Visualizing California's Dynamic Urban Water Use


This site contains interactive visualizations that allow the user to explore how effectively California water utilities conserved water over 2014-16, a critical period of historic drought, as well as the first half of 2017. By visualizing and exploring the relationships between water conservation in 2014, 2015, 2016, and 2017 in almost 400 utilities from around the state, we provide an interesting perspective of the recent historic drought.

For these visualizations we define "conservation" and "water savings" as the difference in water production between a timeframe of interest in a given supplier's service area, and its respective water production in 2013.

This exploratory analysis includes statewide data to show overall water use trends, as well as data for each of the 10 hydrologic regions in California, individual utilities in those regions, and related public interest and media coverage of the drought.

The comparison between conservation efforts in 2014, 2015, 2016, and 2017 also provides insight on the responsiveness of different utilities to different incentives: a voluntary call for conservation in 2014, a state mandate in 2015, the replacement of the state mandate by adjusted self-certified goals in 2016, and the end of the drought declaration in 2017.

Several key lessons can be drawn from these visualizations:

  • The reporting requirements put in place during the drought created an even platform for water utilities to keep track of important data, which in turn allows tools like this one to identify water use and conservation trends, drivers, and opportunities for enhanced water management at a variety of different scales. Further standardized tracking and reporting methods could facilitate the use of data for decision-making.

  • Water utilities collectively achieved significant water savings in the period between 2014 and 2016. While policies and regulations seem to have been significant drivers of water conservation throughout the state (e.g. higher water savings during the state mandate or in response to local watering restrictions), these visualizations show that water use and conservation are very site-dependent and utility-specific.

  • Many uncertainties remain about the human-water dynamics that made water savings possible between 2014 and 2016. A better understanding of local population behaviors towards water use, and responsiveness to different conservation incentives, could help water planners and managers tailor their conservation campaigns more effectively in the future, not only during drought, but also as a long-term water reliability strategy.

  • Authors

    This web tool was developed by the Urban Water Policy and Innovation Group at Stanford University.

    Patricia Gonzales, PhD student (

    Department of Civil and Environmental Engineering

    Engineering Research Center for Re-Inventing the Nation's Urban Water Infrastructure (ReNUWIt)

    Kim Quesnel, PhD student (

    Department of Civil and Environmental Engineering

    Engineering Research Center for Re-Inventing the Nation's Urban Water Infrastructure (ReNUWIt)

    Newsha Ajami, Director of Urban Water Policy (

    Woods Institute for the Environment

    Engineering Research Center for Re-Inventing the Nation's Urban Water Infrastructure (ReNUWIt)


    Water conservation data

    All water conservation data was obtained from the State Water Resources Control Board (SWRCB) Water Conservation Portal - Emergency Conservation Regulation, and pre-processed using the Pandas library in Python.

    Only utilities who were required to report to the SWRCB, and who had complete data for the summers of 2014, 2015, and 2016 (June-August) are included in this analysis. This includes a total of 408 utilities form around the state. We use summer months for comparison between different years.

    Media data

    Media data was retrieved by a novel open-source algorithm Articulate written in Python by our group. The algorithm uses Google Custom Search Engine API to retrieve and tally news articles that contain terms of interest over a specified time period. For this visualization project we show the number of articles that contain the term “California drought” or a combination of “California”,” drought(s)”, and water-related terms such as “water”, “snow”, or “rain”. Only articles from the following state and national newspapers (chosen for high circulation) are included: Wall Street Journal, New York Times, USA Today, Los Angeles Times, Sacramento Bee, Orange County Register, San Diego Union-Tribune, San Francisco Chronicle (SF Gate), and The Mercury News.

    Google Trends data

    Google Trends data is used as a metric of public interest. Google Trends is a free online tool that shows how often a word or term is searched for over a certain period. The number of searches is in relative terms, with the period with the most searches for that topic having a value of 100. For a comparison between media coverage and public interest, we extracted data for how often the term “California Drought” was entered into the Google search bar over the period January 2012 - December 2016 within the state of California. Search frequencies were tallied at both monthly and weekly time scales.


    All visualizations were written in JavaScript and initially prototyped in jsfiddle .

    Libraries used in these visualization include:

    This web application is powered by Heroku , a Platform as a Service (PaaS) that enables developers to build, run, and operate applications entirely in the cloud.