AWWA
Research
Foundation
SCommercial and
SInstitutional End
Uses of Water
Subject Area:
Water Resources
Commercial and
Institutional End
Uses of Water
The mission of the AWWA Research Foundation is to advance the science of water to improve
the quality of life. Funded primarily through annual subscription payments from over 1,000 utili-
ties, consulting firms, and manufacturers in North America and abroad, AWWARF sponsors
research on all aspects of drinking water, including supply and resources, treatment, monitoring
and analysis, distribution, management, and health effects.
From its headquarters in Denver, Colorado, the AWWARF staff directs and supports the efforts
of over 500 volunteers, who are the heart of the research program. These volunteers, serving on
various boards and committees, use their expertise to select and monitor research studies to ben-
efit the entire drinking water community.
Research findings are disseminated through a number of technology transfer activities, includ-
ing research reports, conferences, videotape summaries, and periodicals.
Commercial and
Institutional End
Uses of Water
Prepared by:
Benedykt Dziegielewski, Jack C. Kiefer, Eva M. Opitz,
Gregory A. Porter, Glen L. Lantz
Planning and Management Consultants, Ltd.
Box 1316
Carbondale, IL 62903
William B. DeOreo and Peter W. Mayer
Aquacraft, Inc. Water Engineering and Management
2709 Pine St.
Boulder, CO 80302
John Olaf Nelson
John Olaf Nelson Water Resources Management
1833 Castle Drive
Petaluma, CA 94954
Sponsored by:
AWWA Research Foundation
6666 West Quincy Avenue
Denver, CO 80235-3098
Published by the
AWWA Research Foundation and the
American Water Works Association
Disclaimer
This study was funded by the AWWA Research Foundation (AWWARF). AWWARF assumes no responsibility for the
content of the research study reported in this publication or for the opinions or statements of fact expressed in the report.
The mention of trade names for commercial products does not represent or imply the approval or endorsement of AWWARF.
This report is presented solely for informational purposes.
Library of Congress Cataloging-in-Publication has been applied for.
Copyright 2000
by
AWWA Research Foundation
and
American Water Works Association
Printed in the U.S.A.
ISBN 1-58321-035-0
Printed on recycled paper
CONTENTS
T A B L E S .........................................................................................................................................ix
F IG U R E S ..................................................................................................................................... xiii
FO R E W O R D .................................................................................................................................xv
ACKNOWLEDGMENTS........................................................................................................ xvii
EXECUTIVE SUMMARY ..........................................................................................................xix
CHAPTER 1 INTRODUCTION....................................................................................................1
STUDY PURPOSE AND OBJECTIVES .................................................................... 1
ORGANIZATION OF THIS REPORT ....................................................................... 2
THE CI SECTOR OF URBAN WATER USE .............................................. ...........2
Defining the CI Sector and Categories of Users....................................... ............ 4
WATER EFFICIENCY PROFILES ................................................................................ 5
CHAPTER 2 ANALYTICAL REVIEW OF EXISTING INFORMATION ............................. ....7
DA TA SO U RCES ......................................................................................................... 7
U utility D ata Request ........................................................................................ 7
Literature Review ............................................................................................... 8
U published Reports .......................................................................................... 9
QUANTITY AND COMPOSITION OF CI WATER USE....................................... .9
Relative Importance of CI Sector ..............................................................9
C om position of CI Sector ...................................................................................... 12
C I E nd U ses ........................................................................................................... 14
VARIABILITY OF CI WATER USE RATES............................. ..........................21
Unit Rates of CI Water Use Measurement ......................................................22
Determinants of CI Use....................................................................................28
LITERATURE SUMMARY.............................................................................................35
CI Classifications and Data ...................................................................................35
CHAPTER 3 SELECTION OF CI CUSTOMER CATEGORIES FOR FIELD STUDIES AND
M O D E LIN G .............................................................................................................................37
IN TR O D U C T IO N ............................................................................................................. 37
PARTICIPATING STUDY SITES AND AGENCIES..................................................37
CI BILLING DATA AND RELATED INFORMATION............................................ 38
Data Analysis and Results..................................................................................... 38
FINAL SELECTION OF CI CATEGORIES FOR IN-DEPTH ANALYSIS ................45
Potential Determinants of Demand in Selected CI Categories ............................ 46
CHAPTER 4 DIRECT MEASUREMENT FIELD STUDIES ............................................49
INTRODUCTION.............................................................................................................49
PROCEDURE ................................................................................................................... 50
Selection of Study Sites by Utilities..................................................................... 50
Site Visits .............................................................................................................. 51
Data Analysis .................................................................................................. 54
RESULTS............................................................................................................... 60
Office Buildings .................................................................................. ............ 60
Restaurants .......................................................................................................... 69
Supermarkets....................................................................................................... 76
Hotels .............................................................................................................. 82
High Schools .......................................................................................................91
CHAPTER 5 GENERALIZED MODELS OF CI WATER USE.......................................... 97
INTRODUCTION.............................................................................................................97
DATA SOURCES AND STRUCTURE............................ ..................................................97
OVERVIEW OF ESTABLISHMENT LEVEL DATA......................................... ........... 98
Office Buildings ..............................................................................................98
Hotels and M otels................................................................................................. 99
Supermarkets...................................................................................................... 100
Restaurants .......................................................................................................... 101
Schools .......................................................................................................... 102
Transition to M odeling........................................................................................ 104
M ODELING APPROACH ............................................................................................. 104
Proposed M ethodology ....................................................................................... 104
Conditional Demand Analysis ......................................................................... 104
Evaluation of CDA M odels................................................................................. 107
M modifications of CDA M odels............................................................... .... 108
M ODELING RESULTS ................................................................................................. 109
Office Buildings .................................................................................................. 109
H otels and M otels................................................................................................ 110
Superm arkets........................................................................................ 111
Restaurants .......................................................................................................... 112
Schools ........................................................................................................... 114
Significant M odeling V ariables ......................................................................... 115
CHAPTER 6 BENCHM ARKING AN ALYSIS ....................................................................... 116
INTRODU CTION ...................................................................................................... 116
Purpose of Benchm parking .................................................................................. 116
Benchm parking M measures and V alues ................................................................. 116
Benchm parking A ssum options .............................................................................. 117
Efficiency-in-U se Benchm arks .......................................................................... 117
DEVELOPMENT OF BENCHMARKING MEASURES ........................................ 118
BENCHM A RKING RESULTS ................................................... .............................. 119
Superm arkets............................................... ..................................................... 119
Office Buildings .................................................................................................. 124
Restaurants .......................................................................................................... 127
Hotels and M otels................................................................................................ 130
Schools ........................................................................................................... 133
EFFICIENCY BENCHM ARKS ..................................................................................... 136
Restaurants .......................................................................................................... 137
Hotels and M otels............................................................................................... 138
Office Buildings .................................................................................................. 139
Superm markets .................................................................................................... 140
Schools ........................................................................................................... 141
CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS ............................................. 143
GENERAL CON CLU SION S .................................................................................... 143
CI Classifications and D ata ......................................... .......................... .... 143
Conservation Experience..................................................................................... 145
vii
CONCLUSIONS FROM DIRECT MEASUREMENT FIELD STUDIES.................... 146
Conclusions Regarding Conservation Potential in Study Categories .............. 147
MODELING CONCLUSIONS................................................................................. 150
Statistical M odels ................................................................................................ 150
W ater Efficiency Benchm arks ............................................................................ 151
RECO M M EN D A TION S ................................................................................................ 154
APPENDIX A SIC CLASSIFICATIONS FOR CI SECTORS............................................... 156
APPENDIX B MODEL VARIABLES...................................................................................... 184
APPENDIX C INTERNAL VALIDITY OF WATER USE MODELS: A DISCUSSION ...... 191
APPENDIX D CI MODEL VALIDITY AND BENCHMARKING TABLES ...................... 201
APPENDIX E CI MODEL DATABASE AND AUDIT TABLES......................................... 211
APPENDIX F CI CONSERVATION OPPORTUNITIES AND EXPERIENCE.................... 223
R EFER EN C E S .................................................................................................................. 251
A B B RE V IA TIO N S.......................................................................... ........................................ 262
TABLES
ES.1 Characteristics of significant CI categories in five participating agencies........................xxiii
ES.2 Efficiency benchmarks for restaurants ..................................................................... xxv
ES.3 Efficiency benchmarks for hotels and motels..................................................................... xxvi
ES.4 Efficiency benchmarks for office buildings ..................................................................... xxvi
ES.5 Efficiency benchmarks for supermarkets ................................................. ................ xxvii
ES.6 Efficiency benchmarks for schools................................................................................. xxviii
2.1 Sectoral distribution of public-supplied water delivered in the U.S....................................... 11
2.2 CI water use in nine Southern California cities..................................................................... 11
2.3 Distribution of commercial water use by category in selected cities.................................... 15
2.4 Distribution of institutional water use by category in selected cities.................................. 16
2.5 End uses of w after in hospitals .................................................................................................. 18
2.6 End uses of w ater in schools.............................................................................................. ... 19
2.7 End uses of w ater in hotels ................................................................................................ ... 19
2.8 End uses of w ater in office buildings ............................................................................... .... 20
2.9 End uses of water in commercial laundries............................................................................ 20
2.10 End uses of w after in restaurants ....................................................................................... 21
2.11 C1 rates of water use for selected cities in Southern California......................................... 23
2.12 Estimated water use per employee by broad SIC classification.......................................... 24
2.13 Selected commercial and institutional water use coefficients ........................................... 26
2.14 Selected commercial and institutional unit use coefficients................................................ 28
2.15 Seasonal water use in select Cl categories......................................................................... 31
2.16 Seasonal water use in the CI sector in selected cities ..................................... ............ .. 32
2.17 Examples of CI conservation technologies.................................................................... 34
3.1 Analysis of selected common CI categories in five study sites............................................. 41
3.2 Characteristics of significant CI categories in five participating agencies............................ 44
3.3 Alternative rankings of CI customers in five participating sites ...................................... 45
3.4 Potential explanatory variables and demand indicators for selected CI categories ............... 48
4.1 Size and occupancy of field study office buildings ........................................... ........... ... 61
4.2 Annual and seasonal water use at field study office buildings................................................ 62
4.3 Water use patterns at field study office buildings during data logging periods ....................62
4.4 Estimated annual end uses of water in field study office buildings ......................................67
4.5 Normalized end uses in field study office buildings ..........................................................67
4.6 Model parameters from field study office buildings ..........................................................68
4.7 General information on field study restaurants ......................................................................69
4.8 Annual and seasonal use in field study restaurants ..............................................................70
4.9 Water use patterns at field study restaurants during logging periods....................................70
4.10 Average annual end uses in field study restaurants...................................................... 73
4.11 Restaurant use normalized on the average number of meals served per day ......................74
4.12 Restaurant use normalized on the number of seats............................................................74
4.13 Modeling parameters for restaurants ............................................................................... 75
4.14 General information about field study supermarkets ....................................................76
4.15 Annual and seasonal use in field study supermarkets .......................................................77
4.16 Water use patterns at field study supermarkets during logging periods..............................77
4.17 Estimated annual end uses in field study supermarkets ...................................................80
4.18 Normalized end use at field study supermarkets................................................................80
4.19 Modeling parameters for field study supermarkets............................................................81
4.20 General information on field study hotels .........................................................................82
4.21 Annual and seasonal demand at the field study hotels..................................................83
4.22 Water use patterns at field study hotels during logging periods....................................83
4.23 Disaggregated indoor use at Irvine hotel.......................................................................85
4.24 Disaggregated indoor use at Phoenix hotel .......................................................................87
4.25 Annual water use in field study hotels by end use .............................................................88
4.26 Normalized annual hotel water use on a per room basis .................................................89
4.27 Modeling parameters for field study hotels................................................................... 89
4.28 General information on four field study high schools......................................................92
4.29 Annual and seasonal uses at four field study high schools..................... .....................93
4.30 High School water use during logging periods ....................................................................93
4.31 Estimated annual end uses at four field study high schools ......................................... ...94
4.32 Normalized annual use per square foot of building area at high schools............................95
4.33 Normalized annual water use per person at field study high schools..................................95
4.34 Model parameters from field study high schools .............................................................96
5.1 Office building characteristics and average water use ............................................... .....99
5.2 Hotel and motel characteristics and average water use................................................... 100
5.3 Supermarket characteristics and average water use ......................................................... 101
5.4 Restaurant characteristics and average water use.................................................................. 102
5.5 School characteristics and average water use....................................................................... 103
5.6 Model for estimating office building water consumption................................................ 110
5.7 Model for estimating hotel and motel water consumption....................................................111
5.8 Model for estimating supermarket water consumption .................................................... 112
5.9 Model for estimating restaurant water consumption .......................................................... 113
5.10 Model for estimating water use in schools....................................................................... 114
5.11 Significant modeling variables in five CI categories........................................................... 115
6.1 Field studies benchmarks for supermarkets........................................................................... 120
6.2 Audit data benchmarks for supermarkets ............................................................................ 122
6.3 Predicted benchmark values for supermarkets......................................... .......................... 123
6.4 Field studies benchmarks for office buildings........................................................................ 124
6.5 Benchmarking values from audit data for offices .............................................................. 125
6.6 Predicted benchmark values for office buildings............................................................... 126
6.7 Field studies benchmarks for restaurants.............................................................................. 128
6.8 Audit data benchmarks for restaurants ................................................................................ 129
6.9 Predicted benchmark values for restaurants ........................................................................ 130
6.10 Field studies benchmarks for hotels ................................................................................... 131
6.11 Audit data benchmarks for hotels.................................................................................... 132
6.12 Predicted benchmark values for hotels ............................................................................. 133
6.13 Field studies benchmarks for high schools....................................................................... 134
6.14 Audit data benchmarks for schools .................................................................................... 135
6.15 Predicted benchmark values for schools ............................................................................ 136
6.16 Efficiency benchmarks for restaurants ............................................................................... 138
6.17 Efficiency benchmarks for hotels and motels...................................................................... 139
6.18 Efficiency benchmarks for office buildings ................................................ ............... 140
6.19 Efficiency benchmarks for supermarkets ......................................................................... 141
6.20 Efficiency benchmarks for schools ..................................................................................... 142
A.1 Nonresidential water use coefficients.................................................................................... 157
A.2 Analysis of selected common CI categories.......................................................................... 168
A.3 Per employee water use in a sample of CI establishments in Southern California.......... 172
B.1 Model variables for restaurants............................................................................................. 185
B.2 Model variables for hotels..................................................................................................... 186
B.3 Model variables for grocery stores ........................................................................................ 187
B.4 Model variables for office buildings..................................................................................... 188
B.5 Model variables for schools .................................................................................................. 190
D .1 Superm arkets.......................................................................................................................... 202
D.2 Hotels and Motels ............................................................................................................ 203
D .3 O office B u ildings..................................................................................................................... 205
D .4 R restaurants ....................................................................................................................... 206
D .5 Schools .......................................................................................................................... 208
E.1 Benchmarking measures for audit data Supermarkets ........................................................ 211
E.2 Benchmarking measures for audit data Office buildings................................................... 212
E.3 Benchmarking measures for audit data Restaurants............................................................. 214
E.4 Benchmarking measures for audit data Hotels and motels ................................................ 216
E.5 Benchmarking measures for audit data for schools............................................................ 219
F.1 Sample of current or planned CI projects in selected states................................................ 226
F.2 Survey costs by CI category ................................................................................................... 237
F.3 Potential savings and costs of CI conservation.................................................................... 238
F.4 Range of audited minimum and maximum savings with and without irrigation................ 240
F.5 Summary of potential savings per site by measure category............................................... 241
F.6 Identified water savings by CI category and measure......................................................... 244
F.7 Identified water savings by institutional category and measure ......................................... 245
F.8 Reported implementation rates per CI category .................................................................. 246
F.9 Conservation measures: implementation rates and water savings ...................................... 246
F.10 Reported reasons for not implementing recommended measures............... .............248
FIGURES
4.1 D ata logger used in this study............................................................................................ 53
4.2 Example of monthly use pattern in an office building with irrigation ..................................55
4.3 Example of a simple CI flow trace from small office building.............................................58
4.4 Example of flow trace from large hotel........................................................................... 58
4.5 Cold water use in 4 sub-metered motel rooms ....................................................................59
4.6 Irrigation and cooling use in large office......................................................................... 59
FOREWORD
The AWWA Research Foundation is a nonprofit corporation that is dedicated to the
implementation of a research effort to help utilities respond to regulatory requirements and
traditional high-priority concerns of the industry. The research agenda is developed through a
process of consultation with subscribers and drinking water professionals. Under the umbrella of
a Strategic Research Plan, the Research Advisory Council priorities the suggested projects
based upon current and future needs, applicability, and past work: the recommendations are
forwarded to the Board of Trustees for final selection. The foundation also sponsors research
projects through the unsolicited proposal process; the Collaborative Research, Research
Applications, and Tailored Collaboration programs; and various joint research efforts with
organizations such as the U.S. Environmental Protection Agency, the U.S. Bureau of
Reclamation, and the Association of California Water Agencies.
This publication is a result of one of these sponsored studies, and it is hoped that its
findings will be applied in communities throughout the world. The following report serves not
only as a means of communicating the results of the water industry's centralized research
program, but also as a tool to enlist the further support of the nonmember utilities and
individuals.
Projects are managed closely from their inception to the final report by the foundation's
staff and large cadre of volunteers who willingly contribute their time and expertise. The
foundation serves a planning and management function and awards contracts to other institutions
such as water utilities, universities, and engineering firms. The funding for this research effort
comes primarily from the Subscription Program, through which water utilities subscribe to the
research program and make an annual payment proportionate to the volume of water they deliver
and consultants and manufacturers subscribe based on their annual billings. The program offers
a cost effective and fair method for funding research in the public interest.
A broad spectrum of water supply issues is addressed by the foundation's research
agenda: resources, treatment and operations, distribution and storage, water quality and analysis,
toxicology, economics, and management. The ultimate purpose of the coordinated effort is to
assist water suppliers to provide the highest possible quality of water economically and reliably.
The "end uses" of water in the commercial and institutional service sector is a
fundamental planning issue. Water conservation and resource planners need an accurate picture
of how private businesses and public institutions consume water. Engineers rely upon end use
information to identify design capacity and other engineering parameters. The limited
availability of "high quality" water resources requires that we understand more precisely how
water is used. Most existing end use information is very limited in scope. Unfortunately,
engineers and planers are left to estimate commercial and institutional facility end uses in the
absence of sound predictive models. This project provides a more comprehensive and accurate
picture of end uses within the commercial and institutional setting.
Julius Ciaccia, Jr. James F. Manwaring, P.E.
Chair, Board of Trustees Executive Director
AWWA Research Foundation AWWA Research Foundation
ACKNOWLEDGMENTS
This study has been supported through the financial and in kind participation of the
following water agencies:
Irvine Ranch Water District
San Diego Water
Santa Monica Water Department
Phoenix Water Services
Los Angeles Department of Water and Power
San Diego County Water Authority
Metropolitan Water District of Southern California
The researchers also gratefully acknowledge the advice and guidance provided by the
project advisory committee: Al Dietemann, Bill Hoffman, Cliff Pugh, Bill McDonnell, and John
Baliew. AWWARF Project Manager Robert Allen was a great help in bringing this project to
fruition. Jon Sweeten of Metropolitan Water District reviewed the draft report and provided
many useful suggestions. The researchers also wish to thank the following individuals and
organizations for their contributions to this study: David Lewis, Doug Kobrick, Jane Ploeser,
Dale Lessick, Luis Generoso, Paul Handley.
Finally we wish to thank all of those individuals who assisted by providing data,
completing survey forms, answering questions, accompanying auditors, and assisting with data
logging equipment. This study would not have been possible without you.
xvii
EXECUTIVE SUMMARY
The purpose of the Commercial and Institutional End Uses of Water study is to:
Summarize and interpret the existing knowledge base on commercial and institutional
(CI) uses of utility-supplied potable water in urban areas,
Present the results of field studies in a sample of 25 establishments in five urban
areas,
Provide econometric end use models for various categories of CI customers, and
Develop a set of efficiency benchmarks for five important CI categories restaurants,
hotels and motels, supermarkets, office buildings, and schools.
The water use of CI customers involves approximately one fourth of the total quantity of
water demanded for an urban area (USGS 1995). Despite the substantive proportion of total
urban water use for CI customers, comparatively little attention has been focused on the water
usage of this sector.
The CI sector consists of a large number of dissimilar customers with regard to the
purposes of water use. The lack of benchmark measurements of the quantities of water used for
cooling, cleaning, sanitary, and landscape uses within subgroups of similar establishments is an
obstacle to designing CI water efficiency programs and to developing reliable estimates of CI
water use and efficiency savings.
CHARACTERIZATION OF THE CI SECTOR
Variability Of CI Water Use
Heterogeneous customers with highly variable use characterize the CI sector. The
variability of CI use can be examined by expressing water use in individual establishments or
categories of users in terms of water use per unit, which removes the effect of establishment size,
however size or scale might be measured. In cases where sector-wide use is compared between
cities, the effect of the distribution of establishment sizes is also important. The total number of
customers within a CI category, the number of employees, and total output (or some other
appropriate volume of commercial activity) can be used to derive and compare rates of water use
per unit and examine potential sources of variability. The unit rates of use are helpful in
examining the relative efficiency of use among individual establishments, categories, or the CI
sector as a whole in different cities.
CI Classifications and Data
The systems of classifying CI customers by water utilities are generally inadequate for
comparing water use for individual categories between cities. At this point in time, only a few
categories are adequately defined and comparable. These include such categories as
hotels/motels, schools, restaurants, laundromats, car washes, and other easily recognizable types
of businesses. Other categorizations are difficult to generalize due to lack of data. Also, a
significant percentage of CI water customers do not fall into these categories and remain within
the generic category of "other CI users."
IDENTIFICATION OF CONSERVATION POTENTIAL
Billing records can be used for identifying the potential for water conservation among CI
customers for a given water utility based on such characteristics of water use as:
Degree of homogeneity of water use types (or composition of end uses) within a
given CI category
Inter- and intra-class variability of per account water use
Total water use by category relative to the CI sector use
Number of customers within category
Presence of seasonal water use
Categories of CI users with high cross-sectional variability of usage rates and/or
variability of usage rates throughout the year are likely candidates for conservation programs.
Another important consideration is the number of customers within the category that have to be
approached during program implementation. Categories with fewer users that account for a
significant percentage of total water use represent more focused conservation targets than
categories with a large number of customers. Single large users such as an oil refinery or state
university are significantly harder to move to implementation compared to a hotel chain for
example.
Conservation Findings
Findings and implications regarding the design and implementation of CI conservation
programs are based on the information on opportunities for water conservation described in this
report:
Some large-water-using categories have been ignored for water audits. Water audit
programs need to include warehouses, correctional facilities, military bases, utility
systems, and passenger terminals.
Potential savings are in the 15 to 50 percent range, with 15 to 35 percent being
typical. In addition, experienced payback periods typically range between one and
four years
Many CI water users do not need to use potable water in all applications. Each
customer and water use should be examined to determine if water of less-than-
drinking-quality can be used or recycled on-site, or if reclaimed effluent could
feasibly be used.
Discussion of the successes and failures of other programs can provide insight.
Cooperation between water and wastewater providers, and energy utilities is essential
in order to improve and optimize demand management programs.
Category-wide benchmarks of CI water use cannot be developed on the basis of
average daily or annual water use per active account (or customer) within a CI
category due to the differences in size of establishments that comprise the category.
In general, water conservation programs that have been implemented are rarely well
documented and evaluated. Many available documents lack direct information for generalizing
water savings. There is a need for more information on program costs, implementation
conditions, and measurement of savings.
SELECTION OF PRINCIPAL USE CATEGORIES
As part of this study, an analysis of CI categories and water use was performed on billing
data from five participating water providers:
1. Los Angeles Department of Water and Power, California
2. Irvine Ranch Water District, California
3. City of San Diego Water Utilities Department, California
4. City of Santa Monica, California
5. City of Phoenix Water Services, Arizona
Table ES.1 presents an initial selection of eleven CI categories, which account for more
than one half of CI use in the five sites. A complete set of billing records for a period of one year
was analyzed including:
Analysis of CI categories and use in each city independently
Comparison of CI categories and use across the five sites
Development of data that can be used to rank the categories according to conservation
potential
To aid in selecting five CI categories for further investigation, these eleven CI categories
were ranked according to scaled average daily use per customer. For a given CI category, this
construct is derived as average daily use per customer (in gallons per day) multiplied by the
fraction of total annual CI use accounted for by all customers in the given CI category. The
scaled average daily use per account balances the rate of water used by customers in a category
with the relative prominence of that category within the total use of the CI sector.
Of the eleven categories, the irrigation, car wash, and laundry categories are comprised of
very specific types of end uses directed at providing specific products or water services.
Although the individual customers in these categories display considerable variance in water use,
it was decided that a study of conservation opportunities for these categories should be narrowly
focused and perhaps better served by independent studies.
xxii
Table ES.1 Characteristics of significant CI categories in five participating agencies
Customer category Average Coefficient Percent of Percent of Percent Scaled
description annual of variation total CI seasonal average
daily use in daily use CI use customers use daily use
(gpdc)* (gpdc)t (%) (%) (o) (gpdc)**
Urban irrigation 2,596 8.73 28.48% 30.22% 86.90% 739.0
Schools and colleges 2,117 12.13 8.84% 4.79% 57.99% 187.0
Hotels and motels 7,113 5.41 5.82% 1.92% 23.07% 414.0
Laundries/laundromats 3,290 8.85 3.95% 1.38% 13.35% 130.0
Office buildings 1,204 6.29 10.19% 11.67% 29.04% 123.0
Hospital/medical office 1,236 78.50 3.90% 4.19% 23.16% 48.0
Restaurants 906 7.69 8.83% 11.18% 16.13% 80.0
Food stores 729 16.29 2.86% 5.20% 19.37% 21.0
Auto shops 687 7.96 1.97% 6.74% 27.16% 14.0
Membership organization 629 6.42 1.95% 5.60% 46.18% 12.0
Car washes 3,031 3.12 0.82% 0.36% 14.22% 25.0
* gpdc: gallons per day per customer
t Percent of CI customers pertains to CI customers in sites that have respective category only.
$ Coefficient of variation in daily use: The ratio of standard deviation of daily use to average of daily use.
Percent seasonal use = [(total annual use 12 x minimum month use ] / total annual use
** Scaled average daily use = average annual daily use in category x percent of total CI use attributed to the
category.
The auto shops and membership organization categories share similar qualities in that
they are comprised mainly of specific purposes (i.e., washing and sanitary uses, respectively).
Further, it was determined that the scope and intensity of water services in hospital and other
health-related settings blurred the distinction between CI and "light industrial" customers.
Therefore, the following five categories were selected for detailed analysis:
Schools
Hotel/motels
xxiii
Office buildings
Restaurants
Food stores
These categories represent CI customer types that are common to most cities, and which
present a diversity of end uses and therefore a good basis for examining conservation.
EFFICIENCY BENCHMARKING, RESULTS FROM FIELD STUDIES, AUDIT DATA
ANALYSIS, AND END USE MODELING
The statistical analysis of establishment level data for the five selected categories of CI
urban water users permitted estimation of models for predicting total water use in establishments
as a function of size, magnitude of operation, specific type of establishment within a broad
category, and presence of specific end uses. Data for a total of 433 establishments among the
five CI categories were used to develop the statistical models. These data were derived from
available water use audit databases.
These audit databases and data collected from field studies of 25 CI establishments (5
from each selected category) were analyzed to determine the benchmarks of average and
efficient rates of water use for each category of establishments. The derived values were
compared to predictions derived from the statistical models. The comparison of results from all
three sources (i.e., audit data, field study data, and modeled audit data) allowed the project team
to derive expected average rates of water use for various purposes as well as approximate values
of efficient use.
The efficiency benchmark was selected as the 25th percentile value for each efficiency
measure. This value does not constitute an absolute measure of efficiency; instead, it represents
an achievable low rate of use as evidenced by one-fourth of the sample establishments showing
usage rates at or below the selected value. The efficiency benchmarks developed in this study are
only approximations of efficient water usage based on the distribution of unit rates of water use
in the samples of establishments in five CI categories. The results of these comparisons are
presented in the following sections.
xxiv
These efficiency benchmarks can be used by utilities, businesses, property managers, and
others to assess the relative level of existing efficiency in establishments within the five
categories. Establishments on the higher end of water use for an efficiency measure are probably
the best candidates for CI conservation programs. More limited savings may be available from
establishments that already use water efficiently.
Restaurants
Table ES.2 shows the comparison data for restaurants. The data suggests that an efficient
restaurant would use approximately 130 to 331 gallons of water per square foot of building area
in a year. Also, an efficient restaurant would use around 6 to 9 gallons of water per meal served.
Furthermore, total water use for an efficient restaurant would fall within a range of use of 20 to
31 gallons per seat per day and 86 to 122 gallons per employee per day.
Table ES.2 Efficiency benchmarks for restaurants
End Use/Benchmark Measure N Efficiency Benchmark Range*
TOTAL WATER USE
Gal./sf/year 90 130-331
Gal./meal served 90 6 9
Gal./seat/day 90 20 31
Gal./employee/day 90 86- 122
*Developed from combined methods (field studies, audit data, and modeling results)
Hotels and Motels
Table ES.3 shows the comparison data for hotels and motels. The data suggests that an
efficient hotel or motel would use about 60 to 115 gallons per day per occupied room for indoor
purposes. Concerning cooling use, an efficient hotel or motel would use around 7,400 to 41,600
gallons per occupied room in a year. Efficient irrigation use would involve 16 to 50 inches per
year depending on the local weather conditions and the type of landscaping material. An
efficient hotel's total water use should fall within a range from 39,490 to 53,960 gallons per
occupied room per year.
Table ES.3 Efficiency benchmarks for hotels and motels
End Use/Benchmark Measure N Efficiency Benchmark Range*
INDOOR USE
Gal./day/occupied room 98 60 115
COOLING USE**
Gal./year/occupied room 97 7,400 41,600
IRRIGATION USE**
Inches per year 97 16 50
TOTAL WATER USE**
Gal./year/occupied room 98 39,000 54,000
* Developed from combined methods (field studies, audit data, and modeling results)
** Appropriate benchmarks will depend upon local climate.
Office Buildings
Table ES.4 shows the comparable data for office buildings. The data suggests that an
efficient office building would use for indoor purposes approximately 9 to 15 gallons per square
foot of building area per year. Also, efficient indoor use would involve 9 to 16 gallons per
employee per day. Concerning cooling use, an efficient office building would use around 9 to 22
gallons per square foot per day. Efficient irrigation use would involve 26-50 inches per year
depending on the local weather conditions and the type of landscaping material. An efficient
office building's total water use should range from 26 to 35 gallons per square foot per year.
Table ES.4 Efficiency benchmarks for office buildings
End Use/Benchmark Measure N Efficiency Benchmark Range*
INDOOR USE
Gal./sf/year 62 9 15
Gal./employee/day 72 9-16
COOLING USE**
Gal./sf/year 49 8.5 22
IRRIGATION USE**
Inches per year 47 26 50
TOTAL WATER USE**
Gal./sf/year 62 26 35
* Developed from combined methods (field studies, audit data, and modeling results)
** Appropriate benchmarks will depend upon local climate.
xxvi
Supermarkets
Table ES.5 shows the comparable water usage data for food stores. The data suggests
that an efficient supermarket would use between 24 to 52 gallons per square foot of building area
in a year. Also, an efficient supermarket would use approximately 0.008 to 0.029 gallons per
square foot per daily transaction. Concerning irrigation use, an efficient supermarket would use
about 30 to 50 inches per year depending on the local weather conditions and the type of
landscaping material. An efficient supermarket's total water use would range from 57 to 80
gallons per square foot bf building area and 3 gallons per transaction.
Table ES.5 Efficiency benchmarks for supermarkets
End Use/Benchmark Measure N Efficiency Benchmark Range*
INDOOR USE (WITH COOLING)**
Gal./sf/year 38 52 64
Gal./sf/daily transaction 38 9 16
IRRIGATION USE**
Inches per year 5 30 50
TOTAL WATER USE**
Gal./sf/year 38 57 80
Gal./transaction 38 3
* Developed from combined methods (field studies, audit data, and modeling results)
** Appropriate benchmarks will depend upon local climate.
Schools
Table ES.6 shows the comparison data for schools. Recall that the field study data comes
exclusively from high schools while the audit data includes information from schools ranging
from elementary to college level. The data suggests that an efficient school would use about 8 to
16 gallons per square foot per year for indoor use. Also, an efficient school would use between 3
to 15 gallons per school day per student for indoor use. Concerning cooling use, an efficient
school would use around 8 to 20 gallons per square foot per year. Efficient irrigation use would
involve 21.5 to 50 inches per year depending on the local weather conditions and the type of
landscaping material. An efficient school's total water use should range from 40 to 93 gallons
per square foot per year.
xxvii
Table ES.6 Efficiency benchmarks for schools
End Use/Benchmark Measure N Efficiency Benchmark Range*
INDOOR USE
Gal./sf/year 142 8-16
Gal./school day/student 141 3 15
COOLING USE**
Gal./sf/year 35 8 20
IRRIGATION USE**
Inches per year 132 22 50
TOTAL WATER USE**
Gal./sf/year 142 40-93
* Developed from combined methods (field studies, audit data, and modeling results)
** Appropriate benchmarks will depend upon local climate.
RECOMMENDATIONS
The results of this study and the insights gained through the field investigations and data
modeling and analysis support the following recommendations for the management of water
demands in the commercial and institutional sector of urban water users.
1. A standardized classification scheme of CI customers should be developed by water industry
to facilitate both demand planning and evaluation of conservation programs. The existing
classification systems are generally inadequate for comparing water use of similar customer
categories between different water providers.
2. Meaningful aggregate benchmarks can be developed by collecting additional aggregate data
on the size of the CI activity represented by a category. The aggregate measures of size
could include the combined square footage of all buildings in the category, total category
employment, school enrollment, seating capacity (in restaurants), number of hotel rooms or
other aggregate measures of business size in the service area. Benchmark values should be
developed by dividing the total category water use by the scaling measure.
3. It is recommended that water agencies institute a routine collection of supplemental data
from their CI customers on the size of their business. Only the relevant information should
xxviii
be collected to minimize the burden on the individual customers. The common measures of
business activity (and size) such as the number of employees, square footage of all buildings
or number of transactions per unit time should be adequate for benchmarking purposes.
Establishment-level benchmarks can be very useful in assessing the conservation potential of
individual CI customers. However, the development of meaningful benchmarking measures
would require good information on both the establishment's water use and the identification
of establishment type and size. Opportunities may exist for partnering with other utilities and
agencies to obtain this supplemental data. Many electric and gas utilities and taxing
authorities maintain extensive databases on CI customers including square footage, number
of employees, sales tax generated, etc. Water agencies should make use of existing data
resources whenever possible and should not bother customers with information requests
unless accurate establishment level data is not available.
4. Utilities should consider developing efficiency benchmarks for their larger CI end uses.
Maintaining this tool may require additional data collection or partnering with other utilities
and agencies, however, the net utility cost should be considerably less than blanketing all CI
customers for detailed conservation assistance. Whether it be for informing customers of
their relative rankings, providing technical assistance, or looking for major savings
opportunities, utilities should make better use of benchmarking as a CI conservation targeting
tool.
5. Future research should expand to other categories of CI customers and should also assess the
actual levels of efficiency in water uses. Establishments for which the individual end uses are
verified to be efficient should be included in the calculation of efficiency benchmarks.
Without such verification, the approximate range of efficiency benchmarks will remain
relatively wide to allow for the analytical uncertainty in deriving the efficient usage values
from data distributions.
xxix
CHAPTER 1
INTRODUCTION
This report presents the results of the Commercial and Institutional End Uses of Water
Study, which has been commissioned by the American Water Works Association Research
Foundation (AWWARF). The following team of consultants has performed work on this study:
Aquacraft, Inc. Water Engineering and Management (prime contractor)
Planning and Management Consultants, Ltd.
John Olaf Nelson Water Resources Management
STUDY PURPOSE AND OBJECTIVES
The purpose of this report is to: Summarize and interpret the existing knowledge base on
commercial and institutional (CI) uses of utility-supplied potable water in urban areas; present
the results of field verification studies in a sample of 25 establishments in five urban areas,
provide econometric end use models for various categories of commercial and institutional
customers, and develop a set of efficiency benchmarks for five selected CI categories. The
specific objectives are:
1. Review and create a database of existing information about CI water use (including
information on various end uses where available) and on previously implemented CI
efficiency programs.
2. Obtain information on the presence of specific end uses of water use, and quantities
of end uses from CI customers within selected categories by examining a sample of
CI customers for each of the categories and participating utilities.
3. Identify and develop water efficiency profiles or benchmarks of specific CI customer
categories found significant in terms of levels of water use and numbers of customers.
ORGANIZATION OF THIS REPORT
This report is organized into seven chapters. In this chapter, the CI sector is defined
including a description of the sources of data and information that were used to meet study
objectives.
Chapter 2 presents an analytical review of previously existing information on water use
in the CI sector. It describes the quantities of water that are used by the CI sector and identifies
principal categories of CI users; presents compilations of existing data on average rates of water
use in the CI sector and individual CI categories; provides a discussion of the determinants of CI
water demand and describes various measurement and analysis techniques. An associated
Appendix (Appendix F) describes the results of available studies of implemented CI
conservation programs and examines the potential for the future development of effective
conservation programs in the CI sector.
Chapter 3 describes the methodology and procedures used to select five specific CI
categories for examination in this study. The five categories that were ultimately selected by the
Project Advisory Committee (PAC) include: Office buildings, hotels and motels, restaurants,
supermarkets, and schools.
Chapter 4 presents the results of the direct measurement field studies. The chapter
describes the selection of the study sites, site visits, data analysis, and results from the five
selected CI categories.
Chapter 5 presents the generalized models of CI water use.
Chapter 6 presents the benchmark developed from three sources in this study direct
measurement field studies, audit data, and modeling results.
Chapter 7 includes conclusions and recommendations based upon the results and
experience with this study.
Detailed sets of appendices provide supporting data, additional methodology, and
discussion.
THE CI SECTOR OF URBAN WATER USE
Urban water supply utilities provide water for most uses that are found in urban areas.
Generally, a central water supply system is constructed to deliver water to individual customers
through metered connections. Because the cost of serving individual customers depends on their
water-use characteristics, water utilities group their customers into various classes that exhibit
similar characteristics.' Typically, the customers are grouped into residential and nonresidential
classes. These broadly defined classes of customers are also referred to as user sectors. These
main classes or sectors can be subdivided into subsectors. For example, the residential sector
can be subdivided into single-family, multifamily, mobile home, and other residential subsectors.
The nonresidential sector can be subdivided into commercial, institutional and/or governmental,
and industrial subsectors. Both the sectors and subsectors can be further subdivided into smaller
groups or categories. For example, the commercial sector can be disaggregated into such
categories as restaurants, hotels and motels, food stores, car washes, and others. Because of the
recent emergence of utility-sponsored water conservation programs, these groupings of
individual customers now hold an added significance. A useful system of sectors and categories
of urban water users would be one that subdivides all customers by both their cost-of-service
characteristics to support rate making and their potential for water conservation.
The literature on urban water conservation shows several definitions of the nonresidential
sector. The sector that contains the industrial, commercial, and institutional users of urban water
is designated as the ICI or CII sector. Where significant industrial customers are not present, the
term CI is often used. An alternative acronym, BIG stands for business, industry, and
government. The definitions of CI sectors vary between water utility and from study to study.
For instance, some agencies define the CI sector as all business accounts in the commercial
sector, which may include manufacturing and governmental establishments, while others may
separate industrial and institutional sectors. In addition, residential complexes such as apartment
buildings, mobile home parks, etc. whose accounts may be registered in the name of a business
entity are often considered commercial accounts. For the purpose of this study, the commercial
and institutional sector (or CI sector) is defined as a grouping of all nonresidential and non-
industrial users2 of urban water. In particular, commercial users are defined as retailers,
wholesalers, other services establishments, and institutional users such as public (or quasi-
public) providers of goods and services.
'Many utilities divide customers by meter size, which reflects the hydraulic requirements of the customer
connection.
2Many small manufacturing businesses have participated in CI audits and may contribute to participation in CI
conservation programs in general.
Defining the CI Sector and Categories of Users
The entire CI sector consists of a large number of dissimilar customers, particularly in the
way they use water. The combined water use of all CI customers typically constitutes
approximately 15 to 25 percent of total municipal water demand and more in some locations.
Despite its importance as a sector, CI uses have received less attention than the residential sector
in the development of water conservation initiatives nationwide. This is largely due to the
heterogeneous nature of this customer sector and a lack of knowledge regarding end uses of
water.
The only complete and consistent classification of the CI sector is based on the U.S.
Department of Commerce Standard Industrial Classification (SIC). According to this
classification, CI establishments are classified within the range of SIC codes 40 to SIC 97.
Examples of categories of CI sectors are listed in the first two columns in Appendix Table A.1
by two-digit and three-digit SIC.
The CI sector can be divided into two or more subsectors in most cases. Typically, the
CI sector can be broken down into commercial, institutional, governmental, and public
subsectors. Each subsector is divided into categories that most frequently represent various types
of establishments. The categories with customary names are usually inadequate for a complete
coverage of the entire CI sector. Some establishments cannot be classified and are usually
grouped under a generic name like "other commercial" or "general commercial".
While the SIC system is convenient for classifying the CI sector as a whole, the
groupings of individual establishments within two-digit or three-digit codes may not be helpful
for the design and implementation of water conservation programs. It must be pointed out that
the inconsistent definition of the CI sectors together with a general lack of information impedes
the synthesis of information about CI water use patterns from the literature.
John Boland developed a system of water use types for the CI sector that assigns codes of
water use for special purposes to SIC categories at two-, three- and four-digit levels, with a
significant fraction of water use for a named special purpose (Davis et al.1988). Water use by
employees for sanitary purposes is assumed present in all CI categories; the other nine special
purposes included:
I. Sanitary use by residents, students or other defined non-employee population
2. Sanitary use by patrons or general public
3. Food preparation use with food served to employees
4. Food preparation use with food served to residents, students, or other defined non-
employee population
5. Food preparation use with food served to patrons or general public
6. Water use for boiler feed
7. Water use as input to production of goods or services including water incorporated in
product, not seasonal in nature
8. Water use as input to production of goods or services including water incorporated in
product, seasonal in nature
9. Other water uses including vehicle washing, floor and driveway cleaning, etc.
Water use for seasonal purposes such as cooling, air conditioning, and occasional lawn
and shrub irrigation was not separately specified, since those uses can be present in any SIC
code, depending on the type and location of building.
As of 1997, the SIC system has been replaced by the North American Industry
Classification System (NAICS) which coordinates classifications in the United States, Canada,
and Mexico. The NAICS codes do not correspond with the old SIC system. It should be noted
that the water industry has not integrated the new classification system into general practice.
WATER EFFICIENCY PROFILES
An improved classification of the CI users of water is needed for water conservation
planners to be able to develop information on the quantities of water that would be reasonably
required to support various CI activities. Such quantities are often referred to as profiles or
benchmarks. An example of a benchmark usage is the per capital rate of urban water use, which
is usually obtained by dividing total annual water production (or the total volume of water
delivered to the distribution system) by total population served. Urban water use must be
disaggregated into somewhat more homogeneous groups of users in order to obtain meaningful
benchmarks for comparing efficiency of water use between utilities.3
The appropriate efficiency benchmarks would reflect quantities of water used for specific
purposes (also referred to as end uses), that accomplish the same purpose with less water. For
example, a toilet can be theoretically flushed with 1.6, 3.5, or 5 gallons of water. These three
flush volumes can be combined with the daily frequency of flushing per person to represent
efficiency benchmarks for the end use of toilet flushing. Although more or less distinct levels of
technical efficiency can be defined for some end uses, they do not cover all CI uses, and are
often in the form of design parameters that differ from actual performance in accomplishing the
purpose of water use.
The lack of benchmark measurements of the quantity and variability of water used for
cooling, cleaning, sanitary, and irrigation within categories of similar businesses is an obstacle to
designing CI water efficiency programs and to developing reliable estimates of CI water
efficiency savings. However, the development of benchmark quantities for all urban end uses
would be an enormous task that is not likely to be accomplished in the near future. Furthermore,
even if all existing end uses were "mapped out" in terms of their levels of technical efficiency,
such knowledge would be of limited use for conservation planners who must judge the levels of
efficiency based on total water use of customers within a given customer group.
A practical and meaningful benchmark of efficiency in water use is the rate of water use
that takes into account all variables, which affect the observed volume of water use other than
efficiency factors. For example, if average water use per school is compared between two cities,
it is necessary to take into account such variables as type and number of schools and students, the
size of irrigated play fields, swimming pools, type of cooling system, climate, and other
"normalizing" variables. Once the contribution of these variables to the observed average rate of
use is removed, the adjusted (or normalized) rate can be considered as an indicator of the level of
efficiency. One of the purposes of this study is to identify such measures of efficiency for
selected CI categories of urban water users.
3Disaggregation can also improve the accuracy of forecasting future needs.
CHAPTER 2
ANALYTICAL REVIEW OF EXISTING INFORMATION
This chapter presents an analytical review of previously existing information on water
use in the CI sector. This chapter:
Describes the quantities of water that are used by the CI sector and identifies principal
categories of the CI users;
Presents compilations of existing data on average rates of water use in the CI sector
and individual CI categories;
Provides a discussion of the determinants of CI water demand and describes various
measurement and analysis techniques;
In addition, an associated appendix (Appendix F) summarizes results of selected
available studies of implemented CI conservation programs and examines the potential for the
future development of effective conservation programs in the CI sector.
DATA SOURCES
Three major sources of information have been used in developing this chapter. First, a
survey of water agencies was conducted to obtain existing reports, studies, and data on their
water sales to the nonresidential sector. The agencies were also asked for information regarding
their methods for identifying the categories of CI water users. Second, professional and
academic publications were used to synthesize analytical understandings and perspectives on CI
water use characteristics. Third, industry reports and documents were also referenced as a
knowledge base about the experience of implementing CI water conservation programs.
Utility Data Request
To collect information on the current knowledge of CI water uses, the study team
conducted a mail survey of selected water utilities. A cover letter and a data request form were
mailed to 140 persons representing water utilities and other water agencies. The list of these
persons and their names and addresses was generated from a list of attendees at the Conserv96
conference in Orlando, Florida.
Three items were requested. First, information was requested on the categories
established by utilities to classify their nonresidential water customers and the criteria used for
classification. The purpose of this request was to identify and develop profiles of significant CI
customer categories. Second, data were solicited on the sectoral composition of annual water
use4 and the sectoral shares of total metered water sales in individual water agencies. This item
was requested to accomplish three tasks: (a) analyze the share of CI water uses in total metered
water use, (b) examine the variability of CI water sales within various CI categories and as a
whole, and (c) compile a database of existing information about quantities of CI water uses
across the U.S. Third, general information was requested on the characteristics of the utilities'
20 largest nonresidential (excluding manufacturing) water users. Specifically, these
characteristics included the types of business represented, the types of end uses present, and their
corresponding annual water use. The purpose of this request was to learn more about the
correlation between these characteristics and the presence of specific purposes of water use.
Unfortunately, only ten water utilities (7 percent) returned the mail survey forms with
some information regarding their records of CI water uses. Although the response rate of this
mail survey was low, the information received from water utilities was adequate to help provide
an overview of water demands in the CI sector.
Literature Review
In addition to the direct request of information from water utilities, the study team
reviewed academic and professional literature for existing studies on the characteristics and
databases of CI sectors. Compared with the residential sector, the CI sectors have received
limited attention in the literature. One difficulty in comparing the results from different studies
stems from the varied definition of the CI sector and its subsectors. Available literature tends to
group CI and industrial users into one sector defined as "commercial" or "business" accounts or
the "nonresidential sector." Due to the variability in the nature of CI water use purposes, case
studies of individual establishments are usually performed in lieu of sector-wide studies.
4Water utilities were asked to provide annual water use data of two sectors, namely, the residential and the
nonresidential sectors, in both units of measurement and in percentage of total metered water sales.
Unpublished Reports
Unpublished industry reports and documents commissioned by water utilities were
another important source of information. Though some statistics about water use patterns of the
CI sectors were found, several case studies of Cl conservation programs revealed that many
water utilities have little knowledge about their CI customers, even if they have implemented
some forms of CI conservation programs. Most of the implemented CI conservation programs
have been pilot programs, and represent spin-offs from residential programs. CI conservation
programs have typically focused on the provision of water-efficient sanitation fixtures, such as
ultra-low-flow toilets (ULFTs) or faucet aerators, which make them, at least on the surface, to
appear to be extensions of residential programs. Furthermore, conservation savings in these pilot
programs are usually based on estimates from identified or expected savings rather than on total
realized savings. The general lack of available recent data and statistics about the effectiveness
of existing conservation programs suggests that there is an urgent need to generate knowledge
and improve understanding of CI water use characteristics and conservation program
effectiveness in achieving water savings.
QUANTITY AND COMPOSITION OF CI WATER USE
The following sections describe and analyze available data and statistics collected from
various industry reports and documents, as well as surveys related to water use patterns and
trends of CI sectors.5
Relative Importance of CI Sector
An analysis of general trends in CI water use in the U.S. is limited by an infrequent and
short time series of national level data on CI water use. The U.S. Geological Survey (USGS)
compiles the best and most consistent data source for water use at the national level. The USGS
has developed national estimates of water use in the U.S. at 5-year intervals since 1950. The
5't should be noted that statistics throughout this section are generally site or study-specific.
most recent of these estimates of national water use is for 1990.6 Table 2.1 summarizes the
USGS data on the distribution of public-supplied water7 by sector.
Table 2.1 shows that at the national level, the total volume of public-supplied water
between 1985 to 1990 increased in all sectors except in the industrial and thermoelectric power
sectors. The data suggest that increases in water use in domestic, commercial, and industrial
sectors lag behind that of public use during the period. The increase in public-supplied water
demand comes largely from changes in the public uses and losses sector. Between 1985 and
1990, water use in the category of public use and losses increased by 3.1 percent (or 1,420
million gallons per day (mgd)) of total public-supplied water delivered. The share of the
commercial sector decreased slightly from 15.6 percent of total water delivered in 1985 to 15.3
percent of total water in 1990. Water delivered to the commercial sector in terms of volume,
however, increased from 5,730 mgd in 1985 to 5,900 mgd in 1990, at an annual growth rate of
three percent. Nevertheless, the combined share of the commercial and public sectors accounts
for almost 30 percent of total public-supplied water delivered, and is the second largest
destination of public-supplied water. The domestic sector, on the other hand, still accounts for
more than half of all treated water delivered in the U.S.
CI use varies from region to region. Differences in climate, as well as differences in the
economic and social factors, affect the amount of seasonal water use for irrigation and air-
conditioning, as well as general cooling water requirements. For instance, the American Water
Works Association (AWWA) conducted a survey of 331 large water agencies which estimated
that nonresidential users accounted for 44 percent of the total (metered) urban water use
(AWWA 1986). An independent survey of 28 agencies in Southern California, on the other
hand, estimated that commercial and public uses accounted for 18.8 percent and 5.1 percent of
metered urban water use respectively (Dziegielewski et al. 1990). Table 2.2 shows the quantities
of CI use in nine cities in Southern California. The CI sector in those cities is responsible for 15
to 30 percent of total municipal and industrial use (M&I) except for Anaheim, where large
demands by heavy industry reduce the relative contribution of CI sector to less than 10 percent of
6Subsequent to this analysis, the 1995 issue of USGS Estimated Use of Water in the United States has become
available.
7USGS (1990) defined public-supplied deliveries as water provided to users through a public-supply distribution
system.
M&I use. Table 2.2 also shows significant fluctuations of CI demands in time with significant
growth in some cities.
Table 2.1 Sectoral distribution of public-supplied water delivered in the U.S.
1985 1990
Sector Deliveries Percent Deliveries Percent
(mgd) (mgd)
Domestic 21,000 57.4% 21,900 56.8%
Commercial* 5,710 15.6% 5,900 15.3%
Industrial 5,730 15.7% 5,190 13.5%
Thermoelectric power 96 0.3% 80 0.2%
Public use and losses: 4,040 11.0% 5,460 14.2%
Total 36,576 100.0% 38,530 100.0%
Source: Calculated from U.S. Geological Survey. Estimated Use of Water in the United States. (1985, 1990).
* Commercial water use is defined by USGS as water for motels, hotels, restaurants, office buildings, other
commercial facilities, and institutions. The water is obtained from a public supply.
t Industrial water use includes water used for processing, washing, and cooling.
: Public water use is defined as water supplied from a public water supply and used for such purposes as fire
fighting, street washing, and municipal parks and swimming pools.
Table 2.2 CI water use in nine Southern California cities*
1980 1985 1987
City Deliveries Percentt Deliveries Percentt Deliveries Percentt
(mgd) (mgd) (mgd)
Anaheim 4.22 7.8% 6.15 10.0% 8.66 13.7%
Burbank 5.88 23.9% 5.15 24.8% 4.82 23.6%
Chula Vista 1.62 17.5% 1.68 17.3% 1.62 19.7%
Fullerton 5.21 15.1% 5.41 18.0% 5.41 17.5%
Los Angeles 141.45 26.9% 149.09 24.6% 135.40 --
National City 1.34 26.1% 1.85 27.6% 1.61 27.0%
Orange 6.35 30.1% 7.32 30.2% -- --
Santa Monica -- -- 3.07 18.6% --
South Gate 1.58 15.4% 2.03 18.3% 2.22 --
Source: Adapted from Dziegielewski, Opitz, Rodrigo (1990), Seasonal Components of Urban Water Use in
Southern California. Table 1-1, Table IV-1, and Table VI-I.
* Commercial use is defined as water use by business establishments and institutions except manufacturing plants.
The estimates include some use that is defined as public and other.
t Percent of all urban use in respective cities.
Composition of CI Sector
Water use within the CI sector can be examined in terms of:
The presence and prominence of individual categories of CI users such as restaurants,
hotels/motels, hospitals and others
The concentration of CI sector use among individual CI users
The distribution of use by specific end uses such as sanitary use, cooling, and others
The following sections describe the composition of CI water use and describe the typical
end uses within the individual categories.
Distribution of CI Use Among Categories
One way to analyze the CI user at the category level is to study the distribution of total CI
water use among individual categories of CI users. The distribution of CI use by category could
be analyzed in light of the following questions: What is the presence of CI categories in a
particular city/service area? How much water does each CI category consume? What are the
prominent categories in a particular service area?
Table 2.3 and Table 2.4 provide statistics on water use in CI categories in 12 selected
cities based on data for the period between 1992 to 1995 (EPA 1997). It is shown that
commercial use varies substantially as a percent of total CI water use, ranging between 23
percent in Portland (OR) to 86 percent in Orlando (FL). The overall weighted average of the
commercial share of total CI use among the 12 cities is about 62 percent as shown in Table 2.3.
This means that the institutional sector water use accounts for about 38 percent of the aggregated
CI sector water use as indicated in Table 2.4.
In terms of the share of all reported use associated with the CI categories, it is evident
that a few CI categories dominate total CI use in certain cities. For example, of all the identified
CI categories, the car wash class is absent or minor in the distribution of CI water use in Austin
(TX), Miami (FL), and Portland (OR). On the other hand, hospitality and offices account for
over 10 percent of total CI uses in most of the cities. Also, the communication and research
category accounts for almost one-third of total CI use in Burbank (CA), warehousing accounts
for about 31 percent in East Bay Municipal District (CA) and Orlando (FL), and over 10 percent
in Buffalo (NY) and Santa Monica (CA). Among the institutional categories, the utilities and
infrastructure category users dominated in Austin (TX) and Portland (OR), respectively.
Note that Table 2.3 and Table 2.4 data are sorted in descending order of weighted
category use, which allows one to clearly see which categories have the greatest influence in
overall CI sector use. Based on the weighted average percent share shown in the last column of
Table 2.2, the CI category designated as "hospitality" (which includes hotels and motels)
represents nearly 15 percent of CI use. A close second, warehousing, represents more than 12
percent of CI use. Warehousing includes significant uses of water for cooling and refrigeration
of perishable food products as well as sanitary uses of employees and visiting truck drivers who
also use water outdoors for washing trucks and containers. Other significant categories include
office buildings, health care facilities and urban irrigation of parks and landscapes in public
buildings.8 The largest institutional category is designated as "utilities and infrastructure" and
accounts for nearly 23 percent of CI use, mostly because of high shares for this category shown
by Austin and Portland (OR). Health care facilities are the second largest group and account for
over 10 percent of CI water use in eight cities. The education category accounts for nearly 6
percent of CI use and is the most consistent category among the 11 cities.
Data presented in Table 2.3 and Table 2.4 also demonstrate the geographic variability of
water use for a given CI category. For example, the education category share of total CI use
ranges from 0.97 percent in Buffalo (NY) to 11.96 percent in Santa Monica (CA). The share of
CI water use of hospitality ranges from 5.5 percent in Portland (OR) to 39 percent in Santa
Monica (CA). This variability can be attributed to both the geographic location and the method
of classifying the users into each category.
The geographic variability and prominence of a few CI categories in particular cities can
also be due to regional economic factors such as commercial policies, employment requirements
for particular skills, available resources, climatic conditions, population characteristics and
others. For instance, the warm climate conditions in the Sunbelt region fosters not only higher
use for irrigation and landscaping, but also contribute to growth in the senior population which
spurs the concentration of nursing homes, hospitals, and other healthcare categories.
8Landscape water use, particularly in the West and Southwest is often metered separately. At low-use CI sites there
is often only one meter. Therefore landscape water use can show up as "irrigation" water as in Table 2.3 and is also
embedded in all CI categories to a lesser or greater degree.
Concentration of Use Among Top Users
On the CI user level, one of the major characteristics of CI water use patterns is the
concentration of use. It is known that the distribution of use of the CI sector is highly skewed.
That is, a significant amount of sales are concentrated in the water use of a relatively small
number of customers. For example, a recent citywide CI sector survey performed by the City of
Albuquerque (Kiefer et. al., 1998) shows that approximately 80 percent of the total CI water sold
is used by only 20 percent of CI users.
CI End Uses
Because water conservation savings are achieved at the end-use level (i.e. at the specific
activity and/or apparatus that uses water), it is important to know the purposes for which water is
used in the CI sector and to examine the quantities of water for each purpose. This section
addresses the CI end-uses by focusing on the following questions: What types of CI end use are
present in a particular CI category? How much water is applied toward each end use? What are
the significant end uses in a particular CI class?
Compared to the residential sector, the specific types of uses in the CI sector and the
relative importance of these end uses in a particular CI category are complex. The variation in
the purposes of water use depends on the nature of business and the levels of technology and
water use efficiency in different business establishments. Domestic uses in kitchen and
bathroom are only some of the types of end uses present in the CI sector. Other uses such as
cooling towers, process rinsing, water treatment activities, and landscape irrigation are also
present.
Table
2.3 Distribution of
commercial water use by category
in selected cities (percent of CI sector use)
Commercial Users
Austin
Buffalo
Burbank
Glendale
Miami
Orlando Portland
MUD
Santa
Diego
Monica
St. Pau
MN
Reporting Year
1994-9!
Hospitality
Warehousing
Offices
* 13.26
t 13.97
20.94
10.83
15.81
13.45
30.77
11.37
34.86
30.94
34.28
12.29
Irrigation*
Miscellaneous commercial
6.82
Services
tt 5
21.94
10.32
31.05
18.15
Laundries
Vehicle dealers
and services
Meeting and recreation"t
Communication and research
27.84
Landscape*"
Transportation and fuels
Car wash
Passenger terminals
Share of Reported CI Use 46.13 84.67 76.64 81.86 62.87 79.99 86.43 22.77 75.27 67.38 71.9
.n l
source:
Derived from U.S. Environmental Protection Agency
1991). Table
Tabular values
are in percentages.
Hospitality includes restaurant/bar, overnight
accommodations, and other group shelter.
Office includes finance, insurance, real estate, and government.
Irrigation includes parks, gardens, botanical, zoological, cemeteries, and open land.
Miscellaneous commercial includes warehousing, warehouse-cold storage, and boat dock.
Sales include grocery stores, convenience stores, and dry goods.
Services include miscellaneous repair
services,
crematories, funeral homes, laboratories, and printing.
Meeting and recreation include convention center, recreation and theaters, and amusement parks.
Landscape includes landscape horticultural service, agriculture, soil preparation, crop services, veterinary, equestrian, livestock, poultry, a
38.55
15.9
16.8
13.0
11.9
Table 2.4 Distribution of institutional water use by category in selected cities (percent of CI sector use
Institutional users Austin Buffalo Burbank EB- Glendale Miami Orlando Portland San Santa St. Paul
TX NY CA MUD CA FL FL OR Diego Monica MN
CA CA CA
1992 1995 1995 1994 1995 1995 1995 1995 1995 1995 1994-95
Utilities and infrastructure* 32.34 0.67 0.77 1.88 8.49 5.59 73.04 0.98 0.0(
Health caret 5.83 12.03 16.73 5.62 18.21 11.5 4.8 3.5 10.94 20.43 17.1
Education* 11.14 0.97 10.19 8.30 7.16 7.33 1.55 0.27 11.41 11.96 8.5-
Church 1.43 0.31 0.67 2.70 1.18 0.70 0.42 1.19 0.21 1.45
Non-profit service and org.* 1.42 2.34 0.59 0.76 0.20 0.73
Military 2.42 0.02
Share of reported CI use" 53.16 15.4 28.36 18.14 37.15 20.01 13.42 77.23 24.72 32.6 28.0(
Source: Derived from U.S. Environmental Protection Agency (1997). Table 2. Tabular values are in percentages.
* Utilities and Infrastructure include police and fire station, public works/utility, electric steam, natural gas,
gas production and distributic
and disposal, construction, fumigating, and septic tank cleaning.
Health care includes health services, hospitals and nursing homes.
Education includes schools, museums and libraries, colleges/other schools, and social services.
Non-profit service and organizations include professional, labor, civic, political social organizations except churches.
The sum of percent of CI use in each city may not equal 100 percent due to rounding errors.
Table 2.5 through Table 2.10 present an allocation of end uses in hospitals, schools,
hotels, commercial office buildings, commercial laundries, and restaurants based on
measurements and estimates from water audits of six U.S. water service areas.9 The tables
further demonstrate the complexity and variation in the proportion of water used in each end use
as a percentage share of total water use in these CI categories. Note that in certain water-
intensive commercial categories such as car wash and laundry categories, water is the primary
ingredient of services these categories provide. Therefore, water consumption in these
establishments is highly concentrated in one particular activity. For example, the end use of
clothes washing in laundry accounts for the range of 81 to 90 percent of commercial laundry
water used (Table 2.9).
Water used for domestic purposes, via plumbing devices such as faucets, toilets, urinals,
and showerheads, is usually in the range of one-quarter to half of total water use by most CI
categories; except for the car wash and laundry categories mentioned above. The combination of
domestic and kitchen water use accounts for 80 percent in restaurants in the two service areas
listed in Table 2.10. The majority of water use in schools (Table 2.6) is for domestic purposes
(49%) and landscaping (37%). The combination of average water use for domestic and cooling
uses ranges from 47 percent in hotels (Table 2.7) to 64 percent in commercial office buildings
(Table 2.8). Water use in office buildings is concentrated primarily in the restrooms and kitchen
facilities. Landscape irrigation is also a significant use in the western sites.
9 Note that the precise definitions of CI categories are subject to variation in the reference and definition of the study
cited.
Table 2.5 End uses of water in hospitals (percent of total hospital use)
General Specific purpose Phoenix Denver Mesa Ventura Los Weighted
purpose Angeles averaget
Domestic Plumbing 24.33 39.7 22.95 37.87 18.65 27.05
Kitchen 8.5 4.53 2.86 4.51 6.51 6.04
Cooling Cooling tower 27.43 7.22 32.63 8.11 31.29 23.66
Evaporative coolers
(single-passing 5.08 8.8 7.76 na na 4.88
cooling)
Boilers 2.32 3.61 3.25 1.02 0.31 2.24
Process Rinses Photographic 2.00 4.91 13.99 3.42 7.26 5.78
processing
Product water
(miscellaneous na 5.43 0.58 na 10.85 3.12
rinses)
Cleaning Clean-in place (plant na 4.78 na na na 0.89
cleaning)
Sanitation Sterilizers autoclaves 6.04 4.91 na 16.95 4.65 5.42
Ingredients cleaning na na na 0.31 na 0.03
Laundry 7.68 12.33 na 8.43 0.5 5.91
Water treatment
Water treatment 3.42 na 2.4 6.48 16.18 5.22
regeneration
Landscape 13.16 3.77 9.35 11.59 3.3 8.77
Miscellaneous 0.04 na 4.22 1.30 0.50 0.97
Number of establishments 3 4 2 1 2 12
Average water use per establishment
Average water use per establishment 314,640 160,550 154,000 73,330 159,320 172,390
(gpd)
Source: Adapted from Journal ofA WWA, Vol. 84, No. 10 (October 1992), by permission.
Copyright 1992, American Water Works Association.
na = information not available
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t The average is weighted by the proportion of each service area in the combined total use of this category.
Table 2.6 End uses of water in schools (percent of total school use)
General Specific purpose Phoenix Denver Weighted
purpose average
Domestic Plumbing* 33.14 47.79 43.47
Kitchen 6.27 5.35 5.62
Cooling Cooling tower 1.51 5.21 4.13
Evaporative coolers (single-passing 0.16 na 0.05
cooling)
Boilers 0.80 na 0.24
Process rinses Photographic processing 2.09 5.30 4.35
Sanitation Ingredients cleaning na 2.93 2.07
Laundry 1.92 3.88 3.30
Landscape 54.11 29.54 36.77
Number of establishments 4. 5 9
Average water use per establishment (gpd) 36,390 87,110 61,770
Source: Adapted from Journal ofAWWA, Vol. 84, No. 10 (October 1992), by permission.
Copyright@ 1992, American Water Works Association.
na = information not available
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t The average is weighted by the proportion of each service area in the combined total use of this category.
Table 2.7 End uses of water in hotels (percent of total hotel use)
General Specific purpose Phoenix Denver Ventura Weighted
purpose average
Domestic Plumbing 17.08 30.62 33.72 23.97
Kitchen 18.31 9.96 na 13.26
Cooling Cooling tower 0.64 18.43 na 7.49
Evaporative coolers (single- 0.25 na na 0.13
passing cooling)
Process rinses Product water (miscellaneous na 6.41 3.62 2.85
rinses)
Sanitation Ingredients cleaning 4.67 17.25 29.76 12.03
Laundry 16.82 3.10 22.65 12.07
Water treatment regeneration 0.71 na na 0.37
Landscape 41.32 na 10.25 22.2
Miscellaneous 0.20 14.25 na 5.63
Number of establishments 4 2 1 7
Average water use per establishment (gpd) 202,140 153,070 38,940 131,390
Source: Adapted from Journal ofA WWA, Vol. 84, No. 10 (October 1992), by permission.
Copyright@ 1992, American Water Works Association.
na = information not available
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t The average is weighted by the proportion of each service area in the combined total use of this category.
Table 2.8 End uses of water in office buildings (percent of total office building use)
General Specific purpose Phoenix Denver Weighted
purpose average
Domestic Plumbing" 22.35 40.39 37.21
Kitchen 1.54 na 0.27
Cooling Cooling tower 56.05 20.97 27.15
Evaporative coolers (single-passing 1.77 1.61 1.64
cooling)
Boilers 0.68 5.24 4.44
Process rinses Photographic processing 0.25 0 0.04
Product water (miscellaneous na 0.10 0.08
na 0.10 0.08
rinses)
Sanitation Cleaning ingredients, containers 0.23 na 0.04
Laundry 1.54 na 0.27
Water treatment regeneration 4.13 na 0.73
Landscape 12.87 21.60 20.06
Miscellaneous 0.13 na 0.02
Number of establishments 13 3 16
Average water use per establishment (gpd) 55,930 261,530 139,150
Source: Adapted from Journal ofAWWA, Vol. 84, No. 10 (October 1992), by permission.
Copyright@ 1992, American Water Works Association.
na = information not available
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t The average is weighted by the proportion of each service area in the combined total use of this category.
Table 2.9 End uses of water in commercial laundries (percent of total commercial laundry use)
General Specific purpose Phoenix Denver Weighted
purpose average
Domestic Plumbing 2.49 3.53 2.92
Cooling Cooling tower 6.42 0.31 3.95
Evaporative coolers (single- 1.97 1.58 1.81
passing cooling)
Process rinses Product water (miscellaneous
na 0.31 0.19
rinses)
Sanitation Ingredients cleaning 80.73 89.78 84.38
Water treatment regeneration 8.26 na 4.91
Miscellaneous 0.13 4.34 1.84
Number of establishments 13 3 16
Average water use per establishment (gpd) 76,300 51,850 64,090
Source. Adapted from Journal ofA WWA, Vol. 84, No. 10 (October 1992), by permission.
Copyright@ 1992, American Water Works Association.
na = information not available
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t The average is weighted by the proportion of each service area in the combined total use of this category.
Table 2.10 End uses of water in restaurants (percent of total restaurant use)
General Specific purpose Denver Tri-county, Weighted
purpose FL average
Domestic Plumbing 27.75 35.33 31.05
Kitchen 48.48 50.00 49.14
Cooling Cooling tower 0.10 0 0.06
Evaporative coolers (single- 3.20 0 1.81
passing cooling)
Sanitation Ingredients cleaning 4.40 0.22t 2.58
Laundry 0.70 0 0.40
Landscape 4.30 2.45 3.49
Other 2.30 12.03: 6.54
Unaccounted 8.70 0 4.91
Number of establishments 3 6 9
Average water use per establishments (gpd) 7,524 5,800 6,773
Source: Derived from: Black & Veatch (1991). Nonresidential Water Audit Program. Aurora, CO: Black & Veatch.
Table 3-17; Southwest Florida Water Management District (1997). ICI Water Conservation in the Tri-County Area
of the Southwest Florida Water Management District. Brooksville, FL. Table 7.
* Plumbing includes lavatory faucets, toilets, urinals, and showerheads.
t Included also laundry.
$ Included also unaccounted use.
Tri-county area includes Hillsborough County, Pasco County, and Pinellas County.
** The average is weighted by the proportion of each service area in the combined total use of this category.
VARIABILITY OF CI WATER USE RATES
Heterogeneous customers and highly variable use characterize the CI sector. The
variability of CI use can be examined by expressing water use in individual establishments or
categories of users in terms of water use per unit, which removes the effect of establishment size,
however size or scale might be measured. In cases where sector-wide use is compared between
cities, the effect of the distribution of establishment sizes is also important. The total number of
customers within a CI category, the number of employees, and total output (or some other
appropriate volume of commercial activity) can be used to derive and compare rates of water use
per unit and examine potential sources of variability. The unit rates of use are helpful in
examining the relative efficiency of use among individual establishments, categories, or the CI
sector as a whole in different cities.
Unit Rates of CI Water Use Measurement
Unit Use Per Account
This unit usage rate is obtained by dividing total annual (or monthly) water use in a CI
category by the number of active accounts in that category. This measure is easy to obtain from
data that are routinely collected by water utilities. It measures the average water use rate by a CI
account in a day, and the unit is gallons per account per day (GAD). There are several variations
in the literature that are related to this term such as use rate per site (gallons per site per day) or
per working day.10 In this report, GAD is defined as:
GAD= (TGD / TA) (2.1)
where GAD is gallons per account per day in a category;
TGD is the total gallons used per day in a category;
TA is the total number of accounts in a category.
One problem with this definition stems from variation in the numbers of accounts within
a particular CI category from time to time. Further, growth and recession within a particular CI
category, influenced by other economic activities such as mergers, acquisitions, and
consolidations, may affect the numbers of employees and /or activity, and, hence use, from one
year to the other. In addition, the number of accounts per facility is often difficult to compile.
Table 2.11 demonstrates the regional and temporal difference in GADs for the CI sector
in Southern California. The range of GADs in these cities is between 219 in Chula
'1The distinction between GAD and gallons per account per working day (GAWD) calls for an appropriate choice of
the unit of measurement in different CI categories. For example, commercial offices and schools are usually in
operation only during weekdays (a yearly total of 260 days = 5 working days per week no. of weeks in operation
per year). GAWD will be a more representative measurement. On the other hand, facilities such as restaurants,
hotels, and hospitals are usually in operation during all days of a week, then GAD will also equal GAWD and will
be a more appropriate choice of unit. In the literature, GAD is most often reported as a measurement for all CI
categories because there has not been a correction made for the facilities' work schedules.
Table 2.11 CI rates of water use for selected cities in Southern California
City Year No. of Average use Non-manufact. Average use
accounts per account Employment per employee
(GAD) (GED)*
Anaheim 1980 538 5,822 92,174 34
1985 914 5,066 111,504 42
1987 -- -- 119,236 60
Beverly Hills 1985 872 2,515 59,589 37
1987 879 2,632 59,794 39
Burbank 1980 3,018 1,343 48,509 84
1985 3,056 1,421 54,005 80
1987 3,041 1,405 56,204 76
Chula Vista 1980 5,515 235 -- -
1985 4,917 219 -- --
Fullerton 1980 2,428 1,485 43,175 84
1985 1,683 2,904 42,203 116
1987 1,401 3,372 41,814 113
Las Virgenes 1985 382 5,093 18,625 104
1987 -- -- 22,302 73
Los Angeles 1980 52,270 1,868 1,433,941 71
1985 54,912 1,901 1,519,956 69
1987 -- -- 1,554,362 71
National City 1980 1,258 901 21,714 52
1985 982 1,507 29,478 50
1987 -- -- 30,011 41
Ontario 1987 1,624 1,863 31,100 97
Riverside 1980 -- -- 76,757 184
1985 3,489 4,178 89,102 164
1987 -- -- 94,040 169
San Diego 1980 15,004 3,141 425,901 111
1985 13,813 4,071 501,194 112
1987 13,723 3,956 517,537 105
Santa Monica 1985 1,543 1,602 62,990 39
South Gate 1980 1,465 909 14,151 94
1985 1,599 1,093 13,200 132
1987 -- -- 12,820 146
Source: Adapted from Dziegielewski et al. (1990). Seasonal Components. Table 1V-I.
* GED is defined by dividing the total gallons water used per day by the total number of employment in
an establishment or category.
Vista (CA) and 5,822 in Anaheim (CA). In this range, the highest usage per account is nearly 27
times higher than the lowest value. Further, there are considerable changes in GAD rates in
individual cities over time. This implies that factors other than the number of accounts affect CI
water use.
When the total number of employees in each city is applied to normalize water use, the
variation in usage rates in Table 2.11 diminish. The range of per employee rates is narrowed to
between 34 gallons per day in Anaheim (CA) and 184 gallons per day in Ontario; with the
maximum-to-minimum ratio of 5.4.
Table 2.12 shows per employee use rates that were estimated by Mercer and Morgan
(1973) for establishments designated by two digit SIC classes. The results show that in 1970 the
highest water use per employee" occurs in the category of arboreta, botanical and zoological
gardens (SIC 84) which is imbedded in the services category of Table 2.12. The second highest
water use per employee was found in the food preparation and processing industry, which is not
technically in the CI sector as defined for this study. The order of magnitude of change in a 2-
year span indicates the shortcoming of using this unit as an indicator of water use.
Table 2.12 Estimated water use per employee by broad SIC classification
Category SIC 1968 1970
Acre GED Acre GED
feet/year per feet/year per
employee employee
Transportation & utilities 41-49 0.390 348 0.040 36
Wholesale & retail trade 50-59 0.860 768 0.097 87
Finance 60-67 0.033 30 0.030 27
Services 70-89 0.226 202 0.228 204
State and local government 92-93 0.217 194 0.242 216
Source: Mercer, Lloyd, and Douglas Morgan (1974). "Estimation of Commercial, Industrial, and
Governmental Water Use for Local Areas." Water Resources Bulletin 10 (4). 794-801. Table 1.
Per employee rates for more disaggregated categories are shown in Table 2.13 for 22
selected CI categories. These rates of use have been used as library values for the IWR-MAIN
(DOS version) water use forecasting software program (PMCL 1995). The table shows use
among the two-digit SIC categories ranging from 29 gallons per employee per day in the
"Mercer and Morgan (1974) calculated SIC category water use per employee. That is, water use per employee in the
ith two digit SIC is equal to the sum of water use by sample and census firms in the ith two digit SIC divided by the
sum of employment in sample and census firms in the ith two digit SIC.
wholesale durable goods category to 462 gallons per employee per day in the personal service
category. The per employee rates weighted by employment were obtained by dividing the
combined water use in all sample establishments by the combined employment. This method of
deriving per employee rates compensates for the skewness in the distribution of per employee
rates in a sample of individual establishments. The mean establishment GED rate in Table 2.13
is significantly higher than the rate weighted by employment in almost all SIC categories (with
the exception of dentist offices). In the case of amusement and recreation services, the skew is
so severe that the mean GED is more than ten times higher than the weighted mean. In addition,
all categories show high variability of GED rates between the establishments as indicated by the
standard deviation.
Because of the high variability of per employee rates within the 2-digit SIC categories, it
is impossible to derive benchmark rates of usage at this level of disaggregation. In most
categories, 3-digit or even 4-digit SIC disaggregation may be necessary in order to reduce
significantly the variability of per employee usage rates within a particular group. Average per
employee rates in a in a sample of 1,405 establishments drawn from a six-county urban area of
Southern California and in a national sample of more than 3,000 establishments for 3-digit SIC
codes are included in Appendix A. The range and standard deviation in per employee usage
within individual categories shown on those tables indicate that higher levels of SIC
disaggregation continue to show high variability of water use for some categories. For those
categories, the use of employment as the "normalizing variable" may not be appropriate because
the volume of water used by the establishment depends on the presence of activities or
establishment features that are often independent of the number of employees.
Table 2.13 Selected commercial and institutional water use coefficients
SIC Description Sample Employment Mean Standard
size weighted customer deviation
GED use, GED GED
50 Wholesale durable goods 517 29.0 85 492
51 Wholesale nondurable goods 233 86.7 134 409
54 Food stores 90 97.9 194 490
55 Auto dealers & service stations 198 48.9 83 205
56 Apparel & accessory stores 48 67.7 132 302
57 Furniture & home furnish stores 100 41.7 169 169
58 Eating & drinking places 341 156.2 216 553
60 Depository institutions 97 58.9 97 189
61 Non-depository institutions 12 156.4 489 1,702
70 Hotels & other lodging places 197 229.8 740 1,491
72 Personal services 300 462.1 717 1,174
75 Auto repair, services & park. 108 216.6 685 1,805
79 Amusement & recreation service 106 427.1 5,552 30,565
80 Health services 354 90.6 308 743
802 Dentist offices 22 258.7 216 242
805 Nursing home facilities 106 196.7 237 291
806 Hospitals 122 75.4 95 94
82 Educational services 296 115.8 419 1,060
83 Social services 55 106.4 251 568
84 Museums, botanical, zoo, gardens 9 208.0 269 356
86 Membership organizations 45 212.3 257 442
91-97 Public administration 25 105.7 210 282
Source: Adapted from Planning and Management Consultants, Ltd. (1995). IWR-MAIN 6.1: User's Manual and
System Description. Appendix D.
Use Rate Per Unit of Output or Size
Another measurement of unit water use is the mean water use per unit of output in
gallons per unit of output per day (GUOD). The GUOD formula is written as:
GUOD = TGD / TUOD
(2.3)
where TUOD is total units of output per day at an establishment or within a category.
The problem with this definition is that it assumes all output produced requires an equal
rate of water use as an input. Another problem with this definition is the difficulty in measuring
the quantity of output in most commercial and institutional categories. The GUOD measure is
easier to construct in the industrial sector, where numbers of output such as the number of
widgets is easily countable.
Table 2.14 provides some examples of average rates of water use in selected CI
establishments based on various normalization factors related to output (Miller et al. 1983). The
data in Table 2.14 are based on information from surveys of CI establishments conducted during
the 1960s (Linaweaver et al. 1966).
Unit Use Per Floor Area
Table 2.14 also shows ten categories that use square footage as the unit for normalizing
use. For the CI sector, customer use of water is a significant part of total water use in addition to
employee use of water, particularly for customer-oriented establishments such as restaurants and
bars, lodgings, schools, personal services, and others. Under these conditions, an indicator of
water use by customers is the gross area of establishment.
McCuen, Sutherland, and Kim (1975) used a linear regression to model water use as a
function of gross area of department stores. The regression results can be interpreted to suggest
that for every one square foot increase in the gross area of the store, water use of the store would
be expected to increase by 0.056 gpd for retail department stores. The equation provides a
standard error of estimate of 1,520 gpd, which represents a significant reduction from the
standard deviation of 3,262 gpd for water use data, and the variability of estimated water use.
They argue that it is expected to provide reasonable estimates of water use. Also, Kim and
McCuen (1979) used a multiple correlation analysis and a principal components analysis to
estimate commercial water demand, which resulted in a coefficient of 0.0147 gallons per square
foot of gross area. In another study, Behling and Bartilucci (1992) estimated that the seasonal
rate of office building water use is about 0.045 gallons per day per square foot in the fall and
winter, and 0.106 gallon per day per square foot in the spring and summer.
Table 2.14 Selected commercial and institutional unit use coefficients
CI category Unit Gallons/unit/day
Barber shops Chairs 54.60
Beauty shops Station 269.00
Bus/rail depots Square foot 3.33
Car washes Inside square foot 4.78
Churches Member 0.14
Golf/swim clubs Member 22.20
Bowling alleys Alley 133.00
Residential colleges Student 106.00
Hospitals Bed 346.00
Hotels Square foot 0.26
Laundromats Square foot 2.17
Laundry Square foot 0.25
Medical offices Square foot 0.62
Motels Square foot 0.22
Drive-in movies Car stall 5.33
Nursing homes Bed 133.00
New office buildings Square foot 0.19
Old office buildings Square foot 0.14
Jails and prisons Person 133.00
Restaurants Seat 24.20
Drive-in restaurants Car stall 109.00
Night clubs Person served 1.33
Retail space Sale square foot 0.11
Elementary schools Student 3.83
High schools Student 8.02
YMCA/YWCA Person 33.30
Service stations Inside square foot 0.25
Theaters Seat 3.33
Source: Crews, James E. and Mary Ann Miller. 1983. Forecasting Municipal and Industrial Water Use. IWR
Research Report 83R-3. U.S. Army Corps of Engineers, Fort Belvoir, Virginia.
Determinants of CI Use
The variability of unit rate of use indicators such as GED, GAD, GUOD, within a Cl
category, across CI categories, and across time and region (given uniform definitions), is a
function of economic, climatic, and technological factors.
Economic Factors
Employee/Patron Requirements. The number of employees is a crucial factor
contributing to the total demand for water in a particular CI establishment. The definition of
employee requirements for water extends beyond the sanitary needs of salaried personnel to
include any persons that are regularly present in a business facility, such as students in schools,
patients in nursing homes and hospitals, and patrons of business. A large number of employees
and patrons results in a larger use of water in domestic uses such as kitchen and toilet uses.
Larger employment can also be associated with a larger floor area of an establishment, which
may increase the need for other water-consuming equipment such as cooling towers and other
end uses. Ultimately, a larger employment requirement tends to reflect a larger scale of services
and products provided to the public, generally leading to a larger demand for water.
Growth or Recession of Industry. The impact of economic growth and recession of an
industry over time changes the numbers and types of employees and business establishments in a
particular service area. If the change in the number of employees and business establishment is
not reflected in a proportional change in water usage, there will be a change in GED or GAD
over time in a particular CI category. For example, explosive growth in the semi-conductor
industry in Boston (MA), San Jose (CA), and Seattle (WA) not only increased the number of
employees in those areas, but also the number of auxiliary business services such as grocery
stores, laundries, and car washes to support the growth of employment. Then the demand for
water in the CI sector of a particular service area will increase, given that other usage remains
unchanged. Thus, the forecast of water demand of a particular category or the CI sector depends
on the projection of the future capacity of the category or sector.
Price of Water. The pricing of water has been one of the most debated issues in the water
resources industry, as well as in academic literature. The design of an optimal rate structure is
believed to be important to effective water demand management strategies. The various rate
structures debated include marginal rate, uniform rate, increasing block rate, decreasing block
rate, seasonal rate, and others. Most of these pricing structures have been studied with regard to
residential sector water demand, however, the effectiveness of various pricing strategies for the
nonresidential sector has not been settled.
Lynne, Luppold, and Kiker (1978) used a derived demand model to estimate the price
elasticity of commercial demand in the Miami (FL) area. The price elasticity was generally low
(inelastic) for all groups studied except for department stores. This group was found to have an
elastic demand for water at all prices above $0.93 per thousand gallons purchased per month,
where the mean price for the sample (including grocery and supermarket, motels/hotels, eating
and drinking establishments, and other commercial) was $1.24.
In another study (McCuen, Sutherland, and Kim 1975), water consumption in the CI
establishments in general is price inelastic because the water users (employees and customers)
are not directly responsible for water costs. In addition, water cost is just a small component of
the overall expenses of CI establishments when compared to energy, labor costs, and capital
investment. Thus, at least in the short run, a change in water rates will likely have a small impact
on most establishments, except those establishments such as car washes and laundries, which use
water as a major ingredient of services they provided. The major implication of these studies is
that commercial establishments may be more responsive to price changes over the long run.
Climate Conditions
Seasonality. A normalized rate of use in a given CI category in different regions may be
different if the weather conditions are different. Seasonality in water use refers to a
characterization that describes variation in water use due to weather conditions and other
seasonal cycles in business.12 The components of water use that are affected by seasonality may
include landscaping, cooling, and other uses. These uses fluctuate with weather and climatic
conditions in different seasons and in different geographical locations. For example, some
commercial establishments may use water for irrigation, air-conditioning or dust control during
the summer season. Furthermore, some commercial end uses, such as cooling water
requirements, are much higher during hotter, drier days, and in hotter and drier climate regions.
A survey study by Dziegielewski et al. (1990) found that 25 percent of commercial use in
Southern California is seasonal. The presence of seasonal variation in a CI sector may also
depend on the nature of business establishments. Sport clubs, for example, require more water in
the summer than winter due to their seasonal business cycles. Table 2.15 reports the shares of
seasonal use in selected CI categories by Dziegielewski et al. (1990). Seasonality of water
consumption in the CI sectors varies by type of category. For example, seasonal water use
12 Weather seasonality does not necessarily coincide with business seasons. For example, peak resort demand is in
the winter season in Phoenix (AZ) and in Florida.
accounts for 72 percent of water use in sport clubs, significantly higher than that of restaurants
(19 percent).
Table 2.15 Seasonal water use in select CI categories
CI establishments Sample size Percent seasonal use
Restaurants (SIC 5812) 9 19.1
Hospitals (SIC 8062) 15 20.1
Laundromats (SIC 7215) 14 22.8
Hotels/motels (SIC 7011) 15 27.1
Colleges (SIC 8221) 7 40.5
Schools (SIC 8211) 8 42.8
Sport clubs (SIC 7997) 6 72.4
Source: Dziegielewski et al. (1990). Seasonal Components. Table IV-7.
Seasonal use determined based on the minimum month method.
Because of climate differences and variation in the cost of water supply and wastewater
disposal, the degree of seasonality should be expected to vary regionally (Dziegielewski et al.
1990). As shown in Table 2.16, in Southern California, seasonal use as a percent ranges from 10
percent of total commercial water use in Santa Monica (CA) to 39 percent in Las Virgenes (CA).
In seven out of 14 of these Southern California cities, seasonal increase is more than one-third of
annual average demand.
Cooling Degree-Days. Requirements for cooling depend on the amount of heat that has
to be removed. The number of cooling degree-days is a measure of the amount of heat, which is
related to the temperature of the ambient air during the warm season. Davis et al. (1996) studied
the effect of cooling degree-days on municipal and industrial water use and found that the
elasticities of cooling degree-days were estimated to range from 0.016 to 0.021 for paper and
pulp plants, to 0.022 to 0.091 for poultry processing plants. Kiefer et al. (1995, 1996, and 1998)
have estimated significant relationships between nonresidential water use and the departure of
the monthly number of cooling degree-days from long-term monthly normals. Yet, there is no
study of the effect of cooling degree-days on CI water demand only.
Table 2.16 Seasonal water use in the CI sector in selected cities
City Seasonal use (%)
Anaheim 34.0
Beverly Hills 20.2
Burbank 14.1
Chino 36.6
Chula Vista (Sweetwater) 13.8
Fullerton 38.2
Las Virgenes 39.3
Los Angeles 13.2
National City 17.5
Ontario 30.6
Riverside 37.7
San Diego 38.6
Santa Monica 10.3
South Gate 16.6
Source: Adapted from Dziegielewski et al. (1990). Seasonal Components. Table IV-3.
Precipitation. Increases (decrease) in the amount of local rainfall will not only usually
increase (decrease) the available supply of water, but will also decrease (increase) the demand
for water for the purpose of irrigation and landscaping. In addition, an increase (decrease) in
amount of local rainfall will likely decrease (increase) the need for the use of air-conditioning in
the summer season, and thus result in the change in water use by cooling towers.13 Kiefer et al.
(1995, 1996, and 1998) have estimated significant relationships between nonresidential water use
and the monthly amount of precipitation, as well as departures from long-term normal
precipitation.
Technology
The presence, diversity, and efficiency of various end uses in an establishment affect the
quantity of water used. For example, a restaurant using icemaking machines is likely to consume
more water than a restaurant without an icemaker, other things being unchanged. As another
example, office buildings are likely to consume less water than restaurants for a given number of
employees and floor size. Water use in office buildings is likely concentrated in domestic
fixtures and landscaping while use in restaurants may be primarily in icemakers, dishwashing,
food preparation, and others, as well as domestic fixtures and possibly landscaping. Hospital or
hotel establishments with the presence of in-house laundry facilities, landscaping, and variations
in capacities of cooling and heating will consume substantial amount of water compared to
establishments without these facilities.
The efficiency level of technology a Cl establishment has adopted could significantly
affect the demand for water. For example, a water-cooled icemaker uses a larger amount of
water than an air-cooled icemaker. Likewise, a once-through cooling system consumes more
water than a more sophisticated recirculating system.
Recent increased interest in conservation has brought about significant growth in the
development of water conservation technology. There is considerable evidence proving that an
establishment adopting water-efficient technology can significantly reduce the demand for water.
Conservation technologies, which include water-efficient end use equipment such as ultra
low flow toilets, and water-efficient practices such as Xeriscaping, have been relatively effective
as tools of conservation in the residential sector. Conserving technologies reduce overall
demand for water by minimizing volumes of water per usage (low-flow fixtures), or maximizing
cycle per usage (recycling). Table 2.17 lists possible water-efficient technologies for selected end
uses.
13 This situation applies to where a facility automates indoor temperature control that adjusts to daily temperature.
Table 2.17 Examples of CI conservation technologies
End uses
1. Domestic Use
A. Kitchen
1) Faucet
2) Distilled/drinking water
3) Dishwasher
4) Ice machines
5) Garbage disposers
6) Food preparation
B. Bathroom
1) Faucet
2) Toilet
3) Urinal
4) Bathtub
5) Shower
2. Recirculating Cooling
A. Cooling towers
B. Evaporative coolers
C. Boilers
3. Once-Through Cooling
A. Air conditioners
B. Air compressors
C. Hydraulic equipment
D. Degreasers
E. Rectifiers
Conservation Technologies
Faucet aerators
Recycle for garbage disposers, automatic shutoffs
Air-cooled machine, multiple-pass cooling, recirculating for
landscaping
Garbage strainers
Faucet aerators, automatic shut-off, infrared faucet, self-
closing faucets,
Ultra-low-flow toilets
Ultra-low-flow and "waterless" urinals
Low flow showerheads
Conductivity control, total dissolved solids (TDS)
Sidestream filtration, Maintenance of make-up valves
Recycling, incorporating sulfuric acid to reduce carbonate
scale
Ozone disinfection
Reclaimed water for water makeup
Recirculating pumps
Eliminate excessive bleed-off
Eliminate excessive blowdown, eliminate mixing valve water
Ion exchange
Air-cooled equipment, reduce flow rate
Connect to recirculating cooling system
Recycle cooling water
F. Vacuum pumps Convert to mechanical vacuum pumps
4. Process/Rinse
A. Photographic processing Automatic shut-off, eliminate "tempering" flows; water-
efficient equipment timers; conductivity control
B. Product water rinses Countercurrent rinsing, solenoid valves
C. Ingredient/material rinses Squeegees
D. Conveyance Flow-metering, auto-control valves
E. Purification equipment regeneration
F. Rinse baths Waterless equipment
G. Lubrication systems Sequential rinsing, recycling
5. Sanitation
A. Facility cleaning
B. Sterilizers/autoclaves
C. Equipment washing
Dry extraction carpet cleaning system
Automatic shut-off valves, pressure-reducing valves
Flow-metering, control valves
Air pressure host
Wastewater reclamation
(continued)
Table 2.17 (Continued)
End uses Conservation Technologies
D. Vehicle washing Use a nose nozzle
E. Dust control Re-use water from another process, e.g. ice machine discharge
F. Container washing
6. Laundry
Washing machine Horizontal-axis washing machine
Continuous-batch washers
Rinse water reclamation; wash water reclamation
Computer-automated control system
7. Irrigation
A. Spraying Moisture sensors and timers, rainfall sensors
B. Planting Xeriscaping, evapotranspiration, drought resistant shrubs
Recycle, reclamation
C. Decorative water feature
8. Leaks Leak detection system
9. Other/Miscellaneous Recycling
10. Unaccounted Use Monitoring meters
Source: Derived from Black & Veatch (1989). Best Available Technologies" Program: Phase 1 Report:
Industrial/Commercial Water Uses Conservation Opportunities. Phoenix, AZ.
LITERATURE SUMMARY
From the review of literature, one may derive a number of conclusions regarding the state
of CI water use and consumption. Some particular conclusions are offered below in preparation
for the in-depth analyses of Chapters 4 and 5 of this report.
CI Classifications and Data
The systems of classifying CI customers by water utilities are generally inadequate for
comparing water use for individual categories between cities. Only a few categories seem to be
adequately defined and comparable. These may include such categories as hotels/motels,
schools, restaurants, laundromats, car washes, and other easily recognizable types of businesses.
Thus, a significant recommendation of this review is to analyze and develop a standard CI
customer classification scheme, which will facilitate both demand planning and conservation
evaluation activities. The definition and consistency of CI categories is explored in the next
chapter for the utilities that participated in this study.
Availability of Water Usage Benchmarks
Because the system of classifying CI customers is not standardized, it is impossible to
develop benchmarks from billing records of water utilities for comparing water usage rates for
the same categories of CI customers. Other obstacles to developing meaningful benchmarks of
CI water use from billing records include:
The distribution of CI customers by size is usually skewed with a small number of
customers accounting for a majority of water use. This characteristic makes the average
use of water per customer within a CI category very sensitive to the degree to which
water use is concentrated within top accounts. Under this circumstance, the mean use per
customer is not a reliable measure of water use. It can vary in time as the concentration
of usage shifts and it can differ between cities with different degrees of concentration of
use.
The most appropriate variables for normalizing water use depend on the type of CI
category and often cannot be easily measured. The only normalizing variable that is
available to all utilities is the number of active accounts within a CI category. Another
variable, the number of employees can be obtained from government statistics but it is
usually only available in an aggregate form for only a few points in time. The
employment data for individual establishments are usually confidential and imprecise.
Other measures of size such as the number of meals served in a restaurant or the square
footage of a retail store cannot be obtained from secondary sources and require on-site
data acquisition.
In some categories of CI customers demand is concentrated in one or two end uses, and
in order to develop benchmarks, both the size of the establishment and characteristics of
the main end uses have to be known. Establishments with significant water use for
landscape irrigation or cooling fall into this category. Irrigation use is separated for CI
customers in a few utilities and embedded in CI use in others.
CHAPTER 3
SELECTION OF CI CUSTOMER CATEGORIES FOR FIELD STUDIES
AND MODELING
INTRODUCTION
Because of the size and scope of the CI customer sector, as described in the Chapter 2,
this research project was designed to focus in on a selected group of five CI categories. Water
use in the five selected sectors would then be studied in detail through the direct measurement
field studies and end use modeling components of the research project. Because of this required
focus, it was desirable to select five CI customer categories that are present in most service areas
and comprise a significant component of total CI demand. This chapter describes the process by
which these five CI categories were selected.
To assist with this process, each of the five participating cities provided historic water
billing data from their population of CI customers as well as any classification system in use.
These billing data were summarized (some of the results were presented in Chapter 2) and then
used to develop lists of CI categories and their relative importance in the five selected service
areas. The study team then made an initial selection of candidate CI categories and made
recommendations to the Project Advisory Committee (PAC) for the final selection. The PAC
determined the final list of five categories during a conference call in May 1998.
PARTICIPATING STUDY SITES AND AGENCIES
The five participating study sites and two contributing agencies were selected based upon
a willingness to participate in this tailored collaborative study and to contribute to the cost of the
research. Because this study was an extension of the Residential End Uses of Water Study
(Mayer, et. al. 1999) many of the participating study sites in the two studies were the same. The
five participating study sites were:
1. Irvine Ranch Water District, California
2. San Diego Water Department, California
3. Santa Monica Water Department, California
4. Phoenix Water Department, Arizona
5. Los Angeles Department of Water and Power, California
Participation of two of these utilities was sponsored in part by the San Diego County
Water Authority and Metropolitan Water District of Southern California, who also participated
in the Residential End Uses of Water Study
All empirical research including the direct measurement field studies and end use
modeling was conducted in cooperation with the agencies noted above.
CI BILLING DATA AND RELATED INFORMATION
To assist with the selection of candidate CI categories the participating utilities provided
24 months of raw billing data and customer classification data for their non-residential
customers. Requested information included: account number, customer class identifier,
customer name, customer contact, contact phone number, customer address, customer
description fields, meter number, meter size, billing dates, number of days in each billing period,
and water consumption in billing periods. All five utilities cooperated with this request and
submitted raw data for analysis.
Data Analysis and Results
In order to verify the information on the distribution of CI use among significant
categories, the study team performed an analysis of billing records from the five utilities that
were partners in this research project.
A complete set of billing records for a period of one year was analyzed including:
Analysis of CI categories and use in each city independently
Comparison ofCI categories and use across the five sites
Development of data that can be used to rank the categories according to conservation
potential
Three general criteria were used to screen CI categories for each city independently.
First, all categories used by each utility were ranked according to scaled average daily use per
customer, excluding any industrial categories. For a given CI category, this construct is
calculated as the average daily use per customer (in gallons per day) multiplied by the fraction of
total annual CI use accounted for by customers in the given CI category. The scaled average
daily use per account balances the rate of water used by customers in a category with the relative
prominence of that category within the total use of the CI sector.
Next, under-identified CI categories were excluded. Under-identified CI
categories are those which clearly contained heterogeneous groups of customers such as "general
commercial" categories and other catch-all designations and vague SIC groupings such as
"miscellaneous retail". Lastly, only those categories that accounted for 1 percent or more of total
CI water use were retained. This analysis resulted in the selection of eleven categories to be
considered for final consideration by the PAC. These categories included:
1. Irrigation accounts
2. Schools and colleges
3. Hotels
4. Laundries and laundromats
5. Office buildings
6. Hospitals and medical offices
7. Restaurants
8. Food stores
9. Auto shops
10. Membership organizations
11. Car washes
The water use associated with these categories were then compared across all five study
sites. Fortunately, the eleven categories listed above were common to two or more of the five
participating utilities. However the fact that utilities classify their CI users differently made
comparison across study sites difficult and to some degree qualitative. Although the eleven
categories are not used and defined similarly in all study sites, they provided an adequate basis
for selecting significant categories of CI users for this study.
A comparison of the water use statistics from these eleven categories is presented in
Table 3.1. Possible selection criteria included the scaled use measure described above and
statistics related to water use variability and seasonality of use. The data for each category were
combined across all cities to further characterize water use among and within the eleven
categories. The rows of Table 3.1 titled "Logical-weighted average/total" report values for each
category relative to CI use across all of the cities that had the common category, and, in essence,
treat the categories as belonging to one "virtual" CI sector in one "virtual" city.
All eleven categories make significant contributions to total CI use, and in total likely
account for more than 50 percent of CI use (particularly since the types of customer probably
comprise a large portion of under-identified categories). These categories also show high
variability of use among individual establishments and a significant seasonal component of
annual use. These characteristics make them good candidates for targeting CI conservation
programs.14
Table 3.2 shows the eleven categories sorted by their likely potential for water
conservation. The categories are ranked in joint consideration of scaled average use, percent
seasonal use, and variation of use among category establishments. If the total category use was
used as a sole criterion for ranking, then office buildings and restaurants would be among the top
four categories. Their lower rank on stems from the fact that these categories consist of a
relatively large number of individual establishments and do not exhibit a high seasonality of use.
The top ranked category of urban irrigation represents water use by irrigation accounts, which
are becoming increasingly common among utilities.
14 It is important to note that for individual facilities, other data must be considered to analyze the amount of
variability in water use between facilities that is due to differences in water using efficiencies.
Table 3.1 Analysis of selected common CI categories in five study sites
Applicable Customer category description Number of Total Percent of Percent Average Std. Dev. of Sc
City customers 1997 total CI of total annual daily avg. annual avg
Use (ccf) customers CI use use (gpdc) daily use use
Irvine
S. Monica
IRRIGATION
Irrigation
Landscape commercial
Landscape municipal
Landscape public schools
Logical-weighted average/total
6471968
21970
87539
1750
6583227
30.0%
1.4%
5.7%
0.1%
28.5%
5517.
1023.
2086.
1793.
2595.
22120.1
1849.6
4589.3
303.9
22668.8
Phoenix
Los Angeles
San Diego
S. Monica
Los Angeles
Phoenix
San Diego
S. Monica
Phoenix
Los Angeles
San Diego
S. Monica
SCHOOLS AND COLLEGES
Schools
Schools
Colleges and universities
Educational services
Educational services
Schools
Logical-weighted average/total
HOTELS
Hotels and motels
Hotel, motel
Hotels and other lodging places
Hotel w/dining facility
Hotels w/o dining facilities
Logical-weighted average/total
LAUNDRIES AND LAUNDROMATS
Laundries; commercial
Laundry; self service
Coin-operated laundries
Laundry, cleaning and garment services
Coin-operated laundries and cleaning
Laundromats
Commercial laundry
Logical-weighted average/total
2066256
2494943
876642
148975
1010304
70309
6667429
6700.1
3590.5
7100.9
1344.9
3974.0
1549.3
2116.6
2086331
1856730
265255
121525
64274
4394115
5324.
1257.
6629.
2452.
5726.
7112.
81845
308501
1923703
232408
399514
29656
1627
2977254
5241.4
7709.9
7610.6
4451.2
5765.7
2762.5
1667.1
3289.8
10434.9
4454.9
10527.9
3131.4
20054.1
2846.3
25683.5
7781
27854
19273
12928
10268
38476
6514.2
7276.8
6353.0
25987.6
4587.3
3484.3
2073.5
29130.5
Table 3
(Continued)
Applicable Customer category description Number of Total Percent of Percent Average Std. Dcv. of Sc
City customers 1997 total CI of total annual daily avg. annual avg
Use (ccf) customers CI use use (gpdc) daily use use
OFFICE BUILDINGS
Irvine
Office
Office
Phoenix
Los Angeles
Office/bank building (non dining,
medical, nursing home)
Non residential building operators
Logical-weighted average/total
6277
1340990
2232526
3983665
7559933
24.4%
0.1%
14.1%
6.2%
0.0%
18.9%
10.2%
1796.2
1879.9
3639.7
2341.2
1204.4
2844.9
5550.2
4298.8
7576.3
HOSPITALS AND MEDICAL OFFICES
Medical lab
Los Angeles
S. Monica
San Diego
Health
services
Hospitals
Medical office facilities
332347
1458261
106638
6.9%
49541
Health and allied services, nec
Health
services
Hospitals
Medical and dental labs
veterinary
services
Logical-weighted average/total
2066
2245
253048
306436
33158
12847
2554521
0.2%
0.3%
0.2%
4.2%
0.1%
0.0%
3.9%
18407.7
2124.0
10406.4
1586.3
1150.2
1258.7
17444.0
1544.3
642.1
1236.5
77908.8
4526.5
17370.8
2127.5
1319.1
2390.7
54847.8
2651.0
1083.7
97059.1
RESTAURANTS
Phoenix
Los Angel
San Diego
Monica
restaurant, bakery
eating and drinking places
drinking places
eating and drinking places
retail bakeries
restaurant & bars w/ food
bakery or bakery w/ deli
bars w/o dining facilities
Logical-weighted average/total
2039912
3052052
17.6%
42457
14.5%
0.2%
1303088
0.0%
14.1%
217727
6660907
0.2%
0.4%
11.2%
3670.3
1437.8
654.2
1766.2
360.2
2166.0
901.2
822.0
905.5
4726.7
2111.4
589.8
2707.5
100.6
3720.5
149.4
405.6
6965.4
FOOD STORES
Irvine
Los Angeles
San Diego
S. Monica
stores
food stores
grocery store
Convenience
356538
1222326
243385
& liquor stores w/ deli
40316
7387
8.0%
5.8%
4683.7
1269.6
2159.2
437.1
946.1
10830.3
2398.1
3941.2
585.6
1463.7
Irvine
Table 3.1 (Continued)
Applicable Customer category description Number of Total Percent of Percent Average Std. Dev. of Sc
City customers 1997 total CI of total annual daily avg. annual avg.
Use (ccf) customers CI use use (gpdc)* daily use use (
Logical-weighted average/total 2565 1869952 5.2% 2.9% 729.0 11877.1
AUTO SHOPS
Phoenix service station; auto repair 634 278114 2.8% 0.9% 899.0 2442.4
Los Angeles automotive dealers and service stations 2203 657760 8.9% 3.1% 611.9 1123.9
San Diego automotive dealers and service stations 4 246 0.0% 0.0% 126.0 245.2
automotive repair shops 1304 346333 7.4% 1.4% 544.3 1110.1
automotive repair shops, nec 1 30 0.0% 0.0% 61.5
automotive services, except repair 89 152213 0.5% 0.6% 3504.9 4123.8
automotive services, nec 4 3882 0.0% 0.0% 1988.9 1951.9
S. Monica auto repair, sales, and service stations 194 47067 9.1% 3.0% 497.2 717.3
Logical-weighted average/total 4433 1485645 6.7% 2.0% 686.8 5463.6
MEMBERSHIP ORGANIZATIONS
San Diego religious organizations 507 221715 2.9% 0.9% 1008.8 1974.1
Irvine church rate 826 627517 3.6% 2.0% 1559.8 3064.4
S. Monica Membership organizations 44 16420 2.1% 1.1% 764.8 1734.1
Logical-weighted average/total 1377 865652 5.6% 1.9% 628.7 4036.7
CAR WASH
Phoenix car wash 85 247814 0.4% 0.8% 5974.7 5784.3
S. Monica car wash 4 21957 0.2% 1.4% 126962.2 7467.4
Logical-weighted average/total 89 269771 0.4% 0.8% 3031.1 9445.7
Gallons per day per customer
Table 3.2 Characteristics of significant CI categories in five participating agencies
Customer category Average Coefficient Percent of Percent of Percent Scaled
description annual of variation total CI seasonal average
daily use in daily use CI use customers use daily use
(gpdc)* (gpdc)t %) (%) (%) (gpdc)**
Urban irrigation 2,596 8.73 28.48% 30.22% 86.90% 739.0
Schools and colleges 2,117 12.13 8.84% 4.79% 57.99% 187.0
Hotels and motels 7,113 5.41 5.82% 1.92% 23.07% 414.0
Laundries and
laundromats 3,290 8.85 3.95% 1.38% 13.35% 130.0
Office buildings 1,204 6.29 10.19% 11.67% 29.04% 123.0
Hospitals and medical
offices 1,236 78.50 3.90% 4.19% 23.16% 48.0
Restaurants
Restaurants 906 7.69 8.83% 11.18% 16.13% 80.0
Food stores
Food stores 729 16.29 2.86% 5.20% 19.37% 21.0
Auto shops 687 7.96 1.97% 6.74% 27.16% 14.0
Membership
organizations 629 6.42 1.95% 5.60% 46.18% 12.0
Car washes
Car washes3,031 3.12 0.82% 0.36% 14.22% 25.0
* gpdc: gallons per day per customer
t Coefficient of variation in daily use: The ratio of standard deviation of daily use to average of daily use.
$ Percent of CI customers pertains to Cl customers in agencies that have respective category only.
Percent seasonal use = [(total annual use 12 x minimum month use ] / total annual use
** Scaled average daily use = average annual daily use in category x percent of total CI use attributed to the
category.
An alternative ranking priority may be developed to emphasize the share of indoor uses.
Table 3.3 shows a comparison of the eleven categories above based on Scaled Inside Use Factor
per customer. For a given CI category, this construct is derived as annual non-seasonal use per
customer (in gallons per year) multiplied by the fraction of total annual non-seasonal CI use
accounted for by all customers in the given CI category. The scaled inside use factor variable
balances the total amount of non-seasonal use by the category with the relative prominence of
that category within the total non-seasonal use of the CI sector. By ranking according to the
scaled inside use factor, the top five categories become: hotels, laundries, office buildings,
schools, and restaurants, respectively.
Table 3.3 Alternative rankings of CI
Initial Table 2.5 ranking
Urban Irrigation
Schools and colleges
Hotels/motels
Laundries and laundromats
Office
Hospital and medical office
Restaurants
Food stores
Auto shops
Membership organizations
Car wash
customers in five participating sites
Scaled inside use factor ranking
Hotels
Laundries and laundromats
Office buildings
Schools and colleges
Restaurants
Hospital and medical office
Car wash
Food stores
Auto shops
Membership organizations
Irrigation
FINAL SELECTION OF CI CATEGORIES FOR IN-DEPTH ANALYSIS
The list of eleven CI categories and different ranking schemes and recommendations
were presented to the PAC and the final five categories were selected during a conference call in
May 1998. Of the eleven categories, the irrigation, car wash, and laundry categories are
comprised of very specific types of end uses directed at providing specific products or water
services. Although the individual customers in these categories display considerable variance in
water use, the PAC decided that a study of conservation opportunities for these categories could
be narrowly focused and perhaps better served by independent studies.
The auto shops and membership organization categories share similar qualities in that
they are comprised mainly of specific purposes (i.e., washing and sanitary uses, respectively). In
consultation with the PAC, it was also determined that the scope and intensity of water services
in hospital and other health-related settings blurred the distinction between CI and "light
industrial" customers. Therefore, the following five categories were selected for further analysis:
Ranking
priority
1
2
3
4
5
6
7
8
9
10
11
Schools
Hotel/motels
Office buildings
Restaurants
Food stores
However, these five categories were not selected solely because of the elimination of
other categories. To the contrary, these categories represent CI customer types that are common
to most cities, and which present a diversity of end uses and therefore a good basis for examining
conservation.
Potential Determinants of Demand in Selected CI Categories
Table 3.4 presents a list of variables that may be considered a partial set of determinants
or indicators of water demand in the final selected CI categories. As shown in the table, one may
identify variables that affect water use of CI customers in general and variables that would be
unique to specific CI customer groups. For example, the number of employees and
establishment size could be used to explain or normalize water use for each CI category.
Furthermore, the presence of restrooms and other sanitary fixtures indicates their effectiveness
would be pertinent to most CI establishments.
Other variables listed in Table 3.4 reflect indicators of (1) the specific types of goods and
services that are supplied by the five CI categories, (2) the types of patrons that demand these
goods and services, and (3) the types of end uses that are used at these establishments. For
example, water use at a restaurant may be defined as a function of the number of visiting
customers, number of meals served, and the presence and efficiency levels of kitchen fixtures.
The other categories have a more complex set of water demand parameters because of the
diversity of services and/or features that may be found on the premises. For example, a hotel
may have a restaurant on site as well as laundry and spa facilities, all of which will have specific
associated water end uses. Similarly, the modem supermarket may offer a variety of products
and services distinct from the products found among its aisles. These services, such as hair
salons and photo processing, add a different set of end uses than would have otherwise been
found and will contribute to a different pattern and quantity of water use at the establishment.
Finally, the water use associated with office buildings can represent the use of various
establishments and business types typically found on the first floor of the complex. The range of
water-using activities in first floor businesses might be expected to differ from the activities
present in office space typically found in higher floors.
Table 3.4 Potential explanatory variables and demand indicators for
selected CI categories
Common variables Restaurants Hotel/motel Supermarkets Schools
* No. of
employees
* Square footage
* Price of water/
wastewater
* Air temperature
* Precipitation
* Sales $
* No. of
restrooms and
sanitary fixtures
by type
* Presence of
irrigation and
type of system
* Irrigable
landscape area
* Type of cooling
system
* No. of meals
served
* S
eating capacity
* Patron counts
* Menu/types of
meals
* Operating hours
* Type of
restaurant
* Types of kitchen
operations
* No. and types of
water using
kitchen fixtures
* Average meal
price
No. of
occupants
Occupancy rate
No. of rooms
Presence of
restaurant and
lounge
See restaurant
variables
Presence of:
Kitchen
Laundry
Swimming pool
Spa/sauna
Clubhouse/gym
Type of
icemakers
Type, number,
and volume of
water cooling
* Sales
* Presence/types of
services
Deli
Meat shop
Photo
Hair salon
Bakery
* No. of aisles
* Presence of
public restroom
facilities
* Mist sprayers on
vegetables
* Hours of
operation
No. of pupils
No. of sporting
events
Gym seating
capacity/
facilities
No. of showers
Presence of
swimming pool
Size and types of
play fields
Cafeteria/kitchen
See restaurant
variables
Hours occupied
CHAPTER 4
DIRECT MEASUREMENT FIELD STUDIES
INTRODUCTION
Projecting water savings for commercial and institutional customers requires detailed
information on how much water is used by the group for various specific purposes. The current
state of information on the end uses of water in CI customers remains vague. One reason for this
is the highly diverse nature of CI users, which include the wide array of all commercial and
institutional establishments. The typical approach has been to perform a water audit of the site, in
which a trained individual visits the site and collects information on all water using fixtures and
appliances. The auditor then tries to estimate the total daily use for each piece of equipment,
water appliance, or plumbing fixture based on interviews with the staff and measurements taken
on-site. If done properly, this method can provide good results, but auditing is time consuming,
labor intensive, and subject to errors if the information on which the estimates are based proves
faulty.
As part of the AWWARF Commercial and Institutional End Uses of Water Study
(CIEUWS), a concerted effort was made to determine if the data logging and flow trace analysis
technique used in the Residential End Uses of Water Study (REUWS) could be applied to CI
customers in order to provide a new tool for use in the audit process. Using this technique on a
single-family residential account it is possible to disaggregate demand into component end uses
(toilets, showers, faucets, clothes washers, etc.). This is done by collecting a continuous flow
trace from the water meter using a data logger, and then identifying each fixture and appliance
use with signal processing software. Clearly, the same degree of disaggregation was not
anticipated for any but the smallest CI customers since CI use patterns are much more complex
than those of residential customers. At the minimum, it was thought that flow trace analysis
could provide better estimates of water use by each fixture than monthly or bi-monthly billing
data. The researchers anticipated disaggregating at least indoor, outdoor and continuous uses.
This, in itself, is a major advantage for the auditor since it reduces the amount of variability that
must be accounted for and makes it possible to spot anomalous uses for further analysis.
The objective of the direct measurement field studies portion of the CIEUWS was to
combine information obtained from three sources: (1) surveys, (2) water billing data, and (3)
flow traces to develop more accurate estimates of where water is put to use in the five selected
CI categories. The small size of the sample in the direct measurement field study portion of the
CIEUWS limits the ability to generalize from these results to large populations of CI customers.
However, the results from the field studies, especially when compared to the larger audit group,
provide a good indication of the expected range of demands within each end use category and
provide detail about variety of demands which may be found in these five categories of CI
customers.
PROCEDURE
Because the primary focus of the CIEUWS was on the modeling effort, the direct
measurement field studies were limited to collecting data from 25 commercial and institutional
sites spread across the five participating utilities. Each utility was given a set of criteria, and
asked to select five customers from their population of CI customers one from each of the five
selected CI categories: hotels, high schools, restaurants, office buildings, and supermarkets.
Once these sites were selected, technicians visited each of the sites, obtained historic billing
records from the utility, conducted an on-site audit, and installed a data logger to collect a flow
trace from the water meter. During the initial visit, on-site personnel were queried and the site
was inspected for prospective sub-meter locations.
The data were carefully analyzed and a detailed end use report for each of the 25 study
participants was developed. The final research report contains only summary results from the
direct measurement field studies, but to preserve the privacy of study participants copies of the
individual site water use reports may only be available through AWWARF.
Selection of Study Sites by Utilities
In February of 1998 a set of selection criteria were sent out to the five participating
utilities to assist them with selecting study sites for each of the five categories of CI customers
included in this study. Each utility was requested to locate one customer from each category
who would be willing to participate in the study. The small size of the sample being studied
made it impossible to make it representative of the entire group so participants were selected
based on how well they matched the criteria and their willingness to participate. In several cases
it proved impossible to find willing participants who also met the selection criteria. In those
cases the willing participants were used irrespective of the criteria.15 The practical effect of this
was that many of the sites were much larger than originally planned, which led to a reduction in
the resolution of some of the flow trace data.
The process of locating willing participants proved more difficult than anticipated and the
process continued through the summer of 1998 even as the field studies were being conducted.
With the exception of the high schools, sites had been selected for all categories by September.
Work on the selection of high schools was not complete until December 1998 and only four
schools were found which could participate.
Site Visits
Most of the 24 CI study sites were visited once during 1988. Each site visit included
installation of a data logger on the site's water meterss, a meeting with the site superintendent or
building manager to discuss water use at the site, and an detailed inventory of all water using
fixtures and appliances on-site. It was found that water use at many of the smaller sites could be
disaggregated from the flow trace obtained from the main meter and the information obtained
during the site visit, so sub-metering was not required.
The original project work plan called for use of the main flow trace data to disaggregate
water use at the smaller sites, and to attempt to supplement the main traces with installation of
sub-meters at the larger sites where this proved practical. Experience in the field quickly
showed that installing any device into the plumbing systems of large CI customers is difficult
and expensive in most cases. Water pipes where sub-meters could be installed were typically
inaccessible, or the pipe network was designed in such a way as to require numerous sub-meters.
Facility managers were generally cooperative with the auditors, but often made it clear that they
had no interest in participating in any monitoring program that might temporarily shut off the
water supply.
15 The only case in which it proved impossible to obtain a study site was for the high school in San Diego. In that
case the City was unable to locate a school which had compatible meters and was willing to participate in the study.
Many of the schools had old mechanical water meters that are incompatible with the data logger equipment used for
the study.
Site Visit Techniques and Procedures
During the data collection portion of the direct measurement field studies, each CI
customer was first visited in order to install the data logger on their water meters) and record the
flow through the water meter for a period of time. With data logging technology now available,
precise data on where water is used can be collected in a simple non-intrusive manner, directly
from the water meter (DeOreo, Heaney, and Mayer 1996; Mayer and DeOreo 1995; Mayer 1995;
Dziegielewski et al, 1993). Each logger is fitted with a magnetic sensor that is strapped to the
water meter of each study site. As water is used at the site, it flows through the water meter
causing the internal magnets of the water meter to spin. The sensor picks up each magnetic pulse
as water moves through the meter and the logger counts the number of pulses detected and stores
the total every 10 seconds. The logger has sufficient internal memory and battery life to record
for more than 14 days at the 10-second interval.
Using the physical characteristics of each specific brand and model of water meter, the
magnetic pulse data is transformed into an average flow rate for each 10-second interval. This
flow trace is precise enough to detect the individual flow signatures of water using equipment
and appliances (such as clothes washers and cooling towers), and plumbing fixtures in the
building, and that of the irrigation system. Using a custom signal processing software package
called Trace Wizard, each flow trace was disaggregated into the identifiable component end
uses.
The loggers used in this study were the Meter-Master 100EL manufactured by the F.S.
Brainard Company of Burlington, NJ. The Meter-Master 100EL logger, shown in Figure 4.1,
offered the essential combination of data storage capacity, water resistance, and ease of use.
The basic assumption behind the data logging system in that the water meter is accurately
recording flow volume. The logger is not truly measuring flows, but rather recording the
movement of the magnets that link the meter to the register and spin as water flows through the
meter. The logger records the number of magnetic pulses counted in a 10-second interval and
once the data is downloaded, the data logger control program automatically converts the pulse
count into flow using the exact specifications of each water meter. Most of the water meters
used in this study provided resolution of between 20 and 80 magnetic pulses per gallon. When
the logger is downloaded, the logged volume is compared to meter readings taken at the time of
installation and removal to check the accuracy of the flow trace.
Each logger was initialized to local time and synchronized to the watch of the analyst,
who was the member of the research team who was responsible for performing the site audit.
The synchronization process allowed the analyst to record the precise time of events noted
during the visit which were then compared to data obtained from the flow trace.
During the field trip all of the loggers were installed on the first day of the visit. Once
this was complete a second visit was made to each site in order to interview the building
manager and catalogue the water using fixtures present in the building. The technician
completed a detailed site survey during this visit and noted the presence of water using
equipment, appliances and fixtures, building data, occupancy, irrigation systems, and other water
using devices. During each site various appliances were operated and the time was noted down
so that the flow trace could later be inspected for a signature.
Figure 4.1 Data logger used in this study
Possible locations for sub-meters were also sought during the site visits. There were
many obstacles to sub-metering encountered. Finding accurate plumbing plans or individuals
that knew about the plumbing was a major problem. Even where plans were available, it was
rarely possible to find access points that would allow meters to be installed in a practical manner.
Even where access was available, in many cases the actual lay out of the pipes made it
impossible to locate a meter where the desired water use would be registered. For example, a
logical place to sub-meter would be the laundry rooms in the hotels. These were frequently fed
from several directions with piping built into the walls. In none of the hotels was it possible to
locate a place where a single meter would register all water use in the laundry, and could be
installed without disrupting the operations of the facility to an unacceptable degree.
There were only three sites found where it made sense to employ sub-meters. The two La
Quinta Inns (in Phoenix and Irvine) had plumbing lines that were ideally suited for sub-metering.
Tiers of rooms are fed from separate /4" hot and cold supply lines which come off of distribution
lines in the attic. These were easily accessed and required only small meters. Each pair of meters
supplied a group of from 4 to 8 rooms. The University High School in Irvine also had an ideal
system. There, each set of bathrooms was supplied from a single cold water line that entered
through a amply sized janitor room. A single 1" meter could then be used to monitor all water
use for the bathroom. With the exception of these sites, while it would have been theoretically
possible to install more sub-meters, it would have required much more time and money than was
available for this portion of the project.
Data from the sub-meters were found to give highly detailed information about the end
uses at each specific site, and showed that when sub-metering is possible, more detailed end use
data can be obtained using data loggers and flow trace analysis techniques. The project
experience shows that sub-metering would make most sense in a detailed study of a few sites
over a longer period of time. Such a study could justify the cost of the installation.
Once all the site visits in a particular city were completed, the technician returned and
removed the data logger from the water meter, typically after a recording period of 5 days. The
data from each logger were downloaded to a laptop computer, the volume recorded by the meter
and logger were compared and verified, and the data were stored for analysis.
Data Analysis
A complete data set from each site consisted of billing and customer information supplied
by the utilities, site survey data obtained during the sites visits, and flow trace data obtained from
the water meters) during the site visitss. These data were analyzed and the results combined to
create as detailed a picture as possible of the water use at the site over the year of record.
The fundamental goal of the analysis was to derive good estimates of annual water use at
each site for three large categories: indoor use, outdoor use and continuous uses. Indoor use
included all domestic sanitary, process, mechanical equipment, cleaning uses, and periodic leaks;
outdoor use included irrigation, pool filling, driveway/patio washing; continuous use included
leakage and cooling water demand that never ceased during the data logging period. It was also
desired, whenever possible, to disaggregate the indoor uses into individual end-use categories
such as toilet/urinal flushing, sinks, showers etc.
Use ofBilling Data
Billing data were used to determine the annual, monthly and seasonal water use at each
site. Some sites had irrigation meters which allowed accurate disaggregation of indoor/outdoor
use to be made directly from the billing data. In most cases, however, the billing data included
water used for indoor and outdoor purposes and data from the recorded flow traces were used in
conjunction with billing data to estimate the indoor/outdoor use split. An example of a typical
monthly analysis performed for the direct measurement field studies can be seen in Figure 4.2,
which shows the total monthly water use for a small office building. The increase in water use
during the irrigation season at this site is evident in the March through November data.
60 -- -------- -
50
S40
u 30
0 20
10
1997-98
Figure 4.2 Example of monthly use pattern in an office building with irrigation
Flow Trace Analysis
Overview. It is often inadequate to determine important use parameters such as indoor
and outdoor use or cooling demand using only monthly or bi-monthly billing data. In many
locations both irrigation and cooling water use occur throughout the year making estimates of
these end-use categories difficult when based on periodic or seasonal use data.
A goal of the CIEUWS was to use a combination of billing data and information from
water audits to derive end-use estimates using statistical approaches pioneered in the electrical
industry such as conditional demand analysis. As a secondary matter, the study also sought to
use flow trace data, obtained from a customer's water meter, to assist with the disaggregation
process. It was shown in the Residential End Uses of Water Study (REUWS) that in single
family homes flow trace data could be disaggregated down to the level of individual toilet
flushes, showers, irrigation events and clothes washer cycles.
There are two main reasons for the success of this technique in residential customers: the
water meters provide high resolution (up to 120 pulses per gallon), and in a single-family home
the vast majority of use events are discreet rather than occurring simultaneously with several
other uses. This allows the data logger to pick up very small flows and insures that the flow
trace will have sufficient resolution to show the true flow patterns of the individual fixtures and
appliances in the home. The second essential feature that allows the successful disaggregation of
water use is that not too many events occur at the same time. In single-family homes the
majority of all events occurred by themselves as single flushes, washing machine cycles etc.
Many additional events were mixed with small faucet use or toilet flushes that could be
identified as separate events by the software used for the analysis. A small minority of events
were clearly a mixture of several events which required that the analyst make a judgment call on
how best to disaggregate the event, or whether to place them into the "unknown" or
"miscellaneous" category. Generally, in single-family residences the disaggregation from flow
trace data occurred with a high degree of confidence about the accuracy of the results.
As the water meters get larger and more simultaneous uses of water are occurring it
becomes more difficult to use a single flow trace to disaggregate water use behind the meter.
However, even in the largest meters it is possible to accurately identify indoor, outdoor and
continuous uses. In smaller customers, with correspondingly smaller water meters, it was also
possible to identify many individual indoor water uses.
Analysis Process. The analysis of the recorded flow traces aimed to generally
characterize the water use at the facility and included the determination of daily, hourly, and
peak instantaneous water use. Where possible peak indoor and peak outdoor use was separated.
The daily use during the logging period was compared to the average daily use obtained from the
historic billing data in order to determine the degree to which the logging period was typical of
the recent annual use. Then the flow trace was examined in detail in conjunction with the
information obtained from the site survey in order to attempt to identify specific water uses
occurring at the site. In some study sites a high degree to detail was available from the flow
trace with little or no judgement or interpretation required. In other cases it was impossible to
use the flow trace to disaggregate water use beyond the three major categories. In some cases
end uses could be seen during a portion of each day, which were then used to estimate end uses
during the remainder of the day when use was masked by large continuous uses.
At three sites, sub-meters were installed to separate water use in a portion of the building.
Flow trace data were recorded from these sub-meters which yielded information on end uses
which could then be extrapolated to the rest of the facility. Where multiple meters were present,
the simultaneous flow traces were combined to create a virtual single flow trace for the facility.
This occurred in the hotels, where separate meters were installed on the hot and cold water lines.
Each logger gave either the hot water or cold water portion of the event (such as a shower or
bath). The entire event required the combination of both flow traces.
Figure 4.3 shows a two-hour section of a fairly simple flow trace recorded from a small
office building. A constant flow of approximately 3.5 gpm is evident across the entire two-hour
interval, which is due to leakage. There are many toilet and urinal flushes, characterized by their
high flow rate and short duration and smaller faucet uses seen as well. In this trace, the entire
flow trace could be broken down into the desired end uses without difficulty.
40-
30-
S20- "
"" "i : iii: ', i
l .rI : : i i i, i : ,: 1
0
7/14/98 (7:25:01 AM 9:25:01 AM)
Figure 4.3 Example of a simple flow trace from a small office building
A more complex flow trace taken from a large hotel is shown in Figure 4.4. In this flow
trace, recorded from a large meter, the flow rate interval is 25 gpm. This means that because of
the low ratio of magnetic pulses to gallons each pulse in a ten-second interval represented a flow
of 4.2 gallons or 25 gallons per minute. When meters provide a single pulse for many gallons
the flow appear as multiples of that flow, and small events are averaged into the overall pattern.
When large volumes pass through the meter, as is the case with this large hotel, all that is seen is
a series of step flows. These can reveal very large or continuous uses such as irrigation or
cooling use, but no individual uses for showers, toilets or sinks etc can be discerned. This is why
the selection criteria were sought to limit the size of the facilities logged.
150 -
50
7J15/98 (6:00:47 AM 12:00:47 PM)
Figure 4.4 Example of flow trace from large hotel
A portion of a flow trace recorded from a specially installed sub-meter is shown in
Figure 4.5. The sub-meter was installed on a cold water line feeding 4 rooms at a La
Quinta Inn (a second meter was also installed on the hot water line). This figure shows toilet
flushes, which are the 4-5 gpm events, the lower flow sinks and the cold water portion of
showers. Total room use was determined by adding in the simultaneous trace from the hot water
line. The manager provided researchers with occupancy information during the logging period,
so from this sub-meter trace it was possible to make good estimates of daily use per occupant,
which could then be applied to the remainder of the hotel.
10-
CQ 6-
3
4-
3/20/99 (5:04:11 AM 11:04:11 AM)
Figure 4.5 Cold water use in 4 sub-metered motel rooms
A portion of a flow trace, which contains irrigation and cooling water flows, is presented
in Figure 4.6. This trace was recorded from a large office, but since the irrigation occurred late at
night (1:00-4:00 a.m.) and the volumes and flow rates were large (20-80 gpm) it stood out
clearly. Refilling of make-up water for the cooling tower caused the small green spikes that can
be seen before, after and on top of the irrigation event.
8 (1:40:59 AM 3:40:59 AM)
80
$7118f48 (1:40:59 AM 3:40:59 AM)
Figure 4.6 Irrigation and cooling use in large office (nighttime)
Summaiy Analysis
After the flow traces from an individual site were analyzed, daily estimates were made
for all of the identified categories during the logging period. These daily estimates were used in
conjunction with the billing data and other information collected in the site survey and
discussions with the building manager to create estimates of average annual use for each of the
identified end-uses. These annual use estimates were made both in terms of simple volumes
(gallons per year) and normalized on the basis most appropriate to the type of customer. A
summary table was prepared for each customer and included in the individual water use reports.
This chapter provides summaries for each category of customer with notes on individual
customers as necessary.
RESULTS
The detailed information collected from each site was assembled into a set of 24 detailed
individual water use reports. Much of the information that was obtained through data logging
such as daily, hourly and instantaneous demand patterns, is of interest, but is not directly related
to the main objectives of this research report. The purpose of this section is to provide a
summary of the findings that are relevant to the analysis of the annual average day end use
patterns of each of the CI institutions. This section also provided information on as many of the
end use modeling variables as possible so that these results could be use for calibrating and
evaluation the results of the conditional demand analysis modeling.
Office Buildings
General Information
Five office buildings were visited during the study. Table 4.1 shows the basic site
information about each of the offices. The original plan was to limit the sites to offices of no
more than 15,000 square feet, but as can be seen from the table the size of the buildings ranged
from 8,800 to 186,000 square feet. It was impossible to distinguish individual indoor events in
the larger buildings, but disaggregation of the use in the smaller buildings was possible to a great
extent. Some of the information on the total number of offices and persons working in the
buildings was not known by the owners and so was reported as unknown.
All of the sites listed in Table 4.1 were used for standard office purposes except for the
clinic building which contained a pediatrician's office, a reproductive health clinic, and an
optometrist's office. The office buildings in Irvine, Los Angeles, and Santa Monica were
commercial for-lease office buildings, and the San Diego building was occupied by a
government agency.
Table 4.1 Size and occupancy of field study office buildings
Irvine Los Angeles Phoenix San Diego Santa
Monica
Type General General Clinic Govt. General
Meter size (in) 2 4 2 1.5 4
Flow trace resolution High Low High High Low
No. of housed unknown 70 3 1 unknown
businesses
No. of floors 4 6 2 3 11
No. of workers in unknown 650 Unknown 110 unknown
building
Building area (sq. ft.) 57,785 176,500 10,000 8,800 186,000
Irrigated area (sq. ft.) 23,500 4,000 2,400 100 5,000
Annual and Seasonal Use
Billing data for the sites were first analyzed to determine the total annual water use. This
information is provided in Table 4.2. These data represent total water use for each site and have
not been normalized on the basis of either area or occupants. The seasonal water use was
estimated from the billing data using the average winter consumption method, which
extrapolates the minimum month (or bi-monthly) use over the entire year and classifies this as
non-seasonal use. Seasonal use is the difference between the non-seasonal and total annual use.
Seasonal use normally includes increased water use for cooling and irrigation during the
summer. It should be kept in mind that in the warm climates of the study sites there is still
significant cooling and irrigation use during the winter months, so non-seasonal use cannot
necessarily be taken as equivalent to indoor use.
Table 4.2 Annual and seasonal water use at field study office buildings
Irvine Los Angeles Phoenix San Diego Santa
Monica
Total annual use (kgal) 2,039 10,455 406 374 3,903
Average daily use (kgal) 5.6 28.6 1.1 1.0 10.7
Non-seasonal use (kgal) 562 7,423 210 374 2,694
Seasonal use (kgal) 1,477 3,032 198 0 1,209
Logging Data
Data loggers were installed on the water meter at each office building in order to collect
continuous traces of the flow at the site. Each of the flow traces was analyzed, and to the extent
possible, end use information was disaggregated.
Table 4.3 shows the daily and peak instantaneous water use at each office during the
logging periods. Average daily use during the logging period did not differ much from the
average daily use calculated from the billing records and shown in Table 4.2. The only
exception was the Irvine building, which had a leak during the logging period.
Table 4.3 Water use patterns at field study office buildings during data logging periods
Irvine Los Phoenix San Diego Santa
Angeles Monica
Logged average daily use (kgal) 12.4 21.4 1.2 1.2 8.3
Billed average daily use (kgal) 5.6 28.6 1.1 1.0 10.7
Indoor peak instantaneous 67.7 100.0 11.5 20.3 22.8
demand (gpm)
Outdoor peak instantaneous 97.6 125.0 20.9 3.0 24.5
demand (gpm)
Length of flow traces (days) 9.8 4.9 2.6 2.6 7.7
Disaggregation of Flow Traces
A brief description of the results of the attempted disaggregation of water use at each
office is provided below. Each site is discussed with respect to the level of detail revealed by the
flow trace analysis of indoor, outdoor and continuous uses.
Irvine Office. This site had a 2 inch water meter which provided good resolution to the
data logger and made it possible to disaggregate and identify the majority of the indoor uses at
the site. Toilet and urinal flushes and a large leak constituted the majority of all indoor use at the
building. The only other observed indoor uses were for faucets in bathroom and kitchenette
sinks and utility sinks in the janitor closets.
The majority of water use at the Irvine office building was for outdoor irrigation.
Outdoor use was separately metered which made it possible to obtain direct measurements of
outdoor use from the billing database. During 1997, a total of 2039 kgal were used at this
facility 562 kgal (22 percent) for indoor uses and 1477 (78 percent) for outdoor use, primarily
irrigation.
The data logger was placed on the water meter serving indoor purposes at the site and a
continuous demand of approximately 3.33 gpm was observed during the entire logging period.
This demand was probably not due to any process use within the building. There were no
processes identified during the site visit and billing records show that historical indoor demands
are closer to the disaggregated indoor uses rather than the total of the indoor and continuous
uses. The historical average daily indoor use was 1540 gallons per day (gpd) and the logged
indoor uses was 1971 gpd. The continuous use identified from this trace was appears to be due
to a continuous leak somewhere in the building's pipe network.
Los Angeles. Due to its large size (4") and high water use, it was not possible to
disaggregate water use at this site beyond the indoor use category. Total indoor demand, which
included both cooling water and domestic indoor, was estimated at 26,482 gpd across an entire
seven-day week. Cooling water usage accounted for 15,360 gpd (58 percent) of the total indoor
usage and domestic indoor (toilets, faucets, cleaning, etc.) accounted for 11,122 gpd (42 percent)
of the total. The measured average daily indoor usage during the logging period (two business
days and two weekend days) days was 19,602 gpd.
The primary outdoor water use at the site was the irrigation of 4,000 sf of landscape. The
total outdoor use during the logging period was 6,740 gallons. The average irrigation application
rate for landscape during the logged period was 1,685 gpd or 0.42 gallons per square foot (gpsf)
of landscape. Based on 250 watering days per year, the estimated annual irrigation demand at
the site is 420 kgal. On an annual basis this implies that the total irrigation rate was
approximately 105 gallons or 168 inches (14 feet) of application per square foot. However, it is
not known if the irrigation measured during the logging period is representative of irrigation use
throughout the year.
During the logging period a continuous demand in the form of cooling water and possibly
leakage was observed in the flow trace. The cooling system at the building included an open
looped 425-ton evaporative cooler and a closed loop 90-ton cooler. The estimated summer peak
cooling water load, based on 75 percent of the cooling capacity, is approximately 20,200 gpd.
The measured average business day cooling water usage during the logging period was 13,500
gpd. The average daily indoor demand (uses other than cooling) based on a 7 day week was
13,275 gpd. Winter cooling water usage was estimated at 25 percent of cooling tower capacity or
approximately 6,000 gpd.
Phoenix. This site had a 2-inch water meter that provided good resolution to the data
logger and made it possible to disaggregate and identify the majority of the indoor uses at the
site. Toilet flushes constituted the majority of all indoor water use. The other large observed
indoor water uses were for faucets and a dishwasher.
The majority of water used at this clinic office building during the logging period was for
outdoor purposes. This building was equipped with a single water meter, so to get an accurate
determination of the annual outdoor water usage it was necessary to extrapolate monthly indoor
use from the logging data and then subtract annual indoor demand from total annual use. This
approach for estimating outdoor demand has proven accurate because indoor use generally does
not fluctuate with the seasons like outdoor use. The office building is located in a region where
irrigation can occur during all months of the year so that minimum month techniques tend to
underestimate irrigation demands.
During the logging period there was no continuous demand in the office building. The
building used no water for cooling and did not have any continuous leakage.
San Diego. This site had a 1.5-inch water meter that provided good resolution to the data
logger and made it possible to disaggregate and identify the majority of the indoor uses at the
site. Toilet flushes constituted the majority of all indoor water use. Other observed indoor water
use included faucets, showers, leakage and other miscellaneous demands. The miscellaneous
indoor usage appeared to be primarily toilet flushes and faucet use that could not be
disaggregated into discrete events.
Outdoor use represented a small percentage of total water use at this site. The total
irrigated area at the site is less than 100 sf. The annual irrigation rate was estimated at 20 gpsf
per year or 32 inches of water per square foot of landscape.
During the logging period there was a small amount of continuous leakage at the site. The
total leakage during the logging period was 116 gallons. The average daily leakage during the
logging period was 58 gpd or 2.5 gallons per hour (gph).
Santa Monica. This building was unique among those studied in the degree to which
systems were monitored and controlled by the facilities manager. It was the only building in
which all of the fixtures and appliances had been updated with high efficiency devices. It also
was the only building in which the cooling tower was equipped with sand filtration, pH and
disinfecting controls, and meters on the bleed-off lines. A water sampling laboratory had been
constructed in a room adjacent to the air conditioning equipment. At this location samples were
taken from all of the cooling tower flow streams for chemical analysis and control. As a result,
the indoor usage per square foot at this building was the lowest of all five buildings, and its
cooling use was approximately one third of that in the other buildings in the study equipped
cooling towers.
Because of the size of this building, the large amount of daytime cooling water use, and
the fact it had a 4" meter which provides relatively poor resolution of low volume, short duration
water use events, it was not possible to identify all of the individual indoor water use events. The
logged average daily indoor use was 8250 gpd. By far the largest indoor use was the building's
cooling demand. The summer cooling water accounted for 6200 gpd and 75 percent of all indoor
water use. Toilets, urinals, faucets, showers and dishwashers used an average of 2050 gpd.
Based on the logging results it was estimated that there were approximately 60 toilet/urinal
flushes an hour during the 10 hours per day that the building was occupied. The estimated
average daily toilet/urinal usage was 1350 gpd. The 700 gpd of the other domestic indoor usage
was composed mostly of faucet usage.
The primary outdoor water use at the site was the irrigation of 5000 square feet of
immaculately tended landscape. The average irrigation application rate for the landscape during
the logged period was 1660 gpd or 0.33 gallons of water per square feet. The annual irrigation
rate was estimated to be 181 gpsf or 290 inches (24 feet) of irrigation application per year.
However, it is not known if the irrigation measured during the logging period is representative of
irrigation use throughout the year. The building also had a 1000 square foot fountain that used
400 gpd to fill and clean.
The flow trace data showed a continuous demand of 1.5 gpm apparently from a
combination of the 200-ton closed loop cooling tower and whatever leakage was present in the
building. This cooling tower operated continuously. In addition, there was a larger continuous
demand (5.5 gpm) during the daytime business hours from the two open loop cooling towers that
are operated 10 hours per day.
Average daily demand during the logging period was 80 percent of the average annual
use rate. Correcting for this difference, the annual make-up water for the cooling towers was
estimated at 2.25 million gallons. This cooling demand estimate was confirmed using a
theoretical calculation based on the metered bleed off rate, the measured concentration ratio, and
average capacity use estimates.
Estimated Annual End Uses
The estimated annual end uses for each office building are shown in Table 4.4. Estimates
of demand in individual use categories are shown for sites where it was possible to disaggregate
demand from the logging data. The idea of using engineering estimates or event data from one
building to disaggregate demand in another was rejected because the goal of the direct
measurement field studies was to base end use demand on direct measurements rather than on
inferences or induction.
It is difficult to use the results shown in Table 4.4 to compare demand between office
buildings because the range in building size is so great. To properly compare between sites, the
data were normalized on the basis of use per square foot of floor space (for indoor uses) and
irrigated area (for outdoor uses). The normalized end use results are presented in Table 4.5.
Table 4.4 Estimated annual end uses of water in field study office buildings (kgal/year)
Irvine Los Phoenix San Diego Santa
Angeles Monica
Demand (kgal/year)
Indoor
Faucets 32 49.7 46.5
Ice machines 0 0.0
General indoor 0 0.0
Other/misc. uses 215 2.0 63.5 255
Shower 0 0.0
Toilet 315 83.3 242 500
Total Indoor 562 4,035 135 352 755
.--.-.. -----In..... --r.-... ...................................... 562 .............4,035" ...13..5.2 ...755.....
Continuous
Cooling 0 6,000 0 0 2,250
Other continuous (leaks, etc.) 0 0 0 20 0
Total indoor + continuous 562 10,035 135 350 3,005
Outdoor Use
Outdoor 1,477 420 273 2 800
Fountain 0 0 0 0 100
Table 4.5 Normalized end uses in field study office buildings (gal/sf/yr)
Irvine Los Phoenix San Diego Santa
Angeles Monica
Demand (gal/sf/year)
Indoor*
Faucets 0.55 4.97 5.30
Ice machines 0.00
General indoor 0.00
Other/misc. uses 3.72 0.19 7.22 1.37
Shower 0.00
Toilet 5.45 8.33 27.50 2.70
Total Indoor 9.72 22.86 13.49 40 4.07
Continuous
Cooling 0.00 34.00 0.00 0.00 12.10
Other continuous (leaks, etc.) 0.00 0.00 0.00 2.27 0.00
Total indoor + continuous 9.72 56.86 13.49 39.77 16.17
Outdoor Use
Outdoort 63.00 105.00 113.75 20.0 160.00
Fountain 0.00 0.00 0.00 100.00
Based on square footage of office space
tBased on irrigated area
tBased on square footage of fountain
Modeling Parameters
To permit the results from the direct measurement field studies to be correlated with the
conditional demand models, information has been extracted for values of the model variables
where available from the field studies. Only those variables for which positive, non-zero, data
exist were tabulated. Variables that exist at the sites, but have unknown values, have been listed
as unknown. The modeling parameters are presented in Table 4.6.
Table 4.6 Model parameters from field study office buildings
Model Parameter Irvine Los Angeles Phoenix San Diego
Kitchenettes
Kitchens
Display fountains
Dishwashers
Laboratories
Public Restrooms
Tank Toilets
Valve Toilets
Public Urinals
Sanitary Faucets
ULF Toilets
Showers
Wash Stations
Cooling Towers
Cooling Tons
Chillers
Chiller Capacity
2
0
0
1
0
9
0
24
4
9
0
0
4
0
0
1
unknown
12
0
1
0
0
12
0
50
15
44
0
0
6
1 open loop
1 closed
loop
428 open loop
90 closed
loop
1 1 1
unknown Unknown unknown
Santa
Monica
11
0
1
6
0
22
0
64
15
67
64
0
11
2 open loop
1 closed
loop
360open
loop
200closed
loop
1
unknown
Ice Machines
Feed Water TDS
Concentration Ratio
0
NA
NA
1
290
3.8
NA
NA
600
3.3
3.3
|