Building an Integrated Water–Land Use Database for Defining Benchmarks, Conservation Targets, and User Clusters Rebecca Dziedzic 1 ; Katelyn Margerm 2 ; Jeff Evenson 3 ; and Bryan W. Karney, M.ASCE 4 Abstract: Water utilities have large amounts of data at their disposal, which are seldom being used to their full potential. Integrating water billing records with land-use and demographic data organizes information and makes inherent correlations easier to understand, facilitating communication to stakeholders. This data was integrated for three Ontario (Canada) municipalities, Barrie, Guelph, and London. A summary tool was created, with proposed metrics and charts, that facilitates comparisons between cities, definition of benchmarks, and identification of targets for conservation. More than 60% of consumption in these cities is residential, and mostly lies below the Ontario average of 267 L=cap · day. Water user clusters were created through self-organizing maps, K-means, and hierarchical clustering, and selected according to their pseudo-F and Rand statistics. Users within the same or similar property codes were found to cluster together. The application of data-mining methods provides actionable information for utilities seeking to reduce demands and increase system sustainability. DOI: 10.1061/(ASCE)WR.1943-5452.0000462. © 2014 American Society of Civil Engineers. Author keywords: Water management; Conservation; Data analysis; Databases; Water demand; Land use. Value of Integrating Data “Divide and conquer, ” specialization, is a common motto in solving complex problems, which splits complex realities into a variety of disciplines, sectors, departments, etc. This means, however, that interactions and synergies between the segments are either ignored or downplayed. According to Hussey and Pittock (2012), three major barriers prevent greater integration between sectors and policy domains: data deficiencies (missing or disorganized), weak existing policies and frameworks (fragmented, inconsistent, lacking review), and cultural inertia/path-dependency (silo mental- ity). A consistent system for collecting data in an easily retrievable, standardized, and comprehensive fashion is instrumental in managing water demand (Cahill and Lund 2013). The present study focuses on data as a pathway to resolve the second and third types of barriers. It integrates water, land-use, and demographic data with the objective of facilitating both understanding and demand management. The United Nations Environmental Programme (2012) reviewed worldwide applications of integrated approaches to water resource management and recognized the need for better information management, stating, “Information is the foundation of good decision-making and planning, ” with reference to integrated water resources management in Agenda 21 (United Nations Conference on Environment and Development 1992). Although progress has been slow (Muste 2013), this type of effort has been facilitated in recent years due to advances in technology and information exchange, fostering a greater commitment to initiatives in data collection (Maidment 2008). Bringing together data from differ- ent disciplines fills in gaps of information and knowledge (Muste 2013). Boyle et al. (2011) stress the fact that utilities already have valuable data at hand. Utility billing data can be used to inform many types of management decisions, such as pricing, conservation marketing, and peak planning. The use of this data is supported by three characteristics: it is available to all utilities; it can be used to target specific customer groups with customized messages that are more cost-effective than broad public outreach programs; and it can enable an understanding of specific customers, leading to localized utility policies and strategies. Using more detailed data can provide utilities with greater insights on to how water is consumed over space and time (Polebitski and Palmer 2010). Jorgensen et al. (2009) indicate that demographics, dwelling characteristics, and household composi- tion all directly impact water consumption, conservation intention, trust, perceived behavioral control, perception, and habits. Polebitski and Palmer (2010), as well as Morales et al. (2011) joined utility billing data with census demographic and property appraisal data to forecast residential and non-residential water use, respectively. Shandas and Parandvash (2009) suggest that, given current population growth and urban development, approaching water use through the lens of urban planning, namely the structural and demographic drivers of consumption, can improve the effectiveness of water conservation. Brooks (2006) defines water demand management operationally according to five motivators: (1) reducing the quantity or quality of water required for a specific use; (2) adjusting the nature of the task so it can be accomplished with less or lower-quality water; (3) reducing loss in quantity or quality of water in the distribution system; (4) shifting time of 1 Ph.D. Candidate, Dept. of Civil Engineering, Univ. of Toronto, 35 St. George St., Toronto, ON, Canada M5S 1A4 (corresponding author). E-mail: re.dziedzic@mail.utoronto.ca 2 Senior Engineering Researcher, Canadian Urban Institute, 555 Richmond St. W., Suite 402, Toronto, ON, Canada M5V 3B1. 3 Vice President of Urban Solutions, Canadian Urban Institute, 555 Richmond St. W., Suite 402, Toronto, ON, Canada M5V 3B1. 4 Professor, Dept. of Civil Engineering, Univ. of Toronto, 35 St. George St., Toronto, ON, Canada M5S 1A4. Note. This manuscript was submitted on January 22, 2014; approved on May 15, 2014; published online on July 21, 2014. Discussion period open until December 21, 2014; separate discussions must be submitted for in- dividual papers. This paper is part of the Journal of Water Resources Plan- ning and Management, © ASCE, ISSN 0733-9496/04014065(9)/$25.00. © ASCE 04014065-1 J. Water Resour. Plann. Manage. J. Water Resour. Plann. Manage. Downloaded from ascelibrary.org by UNIVERSITY OF TORONTO on 10/30/14. Copyright ASCE. For personal use only; all rights reserved.