices that work on top of existing meters, without disrupting their operation, to collect, manage, and analyze sub-minute resolution data. Specific products developed include:

  • The CIWS-Datalogger (Bastidas Pacheco et al., 2020): a device that measures the magnetic field outside a magnetically-driven residential water meter and can collect data at a variable time resolution interval. The system couples an Arduino Pro microcontroller board, a datalogging shield customized for this specific application, and a magnetometer sensor. The battery life (which is the major constraint in the development of this type of hardware) for the device was estimated to be over five weeks with continuous data collection. Data collected using this system is typically within 2% of the volume recorded by the register of the meter on which they were installed and is ideal as input to algorithms for water end use disaggregation and classification (i.e., computing the distribution of water use across different fixtures, e.g., toilets, showers, etc.).

  • The CIWS-Node (Attallah et al., 2021): a device that builds on the CIWS-Datalogger, adding a Raspberry Pi for communication and computational capabilities for edge computing applications. The device can execute end use disaggregation computations and transmit summarized values using WiFi, reducing the burden of large data transmissions. This device can collect data for over three weeks, but it can be connected to a solar panel to extend its autonomy.

  • The CIWS-Pulse-Logger (Bastidas Pacheco et al., 2022): a datalogger device that measures water use at the full volumetric pulse resolution of the water meter instead of recording time aggregated water use data. Measuring water use using this datalogger can provide more accurate estimates of event occurrence, timing, and features along with producing more discriminative event features that can only be estimated from full pulse resolution data to make end use classification easier and more accurate.

  • CIWS Software (Bastidas Pacheco et al., 2021): an open-source, modular, generalized architecture to automate the process from collection of high temporal resolution residential water use data through analysis and presentation to data users. The CIWS software architecture was prototyped and tested: (1) on a single family residential property and (2) in multi-unit student residential buildings on Utah State University’s campus to demonstrate push and pull data communication models, respectively. CIWS was tested for scalability and performance within our design constraints and proved to be effective within both instances.

Results and Conclusion

The CIWS data collection hardware and data management software serve as examples for how the cyberinfrastructure for managing residential water supply can be advanced. The residential water use studies we conducted using these developments have provided important insights into how and when people use water, including opportunities for conservation, which is incredibly important for regions of the U.S. and other countries where growing urban populations and persistent drought conditions have strained existing water supplies.

Additional Resources

The data collection hardware and software described in this publication are open source. Instructions for its use, including hardware designs that can be used to construct or manufacture the data logging devices, instructions for deploying them, and source code for the software are available on GitHub: https://github.com/UCHIC/CIWS-Loggers


This material is based on work supported by National Science Foundation Grant CBET 1552444. Any opinions, findings, and conclusions or recommendations expressed are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.

About the authors:

Camilo J. Bastidas Pacheco is a Postdoctoral Researcher at the Utah Water Research Laboratory at Utah State University. He was one of the CIWS developers and is currently working on a project evaluating the impact of landscape irrigation audits on outdoor water use in residential users.

Jeffery S. Horsburgh is an Associate Professor in the Department of Civil and Environmental Engineering and Utah Water Research Laboratory at Utah State University. He led the CIWS project and pursues research focused on monitoring systems, data management, and cyberinfrastructure for environmental data.

Nour A. Attallah was also a CIWS developer and recently received his PhD degree from Utah State University. He has now joined the South Central Connecticut Regional Water Authority in Connecticut as a lead water data analyst.