Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Towards a national infrastructure strategy: water distribution systems : a case study Boras, Randall James 1994

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1994-0269.pdf [ 4.85MB ]
Metadata
JSON: 831-1.0094724.json
JSON-LD: 831-1.0094724-ld.json
RDF/XML (Pretty): 831-1.0094724-rdf.xml
RDF/JSON: 831-1.0094724-rdf.json
Turtle: 831-1.0094724-turtle.txt
N-Triples: 831-1.0094724-rdf-ntriples.txt
Original Record: 831-1.0094724-source.json
Full Text
831-1.0094724-fulltext.txt
Citation
831-1.0094724.ris

Full Text

TOWARDS A NATIONAL INFRASTRUCTURE STRATEGY:WATER DISTRIBUTION SYSTEMS- A CASE STUDYbyRANDALL JAMES BORASB.Sc., The University of Alberta, 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESTHE SCHOOL OF COMMUNITY AND REGIONAL PLANNINGWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1994© Randall James Boras, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(Signatur____________________________spartment of____________The University of British ColumbiaVancouver, CanadaDate Z9 FvJLDE-6 (2/88)AbstractTreated water distribution systems in North America represent a majorcomponent of the physical infrastructure in dire need of closer attention by theregulatory bodies, local, regional and national governments, and the public in general.The problems identified by the media over the past decade have been limited to picturesof collapsing or deteriorating pipes. The real problems run much deeper. Reducedgovernment funding over the years, changing public priorities, and a lack ofcomprehensive information required to accurately define the problems have plagued theoverall management of water systems in Canada and the United States.This thesis provides an overview to municipal water distribution systems inCanada, investigating not only the physical processes responsible for the deterioration ofsuch systems, but the historical impetus associated with the development of such systems,the physical profile of the systems unique within Canada, the changing socialenvironment surrounding aged systems, and the real costs associated with repairing wornout systems. Existing historical information is gathered from a variety of sources toprofile the Canadian systems. Research by governments and lobby groups, especially theFederation of Canadian Municipalities, is reviewed and summarized. Technicalinformation and current techniques for managing individual water systems are alsoreviewed. The information is then sysnthesized into a number of policy suggestionsaimed at effective solutions to the current crisis and reviewed within the context of asmall community in Greater Vancouver.There is no single magic formula to solving the problems, but rather a wide and11varied combination of improvements which must be made over the broad spectrum ofwater distribution system management. National, provincial, and local bodies are allinvolved to some degree in the management decisions and all could utilize moreeffective management techniques which focus on better implementing already availabletechnologies rather than developing new technologies. Rehabilitation decisions must bebased on sound principles aimed at effectively protecting public interests, rather thantechniques which are often geared more to the availability of grants rather than theactual condition of the pipes.Information important to the decision making process must, however, not berestricted to the decision makers. Polls have shown that public interest and concern overdrinking water issues is typically very low, and is only heightened by crisis-type situations.This was very evident in the recent federal election, where suggestions by the Liberalparty to implement an infrastructure program were often met by ridicule and cynicism,considered as opportunistic spending aimed only at securing the votes of theunemployed, rather than any legitimate concern for public health or economic benefit.In recent months the Liberal party has formed the new federal government andhas committed an infrastructure program, having set aside $2 billion to solving what maynow be a $30 billion infrastructure problem in Canada. To maintain the public faith andto assure these limited funds are used effectively, there is a real need for improvedpolicies. This thesis will endeavour to provide the basic framework for a national policyto better manage the present and the future of the conduits which carry the gift ofhuman life through our towns and cities - good, wholesome, clean, drinking water.iiiTABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS ivLIST OF TABLES viiLIST OF FIGURES viiiACKNOWLEDGEMENTS ixCHAPTER 1: INTRODUCTION1.0. Introduction 11.1. Methodology 51.2. Organization 61.3. An Introduction to Infrastructure 71.3.1. Characteristics Unique to Water Distribution Systems 81.3.2. Information and Research into Water Distribution Systems 101.3.3. Dealing with the Physical Nature of Underground Systems 131.3.4. Overview 16CHAPTER 2: DEVELOPMENT OF CANADIAN WATER SUPPLY SYSTEMS2.0. Introduction 182.1. History of Canadian Water Supply 192.1.1. The Early Canadian Urban Centres 192.1.2. The Early Development of Water Works 212.1.3. The Push for Publicly Owned Utilities 292.2. Funding for Water Systems 312.2.1. Federal Programs 362.2.2. Provincial Programs 432.2.3. Local Programs 452.3. Legislation 482.3.1. Federal 482.3.2. Provincial Legislation 492.4. Factors Prompting Rehabilitation 522.4.1. Increasing Maintenance and Operation Costs 532.4.2. Fire Protection and Liability 572.4.3. Health Concerns 612.4.3.1. Water Quality Standards 612.4.3.2. Sources and Impact of Contaminants 632.4.4. Impending Changes in the Rules 682.4.5. Environmental Concerns 702.4.6. Economic Development Considerations 722.4.7. Public Concern 74ivCHAPTER 3: CANADIAN WATER DISTRIBUTION SYSTEMS INVENTORY3.0. Overview 763.1. Data Sources and Methodology 773.2. Results of the National Inventory Estimate 793.2.1. Pipe Length and Service Density 793.2.2. Age of Distribution Systems 823.2.3. Pipe Sizes 923.2.4. Pipe Material 973.2.5. The Physical Environment 1023.2.6. System Demand 1083.2.6.1. Community Demand Profile 1103.2.6.2. Regional Demand Profile 1123.2.7. Water System Staffing Levels 113CHAPTER 4: WATER MAIN DETERIORATION4.0. Overview 1144.1. Defining Water Main Deterioration 1154.2. Stresses on a Pipe 1174.3. Age as an Indicator of Pipe Condition 1304.3.1. Historical Development of Piped Systems 1304.3.2. Development Modern Pipe Materials, Construction and Design Techniques 1324.3.3. Modern Pipe Design: Rigid and Flexible Pipe 1344.4. Other Factors Influencing Pipe Performance 1354.5. Corrosion Processes 1374.5.1. Internal Corrosion 1384.5.2. External Corrosion 1424.5.3. Stray Current Corrosion 1454.5. Deterioration Criteria 146CHAPTER 5: CONDITION ASSESSMENT AND MITIGATWE TECHNIQUES5.0. System Condition Assessment 1495.1. System Inventory 1495.2. Condition Assessment Techniques 1505.2.1. Descriptive Analysis 1515.2.2. Predictive Analysis 1525.3. Decision Analysis 1575.3.1. Rules of Thumb 1585.3.2. Economic Methods 1595.3.3. Reliability Analysis 1625.3.4. Physical Models 1645.3.5. Hydraulic Performance 1685.3.6. Field or In-Situ Inspections 1695.4. Mitigation Techniques 1715.4.1. Pipe Replacement 1735.4.2. Trenchless Methods 176V5.4.3. Demand Management 1795.4.3.1. Economic Techniques 1815.4.3.2. Structural Techniques 1855.4.3.3. Sociopolitical Techniques 1865.4.4. Land Use Planning 189CHAPTER 6: DEVELOPMENT OF A POLICY FRAMEWORK6.0. Goals of a Comprehensive Capital Works Management Policy 1926.1. Maintaining high levels of service in Canada 1966.1.1. Eliminating the Current Capital Works Backlog 1966.1.2. Developing Long Term Goals and Plans 1986.1.3. Reacquaint the Public with Lifeline Systems 2006.1.4. Reassessing the Major Roles 2026.2. Improved Replacement and Decision Guidelines 2056.2.1. Determining when to replace a pipe 2116.2.2. Condition Assessment and Monitoring 2196.2.3. Development of National Standards 2216.2.4. Varying Communities 2236.2.5. Program Strategies for Various Sizes of Communities 2256.3. Long Term Fiscal Policy Restructuring 2276.3.1. Costs of Capital Replacement Programs 2296.3.2. Impact on the Average Canadian Water Bill 2326.3.3. Funding Sources 2356.3.4. Economic Benefits of Water Distribution System Investment 2366.4. Summary 239CHAPTER 7: CASE STUDY: PITf MEADOWS7.0. Background 2407.1. Water Consumption 2427.2. Water Rates 2437.3. The Water System 2457.4. Record Keeping 2507.5. Staffing Levels 2517.6. Funding 2517.7. Rehabilitation and Replacement Programs 2537.8. Discussions 2547.9. Summary Recommendations 258CHAPTER 8: CONCLUSIONS AND RECOMMENDATIONS8.0. Overview 2618.1. Application 2618.2. Recommendations and Solutions 262BIBLIOGRAPHY 268APPENDIX 282viLIST OF TABLES1.1: Classification of civil engineering infrastructure 81.2: Classification of infrastructure condition 142.1: Water supply services in surveyed Canadian communities in 1986 272.2: The extent of unit treatment processes in Canada in 1986 282.3: Total value of construction work on water distribution systems Canada 1973-1983 322.4: Comparison of average international water prices in 1986 472.5: Typical prices for popular liquids 472.6: Areas of primary concern in Ontario’s water distribution systems 523.1: Data sources used in estimating the Canadian water pipes inventory 773.2: Population served per kilometre of pipe in Canada 813.3: Estimated average age of distribution systems by region in 1991 873.4: Estimated age of distribution systems in selected provinces in 1991 903.5: Largest and smallest pipe sizes in Canadian communities in 1961 943.6: Pipe sizes in Surrey, B.C. in 1978 963.7: Pipe material in selected B.C. utilities 1003.8: Average frost penetration in selected cities (1964-1971) 1063.9: Domestic per capita water use in selected countries 1083.10: Municipal water use in Canada 1093.11: Domestic per capita water use in Canadian communities in 1989 1103.12: Total per capita water use in Canadian communities 1113.13: Occurrence of domestic water use rates in Canadian communities in 1989 1113.14: Provincial per capita water use in Canada in 1989 1123.15: Regional per capita water use in Canada in 1989 1134.1: Reported reasons for water main failures in Ontario 1234.2: Stresses associated with common causes of breaks 1254.3: Water characteristics related to internal corrosion 1394.4: Soil corrosion evaluation rating for cast iron pipes 1444.5: Criteria used to evaluate water main deterioration 1475.1: Inventory information for water distribution systems 1505.2: Descriptive analysis factors 1515.3: Typical residential price elasticities 1826.1: Roles and organization 2036.2: Examples of organizational input and involvement at various levels 2056.3: Summary of renovation and replacement costs in Ontario in 1983 2306.4: Estimated national renovation and replacement costs in Canada 2326.5: Cost comparison between various options for national rehabilitation 2346.6: Costs and benefits of a combined replacement/renovation program 2387.1: Pitt Meadows recent population growth 2407.2: Pitt Meadows water consumption in 1990 2427.3: Comparison of water consumption with national and regional data 2437.4: Pitt Meadows commercial water rates 2447.5: Historical information on the Pitt Meadows water system 2477.6: Breakdown of Pitt Meadows water system by pipe size in 1981 2477.7: Estimated staff work load split among systems in Pitt Meadows 252viiLIST OF FIGURES1.1.2.1.2.2.2.3.3.1.3.2.3.3.3.4.3.5.3.6.3.7.3.8.3.9.3.10.3.11.4.1.4.2.4.3.4.4.4.5.4.6.5.1.5.2.6.1.6.2.6.3.6.4.6.5.7.1.7.2.Canada’sCanada’sCanada’sCanada’sCanada’sCanada’s11243340808384858689919598101104116120124126127129156161194215217220224241246Per capita costs by facility in CanadaCanada’s water works: Number of systems by yearTotal value of construction works on Canadian water distributions systemsAverage levels of federal infrastructure fundingCanada’sCanada’sCanada’swater distribution systemswater works: Persons per kilometer of watermain by yearwater works: Persons per kilometer of pipe in selected citieswater works: Population served in surveyed communities by yearwater works: Kilometers of pipe in surveyed communities by yearwater works: Percentage of main by age and by regionwater works: Length of main by age and by regionwater distribution systems: Diameter of mains by size of communitywater distribution systems: Percentages of material types by yearCalgary’s distribution system: Kilometers of watermain by material and yearAreas sensitive to acid rainConceptual model of water main structural conditionTypes of water main breaksStresses on a buried water mainDegree of resistance to various types of stressDegree of resistance to various failure typesDegree of resistance to leakage sourcesTypical hazard functionOptimal pipe replacement schedulingNational water distribution policyWater main structural lifecycleWater main replacement schedulingCriteria involved in a rehabilitation decisionConceptual model of overall system break ratesPitt Meadows population growthPitt Meadows existing pipeline networkviiiAcknowledgementsI am deeply grateful to a number of people who guided, questioned, challenged,and supported me throughout the last few years. I am especially thankful for theassistance of the faculty at U.B.C. especially Prof. Setty Pendakur, in the School ofCommunity and Regional Planning, and Prof. Frank Navin, in the Department of CivilEngineering, who were not only valuable in guiding my work, but also assisted me insecuring financial assistance. I am grateful to the Natural Sciences and EngineeringResearch Council for having the confidence in me to provide the scholarship to carry outmy research. I am also indebted to the late Bill Curtis, the former City Engineer at theCity of Vancouver, whose total and long-time commitment to solving the currentproblems and thoughtful insight and vision for the future guided my course. I am deeplygrateful to the numerous faculty members, municipal employees, and governmentofficials who provided data, interviews, and spare time to assist me in putting this thesisinto perspective. I would also like to thank MPE Engineering Ltd. for providingassistance and understanding along the way. Finally, to my friends and family I owe mydeepest thanks. Mom and Dad this is especially for you.ix1CHAPTER 1: INTRODUCTION1.0. IntroductionMedia exposure in North America over the past decade has presented images ofdeteriorating bridges, clogged roadways, overcrowded airports, and collapsing watersystems, graphically defining the term “infrastructure deterioration.” A combination offunding shortfalls, political shortsightedness, and poor planning is blamed for many of theproblems. In Canada, serious concern over the unacceptable decay of urban publicsystems has surfaced over the past eight or nine years. Much of the current problem isrooted in policy shifts dating back to the late 1960s and early 1970s which have led toreduced spending on system rehabilitation and replacement.In 1984 the Federation of Canadian Municipalities (FCM 1984) conducted asurvey of 63 municipalities and estimated that $598 per capita, or over $10 billionnationally, was required to rehabilitate six major urban public facilities: roads, sewagecollection, water distribution, sewage treatment, water treatment, and bridges. Since thattime, from a provincial perspective, only Alberta has expanded its capital fundingprograms to include rehabilitation of existing works, leading the FCM to conclude in1991 that overall the situation is not improving, but rather becoming worse (Curtis1991b). Taking into account inflation and the additional deterioration since the 1984report, there may now be as much as $15 to $20 billion worth of work waiting to be doneand some feel the number may be as high as $30 billion (Mavinic 1990).The recent two year $2 billion federal contribution to a new infrastructureprogram announced by the newly elected Liberal government represents a significantshift in policy, which over the past decade has been dominated by the thinking thatinfrastructure problems were largely provincial and local matters. Prime Minister Jean2Chretien’s new program will adopt the funding formula advocated by the Federation ofCanadian Municipalities, a formula based on equal contributions by federal, provincial,and local govermnent. Unfortunately, the reality of increasing debt loads at both theprovincial and local levels may make it difficult to carry out the required work. From anoverall perspective, the new program will only address a fraction of the problemsidentified almost a decade ago by the FCM. For this reason, the limited resources nowbeing directed toward the problem must be used in the most effective way possible. Agood understanding not only of the physical engineering aspects of deterioratinginfrastructure systems, but of the overall political, environmental, and financial climatesis key to making the most of the current opportunity. Unfortunately, even with therecent work over the past decade by the FCM, there remains major gaps in even themost basic information regarding the condition of our infrastructure systems and thereremains major inadequacies in the management of such systems.The aim of this thesis is to paint a clear picture of current infrastructuremanagement in Canada. To date, a very broad brush has been used to illustrate theproblems, concentrating more on identifying the multi-billion dollar backlog of work andhow to pay for it, with less attention to the details of how we got in this mess, where weare going now, and how we can do things better to get out of it. This thesis willcontribute by exploring the scale, context, and nature of the problems through a detailedstudy of one particular component of our overall physical infrastructure: treated waterdistribution systems.Currently, deteriorating systems are being maintained far past their economicallyeffective life, with already scarce time and monetary resources being used to holdtogether system components that should have been replaced long ago. A shift away3from the prevalent social, political, and financial emphasis on growth and expansion ofnew facilities which began after World War II, to one of effective maintenance,reconstruction and renovation of existing facilities is now required. This can and shouldbe achieved by a number of means, apart from the most obvious infusion of hugeamounts of capital from debt-burdened provincial and federal sources: application ofbetter information systems to allow effective monitoring of infrastructure condition;improved techniques for determining optimum rehabilitation and replacementscheduling; more effective pricing of water to reflect the true value of the resource; andincreased public awareness to both reduce the demand on overworked systems and toincrease the funds available for maintenance and repair. The current philosophy ofdesign which aims at satisfying ever increasing user demands through ever expandingsystems must shift to one which recognizes the need for conservation and more effectiveuse of both the water and the system carrying it. Historically a local concern, waterdistribution systems still serve sufficient interests at both the provincial and federal levelsto merit continued involvement by all levels of government.Technical, financial, political and social concerns are reviewed to develop arational framework which can be utilized to help manage the problems and improve thesystems. At the local level, the framework stresses the importance of improving bothsupply side and demand side management of water distribution systems. At theprovincial and national level the framework emphasizes increased technical support andmanagement guidance, as well as appropriate financing assistance.The water distribution system of a small Canadian community is then profiledwithin the context of the framework to investigate the potential application and specificproblems which must be resolved.4This thesis, while ultimately aimed at senior federal and provincial policy makers,is written and presented to allow a basic understanding of infrastructure managementproblems by post-graduate planning and engineering students, professional planners andengineers, and decision makers involved on a day to day basis with managing andmaintaining infrastructure systems. The thesis purposely focuses on both technical andnon-technical factors which equally contribute to today’s situation.The thesis will concentrate on improving existing water systems which requirerehabilitation or replacement because of deterioration or regulatory obsolescence. Someattention will be given to the problems encountered by communities experiencing rapidgrowth where diminishing system capacities and continued system expansion are majorconcerns. However, the major emphasis will be on planning for the continued use ofexisting systems rather than on the problems associated with growth, a subject which initself is sufficiently complex to merit study on its own.While this thesis admittedly does not provide all the solutions to waterdistribution infrastructure in Canada, it does examine a number of the problems withinfrastructure management in general. This thesis is meant to provide a clearer pictureof the problems being dealt with, rather than represent the final solution. As such, itsymbolizes a small, though not insignificant, piece to the whole puzzle.51.1. MethodologyThe methodology of this thesis consists of five basic components pertaining towater distribution systems both in general terms and terms more specific to the Canadiancontext:1) a literature review of recent trends and developments pertaining to water systems,2) establishment of baseline information for discussion of Canada’s water distributionsystems,3) development of a rehabilitation policy framework for application within Canada,4) discussion of a case study within the context of the policy framework, and5) final conclusions and reconmiendations.The literature review includes an examination of the water distribution systems inCanada: the historical development of them, the current legislation and standardsregulating them, the major concerns regarding them, the physical mechanismsdeteriorating them, and the present mitigative techniques and technology available toimprove them.The estimate of Canada’s total water distribution systems draws on historical andcurrent data to develop a national inventory of systems. The total length of the systems,as well as information on pipe sizes, material types, and age have been compiled fromexisting sources to provide a baseline for discussion of policies aimed at the particularproblems and specific needs of different communities and regions in Canada.A general framework for water distribution system rehabilitation applicable to theCanadian context is then formulated from the information reviewed. The frameworkbasically involves the application of demand management principles and appropriatestate-of-the-art rehabilitation and replacement models to Canadian systems.6Terms of the framework are then discussed relative to Pitt Meadows, a smallcommunity in metropolitan Vancouver. The discussion focuses on the applicability oftechniques, the impediments to implementation, and highlights by example the specificproblems which can be encountered at the community level.The final section of the thesis includes the conclusions drawn from this study aswell as recommendations aimed at the general application of such a framework withinCanada.1.2. OrganizationThe thesis is organized into eight chapters. Chapter one outlines the scope of thethesis, the problem statement, the methodology, and the organization. Chapter twoincludes a history of Canada’s municipal water systems, a review of the currentlegislation and standards, and a look at the factors behind the need for water distributionsystem rehabilitation. Chapter three includes an estimate of the total water distributionsystems in Canada. Chapter four is more technical in nature and outlines thedeterioration mechanisms which affect pipe life. Chapter five is an overview of currentstate-of-the-art system condition monitoring and mitigative techniques. Chapter sixoutlines the development of a general framework for water distribution rehabilitation inCanada. Chapter seven includes a case study on the application and assessment of thisframework to Pitt Meadows, a small municipality in Greater Vancouver. Chapter eightsummarizes the findings, draws conclusions, and makes recommendations on theapplicability of the framework to the Canadian scene.71.3. An Introduction to InfrastructureThe term “infrastructure” is a generic label given to the structure of a host ofsystems which connect or serve human activities in some way. The systems are typicallyquite large and complex, and are often so deeply entrenched in the workings of everydaylife that they are assumed to exist and often are taken for granted. The systems can beprivately held, as are many telephone utilities and gas companies, or publicly held as arethe provincial health care and highway systems.A further distinction is made between social and physical infrastructure (FCM1984). The social infrastructure is comprised of formal connections based less onphysical structures such as pipes and cables, and more on social structures such as laws,bureaucracies, or political systems. Functions such as health care, education, policeprotection, and recreation are included in the social infrastructure.The physical infrastructure, or as Adams and Heinke (1987) term it, the “civilengineering infrastructure” refers to “the extensive and costly physical facilities thatprovide the goods and services necessary for the functioning of modern society.” Itincludes a vast array of systems associated with communication, transportation,transmission, production, and extraction in a variety of regional contexts as illustrated byTable 1.1.Such systems are not only characterized by their massive scale and complexity, butalso by their functional time dependence; they experience loads or demands which varyover time and as such require continuous or intermittent operation (Adams and Heirilce1987). Further, all of these systems, regardless of the degree of maintenance, willeventually wear out and need to be replaced. When this is overlooked in the initialplanning and operational stages of a system, inadequate replacement programs often8result. As the FCM points out, this is the case with many urban systems which are in anunchecked mode of slow decay.Clearly an assessment of the current condition of all forms of even the publicinfrastructure would be an enormous task. The FCM (1984) report concentrates on thecondition of urban infrastructure, which is generally within the responsibility of localgovernment, and finds that the systems associated with the urban social infrastructure inCanada are in a relatively better condition than the physical systems.Table 1.1: Classification of civil engineering infrastructureRegion Public Works Private WorksUrban Water Distribution StructuresSewage Collection Electrical SystemsRoads, Bridges, etc. Gas Distribution, etc.Interurban Highways, Railways Pipelines, RailwaysTransmission Lines, etc. Aqueducts, etc.Nonurban Dams, Reservoirs Dams, Power PlantsRural Roads, Bridges, etc. Mining Structures, etc.Source: Adams and Heinke (1987)1.3.1. Characteristics Unique to Water Distribution SystemsThe problems of water distribution system deterioration and infrastructure arecomplex, involving all levels of government and many interests, and as such requirestrategies based on an interdisciplinary approach including engineering, management,and policy skills (Grigg 1988, p.7).Even other seemingly similar underground systems, such as the sewer and9drainage systems, have significant functional and operational differences which make forvery different deterioration and failure criteria and which require much differentmaintenance and rehabilitation procedures. In addition, water pipes operate underpressure and transport drinking water, two characteristics which imply a set ofoperational and health criteria very different from any other system.The FCM (1984) outlines many factors which have led to the general problem ofinfrastructure in Canada over the past twenty years, including increased competition fromexpanded social programs for even scarcer funds, reduced local debt financing due tohighly unstable interest rates, spiralling costs due to inflation, and increased publicconcern about disrupting the urban environment. But there are also characteristics ofthe individual systems that make each’s “crisis” unique. For instance, the FCM (1984)noted that road systems and sidewalks in Canada were perceived to be in the worstcondition partially due to the fact that they are highly visible, but more because they arefunded from local general revenues, where competition for funds is intense. Waterdistribution systems on the other hand are perceived to be in relatively better conditionmostly because they have their own independent source of funding in the form of “user-fees” (FCM 1984). But the fact that underground systems are buried means their truecondition is not always under public scrutiny and utility managers are not faced by thebarrage of day-to-day complaints associated with highly visible defects. This may tend tounder-estimate the perceived needs of buried systems.But assuming the perception is correct that the roads in Canada are in a muchworse condition than the other systems, water distribution systems still represent asignificant problem. Of the FCM’s (1984) estimated $598 per capita (1984 dollars)required nationally to improve the six most critical physical systems, roads will require10the largest share ($249 per capita or 41.7 percent), followed by sewage collection systems($97 per capita or 16.2 percent) and then water distribution systems ($76 per capita or12.7 percent) as can be seen in Figure 1.1. After applying inflation to the 1984 FCMfigures, approximately $2.2 billion is currently required just for water distributionimprovements.1.3.2. Information and Research into Water Distribution SystemsConsidering that Canadians have invested heavily in their underground systems,information and research is both sadly lacking and uncoordinated (Grover and Zussman1985). The replacement value of the municipal water supply systems in Canada has beenestimated at $62 billion, or nearly $3,000 per person served (1984 dollars), withapproximately 80 percent invested in the distribution system and 20 percent in treatment(MacLaren 1985). While the large scale of this investment indicates that Canadianshighly value such systems, only a small percentage of the current research in watersupply is even devoted to the study of the distribution systems. In addition, while acomprehensive inventory of the water treatment facilities has been compiled by theFederation of Associations on the Canadian Environment (FACE 1987), there does notexist a comparable national inventory of the 2,887 underground systems which FACEestimated in 1986 to be serving over 21.5 million Canadians.Research into roads rehabilitation using modern maintenance managementtechniques has been ongoing for the past few decades. Unfortunately, the same can notbe said for the water distribution systems where techniques have not evolveddramatically in the past 50 to 60 years. Rehabilitation by cleaning and mortar lining andreplacement of a deteriorated pipe with a new one by open trench methods remain theFigureLi:PercapitacostsbyfacilityinCanada0 ci 0 C-) L. U) ci C!) 0 0SewageCoIectionWaterDistributionBridgesFacility300250200150100 50Roads(%,WaterTreatmentSewageTreatmentSource:FCM198412norm. While the development of new light-weight plastic pipe materials havemadeinstallation easier and more economical, procedures for improving the old pipes hascome slowly. New trenchless technologies utilizing the existing “hole in the ground” haveproved promising, but have not been applied extensively to pressure mains.The American Water Works Association has in recent years compiled a guidancemanual which summarizes the condition assessment and replacement managementtechniques developed and utilized by a few large North American utilities (AWWA1986a). In England the Water Research Council has also been developing guidancemanuals, but even with these recent developments, adequate standards and the widespread implementation of new techniques are slow in coming. Improvements arerequired in information sharing, research, and especially water rate structures aimed atproper system financing. Currently, very few communities plan for the eventualreplacement of their systems by including in the price for water a charge for long-termcapital requirements.The very nature of underground systems make them especially prone to bothresearch and funding neglect: 1) they are physically buried so their true condition cannot be seen first hand by their users nor their managers; 2) they can provide satisfactoryeveryday service for long periods of time without having to be replaced; 3) their massivescale combined with public control means major decisions on development and fundingare ultimately made at a political level. Traditionally water systems have beencategorized as one of many public services, often contributing funds to general revenues,but not always getting them back when required. While the recent trends toward selfsustaining water systems through user-pay principles have been developing, competitionfor scarce general revenues is still ongoing for major upgrades.13When in direct competition for general revenue funding, the characteristicsinherent to underground systems put them at a disadvantage. Low visibility and theirmundane nature make them low political priorities; they are “out of sight, out of mind”.This is made worse by a political system which demands quick, highly visible benefitsover relatively short 4 or 5 year cycles, whereas the full benefits of infrastructureimprovements are rarely realized in less than 25 to 50 year cycles. There are very fewplaques or photo opportunities commemorating the opening of a new stretch of watermain.Part of the problem lies in modern societies emphasis on growth and things whichare new and novel. Political, financial, educational, and research institutions all tend tofocus on the construction, development, and eventual expansion of new works, payinglittle heed to the need for eventual repair or replacement of existing works. Ultimatelyimprovement programs are more often than not crisis-driven rather than preventative(Bradley 1987, MacLaren 1987).1.3.3. Dealing with the Physical Nature of Underground SystemsPerhaps contrary to popular belief, water distribution systems are more thansimple pipes that once buried in the ground will operate for a couple hundred years withonly the minimum of care needed to fix a few breaks each year. All systems eventuallybecome inadequate over time due to either structural inadequacies or capacityinadequacies or a combination of the two (see Table 1.2).The design life of any system depends largely on the system’s purpose and may berationally and economically determined to be quite long or relatively short. In the caseof water distribution systems, the useful life is typically set anywhere from 50 to 10014years, and is often based more on the time required to pay off the system through debtfinancing rather than any performance criteria. Still, even with proper design, manysystems develop significant problems in a considerably short period of time while othersmay function satisfactorily for considerably longer.Table 1.2: Classification of infrastructure conditionStructuralCapacity! Adequate InadequatePerformanceAdequate A CInadequate B DA- does not require rehabilitationB - requires rehabilitation for capacity inadequacy onlyC - requires rehabilitation for structural inadequacy onlyD - requires rehabilitation for both capacity and structural inadequacySource: Adams and Heinke (1987)The design environment surrounding a water main is full of uncertainties, withmany complex mechanisms which can significantly shorten its life. Corrosion, soilmovement, temperature changes, and fluctuating internal and external pressures are onlya few of the factors which must be accounted for in design, but which can varyconsiderably even over short lengths of pipe in the field. When such anomalies increasethe incidence of leaks and breaks and thereby the need for repair or replacement, theresult may be a serious financial burden on a community.15To manage the risks associated with pipe deterioration, some utilities formulategeneral repair and replacement policies, which are often based on simple criteria wherea pipe is replaced when it reaches a certain age or has had a certain number of breaks.The development of such policies is often not rigorous and is largely based on broadgeneralizations. O’Day (1983) notes that while age may be a good indicator of thegeneral condition of a system, it is not a useful tool in determining the need to replacean individual pipe, and Andreou and Marks (1987) note replacement based on simplebreak counts is much less than optimal. Thus simple replacement policies based on suchgeneralizations cannot guarantee optimal allocation of scarce local funds.The most effective approach being promoted in the literature seems to be oneaimed at assessing the condition of the individual pipes based on performance orstructural criteria, and replacing only those pipes that need it. A number of authorshave developed statistical techniques for determining optimum replacement strategiesbased on break and repair trends (Shamir and Howard 1979, Clark et al 1982, O’Day1983, Andreou and Marks 1987) while others have applied time-to-failure analysis andrisk assessment techniques (Bratton et al 1986). However, the practical application ofthese techniques has been limited to mostly large utilities which generally have largestaffs and many resources, and has yet to “trickle” down to the smaller centres.Part of the reason for the slow movement of such techniques out from the largercentres is the perception that infrastructure deterioration, and more specifically waterdistribution system deterioration, is a big city problem and not one experienced byyounger and smaller communities, which developed primarily after World War II. Inreality, the problem is not limited to the older central cores of cities such as Halifax orMontreal. In Calgary, a relatively young city, a corrosive underground environment has16been especially harsh on the water system. In smaller communities, which have beenenduring slower growth in recent years, the average system ages are only 10 to 15 yearsyounger than the major centres, indicating that problems first noticed in the large cities adecade ago are probably now becoming more acute in these smaller centres.Unfortunately, many small communities lack the information, staff, and financialresources not only to deal with the problem, but to simply recognize it. Large cities suchas Vancouver and Calgary have independently developed sophisticated pipe replacementmodels which require extensive maintenance records as well as large competent staffsspecially trained in water supply. In some of the smallest communities even the mostrudimentary system information has not been compiled, as-built drawings have been lostor are inaccurate and the only accurate records are in the head of a retired systemoperator. Staffing is also a problem in small centres, as the same person that picks upthe garbage often maintains the water system (Grover and Zussman 1985).All of these factors combine to present a picture of potentially varying levels ofservice among communities, less than ideal system operation and waste both in the waterand financial resources.1.3.4. OverviewWhile this thesis cannot come close to providing all the answers to the problem,there are a number of reasons why its approach has some merit and why waterdistribution system deterioration is important to look at now.Intuitively, underground systems should be considered in the earliest stages of anygeneral infrastructure program such as that suggested by the FCM. The reliability of thesubsurface systems should be established prior to any major outlay of money directed17toward the rehabilitation of surface systems such as roads and sidewalks.Local government in Canada is currently faced with reduced levels of funding forthe rehabilitation and replacement of their deficient piped water systems and whileconservation and improved rate schedules promoted by the federal government over thepast decade may be critical in helping to reduce the demand on such systems, they alonewill not solve the entire problem, and at best can only postpone the eventual failure andrepair of the piped systems.From a national perspective, rehabilitation of water distribution systems can beconsidered an appropriate opportunity for local initiatives aimed at more effectiveresource management. Canadians are currently one of the highest per capita users ofmunicipal domestic water in the developed world, yet pay among the lowest water rates(Tate 1990). In addition, a few water distribution systems in Canada can leak up to 30percent of their treated water into the ground (Environment Canada 1990a). With thecurrent concern for sustainability in our consumption conscious world, measures such asconservation, effective pricing mechanisms, leak detection, and rehabilitation /replacement strategies can all help in better managing an essential resource, our water.18CHAPTER 2: DEVELOPMENT OF CANADIAN WATER SUPPLY SYSTEMS2.0. IntroductionA typical urban water supply system is made up of three major components: asource, a treatment and pumping facility, and a distribution system. Sources can varyfrom groundwater wells to surface supplies such as a lake, a large impoundment or ariver. Treatment and pumping facilities vary with the type of source and the incomingwater quality, but typically include surface structures which can house a number of unitprocesses to treat raw water and make it chemically and bacteriologically safe for humanconsumption. Combinations of processes can include disinfection, clarification, filtration,taste and odour control and pH adjustment, depending largely on the raw water qualityand the standards being sought.Distribution systems carry the treated water to the consumer and consist of anumber of components, including transmission mains, distribution mains, storagefacilities, and a variety of associated appurtenances such as valves, hydrants, and serviceconnections.Although this thesis concentrates on the problems associated with Canada’s waterdistribution systems, and more specifically the “mains” in this system, it is difficult todiscuss all aspects of their development in isolation from the overall water supplysystems. In Canada, often the same funding programs which paid for the development ofthe water distribution systems also paid for the development of treatment facilities,sewers, and storm systems. In addition, some of the advances in water treatment hadenormous impacts on the distribution systems, such as the development of watersoftening which reduced capacity losses due to the build up of deposits within pipes.This chapter will first outline some of the general aspects related to Canadian19water supply systems, the history, the funding programs, and the water quality standards,and then concentrate on some of the concerns which are more specifically related to thedeterioration of the mains, such as reduced fire flows and increasing liability due tobreaks.2.1. History of Canadian Water SupplyThe distribution of water through massive engineering networks is not a modernphenomena. The great aqueducts and the intricate piped systems which survived fromthe ancient Romans are testimony to this. In North America the development of waterdistribution systems may be considerably more recent, though no less impressive. InCanadian municipalities alone there are approximately 130,500 kilometres of water pipein the ground, or enough to circle the globe 3 times. The condition, age, material typeand size of this pipe is as varied as the communities it serves and the history of itsinstallation.This section will first describe the history of municipal water systems in Canada,outlining the rationale and extent of their development. This historical perspective isparticularly important not only in setting the context for further discussion but also in therealization that many of the pipes installed in the early days of these systems are still inoperation today.2.1.1. The Early Canadian Urban CentresMunicipal waterworks and sewerage systems in Canada were developed relativelylate as urban services, but were made essential by the rapid pace of urban growth(Bloomfield et al 1983). The first Canadian waterworks systems did not appear until the20early 1800s, having been preceded by over 200 years of community development.The first urban communities in Canada started as garrison points and warehousebases for the trans Atlantic trade in fish and furs during the 17th and 18th centuries(Careless 1978). European fisherman had gathered at what is now the location of St.John’s since the early 16th century, predating any other Canadian urban centre. The firstmajor urban centres were developed by the French; Quebec City began as a staplewarehouse in 1608 and by 1750 was a substantial walled city of 7,500 while Montreal wasinitially a mission centre to inland tribes in 1642 and with the establishment ofcommercial and trade functions grew to a population of 3,500 by 1750 (Careless 1978).The British settlements followed, with a garrison centre and naval harbourestablished at Halifax in 1749. During the 18th century fur trade, the site of what is nowToronto was a minor post and warehouse, later to be established as York in 1793,eventually becoming the capital and military base of the new province of Upper Canada.It grew from 1,200 people in 1820 to 9,200 in 1834 when it was incorporated as the Cityof Toronto. Other settlements at Saint John and Kingston also emerged in the 18thcentury. Bytown, which began as a base for the building of the Rideau Canal, becamethe City of Ottawa by 1854 with a population of approximately 8,000 (Careless 1978).The Irish migration of the 1840s and early 1850s rapidly increased the size of theeastern communities. By 1850, the population of Montreal had grown to 78,000; QuebecCity to 45,000; Toronto to 31,000; Saint John to 23,000; and Halifax to 21,000 (Anderson1988). The burgeoning growth of these centres, and the problems that accompanied it,was the main impetus behind the need to develop some type of comprehensive watersupply systems.In 1850 western Canada was still largely unsettled, but the coming of the railway21in the 1880s spawned rapid growth. On the prairies, the City of Winnipeg incorporatedin 1873 with a population of 1,600, growing to 25,000 by 1891. Regina first incorporatedas a town in 1883 with a population of 900, followed by Calgary in 1884 with 506, andEdmonton in 1892 with 700 (Artibise 1981). Substantial growth in these communitiesdid not come until after the turn of the century.On the west coast, Victoria was founded by the Hudson’s Bay Company in 1843,growing from a population of 4,000 in 1864 to 17,000 by 1891 (Careless 1978).Vancouver, initially laid out as Granville in 1870, incorporated as a city in 1886, one yearbefore the arrival of the C.P.R., and by 1891 had a population of 13,000 (Careless 1978).2.1.2. The Early Development of Water WorksThere were two major factors which compelled communities to develop the firstwaterworks systems: protection from fire and preservation of health (Anderson 1988).Indeed, early communities were frequently plagued by conflagration as the main buildingmaterial was timber. Two fires in Montreal in 1852 levelled over 1,000 buildings(Anderson 1988) while a disastrous fire in St. John’s in 1892 left 11,000 people homeless(Careless 1978). Major fires also occurred in Saint John in 1877, New Westminster, B.C.in 1898, Ottawa/Hull in 1900, and Trois-Riveres in 1908 (Anderson 1988). Concern overfire protection is argued by many historians to be the prime impetus behind the eventualdevelopment of waterworks systems, with concern for public health coming much later(Anderson 1988). As J. Grove Smith wrote in 1918:“Apart from the importance of a public water supply from domestic, sanitary andindustrial standpoints, its economic value in furnishing a ready means ofcontrolling fires is unquestionable”Many systems were constructed to carry water into the centre of town expressly for fire22protection, and later expanded to provide domestic service only as funding permitted.In North America, the year 1652 marked the first recorded use of water pipes whenwooden conduits were used to carry water from wells to storage tanks in Boston, Mass.Nearly one hundred years later, in 1746, Schaeffertown, Pennsylvania became the firstcommunity to supply all its residents with water via a piped system. But by 1800 therewere still only 17 public water supplies in North America, 16 in the U.S. and one inMontreal (Grover and Zussman 1985).The Montreal system was provided by a private company, as were many of theearly systems, and consisted of wooden pipes which delivered water from springs at theback of Mount Royal to two cisterns downtown. This early system was very unreliable asthe pipes frequently burst and the system was eventually abandoned. By 1816 a new firmhad installed a 4” (100 mm) diameter cast iron main to bring water into downtown froma new source on the St. Lawrence River. Control of this system changed hands in theearly 1820s and eventually the City took it over in 1845, extending the system throughthe addition of more cast iron pipes. Still the improvements to the much overworkedsystem were inadequate, as the devastating fire of 1852 proved, and by 1856 a moreefficient public system was completed (Anderson 1988).Although Montreal had installed a marginal system as early as 1800, Saint John,N.B. is often credited with developing the first comprehensive public waterworks systemsin Canada in 1837, again provided by a private company. The system brought waterthrough a wooden duct to a steam pumping station and then through a 10” (250 mm)cast iron main to a reservoir. From the reservoir a 12” (300 mm) cast iron pipe ran to afire hydrant at Market Square. But as with many of the early systems, it was plagued bylow pressure in the cold winter months, providing only marginal fire protection. This23plus the cholera epidemic of 1854 eventually forced public control of the system in 1855(Anderson 1988).Systems followed in other eastern cities; Toronto’s system was initiated in 1841;Halifax’s in 1848; Kingston’s in 1850; Quebec City’s in 1854; and Hamilton’s in 1869.Extensive growth in the number of systems in Canada did not really occur untilthe 1870s. By 1850 there were only the three systems in Saint John, Toronto, andHalifax; by 1860 the addition of Montreal, Quebec City, and Kingston made six. Theonly system added in the decade of the 1860s was Hamilton, probably reflecting thedifficulty in importing pipes, particularly cast iron during the years of the American civilwar (Anderson 1988). After 1870, increasing development of water systems in the largecentres became substantial enough to reduce material costs, and when combined withadvances in technology, allowed system development to trickle down to the smallercommunities.Insurance companies expanded greatly between 1830 and 1880 and offered muchmore favourable rates to clients located in communities where the building codesforbade wooden structures and where professional fire departments and waterworkssystems existed (Bloomfield et al 1981). In 1888 a member of the American WaterWorks Association reported that towns with waterworks could expect 20 - 30 percent ratereductions for fire insurance and by 1900, claims were made that a system could pay foritself within five years just from savings in fire loss (Anderson 1988). As can be seen inFigure 2.1, the increase in the number of systems since then has been substantial.Fire protection remained the primary focus of water system development until thelate 19th century when the suspected link between water and disease was verified. Therelationship between pure water and public health, though suspected as early as theFigure2.1:Canada’swaterworksNumberofsystemsbyyear30002500200015001000 500105367300±ItJIII18501860187018801890190019101920193019401950196019751986Source;Mi!.1950,1960;FACE1975,1986;Anderson1988.t’a251790s, was not totally understood until the discovery of bacteria in the 1880s and theuncovering of the cholera and fecal contamination link by Dr. John Snow following acholera epidemic in London which claimed 250,000 lives between 1845 and 1849 (Groverand Zussman 1985, Anderson 1988). This discovery displaced the earlier “miasmatic”theory of disease which attributed disease to filthy urban conditions; miasma being thepoisonous atmosphere that can arise from swamps, marshes, urban gutters and streets(Baldwin 1988). In the 1850s, Dr. Snow’s new “contagion” or “germ” theory of diseasespawned the argument that a pure system of water must be paralleled by a separatesystem dedicated to the disposal of human and industrial wastes (Bloomfield et al 1981).Even after the implementation of sewage collection works, many Canadian citieswere still ravaged by typhoid and cholera epidemics throughout the 19th century and intothe 20th, as the early water systems had no means of disinfection. The problem wasworsened by the fact that many communities used common waters for both consumptionand sewage disposal, such as Toronto on Lake Ontario.Treatment was to solve this problem. Some of the rudimentary techniques ofwater treatment such as simple sedimentation were known by ancient cultures whichdiscovered that some solids can be removed from turbid water if the water is allowed tosit undisturbed. The first modern attempts at treatment began with filtration, with thefirst plant being built in Paisley, Scotland in 1804, with similar systems constructed inParis in 1806, London in 1829 and finally in North America in Poughkeepsie, New Yorkin 1871. Early attempts at filtration in Canada at Kingston in 1849 and Hamilton in1859 were less than successful, and the first operational treatment plant in Canada wasJinks Filter in Fredericton in 1891 (Grover and Zussman 1985).Probably the most effective means of treatment for disinfection was by chlorine,26which was found to destroy a wide range of pathogenic bacteria. The first large scaleuse of chlorine was in Middlekerke, Belgium in 1902 and the technology was rapidlyadapted after its first use in North America in Jersey City, New Jersey in 1908.In 1909, Toronto completed the construction of its sewage outfall on LakeOntario, 7 kilometres from the City’s water intake, a separation thought to be more thanadequate considering that the predominant lake currents were in a direction whichshould have carried the sewage away from the water intake. But the typhoid feverepidemic of 1910 proved this wrong, and the city was forced to provide chlorination of itswater. The effects of treatment were significant, reducing the death rate from 44 downto 22 per 100,000 almost immediately, and by half again after completion of a filtrationplant in 1912 (Anderson 1988). By 1918 the rate was down to only 0.9 deaths per100,000 (Grover and Zussman 1985). Montreal implemented chlorination in 1910following a similar typhoid outbreak with results similar to those in Toronto and thefuture of such treatment in Canada was sealed.Other Canadian cities followed the early leads of Toronto and Montreal in watertreatment, with Fredericton installing a rapid sand filter in 1906 and chlorination in 1931.But even with these convincing reductions in water-borne diseases, some communitiesresisted treatment, priding themselves on the purity of their water and perceivingtreatment as an unnecessary evil. In Vancouver during World War lithe U.S. Navy wascontracting for port facilities and insisted on chlorinated water. Dr. E. A. Cleveland, incharge of metropolitan Vancouver’s water system at the time, resisted pressure from thefederal government to chlorinate, claiming “No case of disease has ever been traced tothis city’s water supply.” It was only after much public debate and political pressure thatVancouver eventually did introduce chlorination, and maintains it today (Cain 1976).27Even so, not all communities in Canada have, nor require, treatment facilities tomaintain their water quality standards. In 1987, FACE surveyed the water and sewersystems in 3,650 municipalities with a total population of 22,032,162; representing about88 percent of the national population, with the remaining 12 percent being largely ruraland for the most part served by private supplies. The survey shows that only 57 percentof the communities surveyed maintain water treatment facilities, though they represent89 percent of the surveyed population, while 89 percent of the communities have waterdistribution systems, serving 97 percent of the total surveyed population. These numbersimply that the majority of communities without water treatment are very small, withmany of them located in Quebec, New Brunswick and Prince Edward Island, thoughmost still have some type of shared distribution system (see Table 2.1).Table 2.1: Water supply services in surveyed Canadian communities in 1986Percentage of Percentage of Percentage of Percentage ofCommunities Communities Population PopulationSurveyed with Surveyed with Surveyed with Surveyed withDistribution Treatment Distribution TreatmentSystems Plants Systems PlantsB.C. 93 61 100 85Alberta 89 77 100 99Saskatchewan 94 92 100 97Manitoba 70 60 97 94Ontario 96 86 100 90Quebec 69 36 94 87New Brunswick 74 24 89 62Nova Scotia 79 68 95 93P.E.I. 38 4 82 36Newfoundland 77 64 93 76N.W.T. 100 96 100 94Yukon 100 73 100 83CANADA 79 57 97 89Source: Adapted from FACE (1987)On a regional basis, the degree of treatment among Canadian municipalities is28quite variable. For instance, approximately 60 percent of the communities in B.C. andManitoba are served by water treatment, but as can be seen in the first two columns ofTable 2.2, the average number of unit processes required for treatment in Manitoba is 21/2 times that of B.C. In B.C. most water treatment plants only carry out disinfectionwith a few carrying out clarification. By comparison, in Manitoba most plants carry outdisinfection, clarification, chemical removal, and some pH control.Disinfection Clarification Removal pH Control Fluoride1.0 0.2 0.20.9 1.1 0.5 0.2 0.21.0 1.3 1.2 0.1 0.41.0 1.2 0.7 0.1 0.21.1 0.9 0.2 0.1 0.20.8 0.7 0.1 0.2 0.11.0 0.4 0.1 0.10.9 0.5 0.1 0.4 0.21.0 1.01.0 0.11.1 0.2 0.1 0.11.1 0.20.9 0.8 0.3 0.1 0.2 0.1In general, the most intensive treatment is carried out on the Prairies, whereground water supplies contain high levels of dissolved minerals and where surfacesupplies are frequently contaminated by runoff across silty, organic soils. The leastAvg. Unit Processes per Treatment Plant by TypeTable 2.2: The extent of unit treatment processes in Canada in 1986Avg. Unit Avg. UnitProcesses Processesper perCommunity TreatmentSurveyed Plant OtherB.C. 0.9 1.4Alberta 2.2 2.8Saskatchewan 3.8 4.1 0.1Manitoba 2.1 3.5 0.1Ontario 2.3 2.7 0.2Quebec 0.7 1.9 0.1NewBrunswick 0.4 1.7Nova Scotia 1.4 2.1P.E.I. 0.1 2.0Newfoundland 0.7 1.1N.W.T. 1.4 1.4Yukon 0.9 1.3CANADA 1.4 2.4Notes:“Disinfection” can be by chlorine gas, chlorine compounds, and/or ozone.“Clarification” can include filtration, sedimentation, coagulation, and/or carbon absorption.“Removal” can include aeration, iron and manganese removal, softening, and/or taste and odour control.Source: Adapted from FACE (1987)29intensive treatment is in the Atlantic provinces, in Quebec, and in B.C., where thequality of supply is good.But water treatment is not the only means of maintaining water quality. Some ofCanada’s largest urban centres maintain restricted access to watersheds and require onlyminimal treatment of their potable water; for instance, the cities of Victoria, Vancouver,Winnipeg, and Saint John still rely on impounded surface supplies to provide quality rawwater, with chlorination the only additional treatment performed (MacLaren 1985).2.1.3. The Push for Publicly Owned UtilitiesThe provision of domestic water has historically been a municipal responsibility,versus a regional, provincial, or federal one. The earliest systems in North America wereprovided by private companies, but by the early 20th century there emerged a trendtoward municipal take-overs of many of the privately owned systems.A number of factors led to this trend. It was becoming very difficult for privatecompanies to operate profitable operations while at the same time providing the level ofservice required by the expanding demands of both fire protection and domestic use. Inaddition, these companies were not compelled by law to provide either service and oftendelayed action until forced by some crisis such as fire or an epidemic (Bloomfield et al1981).A general “public ownership” movement was another factor influencing the shift tomunicipal control of utilities. Supported by a powerful rhetoric, the movement hadinitially gained appeal in the 1890s claiming that “municipal socialism” would be farcheaper and more efficient than private enterprise, citing successful models in Britainand elsewhere (Bloomfield et al 1981).30In 1882 privately controlled waterworks systems out-numbered the municipallycontrolled systems and the government of Ontario passed legislation to guide in thegranting of new franchises to private companies. But by the turn of the century, over 50percent of Ontario’s water systems were municipally initiated, while another 19 privatewaterworks had been taken over by municipalities. By 1950, 52 municipal takeovers hadbeen recorded leaving very few private waterworks companies in Ontario (Bloomfield etal 1981).Although supporting the same types of controls as the “public ownership”movement, another movement in western Canada based its arguments on the oppositeend of the ideological spectrum. In the early 1910s, municipal ownership of utilities wascompelled by the business community and its desire to keep input costs as low aspossible. At the time, a University of Toronto political economist determined thatprivately owned utilities charged up to 50 percent more than publicly owned waterworkssystems for the same level of service (Anderson 1979). In Winnipeg and other cities, thedesire for more “business-like efficiency” in the operation of the utilities was directlyrelated to plans for the restructuring of local government which incorporated manyfeatures of private business corporations (Anderson 1979).Regardless of the impetus, once a municipality had taken over control of a watersystem, it almost invariably undertook a major expansion of service, improving both thequality and reliability of the systems.The results of all these early efforts to privatize and expand can be seen today.Of the 2,923 water treatment plant operating authorities across Canada in 1986, only 200were private, 213 were provincial, and 2,510 were municipal (FACE 1987). Of theprivate authorities, most were in Quebec (185), and the remainder were in31Newfoundland (7), Manitoba (4), Nova Scotia (2), B.C. (1), and the N.W.T. (1).Today many of the larger Canadian systems have now expanded to provide waterto vast surrounding regions. In the metropolitan areas of both Vancouver and Torontofor example, regional authorities deliver treated water to a number of individualmunicipal distribution networks which in total serve 1.5 million and 2.5 million peoplerespectively.2.2. Funding for Water SystemsThe FCM (1984) report on the physical condition and funding adequacy ofCanada’s Urban Infrastructure paints a picture of dwindling funds from higher levels ofgovernment in the provision of physical infrastructure. Funding shortfalls were ranked asthe number one impediment to infrastructure renewal in the 1984 FCM survey, withinadequate staffing levels a close second, especially in the smaller communities surveyed.Except for the most recent change in policy by the federal government in the last fewmonths, the situation had not improved drastically over the past decade.The policy of the federal government had been to move away from massiveinjections of federal funds to solve local problems and to move toward improvements inlocal “user-pay” financing by offering technical support. Access to federal funding foractual water and sewer system improvements would only exist for projects that meet theobjectives of other federal programs such as economic development and job training(Environment Canada 1987, Environment Canada 1990b). The main reasons behind thisshift was the view that federal funding combined with local “flat rate” pricing schemessubsidize water systems to a point where water is artificially under-valued and over-usedby the consumer, thereby causing wastage and system over-design (Tate 1990).32Although federal funding has never risen to cover more than 35 percent of thetotal capital costs of water and sewer development in any given year, the withdrawal ofthese funds over the past decade has nevertheless resulted in a net funding shortfall forlocal works. Diminishing federal funds have been paralleled by slow or negligiblechanges in local funding and slowly declining or relatively constant funding levels fromthe provinces. The impact of federal government funding on construction, which peakedin the late 1970s can be seen in Table 2.3 and Figure 2.2.Table 2.3: Total value of construction work on water distribution systems inCanada 1973 - 1983DATE WATER MAINS, HYDRANTS, AND SERVICES ($ millions)Current Year Current Prices 1983 Prices1973 199 5101974 299 6651975 311 6231976 380 6921977 439 7741978 516 819 (Peak)1979 528 7651980 543 7051981 606 7221982 684 7351983 732Notes: 1) data includes repairs as well as new construction.2)1983 prices based on Energy, Mines and Resources Canada index.Source: Grover and Zussman (1985)Due to the reduced spending, capital works have been deferred and the backlog is nowat the point where it merits a review of the involvement both the past and the presentplayers have had in infrastructure development.cn z 0 -J -J900800700Figure2.2:TotalvalueofCanadianwaterconstructionworkondistributionsystems•77N/ /600500400 1972197419761978198019821984YEARw34Amborski and Slack (1987) outline the historical and current roles of the threelevels of government - municipal, provincial, and federal - which have been involved inthe financing of urban infrastructure. After World War II, funding for public works waskept up to date by the post-war development boom and the need for new facilitiesbrought on by a burgeoning population (FCM 1984, Amborski and Slack 1987). Allthree levels of government were very involved in the financing of infrastructure up untilthe end of the 1970s, but a combination of factors have contributed to the decline sincethen (FCM 1984):1) higher inflation in the 1970s meant higher taxes and increasedconstruction costs,2) governments provided an increasing array of services to improve thequality of life thereby increasing the burden on the tax payer andincreased competition for funds,3) interest rates soared in the early 1980s, reducing the capacity anddesire of local government to fund through debt financing,4) community and environmental concerns often made the political climatedifficult, and5) government restraint was introduced to reduce the deficit.These factors have all contributed to a slow, though steady, deterioration in facilities anda lack of an effective strategy to deal with it.Currently the funding to finance water treatment and distribution systems comesfrom two major sources:1) recurring revenues such as from user charges (tariffs) and municipal taxes, and2) local sources of capital, such as bond issues and grants from developers, orfrom provincial sources, through loans and grants.Operation and maintenance budgets come largely from the first source (Curtis 1991a),while a good share of capital cost financing comes from the second (Grover and35Zussman 1985). The FCM survey (1984) notes that most funding for operation,replacement, and rehabilitation comes from local user fees which provide 86 percent ofthe funding for water distribution systems and 83 percent for water treatment.For the past decade, prior to the recent federal election, local government haslobbied to get the higher levels of government, especially at the federal level, re-involvedin funding. Amborski and Slack (1987) outline three basic rationales which are typicallyused to justify involvement beyond the local level:1) spillover,2) fiscal equity, and3) the national character of urban infrastructure.Generally spillover is the strongest of the three and includes financing local works wherethe costs and benefits of the works “spili over” to other jurisdictions. The level ofassistance is usually proportional to the benefits which will be felt outside thejurisdiction. Unfortunately, in the case of water distribution systems, the spilloverrationale is not easily applied. Unlike road systems which can connect to form vastinter-jurisdictional networks or sewage treatment plants which can protect downstreamusers from pollution, most of the costs and benefits of water distribution systems arecontained within the municipality.Fiscal equity includes funding assistance to municipalities which would otherwisehave to bear unduly high tax rates in order to maintain some national or provincial levelof service. This is more applicable to water distribution, especially at the provincial levelsince the maintenance of water quality and distribution standards is constitutionally aprovincial concern.The final rationale applies more at the federal level and uses the argument that36infrastructure forms the backbone of the national economy. Although a popularargument, in many ways it is too generic. From the federal perspective, many otherfunctions in our society also contribute indirectly to our national productivity, yet stillremain outside the immediate jurisdiction of the federal government and direct federalsupport.2.2.1. Federal ProgramsHistorically, federal programs aimed at the provision of water distribution systems,and infrastructure in general, have neither been constant over time nor consistent inpurpose. There does not exist a precedent to fund infrastructure exclusively for thepurpose of rehabilitation and the position of the federal government through the 1980swas that is did not want to set one. While the federal Liberals now in power haveintroduced a new infrastructure program, previous federal programs have contributed tothe development of Canadian systems.It has been suggested that large scale investment in infrastructure by the federalgovernment was initiated in response to the economic depression of the 1930s(Environment Canada 1975). Many municipalities and several provinces, as well as thecountry as a whole, were experiencing severe economic problems when Canada,following the lead of the United States, provided funds for large public works projects tocreate jobs and stimulate the economy in the late 1930s. But such involvement was shortlived and despite massive urbanization immediately following World War II, the federalgovernment largely ignored the area of water supply and wastewater funding. However,concern regarding the pollution of the Great Lakes in the early 1950s again promptedfederal involvement and legislation was eventually passed which allowed for the37implementation of programs to improve water supply treatment and wastewater systems(MacLaren 1985).Although the federal government has supplied billions of dollars to aid localauthorities in the provision of their public systems, nowhere can be found a completerecord of these programs, their objectives, the actual expenditures and the resultsachieved (Grover and Zussman 1985). MacLaren (1985) notes that between 1961 and1980 the Central Mortgage and Housing Corporation (CMHC) contributed considerablecapital to the assembly and servicing of raw land for urban expansion, yet the value ofthis capital is not readily forthcoming but is known to be considerable with respect tolocal water, sewerage, and drainage facilities. Still, a partial listing of the major federalprograms and the corresponding funding is important to illustrate the nature of federalinvolvement.There were four major national programs which contributed in some part to thedevelopment of Canadian water supply systems between 1961 and 1980:1) Municipal Infrastructure Program (196 1-1978)2) Neighbourhood Improvement Program (1974-1977)3) Municipal Incentive Program (1975-1978)4) Community Services Contribution Program (1979-1980)All four of the programs involved contributions made by the federal government throughthe CMHC.The Municipal Infrastructure Program contributed a total of $2.5 billion between1961 and 1978, with $500 million as grants and $2 billion as loans (Grover and Zussman1985). The first 14 years of the program focused strictly on sewage treatment, but from1975 to 1978 funding was also provided to build new water works systems in areas that38were previously unserved. The portion of the total funding which was applicable to bothwater works and sewerage systems between 1975 and 1978 was $1.4 billion, with $395million as grants and $1 billion as loans. This program remains the historical peak infederal funding and at the time represented 35 percent of all capital expenditures onwater supply and sewerage in Canada (Grover and Zussman 1985).The Neighbourhood Improvement Program was implemented in 1974 as anamendment to the National Housing Act. It authorized CMHC to enter into anagreement with a province to assist a municipality for up to 50 percent the cost of anumber of items related to neighbourhood improvement developments, such asformulating plans and acquiring land, and up to 25 percent of the cost of improvingmunicipal and public utility services. The federal commitment over the life of theprogram was $199.5 million, with additional contributions from the provinces of $108million and from the municipalities of $184 million. The program was terminated in1978 in favour of the Community Services Contribution Program (MacLaren 1985).The Municipal Incentives Grants Program was instituted in November 1975 as anamendment to the National Housing Act. The program was implemented to encouragethe development of medium density housing in Canada by permitting CMHC tocontribute $1,000 to a municipality for each eligible housing unit constructed within itsboundaries, given the units were connected to municipal services and that certain densitylimitations were met. The program operated on a total federal contribution of $128million and was terminated in 1978, again in favour of the Community ServicesContribution Program (MacLaren 1985).The Community Services Contribution Program replaced the three previousprograms under the National Housing Act and authorized the CMHC to enter into39agreements with the provinces in order to reimburse the expenditures of municipalimprovements as set out in the agreement. The program was designed to be moreresponsive to the needs of local municipalities, while at the same time shiftingadministrative responsibilities over to the respective provinces. Each province receivedfunds as calculated by a formula based on its urban population and municipal taxes. Theprogram lasted two years, with $150 million allocated in 1979 and $250 million in 1980.Although it financed 10 percent of the national expenditures on water and sewer servicesat the time, these were not the only community services eligible for funds; a few of theothers included neighbourhood conservation and the provision or improvement of social,cultural, and recreational facilities (MacLaren 1985).By the early 1980s, all these nation-wide programs which assisted in thedevelopment of water supply systems had ceased. Figure 2.3 shows the tremendousimpact of the federal government’s withdrawal of support, and provides some insight intothe current funding crisis. The main reasons given by the federal government for thewithdrawal included the desire to hold down the deficit, the consideration that thefederal government should discontinue funding services that were the jurisdiction of theprovinces and their municipalities, and the determination that job creation money couldbe better spent elsewhere (MacLaren 1985).There have been and still are a number of smaller federal programs whichprovide funding for water and/or sewer works. The federal government conducted aWinter Works Subsidy Program through Public Works Canada in the mid 1960s toencourage winter work on municipal works, including water and sewer, thereby reducingunemployment during the colder period of the year.In 1968 the Department of Regional Economic Expansion initiated a program ofFigure2.3:Averagelevelsoffederalinfrastructurefunding140120Source:AdaptedfromFCM1984Average‘AnnualMunicipalInfrastructureFundinginCurrentDollars$2$199.5$128$400$230BillionMillionMillionMillionMillionFundinghadceasedwiththeendofthe1984fiscalyearMillion$10080 60 40 20 0II19611965197019751980Includedarethefollowingprograms:1)MuncipalInfrastructureProgram(1961-1978)2)NeighbourhoodImprovementProgram(1974-1977)3)MunicipalIncentiveProgram(1975-1978)4)CommunityServicesContributionProgram(1979-1980)5)UrbanTransportationAssistanceProgram(1978-1984)1984YEAR41grants and loans to assist municipalities through their provinces with the expansion andimprovement of their infrastructure including water and sewer systems. GeneralDevelopment Agreements with the provinces were responsible for the expansion of thewater distribution system in Saint John, N.B. and advanced water treatment in Regina(MacLaren 1985). The Prairie Farm Rehabilitation Administration has been responsiblefor the development of rural water systems in the prairie provinces, supplying technicalsupport as well as funding through grants and loans to small communities.A number of other departments within the federal government also providedfunding. Environment Canada in 1983 provided $9.5 million to various communities forwater and sewer plants, while the Department of Indian and Northern Affairs providesthe infrastructure on Indian reservations (Grover and Zussman 1985).While all the federal programs outlined were aimed at providing moneyexclusively for new capital works, the new program recently introduced by the federalLiberal government will include the rehabilitation of existing works. A report preparedby the Lands Directorate of Environment Canada (Birchan and Bond 1984) suggests thatfunding programs undertaken by the federal government should give preference toservicing land for infill, redevelopment, or revitalization where choices exist and requirecommunities to follow plans which support compact development.The Federal Water Policy (Environment Canada 1987) outlines the federalgovernment’s position throughout the 1980s and early 1990s on matters concerned withthe provision of water supply systems. The policy suggests that municipal water inCanada is currently under-valued and, in order to allow Canadians to fully realize thevalue of this scarce resource, is supporting pricing schemes based on the philosophy of“user-pay”. The means with which to achieve this are through demand management and42metering (Tate 1990), and a reduction in direct funding which tends to over-subsidize(Environment Canada 1990b).The former federal government reiterated its lack of interest in funding any largescale rehabilitation programs despite calls from the FCM for a cost shared 1/3 - 1/3 -1/3 federal, provincial, municipal program. According to the FCM, the capitalrequirements of all forms of infrastructure, if spread out over the next ten years withcosts borne entirely by the local government, would result in a 7 percent increase in totalmunicipal budgets. The cost shared program proposed by the FCM and now adopted bythe new federal government could reduce the local budget increase to only 2 to 3percent (Curtis 1987).The new found support of the federal government suggests agreement with theFCM’s view that federal involvement can be justified when new federal legislation ispassed which directly requires the upgrading of existing facilities (Curtis 1991b). In theU.S. for example, the federal government has provided some financial assistance fornecessary improvements following the passage of stricter environmental legislation. Thecurrent political position of the FCM expressed by its political Task Force onInfrastructure is to continue pressuring senior governments for funding, even thoughmunicipal governments have generally not provided their one-third share (FCM 1991b).Although the details of the new two year $2 billion federal program, with equalcontributions by the provincial and local bodies, would amount to a total nationalprogram of $6 billion, the details of the program have yet to be announced as to theextent of funding for water distribution systems. It is known that the program willinclude roads and public buildings, meaning the share of funding for water distributionsystems will be only a fraction of the total amount. While this funding program is43unprecedented, at least over the past decade, it is only a $6 billion start to solving whatmay be a $30 billion problem and, undeniably, from the federal perspective much of thereasoning behind the program is job creation rather than effective infrastructureimprovement.2.2.2. Provincial ProgramsProvincial grants and funding programs have also played a major role in thedevelopment of water systems. Theoretically the ability of provincial governments toprovide funding for infrastructure is quite large. Local municipalities often receiveunconditional grants from the provinces which may be used for the provision ofmunicipal infrastructure, with the decision of where to spend being a local one. Butsince these unconditional grants have not generally increased over the years, there arefew funds to transfer to infrastructure rehabilitation. In addition, over the past twentyyears local governments have chosen to fund more politically advantageous socialprograms rather than the necessary, though less prominent, infrastructure requirements.So even with the presence of unconditional funding there is little guarantee it will bedirected toward rehabilitation (Grover and Zussman 1985).Conditional grants to local governments aimed at infrastructure provision varyfrom province to province, but in general still represent only about 2 percent of totalrevenues for water supply and 18 to 19 percent for roads and bridges (Amborski andSlack 1987). So for now, in most provinces, the decision to fund water distributionrehabilitation remains exclusively a local responsibility.The most recent movement toward full fledged rehabilitation funding from aprovincial level has occurred in Ontario and Alberta. Following the initial survey by the44FCM in 1984, most provinces embarked on similar surveys to include those communitieswhich were missed in the national survey and thereby allow a better assessment of theirgeneral infrastructure (Curtis 1991a). But both Ontario (McIntyre and Elstad 1987) andAlberta (Grover 1990) have gone one step further and completed even more detailedprovince-wide age and condition surveys of their underground utilities, with the intentionof developing programs to encourage the rehabilitation of water main and sewerinfrastructure.In Ontario, the initial age and condition survey has been followed up by a 50percent municipal / 50 percent provincial cost shared “needs studies” program to allowindividual municipalities to identify their specific needs and appropriate courses of actionshould an actual rehabilitation construction program be implemented (McIntyre andElstad 1987). To date approximately 200 of the province’s 430 water systems havecompleted the needs studies (MacLaren 1991). The framework for the rehabilitationprogram was outlined in 1983 for the Ministry of Environment (MacLaren 1983), but todate a full-fledged funding program has not materialized.In Alberta, funding programs which allow for infrastructure rehabilitation havealready been approved. The Alberta Department of Municipal Affairs has implementedtwo unconditional grant programs to Alberta municipalities which may be used forinfrastructure upgrading. The Alberta Municipal Partnership in Local Employment(AMPLE) program which started in 1987 is providing $500 million in unconditionalgrants over a 7 or 8 year period to local government. The second program is the AlbertaPartnership Transfer (APT) program which in 1988 combined previously individualpayments for municipal assistance into a single unconditional grant and is providingapproximately $150 million per year (Grover 1990).45Alberta has also expanded the availability of its conditional provincial capitalgrants to include not only new construction, but infrastructure rebuilding as well. Thesegrants are administered on a 25 percent municipal and 75 percent provincial cost sharingbasis. The Alberta Municipal Water Supply and Sewage Treatment Assistance Programsadministered by the Department of Transportation and Utilities provided $40 million in1988/89 and $35 million in 1989/90 (Grover 1990).The programs implemented by Alberta represent the first move in Canada towarda bilateral provincial-municipal funding model which has become common in the U.S.where a variety of states are implementing similar bilateral programs rather than waitingfor federal involvement. Some of the states involved include California, Washington,New Jersey, Connecticut, Oregon, Georgia, New Mexico, Massachusetts, and Ohio(Curtis 1991b).2.2.3. Local ProgramsAs outlined previously, the bulk of funding for water distribution rehabilitationnow comes from local user-fees, with smaller amounts coming from capital worksbudgets that draw upon unconditional provincial grants or general revenues.In general, maintenance programs in Canada are not underfunded to the extentthat capital works are (Curtis 1991a). But the backlog in capital works does imply thatmaintenance dollars are probably not being spent as effectively as intended, and areoften being used to hold together systems which, from an economic point of view, shouldhave been replaced or rehabilitated years ago.“User-fees”, which in the case of water supply are administered through water rateschedules, are a double edged sword to municipalities. On the one hand, they provide46direct funding for systems thereby allowing local autonomy and less reliance on higherlevels of government; theoretically, they are easy to change as demand requires. But onthe other hand, the reality of the political process, the mood against any increased taxesor fees for municipal services, and a competitive marketplace dictates that user-fees, aswith local tax rates, must be kept as low as possible both to attract new residents andindustry and to keep the existing ones happy. The reluctance among local governmentsto increases water rates is evident in the fact that, while water systems are in bettercondition than say roads which are not funded by direct fees, there still remains a multi-billion dollar funding shortfall in water systems.Tate (1990) suggests that the extremely low municipal water rates in Canada donot reflect the true value of our water resource nor the huge investments made in ourwater systems. As can be seen in Tables 2.4 and 2.5, Canadian drinking water is indeedcheap.Tate argues that these unrealistically low prices invite problems. The inverserelation between price and water demand is well documented by Howe and Linaweaver(1967), Grima (1972), and Hanke (1978), from which Tate concludes that the undervaluing of treated water leads to increased wastage, inflated demand, and the ensuingover-design of treatment and distribution facilities.The total cost an individual consumer pays for water service depends on both thewater rate and the rate structure itself. There are two main types of water ratestructures: volume charges, which require metering of each consumer, and flat rateswhich do not. Both rate types possess characteristics which can either promote ordiscourage excessive water demand.Table 2.4: Comparison of average international water prices in 1986Country Price ($/1000L)Australia 1.65Germany .99France .75Belgium .70United States .53United Kingdom .50Sweden .50Canada .25Italy .17Note: does not include the cost of waste treatment; in Canada this would raise the price to $0.47.Source: Tate 1990Table 2.5: Typical prices for popular liquidsLiquids Cost ($/1000L)BeveragesTap Water 0.47Cola 787.00Milk 900.00Perrier Water 1,333.00Beer 2,000.00Wine 6,000.00Whisky 18,000.00Gasoline 550.00Note: Tap Water price based on the average Canadian monthly residential water price for 35,000 Litres.Source: Tate 19904748A 1986 survey of 470 municipalities by the federal government reveals that, of the 1,110rate schedules submitted, over 70 percent are such that they do not discourage excessivewater demand (Tate 1989).2.3. LegislationWhile legislation in the general field of water is extensive, with 15 major waterrelated pieces of federal legislation and 82 pieces of provincial legislation estimated in1985 (Grover and Zussman 1985), the legislation governing municipal water distributionsystems is limited.2.3.1. FederalFederal legislation impacting water distribution systems is almost non-existent.Federal involvement in the broad area of water is derived from the powers outlined inthe 1982 Constitution related to navigation and shipping (s.91 [10]), seacoast and inlandfisheries (s.91 [12]), and criminal law (s.91 [27]). On a broad basis, the federalgovernment also maintains the right to pass legislation for the peace, order, and goodgovernment of Canada.The majority of federal water legislation is aimed at the protection of receivingwaters. For instance the Fisheries Act allows for compensation of fisherman whoselivelihoods may be affected by pollution deleterious to the health of the fish resource.The Pest Control Act deals with the control of pesticides especially related to thecontamination of groundwater. The Clean Water Act is designed to allow cooperationwith the provinces in managing water resources on a nation-wide basis, allowing fornation-wide information gathering and scientific transfers. The Environmental49Contaminants Act protects human health and the environment from substances thatcould potentially be harmful and is meant to be used as residual legislation where otherprovincial or federal legislation may come up short.Probably the most significant federal function is not related to legislation orregulation, but rather entails the continuing development of drinking water qualityguidelines. These guidelines are not currently enforceable through federal law, as waterquality is a responsibility of the provinces, but most provinces have adopted theguidelines with some provinces maintaining additional standards for locally significantwater contaminants.The first federal water quality standards were instituted in 1923 to regulate thebacteriological quality of water on ships in the Great Lakes. In 1968 the first nationaldrinking water guidelines were developed through the cooperative efforts of the federaland provincial governments as well as the Canadian Public Health Association (Morrison1984). These initial guidelines covered certain microbial and physical characteristics,radio nuclides, and some inorganic and organic chemicals. Since then, improvements inanalytical chemistry resulting from the development of gas chromatographs and massspectrometers have increased detection capabilities to parts per billion and evenquadrillion (Mannion 1988). This, as well as increased information on the safety ofsubstances, has resulted in updates of the guidelines in 1978, 1987, and 1989 (Health andWelfare Canada 1989).2.3.2. Provincial LegislationThe rapid developments in the medical field of the 1850s compelled provincialgovernments to enact the first legislation which ensured the proper development of water50systems. Provincial boards of health were created in Ontario in 1882 and in Quebec andNew Brunswick in 1887. Although their powers were initially advisory, in Ontario andQuebec they were expanded such that all plans for water and sewer systems had to beapproved (Anderson 1988).Today, the Canadian Constitution of 1982 gives a province nearly exclusivecontrol over its water resource (Grover and Zussman 1985). With respect to watersupply systems, a province typically delegates some of this control to a municipalcorporation through some form of enabling legislation (ie. a Municipal Act or a PublicUtilities Act). Such legislation empowers a municipality to acquire, establish, andoperate water works for the supply of drinking water to its inhabitants. The power isoften a permissive one, but most municipalities have undertaken to supply this publicutility (Wood 1987).The province usually reserves the right to regulate water quality and to controlgeneral design and construction standards. Among the provinces in Canada there is awide array of legislative and persuasive methods of controlling drinking water quality, butas of 1985 only Quebec had entrenched in its legislation enforceable rules governing thequality of drinking water. Assuming this remains the case today, outside of Quebec aperson would have no legal recourse against a water supply utility which is providinglower quality water than specified in the national guidelines (Grover and Zussman 1985),that is, unless it can be proven that the utility was negligent in its common law duty tosupply wholesome water in which case it could be liable for damages (Wood 1987).Quebec’s Environmental Quality Act of 1984 stipulates that utilities must takeregular tests of the drinking water quality, and that the tests will be at the utilitiesexpense. Every other province has adopted some form legislation aimed at regulating51the quality and character of its environment as well, however, except for Quebec, a goodportion of the regulatory power at each province’s disposal is rarely exercised. Mostprovinces rely only on non-enforceable water quality guidelines (Grover and Zussman1985).That is not to say there is no monitoring or control of water quality. In mostprovinces, water quality tests are regularly conducted by the department responsible forpublic health or the environment. In provinces such as Ontario, public water suppliesmay be shut down by local health officials acting under public health legislation ifcommunicable disease organisms are detected in the water.Standards for new water works construction can vary among the provinces both insubstance and in purpose, with differences among provinces due to such things asclimate, material availability, water quality, and ability to pay. In B.C., the Ministry ofMunicipal Affairs has developed a set of general development standards which setguidelines for both site planning and the engineering of roads and utilities (B.C.Municipal Affairs 1980). However these standards have not been adopted intoregulations and act only as guidelines for municipalities. In Alberta however, under theClean Water Act (1988) and the accompanying Clean Water Regulations (MunicipalPlant), municipal water supply systems must now be designed to meet a variety ofmandatory requirements (Alberta Environment 1988). The mandatory requirementsrelated to water distribution design are as follows:- pipe materials must adhere to Canadian Standards Association (CSA) standardsfor potable pipe or where no standards exist for the specific material, adherenceto other standards (eg. AWWA, CGSB, NSF) may be considered by the Directorof Standards and Approvals,- all thermoplastic pipe must be manufactured from a resin conforming to CSAstandards,52- water mains designed to carry fire flows must be a minimum of 150 mm diameter,- minimum depth of cover must be 2.5 meters or more where required,- specifications for shut-off valve and air release valve locations, minimumseparation of water and sewer lines and manholes, and main disinfectionprocedures must be adhered to.The Alberta regulations also provide non-mandatory guidelines for pipeline configurationand suggested operating pressures.2.4. Factors Prompting RehabilitationReduced funding levels, combined with steady deterioration of the waterdistribution systems, has resulted in the current backlog of necessary work and hascaused many communities to become very concerned with the condition of their watersystems. A recent survey of 300 Ontario municipalities show the areas of primaryconcern are with water quality, high break frequency, poor pressure, and system leakage(see Table 2.6).Table 2.6: Areas of primary concern in Ontario’s water distribution systemsConcern Percentage of Communities with Concern as PriorityWater Quality 29.5Break Frequency 21.9Pressure 19.7Leakage 14.5Other 14.5Source: Adapted from Mcintyre and Elstad 1987There are three primary sources of the concerns in Table 2.6 which, to varying degrees,53drive a utility to consider the need for rehabilitation:1) the desire to minimize costs,2) the desire to maintain adequate levels of service, and3) the desire to satisfy public demands and minimize public concerns.The concerns with breaks and leakage are related primarily to the desire to keepdown costs due to repairs, liability, and lost water. The concern over reduced pressureand flow is closely related to the desire to maintain service standards related to domesticuse and fire flows. The desire to maintain water quality is aimed at instilling confidencein the public regarding health concerns.Frequently the desire to maintain the cost and service level criteria areoutweighed by public concern, which can rapidly be transformed into action via thepolitical process. This is probably the most variable factor affecting a utility and thereare a number of indicators that it may becoming more important. Today’s concern withthe health, the environment, and economic development all factor into the decisionprocess, even when there are only perceived effects which are not validated by conclusiveevidence, as is the case with the debatable concern over asbestos-cement pipe.The following section will briefly outline some of these concerns to allow agreater appreciation for today’s situation which in many ways is much different than thatwhich existed when many of the systems were installed.2.4.1. Increasing Maintenance and Operation CostsDeterioration of a piped system is commonly measured in terms of themaintenance required to repair pipe failures, which are usually in the form of leaks or54breaks. While all piped systems, regardless of age, experience such failures, unusuallyhigh failure rates and increasingly expensive maintenance costs indicate the need forrehabilitation. McIntyre and Elstad (1987) and MacLaren (1983) both estimate theaverage cost of a break in Ontario at $3,000.Before continuing on, it is important to introduce a point of confusion common tothe literature and statistics dealing with pipe failure. While it may seem intuitivelyobvious what a break and a leak are, there are a number of interpretations in theliterature. For instance, Moruzzi (1987) defines a break, which he refers to as a“failure”, as a macroscopic discontinuity which interrupts the regular flow of a pipe, whileleakage refers to damage which causes a loss of water without interruption of the flow.Leaks and breaks are then mutually independent, meaning a particular system couldhave a large number of breaks, yet little leakage and visa-versa. In contrast, the AWWA(1986a) defines a break as a special type of leak, with “leak” being a more universal termfor pipe failure. This incongruence pervades much of the literature and the statisticsregarding leaks and breaks; in many cases the definition being applied is notforthcoming. The FCM (1984) estimated the historical increase in breakage rates inCanada, and although not explicitly stated, probably based the definition of a “break” ona repair being carried out due to the noticeable presence of water on the surface, be itfrom a leak or a break.From 1968 to 1983, the FCM (1984) estimates the average breakage rate inCanadian systems nearly doubled, increasing from 16.6 breaks per 100 kilometres to 30.4.This is comparable to the 25 breaks per 100 kilometres per year now experienced in theU.S. and Britain, where the excessive deterioration of water systems have already beenrecognized (FCM 1984).55During this same period, emergency and scheduled repairs in Canada wereestimated to have increased by 27 percent, from just under $5 per capita toapproximately $6.25, while per capita pipe replacement expenditures only increased by15 percent, from $5.70 to $6.56. Also during this period, average operation andmaintenance expenditures increased by approximately 42 percent, from just over $7 percapita to $10. All figures were based on 1983 dollars and utilized the ENR index forcomparison purposes.While it is tempting make generalizations from these types of figure, it must berealized there is a great deal of variance among communities. McIntyre and Elstad(1987) found the current average annual break rate in Ontario to be approximately 25breaks per 100 kilometres, which is close to the national average, but of the 436communities supplying system length data, 111 were experiencing no significant breakageproblems. Of the remaining 325 communities, only 47 percent were experiencingbreakage rates above the provincial average. Thus, while breakage problems aresignificant, they are often concentrated in a relatively small number of communities.It is interesting to note that of the 35 Ontario communities which experiencedbreakage rates in excess of 100 breaks per 100 kilometres per year, 34 were in smallcommunities under 7,500 in population. McIntyre and Elstad (1987) feel that thisindicates a large number of small systems are located in adverse conditions which leadsto a disproportionately high share of the local resources being spent on the almostconstant repair of the water distribution system.The increase in break repairs is not the only source of increased expenditures.Excessive leakage from deteriorating systems means pumping energy, treatmentchemicals, and raw water are wasted. Tate and Lacelle (1987) estimate that 12.4 million56cubic meters (MCM) flowed through municipal systems on an average day in 1986, with26 percent of the volume not accounted for. Of this, they estimate 20 percent isattributable to leakage or other waste and represents 240 million cubic meters per year,which is enough to fill up a 4 meter square by 1.5 meter deep swimming pool behindevery household in Canada.Dealing with the leakage problem is difficult due to the inability to easily detectleaks. Over time, even the smallest leaks can lose enormous amounts of water withoutany noticeable signs on the surface. The fact the pipes are buried hides much of theloss. To illustrate the scale of leakage that can occur and can be detected when thepipes are easily accessible, take the case of Paris, France presented by Versanne (1987).The Paris water system is unique in that it consists of two parallel systems, a drinkingwater system with 1,800 kilometres of pipe, and a non-drinking water system with 1,600kilometres of pipe. The system is also unique in that 96 percent of the pipes areaccessible, being located either in special water main galleries or sewer tunnels, and leakinspection can be carried out visually. The breakage rate on the drinking water systemsreported by Versanne (1987) is equivalent to the Canadian average of 30 breaks per 100kilometres per year, yet the system develops an average of 15,000 leaks per year whichtranslates into 833 “leaks” per 100 kilometres per year. While it must be realized thatonly 10 percent of the Paris system was installed since 1955, compared to 70 percent inCanada, this case still emphasizes the scale of the problem that can potentially develop.While there is little historical data to suggest the leakage problem has beenworsening in Canada, there is evidence which suggests today’s leakage from municipalwater systems may is too high. Part of the information problem is rooted in the presentattitude toward leakage. Hennigar (1984) notes that while the AWWA suggests system57leakage in the order of 10 to 15 percent is reasonable, in Europe, after extensive leakagesurveys, rates of 5 percent are now considered high. Outside of the municipal waterindustry, pipeline leakage rates are also much lower. The Energy ResourcesConservation Board, which is responsible for monitoring over 204,000 kilometres of gas,oil, and water lines used by the energy industry in Alberta, reports average water lineleakage rates of only 1 per 100 kilometres per year, and gas lines leakage rates of only0.2 per 100 kilometres per year (ERCB 1983). Leakage as defined by the ERCBincludes both ruptures (ie. breaks) and leaks.The third significant result of system deterioration is capacity loss. This istypically due to corrosion products on the interior of the pipe wall or increases indemand needed to meet fire flows or system expansion. In any case, the typicaloperational response is to increase the flow and pressure through increases in pumpingcapacity. Such action can increase pumping costs as both increased energy is consumedand increased capital may be required. In addition, further stresses due to increasedpressures are placed on existing distribution systems.2.4.2. Fire Protection and LiabilityWood (1987) reviews some of the legal aspects related to the provision of watersupply systems as they apply to Ontario law. If a municipality in Ontario exercises itsright to establish a water system, it has a statutory duty to supply an uninterrupted supplyof water to all of its customers, it has a common law duty to supply a “wholesome andpure supply of water to the consumer”, and it may at times be required to improve itswater supply system in accordance with certain water resource statutes. Failing tocomply with any of these duties can result in the risk of liability (Wood 1987). Although58the points outlined specifically apply to Ontario, they do illustrate principles common toother parts of Canada.The recent trend in North America toward increased litigation has not left publicutilities untouched. Lawsuits resulting from water distribution system operation havehistorically resulted from flood damage due to main breaks, though more recently aclaim has succeeded against a municipality for failing to provide adequate fire protection.The potential for liability due to flooding damage is a key determinant in themaintenance strategy of some utilities. As Bratton et al (1986) notes:“In recent years, the courts in British Columbia have held the water utilities responsibleunder nuisance law for all private property damage. This imposition of absolute liabilitynow requires that the decision to replace or rehabilitate a water main must consider theprobability of a major damage claim with a water main failure. To wait for thedevelopment of a failure history upon which to base a replacement decision could resultin a utility facing a multi-million dollar lawsuit.”The increasing values of real estate, especially in downtown cores, makes theconsequences of such liability even more pronounced. Bratton et al (1986) estimatesthat a water main break in Vancouver’s downtown business district will result in $48,500worth of damage when it is on a high pressure main (100 to 150 psi), and $21,000 whenon a low pressure main (40 to 70 psi). Conversely, a break in a residential area willresult in an average of $6,000 worth of damage regardless of the pressure. All of thesevalues are calculated using a weighted average based on the probability of occurrence ofspecific damages and historical data using a 300 mm diameter pipe.The provision of fire protection is an aspect of water distribution that is solely thedecision of the local authority. However, once a municipality makes a decision toprovide fire protection, it is now obligated to provide reasonable service or face potentialliability. A 1988 decision of Supreme Court of Canada (Laurentide Motels v. City of59Beauport) deemed the City of Beauport, Quebec (pop. 60,000) liable for $2.5 million inadditional fire damages to a hotel caused by an inoperable fire hydrant. In the court’sconclusions, L’Heureux-Dube J. (SCC 1989) commented,“In the case at bar the city of Beauport exercised its discretionary power to createa firefighting service (a policy decision) and, in light of the by-laws adopted in thisregard and the other facts presented in evidence, the municipality undertook, atthe very least by implication, to maintain this service and ensure that it was ingood working order (an operational decision).”“...the fact that a municipal corporation makes a policy decision or refuses to doso does not entail its civil liability. If, however, the municipal corporationexercises its powers, discretionary or otherwise, so as to make its decisionoperational, subject to public law, it can be held liable for any damage caused toanother through its fault, or through that of its employees in the course of theirduties, unless the enabling legislation expressly excludes such liability orauthorizes the municipal corporation to exonerate itself from liability.”While the liability regarding the proper provision of fire protection has been lesspronounced, there exists the potential for high liability costs barring any changes inlegislation. Excluding farm and vehicle fires, in 1987 there was approximately $821million in property losses due to 41,405 structural fires in areas potentially served bymunicipal water systems (IBC 1989). This translates into an average loss of almost$20,000 per fire.The recent movement by the courts regarding such liability have become a sourceof concern for many Canadian communities. In many communities, the distributionsystems do not meet the fire flow requirements as set out by the Fire UnderwritersSurvey, partially due to deterioration and partially due to the presence of small diametermains (FUS 1981).A survey of 12 large (> 100,000), 15 medium (20,000-50,000), and 35 small (<10,000) Canadian communities reported on by Wareham and McBean (1985) concludes60approximately half of the communities have difficulties meeting the FUS requirements.The problems are most pronounced in the small and medium sized communities, ofwhich an estimated 33 percent are expected to have future difficulties. In especially thesmall communities, relaxation of the FUS guidelines is common. Dead-end hydrants andinadequate fire flows due to small diameter mains are typical. Overall, the three mostcommon reasons for not meeting the requirements are:1) insufficient additional storage,2) the continued replacement of old buildings with larger, new buildings withoutreplacement of the old mains,3) small diameter mains in the distribution network.Wareham and McBean (1985) conclude that the problem is not easily solved. There isoften little incentive on the part of a community to upgrade partly due to the slowness ofinsurance companies to compensate or reduce premiums in accordance with acommunity’s endeavour to meet the FUS requirements.Still, the inability of a system to provide adequate fire flows is only oneconsideration with respect to fire protection. System reliability in the event of anemergency is also usually not well known, though there have been recent developmentsin reliability analysis over the past few years to improve this situation (ASCE 1989). Thepresence of deteriorated mains which are susceptible to collapse from negative pressuresduring high flow fire events remains a factor in the provision of service.The FCM (1991a) also suggests that for communities located in seismic areas, theability to fight widespread fires and provide necessary drinking water after an earthquakeis compromised by the presence of old outdated cast iron water mains which are subjectto damage. Cast iron mains are very brittle and have only one-tenth the ability to61withstand earthquake shocks (FCM 1991a), yet are prevalent in many Canadian systems.2.4.3. Health ConcernsWater is the ultimate solvent and as such it can effectively dissolve substancesboth beneficial and deleterious to human health. As Morrison (1984) notes, whenassessing drinking water, one must keep in mind “pure” water really means ‘wholesome”water, and that all drinking water contains chemicals and ions that actually make it tastegood, rather than flat and insipid like distilled water. When considering the implicationsof a toxic substance in drinking water as it relates to human health, what is important isnot only the presence of the substance, but its concentration. In setting water qualitystandards, the maximum allowable concentrations of such substances are based onlifetime exposures with average daily consumption using experimental data and appliedfactors of safety for risk control.2.4.3.1. Water Quality StandardsThe responsibility of the public health community is typically threefold: firstly, forassessing the risk associated with potentially harmful substances; secondly, fordetermining the maximum allowable concentrations which can be taken in by a humanwhile still maintaining good health; and finally, recommending standards whichincorporate the results. In theory, the process is straightforward and scientific, but anumber of factors make the process less than ideal.In past 25 years, the development of mass spectrometers and gas chromatographyhave rapidly advanced the science of analytical chemistry. Today, not only can newsubstances in water be detected which were undetectable in the 1950s and 1960s, but the62substances can be detected at incredibly low concentrations of parts per billion and partsper quadrillion. However, many authors have stressed that analytical chemistry is stillrunning orders of magnitude ahead of the ability of toxicologists and epidemiologists todetermine the consequences of exposure to such minute quantities (Cotruvo 1984,Morrison 1984, Mannion 1988). Toxicology has strict limitations in that it simplifies thereal world, looking at one chemical in isolation and its affects upon one animal, usually alaboratory mouse. Standard setting is based largely upon a health agency taking thisoften marginal information and extrapolating it to determine the impact of life-longexposure on humans (Hall 1984). In the cases where there is little conclusive data,agencies typically try to err on the safe side and draft conservative standards. As McColl(1985) points out, factors of safety can range from 2 to 4 for most inorganic toxins to 500to 1000 for non-carcinogenic organic toxins (such as pesticides).The Canadian drinking water standards are exceptional in that they are moreextensive than those of many other countries, including the U.S., and the sampling ofwater supplies is more frequent (McColl 1985). One difference between American andCanadian standards which has a direct impact on the rehabilitation of piping systems isin the level set for the presence of asbestos fibres, which can originate from the leachingof asbestos cement pipes by soft or pH depressed waters. Studies on the occupationalhazards associated with breathing in asbestos fibres are quite conclusive, yet there islimited information on the cancer-causing effects of ingestion (Hickman 1984).Canadian water quality guidelines do not include a provision for asbestos; as Health andWelfare Canada (1989) point out, the “assessment of data indicates no need to set anumerical limit.” In the United States the EPA has proposed a recommended maximumcontaminants level (RMCL) of 7.1 million fibres per litre for asbestos fibres longer than6310 urn, but this has proven to be quite controversial, as the limits seem to have beenbased only on the proven carcinogenicity of inhaled fibres (AWWA 1986b).A number of other substances directly related to pipe systems are or may soon beincluded in the Canadian guidelines. Vinyl chloride, a constituent of P.V.C. pipe, iscurrently under consideration to be added to the guidelines. The Canadian guidelinesalso make note that where lead is still a component of many plumbing systems, faucetsshould be thoroughly flushed prior to use.Still, while there is concern over piping materials and their impact on humanhealth, for the most part concern is limited and the public perception is that the dangeris negligible.2.4.3.2. Sources and Impact of ContaminantsAlthough a number of contaminants are identified in the water quality guidelines,the guidelines are not enforceable and in some communities the contaminant levels maybe and often are exceeded. It is difficult to know whether this is causing serious healtheffects for a number of reasons: first, in Canada there is a perception that water bornediseases and the ill effects of contamination are at extremely low levels, but Grover andZussman (1985) note that reported deficiencies in the national reporting systemresponsible for detecting such effects have led many health officials to believe a largenumber of local incidents go unreported. Secondly, drinking water is not the solecontributor of contaminants to the human body as food, air, and even absorption throughthe skin also contribute. Thus drinking water cannot be looked at in isolation as it iscontributing only a portion of the total amount of total daily intake of certain substances;for instance drinking water contributes an estimated 45 percent of a Canadian’s total64daily intake of trihalomethane (THM), but only 3 percent of the mercury, 1.2 percent ofthe lead, and 0.06 percent of the cadmium (Hickman 1985). Finally, doses ofcontaminants and their effects on the population are highly variable. For instance,approximately two percent of Canadians drink more than two times the national averageintake of 1.34 litres per day, which in effect means the dose they are receiving hasincreased by two, or similarly the factor of safety used in setting the drinking waterguidelines has been reduced by one half. On a dose per kilogram of body weight basis,small children are more at risk than adults, who also have the added benefit of drinkinga large proportion of their water boiled (in tea or coffee) which further reduces thecontent of volatile organics (Hickman 1984).The impact on health of deteriorating distribution systems is usuallyovershadowed by the concern over quality of the raw water supply sources. Recent newsarticles have focused on the contamination of the Great Lakes which supply major citiessuch as Toronto with drinking water and which are known to contain levels of mercury,lead, pesticides, PCB’s and other toxic substances in concentrations that exceed both U.S.and Canadian water quality objectives (Macleans 1990). But contaminants do notexclusively originate in the raw water source.Hickman (1984) outlines the three main sources of chemicals commonly occurringin drinking water:1) substances affecting the source quality (raw water),2) substances resulting from treatment,3) substances arising from the distribution and service systems.The substances affecting the source can include: naturally occurring substancesleached from soils, sediments, or geological formations (e.g. calcium, heavy metals),65pollutants derived from point sources such as domestic sewage treatment, industrialeffluent, or landfills (e.g. synthetic organics, metals, cyanide), and pollutants derived fromnon-point sources such as agricultural run-off, urban runoff, and atmospheric fall-out(e.g. fertilizers, pesticides, salt, chlorinated organics, PAH’s).Substances resulting from treatment can include those formed during chlorinedisinfection (e.g. trihalomethanes, chiorophenols) plus treatment chemicals (e.g.chloramines, fluorides) and their impurities (e.g. acrylamide monomer, carbontetrachloride).The third source of drinking water contaminants is the most relevant to thisdiscussion. Substances which arise from the distribution and service systems includecontaminants arising from contact with constructional materials and protective coatings(e.g. lead, vinyl chloride monomer, asbestos fibres from piping, cadmium from fittings,PAH’s from coal tar linings) as well as from point-of-use devices, such as home carbonfilters (e.g. sodium, silver).In Canada, though federal water quality guidelines do not include a provision forasbestos, there still exists concern over the health effects of it. Considering there is asmuch as 13,000 kilometres of A.C. pipe in Canada, the implications of more stringentstandards are broad and expensive.Studies have been carried out in Canada using the U.S. EPA standard of 7.1million fibres per litre (MFL) as a basis for discussion. Hickman (1984) notes thatWinnipeg, which has both aggressive waters and substantial amounts of A.C. pipe, hastaken samples from the distribution system containing 6.5 MFL, while samples at thetreatment plant only contain 0.3 to 0.4 MFL, indicating a substantial amount of the fibresare from the distribution system. Hunsinger et al. (1989) points out that asbestos fibres66are also found in natural waters, which can be removed by conventional water treatment,and of 268 surveyed Ontario communities, only one percent of the samples exhibitedlevels greater than 7.1 MFL for fibres of all lengths, and none had long (> 10 um) inconcentrations exceeding 7.1 MFL. Only three supplies showed treated water as havingmore than 7.1 MFL of all sizes, with the chance of encountering fibres longer than 10urn at these locations calculated to be less than 0.01 percent.Metal pipes can also introduce substances into treated water. Lead pipes andlead joints, though discontinued for use several decades ago, are especially prone toleaching in the presence of soft, aggressive waters and still do exist in some older urbancores (Morrison 1984). Copper and galvanized steel services have also been documentedto contribute excessive lead through the deterioration of solder joints, especially in thepresence of aggressive waters where protective calcium carbonate films can not form,and especially when the water stands for long periods of time in the pipes (McClelland1981, Hickman 1984). Millette and Mavinic (1988) found copper and iron concentrationswhich exceeded the national guidelines by 100 to 300 percent in a number of locations inVancouver, which has an aggressive, soft water with little buffering capacity.Plastics can also contribute to the contamination of drinking water throughleaching, though Hickman (1984) notes that plastic pipes certified by the NationalSanitation Foundation (NSF) are not problematic. McClelland (1981) provides anexcellent overview of the standards and testing procedures used by the NSF and stressesthe fact that plastic pipes are highly regulated and tested, especially when compared topipes of other material (e.g. metal). The NSF, through a voluntary testing program withthe manufacturers, checks for compliance of P.V.C. pipes regarding content of certainmetals (e.g. antimony, arsenic, barium, cadmium, chromium, lead, mercury, selenium,67and tin) and residual vinyl chloride monomers (RVCM) which can be transferred towater in use. In Canada, the CSA (Canadian Standards Association) provides standardscomparable to those provided by the NSF in the U.S.Bacteria can sometimes thrive in water distribution systems. Pathogens such asthose responsible for Legionnaires disease have been found in Canadian water systems(Grover and Zussman 1985). Pathogenic coliform can reproduce and migrate indistribution systems especially if the residual chlorine is low, turbidity is high, or scalingand tuberculation are high (AWWA 1988). Generally, such bacteria are present due totwo mechanisms: breakthrough and growth (AWWA 1988). Breakthrough refers tothose bacteria which pass through the disinfection process, while growth refers to theincrease in bacteria due to their continued growth downstream of the treatment plant.Models aimed at predicting the chlorine residual in distribution systems are beingdeveloped to combat this problem (AWWA 1988).Growing concern over the quality of tap water has proved a boon to the bottledwater industry, which had sales of $150 million in 1989. Recent tests have shown thatsome bottled waters, which are not as stringently regulated as tap water, contain highlevels of barium and bacteria which probably resulted from the contamination of thesource, which is typically a groundwater “spring” (Macleans 1990). As of January 1990,only Quebec had in place regulations for bottled water quality (Macleans 1990).An additional response by consumers has been to purchase point-of-use watertreatment devices, but there are a number of concerns with these devices as wellincluding the accumulation and concentration of metals and pathogenic bacteria in filters(especially ones which are infrequently changed or operating on raw water) and the falseclaims made by sales people as to the effectiveness of such filters (Grover and Zussman681985, Hickman 1984, Macleans 1990). In 1984, an estimated 3 percent of Canadianhomes had point-of-use filters, with half being activated carbon (Grover and Zussman1985).2.4.4. Impending Changes in the RulesVarious levels of government are looking to change some of the existing practicesdealing with water supply in Canada. The federal government is looking to assure waterquality across the country through the introduction of mandatory federal regulations.Local governments concerned with rising liability claims are looking to developconsistent benchmark standards to measure maintenance against. Such changes couldhave a potentially huge impacts within the next five to ten years and merit somediscussion here.Although the quality of drinking water has been traditionally a provincialresponsibility, the federal government through the Department of Health and Welfare ispreparing to introduce legislation for a Canada Drinking Water Safety Act. CurrentlyCanada remains one of the few developed nations without a federal law requiringminimum drinking water quality standards (Environment Canada 1990b).In the U.S. for example, the Safe Water Drinking Act of 1974 set out “toconstitute a cooperative venture among federal, state and local levels of government, byrequiring the establishment of consumer oriented drinking water regulations” (Groverand Zussman 1985). However, implementation of the Act was not without its politicaland administrative problems. While the Act provided for funds to the States to helpcover the administrative costs of implementing the new legislation, there was no moneyset aside to help utilities with the needed improvements to comply. This problems was69particularly acute among small communities which inherently have a very limited taxbase; the economies of scale associated with water treatment meant the smallcommunities had to pay considerably more per capita than the larger centres.In Canada there is no national “cod&’ regulating the design or maintenance ofwater works. The FCM (1989) has recognized the need for standards in Canada to helpprotect municipalities from the resulting risks based on the following rationale:“... having standards in any given municipality has the advantage that, in case of claimsinvolving litigation, the municipality itself will have a set the standards, whereas, in theabsence of standards, a judge may do so, and he may choose a far higher and moreexpensive level of service as his standards. Our advice is that usually courts acceptstandards chosen by municipalities, providing they are reasonable and consistent. Itappears that it is not necessary that they be common across the entire nation, as courtsrecognize that some municipalities have a lower ability to pay for service levels thanothers. Standards must be adopted politically and not merely be technical regulations.”The FCM’s intent is to develop a set of national standards that are stringent enough tosatisfy the courts, yet flexible enough to allow all municipalities to adopt the standards,regardless of their ability to pay. For this reason the standards will be somewhatgeneral, with ranges of service levels based on established practice, rather than on theimposition of strict standards.The FCM is currently developing design and service level standards for sewercollection infrastructure and plans to start development on similar standards for watersystems (Curtis 1991a, FCM 1991b). The FCM hopes the standards, which are based ona survey of nine member municipalities, will lead to the wider adoption by other membermunicipalities and the eventual endorsement of them by the Canadian Public WorksAssociation (FCM 1991b).702.4.5. Environmental ConcernsConcern for the environment and the impact of development on it is growing.Conservation was the key in the 1960s and 1970s. Today, “sustainable development”, aterm introduced by the United Nations in 1980, is entering many facets of our lives.Sustainable development can be summarized as (IUCN 1980):“the modification of the biosphere and the application of human, financial, living andnon-living resources to satisfy human needs and improve the quality of human life. Fordevelopment to be sustainable it must take account of social and ecological factors, aswell as economic ones; of the living and non-living resource base; and of the long termas well as the short term advantages and disadvantages of alternative actions”A number of authors suggest that “water demand management” promotes andencourages the objectives of sustainable development (Postel 1985, Tate 1990). Brooksand Peters (1988) define water demand management as “any measure which reduces orreschedules average or peak withdrawals from surface or ground water sources whilemaintaining or mitigating the extent to which return flows are degraded”.Postel (1985) suggests that demand management in water systems can save notonly through the reduced demand for the water resource, but also through a reduction inthe resources necessary to construct such systems. In the past, planners have tended toproject historical per capita water demands to future population forecasts, rather thantrying to focus on reducing demand over the long term. Postel (1985) suggests thatreduced demands through more rational pricing can reduce the system capacitiesrequired, and thus the total cost of the systems.Tate (1990) outlines the three basic techniques available for water demandmanagement: economic techniques, structural and operational techniques, and sociopolitical techniques. These techniques will be discussed in more detail in later sections,71but all basically involve reducing excess demand through the elimination of subsidies, thereduction in system water losses, and through the promotion of conservation.The potential application of such techniques to municipal water supplies is great;in Canada, municipal systems use over 12 million cubic metres of water, constituting 16percent of the total Canadian water draw, with at least 26 percent of this flowunaccounted for through system leakage, waste, discharges directly to return streams, fireflows discharged directly to storm sewers and to evaporation from lawns and gardens(Tate and Lacelle 1987, Tate 1990).But the current concern for sustainable development can also have indirect effectson water distribution system development. Sustainability concepts promote changes inland use toward denser, more compact urban forms. Sustainability has also been used inarguments for tighter industrial emission controls which can affect both the atmosphereand eventually the quality and pH of the water source through acid rain.Control of acid rain has been a prominent global issue in the past few years. Theinternal corrosion problems of water distribution systems can worsen if acid raindepresses the pH of the supply. Communities situated on the Canadian shield and thecoasts are especially prone since the buffering capacity is very low for lakes andcatchment areas situated on granite and igneous formations. Vancouver for instance hasan average pH well below the Canadian recommended range of 6.5 to 8.5 and has beenexperiencing problems with corrosion of its copper services and cast iron mains (Milletteand Mavinic 1987). A number of other communities in greater Vancouver are alsoexperiencing wall softening and high breakage rates among their asbestos cement pipesdue to the aggressive nature of the water (Robinson 1991, MacLean 1991).722.4.6. Economic Development ConsiderationsThe provision of efficient and cost-effective infrastructure is necessary to attractinvestment in any community both from public health and economic perspectives. Intoday’s global marketplace, there is increasing competition among countries, regions, andmunicipalities to attract new industry which may result in new jobs and new prosperity.In the U.S., Williams (1984) points out the importance of municipal water systems toindustry,according to a U.S. Commerce Department survey, the most criticalcommunity attribute that industry looks for in deciding where to relocate orexpand is the availability of fire protection. That means the most important asseta community can develop is a reliable water system capable of delivering water ata sufficient pressure and flow. This is much more valuable and important toindustry than tax incentives are. Yet Sates and localities continually compete withone another to ruin their tax bases.ttWilliams (1984) notes that in the same U.S. Commerce Department study, taxeswere only one of many variables which influenced industry location, and in many casesmade little difference in the final relocation or expansion decision. Costs of fireinsurance and plant construction can be directly related to the community’s ability toprovide fire protection and the ability of the community to provide quality, reliablepublic services.But the provision of quality services to attract development is not the onlycontributor which can enhance a community’s prosperity. Development andimprovement of infrastructure is labor intensive, and can generate much neededemployment and the benefits of this. Studies have indicated that 35 to 40 person-yearsof employment and one quarter of a million dollars in expanded local business can resultfrom each million dollar investment in water supply or sewer systems (Environment73Canada 1983).Investment and financing in infrastructure are not only beneficial from anemployment and economic perspective, but in many ways represent the heart of thecurrent problems and the required solutions. According to Williams (1984), theinfrastructure crisis is basically a financing problem. The current situation has not comeabout due to technological short-comings nor design or construction failures, but ratherhas developed through unfortunate political and fiscal decision-making. In attempting tosolve the current dilemma in New York State, Williams makes five basic assumptionswhich can be extended to many other geo-political areas:1) taxpayers must not have to pay higher taxes,2) taxpayers can not be subject to further debt liability due to unforeseencircumstances such as municipal default,3) there is no reason a person living in one city should have to pay for repairs inanother city,4) the solution must work within the given constitutional framework, and5) revenues generated by a utility should stay with the utility and not go to supportother general municipal services.While any one of the assumptions may be challenged or debated, together they do forma basis for discussion required in developing a policy framework. Williams concludesthat one solution lies in the creation of a State finance authority set up by legislation toprovide bonding capacity to the local municipalities and to request loans and grants fromfederal programs on behalf of the local municipalities. Three financial options were putforth including loans, leaseback, and a revenue bond option.Although each of the financial and political options just outlined will have adifferent degree of applicability within the Canadian context, one important point is74brought forth: cooperation and coordination from all levels of government and financialinstitutions within the political context of each region is one key to overcoming theproblem and enhancing the communities affected. The nature of the problem is diverse,and so should be the solution.2.4.7. Public ConcernWhile all of the preceding discussions outline the concerns specific to the need forwater system rehabilitation, it must be recognized that the overall importance of theseissues to the public will be the main factor which determines whether or not the politicalwill is there to get something done. Unfortunately, information gauging public opinionson these issues is very limited.Grover and Zussman (1985) noted that while environmental health concerns wereamong the top ten issues covered by the Canadian media, only a fraction of the coverageconcentrated on water pollution matters, and even less on drinking water issues. ADecima survey conducted for Health and Welfare Canada at the time placed concern forhazards from tap water as very low compared to three other categories which includedfood additives, air pollution, and pesticides. These findings led Grover and Zussman toconclude that while the long term interest in drinking water issues is quite low, whenproblems do arise Canadians demand immediate and remedial action. This conclusion isechoed by Anderson (1988) who postulates that the only way to draw attention to watersupply issues is through crisis, yet once the crisis has passed, water supply again becomesa non-issue.More recently there has been growing concern over drinking water quality inCanada, though much of the concern has been related to specific environmental issues75such as the toxins in Lake Ontario. Still, according to a Gallup poii published inOctober 1989, 95 percent of those polled were concerned with the quality of drinkingwater in Canada (Macleans 1990).How effectively this concern is being transformed into political action aimed atthe improvement of water supply infrastructure has yet to be seen. Infrastructurerehabilitation was not a major issue in the 1988 federal election, as the FCM had hopedit could be, but did enter the 1993 election campaign on the basis of job creation ratherthan any deep concerns over health, the environment, or industrial need. Since theFCM’s report in 1984, conferences and briefs have increased public awareness of theproblem, but in the overall scope of day to day life, public concern over infrastructure isstill not great. Indeed, the public sentiment over the proposed infrastructure programannounced by the federal Liberals during the election campaign was one of scepticism,with the perception that the debt-burdened federal government would only be buyingjobs with more borrowed money. Little attention was paid to the benefits that such aprogram might have on health or industry within the average Canadian community, anindication that the education gap is still quite wide.76CHAPTER 3: CANADIAN WATER DISTRIBUTION SYSTEMS INVENTORY3.0. OverviewThis chapter provides a look at the “pipe” beneath Canadian municipalities,representing a current “best estimate” inventory based on data which was accumulatedfrom various sources and transferred by the author onto a spreadsheet program.Unlike water treatment systems, there is no national inventory of the waterdistribution works within Canada even though the distribution system typically representsnearly 80 percent of a water system’s value (Edwards and Cox 1981, MacLaren 1985).Heinke and Bowering (1984) estimate that in 1981 there were 100,000 km ofwater main in Canada based on a served population of 20.3 million (FACE 1981) and anaverage “service density” of 200 persons per kilometre of watermain, a value extractedfrom a report to the Ontario Ministry of the Environment (MacLaren 1983).The need for a comprehensive inventory of Canada’s underground plant has beennoted by a variety of sources (FCM 1984, MacLaren 1985, MacLaren 1991) but to dateno agency keeps such records on a national scale. It is very difficult to discuss theefficient management of water systems in Canada without some basic information on thesystems: how much pipe is in the ground? how old is it? what types and sizes of pipe arepredominant? what are the soil conditions? what are the water conditions? whatoperating pressures are common? Unfortunately in Canada not even the first questioncan be answered with a large degree of accuracy. Although a variety of informationexists within individual municipalities and some provinces, no agency has yet to compilea comprehensive national data base, although the 1984 FCM survey was a first step(FCM 1984).In England, a comprehensive survey was carried out in 1977 from which courses77of action could be taken (DoE and NWC 1977). Similar surveys in Canada have beencarried out at a provincial level in Alberta (Grover 1990) and Ontario (McIntyre andElstad 1987). The U.S. is currently basing much of its rehabilitation policy on systemage (FCM 1984).3.1. Data Sources and MethodologyHistorical data from provincial municipal statistics reports, trade journals, annualreports of various cities, and past system surveys is used to estimate the amount of pipebeneath Canada’s villages, towns, and cities (see Table 3.1). In addition, information oncurrent consumption rates was also compiled.Table 3.1: Data sources used in estimating the Canadian water pipes inventoryEXTENT COMMUNITIES DATA INCLUDED YEARSNationalNationalNationalB.C.AlbertaOntarioNova ScotiaWinnipegMontrealNationalNationalNationalNationalNationalallpop. >pop. >allallallallcitycitylimitedallallallpop. > 1,0001912195119611951-881943-881934-671965-881931-861979-8419841975197719861989SOURCE1,0001,000Denis 1912Municipal Utilities 1951Municipal Utilities 1961Provincial MunicipalStatistics ReportsAnnual ReportsFCM 1984 (survey)FACE 1975FACE 1978FACE 1987IWD 1990Population; Pipe length, type, size rangePopulation; Pipe length, type, size rangePopulation; Pipe length, type, size rangePopulation; Pipe lengthPopulation; Pipe lengthPopulation; Pipe lengthPopulation; Pipe lengthPipe lengthPipe lengthPipe lengthPopulation served; Number of systemsPopulation served; Number of systemsPopulation served; Number of systemsConsumption78The procedure involved accumulating or estimating national water main data forfour available years: 1912, 1951, 1961, and 1986. An estimate for 1991 is calculatedbased on average growth rates. Unfortunately, none of the data sources in Table 3.1 areideal, neither individually nor when combined. Only the data source for 1912 contains acomprehensive listing of the Canadian systems (Denis 1912). While there remains majorgaps in the data, what does exist is still very useful.The data sources for the 1951 and 1961 estimates provide information for allmunicipalities greater than or equal to 1,000 in population (MU 1951, MU 1961), so thepipe in the smallest communities must be estimated by applying average pipe densitiesfrom the available data to the population served in these smallest communities.The data base for the 1986 estimate is the least comprehensive. Comprehensivewater main length information for nearly all communities in Alberta, British Columbia,and Nova Scotia is available from each province’s respective “summary of municipalstatistics” publication.To estimate the pipe lengths in communities outside of these provinces, the datafrom the three provinces is aggregated and the average population served per kilometreof water main, or “service densities”, are calculated for ten community populationintervals. These service densities are then applied to the “populations served by waterdistribution” available in the FACE (1987) publication to give an estimate of the watermain lengths. For some of the larger centres such as Winnipeg, Toronto, and Montrealinformation is extracted from each city’s annual reports, while for some othercommunities, information is extracted from the FCM (1984) survey data base.Although a rigorous statistical analysis is not performed, the sample provided bythe 1986 data sources still represents 40 percent of the entire population served by water79distribution systems in Canada, which constitutes a very good sample. While it is amajor assumption to apply the pipe density for communities in Alberta, B.C. and NovaScotia to other provinces like Quebec or Ontario, especially considering the differencesin land use patterns and urban densities, the results obtained seem reasonable.For a more detailed description of the data sources and the procedure used informulating the national estimate, see Appendix A.3.2. Results of the National Inventory EstimateThe inventory provides some basic information on the length, service density, age,material types, and diameters of the water distribution “pipes” in Canada as well as somegeneral information on system demands.3.2.1. Pipe Length and Service DensityFrom the inventory information, in the order of 23 million Canadians in 2,887communities are served by approximately 130,500 km of water main, with this numbergrowing from 1 to 2 percent annually, or about 2,000 km per year. Thus 85 percent ofall Canadians receive their water from some type of shared water distribution system.Much of the remaining population resides in rural areas of the country and, for the mostpart, is serviced by private wells (FACE 1987).Of all the systems in Canada, only 31 are located in large municipalities(100,000+), 272 are in medium sized communities (10,001-100,000), and 2,584 are insmall centres (< 10,000). As can be seen in Figure 3.1, while there are relatively fewlarge systems, they serve the largest percentage of the total population served (42percent) yet contain the least total pipe length (26.4 percent) of the three populationFigure3.1:Canada’swaterdistrthutionsystemsNumberofsystemsbycommunitysizein1986Populationservedbycommunitysizein1986(37.1%)Kilometersofpipebycommunitysizein1986(107%)-(212%)COMMUNITYSIZE-21Lcrge(100,000+)Medirn(10,001-100,000)[1Smd(<10,000)-(9.42%)A(26.4%)tARCE00 C81groups. Medium sized centres, on the other hand, do not exhibit such a discrepancy; the272 medium sized communities represent 36.8 percent of the population and contain 37.1percent of the pipe. Small communities represent the antithesis of the larger centres,serving only 21.2 percent of the population from 36.5 percent of the nation’s total pipe.The total pipe length calculated seems reasonable when compared to estimatesfrom other sources. An average of 177 persons are served by each kilometre of watermain in Canada, a figure similar to the density reported for the United Kingdom of 175persons per kilometre (DoE and NWC 1977, Edwards and Cox 1981). MacLaren (1983)estimates that the Ontario density is approximately 200 persons per kilometre, while thisstudy estimates it at 202 persons per kilometre.The Ontario density is greater than the national average partly due to Ontario’shigher proportion of large, dense centres. Toronto for instance has a service density ofover 500 persons per kilometre of water main, or nearly three times the nationalaverage.Table 3.2: Population served per kilometre of pipe in CanadaPopulation Interval Persons per Kilometre of Pipe500,001 — 1,000,000 313100,001 — 500,000 26130,001 — 100,000 18010,001 — 30,000 1595,001 — 10,000 1142,501 — 5,000 1121,001 — 2,500 106less than 1,000 68National Average 177Source: Derived from data base in Table 3.182There is a very strong relationship between the pipe density and size ofcommunity in Canada. From Table 3.2, one can see that the largest municipalities inCanada serve nearly five times as many people from the same length of main as do thesmallest communities. Generally, the larger the community, the higher the servicedensity. This general characteristic has existed historically and remains quite evidenttoday as can be seen in Figure 3.2. It is also important to note that there has been anoticeable decrease in the service density across all population intervals over the pasteighty years, indicating a general trend toward less intensive urban land use and higherlevels of service.As can be seen in Figure 3.3, there exists significant differences in servicedensities even among the largest cities in Canada. Older cities such as Toronto andMontreal have historically held higher service densities than younger cities such asCalgary and Edmonton, although the densities in the older cities appears to bedecreasing with the development of less dense suburbs. The apparent increase in servicedensity in Halifax is probably due more to an inaccuracy in the provincial data, ratherthan major increases in the city’s density.3.2.2. Age of Distribution SystemsAs can be seen in Figures 3.4 and 3.5, the provision of water distribution servicein Canada is only 150 years old, with the most rapid expansion coinciding with the “babyboom” following World War II. From Figure 3.5, the average age of Canadian waterdistribution systems in 1991 is calculated to be 31.9 years, which corresponds well withthe average age of 25-28 years estimated in 1984 by the FCM, especially considering thatseven years have passed since the FCM survey (see Table 3.3).Figure3.2:Canada’swaterworks700600500400300200100 0 18001850190019502000PersonsperkilometerofwatermainbyyearE (1) C 0 (I) ci)vCANADAMEDIUSMASizeofUrbanAreaLarge(100,000+)—o—Medium(10,001—100,000)—x—SmdI(<10,001)—v—CanadaYear00Figure3.3:Cancida’swaterworksPersonsperHometerofpipenselectedcities900800700600500400300200ci) ci) E 0 (I) C 0 (1) a) 0100 191019201930194019501960197019801990‘I’ear00Figure3.4:Canada’swaterworksPopulationservedinsurveyedcommunitiesbyyearU) (I, C 0 U, ci) 030 25 20 15 10 5 0SizeofUrbanAreaLarge(100,000+)—a—Medium(i0,00l—lO0,000)—x—Small(<10,001)—v—Canada7 / /—‘..lxx18301850187018901cJ10Year193019501970199000 UiFigure3.5:Canada’swaterworksKilometersofpipeinsurveyedcommunitiesbyyearU) C U) :3 0 U) q) E 0140120100 80 60 40 20 0SizeofUrbanAreaLarge(100,000+)—0—Medium(10,001—100,000)—x—SmaN(<10,001)—v—Canada/ /I,v-.—zz=Z183018501870189019101930195019701990Year0087The specific age of Canadian systems varies from region to region and from cityto city in Canada. Even within a community the age varies from area to area, with theoldest portions of the system usually located near the core. But because pipes wear outsomewhat inconsistently, some sections of a system may contain the original pipe laiddown when the systems was first initiated, while other sections may have been replacedonce or a multitude of times.Table 3.3: Estimated average age of distribution systems by region in 1991REGION AGE (Years)B.C. 33Prairie 28Ontario 34Quebec 30Atlantic 32CANADA 32Source: Derived from data base in Table 3.1The rate and nature of this replacement varies significantly and makes it difficult toobtain an exact picture of the general age of the Canadian system using this data.In Toronto, Doherty et al (1987) notes that only 50 kilometres of Toronto’s 1,227total kilometres were replaced between 1968 and 1987, which translates into only 0.2percent per year. At this rate, it would take 470 years to replace all the mains in thecity. In Vancouver, Bratton et al (1986) notes that 370 kilometres of steel mainsinstalled in the 1920’s were replaced in the 1960’s due to high leakage, representing over26 percent of the city’s current 1,400 kilometres of water main.Because of the difficulty in determining national replacement rates, the age of thepipes determined in this analysis are based on an assumed zero replacement rate. While88this may seem a major assumption, the relatively young average age of the systems inCanada probably means a great deal of the original pipes have not yet been replaced.Considering the close match between the FCM and this study’s calculation of averageage, this may be a reasonable assumption. Even if this assumption were to prove lessthan acceptable, Table 3.3 still can be thought to represent a “worst case” where none ofthe pipe in the ground has been replaced, or an average “time since first installation”.As can be seen from Figure 3.5, the majority of the Canadian system has beeninstalled since 1951. Only about 23 percent of Canada’s water mains were installed morethan 40 years ago, and only 6 percent were installed more than 80 years ago.As a percentage of their current system length, Atlantic Canada and Ontariocontain the largest percentage of pipe greater than 75 years old (see Figure 3.6) andgenerally have the oldest systems (see Table 3.3). As expected, lower percentages of theoldest pipe are found in Western Canada, but perhaps surprisingly, the smallestpercentage is found in Quebec. This may partly be due to the fact that historically theservice densities have been high in Quebec, thus there is little “old” pipe left today.Also, the application of the relatively low densities found in B.C., Alberta, and NovaScotia to Quebec communities may have resulted in a slight over-estimation of thecurrent pipe length in Quebec, and intern the percentage of pipe which is less than 25years old.It is also interesting to note that B.C. holds the largest percentage of pipe whichwas installed between 50 and 75 years ago while maintaining the least amount of pipeless than 25 years old, indicating a large percentage of the community systems wereinitiated or expanded between 1916 and 1966 in the province. This gives B.C. the secondoldest average systems length in the country (see Table 3.3) at 33 years, though this mayFigure3.6:Canad&swaterworks100Percentageofmainbyageandbyregion90 80 70 600 C__________________________________________________________ci) (-) ci) U-40 30 20B.C.PRAIRIESONTARIOQUEBECATLANTICCANADARegionAgeInterval•>7550-7525-50<25Go90be slightly exaggerated considering the large scale replacements in Vancouver during the1960’s (Bratton et al 1986) and considering that Vancouver contains a large percentageof the province’s older water mains.The large percentage of pipes installed nationally in the last 25 years correspondsto the rapid growth both in the population and in the funding from all levels ofgovernment aimed at providing expanded and better quality water distribution service.But still, 47 percent of the current systems are over 25 years old, meaning they are at orapproaching an age where some major rehabilitation may be required.On a pipe length basis, as expected, the longest lengths of old pipes are inOntario and Quebec, which combined retain more than 64 percent of the country’s pipesinstalled over 50 years ago (see Figure 3.7).System age varies not only by province in Canada, but by size of community.Continuous records of pipe length by community and by year were only available throughmunicipal statistics for Alberta, B.C. and Nova Scotia. Using this data, it can be seen inTable 3.4 that, as expected, the average age of the mains in the larger centres is olderthan in the smaller centres.Table 3.4: Estimated age of distribution systems in selected provinces in 1991AVERAGE AGE (Years)COMMUNITY POPULATION (1988) B.C. ALTA. N.S. COMBINED> 100,000 42 30 79 (*) 3710,001 — 100,000 30 23 43 31less or equal to 10,000 28 20 43 28Provincial Average 33 25 46 32*— Halifax is the only community in this categorySource: Derived from data base in Table 3.1Figure3.7:Canada’swaterworksLengthofmainbyageandbyregionAgeInterval•>7550-75El25-50<25(I) ci) ci) E 08000070000600005000040000300002000010000 0_______.//-:I_-,I-B.C.PRAIRIESONTARIOQUEBECATLANTICCANADARegion92However, the average age of the systems in the largest communities are typically only 10to 15 years older than the smallest communities. This can partly be explained by theincreasing growth in the larger centres, such as Calgary, Vancouver and Edmonton,which has resulted in a large amount of new mains being installed and a subsequentdecrease in the overall age of the systems. Smaller communities have not experiencedthe degree of growth of the large cities and, as a result, a large portion of their currentsystems consist of the original pipe, though the age of this pipe is still younger than theoldest pipe in the large cities.The average system age of the smaller communities also varies among theprovinces. The smaller communities in Nova Scotia have average system ages anywherefrom 15 to 20 years older than their western counterparts and there is no indication thatmajor portions of the systems have been replaced. Referring to a recent study of thewater systems in the Maritimes, MacLaren (1985) concludes that the worst problemareas are in small to medium sized communities which have never undergone majorurban renewal nor experienced significant urban expansion.It is also apparent that Nova Scotia maintains the oldest systems of the threeprovinces. As mentioned earlier, there seems to be some doubt as to the accuracy of thedata for Halifax, though overall the anomaly does not seem to influence the resultssignificantly.3.2.3. Pipe SizesThe size of a pipe is an important characteristics when considering therehabilitation of a water main. In Canada, the majority of the pipes in any particulardistribution system generally have diameters in the 100 mm to 300 mm range which are93fed by transmission lines which usually range anywhere from 300 mm to over 1200 mmin diameter. Much of the data used in this section dates from 1961, but since designpractise has not changed dramatically since then, the general information should be aquite reasonable estimate of today’s situation. Probably the only major design changerelated to pipe diameter in the past thirty years has been a move away from smalldiameter pipes in the 100 mm and smaller range due to the need for increased fire flows.As can be seen in Table 3.5(a), the larger a community the larger the pipes thatcan be found in its system. In 1961, nearly 93 percent of all the smallest communitiessurveyed had pipes no larger than 300 mm in diameter. In the largest communities, thelargest sizes ranged from 550 mm to 1200 mm and up.From Table 3.5(b) it can be seen that in most communities, the smallest pipediameter was 100 mm, with only 22.5 percent of the communities having smaller mains.From a sample of 78 small and 10 medium sized communities in the 1961 survey(MU 1961), it can be seen from Figure 3.8 that the most common pipe size is 150 mm,which comprises 55 percent of all the pipes, followed by 200 mm and 100 mm whichrepresent 16 percent and 12 percent respectively. As noted earlier, the smallestcommunities contain very little large diameter pipes with only 12 percent of the pipebeing larger than 200 mm, while in medium sized communities this value jumps to 21percent. Although a breakdown for the largest sized communities is not available, it isreasonable to assume that the proportion of large diameter pipes is even greater.Given these general trends, there can still be significant variations in the range ofpipe sizes which seriously can affect system capacity, even among larger communities. In1978, Surrey, B.C. had quite a large system serving approximately 130,000 people.94(A) Percentage of connunities with largest pipe sizes by diameter:Size of Coninunity< 1,001 1,001- 2,501- 5,001- 10,001- 30,001-2,500 5,000 10,000 30,000 100,000Diameter(inches)48+27-4722-2418-20 .814-16 3.112 8.7 12.210 4.3 24.08 30.4 437*6 52.2* 13.0<6 4.3 3.1Total 00 100.0 100.0Sample Size 23 254Sample Pop. 39,000 527,850Sample km 61.2 1157.8(B) Ppr’ tomunitips with1.7 3.3.5 1.7 15.02.5 5.0 24.212.1 27.3 32.5*26.3 33.1* 18.327.8* 20.7 4.224.7 8.3 1.75.6 1.7 .8.5 .8100.0 100.0 100.0198 121 120777,700 956,300 2,264,7301452.6 1344.7 3199.7--smallest pipe sizes by diameter:Size of Comunity< 1,001 1,001- 2,501- 5,001-2,500 5,000 10,000Diameter(inches)10-14864<4Total (%)Sample SizeSample Pop.Sample km*- denotes**- denotesSource: MU (1961)2.3 33.3 .825.0 40.0* 3.043.2* 26.7 5.718.2 6.59.1 13.92.3 19.119.023.17.61.4100.0 100.0 100.044 15 7752,400,546 5,451,724 12,417,8503525.4 4866.4 15607.8Diameter(m)250-350200150100< 100The total 934 kilometres of mains had an unusually small percentage of largetransmission sized mains (over 300 mm) and an unusually large percentage of very smalldiameter mains (under 150 mm) as can be seen in Table 3.6. The system was plaguedTable 3.5: Largest and smallest pipe sizes in Canadian communities in 1961100, 001+ TotalDiameter(m)1200+675-1175550-600450- 500350-400300250200150< 15010,001- 30,001- 100,001+ Total30,000 100,000.4 1.0 2.5 2.5 1.22.4 2.0 .8 1.4435** 24.4 24.2 22.3 18.3 9.1 20.0 22.734.8 47.6** 56.6** 545** 52.5** 63.6** 46.7** 52.321.7 25.2 16.2 19.8 26.7 27.3 33.3 22.5100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.023 254 198 121 120 44 15 77539,000 527,850 777,700 956,300 2,264,730 2,400,546 5,451,724 12,417,85061.2 1157.8 1452.6 1344.7 3199.7 3525.4 4866.4 15607.8the largest percentage of comunities with this diameter as their largest pipe sizethe largest percentage of comunities with this diameter as their smallest pipe sizeFigure3.8:Canada’sWaterDistributionSystemsDiameterofMainsbySizeofCommunityin1961for78Smalland10MediumSizedCommunitiesUIcJ’ U C (-) L) 080 60 40 20 0SMALLMEDIUMCOMBINEDSizeofCommunityPipeSize•<100100150200250300UI>30096by low pressure and flow. After significant public outcry and recent developmentpressures, the problem has been remedied by replacing many small diameter mains withlarger ones (Lalani 1990). Many communities in Canada have not seen the developmentpressures that Surrey has and could today have serious capacity problems.Overall, there is a predominance of small diameter mains in all communities.Approximately 80 percent of a typical system is made up of relatively small mains under300 mm, with the figure closer to 90 or 95 percent in the smallest communities.Table 3.6: Pipe sizes in Surrey, B.C. in 1978Diameter Percentage of System Length< 100 7.25100 21.95150 46.35200 12.37250 1.09300 6.95> 300 4.03Source: Surrey (1978)The hydraulic basis by which pipes in Canada are sized is fairly straight forward.System design is based on maximum consumption rates, which can vary with the seasons,the day of the week, and the hours of the day. The seasonal peak is invariably in the hotsummer months and the hourly peak is close to noon while the hourly minimum is in theearly hours of the morning; typically, the maximum day exceeds the average day by 1.2 to2.0 times while the maximum hour exceeds the average hour by 2.0 to 3.0 times(MacLaren 1985). While peak hour flows are important in the design, in a typical97municipal system the critical design criteria is the provision of fire flows. The flowrequired for fire fighting purposes can be based on the community size, the land usedensity and type, and the fire fighting equipment and stationing. The Fire UnderwritersSurvey sets guidelines for municipal fire protection and the associated insurance ratings.Unfortunately, as Canadian systems developed, economics and availability ofmaterials often took precedence over the hydraulic design basis, thus in manycommunities the pipes installed have inadequate capacities to handle fire and in somecases even domestic flows. Typically, to handle fire flows a pipe must be at least sixinches (150 mm) in diameter. As evident by this discussion, many communities have asignificant amount of such inadequate pipe.3.2.4. Pipe MaterialIt is unfortunate that extensive national information related to the more recentuse of pipe materials is not readily available, especially since the use of new materialshas evolved rapidly since 1961. The increased use of ductile iron (D.I.) and asbestoscement (A.C.) in the 1960’s and 1970’s and the more recent large scale use of P.V.C. andP.E. plastic pipes have dramatically changed the nature of new distribution works. Themost recent national data source remains the 1961 Municipal Utilities publication.Although the information is significantly out of date with respect to current installations,it is still very applicable when considering current rehabilitation technologies which willgenerally be aimed at pipe installed prior to 1961.Up to 1961, the major pipe material used in Canadian water mains was cast iron(see Figure 3.9). Up to 1951, materials such as steel and wood were primarily used forlarger diameter transmission mains, along with smaller amounts of concrete and steelC) cl)Figure3.9:Canada’sPercentagesofwaterdistributionsystemsmaterialtypesbyyear100 90 80 70 60 50 40 30 20 10 0III zrNj fliL 1912I19511961Year.MaterialTypeCi.STEELWOOD•A.C.MISC.0099reinforced concrete which are included in the miscellaneous category of Figure 3.9. Alsoin this category are small galvanized iron pipes, usually less than 100 mm in diameter,which were common in some communities but not widely utilized. Wood was commonas a pipe material in areas where the supply source was generally reliable and lumberavailable, mostly in B.C. with a small percentage in Quebec.The development and initial use of new light-weight materials such as asbestoscement (A.C.) in the early 1950’s displaced some of the traditional cast iron market.A.C. was first introduced in North America by Johns-Manville Corporation (U.S.A.) in1929. It has been estimated that 2.5 million kilometres of A.C. pipe have been installedworld wide, and in Canada A.C. has been used extensively since the early 1950’s with88.9 percent of the A.C. pressure pipe being located west of Ontario (Nebesar and Riley1984). By 1961 A.C. made up almost 9 percent of the total system mileage, beingespecially popular in the corrosive soils of the Prairies (see Figure 3.9). In 1977, A.C.pressure pipe made up 18 percent of the new pipe market in Canada (Vagt 1980) and in1980 made up 130,000 km or 13 percent of the total 1 million kilometres of water mainin the U.S. (Nebesar and Riley 1984). If similar numbers apply to Canada, there couldbe as much as 13,000 km of A.C. pipe in the ground today, with over 11,000 km inWestern Canada. Since the late 1970’s, however, the use and manufacture of asbestoscement has diminished greatly with the wide spread use of plastic pipes, especially P.V.C.and to a lesser extent P.E., and with the recent public health concerns over asbestos ingeneral.Facing stiff competition from new materials, the iron pipe industry introducedductile iron pipe which was both thinner, stronger, and more impact resistant than itsmore brittle cast iron predecessor. By 1967 ductile iron had all but replaced cast iron as100a material for new pipes in cities such as Calgary (see Figure 3.10). Over the years,cities like Calgary and Vancouver continued the almost exclusive use of iron pipe in theirsystems. In Vancouver today, shallow bury depths and the high probability of third partydamage require a material with high impact resistance such as D.I. (Bratton 1990).Unlike Vancouver, smaller communities in the Greater Vancouver area opted for avariety of newer materials over the years, including significant lengths of asbestos-cementas can be seen in Table 3.7.Table 3.7: Pipe material in selected B.C. utilitiesMATERIAL PERCENTAGE OF TOTAL SYSTEM LENGTHC. I.D. I.A.C.SteelPlasticWoodGalvanized IronConcreteMisc.TOTALSurrey (1978)29.029.326.74.50.32.05.41.81.0100.0Burnaby (1991)34.919.938.85.60.10.00.00.00.0100.0North Vancouver (1991)District *21.020.048.06.00.04.01.00.00.0100.0Total Length 934.3 km 665.7 km 392.0 km ***— estimated material percentages from MacLean (1991)**— estimated from 1988 B.C. Municipal StatisticsSources: Surrey 1978; Robinson 1991; MacLean 1991.Figure3.10:Calgary’sdistributionsystemKilometersofwatermainbymaterialtypeandyear“I)4) S 0-4350030002500200015001000 500 0:/V:V-v—- -TOTALv—CASTRON- -vzz—DUCTLERONOTHER(A.C.,P.V.C,CONC.,P.E•)--><X-1955196019651970197519801985:‘:•Year1023.2.5. The Physical EnvironmentJust as the various systems in Canada vary with age, material type, and pipediameters, so varies the physical environment to which the piped systems are exposed.Outside of the operational parameters such as pressure and pressure cycles whichconstantly strain the pipes over the years, the effects of aggressive distribution waters,aggressive soils, and frost penetration play a major role in determining the design andlifespan of the pipes.The acidification of the lakes and forests along the Canadian shield have raisednational and international concern over the sulphur dioxide (SO2) emissions from nonferrous ore smelting operations and coal fired power plants, and nitrogen oxides (NOr)from automobile exhaust and fuel combustion. East of the Saskatchewan/Manitobaborder, over 4.5 million metric tonnes of SO2 are released annually into the atmosphere,49 percent of which are from Ontario and 24 percent from Quebec (EnvironmentCanada 1986). Particles from these emissions react with water in the atmosphere andfall back to the earth as acid rain, depressing the pH of unbuffered soil and waterbodies. The ten year Canadian Acid Rain Control Program aims to cut this level in halfby 1994 with Quebec and Ontario commencing their respective emission control plans in1985 (Environment Canada 1986).The goal of the program is to reduce the annual acid deposition to less than 20kilograms per hectare per year (18 lbs./acre/yr.), which is the level most lakes and riversin Canada can tolerate without damage. The effects of acid rain have been welldocumented over the past ten years: it increases the acidity of lakes and streams to thepoint where aquatic life is depleted, it increases the acidity of shallow groundwater, it issuspected as one of the causes of forest decline, it erodes buildings and monuments, and103is suspected of contributing to respiratory problems in people (Environment Canada1986).One of the effects of acid rain which receives little or no attention is theacceleration of corrosion in water distribution piping and the elevated levels of dissolvedmetals in the drinking water which may result. As can be seen in Figure 3.11, except forthe Prairies, the Hudson Bay basin, and some relatively isolated areas, most of Canada issusceptible to the effects of acid rain. Sensitive areas are identified by mapping thebedrock with very little buffering capacity; buffering capacity being the capacity toneutralize acid solutions or to increase pH to the natural levels that prevailed in ahydrological system before the incidence of acid precipitation. Carbonate minerals(calcite, dolomite, siderite-rhodochrosite) provide the greatest buffering capacity pervolume of rock, while clay-sized micaceous minerals provide the second greatest (Shuts1981). Bedrock producing soils with quartz and feldspar rich silt, sand and pebbles aremuch more sensitive to acid precipitation. Thus the granitoid bedrock of the Canadianshield has very little buffering capacity while the limestones of the prairies have a highbuffering capacity. Though the type of bedrock is the best overall indicator of soilsensitivity, some caution must be exercised as glacial transport of drift can distort thesensitivity patterns derived solely from the bedrock information (Shilts 1981).While much of attention regarding pH depression of lakes and rivers due to acidrain has focused on eastern Canada where the major sources of sulphur dioxide andnitrogen oxides are located, the effects of acidic rainfall can be found in western Canadaas well. Millette and Mavinic (1988) note that the pH of normal rainwater isapproximately 5.5, while in Vancouver where the rainwater is normally acidic, the pH ofdrinking water is typically in the 4.5 to 5.7 range, well below the neutral pH 7 value.Figure 3.11: Areas Sensitive to Acid Rain104NORTHAMERCA105The geology of the coastal region is such that the waters also contain very lowconcentrations of dissolved species, hardness and alkalinity being the main indicators ofthese. Thus, Vancouver is in a situation of aggressive waters with little bufferingcapacity. Millette and Mavinic (1988) found elevated levels of iron and copper innumerous tap water samples and considered the addition of lime as a solution. InVancouver a large number of steel mains installed in the 1920’s have had to be replaceddue to excessive breaks and leaks caused largely by internal corrosion. In many of thecommunities surrounding Vancouver, asbestos-cement pipe has also been known todegrade due to the aggressive nature of the waters.The immense size of Canada makes generalizations about climate and soil typesnearly impossible. Typical frost penetration varies from nearly zero in Vancouver to 2.2meters in Edmonton. In the north, where continuous permafrost exists, rather thanmeasuring frost depth in the winter, thaw depths are measured in the summer (MOT1973). A number of excellent sources are available for estimating the depth of frostpenetration (MOT 1973, Canadian Geotechnical Society 1985) though the best recordscan usually be obtained from the local Environment Canada weather station.Frost penetration has a huge influence on the design, cost, and maintenance ofpiped systems. In Alberta for instance, provincial design regulations are based on frostpenetration and stipulate that all municipal water pipes must maintain at least 2.5 m ofcover over the crown of the pipe, while in Vancouver, water lines are typically buriedwith only 1.0 m to 1.2 m cover, with such depths based on surface loads rather than frostpenetration. Since a major part of the cost of pipelines is in the excavation andinstallation, the depth of cover is a big factor in overall installation and replacementcosts. MacLaren (1983) notes that breakage rates in American cities, where frost106penetration is much less, are much lower than comparative Canadian rates.Table 3.8: Average frost penetrations in selected cities (1964-1971)LOCATION YEARS OBSERVED AVERAGE FROST PENETRATION (m)Asphaltic Concrete Portland CementPavement Concrete PavementFort Nelson, B.C. 7 3.3Smithers, B.C. 3 1.5Calgary, Ab. 5 1.9Edmonton, Ab. 6 2.2Lethbridge, Ab. 3 1.2Regina, Sask. 7 2.0Saskatoon, Sask. 2 1.8Gimli, Man. 3 3.0Winnipeg, Man. 4 1.9 2.1Lakehead, Ont. 4 1.5Ottawa, Ont. 3 1.6Toronto, Ont. 3 0.9Bagotville, Que. 4 2.1Montreal, Que. 3 1.4Quebec, Que. 3 1.3Fredericton, N.B. 3 1.7Moncton, N.B. 3 1.2St. John, N.B. 4 1.6Halifax, N.S. 4 1.1Sydney, N.S. 4 0.9Summerside, PEI 4 1.2Gander, Nfld. 4 1.4St. Johns, Nfld. 4 0.7Hay River, NWT 2 3.6Yellowknife, NWT 2 4.3Whitehorse, Yuk. 7 3.2Note: Depths are on snow cleared surfaces.Source: Adapted from MOT (1973).Soil types also vary across Canada and even within a given system can vary widely,especially systems which developed in river valleys or on flood plains where silt, clay, and107gravel depositions vary widely and can therefor vary the corrosion rates on pipematerials. MacLaren (1983) notes that Winnipeg has had a tragic break record for morethan sixty years due to the sulphate corrosion of the local gumbo clay on the originalsand cast iron mains.Calgary is also affected by extremely corrosive clay and silty clay soils and in theearly 1980’s implemented an extensive pipe replacement and cathodic protectionprogram (James and Nq 1991). In 1978, the breakage rate in the City peaked at over 50breaks per 100 km, or nearly 1,300 breaks system-wide. This represented a rate nearlytwice that of the national average and five times the target break rate of 10 breaks per100 km per year recommended by MacLaren (1983). Calgary’s cathodic protectionprogram concentrated on:a) electrical isolation from electrical ground grids and unprotected water mainswhere practical,b) electrical continuity of water mains to be protected,c) use of dielectric coatings,d) use of galvanic anodes for protection of coated and uncoated distribution mains,e) use of rectified impressed current system for the protection of large diametertransmission main,f) a monitoring system to check potentials of cathodically protected piping systems.Since being implemented, the program has resulted in a reduction of the break rate byover one-half, with the City reporting 20 breaks per 100 km in 1990, or 583 breakssystem-wide (James and Nq 1991). Using MacLaren’s (1983) estimate of $3,000 perbreak repair, this translates into a saving of over $2 million a year.Although research is continuing into the soil specific nature of water maindeterioration (AWWA 1986a), the problems and the solutions will remain very site108specific and will continue to be handled by each individual utility on the basis ofexperience and trial and error with a variety of existing methods.3.2.6. System DemandBoth the level of service and the per capita water consumption have increasedsince the first systems were installed in Canada in the mid-1800’s. Anderson (1988)notes that the first system in Quebec City was constructed in 1848 to accommodate ademand of approximately 166 litres per capita per day, but in 1924 a special commissionhad found the actual consumption was nearly seven times that amount or 1036 litres perday. Even today, Canada ranks among the highest per capita users of treated water (seeTable 3.9), with an average per capita consumption of 360 litres per day (Tate andLacelle 1987).Table 3.9: Domestic per capita water use in selected countriesCountry Pumpage per capita per day (litres)United States 425CANADA 360Sweden 200United Kingdom 200West Germany 150France 150Israel 135Source: Tate (1990)Tate and Lacelle (1987) estimate that in 1986, the average daily flow of allmunicipal water utilities in Canada serving populations over 1,000 totalled 12.4 millioncubic meters (MCM). Over an entire year this represents enough water to fill LakeOntario more than two and a half times (Canada World Almanac 1990).109Of this water, Tate and Lacelle (1987) estimate that 40 percent is used fordomestic purposes, 34 percent is used for commercial, institutional or industrial purposesand 26 percent is unaccounted for.On average, leakage and waste accounts for at least 20 percent of unaccountedwater (Tate and Lacelle 1987), or more than 5 percent of total daily flows. Tate suggeststhis to be further evidence of the need for major system renovation (Tate 1990).Tate and Lacelle (1987) note that only a small fraction of domestic water is usedfor drinking and cooking, with over 70 percent either used in toilets or on the lawn(Table 3.10).Table 3.10: Municipal water use in CanadaA) MUNICIPAL END USES:Major End Use Percentage of Total PumpedDomestic 40Commercial/Institutional 16Industrial 18Losses/Unaccounted 26B) DOMESTIC END USES:Domestic Use Percentage of Total Domestic UseDrinking/Cooking 5Toilet 40Lawn Watering 30Bathing/Personal 15Laundry 10Source: Tate (1990)Only 1.34 litres of water per day is required per person per day for drinking (Hickman1984, MacLaren 1985), which is only 0.4 percent of the average per capita domesticwater usage of 360 litres per day (Tate 1990).1103.2.6.1. Community Demand ProfileBased on recent data obtained from the Inland Waters Directorate (IWD 1990),water consumption rates vary among different sized communities in Canada. From thedata, the 30 largest communities in Canada have the lowest per capita domestic waterconsumption rates which are on average 15 percent below the calculated nationalaverage of 351 litres/day/capita (see Table 3.11). Small and medium communities bothtend to have similar rates which are typically more than 20 percent above the nationalaverage.Table 3.11 Domestic per capita water use in Canadian communities in 1989Population Communities Per Capita Consumption (l/d/cap)Surveyed High Low Average100,000 + 30 475 95 29210,001 — 99,999 244 1727 93 4020 — 10,000 817 2749 36 434CANADA 1091 2749 36 351Source: IWD (1990)Based on total water usage, there is no significant difference among the varioussized communities. The difference between domestic and total water usage accounts forcommercial, institutional, industrial, or other consumptive uses which occur in acommunity. As can be expected and as evidenced by Table 3.12, in smaller communitieswhere the population base is small and there may be a single large industrial user ofwater, the occurrence of very large “per capita” total consumption figures will be morecommon than in the larger communities.111Table 3.12: Total per capita water use in Canadian communitiesPer Capita ConsumptionHigh Low1310 3682878 1556836 40(l/d/cap)Average698680668Source: IWD (1990)The percentage of surveyed communities which consume water at a significantlyhigher rate than the national average can be seen in Table 3.13. Approximately 44percent of all communities in Canada use domestic water at rates at least 15 percentover the national average. On a national scale, this could amount to over 1,200communities, nearly all of which are small and medium sized.Table 3.13: Occurrence of domestic water use rates in Canadian communities in 1989——Percentages of Canada’s Total——Communities Popn. FlowPopulation CommunitiesSurveyed100,000 + 3010,001 — 99,999 2440 — 10,000 817CANADA 1091 6836 40 688Consumption(l/d/cap)CommunitiesSurveyed0— 200 115 11 18 12201 — 400 497 45 55 54401 — 2750 479 44 27 34CANADA 1091 100 100 100Source: IWD (1990)11232.6.2. Regional Demand ProfileThe same data set as used for the communities can also be used to deriveregional consumption trends. As can be expected, average water consumption rates varyfrom province to province, but as can be seen in Table 3.14 the lowest consumption ratesare in Ontario and on the Prairies, while the highest rates are in Quebec and on the twocoasts.The highest per capita water usage occurs in New Brunswick, Quebec,Newfoundland and B.C. where supplies are abundant and a large volume of municipalwater is supplied to industrial and other processes. Lowest per capita usage occurs inOntario, P.E.I. and on the Prairies where good supplies tend to be more limited. Thelow values for Ontario are undoubtedly due to the heavy weighting of urban areas suchas Toronto, Hamilton, and Ottawa which have quite low per capita consumption rates.Table 3.14: Provincial per capita water use in Canada in 1989Province Consumption (l/d/cap) Percentage of National FlowTotal Domestic Total DomesticNewfoundland 729* 515* 1.4 1.9P.E.I. 525 221 0.1 0.1Nova Scotia 646 357* 1.6 1.7New Brunswick 1136* 514* 3.1 2.8Quebec 844* 420* 32.9 32.2Ontario 605 276 34.1 30.5Manitoba 491 370* 2.8 4.1Saskatchewan 591 317 2.7 2.8Alberta 585 314 8.8 9.3B.C. 722* 429* 12.3 14.4Yukon 737* 414* 0.1 0.1N.W.T. 551 327 0.1 0.1CANADA 688 351 100.0 100.0Note: * — indicates higher than national averageSource: IWD (1990)113Table 3.15: Regional per capita water use in Canada in 1989Region Consumption (l/d/cap) Percentage of National FlowTotal Domestic Total DomesticMaritimes 845* 451* 6.2 6.5Quebec 844* 420* 32.9 32.2Ontario 605 276 34.1 30.5Prairies 565 327 14.3 16.2B.C. 722* 429* 12.3 14.4Territories 642 370* 0.2 0.2CANADA 688 351 100.0 100.0Note: * — indicates higher than national averageSource: IWD (1990)3.2.7. Water System Staffing LevelsWhile the maintenance of water distribution systems has remained at adequatelevels, there is still evidence of gradually reduced staffing levels over the years. Whilesome of the reductions could be attributed to more efficient maintenance and repairtechniques and improved pipe and installation technology, the decrease in staff levels stillexhibits a shift away from management of infrastructure systems.According to Statistics Canada in their annual publications of MunicipalGovernment Employment and Local Government Employment, the total number ofpeople employed by local government, in communities with more than 10,000 people,more than doubled during the period from 1962 to 1984, increasing from 107,294 to inexcess of 244,000. Over the same time, water systems expanded enormously with thetotal length increasing by more than 2.5 times. Yet during this period, the number ofpeople employed in the water works departments increased by just over 10 percent,going from 6,989 in 1962 to 7,738 in 1984. This is further indication of the shift in localgovernment’s support over the years away from basic infrastructure needs in order toserve the ever increasing demands of expanded social services.114CHAPTER 4: WATER MAIN DETERIORATION4.0. OverviewIn order to effectively deal with the deterioration of a water systems, a goodunderstanding of what deterioration is, the mechanisms responsible for it, and theavailable procedures a utility may implement to mitigate it are invaluable. This sectionwill introduce the common indicators of water main deterioration, the major causes of it,the mechanisms associated with it, and the techniques used to deal with it.Although every pipe in the ground has been designed to withstand such loadingswithin a given environment, the design process by nature is one involving a number ofassumptions and general principles to come up with one specific design which mustsatisfy a wide range of environmental loadings developed. With pipelines this isespecially the case. It is economically impractical to constantly change the pipecharacteristics to accommodate every condition along a pipeline’s total run. A typicalpipe design will specify only one or two materials each with specific wall thicknesses anddiameters, and some factor of safety to accommodate the broad range of localizedenvironmental loads to which a high level of uncertainty exists.This process of general design to accommodate specific local loadings works quitewell, but not all consequences can be foreseen and not all conditions adequatelypredicted, especially when long periods of time are involved in the operational life of apipe. Thus failures of the pipeline in the form of breaks or leaks will always occur, withthe frequency of occurrence the indicator of a successful design. The design processinvolves combining imperfect information and an imperfect understanding of all thephysical mechanisms to come up with the best design for the lowest possible cost. Insuch a process there is a trade-off between uncertainty and economics. It would is115possible to design a pipe which could potentially last 300 years without a break, but theeconomics would prohibit such a design. Thus a pipe is inherently designed todeteriorate and eventually wear out.4.1. Defining Water Main DeteriorationThe response of a pipe to its environment is dependent on a wide variety ofcomplex, interacting influences. Soil conditions, pipe material, pipe geometry, externalloadings, internal pressures, temperature, and construction all play a part in the life of apipe as can be seen in Figure 4.1. Unfortunately, not all influences can be predictedwith a great deal of certainty and many conditions can change radically to affect theperformance of a pipe. The situation is further complicated by the fact that there is notan easily distinguishable point in time when a pipe can be considered to have succumbedto its environment and needs replacing due to failure.The deterioration of a water pipe is not a discrete phenomenon, but rather one ofvarying degrees. In general, a pipe is designed to carry a certain quantity of water fromone point to another, and any time the pipe fails to do this can be considered a “failure”.Based on this simplistic definition, a pipe break is most often associated with failuresince a break disrupts flow. But the occurrence of one break is not sufficient to meritreplacement of the entire water main.The determination of when to replace must incorporate some type of criteriawhich defines when a main has truly “failed” to serve its intended purpose. This istypically defined as that point in time where it becomes more economical to replace awater main rather than to continue its operation or the point in time when the risk inmaintaining an increasingly unreliable pipe becomes too great. When theseFigure4.1:ConceptualmodelwatermainstructuralconditionExternalLoadsEarthTemperatureTruckAerationSoilCharacteristicsBeddingConditionLeakageConstructionStresses:hooptensileringbeam1ContractionLoadFrostConstruction—øElectrol/GroundwaterExternal/Galvanic4SoilMoistureCorrosion/ZakaeJInternalCorrosionPressure_IUnitPipeStrengthSafetyFactor117determinants are used, a number of structural and performance factors, other than justbreaks, can be introduced into the formula which determines the replacement time.From a structural perspective, the common indicators of deterioration are mainbreaks and leaks. The costs associated with continued operation include the costsassociated with repairing both the breaks and any detected leaks plus the costs associatedwith the water lost due to the detected breaks and leaks, as well as the undetected leaks.For discussion purposes, a leak and a break are distinguished in that a break interruptsflow and is therefor easily detected by large volumes of water appearing at the groundsurface. It usually does not result in significant water loss over time since it is often ashort-term event. In contrast, a leak does not interrupt the flow and if it is a slow leakor does not surface due to bedding or sub-surface conditions (gravel beds orunderground streams), it can often remain undetected and result in significant waterlosses over time.Because of the difficulty in locating many of the small, less detectable leaks, manyutilities do not concentrate on remedying leakage problems in their systems, but ratherchose to base pipe repair and replacement on the increased costs due to emergencyrepair events caused from breakage. Undetected leaks remain largely ignored, andthough each may be small, all together they can represent sizeable water and revenuelosses over time.4.2. Stresses on a PipeThe degradation of a pipe’s physical properties depends largely on the interactionof the pipe material with its surrounding environment over time. The interaction canresult in changes to the geometric or material properties of a pipe. Geometric changes118can include pitting or thimiing of the pipe wall, or extreme deflections which can causeloss of seal at joints. Changes in material properties usually result in a loss of strengththrough a corrosion or ageing process which results in a chemical change of the material,a hardening or softening of the pipe wall, or a leaching out of constituent materials. Inany case, the mechanical properties of the pipe are changed and the ability of the pipe towithstand loads is reduced. Such degradation is not only limited to the pipe, but canaffect the joints and the various appurtenances such as the service connections, thevalves, and the hydrants.A number of authors have noted a wide variety of interpretations by utilities withrespect to a what is considered a pipe “failure” and have commented that there is apressing need to adopt standard main repair definitions and reporting procedures to helpmonitor and compare experiences (AWWA 1986a). Some utilities distinguish between‘break” and “leak” repairs, while some aggregate the two, while others record“maintenance events” which can include any repairs on a pipe. For the purposes of thisdiscussion, the designations adopted by the AWWA (1986a) will be used.Using the AWWA definition, any loss of water associated with the degradation ofthese material properties is considered a leak. A leak can be through corrosion holes orat joints or through a macroscopic discontinuity (ie. a break). In this sense a break ismerely one type of leak. A break is accompanied by a water loss which approaches theactual flow in the pipe, while smaller leaks lose so little water they cannot bedistinguished from normal demands. Breaks are especially critical to the effectiveoperation of system as they disrupt service, reduce fire-fighting capabilities, damageproperty, pose a public health threat, and can be costly to repair.The AWWA designates four types of leaks: a main leak, a service leak, a valve119leak, and a hydrant leak, with a main break and a joint leak considered special types ofmain leaks. The resulting set of definitions then apply:Leak Repair: all actions taken to repair leaks in mains, line valves, hydrant branches,and service pipes;1) Main Leak: all problems which lead to leakage of water from the main(including joint leaks, holes, circumferential breaks, longitudinalbreaks, defective taps, split bells and not hydrant, service line orvalve related leaks),a) Joint Leak: a loss of water from the joint between adjacent mainsections and not a structural problem, but a separation ofthe main sections caused by expansion and contraction,settlement, or movement of joint materials because ofpressure or pipe deflection,b) Main Breaks: structural failure of the barrel or bell of the pipe due toexcessive loads, undermining of the bedding, contact withother structures, corrosion, or a combination of these.There are four types of main breaks:- circumferential- longitudinal- holes from corrosion or pressure/blowout- split bells including bell failures from sulfur compoundjoint material2) Valve Leak: leaks at valve flanges, valve bonnets, or valve bodies,3) Hydrant Leak: leaks at hydrant branch lines, hydrant valves and hydrant barrels,4) Service Leak: leaks at taps, corporation stops, service pipes, and curb stops.The four types of main breaks are illustrated in Figure 4.2.The nature and characteristics of main breaks merits a more detailed discussion.The type of environment in which a pipe is laid is probably the single most importantfactor which contributes to its deterioration over time. A pipeline is not too differentfrom any other type of structure; it experiences both live loads and dead loads, it hasunique structural properties, and a good foundation is an integral part of the entirestructure. The degree of deterioration may or may not contribute to the occurrence of a120Figure 4.2:LongitudinalTypes of water main breaksCircumferentialBellHoleBreak TypeLongitudinalCircumferentialSplit BellStress AxisTransverseLongitudinalTransverseStructural CausesExcessive Ring LoadsInternal PressureThermal ContractionBeam FailureInternal PressureLead Substitute Joint ExpansionSource: Adapted from AWW4 1986121break, depending on the nature of the loads. Assuming a pipe has been properlydesigned to withstand any anticipated loads, there are three main scenarios which canproduce failures:1) the intensification of anticipated loads above the load carrying capacity of a pipe2) the presence of unanticipated loads above the load carrying capacity of a pipe3) the degradation of the pipe’s physical properties which decreases the load carryingcapacity of a pipe.The combination of the above three scenarios generally indicates the probability of abreak. In general, if a pipe has not experienced loads greater than its design capacityand has not undergone significant deterioration, chances are it will not produce a break.If a pipe has not experienced loads greater than its design capacity, but has incurredsome deterioration, chances are it may form a leak, but unless the deterioration isexcessive, will probably not form a break. If a pipe is exposed to excessive anticipated orunanticipated loads, chances are good it will develop a break, and will be greater if thepipe has been weakened through deterioration.The loads typically anticipated in a pipes design include the dead load due to theearth pressure, the live loads due to traffic, the internal loads due to pressure and inrecent years, the internal loads due to waterhammer. The loads and the effects of theloads can be increased over the life of a pipe. For instance, the increase in the weight orvolume of large trucks over the years can increase the vehicle loads and the number ofcycles respectively, the loss of bedding due to leaks, washout, or settlement can increasethe bending moments induced by the earth pressure, and the improper operation of asystem can increase the operating and water hammer pressures.While these loads may be the most obvious and the most easily determined in122design, there are a number of other loads which can act on a pipe but which can not bepredicted with a high degree of certainty due to the complexity of the undergroundenvironment, especially over the lengthy life of a pipe. Cyclic loadings due to frost, long-term loads induced by soil movement, and intense short term loads due to constructionequipment impacts and earthquakes can result in both localized or more wide-spreadfailures. These loads have typically been accounted for by the designers choice of anappropriate factor of safety, but changes in standards and technical knowledge over theyears have resulted in a wide variety of pipes with an equally wide variety of resistancesto such loadings. The effects of such loads are highly variable and difficult to predict, asthey can act in combination and largely depend on the integrity of the pipe at anyparticular time.A pipes ability to resist such loads can be reduced by corrosion or erosion whichact to degrade its physical and material properties. Internal corrosion from aggressivewaters or external corrosion from corrosive soils or stray current can cause both localizedreductions in wall thickness, which is common to metal pipes, or an overall reduction inpipe wall thickness, which is common to both asbestos-cement and metal pipes. Internalerosion of the pipe wall due to high velocity water can also cause localized weak spots.The effects and rate of such degradation vary with material type and the environmentalconditions.In McIntyre and Elstad’s (1987) survey of Ontario water systems, there were fourmajor causes of watermain failures: frost, construction methods, material failure andground movement (see Table 4.1).123Table 4.1: Reported reasons for water main failures in OntarioCAUSE MECHANISMS INCLUDED PERCENTAGE REPORTEDFrost Frost, Frost Heave, Frost Jacking, Other Temperature Effects 28.9Construction Methods Rocks in backfill, Poor tamping of bedding, Adjacent Construction 19.3Material Failure Corrosion (1 1.7 %), Age (8.2 %) 19.9Loss of Bedding Ground Movement (erosion, differential settlement), Poor Bedding 11.0Other or Unknown Varies 21.0TOTAL 100.0Source: Adapted form Mcintyre and Elstad (1987)Pipe deterioration is directly influenced by the stress regimes encountered. Thevariety of stresses which a pipe must resist can be seen in Figure 4.3. Bending stress maybe induced from soil movements or surface vehicle impacts and is highly dependent onpipe geometry. Axial stress may be induced by temperature changes and is influencedlargely by the properties of the joints. Shear stress may be induced by differential soilmovements and is especially important when pipes are connected to rigid structures.Ring loads may be induced by changes in external soil pressures or internal waterpressure fluctuations and depend largely on the operations in and around the pipe.The major factors influencing pipeline failures are numerous and have beensummarized into four major categories by Shamir and Howard (1979):1) the type of environment in which the pipe is laid and the associated loads,including the corrosiveness of the soil, frost and heaving, external loads;2) the characteristics of the pipe, connectors and other equipment, including quality,age, size, and type;3) the quality of the workmanship used in laying the pipe;4) the service conditions, such as pressure and water hammer.Figure4.3:StressesonaburiedwatermainJOINTbendingmomentduetofloatingstressoroverburdentorueortorstonshearstressaxialtensilestressaxialcompressivestressinternaloverpressureinternaldepressionorexternaloverpressureMBTMTMBaTMTMB:MT:T:A:C:PcIPCSectiona-aSource:AdaptedfromMoruzzi(1987)i125The reasons for a pipe break are diverse and can come from a number of sources,each of which produces a unique combination of stress conditions. Table 4.2 outlinessome of the major causes which contribute to a break and the stresses each can induce.There are four break types which are common to these stress regimes, as can beseen in Figure 4.4. The relative ability of a pipe to resist a particular stress regime islargely dependent on the pipe’s material, characteristics, the joint type, and the pipegeometry. Moruzzi (1987) compares the relative resistance of various pipe material/jointcombinations using a scale from 0 to 10, where 0 represents the least resistance and 10represents the most (Figure 4.5). While the comparison is based on the physicalproperties of new pipes and not the actual field measured performance, it still provides auseful base for discussion.Table 4.2: Stresses associated with common causes of breaksREASON FOR BREAK EXAMPLE OR SOURCE POTENTIAL STRESSESINDUCEDViolent movement of the soil earthquake, landslide, MB, MT, T, A, C, Pbuilding collapse, warSlow movement of the soil subsidence, road settlement, MB, T, Apipe settlementExcessive transmission of poor soil thickness above MB, T,direct surface loads pipe, heavy vehicle impactsLow temperatures poor pipe protection, cold P1. Awinter water temperaturesWater Hammer overpressure or depression P,,A,from various sourcesRoad and adjacent works soil disturbed over, under, M, MT. T, A,and/or adjacent to pipe.Source: Adapted from Moruzzi (1987)Figure4.4:DegreeofresistancetovarioustypesofstressTYPESOFSTRESSSource:AdaptedfromMoruzzi(1987)TYPESOFWATERPIPESANDJOINTSGREYCi.Di.A.C.STEEL’P.V.C.P.E.F.R.P.M8PRESFLNGPRESFLNGSLVEWELDFLNGPRESSQEZSLVEWELDSQEZ56664109657108MT6566610955666T559.59.54.510101.51.51.51.52.5A0609010901.501.52.5C710794.59911.5111.5P17.58910810107.587.588.5P7.5105.55.5933TOTAL38504360.53662592526.527.532.534444.54.55IS I’)Figure4.5:DegreeofresistancetovariousfailuretypesTYPESOFWATERPIPESANDJOINTSGREYC.I.Di.A.C.STEELP.V.C.P.E.F.R.P.PRESFLNGPRESFLNGSLVEFLNGPRESPRESSQEZSLVEWELDSQEZVIOLENTMOVEMENT29.542344528524917.518.52024.525.5OFTHESOILSLOWMOVEMENT101715.5278.530287.588.51313OFTHESOILEXCESSIVESURFACE17.5212123.517.5232211.510.5131615.5ACCIDENTALLOADSLOWTEMPERATURES7.514919820197.59.57.59.511WATERHAMMER152414.524.517232211.513.5121416ROADWORKS22.532273623.5434016.5171923.524Source:AdaptedfromMoruzzi(1987)128As can be seen from Figure 4.6, metallic pipes provide better resistance than nonmetallic pipes. Steel pipe with welded joints provides the best overall resistance, thoughthis is not the most common material used in Canada. Of the more common materials,ductile iron with flanged joints is the strongest metallic pipe. Asbestos cement followedby welded P.E. provide the greatest resistance of the non-metallic pipes. It is importantto note that the use of lower resistance materials such as asbestos cement and the morerecent use of P.V.C. has been widespread since the 1950s. It has also become obvious insome municipalities that these newer materials do not always provide service lives whichmatch those of the more traditional iron pipes.In Canada, the majority of pipes in need of rehabilitation are either cast iron orasbestos cement which respectively make up 85 percent and 9 percent of the pipe in theground over 30 years old (MU 1961), with the remaining 6 percent being either steel,wood, or a variety of less used materials.In A.C. pipe, water which has a depressed pH or is overly soft and thereforelacking calcium can leach the calcium components out of the cement thus releasingasbestos fibres (AWWA 1974, Davis et al 1979, Commins 1979, Toft and Marks 1983,Nebesar and Riley 1984). Thus, some utilities are having to replace A.C. lines after onlytwenty years of service (Robinson 1991, Maclean 1991).Still, when interpreting the results of Figures 4.5 and 4.6, some caution should beexercised. In some applications, lower resistance does not necessarily mean poorerperformance. For instance, in the case of slip on type pressure and sleeve type jointswith rubber ring gaskets, the lack of resistance to axial loads can be an asset, providing anon-rigid discontinuity which can move to reduce axial stresses due to thermal expansionor contraction.Figure4.6:DegreeofresistancetoleakagesourcesREASONSFORLEAKAGETYPEOFSURFACEORTYPEOFPIPETYPEOFSERVICEINPIPES/JOINTSINDEPENDENTOFCOATING(mt.and/orext.)MATERIALJOINTPIPETYPEANYTYPESOFSTRESSBITPLAACSTCIDICIDISTPVCPEFRPWFPSLSQSTCuPLAEXTERNALCORROSION61010345453101010789810679INTERNALCORROSION4109123231101010399810278AGEINGOFMATERIALS3610109109101065?1097661069LEAKAGEOFJOINTS------------109769---TOTAL132629141518151814262520+3035322836182026BIT:bituminousCl:greycastIronW:weldedSI:steelPLA:plastlcDl:ductileIronF:flangedCu:copperAC:cementorasbestosSI;steelP:pressurePLA:plasticcementLEGENDST:steelPE:polyethyleneSL:sleeveCi:greycastironPVC:polyvinylchlorideSO:squeezedDI:ductileIronFRP:flbrereinforcedplasticSource:AdaptedfromMoruzzi(1987)IS130In addition, because of the relatively young age of the plastics used in waterdistribution there is little data on the long term deterioration of these products.4.3. Age as an Indicator of Pipe ConditionA number of authors have suggested that age is not a good indicator of thespecific condition of a water main (Arnold 1960, O’Day 1984, AWWA 1986a). Arnold(1960) notes that many 100 year old pipes still operate effectively, while 50 to 75 yearold pipes require intensive maintenance. The AWWA (1986a) found poor correlationbetween pipe age and anticipated leak/break rates. However, both O’Day (1984) andMcIntyre and Elstad (1987) concede that while age cannot be used to predict individualbreak rates, it can be used to roughly indicate average break rates by age group and thusthe general condition of longer aggregate samples.While age may not be the best indicator of system condition, in many systemswhere maintenance and repair information are lacking, it is often the only indicatoravailable. For this reason, it remains a basis for preliminary national or regionalcondition assessments.4.3.1. Historical Development of Piped SystemsThe transmission of water through pipes is an old technology. In 180 B.C. thePhoenician Hellenistic Pergamon constructed a piped system that brought water from anearby mountain under a pressure of 16 to 20 atmospheres (de Camp 1990). Threehundred years later, the Romans developed the famous aqueducts which carried distantwaters to feed into systems consisting mostly of lead pipes. Lead was used extensivelyfor pipes as wood tended to split and rot; tile and concrete, though durable, were poor in131tension and thus could not withstand the internal pressures; and bronze, while strong,was hard to work and expensive (de Camp 1990). The ill effects of using lead pipes tocarry water, though suspected in the Roman times, were not positively proven untilBenjamin Franklin diagnosed lead poisoning in 1768 (de Camp 1990).The Roman systems suffered from many of the same problems faced by today’ssystems (leakage, cracking, etc.) and required a great deal of maintenance. Even aftermassive repairs through the centuries, the systems along with the Empire had all failedby 1000 AD. In Paris, the Roman aqueducts destroyed by Norse invaders in 900 ADwere replaced by marginal aqueducts around 1200 AD, but it was not until 1600 AD,after many disputes over scarce piped water, that Henry IV decreed that users must payfees large enough to support the water system. Eventually a pumping system was built toraise the river water and carry it to his palace, with the excess water being turned over tothe public. This marked the beginnings of Paris’ modern system (de Camp 1990).Over time, water systems became more elaborate and wide-spread. Early systemsin England utilized wooden “tree trunk” mains with the use of iron pipes as the materialof choice emerging in the early 1800s. Between 1812 and 1819 some 600 kilometers ofwooden “tree trunk” mains were replaced by vertically cast iron mains in London,England, many of which are still in use today (Edwards and Cox 1982).In North America, wood pipes were often installed because of the expense of castiron pipes, though the wooden pipes in many of these early systems would frequentlyleak or burst. Wood pipes were often used in areas where water supplies were plentifuland therefore system reliability less important, such as in British Columbia and Quebec,and were typically installed with the idea that once revenues were sufficient, they wouldbe replaced by cast iron pipes (Anderson 1988).132The large scale use of wood never really developed. In 1961 only 48 communitiesin Canada had wood stave pipes still in their systems, with 24 in British Columbia and 12in Quebec, representing only a small fraction of the 1,878 systems which were in Canadaat the time.4.3.2. Development of Modern Pipe Materials. Construction and Design TechniquesWhile the exact age of a pipe may not be an accurate indicator of its condition,the year or period it was installed still holds valuable clues to its condition. Over theyears, new materials and construction techniques have emerged which have improvedsome characteristics of the pipe, while degrading others.Today the majority of the pipes in the ground are iron, either in the form of castiron or the newer class of ductile iron developed in the 1960s. The relative high costand heavy weight of iron pipes stimulated the development of other materials for pipesuch as asbestos cement (A.C.) which was first introduced in North America by JohnsManville Corporation (U.S.A.) in 1929 and became popular in the late 1950s (Nebesarand Riley 1984). It had the advantages of being light weight, low cost and resistant toexternal corrosion since it did not carry electrical current. Eventually, concern overasbestos fibres and the development of plastics such as Polyvinyl Chloride (P.V.C.) andPolyethylene (P.E.) lead to the demise of A.C. pipe. P.V.C. and ductile iron remain thepredominant material types installed in distribution systems today.A better understanding of construction techniques has benefitted throughcontinued experience and research with piped systems. As an example, the importanceof continuous, tamped, select bedding material under a pipe was not totally understooduntil the 1930s and 1940s. Many of the early mains were often laid in corrosive backfill133and the AWWA’s 1927 “Handbook of Cast Iron Pipe” actually recommended blockingunder the bells, a situation which can cause excessive beam stress since the main is notcontinuously supported (AWWA 1986a).Significant advances have also been made in joint design with the introduction inthe 1930s of the mechanical and roll-on joint to replace the older lead-caulked bell andspigot and the introduction of the push-on type rubber joint in the 1940s which increasedthe seal and pipeline flexibility to reduce leakage (AWWA 1986a).Design techniques have also improved. For iron mains laid between 1908 and1939, wall thickness was based only upon anticipated internal working pressures andbedding conditions, while after 1939 external loads were also taken into account.Modern iron pipe design practises are now based upon theoretical principles such as theapplication of a factor of safety of 2.5 applied to two separate load conditions: the firstwith earth external loads and both working and surge internal pressures, and the secondwith earth and truck external loads and only working internal pressure. The greater wallthickness from these two loadings is chosen.While the development of new materials and the understanding of loading andcorrosion mechanisms have improved vastly since the turn of the century, improvementsin technology have not always resulted in improved pipe performance and longevity. Therefinement of modern design principles have meant that some factors of safety haveactually been reduced over the 1908 standards, which were quite conservative due to thegreater uncertainties of design. The 6” (150 mm) diameter pit cast iron pipe of 1908 bytoday’s standards would actually provide a factor of safety of 7.3 over working and surgepressures, and 14.8 over the external earth and truck loads (AWWA 1986a).Other advances, which resulted in greater reliability and reduced material and134installation costs, also reduced overall durability. For instance, the introduction ofductile iron in the 1960s meant the bursting tensile strength, rupture strength, and tensilestrength measures were increased by a factor of three over the pit cast pipes used from1850 to the 1930s, and by a factor of two over the centrifugally cast pipes used from the1930s to the 1960s. This increase in material strength allowed pipe wall thicknesses tobe reduced by 25 percent over the 1908 pit cast main standards. While resulting inconsiderable construction cost savings, the reduced wall thickness meant that the pipeswere more vulnerable to the effects of corrosion, which is highly dependent on wallthickness. A study has confirmed that pipes installed since the 1950s have performedrelatively poorly compared to earlier installed pipes (Andreou and Marks 1987).4.3.3. Modern Pipe Design: Rigid and Flexible PipeThere are two basic classifications of modern pipe designs: rigid and flexible.Rigid pipe is characterized by a high resistance to crushing loads, but is brittle andcracks under very small deformation of the vertical diameter, while flexible pipes, in theabsence of side supports, will deform progressively without cracking under a slowlyincreasing crushing load (Clarke 1968). Flexible pipes can deflect up to 2 percentwithout structural distress (Moser 1990, p. 4), and through this can transfer part of thevertical soil and traffic loads into a radial thrust, thereby activating the passive earthpressures on the side of the pipe which produce an arcing effect to help support theloads (Moser 1990 p. 21). A rigid pipe, on the other hand, can not produce such aneffect and therefore must be designed with sufficient resistance to carry the entirevertical load itself. Rigid pipes common to water distribution systems include cast iron,135asbestos cement, reinforced concrete, composite steel and concrete, while flexible pipesinclude steel, ductile iron, and most plastics.The different characteristics of rigid and flexible pipes mean different designprocedures are followed for each. In general, the construction techniques used formodern flexible pipes are much more dependent on proper installation versus the olderrigid pipe installations which relied more on the strength of the pipe rather than theproper bedding and compaction requirements of modern flexible pipes. Because ofspace limitations, the modern design methodology for rigid and flexible pipe will not bedescribed in detail, but a number of excellent references which describe the currentdesign practises are available (Clarke 1968, Young and Trott 1984, Stephenson 1989,Moser 1990).4.4. Other Factors Influencing Pipe PerformanceMaterial properties, joint types, and age are not the only characteristics of a pipewhich influence its structural performance. Geometry, seasonal variations, waterattributes, and system operation are also important factors influencing a pipe’s lifecycle.The geometry of a pipe is measured in terms of its wall thickness and diameter.Bending due to overburden or loss of bedding is especially critical in small diameterpipes which are susceptible to circumferential cracking due to the relatively smallmoments of inertia compared to larger diameter pipes (AWWA 1986a). Largerdiameter pipes, in general, are more resistant to external forces than the small diameters,but are susceptible to longitudinal cracking from crushing loads due to heavy vehiclesand frost effects. (O’Day 1983). A number of sources have shown that breakage rates136are highest among the small diameter pipes. In Philadephia and Denver, annual repairrates per mile for pipes 6” (150 mm) in diameter or smaller are typically 2 to 5 timesgreater than for larger pipes up to 16” (400 mm) in diameter and 7 to 25 times greaterthan for the largest pipes over 16” (400 mm) in diameter (AWWA 1986a).Break rates also tend to be seasonal in nature, being highest in the cold wintermonths especially in cities which experience extreme summer and winter temperaturevariations. Within Canada, this includes nearly all communities, with the exception ofthose in the more moderate coastal zones, such as Vancouver and Victoria. In Calgaryfor instance, between 1975 and 1980 the monthly failure rate during the cold months(November to March) was approximately five times the failure rate which occurred inthe summer months (June to September)(Caproco 1985). In the northern sections of theU.S., between 60 and 70 percent of the annual main breaks occur in the four wintermonths of November to February (O’Day 1983).The lower temperatures in the winter increase the tensile stress on mains due totemperature induced contraction, and can increase the external stresses caused by soil-moisture expansion from frost penetration (O’Day 1983). Flexible joints reduce theeffects of temperature induced stress due to contraction, but the presence of valves,services, and structures can effectively restrict movement and increase the stress. Soilmoisture expansion is a significant factor especially in areas where the bedding has beendisturbed or pipes have been weakened by corrosion.The spatial distribution of breaks is also an important characteristic of pipedsystems. Kettler and Goulter (1985) found that pipe breaks in Winnipeg actually occurin clusters, with 22 percent of all breaks occurring within one meter of a previous failure,and 46 percent of the failures occurring within 20 metres of a previous failure. Clark et137al (1982) also notes that in two utilities observed, a small percentage of the pipes havethe most problems; after 40 years, over 52 percent of the pipes had no maintenanceevents. This implies that massive system replacement may not be the best solution;rather programs aimed at the replacement of troublesome sections may be moreeffective.Apart from the characteristics of the individual pipes, factors related to theoverall system environment also contribute to increased maintenance costs. Clark et al(1982) found utilities with soft water (< 60 mg/L as CaCO3)to have 31 percent highertotal unit costs than those with hard water. In addition, higher leakage rates were foundin systems with relatively few pressure zones (such as those served on flat terrain by agroundwater source) probably due to the greater system wide impact of pressurevariances.4.5. Corrosion ProcessesCorrosion is a primary cause of a water main deterioration. Corrosion typicallyrefers to process in iron based pipes such as steel or cast-iron or ductile iron where anelectrochemical reaction between the pipe metal and its adjacent environment causes thepipe to lose its ferrous constituents. But corrosive environments can also attack cementbased pipes, such as concrete and asbestos cement, where the leaching out of the limecomponent of the cement produces a softening of the pipe wall.The corrosion process is a complex phenomena which can not be eliminated butcan be controlled. It may be uniform along a pipe or localized in nature, attacking onlya small area of a pipe. A number of excellent references are available which deal withthe corrosion of water pipes (Parker and Peattie 1984, AWWA 1989, Smith 1989).138There are three major types of corrosion associated with iron based water mains:1) internal2) external3) electrolysis from stray d.c. current4.5.1. Internal CorrosionInternal corrosion is predominantly a problem in older unlined cast iron and steelpipes. It is initiated at a discontinuity such as a scratch or rust where there is anelectrical potential difference along the pipe wall. Water acts as an electrolyte, acceptingiron ions from the anode while electrons flow from the anode through the pipe wall tothe cathodic area where hydrogen ions are combined to form free hydrogen gas or reactwith oxygen to form hydroxide ions. At the anode, ferrous ions react with water to formferrous hydroxide, which is moderately soluble, or, if the water contains highconcentrations of dissolved oxygen, ferric hydroxide which is highly insoluble. If ferrichydroxide can precipitate at the surface in large enough quantities, a tubercule will form,and in doing so will accelerate the corrosion process and cause pitting in the adjacentanodic area. A number of factors related to the characteristics of the water which eitherenhance the corrosion process, or inhibit it are presented in Table 4.3.Inhibitors generally allow the formation of a protective layer on the corrosion site,while the corrosion enhancers limit the formation of this layer and accelerate theprocess. The growth of microorganisms which feed on the nutrients in the water and offthe corrosion products can pose a health concern. There are two common methods forassessing corrosion potential: coupon tests and corrosion indices. Coupon tests involvethe periodic taking and weighing of a sample of the pipe wall.139Table 4.3: Water characteristics related to internal corrosionFactors Which Can Enhance Corrosion: Factors Which Can Inhibit Corrosion:Low pH (< 7.5) High Buffering CapacityHigh Dissolved Oxygen (> 0.3 mg/I) High SilicaHigh Total Dissolved Solids (> 500 mg/I) High CalciumHigh Temperatures (> 60 degrees F.) High PhosphatesHigh Flow Rates (> 4 ft/sec) Hard Water (>150 mg/I as CaCo3)Stagnation (< 0.5 ft/sec)Low Alkalinity (< 30 mg/I as CaCO3)Chlorine (> 200 mg/I)High Sulfates (> 300 mg/I)Hydrogen Sulfide (> 0.1 mg/I)Growth of Microorganisms (on pipe wall)Soft Water (<75 mg/I as CaCO3)High Soluble IronSource: Adapted from AWWA (1986a); Critical values from Ryder (1989) and AWWA (1989).While this is the most direct means of assessing the condition, it is quite costly and timeconsuming and is therefore often practised by only the largest utilities with seriouscorrosion concerns. The predominant corrosion indices are based on the distributionwater’s tendency to precipitate calcium carbonate and therefore resist changes in pH.Levels of hardness, alkalinity, corrosive ions (chlorides and sulfates), and a variety ofother water quality parameters are included in such indexes. The AWWA (1989)outlines many of the indices which have been developed, but the most common remainsthe Langelier Saturation Index, or LSI. The Langelier Saturation Index, or simplyLangelier Index (LI) is essentially a comparison of the observed pH with the calculatedpH at calcium carbonate saturation. While it provides an indication that a water iscorrosive, it does not indicate the degree of corrosiveness (AWWA 1989, Hunsinger et al1989). A negative LI index indicates a potential for corrosiveness, while a positive LI140indicates the potential for deposition of a scale. The Langelier Index is calculated asfollows:LI = pH- pH (4.1)where: LI = Langelier IndexpH = observed pH of waterpH8 = pH at calcium carbonate saturationand,pH8 = T + B - log (Ca2j - log (A)where: T = constant (function of water temperature)B = constant (function of total dissolved residue)A = total alkalinity (as mg/L of CaCO3)The LI index has its limitations in that it does not predict the optimum amount ofcalcium carbonate required to provide an effective protective layer without forming anexcessive layer of precipitation which reduces pipe capacity. Water of pH 6.5 to 9.5 witha positive LI is generally not corrosive, but may be under certain conditions given thepresence of other corrosion enhancers in the water (AWWA 1986a).A modified LI has been developed specifically for asbestos cement pipe which isknown as the Aggressiveness Index, or Al. The Aggressiveness Index is based on thepH, hardness, and alkalinity of a water supply. An Al less than 10 indicates water whichis aggressive, 10 to 12 indicates moderately aggressive, and over 12 indicates nonaggressive. Al is calculated as follows:Al = pH + log (A*H) (4.2)where: Al = Aggressiveness IndexpH = observed pH of waterA = total alkalinity as mg of CaCO3H = calcium hardness as mg of CaCO3141Buelow et al (1980) in a study of 10 distribution systems in the U.S. concludedthat water is not expected to attack AC pipe when the Al is greater than 11, and thathigh concentrations of some metals, such as iron, in the water may inhibit corrosionthrough the formation of a protective layer even if the Al is lower than 11.Water with a low Al, due to a depressed pH or low alkalinity or hardness, willleach the calcium components out of the cement thus releasing asbestos fibres (Davis etal 1979, Toft et al 1981 as cited in Hunsinger et al 1989). Generally soft waters arethose with hardness of 0 to 75 mg/l as calcium carbonate, while moderate waters have75 to 150 mg/i, and hard waters have 150 to 300 mg/i (AWWA 1989).Once problem with corrosion has been identified, various methods of control canbe considered. Three general techniques are common:1) the use of non-metallic pipe materials or protected metallic pipe,2) the cleaning and cement lining of the main,3) the introduction of corrosion inhibitors to the distribution water.Replacement of corroding pipes with plastic or lined ductile iron pipes is now quitecommon. When significant life is left in a metal pipe, it can be economically renovatedusing cleaning and cement lining techniques which can last up to 20 to 30 years even inan aggressive environment. Inhibitors are typically added at the treatment plant and caninclude phosphates and silicates which pacify the corrosion reaction, or lime, limestone,caustic soda, soda ash, and sodium bicarbonate which raise the pH (Hunsinger et al1989).In Canada, internal corrosion is a significant factor in cities like Vancouver whereit accelerates the corrosion of copper and iron pipes (Millette and Mavinic 1987) and inWinnipeg, Burnaby, and North Vancouver where it is attacking asbestos cement pipes142(Hickman 1984, Robinson 1991, Maclean 1991). In the greater Vancouver area, thewater is extremely soft and aggressive, with an LI of -5.1 and an Al of 7.0 (Bratton et al1986) and multi-million dollar plans for the addition of inhibitors are now under way(Mavinic 1990).Failing to deal with internal corrosion effectively can result in health concerns dueto excessive concentrations of corrosion products and metal ions, aesthetic concernsrelated to staining of fixtures and discoloration of the water, and economic effects in theform of repair and premature replacement of deteriorated pipes. The AWWA (1989)outlines an eleven step process to evaluate corrosion which includes reviewing waterquality data, potential effectiveness of treatment processes, and costs and benefitsrealized by both the utility and the consumers.4.5.2. External CorrosionThe electrochemical process involved in external galvanic corrosion, otherwiseknown as “graphitization” is similar to that of internal corrosion. A galvanic cell isproduced when the metal pipe is in contact with corrosive soils of low resistivity. Ironions go into solution at the anodic area to form rust scales on the pipe, leaving only aweak carbon skeleton. At the cathodic area hydrogen ions or oxygen ions are formed.The build-up of rust at the anode or hydrogen ions at the cathode can inhibit thecorrosion process, but the presence of dissolved oxygen in the soil can again acceleratethe process as dissolved oxygen reacts with the hydrogen ions at the cathode to formwater. Pitting of the pipe will then occur at the relatively concentrated anodic reactionsite, which is often only 5 percent of the area of the cathodic site (AWWA 1986a). Theconnection of dissimilar metals, such as copper pipe and brass service fittings attached to143iron mains, can increase the potential for a galvanic reaction. Although iron is anodic tobrass and copper, two factors make this situation less than critical. First the iron pipesare usually much larger than the smaller copper service pipe, thus there is a large anodewith a small cathode which is the weakest of the bi-metallic corrosion cells (Smith 1989).Secondly, soil resistivity may be sufficient to suppress current flow and subsequently theaction of the corrosion cell.Soil resistivity is an important factor in assessing of the corrosive potential of apipe’s environment. Soil resistivity is a measure of the average electrical resistance ofthe soil and decreases with high moisture contents and heterogeneity of the soil. Morris(1967) notes a resistivity of 1500 ohm-cm or less indicates the presence of soluble saltssufficient for an effective electrolyte.Soil resistivity combined with soil to pipe potential are the main predictors ofactual pipe corrosion (Smith 1989). The corrosion process is generally inhibited by highsoil resistivities and accelerated by high soil to pipe potentials. Soil to pipe potential isthe voltage potential between the pipe and its adjacent soil; when this potential is highand the soil resistivity is low, a corroding area is often identified. This is the mostcommon of the electrolysis testing techniques, though others are outlined by Smith(1989).The corrosion potential associated with soil resistivity varies among pipematerials. Soil resistivities above 2,000 ohm-cm are generally not corrosive to cast iron,while for steel pipes, resistivities up to 2,000 ohm-cm are extremely corrosive, between2,000 ohm-cm to 6,000 ohm-cm they are moderately corrosive, and above 6000 ohm-cmthey are non-corrosive (Smith 1989).As in Table 4.4, soil resistivity, pH, redox potential, sulfide content, and moisture144content are important indicators of the need for corrosion protection on cast iron pipes.Table 4.4: Soil corrosion evaluation rating for cast iron pipesPointsResistivity < 700 10700—1000 81000—1200 51200—1500 21500—2000 1>2000 0pH 0-2 52—4 34—6.5 06.5—7.5 0 *7.5—8.5 0>8.5 3Oxygen Reduction Potential > 100 my 050—100 my 3.50—50 my 4negative 5Suif ides positive 3.5trace 2negative 0Moisture poor drainage, always wet 2fair drainage, generally moist 1good drainage, generally dry 0*— if suif ides are present and low or negative redox results are obtained 3points shall be given for this range.Note: a total of 10 points indicates a soil corrosive to cast iron pipe andprotection is indicated.Source: Smith (1989)Calgary has one of the best documented corrosion problems in Canada. Soilresistivities have been extensively measured at thousands of locations in the city, withresistivities of 1,000 ohm-cm to 2,000 ohm-cm common (Caproco 1985). The corrosivesoils resulted in annual breakage rates in the mostly cast and ductile iron system ofnearly 50 breaks per 100 kilometres in 1979 at nearly twice the national average. Afterextensive study and the implementation of an extensive corrosion control program and145main replacement program since 1982, the rate is now down to about 20 breaks per 100kilometres per year (James and Nq 1991).A number of processes may increase external corrosion rates. Leaking pipes orabandoned services can both increase the soil moisture and flush away corrosionproducts from the pipe surface (AWWA 1986a). Soil resistivities may also be depressedby infiltration of road salts in snow melt or high tidal intrusions.4.5.3. Stray Current CorrosionThe final type of corrosion in due to the presence of stray direct electrical current.The process involves the conduction of external current from a power source such as anelectrical trolley with ground rails, through the soil and through the pipe. Where thecurrent discharges from the main, anodic corrosion will occurs.The methods of controlling external corrosion are generally directed at new mainswith considerable life remaining since there is often little economy in the protection ofold mains (AWWA 1986a). The techniques include: electrolysis control, polyethylenewraps, and plastic and epoxy coatings. Electrolysis control includes both galvanic andimpressed current cathodic protection. Galvanic cathodic protection refers to theconnection of a sacrificial anode made of zinc or aluminum which corrodes, thus makingthe entire pipe the cathode. Impressed current protection induces reversed the cathodicreaction and is used especially when significant stray current is present but has a drawback of requiring a constant power source. Polyethylene wrap is a good control, butmakes repairs difficult. Plastic and epoxy coatings are also good, but care must be takennot to chip them and provide a site for a concentrated anodic reaction.1464.5. Deterioration CriteriaThe previous sections outlined a number of factors which make simplecause/effect relationships both difficult to understand and often impossible to applywhen assessing the deterioration of water distribution systems. The complexity of thedeterioration process has resulted in the development of empirical and statisticaltechniques to predict the deterioration, though they often receive only limited applicationby utilities. Often pipe deterioration assessments and the resulting rehabilitation orreplacement programs are based on “seat-of-the-pants” techniques or generalizationsrelated to age or type of pipe, an approach that has been rejected by many authors asbeing inefficient (O’Day 1983, Andreaou and Marks 1987).Four major criteria are recommended by AWWA (1986a) to best evaluate thecondition of a water main:1) structural integrity,2) leakage,3) hydraulic conditions, and4) water quality.Since buried water mains are inaccessible, measures for the above criteria have beendeveloped using indirect indicators. These indicators are listed in Table 4.5.The degree to which municipal water utilities have formally adopted these criteriainto a pipe replacement and rehabilitation program varies widely across North America,depending largely on the utility size, the available resources, the severity of the problem,and the individual philosophy of each manager. In general though, the application offormal programs is lacking. Mays (1989) finds that in a survey of the methods used byutilities in the U.S., the application of computerized information and data base systems147related to water distribution systems is almost non-existent. While many of the surveyedmunicipalities have hydraulic simulation capabilities, none have the networks stored inan accessible data base. In addition, none of the municipalities utilize methodologies forthe optimal upgrading of aging water mains and when new systems develop, simple trialand error procedures aimed at minimum cost designs are still the most common.Table 4.5: Criteria used to evaluate water main deteriorationCRITERIA METHOD INDICATOR POTENTIAL CONSEQUENCESStructural Integrity break record high break rate - high repair costsanalysis - emergency service disruption- consumer service disruption- water contamination (from exposure)Leakage leak survey high water loss - excessive water demand- high pumping and treatment costs- pipe bedding wash-out- increased moisture in corrosive soils- water contamination (from backflow)Hydraulic Conditions flow tests low pressure and - inadequate fire flowsand models inadequate flows - service level complaints- corrosion and tuberculation indicated- undersized mains indicatedWater Quality lab tests high concentrations - poor taste and odor characteristicson water of substances - fixture stains from corrosion products- bacterial contamination in pipes- poor installation indicated- inadequate treatment indicatedIdeally, all four criteria are to be monitored and eventually input into therehabilitation decision process. However, current applications tend to focus on thehydraulic conditions and structural integrity, both of which are easy to monitor and arecharacterized by easily identifiable costs. As such, most utilities base pipe replacementprograms on high maintenance costs or inadequate flows and pressures.148While water quality is monitored on a regular basis, it is sometimes difficult todistinguish between problems originating at the source and those originating in the pipes.Typically, efforts at solving water quality problems focus on the treatment plant ratherthan the distribution system. Still, serious public health concern over water quality, ifresulting from the distribution system, can be a powerful determinant in the decisionregarding when a pipe should be replaced, although its occurrence is probably the leastfrequent.When a rehabilitation program is in place, often the least amount of attention ispaid to leakage criteria as leaks are difficult to detect and the nature of the problem canbe such that many small leaks may be spread over a large area of the system, makingimplementation of a program quite costly. Specific locations where leakage is a problemcan only be determined by sonic or other leak detection tests, while pressure and flowproblems can be easily pinpointed through customer complaints or simple flow tests.Water quality problems can also be identified through customer complaints and throughfrequent water quality tests by health authorities. Pipe sections with high repair rates aretypically identified through the utility’s maintenance records.149CHAPTER 5: CONDITION ASSESSMENT AND MITIGATIVE TECHNIOUES5.0. System Condition AssessmentThere is no accepted standard procedure yet developed which can be used toassess the condition of a water distribution system and then determine the appropriatecourse of mitigation. From the literature, effective rehabilitation programs typicallyinclude five major components:1) development of a detailed database and inventory of piping system elements,2) a detailed condition assessment (including survey-inspection for leakage and flowtesting, and break analysis) using descriptive, predictive, and physical models3) “needs study” to analyze deficiencies and set priorities,4) systematic maintenance, repair, and renovation program,5) systematic construction program to replace pipe beyond feasible repair orrenovation.5.1. System InventoryBefore a utility can even begin to assess the existing condition and future needs ofa system, a detailed inventory of the system must be developed. The development ofgraphical information systems (GIS), digital mapping, computer aided drafting (CAD),database systems and spreadsheets for personal computers has advanced rapidly andcosts have come down over the past ten years to a point where even the smallestcommunities can have access to sophisticated technologies.Rodi (1987) notes that up that to 90 percent of a municipality’s information hassome type of geographical reference. MacLaren (1983) suggests that when setting upsuch system, references should be converted to standard provincial mapping recordsystems and a block plan of each element should include the information outlined in150Table 5.1.Table 5.1: Inventory information for water distribution systems- identification code for each element- the element type (pipe length, valve, meter, hydrant, service, manhole)- the element class (trunk, arterial, local, private)- location according to reference system- pipe size and length- material type- lining or coating- date of installation- burst or leak record and location- repair record (leak, repair, relining)- complaint reports (flooding, low pressure)- initial cost or replacement costs- cost of repairs by year- maintenance and service records (flow tests, hydrant flushing, valve rotation, etc.)- condition ratingSource: MacLaren (1983)While there are programs available which incorporate such information, the useand understanding of them needs to be better promoted. The sheer volume of individualsoftware packages now available has made it difficult for small communities to decide onthe best system to meet their needs. The Government of Ontario has recently developeda computerized inventory program, WIMS, for water distribution systems which can beused as an effective basis for implementing an ongoing maintenance and rehabilitationprogram (Phillips et al 1991).5.2. Condition Assessment TechniquesThe following section will outline the two basic analytical techniques which havebeen identified and developed by a variety of agencies to assist in making optimalrehabilitation decisions. The two techniques include the more basic descriptive151techniques and the more sophisticated predictive techniques.5.2.1. Descriptive AnalysisA descriptive analysis is the most basic and analytically simple method ofdetermining baseline information regarding the condition of a water system. The mostcommon descriptive analysis involves the simple cross tabulation of deterioration criteria,such as leaks or breaks, with pipe characteristics such as diameter, location, soil type,operating pressure, or traffic counts on the road above. This analysis can determinebasic areas where future rehabilitation effort should be concentrated. A tabulation ofbreakage trends versus physical location or soil type may be useful in determining “hotspots”. To assist in determining the performance of newer pipes versus old, acomparison of breakage rates among different pipe installation periods or material typesmay be useful. Factors which are common to such tabulations are given in Table 5.2.Table 5.2: Descriptive analysis factorsDependent Variables Independent VariablesMain Breaks Soil TypeLeaks Soil ResistivityEmergency Repairs Heavy Truck VolumesPressure Loss Average Bury DepthMaintenance Costs Time (Day, Month, Year)Damage Claims Pipe MaterialCustomer Complaints Installation PeriodFailed Quality Tests Intensity of DevelopmentOperating Pressure (Absolute)Pressure DifferentialPipe DiameterPast Renovation Carried OutFreeze/Thaw Cycles Since InstalledThis type of analysis is only as good as the type and format of information152available. For instance, a utility may itself record emergency repairs, but may have totransfer the data to a useable form such as a spreadsheet or database. Other data suchas soil types, traffic volumes, or intensity of development may have to be acquired fromother departments or agencies such as municipal traffic engineering departments,provincial lands departments, or Statistics Canada. Specialized data, such as soilresistivities, must be specially gathered by the utility once the need for such data isjustified.Historical records are also invaluable. In New York City, for instance, old mapsof original stream beds which existed before the City developed are kept on hand asleaks and breaks can often surface blocks from the actual source, having travelled alongthe old stream routings. Such maps can also give important insights into soil conditionswhich can develop into specific corrosion areas.The method of analysis can also vary. It can be as simple as placing colored pinson a system map or it can incorporate widely available spreadsheet programs whichallow graphing of both bar type and linear plots, and which can accommodate moresophisticated techniques involving statistical distribution and multiple regression.Regardless of the method, the analysis is still only as good as the data available.5.2.2. Predictive AnalysisIn the past decade, a number of studies have been conducted which attempt tomodel the breakage mechanism of a water main (Shamir and Howard 1979, Clark et al1982, Andreaou and Marks 1987). Shamir and Howard (1979) found that breakagetrends in the City of Calgary seemed to increase exponentially once an initial break hadoccurred. Based on this observation, an expression for the prediction of future breakage153was formulated:N(t) = N(t0)etto (5.1)where, t time in years,t0 = base year for the analysis,N(t0) = number of breaks per 1,000 foot lengthof pipe in the base year,N(t) = number of breaks per 1,000 foot length of pipein any given year “t”,A = break growth rate coefficient (per year).In this study, growth rate coefficients were found to range between 0.05 and 0.15, whilethe number of breaks in the base year ranged from 0.10 and 0.25 breaks per 1000 feetper year, which is equivalent to between 33 and 82 breaks per 100 kilometres per year.To calculate the expected number of breaks over a given time period, equation 5.1 isintegrated with respect to time.Clark et al (1982) uses a multiple regression technique to predict both time tofirst repair and subsequent repair trends, with a repair being due to either leaks orbreaks. Repair data for the study was provided by a large utility (11.39 m3/day). Fromthe regression analysis, the time to the first repair was estimated by:NY = 4.13 + 0.338D - 0.022P - 0.2651 - 0.0983RES - 0.0003LH + 13.28T (5.2)= 0.23where, NY = number of years to first repair,D = diameter of pipe, in inches,P = absolute pressure with a pipe, in p.s.i.,I = percent of pipe overlain by industrial development ina census tract,RES = percent of pipe overlain by residential development ina census tract,LH = length of pipe in highly corrosive soil, in feet,T = pipe type (1 = metallic; 0 = reinforced concrete)154While such an equation is quite system specific and should not be used in the predictionof trends in other systems, it can still be seen that generally the larger the pipe diameterand the lower the operating pressure, development intensity, and presence of corrosivesoils, the longer a pipe will function without the need for repair.In order to predict the number of expected repairs, Clark et al (1982) carried outanother regression analysis with the resulting equation:REP = (0.1721)(eO.97)T e00D(eO865A012l) E\T SLO4(SH9 (5.3)where, REP = number of repairs,T = pipe type (1 = metallic; 0 = reinforced concrete),PRD = pressure differential, in p.s.i.,A = age of pipe since first break, in years,DEV = percent of land over pipe in low and moderately corrosivesoil,SL = surface area of pipe in low corrosive soil, in sq. ft.SH = surface area of pipe in highly corrosive soil, in sq. ft..The above regression equations can be substituted into equation 5.1 as NY in equation5.2 is equivalent to t0 (the base year) in equation 5.1, and the constant (0.0865) in thefourth term of equation 5.3 is equivalent to A (the growth rate in equation 4.8).Andreaou and Marks (1987) present another method of predicting the breakfailure patterns based on proportional hazards and Poisson-type models. This method isa further refinement in that it recognizes various stages in a pipe’s life, rather thanassuming that after an initial break, a pipe will invariably break at an exponentiallyincreasing rate.The basic approach considered by Andreaou and Marks (1987) is based on thepremise that the operation life of a piped system, like many other multi-component155engineering systems, can be represented by a bath-tub shaped baseline hazard function(see Figure 5.1). Such functions are characterized by three phases: a start-up phase, anormal operation phase, and a wear-out phase (Bury 1975). Applied to a piped system,the start-up phase is generally the one or two year period of initially high repairs whichdiminish as the initial “bugs” are taken out of the system. During this period, thecontractor responsible for the installation of the pipe is typically responsible for therepairs.Once this initial period has passed, the normal operation phase is entered into.Andreaou and Marks (1987) refer to this as the slow breaking state, with pipes havingexperienced few (usually no more than two) and infrequent breaks. The probability offailure can best be described by a proportional hazards type model, similar to equation5.1 with the form:h(t) = h0(t)e (54)where, h(t) = hazard rate (breaks per year) at time t,h0(t) = baseline hazard function, estimated from a regressionanalysis,z = vector of covariates representing pipe and environmentalcharacteristics,b = coefficient estimated from a regression analysis.Eventually a pipe will enter the wear-out phase, often after it has experienced twoor more breaks in relatively close succession. Andreaou and Marks (1986) refer to thisas the fast-breaking stage where pipes experience multiple and frequent breaks andwhere there is no apparent trend of increasing or decreasing break rate with time.Breaks at this stage can be reasonably represented as poisson arrivals at a constantFigure5.1:Typicalhazardfunctionh(x)HazardFunctionLtlBt2TirneXA-Start-UpPhase(initiatedatt0)B-NormalOperationPhaseC-Wear-OutPhaseSource:Bury(1975)I’ (11157arrival rate, with break rate independent of time and the number of previous breaks. Atthis stage, the break rate among various pipes can be highly variable and is best capturedthrough a regression-type model. Andreaou and Marks (1987) found the followingexponential model to be appropriate:p(x) = [(rt)e9 / x! (5.5)where, p(x) = probability of having “x” breaks in time period “t”,r = break-rate, estimated from the following regression:r = e)1z1+ b2z+ ... + bnzn (5.6)where, = covariates reflecting pipe and environmentalcharacteristicsbn = coefficients estimated from the regressionAs can be seen, the regression equations in this analysis are similar to those of equations5.3 and 5.4. by Clark et al (1982).The development of predictive models such as those outlined in this section isimportant in the rehabilitation decision process. Inputs on anticipated pipe behaviourare required to arrive at optimal replacement decisions as will be shown in the nextsection.5.3. Decision AnalysisThere are a number of methods currently used by utilities in making decisions onwhether and when to replace or rehabilitate a water main:1) General “Rules of Thumb”2) Economic Analysis3) Reliability Analysis4) Physical Models1585) Hydraulic Performance6) Field or In-Situ InspectionsThe extent to which a utility uses one or a combination of these techniques varieswidely. One or two techniques may be used in small utilities while combinations of allthe inputs have been developed by larger agencies into replacement scoring systemswhich can be used to rate individual pipe sections for repair or replacement.5.3.1. Rules of ThumbThe general “rules of thumb” is probably the most common method used, thoughit has severe limitations and can result in much less than optimum replacementschedules. Such techniques are typically based on the experience of the manager oroperator and are characterized by usually being quite simple and sometimes arbitrary,for instance replacing all pipes older than 40 years, replacing all pipes with at least twobreaks, replacing all asbestos cement pipes, or replacing all pipes under future roadimprovements. While these rules are often justified by some underlying economic orhealth premise, they are not always arrived at in a rational manner and can sometimeslead to sub-optimal decisions.Common “rules of thumb” based on replacing individual pipes by age, simplebreakage rules, or “gut feelings” have been discounted as inefficient by many authors.Andreaou and Marks (1987) note that the probability of a break depends on manyfactors other than time since installation (e.g. pressure, corrosivity, land development,period of installation, and stage of deterioration) and that for pipes in the fast-breakingstage, the breakage rate is actually independent of age nor previous breaks. In the casestudies investigated, Andreaou and Marks (1987) found that for pipes in the slow growth159stage, replacement decisions based on simple rules of thumb, such as replacing all pipeswith two previous breaks or replacing all pipes over 65 years old, would only reduce thebreakage rate by at most 25 percent, while the replacement of all pipes with a failureprobability greater than 5 percent, as derived from the statistical analysis, would result ina 40 percent reduction in breakage.Another “rules of thumb” example relates to the health concerns of A.C. pipewhich have not proven conclusive and in many ways are in doubt. The perception is thatasbestos fibres cause cancer when inhaled and must be carcinogenic in general. In theU.S., the EPA has seemingly acted on this perception and applied it to water qualityguidelines. Still, unless there is first some concrete evidence that the fibre concentrationis significant and, secondly, that the A.C. pipe is contributing to the fibre concentrationin the distribution water, perhaps resources and replacement dollars could be betterspent on mains which have high breakage incidents and which on a regular basis areresponsible for much higher real public health concerns. That is not to discount thegenuine concern regarding health factors, but it does illustrate how uncertainty canoveremphasize some concerns.5.3.2. Economic MethodsThere are both direct and indirect costs resulting from main breaks. The directwater utility costs include the repair and emergency crew costs, water treatment andpumping costs with lost water, and damage claims. Indirect costs include overhead andemergency police and fire protection during break, damage costs to consumers notreimbursed by the utility, service disruption costs, costs imposed on affected adjacentutilities, and the cost of traffic and public transportation disruption (AWWA 1989).160Economic methods base replacement of a main on a comparison of currentreplacement and maintenance costs to the sum of current repair costs and future costs ofrepair and service disruptions, discounted to the present. Replacement is justified whenthe money spent on repair is greater than the amortized replacement value of the mainin the ground or when the total costs of repair and replacement are minimized (seeFigure 5.2).The optimal replacement time is typically given by:t = ln [Crln(1 + r) / CbEOI / ln(1 + g) (53)where, t* = optimal replacement time,Cr = cost of replacement,Cb = cost of repair,Eo = expected number of breaks during base year,g = estimated break growth rate,r = discount rate.Predictive models such as presented earlier by Shamir and Howard (1979) are used toestimate future breakage trends. Andreaou and Marks (1987) note that pipes in a slowbreaking stage rarely require replacement by such economic criteria and suggest thatmany of the pitfalls which assume exponential or linear increase of break rates with timecould be avoided.Similar economic criteria are used to determine the timing for pipe renovation,which involves cleaning the tuberculation off the interior of a pipe with a scrubbing “pig”and then cement mortar lining the interior of the pipe to prevent further corrosion andtuberculation build-up. Andreaou and Marks (1987) suggest that a pipe should not berenovated if it is expected to develop a large number of breaks in the future. As theeffects of renovation on breakage rates, while not yet documented, are undoubtedlysmall. Thus a pipe should be replaced if the cost of replacement is less than the cost ofFigure5.2:OptimalpipereplacementschedulingCost(dollars)Time(years)t0initialinstallationtroptimalreplacementtimeTotalCostMaintenancett0rON162rehabilitation plus the cost of expected repairs after renovation:CrCreh [(CEQ(1+g))/(1+r)J (5.8)where, Cr = cost of replacing a pipe,Creh = cost of rehabilitation (or renovation),Cb = cost of break repair,E0 = expected number of breaks during the base year,g = break growth rate,r = discount rate,t = time increment.Walski (1982) provides an economic based technique to determine how to renovate linesfed directly from pumps, where there is the option of increasing pumping capacity toincrease the pressure or renovating the main to increase the “C” factor. The method isnot appropriate for small distribution lines which are sized for fire flow and are feddirectly from storage tanks, where it is impossible to significantly increase the head. Themethod entails renovating a line only when the cost for rehabilitation is less than theextra cost for pumping energy and additional equipment (ie. pumps) through thereduced “C” pipe.5.3.3. Reliability AnalysisReliability analyses on new systems are rarely done, primarily because of the lackof a universally acceptable definition or measure of the reliability of water distributionsystems. Design techniques have been developed using optimization methods, but moststill have strict limitations and are based on normal loading conditions such as peakhourly demands or peak daily demands. Very little work has focused on abnormal oremergency loading conditions such multiple fire demands, pump failure, control valvefailure, power outages, and broken links which can be associated with normal daily163operation or infrequent events such as earthquakes. No “optimization-reliability”evaluation or design technique with general application has yet been developed (Mays1989). However, the study into reliability of existing systems especially in seismic zoneshas grown, especially since the 1971 San Fernando earthquake and even more so sincethe 1989 San Francisco earthquake. In North America, there are no seismic codesrelated to water systems, though guidelines have been developed based on pastexperience.With respect to normal system operation, Andreou and Marks (1987) suggest thatreliability criteria would play a more important role than economic criteria indetermining repair versus replacement strategies. Failure probabilities obtained by thehazard models can be used to directly estimate the reliability of given links in a network,while at the same time they represent a single quantitative measure of the risk based onthe interactions of factors like age, break history, operating and environmentalcharacteristics. At the networks level, the implications of a break would also have toinclude the evaluation of a hydraulic model to assess the implications at all points on thesystem. For further developments in reliability analysis, the reader is directed to arecent publication by the ASCE (1989). However, as O’Day (1984) points out, there arestill a number of legal and technical matters to be cleared up before reliability and riskanalysis are widely accepted. It must be remembered funds are limited and prioritizingmay have to be carried out. O’Day suggests in areas of high probability of failure butlow risk, leakage control should be increased while in areas of high risk, the replacementof mains should be accelerated to reduce the impact of main break damage.1645.3.4. Physical ModelsThe estimation of remaining structural life in a main is typically used in largeutilities where human and technical resources are sufficient and pipe corrosion is a majorproblem. Physical models are developed to estimate the pipe age using deterministicrather than probabilistic means. Typically the problem is well defined and the mechanicsof the deterioration process being modeled are relatively well known. Threeinvestigations by the City of Vancouver, the U.S. Army Corps of Engineers, and the Cityof Philadelphia involved modelling the physical processes associated with pipe corrosionand are outlined in AWWA (1986). The Vancouver and Philadelphia studies bothfocused on estimating the “time to failure” of old grey cast iron water mains while theU.S. Army study focused on gas distribution mains.In 1978 the City of Vancouver initiated a program to determine the remainingservice life of 68 kilometres of cast iron mains installed in the 1900s (AWWA 1986) andlater developed models to incorporate steel pipe (Bratton et al 1986). The basis of themodel is Rossum’s general equation for predicting maximum external pit depths (Brattonet al 1986):P = Kn (10 - pH) p t (5.9)where, P = depth of pit (mils),Kn, n = aeration constants,pH = measure of soil acidity,p = soil resistivity (ohm-cm),t = time (years).Internal corrosion of unlined steel and cast iron pipes was estimated using the following(Bratton et al 1986):P = 45 1/3 (5.10)where, P = depth of pit (mils),165t = time (years).A series of 30 theoretical “pit depth versus time” curves were developed and calibratedusing field data for various conditions any particular pipe may be exposed; curves weredeveloped for soil resistivity ranges from 3,400 to 900,000 ohm-cm, pH ranges from 4.6to 9.4, and soil aeration from poor to well aerated. Field data was collected fromvarious sources: break records and staff interviews identified problem areas in the City;potential stray current sources were evaluated; water analyses were done to determinecorrosive properties; over 5,000 soil resistivity measurements were taken to identifycorrosive areas; soil sample analyses at main depth were done to determine pH, sulfides,moisture content, and redox potential; and 20 mains were physically inspected for pitdepths and uniformity of wall thickness (AWWA 1986, Bratton et al 1986). The studyfound that actual pit depths were on average about 9 percent lower than the theoreticalpredicted.The practical application of the model involves determining the age, wallthickness, soil type, and loss of wall thickness due to internal corrosion for a particularpipe. The minimum wall thickness allowed for the pipe is then determined from theforces and stresses on the pipe as well as the pipe’s material properties; for instance thestructural strength of a cast iron pipe is directly dependent on the wall thickness. Theactual versus minimum wall thicknesses are compared and the difference is applied toone of the theoretical curves to derive the “time to failure”. Once any pipe has reachedits theoretical “time to failure” it is put on a replacement program (Bratton et al 1986).Application of such a model requires intensive field investigation but has theadvantage of including factors that directly have an effect on the structural integrity of amain. Unfortunately this model does not consider external loads nor bedding related166structural conditions (AWWA 1986).The U.S. Army Corps of Engineers Construction Engineering ResearchLaboratory (CERL) developed a model which also estimates the depth of externalcorrosion of cast iron mains and the “time to failure” using a corrosion status index(CSI). The index is a numerical rating between 1 and 100, with 1 representing acompletely deteriorated main and 100 representing a newly installed one. The CSI iscalculated as follows:CSI = 100 - 100 (PAV / T) (5.11)where, CSI = corrosion status indexPAV = average corrosion pit depth (inches)T = thickness of main wall (inches)Gas mains were found to leak when the CSI dropped to 30, which correlates to theaverage depth being 70 percent of the original wall thickness. The pit depth can bedetermined by either physically digging up the main and measuring, through electricalpolarization techniques, or through mathematical model estimates. Estimates of the CSIhave also been developed based on factors such as the rate of pit growth with time,prediction of maximum from average pit depths, average number of years to leaks, andthe effects of main coatings as well as soil pH, resistivity, sulfides, and moisture. CERLhas developed a computerized pipe management program for low pressure gas mains,and with modifications to account for internal corrosion and higher operating pressures,could be applied to water pipes.The Philadelphia Water Department has developed a prototype water maincondition assessment model (CAM) which estimates the internal and external loads andthe internal and external corrosion losses, then predicts the remaining wall thickness andcomputes the safety factor. The system was developed for older grey cast iron mains167which represent the majority of the City’s distribution system. Because the model wasdeveloped for grey cast iron pipe, which is structurally rigid, it can not be applied toductile iron pipe. The model includes eight modules: pipe characteristics, environmentalconditions, corrosion losses, estimated loads, maximum allowable loads, design loads,safety factors, and structural condition rating.Each module has a variety of inputs and outputs. The “pipe characteristics”module includes the input of pipe diameter, age, length, depth, original wall thickness,material, joint type, breaks, working pressure, and cleaning and lining history. The“environmental conditions” module includes the input of physical conditions that mighthave an impact on the life of the main such as leakage, number of abandoned services,pavement type (rigid or flexible), proximity to stray current (ie. rapid rail substations, ord.c. rail). The “corrosion losses” module estimates corrosion losses for internal, galvanic,and electrolytic corrosion based on modifying the City’s average corrosion rates for theenvironmental conditions. The “estimated loads” module assesses the surge pressure,external loads, truck loads, thermal stresses which result from the environmentalconditions. The “maximum allowable loads” module determines the maximum ring load,pressure stress, and beam moment the pipe can withstand based upon its unit strengthand remaining wall thickness. The “design loads” module determines the stresses causedby five standard load scenarios as outlined in AWWA C1O1-67. The “safety factors”module estimates the individual safety factors by comparing the computed stressesderived from the design loading to the unit strengths or maximum loads at failure forseven conditions with varying internal and external pressures and loads. The “structuralcondition” ratings are based upon the minimum of the seven safety factors: mains withsafety factors less than 1.0 are classified as “questionable”, while mains with safety factors168greater than 1.0 are classified as “satisfactory”.As can be expected, the previously mentioned models can only be considered asprototypes as every utility has very site specific characteristics and problems. The abovemodels should be used as guidelines, but must be calibrated and verified using local datato assure that all results are realistic.5.3.5. Hydraulic PerformanceWhile structural considerations associated with the previously describeddescriptive, predictive, and physical models are extremely useful in assessing a waterdistribution systems structural condition, many utilities rely on hydraulic performancecriteria rather than structural criteria to assess the condition of their systems. Flow andpressure are the main performance criteria whereas breaks are the main structuralcriteria. Although both structural and hydraulic criteria can be used mutually give themost reliable assessment of system condition, the ease and availability of hydraulicprograms makes them most often and in many cases exclusively used, especially insmaller utilities where neither the records nor the resources required to develop astructural model are readily available. Structural models are typically limited insophistication to the utility deciding from experience that a main is experiencing toomany breaks and should be replaced.The advent of the personal computer has meant that hydraulic models which canmodel a communities water system in a variety of consumption and future scenarios areeasily affordable and relatively simple to operate. Programs such as WATER 1 and1 WATER - Copyright Municipal Hydraulics169SDP 2 allow the user to input pipes, nodes, pumps, and control valving to simulateentire water systems. Once input, limitless demand scenarios can be run and re-run,each taking only minutes versus what used to take hours manually. The output can thenbe calibrated with actual flow and pressure measurements obtained at access pointsthroughout the system, such as at fire hydrants. Any sections of the system which showreduced flow capacity or large pressure drops can be readily identified. Bottleneckswhich can reduce fire fighting potential or future system expansion can be found; pipessuspected of having undergone tuberculation can be verified and scheduled for cleaning;zones of negative pressure which could be a source of system contamination orpotentially produce pipe collapse can be catalogued and rectified. Still these programsrepresent only one of the tools required to properly manage water distribution systems.5.3.6. Field or In-Situ InspectionsWhile network models can help detect sections with gross capacity deficiencies,more specialized techniques are required to pin-point the more subtle deficiencies suchas slow leaks and localized reductions in wall thickness. The technologies involved donot involve excavation but rather are from an analysis with the pipe “in-situ”; as such,detection methods are often more of an art than an exact science.A number of devices and technologies developed by the oil and gas pipelineindustry can and have been successfully applied to water pipelines. Graf (1989) outlineshow ground penetrating radar has been used successfully to detect leaks in gas mains.Price and Wolf (1989) outline and review a number of techniques including visual2 SDP - Copyright Charles Howard & Associates Ltd.170inspections, above ground electrical surveys, acoustical methods, proof testing, and T.V.inspections. Above ground electrical surveys involve the measuring of electricalpotentials, the electrical resistivity of the soils or the localized leakage of an inducedcurrent from a pipeline to detect coating damage or areas of potential corrosion. Suchsurveys rely on a great degree of interpretation and are most effective if carried outsystematically over a large area. Acoustical methods include sonic and ultra-sonicdetection of cracks and voids; while effective they can be hampered by noisy pipelines orhigh background noise in the underground conditions.Proof testing by isolating, pressuring up, and measuring the pressure drop of asection of pipe is also very effective in detecting leaks, though over long stretches it canbe difficult to locate the leak. In addition, the pipeline must be put out of servicetemporarily.The running of remotely controlled video cameras inside small diameter pipes hasfor years been standard practice with sewer lines, though it has had only limited successwith water lines. Water lines lack the easy access points such as manholes and are oftenriddled with obstructions such as valves. The resolution of the video cameras is usuallynot great enough to allow location of the small cracks usually responsible for slow leaksand again, the water line must be taken out of service to allow a video inspection.Visual inspections from inside a pipe can only be carried out in large diameter mains;the mains must be shut down and access gained to allow inspections.Instrumented pigs have also proven popular in the oil and gas industry, though asVerNooy and Jordan (1983) point out the technology is still developing. Instrumentedpigs include any self-contained device which travels through a pipeline propelled by theflow of fluid carried in the pipeline. They measure various parameters and record the171data for later evaluation (ie. pipe geometry, leaks, wall thickness, etc.). With the adventof newer, stronger pipe materials, pipe wall thicknesses have decreased resulting insignificant material savings. However, as VerNooy and Jordan (1983) point out, thereduced wall thickness has meant that pipe breaks now occur after many cycles ofoperational stress reversals rather than in the initial hydrostatic tests. Thus the need forgood leak detection is becoming more rather than less. VerNooy and Jordan (1983) andAilman and Dilay (1983) list a number of such “smart” pigs used in the oil and gasindustry: calliper pigs can continuously measure the inside diameter of a pipe; leakdetection pigs can measure pressure, temperature, or material differences at leaklocations. Although promising there are inherent of characteristics which makeapplication of these techniques difficult; water lines have many service connections whichcan be wrongly identified as leaks; ultrasonic leak detection pigs are promising but againmay be subject to interference in noisy or dirty pipes; a variety of miscellaneous pigswhich can video, measure pressure and temperature, flow detection and magnetic fluxflaw detection have also been developed though their application to water distributionsystems is not yet extensive. Most of these techniques would best be applied to long,straight transmission mains rather than the congested networks of distribution mains.To avoid the perils of crisis management which leads to poor customer serviceand unnecessarily high repair costs, regular leakage surveys and water budgetcalculations can be based on an inventory assessment in larger utilities or on a regularinterval basis such as ten year periods.5.4. Mitigation TechniquesThe development of municipal technologies versus other technologies in our172modern world has not been rapid due in large part to the lack of competing groups andthe aversion to risking new technologies among municipally run utilities. Within thewater distribution industry the two main upgrading technologies used today are by nomeans new; a deteriorated water main can either be cleaned and mortar lined orreplaced. Cleaning and lining is typically limited to those pressure and flowcharacteristics which can still be lined without significantly reducing the hydrauliccapacity below the required demands. Replacement of a pipe is still by far the mostwidely used method of improvement. It is largely carried out using conventional open-cut techniques with an excavator such as a backhoe supplemented by human labor toassist in the placing and aligning of pipe and the compacting of backfill. New“trenchiess” technologies have developed in recent years to take advantage of the existing“hole in the ground” but for the most part practical applications have been in the gravitysewer industry. Still trenchiess techniques hold promise with pressurized waterdistribution systems, at least on long lengths with few obstructions such as valves, elbows,or service connections.Apart from the supply side solutions aimed at improving the structuralcomponents of the distribution system, non-structural techniques are also carried onexclusive of the pipe. Typically they involve operational techniques aimed at reducingever increasing system demands through water conservation methods. While effective inthe longer term, such techniques do not solve the short-term problems of leaking,deteriorating pipes.1735.4.1. Pipe ReplacementThe open trench method involves the excavation of a trench, laying the pipewithin the excavation, and backfilling. The width of trench is governed by safetyregulations aimed at protecting the workers in the trench. Minimum excavation sideslopes combined with the depth required will set the width of surface disturbance. Foran installation depth of 2.5 meters, the width of trench at the surface can vary from 4 to6 metres, with the entire disturbed width being approximately double this with a spillpile. Such excavation can effectively shut down an entire roadway. Shoring theexcavation walls can minimize the excavation width, but the operation can still close oneto two lanes of traffic and significantly disrupt access to commercial or residentialestablishments adjacent to the street.When calculating the overall cost of a project, a utility typically includes the directcosts which will have to be paid out. Such costs include the actual costs of constructionplus any third party costs such as compensation for disturbance of business, the cost ofservice diversions, or the cost of structural damage due to trenching in poor soils orareas with a high ground water table. What typically does not enter the cost analysis arethe intangible or indirect social costs which are borne by the community but are not paidfor by the utility. Such costs include those associated with traffic delays, reducedpavement life, additional wear on diversion routes, and the environmental effects ofincreased noise, vibration, dust, and unsightly excavation conditions.Trenchless technologies which have developed since the mid-1970s aim atminimizing or eliminating altogether the need for such disruptive surface excavations andat satisfying the implicit “Do Not Disturb” sign now being posted by society (Fedotoff etal 1990). Actual construction costs associated with trenchless methods often exceed174those of conventional methods. However, the main argument for the use of trenchiesstechnologies lies in the significant reduction of the indirect costs associated with open-excavations. In a recent report to the FCM outlining guidelines for the use of trenchiesstechnologies on Canada’s sewer infrastructure, various studies were identified whichfound the social costs associated with traffic delays were estimated at 1.25 to 2.6 times,and in one case 10 times, the capital construction costs of the sewers (City of Winnipeg1991). The report sites a case in the Barbados where, even though the capital costsassociated with the conventional method were 25 percent lower, upon inclusion of thesocial and environmental costs which would have negatively impacted on the touristindustry and the local residents, the trenchless technique was found to be more costeffective.Trenchiess technologies are considered to be in the developmental stages in NorthAmerica, having developed largely in Japan, Germany, and the United Kingdom (City ofWinnipeg 1991). A number of “NO DIG” conferences have been held in North Americasince 1985. The ASCE and the City of Montreal have included trenchiess techniques intheir sewer rehabilitation manuals, and a number of projects have successfully utilizedthe methods. In Toronto a 430 metre length of 750 mm diameter storm sewer on KeeleStreet was successfully completed for just under $1,000,000 during the winter of1990/1991 (Vardin 1991). The project involved poor soils and a high water table whichwould have required dewatering and well points had the conventional method beenemployed. The contract tender allowed one of three methods: open cut, conventionaltunnel, or microtunnel. The low tender proposed use of the microtunnel technique, aresult which signals such techniques are becoming competitive from a capital coststandpoint with conventional techniques. The project minimized traffic disruption,175maintained access to all neighborhood driveways, reduced truck traffic through reducedexcavation volumes, minimized noise as only small portions of pavement had to bebroken up, and saved $60,000 in pavement cut repairs over the open-cut method. Inaddition, job safety and working conditions were much improved over conventionalmethods. While Vardin (1991) notes the method may not be applicable to everysituation, it is especially appropriate to deep (> 6 metres) excavations or where trafficdisruptions would be unacceptable.A number of centres have made trenchless techniques common in theirrehabilitation programs. In Singapore for instance, trenchless techniques are the onlyones allowed for sewer construction rehabilitation in built-up areas (City of Winnipeg1991). In North America, the City of Houston has been the main proponent oftrenchless techniques. By early 1990, the City had installed over 24 kilometres in fouryears using the microtunnelling method. This represents over 80 percent of the totallength utilizing this technique in the U.S. (Fedotoff at al 1990).While the trenchless techniques offer significant improvements over traditionalmethods, their application has been limited more to gravity sewers rather than pressuredistribution and transmission mains. Slipliing and cement lining techniques haveremained the most common renovation methods, but have several disadvantagesincluding loss of inside diameter and therefore hydraulic capacity. In communities whichhave a large number of small diameter deteriorating pipes which are already insufficientfor fire flows, such techniques will not be useful. Probably the most promising techniquefor in situ replacement of cast iron pipes is by the bursting method which is especiallysuited to cast iron’s brittle nature. Unfortunately, such methods are not as suitable tothe stronger and more flexible ductile iron pipes.176Unlike sewer lines, water lines are more problematic with respect to suchtrenchless techniques. Water distribution systems contain obstacles such as valves andtight elbows and lack manholes for easy access. In addition, trenchless techniques stillrequire that open excavations be carried out to allow reconnection of services. However,such techniques still hold promise though they will not solve the root of today’s problemswhich is the political will and financial means to carry out such operations on a nationwide and systematic basis.5.4.2. Trenchless MethodsTrenchiess methods includes a wide variety of techniques aimed at therehabilitation and renewal of pipe structures. Rehabilitation techniques aim at leavingthe existing pipe in place and improving the inside surface of the pipe. Renewaltechniques focus on removing, displacing, or abandoning the existing pipe in favour of anew pipe.Rehabilitation techniques for sewers typically involve pipe lining methods andinclude (City of Winnipeg 1991):1) Insituform (polyester resin) lining2) Fibreglass lining3) High density polyethylene (HDPE) lining4) Phenolic resin lining5) Reinforced mortar coating6) Polypropylene liningInsituform lining is a patented process and product where a sock-like product comprised177of a polyester needle felt tubing saturated with a thermosetting resin is pulled into anexisting pipe following an initial pipe inspection and preparation. The “sock” is thenpressurized into final shape with cold water which is subsequently heated to initiate thethermosetting reaction which creates a new pipe within the old.Fibreglass lining is used in structurally sound sewers with diameters typicallygreater than 1100 mm. Segmented fibreglass reinforced plastic liners are inserted andmechanically jointed within the pipe. Any spaces between the liner and the pipe arethen filled with grout.High density polyethylene liners involve the installation of continuous lengths ofHDPE pipe with an outside diameter smaller than the inside diameter of the existingpipe. A variety of installation techniques are used including insertion into an existingpipe of a collapsed or smaller diameter PE pipe which is subsequently heated orpressurized to conform to the existing pipe wall.Phenolic resin lining involves a technique similar to mortar lining which isconimorily done on water mains. The technique involves first scraping and cleaning theinterior wall of a pipe to be treated and then spraying the surface using centrifugalapplication of a phenolic resin rather than a mortar. As with mortar lining, thistechnique is used on pipes which are still structurally sound.Reinforced mortar coating involves the centrifugal application of a mortar to aprepared interior surface of a pipe followed by the insertion of steel reinforcing and afinal application of mortar to hold the reinforcing in place.Polypropylene lining is similar to that of HDPE lining, though the differing linermaterial characteristics make each suitable to slightly different conditions.Trenchless techniques aimed at replacing the pipe in the ground rather than178improving it are quite varied and include the following (City of Winnipeg 1991):1) Tunnelling2) Pipe Jacking3) Microtunnelling4) Auger Boring5) Impact Ramming6) Directional Drilling7) Impact Moling or Pipe Bursting8) Jet Cutting9) Thrust Boring10) Wet Boring11) Slurry BoringTunnelling has been used for decades and involves the large diameter installationswith secondary liners. Tunnel sizes are such that personnel may easily enter.Pipe jacking also involves the use of large diameter installations. A rigid pipe isjacked into place, usually with hydraulic equipment, just behind boring or augeringequipment at the face of the excavation. Waste excavation is hauled via a conveyor backthrough the pipe.Microtunnelling is one of the techniques developed in recent years which allowsthe installation of small, non-personnel entry sized pipe using a remote controlledexcavation, mucking, and steering operation. Standards and techniques are becomingquite well developed. Stein et al (1989) provides a detailed reference for this technique.Auger boring is a common method for railway and highway crossings involving ahorizontal excavation without remote steering. A steel casing is typically installed toallow insertion of the distribution pipe.Impact ramming is similar to the auger bore, except the steel casing is rammedinto place with pneumatic equipment rather than bored.Directional drilling involves the horizontal excavation using remotely controlledsteering equipment. It is distinguished from microtunnelling in that it involves the179smaller diameters of non-entry pipes, typically 150 mm or smaller.Impact moling and pipe bursting involves the insertion of a conical device into anexisting pipe. The device fragments the existing pipe and then displaces the fragmentsinto the surrounding soil. The insertion pipe typically follows immediately behind theoperation and may be equal or greater in diameter than the original pipe size.Jet cutting is a variation of the directional drilling method involving the use of ahydraulic jet to excavate the hole.Thrust boring, wet boring, and slurry boring are variations of the auger boringmethod.Fedotoff et al (1990) provides an excellent summary of the current trenchiesstechniques including the typical pipe diameters, spans, and materials associated with eachtechnique as well as the accuracy, applications, and examples of each.While physical pipe rehabilitation techniques are the most obvious means ofmitigating the current problems, operational techniques can aim at reducing the demandson overworked systems or conversely utilize under-worked systems to reduce the needand costs for expansion. The premise of such techniques is that systems should beutilized to their optimum and that demand on such systems should be reasonable. Thusthe concepts of demand management, which include conservation and realistic, equitableprice structures, fall into this category. In addition, effective land use planning with theobjective of utilizing the existing infrastructure to its optimum level can be a component.5.4.3. Demand ManagementWhile the techniques associated with the efficient and effective upgrading of thepipes in the ground are valuable, they really only represent half the long term solution.180Rather than focusing on repairing the supply systems, demand management focuses onoptimizing consumption and peak flows which run through the system. It is based on animplicit philosophy that the amount of water a consumer uses should not be open-endedand that policies should be put in place which discourage such a case. Municipal watersystems would represent only one component of an overall scheme which would includeother major water uses such as irrigation, industrial, and power generation.Canada does not have an over-abundance of fresh renewable water. Canadaoccupies 7 percent of the world’s land mass and nearly an equal proportion of itsrenewable water at 9 percent, but of this about 60 percent drains north and 90 percent ofthe population is within 300 kilometres of the southern border (Environment Canada1987). In essence, while Canada is not short of water, most of it is not necessarily whereit is needed. This situation is made worse by the fact that while Canadians enjoy theamong the lowest prices in the world for water, they are among the highest consumersper capita.Demand management stresses the reduction of unrealistically high consumptionrates and a more effective distribution of human, financial, and natural resources withinthe water sector. Traditional water system development has largely hinged on a viewthat water is a requirement to be met rather than a resource to be conserved. As Postel(1985) argues:“Planners have typically projected future water demands based on the historicalrate of growth in per capita water use and the projected population. They thenplan to meet this estimated demand by drilling more wells or building newreservoirs, and expanding the capacity of their water and wastewater treatmentplants. Rarely have planners focused on reducing water demand as a way tobalance the long term supply/demand equation.”The tools required for an effective reversal of this trend vary widely and would181target a broad range of groups, from the individual consumer in their home to the utilityitself to the local, provincial, national, and even international political forums. Tate(1990) summarizes the techniques into three basic groupings:1) Economic Techniques2) Structural and Operational Techniques3) Socio-political Techniques5.4.3.1. Economic TechniquesEconomic techniques as applied to the municipal water management aim at usingwater pricing policies to influence the level of water demand. The techniques arecentred around the economics of price elasticity. In general, the higher the absolutevalue of elasticity, the greater a given change in price will have on demand. Thedemand curve for domestic water is inelastic over the initial quantity of water used,meaning any price changes over this range of consumption will be very ineffective atreducing the water demand. Intuitively this makes sense as the initial water used in ahousehold is considered essential to life. Above this, as water demand increases, thewater uses become less essential and the price elasticity of demand increases. Waterpricing methods can have the greatest impact in the form of reduced demand with useswhich are both a highly consumptive and highly elastic such as lawn watering. Tate(1990) notes that lawn watering accounts for 30 percent of total domestic usage and canhave elasticities in the order of - 1.0, meaning a 50 percent increase in the price forwater would result in approximately a 50 percent decrease in consumption. Some typicalprice elasticities related to water supply can be found in Table 5.3.182Table 5.3: Typical residential price elasticitiesWATER USE ELASTICITYResidential (composite) - 0.225Domestic - in—house - 0.260- lawn watering (western Canada) - 0.703Average Day - 0.395Maximum Day - 0.388Source: Tate (1990)While the results of pricing changes may vary considerably among communities, theinformation presented does show some significant savings could be realized. As Tate(1990) emphasizes “Realistic water pricing, in the sense of recovering the full costs ofwater infrastructure, including repair, upgrading, and expansion costs, is the key factor inestablishing demand management as a major tool in managing water resources.”Currently, not only the price but the price structures are thought to contribute tothe high demand for water in Canada. Flat and declining block rate structures do notpromote conservation and actually promote waste, as the consumer is paying either nomore or less per unit volume used respectively for increasing volumes of water Tate(1990) notes that in Canada, of the 591 residential rate structures surveyed incommunities, 442 were flat or declining block representing 75 percent of the total; of the532 commercial rate structures surveyed, 357, or 67 percent, were also flat or decliningblock. The remainder included constant unit rates, which charge the same per unitregardless of volume used, and increasing block rates, which charge more per unit forlarger volumes of water used. Increasing block rates, which are considered to be thebest from a conservation view-point, were incorporated into less than 2 percent of thetotal commercial or residential rate structures.183Tate (1990) goes on to argue that low water prices and rate structures whichpromote inefficiency and waste result in the undervaluing of the water resource. Suchundervaluing has hindered technological advances within the water industry and Tateconcludes that realistic pricing would make savings due to technological improvementsmore valuable, thus prompting a greater emphasis on such improvements. The basis forsuch a realistic water rate structure would:1) assure no revenue shortfall,2) meet an efficiency criterion and treat customers in an equitable manner.Tate suggests a two-part tariff, with a fixed charge to cover overhead and administrationto be shared by all customers, and a constant commodity charge per unit of use, basedupon the marginal cost of water supply and theoretically meeting the efficiency criterion.The mechanism required to implement realistic pricing for the water commodityis the water rate schedule. Water demand management focuses on making water rateschedules reflect the true value of water and recovering the true costs which are incurredwhen delivering it. There are two basic types of water rate structures which are used bymunicipalities and other water agencies: flat rates and volume based rates.Flat rate schedules require no metering as rates are fixed, regardless of thecustomer or the usage, and are generally based on covering operation costs. Flat ratescontribute to the greatest amount of wastage and excess demand since the marginal priceof water is zero (price of and additional unit of water over and above current use), thusthere is no incentive to conserve. In the 1986 survey across Canada, 47 percent of theresidential schedules were flat rate and were concentrated in smaller urban sized groups(Tate 1989).There are three main types of volume based rate structures a community can184adopt: constant rate, declining block, and increased block. All three types requiremetering of the consumers water, but not all are effective at promoting conservation.The constant rate structure charges a fixed rate per unit volume of water regardless ofamount used. Declining block rates have successively reduced unit prices for higher“block” volumes of water consumed. Increased block rates have successively increasingunit prices for higher “block” volumes of water consumed. The volume of these blocksare set locally and can vary by class of consumer (eg. commercial, residential). Of thethree methods, the declining block is the least effective at encouraging conservation, yetit is used in 52 percent of the residential rate schedules surveyed in 1986, whileincreasing block schedules, which are the most effective at promoting conservation, areused in only 2 percent of the surveyed schedules (Tate 1989).But the effectiveness of rate schedules can not be compared on the criterion ofconservation alone. The American Water Works Association (AWWA 1983)recommends that rate making practices be assessed against the four criteria of:1) cost recovery2) equity3) economic efficiency, and4) local acceptability.Cost recovery includes the municipalities effectiveness at recovering the complete cost ofmaintaining, operating, and upgrading their systems through water rates.Equity, though a somewhat ambiguous concept to define, basically incorporatesthe need to assess whether the consumers are being charged in proportion to theirreceived benefits. Based on the equity criteria, the AWWA recommends declining blockrates, as administrative costs which are incurred by all consumers are recovered in the185higher cost initial blocks. However, Tate (1989) argues that other interpretations of theconcept of equity can result in flat rates, where everyone is charged equally, or increasingblocks, where the extra system capacity required by large users is paid for by these usersthrough the higher rates charged in the increasing blocks.The economic efficiency criterion states that a systems must meet its given needsat the lowest cost. Tate (1989) points out that both flat and declining block rates failthis, as the marginal cost of water is zero or declining, thus producing overuse andartificially high system costs over time.The final criterion, local acceptability, is probably the most important factor in thesetting of rates and accounts for the widest variability in rate schedules across Canada.This is where the politics of local government comes in, as most municipalities choosethe more acceptable pricing schemes such as flat rate or declining block, which areintuitively simple and perceived to be most equitable (Tate 1989).There is a broad variation in rate structures and the resulting prices acrossCanada. Volume based rate schedules are common on the Prairies, accounting for 86percent of the residential schedules and 95 percent of the commercial. To illustrate theeffects of this on price, the mean price for 35 cubic meters of water is only $7.97 inNewfoundland and $31.91 in Manitoba. As can be expected, water rates are generallymuch higher on the Prairies than in the east or on the coasts (Tate 1989).5.4.3.2. Structural TechniquesStructural and operational techniques include metering, retrofitting, using dualsystems, and repairing infrastructure leaks for example. In Canada, only about 50percent of connections to municipal water systems are metered, a great impediment to186instituting pricing schemes based on water use. Based on an estimated 6.7 millionconnections in Canada, this would mean over 3 million meters would have to be installedat a cost of between $600 and $700 million dollars (Tate 1990). Tate feels this would bea good value compared to the $8 to $10 billion required for current infrastructure repairand upgrading and when coupled with price re-structuring would result in significant costbenefits to communities. Several cases have shown water metering and realistic pricingto result in reductions in water use by 15 to 30 percent (Flack 1981, Loudon 1986, Davey1987).Retrofitting high water use devices such as shower heads and toilets with watersaving devices also falls under this category. Such techniques have been shown to savean average household 20 percent in water consumption over conventional devices and upto 40 to 50 percent on individual devices such as low-flow showers (Robinson 1980,Barclay 1984). In addition the energy savings associated with heating the water can besubstantial (Postel 1985).Another useful structural development is the dual water system which splits thehousehold water into two separate sets of piped supply, one potable for cooking,drinking, and other functions requiring high quality water, and one sub-potable forsanitation, irrigation, or fire protection. Such systems can save on the water treatmentchemicals required for potable water. “Grey water” systems which recycle water kitchenand shower water for use in flush toilets and lawn irrigation have been found to save upto 39 percent over conventional systems (Haney and Hagar 1985).5.4.3.3. Sociopolitical TechniquesThe final category of demand management tools which Tate (1990) outlines are187the sociopolitical techniques, the major ones being:1) water pricing2) public education3) privatizationWater pricing, as outlined earlier, forms the backbone of any demand managementpolicy and works toward self-sufficient utilities rather than ones supplemented by generaltax revenues.Public education through the media, educational institutions, and action groups isan important factor in bringing such programs to fruition. The most difficult change willbe a change in attitudes and a realization that water and the infrastructure associatedwith it is valuable. Robinson (1980) estimates that education programs alone couldaccount for a 10 percent decrease in consumption.Privatization is a third technique being investigated in North America to assistailing municipal utilities. Tate (1990) notes that some studies have shown thatprivatization could save municipalities 10 to 30 percent. Proponents argue that theprivate sector would be more efficient and effective at managing and improving systems,and more importantly would be more effective at raising capital than the public sector.Precedence are cited in France where private utilities have operated systems since WorldWar II which now serve over 60 percent of the French population. These systems havebeen operated efficiently and substantial innovation has been associated with themanagement and development throughout the years. Tate (1990) notes that, while notmaking an assessment of privatization within the Canadian context, a number ofadvantages and disadvantages can be found. The advantages for a municipality typicallyinclude a decrease in the financial burden, cuts in administrative costs, and the188opportunity to take advantage of technological innovations. Disadvantages include theloss of direct control by municipal authorities, environmental quality control, and publicownership of water systems. However, Tate notes that some proponents stress that suchdisadvantages could be overcome through effective contract negotiations.Water demand management in practice will not be without its problems. Currentinfrastructure needs require a significant amount of capital and new pricing structureswill be required to assist in financing such. In addition, the costs associated with thesupply of water are relatively inelastic to demand (Loudon 1986), and as such will not bereduced a substantial amount even for significant reductions in demand. Given that themajor financing requirements will not decrease in the short term, any program which iseffective at reducing demand will not necessarily be effective at reducing the revenuesrequired by the system. Since charges on consumptive flows are the only means a utilityhas to raise revenues, it is reasonable to conclude that every unit decrease in demandmust be accompanied by a similar increase in price to generate the same amount ofrevenue. To meet current capital requirements plus the additional requirements neededfor upgrading, revenue levels must increase rather than decrease. To accomplish this,the average consumer will see an increase in their overall water bill regardless of theirconservation effort and decreased consumption level. As Loudon (1986) notes,“Customers will understandably not be pleased to see their efforts rewarded with waterbills which don’t decrease.” Loudon termed this entire situation the “conservationconundrum”, and emphasized that this lack of real short-term financial reward toconsumers can be an obstacle to the implementation of conservation programs aimed atlonger-term benefits and consequently make municipalities wary of adopting them. AsTate (1990) notes these problems may require short-term financial assistance from senior189levels of government.Tate (1990) feels the practical application of water demand management based onthe installation of water meters and realistic pricing mechanisms will lead to a greaterdesire to conserve. Depending on the magnitude of price increases and the elasticity ofdemand, short term water usage would fall by about 25 percent and level out around 15to 20 percent in the long-term. Financial capital allowing repairs and leak detectionwould be generated, though in the short-term some temporary financial assistance tomunicipalities would be required to overcome the initial effects of reduced demand.Eventually, demand forecasts used in facility planning would be lowered, delaying orpostponing the need for infrastructure expansion, and lowering the operation andmaintenance costs as pumping energy costs and treatment and waste treatment costs arereduced. Tate hypothesizes that urban centres might become more dense as newdevelopers and new developments become responsible for the full marginal cost of watersupply. Eventually, with a more realistic value of the water resource realized, residentsand industry would be prompted into more conservation conscious measures.5.4.4. Land Use PlanningUndoubtedly the single most important factor influencing both short and longterm infrastructure requirements is how people want to live in their community. Howpeople want to live is nearly solely reflected in the communities’ land use policies andprominent zoning classes, be they low density, sprawling, large lots carrying single familyresidences or intense high rise developments. As such, infrastructure in itself is a means,not and end, to achieving these policies and as such infrastructure development is nearlywholly determined by external factors.190Much has been written in planning about the efficiency of dense developments,especially residential developments and the savings which can be realized from reducedinfrastructure requirements relating to transportation and utility services. Infilling ofvacant lands within urban areas and increased subdivision densities can reduce watersystem requirements on a per capita basis. Theoretically, densification can increaserevenues for a utility granted the existing system capacity is such that it can handleadditional users. Undoubtedly, in small communities where densification could have thegreatest impact there would be resistance toward intensified land use from residents whoare typically living in the community for the very reason it is not as urban or dense. Inthe larger cities the potential for increasing densities varies, with many cities possessingdensities which are already quite high. Still, infilling and increasing densities wherepossible while still retaining the character of a community can help alleviate some of theproblems.Revenues from development cost charges per meter of water lines installed can beincreased from increased development densities. In some cases, the newer parts ofexisting systems could be densified to help pay for the rehabilitation of the older parts ofthe system. With proper planning, the older parts of the system could be rehabilitated toa point where they can support additional capacity and densification to eventually fundsfor other parts of the system. However, any increases in density must be handled withcare so as to avoid straining existing systems which are already overworked or in adeteriorated state. This type of solution could work best in areas where the existingsystem does not immediately require significant rehabilitation, where sufficient capacityexists for domestic demand and fire flows, and where the community will accept it.Unfortunately, the latter citeria often dictates land use densities, and the issue of191efficient water system management does not enter the debate.Another major drawback to this approach when looking at the current situation isthat it is a long-range solution and not a quick fix. It may require years of debate and achange in attitude for a community. In many cases it is likely people would rather payhigher water rates rather than have the character of their community changedsignificantly.Still, densification is not a guarantee that it will solve the problem. Waterdistribution system deterioration still remains a significant problem even in the largerdenser cities, where service densities can be 5 or 6 times those of smaller communities,but it can reduce the per capita impact of system rehabilitation when needed. It wouldstill be beneficial for small communities to adopt this type of policy so that when systemdeterioration becomes significant in the next 10 or 15 years, the impact can be lessened.192CHAPTER 6: DEVELOPMENT OF A POLICY FRAMEWORK6.0. Goals of a Comprehensive Capital Works Management PolicyThe current infrastmcture funding crisis requires a comprehensive capital workspolicy which is sensitive to the realities of the political atmosphere, the physical nature ofthe piped systems and the historical developments which led to the current situation.Policies must ultimately involve all levels of government, associated interest groups, andthe public. With water distribution systems, the main focus must be on improvingmanagement at the local level. This chapter will formulate a framework to illustrate theshort-comings of the existing situation and develop a rationale to improve it.From the review in the previous chapters, there emerges three major goals whichshould ultimately form future water utility policy in Canada, especially as it relates tocapital funding:1) Maintain high levels of service: in order to maintain existing health, economic,and lifestyle standards, the level of service Canadians have grown accustomed tocannot be allowed to diminish in the future; the immediate problems includingthe current capital works backlog and the reduced public interest in “lifeline”systems must be overcome; demand management can assist in conserving thewater resource and preserving the longevity of existing systems already in theground; long term renovation and replacement plans can assure adequate servicelevels in the future;2) Promote greater self-sufficiency among utilities: the reality of the currenteconomic situation, with all levels of government burdened with heavy debt loads,means utilities will have to look less to senior government for capital financingand more toward the users; the undervaluing and continued subsidization of water193systems will eventually have to stop and the true costs of system upkeep will bepassed on through increased water rates both in the short and long terms; thechallenge will remain one of maintaining service in an equitable manner bothamong individual consumers and their communities, recognizing each has differentabilities to pay; depreciation of water lines will have to be recognized and systemreplacement planned for;3) Promote increased research: research into the implementation of replacementscheduling, decision making, and rehabilitation techniques must be continued; theimprovement and application of already developed technologies and models mustbe promoted especially in the many communities still using inefficient “seat of thepants” techniques to manage the water infrastructure; new constructiontechnologies which are less disruptive and more cost effective must also be furtherdeveloped to aim at the specific needs which will develop in Canadiancommunities in the near future;4) Develop rational service standards and rehabilitation programs: both must beflexible enough to meet the needs and characteristics unique to Canadiancommunities: condition assessment standards and regional condition monitoringprograms need to be rationally developed to both guide communities in their owncapital improvement programs as well as protect communities from increasedliability; programs and standards must be flexible enough to adapt to communitiesof varying size, resources, and environments; technical and funding assistancefrom senior levels of government will still be sought depending largely on theresources available and ability to pay of the respective communities.As can be seen in Figure 6.1, any national policy aimed at solving the current problems194Figure 6.1: National Water Distribution PolicyONPUTSOUTPUTAPPLACATOON IFEEDBACK- flexible- inventory- standards- local— lawfulIIsoCo =POLTD CAL- equitable- accessible— sustainable- practical1ECM CAL- reflectiveof systemlife cycles- p.c. basedCAACflANFiSCAL IECONOMC— self—sustaining- long andshort term+ + INATBONAL WATERDDSTRDBUTON SYSTEMPOUCYjr fNAflONAL- standard setting, monitoring, fundingIIPROVNCUAL- standard adoption, monitoring, fundingLOCAL- administration, reporting, construction195with water distribution systems should be based on a number of inputs:1) Socio-political: equity among users paying for the service must be balanced withthe political desire to provide access to good, wholesome water for both individualusers, who require a minimal amount of water for good health, and individualcommunities, many of which must deal with lesser abilities to pay and extremeenvironmental conditions; policies must also recognize the current sensitivity toenvironmental impacts and be practical from a cost, standards, andimplementation perspective;2) Technical: policies must be reflective of the physical lifecycle of water systems andincorporate current management technologies developed to manage andrehabilitate water systems; the broad availability of p.c. based data andmanagement systems in even the smallest communities can be taken advantage of;3) Fiscal / Economic: policies must take into account the current fiscal reality inCanada while trying to improve the funding and revenue generation mechanismsto assure adequate resources will be available to maintain adequate levels ofservice in Canada in both the long and short terms;4) Canadian: policies must reflect the inherent environmental, communitydevelopment, economic, political, health, and constitutional characteristics uniqueto Canada; programs developed in the U.S. or Europe, while useful as guidelines,will not be directly applicable to the specific problems in this country.All the above inputs will be incorporated into a national policy which delineates theroles and responsibilities of the national, provincial, and local authorities. Feedback andmonitoring of the programs implemented under the policy will be integral to the policyin order to gauge the progress of programs and to improve the policy itself.196This chapter will elaborate further on each of the four main goals and discuss thebasic factors for input into a progressive national policy aimed at alleviating theproblems of treated water distribution systems in Canada.6.1. Maintaining high levels of service in CanadaWhile there is a real short-term rehabilitation backlog due to the pulling back offunding assistance by senior levels of government from areas where traditionally all levelsof government once shared, new policies must not only focus on eliminating this crisisbut must also address impending long term problems. If current policies are anyindication, as a society we seem to be either very naive to the fact that our infrastructuresystems are going to eventually wear out or in a state of perpetual denial that problemswill ever really develop. The attitudes, roles and involvement of the public, government,and interest groups must be reassessed to optimize use of the current physical systemsand assure adequate levels of service in the future.6.1.1. Eliminating the Current Capital Works BacklogMuch of the publicity over the past decade has focused on the enormous capitalimprovement backlog which has now developed. This problem has developed over theyears due to a combination of policies at all levels of government and thus must besolved in as rational and shared a manner as possible since it did develop out of ashared course of events, decisions, and responsibilities. The short-term solution willultimately involve increasing expenditures to clear up the backlog. Unfortunately, thesources of funding are by no means assured and up until the recent change ingovernment, no commitment had been made at the federal level. While the FCM197recommended a 1/3 1/3 1/3 split among the local, provincial, and federal lines, theprevious federal government resisted based largely on reduced spending policies and arationale that water systems and indeed municipal infrastructure rehabilitation were notpart of the federal mandate. There was no precedence for such assistance to an areatraditionally considered to be local and provincial domain.However, insomuch as the federal mandate would be served in such areas as jobcreation, regional economic development and protection of the national water resource,the federal government could justifiably assume a portion of the required costs. The newLiberal government has recognized this and has announced a program which is a start tosolving the problem, but judging by the scale of the commitment, it is by no means theultimate solution. While the cost to be shared by local government should be significant,any cost-sharing program must recognize that part of today’s problem stems from a lackof direction in the past by senior government whose programs of loans and grants rarelyencouraged local municipalities to manage or save for eventual replacement. Much likea parent handing out weekly allowances to their children without giving any advice onhow to save for a rainy day or on how to properly spend, or how to prepare for theeventuality the allowance would be cut off, the federal and provincial funding programsinstilled a perception that senior levels of government would always be there to provideneeded capital. Local governments have come to depend and even expect such grantsand loans and, intern, undercharge for water service. Although it can be argued thatmuch of the assistance came when times were good and that no one, including the seniorlevels of government, could have foreseen the cuts adopted over the past years in thename of fiscal responsibility, it is obvious the mounting problems of system deteriorationare not wholly a local problem. Any programs aimed at clearing up the current funding198shortfalls therefore should involve a level of federal involvement and even moreprovincial involvement.Regardless of the commitment or the cost sharing formula, the basic philosophy ofany program should be to clean up the mess made by all parties in the past and worktoward proper long-term care in the future. In the short term, which may be consideredas ranging anywhere from 3 to 10 years, three major problems must be dealt with:1) Financial Assistance to Municipalities - coming to grips with the financing of thebacklog of capital improvements already identified by the FCM and assisting localmunicipalities in managing financial resources to do so;2) Improving Information Systems - setting common standards and guidelines forpractice by municipalities to better identify required rehabilitation and betterestimate the overall scale of the impending problems.3) Providing Information to Consumers and Managers - information on the currentproblems and the required solutions must be introduced into the psyche of bothusers and providers.6.1.2. Developing Long Term Goals and PlansLonger-term goals should focus on utilities becoming more self-sustaining throughimproved long term planning. Consideration must be given to better systemmanagement, the setting up of long-term capital replacement funds for both new andexisting works, and the overall reduction in consumption and therefor the demand onalready taxed systems. National and regional monitoring should be enhanced to allowtracking of trends and the setting of national bench marks for improvements. Educationmust be continued to allow consumers to realize the benefits and allow for futuremodifications.Regardless of the course of action taken, two salient points can be drawn from199the reality of physical deterioration of systems over time:1) Work must eventually be done: capital work to improve existing systems must bedone and be continually done in the future, over and above current levels, to clearup the backlog of existing problems and to maintain the existing systems at properlevels of operation within the context of increasing standards and accelerateddeterioration;2) Money must eventually be found: more rational funding policy will be required toproperly carry out the additional capital work as existing levels of funding areinsufficient to keep ahead of the deterioration, resulting in maintenanceoperations which are not being effectively utilized.From the previous discussions, there appears to be five major areas which need tobe improved and incorporated into longer range water system management plans:1) Information: system-wide monitoring of the physical nature and state of repair ofindividual water distribution networks must be improved; effective informationmanagement must be integrated to monitor problems locally and to chart programprogress on both a regional and a national level;2) Standards: policies should maintain minimum national levels of service as theyrelate to health and economic development; they should recognize that historicallocal levels of service can vary with respect to non-health related concerns andstandards;3) Funding: policies should incorporate rate structures which balance ability to payand equity concerns with the concepts of user-pay and demand management; theaim should be at both curbing waste of the water resource and maintaining properlevels of funding for the inevitable future requirements; program cost sharingamong all three levels of government should be implemented based upon eachlevel’s individual mandate served and their historical role in the development ofthe individual systems;4) Education: information on water conservation, actual costs of service, and thecurrent state of repairs associated with these systems is necessary. The “out ofsight, out of mind” and “water is cheap” perceptions must be put to rest; policymakers, consumers, and the public in general must all be aware of the true valueof the water resource and the extensive infrastructure required to bring qualitywater into the home;5) Research: technical improvements in both new construction and rehabilitationtechnologies must be sought; management programs should be improved,promoted, and implemented.200Any new policies should recognize that Canadians pay among the lowest waterrates in the world, and are second only to the U.S. in per capita water consumption.Over the years, the design community has accommodated this high demand bydeveloping standards which satisfy it. However, with rising construction costs and fewerfunds to build, it only makes sense to reduce demand and thereby reduce the systemrequired to meet it. In the long term there could be a down-sizing of some of thestandards and an extension of the service life of many existing systems. Demandmanagement would be an invaluable tool to meet this end.6.1.3 Reacquaint the Public with Lifeline SystemsThe real solutions must focus on the real problems. The real problem with waterdistribution systems is not characterized by rapidly collapsing structures, nor one ofimmediate and widespread risks to health, nor one of fraudulent neglect. Rather, theproblem emanates out of many incremental decisions, physical processes, and changes instandards and attitudes over the years. The problem is one of slow decay versusimminent catastrophe. Public and political attitudes have shifted priorities and fundingaway from basic infrastructure needs.Accurate recognition of the need to rehabilitate a water system by a utility doesnot necessarily guarantee the work will be carried out. The current funding shortfallsare not only due to increases in the need to rehabilitate, but political and public concernwhich does not match this increasing need to rehabilitate the systems. As Grigg (1988,p. 62) notes, the natural force in the political marketplace is to defer capital needs. Asin most publicly controlled utilities, the criteria for making decisions about when to carryout capital works used by the operational side of a utility do not necessarily match those201of the funding side.Utilities should raise the awareness of the true value of such “lifeline” systems andthe need to at least maintain existing service levels. When water distribution systemswere first developed, the users were aware of the huge benefits and were willing toallocate significant resources to develop them. However, today we have become soaccustomed to the presence of these systems, which have been maintained without agreat deal of expense, that it is very distasteful to think a water bill should ever have toincrease by two or three fold. The realization of the true benefits and the real costshave been lost over time.Introducing information to consumers and system managers on the currentproblems and the required solutions must be carried out through multi-media educationprograms. On water bills and in the media, issues must be raised including the nature ofthe capital works backlog, the undervaluing of water, and the need for conservation,especially within the current public context where ecosystem impacts and sustainabledevelopment issues are forcing us to rethink our past habits, especially as they apply togrowth and new development. To mobilize support for such programs, both thepoliticians and the voters must recognize the facts of the issue.Competition among public programs for funding and the general trend away fromincreased levels of spending means future funding levels from higher levels ofgovernment will not increase drastically, though there an indication they will increasemarginally. To better assure that there is going to be some increase, utilities in allcommunities, and especially the small ones, should consider improvements to systemmanagement which allow better identification and more effective mitigation of systemdeficiencies. The adoption of such techniques at the local level can increase the202confidence of higher levels of government concerned that any moneys made availablewill be spent properly. Better management will also instill confidence in the users thatsystems are being operated in the most effective manner and any users fees are beingspent wisely.One fact which must be realized in any policy discussion must be the underlyingreason we have such extensive water distribution systems in Canadian communities.These systems did not merely develop out of a need to increase our everyday comfortand convenience or our standard of living, nor was their initial development necessarilyspawned from public health concerns. Although these were indirectly satisfied, the earlydevelopment of water systems, as well as much of our existing infrastructure, wasspawned out of economic concerns, particularly protection from fire. The developmentof the physical infrastructure produced employment and contributed to the prosperity ofthe nation. The importance of this cannot be diminished as any investment in waterdistribution systems is an investment in the economic fabric of Canada. Frequently thisis overlooked and, as gauged by public opinion, and recent media coverage, is often lostin the discussions.6.1.4 Reassessing the Major RolesWhen identifying the problems associated with water distribution infrastructure, itis not difficult to implicate nearly all levels of the political spectrum, not to mention thedesign, planning, and educational communities. Rather than lay blame for the problemsof the past, it is going to become more and more evident that perhaps the roles of eachmust be rethought. Any national policy development should consider matching specifictasks and responsibilities to appropriate bodies; Table 6.1 outlines the potential roles of203agencies currently involved with the provision of water infrastructure.Table 6.1: Roles and OrganizationSupply ManagementArea Sub-Area Major Role Secondary RoleInformation Gathering/ Standards and Guidelines Provincial NationalCondition Monitoring Systems Overall Monitoring of Systems Provincial NationalInformation System Development Provincial National/LocalAssessment of Existing Packages Provincial LocalImplementation of Information System Provincial LocalTraining and Technical Support Provincial -Feedback Provincial Local/NationalInformation Uploading - Regional Provincial -Information Gathering - Regional Provincial -System Monitoring - Local Local -Information Uploading - Local Local -Information Gathering - Local Local -Rehabilitation/Replacement Standard Setting:Models for Decision Making - National: ie. health, economic dev. National -- Prov: ie. construction, design Provincial -Assessment of Inputs:- System Condition Data Provincial Local- Construction Techniques and Costs Provincial LocalPrioritizing of Needs Provincial -Model Development Provincial National/LocalFeedback Provincial Local/NationalModel Implementation Local ProvincialProvision of Capital Major Upgrades to National Standard National Provincial/LocalMajor Upgrades to Provincial Standard Provincial LocalLong Term Capital Replacement Local Prov./NationalMinor/Normal Upgrades Local -Demand ManagementSub-Area Major Role Secondary RoleRate Structure Setting Guidelines and Standards Provincial NationalCost Analysis (O&M, replacement) Local ProvincialRate Setting Local ProvincialConservation Public Awareness Provincial Local/NationalIncentive Guidelines Provincial -Implementation with Incentives Local ProvincialEducation Conservation National ProvincialProvincial/Local True Cost of Water and Systems Provincial National/LocalUser Pay with Equity Concerns Provincial National/Local204As can be expected, all encompassing tasks such as standard setting and overall programmonitoring should be carried out at the national and provincial levels, with raw datacollected locally via individual system monitoring. Program implementation, though itmay be coordinated at the provincial level, should be carried out locally by operationsstaff familiar with the water systems.Senior levels of government can provide funds and technical assistance in theareas associated with overall national interests. Small utilities which have neither theresources not the technical ability to develop and implement independent programs ontheir own could benefit most from technical agencies within the government or fromfunding to allow consultants to develop programs at the local level. Provincial andfederal governments should play a leading role in developing standards and guidelinescritical to the standardization of any programs. Senior governments can also helpthrough seed money or administrative assistance in the pursuit of innovative fundingmechanisms. Still, any financing from senior government should be conditional upon theapplicants completing an overall system assessment and a general plan for managing thesystems in the future.Involvement need not be limited to government agencies but should includeprofessional and technical bodies such as the AWWA, the Western Canada Water andWastewater Association, or special interest groups such as the FCM or FACE asoutlined in Table 6.2.Regardless of the organizational input into any national policies, the aim of allorganizations should be to optimize the use of the existing pipes in the ground. Whilestructural techniques can increase the useful life of pipes, non-structural demandmanagement techniques can effectively involve the individual consumer to reduce205demand and thereby postpone or even eliminate the need for major capital expansions.In doing so, the usefulness of system components are maintained, eliminating the need toabandon components that had not reached the end of their physical lifecycle.Table 6.2: Examples of organizational input and involvement at various levelsType Local Provincial/Regional NationalGovernment Municipal: Provincial: Federal:- council - Dept. of Environment - Dept. of Environment- waterworks staff - Dept. of Health - Dept. of Health/Welfare- Dept. of Municipal Affairs - CMHC, PFRA- Dept. of Public Works- Funding/Grant AgenciesWorkshops/meetingsAWWA affiliatesUDI affiliatesPublic Local surveys/meetingsInterest Groups Chamber of CommerceUniversitiesWorkshops/meetingsFCM, FACECSADemand management can also increase consumer awareness of the true value ofthe water resource through conservation-based advertising and through morerepresentative water pricing schemes. Reduced flows will soften the impact andoccurrence of crisis type situations, and reduce the dependence on crisis typemanagement in the water supply field. Reduced demand can also result in loweroperational pressures and reduced breakage occurrences.6.2. Improved Replacement and Decision GuidelinesWhile increasing the levels of spending on water distribution, and indeed allinfrastructure systems, will ultimately solve problems, the effectiveness of any spendingprogram will depend largely on the effectiveness of the decision making tools andproblem identification methods adopted. Within water distribution system management,206the development, implementation, and standardization of such decision making tools issadly lacking.From the literature it is evident that there are two basic shortcomings of thecurrent “seat of the pants” approaches:1) Non-representative of the physical systems: utilities have not necessarily adoptedpipe management techniques which reflect the characteristics of the pipe’slifecycle; in order to maintain credibility and assist in the competition for scarcerfunds, rational approaches to “what to fix when” must be based on an evaluationof accepted performance, structural, and external factors which can accuratelycharacterize the condition of a pipe; in addition, the real consequences and risksassociated with inaction must be demonstrated to the decision makers.2) Inadequate allocation of resources to meet the identified rehabilitation demand:once a component is rationally designated for rehabilitation, there must besufficient funds in place to allow the required work; currently due to reducedpublic and political interest, much of the work which must be done is not beingdone, with the public resources being directed elsewhere.Under ideal conditions, a water distribution system would involve an ideal system servingan ideal consumer in a ideal world. The system would be constructed with idealinformation about the users and the expected loads and conditions; it would be operatedand maintained in the most economically and socially acceptable manner, supplyingwater of the highest quality to consumers who use it only in optimum amounts. Ofcourse, even in an ideal world, the system would wear out, but would be rehabilitated or207replaced in the optimum fashion at the optimum time. Unfortunately we do not live in aperfect world, so less than optimal circumstances always exist.Marks (1987) identifies six approaches commonly adopted by utilities in effectingcapital replacement decisions:1) minimalistic: “if it ain’t broke, don’t fix it”2) crisis management: “fix the worst first”3) opportunistic scheduling: “if someone else will pay, fix it”4) condition monitoring: using pre-specified maintenance standards5) risk management: repair facilities at risk6) preventative maintenance: cradle to grave concern for maintenance.In communities today, the first three strategies are the most prevalent, though they arethe least effective. The last three strategies, while more effective, are generally limited tolarger municipalities which have the substantial management resources required todevelop and implement such management techniques. That is not to say mostcommunities have not necessarily made a conscious decision to adopt less than adequatestrategies, but, as is often the case, they have not because of resource constraints orbecause of relatively minor problems to date. However, in order to maintain publicconfidence in the management of such systems, it is prudent that communities adoptimproved techniques especially considering the inevitable fact that the systems will wearout and that help from senior levels of government may not always be readilyforthcoming.Much of the current problem in infrastructure financing can best be characterizedas a distancing between the needs of the operational side identified at the local publicworks level and the reality of the resources allocated by the funding side as represented208by the political process. The operational side’s mandate is to provide the user with thebest possible service from the best possible system, while the funding side has to factorin a fiscal limit set by public and political bodies. This limit is directly related to theimportance of the services provided relative to the other services provided by thecommunity. When the reality is as it is today, where water supply is rarely at the top ofthe public priorities list, there will always be a discrepancy between the operationalneeds and the financial resources made available.In this age of competition among programs, the operational side has to strengthenits case and raise the profile of the needs in water supply. The criteria used by theoperational side and the funding side must not only be brought closer together, but atthe same time must be improved to assure optimum rehabilitation decisions. A numberof options already discussed can be at the individual utility’s disposal to affect thedecision making process.A utility can improve its management image by focusing on the improvement oflong-term rehabilitation plans rather than just short-term patching, and by attempting tominimize the impacts and costs of maintaining such long-term plans. New technologiessuch as trenchless technologies can help to alleviate the negative image of water systemrepair, which is often one of large scale disruption and inconvenience. The ultimate goalof future programs would be to assure that at all times adequate levels of service aremaintained.All too often, the only time level of service becomes an issue is “after the fact”following a major crisis. Unfortunately, the wrong time to discuss effective managementis following a catastrophic fire or an earthquake, though it is often the most common.There needs to be a change in attitudes regarding water systems away from the209perception they are merely pipes in the ground toward one which recognizes their trueimportance as “lifelines”.The increase in information can be accompanied by public promotion to raise theprofile of water utilities and promote awareness through direct educational involvement.Both the costs and the benefits of large scale programs must be delineated. The settingand adoption of service level standards can be enhanced by public awareness and directpublic involvement.Together, the options outlined can help to bring the funding agency criteria andthe operational criteria closer together, and closer to a rational optimum. Whenanalyzing the long-term management needs of water distribution systems, mechanismsmust be put in place to track three important criteria:1) the current state of the existing systems from both a physical and a fundingperspective;2) the optimum replacement and/or rehabilitation schedule on local and regionallevels;3) the required need defined as the difference between the current state and theoptimum.The current state includes the specifics related to the physical properties of the systems,including pipe sizes, material types, and age, the environmental concerns specific toinstallations within Canada, such as soil types, frost concerns, and properties of the watercarried, and finally the political and social environment related to funding and theprominence of such systems in the social psyche.Effective system management techniques which recognize the inherent nature of awater distribution systems life cycle have been developed but now need to be publicly210adopted on a large scale over communities of varying sizes, needs, and levels ofsophistication. The development and wide-spread implementation of such programs willnot occur overnight. For example, while the development of pavement managementsystems (PMS) started in the late 1960s and were quite well developed by the mid 1970s(Hudson and Haas 1976), many municipalities have just begun to implement them. Still,the implementation of computer based information and decision systems for watersystems can be accelerated due to the presence of existing models such as PMS, and dueto the rapid advances in personal computer software regarding database management,graphics programs, and geographic information systems (GIS). Today, most of even thesmallest communities have personal computer capabilities for at least administrativepurposes.Currently such optimization techniques are being used successfully only in a fewof the largest utilities. Real innovation can come from widespread implementation ofsome of these methods rather than through the development of new techniques whichessentially re-invent the wheel. As Grigg (1988) points out, “the challenge in publicworks is mostly making the best use of what is available, rather than developing newapproaches that are not already available to business and industry”.Once the current state and the optimum requirements are identified, the systemneeds can be calculated and resource allocation planned. Comprehensive programswhich involve the monitoring and feedback both at local and regional levels must bedeveloped such that the unique resources available at each level are best utilized. Whileresearch into technologies which minimize the costs and negative impacts of constructionand rehabilitation needs to be promoted, appropriate technologies must be matched tohighly variable and typically unique rehabilitation circumstances which can vary greatly211depending on the size, geography, public attitudes and perceptions, and monetaryresources of any particular community.6.2.1. Determining when to replace a pipeUtility managers are responsible for managing the built systems within theorganizational and fiscal constraints as set out by the governing body. As Grigg (1988, p.264) points out, any solutions to the current infrastructure management problems will bemulti-disciplinary and multi-faceted involving organizational management of theorganization, project management of the projects, and operational management of thebuilt systems.In all facets of the management of a water distribution system, there must be animplicit understanding of the nature of such systems. Systems are characterized byincremental development over the years with different people, contractors, agencies, andstandards involved in each stage of the development. With such a mode of development,gaps, inaccuracies, and imperfections in the information will develop. In addition, eachindividual component added to a system has a unique lifecycle which, combined with allthe other components, contributes to the aggregate lifecycle of the overall system.Unfortunately, due to the incremental development of continuous systems such as waterdistribution networks, identification of overall system lifecycles is not easy. Still, within awater system or any individual component, there are three major phases whichcharacterize the lifecycle:1) Phase I: planning, design, construction, and start-up,2) Phase II: normal operation and maintenance,3) Phase III: wear-out, rehabilitation, and replacement.212A basic recognition of these three phases is essential to determining the optimum timeand sequence to replace such systems. Unfortunately, professional technical andengineering education programs tend to focus heavily on the initial design phase, withlittle effort directed at how to properly operate and maintain such systems for long lifeand even less effort at rationalizing the overall lifecycle.The planning, design, and construction technologies associated with phase I arewell entrenched and typically well funded in Canada. In the planning stages, empiricalrelations based on historical trends in demand are applied to population and economicforecasts to estimate anticipated demands. These demands are then used to calculatedsystem requirements given a set of performance criteria or “standards.” For a waterdistribution system, typical performance criteria may entail maintaining a residualpressure at the curb of 45 m of head under maximum daily demand, or maintaining 30 mof head under maximum hour demand or maximum daily demand plus fire flows (Adamsand Heinke 1987). A hydraulic analysis then determines pipe sizes and pumping andstorage requirements. Once these components are sized, they are designed to satisfystructural standards given the anticipated set of environmental loads. Structural designstandards for the pipes are typically set by the Canadian Standards Association or asimilar body (AWWA, NSF, CGSB), while structural standards for facilities are designedto meet various national, regional, and/or local building, electrical, and structural codes.Construction entails a number of uncertainties, ranging from quality control, totechniques and materials incorporated, to environmental conditions encountered. Inmany cases, the quality of construction is the overriding factor which determines theproper functioning and continued life of a system.Phase II follows the construction and breaking-in of a new system, and includes213general maintenance and normal operation. According to the FCM, attention to theseoperations is not an issue as it has remained at an adequate level over the past 20 years.Components are inspected, breaks are repaired, faulty components are repaired, etc.Phase III involves the wear-out of the system and is typically the most overlookedphase in a water distribution systems life. Proper identification and managementprocedures at this phase are significant not only because they assure the properfunctioning of existing systems, but they can provide valuable information andtechnologies which can refine both the design and maintenance phases.Both Phase I and Phase III are similar in that they often involve large capitalexpenditures over short periods of time and as such require more time and site specificplanning inputs than the maintenance phase. However, Adams and Heinke (1987) notethere are important characteristics which do distinguish the two phases: Phase I involvesthe application of deterministic standards or criteria to uncertain conditions, while PhaseIII involves the application of probabilistic standards or criteria to certain conditions.This distinction is based on the premise that in developing a new system, uncertaintyexists about actual system growth or environmental conditions, but once a system isinitiated and operated, the actual loads and demands are realized. Unfortunately,without a program to monitor or provide feedback, the benefit of increased certaintyfollowing phase I can be lost.Monitoring is an important management component in all three phases and is theonly means of accurately distinguishing between the phases. For instance, the boundarybetween the maintenance and rehabilitation phases can be somewhat arbitrary, basedlargely on predetermined scales of work or budgetary limits. Monitoring can also beimportant in the setting of overall design standards for future new works, and to the214operation of effective maintenance programs, but undoubtedly the one area where it iscritical is in the setting of effective rehabilitation criteria and the effective allocation ofrehabilitation resources.A broad base of inputs and criteria (ie. economics, service, reduced frequency ofcrises) can be included to standardize when a pipe moves from phase II into phase IIIand what action is specifically required to improve the pipe. Regardless of the inputs, inorder to determine when a water distribution system or any one of its componentsdemands attention, three basic factors should typically be monitored:1) Structural Adequacy of the system,2) Performance Adequacy of the system,3) External Factors.The “structural adequacy” of a system relates to the actual pipe being able to maintain itsphysical integrity, and is often measured in terms of high breakage rates or leaks and isrealized in terms of high maintenance costs. The nature of the structural adequacy isbest illustrated by the “bath-tub” type hazard function discussed Andreou and Marks(1987). As can be seen, this function closely resembles the overall lifecycle of a pipedsystem, a lifecycle very much determined by structural considerations. Water pipes wheninitially installed often have a high repair event for the initial years when in the break-instage (see Figure 6.2). This is followed by a long period of regular maintenance wherenone or one or two breaks may be experienced by the pipe. Eventually, the pipedeteriorates to a point where it enters the “wear-out” stage or the “fast-break” stagewhich Marks (1987) notes occurs after about 40 years and is typically signalled by twoFigure6.2:BreaksorRepairsh(x)Watermainstructurallifecyclex-Start-UpPhase(normally1to3years)II-NormalOperationPhaseIll-Wear—OutPhaseIIIIIITime(years)IS216breaks occurring in relatively rapid succession. In this stage, the break rate no longer isrelated to the age of the pipe, but becomes relatively constant.When determining the optimum replacement time for a pipe, ideally one wouldbe able to locate the condition of the pipe on the hazard curve and through an economicanalysis similar to the one presented by Shamir and Howard (1979) determine theoptimum replacement time, Unfortunately, common “seat of the pants” techniquesovershoot or undershoot this optimum. Further, since the point at which a pipe entersthe fast breaking stage varies with pipe material and conditions, replacement schemesbased simply on age or maximum breakage rate criteria can result in prematurereplacement, which represents an under-utilization through loss of potential use, or latereplacement which results in a loss of maintenance resources. The combination of lessthan optimum replacement scheduling combined with public funding which does notnecessarily correspond with the utilities identified needs can result in less than optimumreplacement timing (see Figure 6.3).“Performance criteria” are typically represented by system pressure for the givensystem demand, typically measured as flow (Adams and Henke 1987). Over the life of apipe, deterioration due to tuberculation can reduce the effective diameter of a pipe andincrease the roughness. Often these effects are combined into an equivalent reducedHazen-Williams “C” value.The basis and extent for measuring performance criteria must be unambiguous toavoid ineffective or inefficient actions being taken. For instance, the criteria may bebased on providing a minimum service pressure on a peak hour with a high degree ofreliability over the entire system, in which case the measurement should probably betaken at the far reaches of the system or points where pressure is known to be poor.Figure6.3:WatermainreplacementschedulingA-replacementlevelcorrespondingtothefundingfunctionDeteriorationNumberB-optimumreplacementlevelofC-replacementlevelcorrespondingtoutility’sfunctionFunctionBreaksRangeofPubiicFundingFunctionA B C_____RangeofUtility’sReplacementFunctionTA-actualreplacementtimeTCTBTAYear:TB-optimumreplacementtimeTc-utilitiescalculatedoptimumreplacementtime218Similarly, the criteria may be to provide fire flow on a peak day or to maintain aminimum average pressure over the entire system, realizing that the distributionalaspects must be recognized so that the average is not made up of extreme highs andlows. In any case, such performance criteria must adequately reflect the true nature ofthe communities level of service expectations.Probably the least predictable and most difficult to systematize are the “externalfactors”, or the system criteria external to the actual physical performance and operationof the pipe. External factors are often one-time events and are characteristicallyunpredictable; they can influence decisions to such an extent that they completelyovershadow any structural or performance concerns and as such they can include:1) increased public concern over pipe materials related to health and water qualityreflected in increased water quality standards,2) increased availability of cheap rehabilitation money,3) increase in standards related to safety (fire flows),4) increased concern regarding liability due to fire and flood damage,5) increased demand above anticipated design capacity due to unforeseen growth,6) increased political involvement in the decision making process,7) increased need to replace based on coordination with other programs (ie. surfaceimprovements, etc.),8) increased legislation or regulation.Due to the unpredictable nature of the such external factors, water utilities tendto base capital plans on the more easily measured structural and performance criteria.External factors are also difficult to quantify and must be expressed in some type ofqualitative index to be included in any analysis. Still, because of the longevity of water219systems, external factors are often the over-riding factors influencing decisions torehabilitate.As mentioned, any one of the three criteria can be the primary influence in theneed to rehabilitate or two or all three may combine to promote the need to rehabilitateas can be seen in Figure 6.4. Policies should be developed which promote thedevelopment of standardized structural and performance criteria, and assist in identifyingtrends related to external factors which could influence rehabilitation decisions. Externalfactors, since they typically emanate outside the normal operational realm of the utility,will best be tracked through better communication between the utility and outsideagencies responsible for planning, funding, public health, and insurance for instance.6.2.2. Condition Assessment and MonitoringAs discussed, as current comprehensive information is sadly lacking, infrastructurerehabilitation policies should first focus on identification of the extent and nature of thecurrent problem prior to generating any large scale mitigative solutions.O’Day (1983) acknowledges that resistance to spending money on information andmanagement systems can be expected, especially when money is scarce, but argues thatsuch systems will allow scarce resources to be stretched even further. In the privatesector, financial consultants, investment brokers, and management professionals arevaluable primarily because they help in deciding where and when to invest limitedresources in order to maximize the desired benefits. The decisions are invariably basedon systems which effectively gather and assimilate the best, rather than the most,information. The value of such information systems can be potentially just as great tolocal government.220Figure 6.4: Criteria involved in a rehabilitation decisionStructuralBreaksperYearPerformanceYearPressureorC FactorMm.ExternalStandardsLevelsTSMax.TpYearMax.TeYear221These information systems should not remain pure data gathering systems whichare designed for the sole purpose of supplying statistics, but should be full fledgedinformation systems which are readily useable and accessible to the decision makingprocesses regarding all facets of system management, including planning, design,construction, maintenance, and rehabilitation.One area which these systems would prove invaluable is in the area of fundingallocation, especially for grant programs aimed at improving water distribution systemcondition. Condition assessments of existing water distribution systems would be a prerequisite for funding and could include a review of the following in any particularcommunity:1) leakage survey,2) identify repairs for deficiencies (not day to day maintenance),3) identify procedures to extend system life (ie. cleaning and lining),4) identify replacement requirements,5) review the current rate structure and development policies for potential futurefinancing options (ie. user charges, development cost charges, etc.)6.2.3. Development of National StandardsMaintenance and condition monitoring criteria must be standardized andimplemented nationally to assure comparable system conditions throughout all utilities.Broad and flexible condition assessment criteria must be developed to allow assessment,application, modification, and implementation among greatly varying communities at thelocal level.As illustrated in Table 6.1, programs must ultimately be carried out at the local222level, but senior levels of government or national groups such as the FCM or theAWWA should coordinate and manage the standard setting and monitoring programover the entire country. National bodies would be instrumental in identifyingcommunities with similar problems and networking such communities to assist each othersharing common experiences.Such information and decision making systems should aim at standardizing systemcondition criteria to allow a more accurate assessment of system needs and to allow easeof gathering and comparison among different utilities. Previous large scale regionalsurveys (FCM 1984, McIntyre and Elstad 1987) have only been able to collect systemcondition information based on local criteria, which can be quite subjective, being basedon the managers perceptions rather than actual predetermined standardized criteria.This adds a great degree of uncertainty to national or regional estimates of cost andcondition as some communities may choose to adopt very high standards and wouldtherefore estimate system condition to be much lower than average and the resourcesrequired much higher.National rehabilitation standards for water distribution systems are currently beinginvestigated by the FCM (Curtis 1991a) and will provide a point of reference which canbe debated publicly and politically, and eventually adopted to help standardize thecriteria for determining system condition. Through the development, debate, andadoption process, the increased profile of the levels of service in Canada may help toeducate both system users and managers on the issues and crystallize their positions ondesired levels of service.More information on breakage rates, leakage rates, and repair costs for thevarious sized communities would be helpful in such analyses as this information is223lacking right now. Long term monitoring of the newer pipe materials now beinginstalled such as polyethylene and p.v.c. would also be beneficial.Technical research into improved construction techniques is also needed,particularly trenchless techniques which concentrate on the small diameter cast iron andductile iron mains which will be the prevalent materials requiring replacement inCanadian town and cities. Application of trenchless technologies to water mains, withfrequent obstacles such as valves and services, will be a major challenge.6.2.4. Varying CommunitiesAs reviewed in earlier chapters, Canadian water distribution systems include avaried mix of material, pipe sizes, age, and physical conditions. The systems aregeographically and physically isolated from one another though each has one majorpurpose, to effectively deliver water of good quality at adequate flows and pressures.While new materials and construction techniques over the years may have reduced theinitial costs of construction and allowed for the extensive development of water systemsin even the smallest of communities, some of the newer materials and techniques haveresulted in reduced pipe life. System deterioration is therefore a small town problemjust as much as it is a big city problem.No one management policy will be entirely appropriate for all communities norall systems. System rehabilitation “needs curves” will vary among utilities and even varywithin a utility among the individual pipes. Figure 6.5 illustrates the difference whichcan exist between the structural needs curves of a small and a large utility. In a typicalsmall Canadian community for example, assuming most the of the water distributionsystem was installed over a ten year period between 1950 and 1960, the environment and224Figure 6.5: Conceptual model of overall system break ratesSmall UtilityTOTAL/iv Id ual Pipe S ectionYearI n stall at ionLarge UtilityBreakRateInstallation Year225construction can be expected to be relatively constant over the town. Assuming therehas been relatively slow growth since installation, the structural deterioration will not bea major problem for the initial 30 to 40 years after installation, but can be expected toincrease relatively rapidly after that given the general hazard function. The smallcommunity will require a relatively large investment in a relatively short period of timeto replace the deteriorated system.In larger communities where the distribution system has been installed in a moreincremental fashion, the problems associated with structural deterioration will be moreconstant over time, making the task of dealing with it somewhat easier since resourceand labor levels can also be maintained at a more constant level. In most largeCanadian cities therefore, information and management systems are well developed,though the techniques used to determine rehabilitation needs still vary widely. In manysmaller Canadian communities, not only have the rehabilitation needs not beeninvestigated in detail, but even the most basic system information, such as pipe sizes orsystem length, are not available or in a useful format to facilitate even the most simpleanalysis.6.2.5. Program Strategies for Various Sizes of CommunitiesThe size of the community will typically dictate the level of sophistication of anyrehabilitation program and the resources available to implement it. In Canada the 31largest communities contain the least amount of pipe per capita, have the lowestdomestic consumption rates, and serve the largest percentage of the population. Effortsto reduce leaks and breaks and subsequently the costs associated with repair and lostwater will be very effective here where systems are concentrated and therefore more226manageable to assess on a systematic basis. Since the largest volumes of water flowhere, it is reasonable to assume the largest volume of lost water can be found. Inaddition, with the large centres growing at increasing rates, work on reducing theproblems within today’s existing distribution systems will reduce future problems whichcould arise as larger populations connect.The 272 medium sized communities in Canada use a volume of water fordomestic use which is almost equal to that of the larger communities, but at a domesticper capita rate which is 38 percent higher than in the larger centres. Undoubtedly, someof this could be accounted for by the relatively large number of traditional single familyresidences which have significant lawn watering requirements. Still, water conservationmeasures in these communities could reduce the consumption rate and the overall waterconsumption significantly. Programs aimed at reducing leakage and breaks would alsobe effective here as total flows are high, though the program would have to be spreadout to nearly ten times the number of communities as the larger centres, therebyincreasing administration costs. Recent growth trends in such communities signify agrowth rate which is stabilizing.In the smallest centres, conservation methods may be an effective means ofreducing the very high per capita domestic consumption rates. Programs aimed atreducing breaks and leaks could be very effective on a per capita basis as thesecommunities contain the largest amount of mains per capita, thus breakage and leakagerates could be high per capita. However, the administrative and training costs associatedwith implementing such a program in nearly 2,600 communities could be high. Inaddition, with staff size very small, there may not be the resources in each community tomanage such a program on a day to day basis. Thus, it may be easier to have small227communities gather the data and monitor the day to day activities, but have each submitreports to a larger agency, perhaps a provincial government department or a regionalplanning board, which would then be responsible for synthesizing longer range plans andanalyses. With growth rates in many of these small communities decreasing, it is likelythe tax base will not be growing substantially. Thus the emphasis in these communitiesshould be on getting the most out of existing systems rather than planning for majorfuture expansions. Land uses should be assessed to assure an optimum number ofpeople are being served by existing system. Large areas of undeveloped lands or vacantlots adjacent to existing pipes should be promoted for development versus areas outsideor on the fringe of existing services. While total water volume savings may not besignificant in small communities which only account for 17 percent of national domesticwater flows, on a per capita basis these communities represent the greatest potential forimprovements. From a piping perspective, the systems in these small communities wereinstalled in the last 30 to 40 years using new materials that have proven to be lessdurable than older pipes. Thus, the problems of system deterioration in thesecommunities can not be discounted as they will not be far behind that of the larger oldersystems of larger centres.6.3. Long Term Fiscal Policy RestructuringWhile elaborate condition assessment programs and pipe deterioration modelsmay easily identify a pipe as physically in need of major renovation or replacement, therehas not yet been a program developed which guarantees that the funds will beappropriated to actually carry out the work. While the physical processes of corrosionand fatigue are complex, the complexity of these physical processes pale by comparison228to the decision making processes which determine whether the will, determination, andmost importantly the financial resources will be available to carry out the needed work.Because water systems are so extensive, it can be expected that the costs of any newprograms will be great, and as such the speed at which decisions will be made to spendmoney will undoubtedly be very slow. Still, in relative terms the cost of solving theproblems are not excessive nor prohibitive. Increasing water rates to levels notuncommon in Europe would greatly assist in solving many of the problems. While suchchange may involve a slow and arduous process of informing consumers of the true costof water service and the need to phase out unnecessary subsidization, the end result willbe a more enlightened consumer and more effective and efficient water systems.At the root of any new national public utility management strategies must bemore rational fiscal policies. Such policies must be far reaching and based on aphilosophy which includes three major ingredients:1) Implementing more cost-effective and value-reflective programs: cost-effectiveinformation management systems which optimize overall maintenance andrehabilitation dollars as well as realistic water rate structures which reflect thetrue value and costs of water systems must be adopted; a recognition of the long-range capital replacement needs should now be an integral part of all localprograms;2) Opening the books: the cost-effective and value-reflective programs must bepreceded and eventually accompanied by a public program aimed at identifyingand advertising the true costs and the true value of water infrastructure. Prioritiesand planning must be set through public processes to gain acceptance;3) Developing of rational long-term fiscal objectives: local utilities must aim toward229increased funding self-sufficiency with senior levels of government providingassistance based only on fulfilment of respective mandates; senior assistanceshould aim at supplementing communities with lesser abilities to pay to meetnational standards and assisting with the implementation of any new, morestringent standards; the focus is to move away from the continued widespreadsubsidization of capital works.6.3.1. Costs of Capital Replacement ProgramsWhen discussing the dollars associated with rehabilitation of water systems, thereare two separate issues which plague Canadians:1) clearing up the current backlog of work;2) implementing continuing long term programs of renovation/replacement.According to the FCM (1984), the cost required to upgrade the existing systems inCanada and eliminate the backlog of work amounted to $76 per capita or about $1.75billion in 1983. Today, accounting only for inflation, this figure would be closer to $96per capita, or $2.2 billion nationally, which over a ten year period as the FCM suggests,amounts to $220 million per year in additional spending. According to the FCM (1984),of the total average annual water distribution system budget of $24.21 per capita in 1983,approximately $9.16 (or 38 percent) is for scheduled repairs and replacement. In 1992dollars, this represents approximately $263 million per year nationally. Combined withthe $220 million estimated by the FCM, spending levels near $483 million each year overthe next ten years would be required just to clear up the problems which existed before1984. This represents an 84 percent increase over current repair and replacementspending levels.230The costs and nature of longer-term distribution system replacement programs canvary widely and would require investigation for the cost versus benefits of each.MacLaren (1983) notes that in Britain the replacement program is based upon replacingdifferent sized pipes at the end of their effective life: 100 years for pipes and servicelines smaller than 300 mm in diameter; 120 years for 300 mm to 600 mm; and 150 yearsfor pipes larger than 600 mm. Based on a similar 100 year replacement program andestimating the average system age to be 40 years, MacLaren (1983) calculates that totalsystem replacement cost of $2,350 per capita for mains and services in Ontario would bespread over 60 years, resulting in a $40 per capita per year program (1983 dollars).However, MacLaren estimates that a combined program of renovation with minorreplacement would be more realistic and only cost in the order of $25 per capita peryear. MacLaren estimates current spending in Ontario to be only 25 percent to 30percent of this $25 level. He goes on to estimate that initial inventory system and surveycosts would amount to an average of about $10 per capita in Ontario with the costs insmaller communities closer to $15 per capita and in larger centres $5 per capita(MacLaren 1983). The costs outlined by MacLaren (1983) are summarized in Table 6.3.Table 6.3: Summary of renovation and replacement costs in Ontario in 1983Option Program Total Per Per Capita Monthly CostPeriod Capita Cost Annual Cost Per Household(1983 dollars) (1983 dollars) (1983 dollars) (1983 dollars)Total system replacement 60 years $ 2,350 $ 40 $ 11.70Renovation with minor replacement 60 years $ 1,500 $ 25 $ 7.30System inventories 5 years $ 10 $ 2 $ 0.60Source: MacLaren 1983231Combining the estimated national pipe quantities and typical sizes in this studywith the unit prices for replacement provided by MacLaren (1983), replacement of the130,300 km of water mains (including valves and fire hydrants) in Canada would amountto approximately $30 billion or about $1,430 per capita in 1983 dollars. The cost toreplace the nearly 7 million services would amount to an additional $20 billion, or $929per capita. Thus the total cost of replacement of the municipal distribution systemwould be approximately $2,355 per capita in 1983 dollars.Based on an ENR 20 city construction cost index of 5,000 in 1992 and 4,000 asused by MacLaren in 1983, construction costs increased approximately 25 percentbetween 1983 and 1992. Thus the $2,355 per capita in 1983 dollars is closer to $2,940 in1992. Based on the national average system age of 32 years as calculated in 1991, thecost per year for a total replacement program for all Canadian distribution systemswould be $43 per capita, which is comparable to MacLaren’s estimate, especiallyconsidering inflation and the relatively young overall age of Canada’s water systemsoutside of Ontario. Based on the same inflationary adjustment, a national combinedrenovation and replacement program would cost $27 per capita in 1992 dollars, orapproximately $620 million per year nationally. The cost to set up the nationwide systemsurveys would amount to approximately $290 million based on a 25 percent increase incosts since 1983 and a per capita cost of $10 in 1983. Again, the costs per capita insmall centres would be more than those in larger centres. Assuming a nation-wideprogram of setting up inventory systems was also set up over the next 5 years at a costper year of $58 million per year, this would amount to an additional $0.73 per month perhousehold water bill. These costs are summarized in Table 6.4.232Table 6.4: Estimated national renovation and replacement costs in CanadaOption Program Total Per Per Capita Monthly PerPeriod Capita Cost Annual Cost Household Cost(1992 dollars) (1992 dollars) (1992 dollars)Total system replacement 68 years $ 2,940 $ 43 $ 12.50Renovation with minor replacement 68 years $ 1,836 $ 27 $ 7.90System inventories 5 years $ 13 $ 3 $ 0.73While the required spending increases are large, there will be significant savingsover time due to reduced breakage rates. Breakage repairs according to the FCM(1984) amount to $3.63 per capita per year for 30.4 breaks per 100 kilometres year.Based on an estimated 113,300 km of pipe in Canada in 1984, this amounts 34,400breaks with a total repair cost of $76 million, or approximately $2,220 per break (1983dollars). Accounting for inflation, the cost per break would be closer to $2,800 in 1992.Assuming the repair and renovation program reduces the average breakage rates downto 10 breaks per 100 kilometres per year as anticipated by MacLaren (1983), a saving of$53 million annually could be realized.On the demand management side, Tate (1990) argues that long term waterdemand reductions can be realized through the installation of meters in the half ofservices not currently metered in Canada. According to Tate, the cost to install metersnationally would be in the order of $700 million based on $200 per meter and 3.5 millionhouseholds.6.3.2. Impact on the Average Canadian Water BillWater distribution systems are somewhat unique in that the two underlyingmanagement goals now being promoted tend to work in opposition to each other. Major233policies aim at both solving the problem of over-consumption and eliminating pricesubsidies through increasing water rates. While the goal of water conservation to reduceconsumption is honourable, it does create a dilemma for utilities whose only source ofrevenues is directly tied to water consumption; the greater the consumption, the greaterthe revenues.In order for a utility to simply maintain revenues, any conservation efforts whichreduce consumption will have to be accompanied by increases in the unit rates chargedfor water. However, to eliminate the current price subsidies and to provide neededcapital for the anticipated replacement and renovation programs, water revenues cannotsimply be maintained, but must increase. Thus, from a consumers perspective, the factthat increased conservation efforts will be met by ever increasing water rates will resultin a perception of being penalized for actually conserving water! It is for this reason thatpublic awareness of the reality of the current situation must be raised.To illustrate the impact on the average Canadian’s domestic water bill of acombined replacement and renovation program as promoted by MacLaren (1983),assume spending levels are raised to $620 million per year as illustrated in Table 6.5from the current $263 million, with the consumer picking up the full tab. In Canada, thetotal annual water pumpage is 12.4 million cubic metres in communities larger than1,000 in population. Assuming the $357 million per year cost increase will be borneequally among all users based on percentage of water used, then the domestic users’share will be $182 million, or 51 percent of the total which represents domestic flows asa percentage of the national total. Assuming the average Canadian household has 3.5persons and uses 360 litres per capita per day, then on a monthly basis, the residencecould be expected to use 37.8 cubic metres of water. Based on an Tate’s (1990) average234cost of $0.47 per cubic metre, the average Canadian monthly domestic water bill iscurrently in the order of $17.75. Should Canada’s 23,000,000 domestic users be requiredto pick up the $182 million rehabilitation tab, the average Canadian would pay $7.91more per year which translates into $27.70 per household annually, or $2.31 more permonth on their water bill. The new monthly water bill would be $20.06, a 13 percentincrease over the current billing.Table 6.5: Cost comparison between variousDescriptionCurrent scheduled repair/replacement budgetSpending level over 10 years to clear up backlogaLong-term combined replacement/renovation programbLong-term replacement of all distribution systemsa - based on FCM (1984)b - as introduced by MacLaren (1983)options for national rehabilitationTotal Annual Cost Annual Cost per Capita$263 million $11$ 483 million $ 21$ 620 million $ 27$ 987 million $ 43However, as Tate (1990) suggests, increased rates, combined with extensivemetering and conservation programs, can be expected to decrease long term waterdemand in the order of 20 percent. For a utility to maintain revenues at the $20.06 permonth level, the reduced consumption would mean the average unit price of water wouldultimately have to be raised to $0.66 per cubic metre, a 41 percent increase over andabove the current unit price of water. As can be seen in Table 2.4 of Chapter 2, thiswould put Canadian water prices about 25 percent above those in the U.S., but still lessthan many European countries.2356.3.3. Funding SourcesWhile the previous sections focused on the users picking up the full cost of anynew programs directly, there is some justification for shared funding with higher levels ofgovernment. Such justification relies heavily on the degree to which any particular levelof government’s constitutional mandate is fulfilled by the funding program. Involvementof higher levels of government should come in two forms: direct funding and newlegislation to support innovative funding mechanisms.Direct funding is the most commonly used approach. In the future, financialassistance should continue to include conditional grant or low interest “infrastructureimprovement” loans which aim at clearing up the worst of the current backlog. Costsharing programs similar to those now being implemented in Ontario and Alberta can bedeveloped nationally based on the rational benefits realized by the investment. Ideally,the goals of such programs should look beyond the short political time horizons andshould provide guidelines for assessing system conditions and a fundamental groundworkfor future capital works.Any large scale funding programs should make funding or loans contingent upon adetailed assessment of the community’s current water distribution system condition. Theassessment would look not only at the physical condition of the water distributionsystems, but the financial outlook based on the municipalities current rate structures andcapital financing plans. Information system improvements such as defining commoncriteria for rehabilitation and replacement as well as common condition assessmentcriteria would allow comparison and overall monitoring of the progress of such aprogram.The repayment of any local “water infrastructure improvement” loans could be236through surcharges added to the water bill, but should be accompanied by an explanationto the consumer of the loans and improvements. Demystifying such loan programs andthe necessity of them would be essential to prevent probable public backlash.Consumers must be made aware that in the long-run they will be paying less for the newcapital programs now versus the higher maintenance, repair, and future capital workswhich would ultimately have to be paid later.Williams (1984) suggests another option would be community contributions torevolving loan funds. Such a fund could be set up which every community in a specifiedgeographic area could pay into, based on a provincial capital recovery surcharge. Whena rehabilitation need arises, a community could draw on the fund for required upgradingmonies. Although such resource pooling may be attractive, the bottom line remains thatcommunities will have increase revenues to make funds available to pool.One other popular source of revenue is the development cost charge leviedagainst developers wishing to tie into existing systems. Although a justifiable source offunding, the charges must remain equitable so that the developers pay their true share ofthe actual rehabilitation, especially considering that these charges are ultimately chargedback to the new residents in the form of higher housing prices. Any over-charging forthe simple purpose of convenient revenue generation would result in new residentspaying a disproportionate share of other’s past problems.6.3.4. Economic Benefits of Water Distribution System InvestmentWhile the costs may seem high, reinvestment in our water infrastructure can belooked on in a positive light as the natural continuation of a process that was startedafter World War II. From a national perspective, development improved the standard of237living in North America and provided both the infrastructure backbone required fordevelopment of industry as well as much needed employment in public works projects.While this course of action has largely been abandoned as evident in the shifting offunds out of federal infrastructure programs, recent political changes in both the UnitedStates and Canada signal hope that the situation may be changing. President Clinton hasannounced publicly on numerous occasions that he plans on introducing a $20 billionprogram aimed at improving America’s roads, bridges, and lifeline systems, andestimated to create approximately 25,000 jobs per billion spent. While this will not gofar considering the American’s annual funding shortfall of $9 - $11 billion estimated byDunlop (1983) it is a good start. In Canada, the new Liberal govermnent under JeanChretien has announced a similar $2 billion program aimed at all physical infrastructuresystems. Federal funds would be matched by equal provincial and local contributionsresulting in a total program approaching $6 billion over the next two to three years.Again, while a promising start, the program will only begin to solve the current $30billion backlog of public works and the contributions of the debt-burdened provincial andlocal governments are by no means assured.At the heart of the new federal program is job creation. A 1983 study shows thatinvestment in public works results in 35 to 40 person years of employment and $250,000in expanded local business for every million dollars invested (Environment Canada1983). After accounting for the fact that a million dollars in 1983 is equivalent to $1.25million in 1992 dollars, an annual spending increase of $357 million per year toimplement a national combined replacement and renovation program for waterdistribution systems would still create approximately 10,000 jobs and generate $71 millionworth of local business. In addition, a savings in unemployment insurance benefits paid238out for the 10,000 jobs in the order of $105 million could be realized annually, based onan average of $202.75 per capita per week paid out in unemployment benefits (U.I.C.) in1988 (Canada World Almanac 1990).As can be seen in Table 6.6, the $357 million increase in expenditures could resultin long term benefits totalling $229 million from reduced breakage rates, annual U.I.C.savings, and local economic development. Of course the benefits from reduced breakagerates would not be realized until later years when the overall system characteristics areimproved. However, economic impacts from job creation would be realized early.It is also from this job creation aspect that local and provincial bodies shouldcontinue to lobby the federal government since employment and regional developmentare both within the federal jurisdiction. From this perspective cost sharing in the orderof 25 to 35 percent can be justified.Table 6.6: Costs and benefits of a combined replacement/renovation programDescription Total Annual CostCombined National Renovation / Replacement Program $ 620 millionCurrent National Scheduled Repair / Replacement Spending $ 263 millionNet increase in spending required $ 357 millionAnnual savings in water main breaks ( $ 53 million)Annual savings in U.I.C. benefits paid out ( $ 105 million)Local economic development ( $ 71 million)ACTUAL NET ANNUAL COST OF PROGRAM $ 128 million2396.4. SummaryThis thesis has focused on improving the information on existing systems throughthe use of historical data which may have been overlooked and to outline the mainproblems and available solutions, eventually synthesizing them into a framework whichcan be used in Canada. There is nothing new in the information, and as Grigg (1988)points out the current problems do not necessarily require the generation of new ideasbut rather the effective application of already developed ideas. There is nothing magicalabout the solutions; the real trick will be to get these things up and running by matchingthe appropriate technologies to the appropriate situations. Levels of sophisticationshould be considered in the management of systems: the small village of 500 will notneed nor be able to operate the sophisticated models of the large cities of 500,000, yetthe small villages of 500 should still be included in a larger global model which may beoperated on a regional or provincial basis. The current information age brought aboutby the new “personal computer generation” and improving education levels of operatorsmeans that both large and small communities can expect higher levels of managementsophistication than what exists today. The tasks are not impossible, but the political willmust be there to implement them.240CHAPTER 7: CASE STUDY7.0. BackgroundThe Corporation of the District of Pitt Meadows is situated on the confluence ofthe Fraser River and the Pitt River on the eastern edges of the Vancouver CMA. TheDistrict was incorporated on April 1, 1914 and encompasses and area of 5004.5 Hectares,with a 1989 population estimated at 9,546 (GVRD 1990). Compared to other smallcommunities in Canada, the District is growing extremely fast (see Table 7.1), as is muchof the Lower Mainland, and can expect between 200 and 400 new residents per year.Table 7.1: Pitt Meadows recent population growthPERIOD POPULATION GROWTH IN PERCENTPitt Meadows National Average (Small Centres: pop. < 10,000)1971 - 1976 69 111976-1981 32 81981 - 1986 29 2Source: Adapted from population data; B.C. Municipal Affairs 1951 - 1988Although a substantial percentage of the current distribution system was in placein the 1950s and 1960s, rapid growth did not occur in the District until the 1970s as seenin Figure 7.1. In 1971, while it had less than 30 percent of today’s population, theDistrict had over 65 percent of the current pipe in the ground. Thus, since the 1970s thewater distribution system has been undergoing infilling and relatively slow expansion.Compared to other municipalities in the Greater Vancouver area, property valuesin Pitt Meadows remain relatively low, reflecting a large rural influence. The 1989 taxroles included 3,808 assessed properties with an average value of approximately $87,000.z 0 F -J 0 0 0Figure7.1:PittMeadowspopulationgrowth1500012000900060003000 019211931194119511961197119811991200120112021YEARSource:BCMunicipalStatistics(vary);CVRD1990File:PITTPOPCAL242This compares to the total of 455,286 properties in the Vancouver CMA with an averagevalue of $174,178, and 116,677 properties with an average value of $232,365 in the Cityof Vancouver itself (GVRD 1990).7.1. Water ConsumptionThe water system is somewhat unique in that it serves an urban area representingabout 76 percent of the population and a rural area with 24 percent. The connectionrate in Pitt Meadows is 100 percent, meaning the entire population is served from theDistrict’s distribution system.Table 7.2: Pitt Meadows water consumption in 1990.Average Daily Flows:WATER 7336 m3/d - Domestic 5135 m3/d- Commercial/Institutional 734 m3/d- Industrial 1100 m3/d- Other 367 m3/dSEWER 3260 m3/dDIFFERENCE 4076 m3/d (UNACCOUNTED FLOWS)Unaccounted Flows: (as Dercentage of water flows)Pitt Meadows 56 %National Avg. 26 %Source: Data adapted from IWD (1990)The average daily flow for the District is 7,336 m3/d with only about 44 percent of243this returned to the sewer system, leaving 56 percent unaccounted for (see Table 7.2).Some of the unaccounted water can be attributed to high consumptive uses such asgreenhouses and agricultural type uses. However, the most obvious reason is the lack ofsanitary sewers in areas which are served by the water distribution system. In PittMeadows in 1988, there were 82 kilometres of water mains, yet only 26 kilometres ofsewer mains. This is a reflection of the fact that the water system supplies a largenumber of rural users with individual sewage systems and highly consumptive agriculturaluses. Thus a significant flow is not returned to the District’s sewer system.Accordingly, per capita water consumption in the District is quite high as can beseen in Table 7.3. Still, when factoring out the high commercial and industrialconsumption, domestic consumption in the District is about 36 percent above both theprovincial average and the national average for small communities, and 68 percent abovethe overall national average.Table 7.3: Comparison of water consumption with national and regional dataCOMMUNITY CONSUMPTION (l/d/capita)Total DomesticPitt Meadows 843 590B.C. Average 722 429Canadian Average (all communities) 688 351Canadian Average (< 10,000 popn.) 668 434Source: Data adapted from IWD (1990)7.2. Water RatesWithin the urban area, the majority of consumers are residential or commercial244users and are governed by a very low flat rate for water. Based on 1991 rates, singleoccupant residents are charged $27 per year and multi-occupant residences $36.Commercial users rates vary among type of use, ranging from $12 per year for a church,to $90 per year for a slaughterhouse, with the average rate being about $50 annually.The rural area is metered and the rate structure includes a minimum charge fordomestic use with a constant unit rate above a set volume, and a declining block forcommercial and industrial users. Domestic users pay a minimum of $18 for every sixmonth period for 53,000 Igal (240 m3 or 40 m3 per month) from January to June or60,000 Igal (272 m3 or 45 m3 per month) from July to December. Consumption abovethe set volume is charged at a constant rate of $0.24 per 1,000 Igal ($0.053 per m3).Except for golf courses which pay a minimum 6 month charge of $240, commercial andindustrial users pay a minimum of $18 per 6 month period. Consumption chargesdecrease with increasing blocks, ranging from $0.40 per 1,000 Igal for the first blockbetween 0 and 20,000 per 6 month period, to the highest block where the charge is $0.24per 1,000 Igal for any volume used over 175,000 Igal (see Table 7.4).Table 7.4: Pitt Meadows commercial water ratesCONSUMPTION BLOCK WATER RATEPer 1,000 Igal Per rn30 — 20,000 Igal $0.40 $0.0920,001 — 50,000 Igal $0.35 $0.0850,001 — 100,000 Igal $0.30 $0.07100,000 — 175,000 Igal $0.26 $0.06over 175,000 Igal $0.24 $0.05Source: District of Pitt Meadows 1991 Water Rates2457.3. The Water SystemThe District receives its water from the Greater Vancouver Water District, aregional supply and transmission system that supplies water to approximately 1.5 millionpeople in 20 municipalities and 3 electoral areas. The GVWD, by virtue of a number ofhigh elevation sources in protected water sheds, distributes water requiring onlychlorination and little or no pumping at wholesale rates which are among the cheapest inCanada. Pitt Meadows receives its water from three connections to a 750 mm (30”)GVWD main along the Lougheed Highway which also serves Maple Ridge to the east.Delivery pressures during peak periods at the three locations vary from 76 to 108 psi andare expected to decrease in the future with increasing demands from Maple Ridge (CBA1981). Because the District is supplied from a pressurized and treated supply, there areno pumps, treatment works nor reservoirs in the District.The urban area is served by a system which provides flows and hydrants for fireprotection (Figure 7.2). The rural system consists of many small diameter pipes,including 50 mm and 100 mm diameter, and as such provides only domestic andindustrial water service with no fire protection. The system age is typical for acommunity of this size.Most pipe was installed after World War II and based on the Municipal Utilitiessurveys of 1951 and 1961 and provincial municipal statistics, the data in Table 7.5 hasbeen accumulated. Based on this information, the average age of the overall system is31 years. Of the 82 kilometres of pipe in the ground in 1988, over 50 kilometres is oldercast iron or small diameter galvanized iron. The majority of the remainder is ductileiron with a small fraction of asbestos cement.Figure7.2:PittMeadowsexistingpipelinenetworkIIUNICW*LIOUNDARYWILOLIFIREICRORROUNDARYPAVEDROADUNPAVEDROADPIPR(ZELrt’JPitt Meadows water systemPIPE DENSITY ---- SYSTEM BREAKDOWN(persons/km) Length Material Sizes27.2 40.4 Cast Iron 1 - 6”33.9 40.4 Cast Iron 1 - 611.2 Galvanized 1”1971 2771 - 58.2 47.6 - - -1981 6209 - 72.0 86.2 - - -1988 8700 843 82.0 106.1 - - -Source: MU 1951; MU 1961; B.C. Municipal Statistics 1971, 1981, 1988As can be seen in Table 7.6, the pipe sizes vary widely from the urban to the ruralareas.Table 7.6: Breakdown of Pitt Meadows water system by pipe size in 1981PIPE SIZE PERCENTAGE OF LENGTH IN RESPECTIVE AREA CANADIANURBAN RURAL TOTAL SYSTEM AVG. IN 1961<100mm 0 43 37 2100mm 0 22 19 12150 mm 16 20 20 55200mm 36 5 10 16250mm 6 4 1 4300mm 0 1 1 8>300mm 42 8 13 3TOTAL 100% 100% 100% 100%Table 7.5: Historical information on theYEAR POPULATION PUMPAGE LENGTHSERVED (lid/capita) (km)1100 40.41750 51.619511961247822Source: Pitt Meadows data extracted from CBA (1981); Canadian estimate from data base MU (1961).248The District maintains large mains in the urban area to provide fire flows, while in therural area nearly 70 percent of the pipes are smaller than 100 mm, with nearly 17kilometres being 25 mm and 50 mm.At the present time, the District does not have long range plans for the eventualreplacement of the existing water system and as of 1991 was putting together its first ever5 Year Capital Improvements Plan. The most current assessment of the existingdistribution system was a study done by a private consultant in 1981 (CBA 1981). Thestudy was undertaken for the B.C. Ministry of Environment, Inventory and EngineeringBranch pursuant to the Canada - British Columbia Agricultural and Rural DevelopmentSubsidiary Agreement from July, 1980 to January, 1981.The main focus was not on a system wide assessment of condition, but ratherconcentrated on the system improvements which would be required to removeconstraints on future rural agricultural and commercial developments and the benefitswhich would be associated with such improvements. In fact, the Highland Area which isthe District’s main urban area and town centre, was excluded from the study area as itwas outside the Agricultural Land Reserve, and was assumed to have an adequate waterdistribution system.The existing rural economic base relies heavily on water and includes dairying,beef feeding, hog production, nurseries, specialty crop production, and a large number of5 and 10 acre hobby farms. Future increases in hog production, nurseries andgreenhouses, and rural residences are anticipated. The 1981 study includes:1) a projection of rural development over a 5 year period;2) an evaluation of future water demand;3) preliminary design and costing of the water supply system;2494) benefit: cost analysis of project alternatives.Future demand estimates were extrapolated from the existing consumption records andapplied to the future development scenarios. A computer analysis of the pipe networkwas used to determine sizes for water mains requiring upgrading or extending and costestimates were determined.The projection of future rural development and the benefit : cost analysis wereboth based on a questionnaire circulated in the District as well as census and economicinformation gathered from various government bodies. From the questionnaire, benefits,on-farm costs, and probable future development for each respondent could bedetermined. Direct benefits such as those resulting in increased commercial output orresidential development were identified, as well as secondary benefits associated withimproved fire protection and potential industrial development.Much of the system in the study area is undersized with 1” to 4” (25 mm to 100mm) pipes and the District has had to refuse connections due to lack of peak periodpressure and capacity. In addition, one part of the system is isolated from the rest of thesystem fed from the GVWD: the source being a lake susceptible to large fluctuations inwater quality, the water being untreated and unmetered, and the distribution systemconsisting of some unburied PVC pipe and many 1” to 3” (25 mm to 75 mm) mains. TheDistrict has proposed connecting any isolated sections of the system to the main systemand upgrading any 4” (100 mm) trunk mains with 8” or 10” (200 mm to 250 mm) pipes.The study recommends three pipe upgrading scenarios based on the hydraulic andeconomic analysis done. The estimated costs of the three scenarios proposed rangedfrom $2.1 to $5.1 million (1981 dollars) and result in benefit:cost ratios ranging from1.3:1 to 2.6:1.250According to the District Engineer (Lowry 1991), the District intends to direct itsefforts toward the continued upgrading of capacity as recommended by the study.Funding is through the Canada - B.C. Agricultural and Rural Development SubsidiaryAgreement with the District paying 25 percent and the senior levels of governmentpaying an equal share of the remaining 75 percent. Negotiations are continuing tosecure future funding under the program as the District is highly dependent on theprogram to complete the rural works.7.4. Record KeepingRecord keeping of breaks and system problems in the District is not systematic,with most of the records in the hands, heads, and daily journal of the public workssuperintendent. The District practises a loose “rule of thumb” method of pipereplacement similar to many smaller communities. In general, when a section of pipe isdeemed by the superintendent, based on his discretion and experience, to have too manybreaks, the section is put on the list for replacement. While leak and breaks arerecorded in daily journals and repair reports, the information is not systematicallytracked (Cross 1991).The district has recently purchased a p.c. based computer G.I.S./C.A.D. system forapproximately $20,000 but, due to the enormous time demands placed on the staff due torapid growth of the District, formal plans regarding what to do with the system and theinformation input are only slowly being formulated. In addition, the enormous staff andman-hour commitment required to implement and get data into the system were underestimated at the time of purchase, so the system will probably not be fully operationalfor a few years (Lowry 1991). As of late 1991, about 60 percent of the system was251recorded on hard-copy as-built drawings, with about 40 percent of these now on theC.A.D. system.The District’s record keeping is for now limited to written reports and journals.Reports of leaks and breaks and their subsequent repairs are recorded, but are not easilyaccessible for planning or system monitoring purposes. Overall system maps have beendeveloped as well as system future plans for looping and upgrading. No overall systemassessment for breaks, nor any system wide leak survey has ever been carried out nor isanticipated within the near future. As can be expected, no replacement strategy models(descriptive, predictive, nor physical) have been developed nor utilized. For now, themain indicators of system deficiencies are inadequate system flows and/or pressures.7.5. Staffing LevelsFrom a staffing perspective, the District employs 13 people in its Public WorksDepartment and 3 in its Engineering Department. The Public Works Departmentincludes one professional/technician in a supervisory role and 12 field personnel. TheEngineering Department employs 3 professional/technicians and no field personnel. Thestaff works to operate and maintain a variety of systems, with none dedicated to anyparticular system. In general, work on the water distribution system accounts for about20 percent to 25 percent of the staff time (see Table 7.7).7.6. FundingFunding for the water system operations and day-to-day maintenance comes largelyfrom user charges with any surplus going into the District’s reserve fund. The reservefund is a sizeable general purpose fund used by the District, not necessarily for utility252improvements, with the surplus contributed by the utility varying from year to year(Lowry 1991).Table 7.7: Estimated staff work load split among systems in Pitt MeadowsSYSTEM PUBLIC WORKS DEPT. ENGINEERING DEPT.Roads and Storm 50 % 50 %Water Distribution 20 % 25 %Water Treatment 1 % 1 %Sewer Collection 20 % 20 %Garbage Collection 0 % 1 %Miscellaneous 4 % 2 %TOTAL 100 % 100 %SOURCE: Lowry 1991 (estimate)Fiscally, the District is in relatively good shape. Debt financing is not frequentlyutilized to finance system improvements and at present the only outstanding debenturesare on some relatively minor sewer system improvements. Provincial subsidies in theurban area of the District account for approximately 25 percent of system revenuesthrough revenue sharing water/sewer capital grants. General revenues through taxesaccount for an additional 30 to 35 percent, while development cost charges on newsubdivisions can pay up to 30 percent of upgrading costs. User charges account forapproximately 20 to 25 percent of total system revenues, with most directed towardoperations rather than capital improvements. In rural areas, the ARDSA grants coverup to 75 percent of system improvement costs, development cost charges can account for253up to 15 percent, and the remainder typically comes from general revenues. Assumingthe needed improvements have been sufficiently identified and funding sources do notsignificantly change, revenues are expected to meet the desired expenditures (Lowry1991).The District does have a program in place for the systematic cleaning and lining ofproblem mains, but to date the program has never been implemented (Cross 1991). TheDistrict does however carry out annual flushing operations.7.7. Rehabilitation and Replacement ProgramsApart from the need to replace pipes due to limited capacity, the District has notidentified a need to replace any significant lengths of pipe due to deterioration and/orhigh breakage or leakage rates. Currently, approximately 13 kilometres of pipe need tobe replaced due to capacity concerns, while only 1.3 kilometres need replacing due todeterioration, with approximately half being asbestos-cement pipe. As in other parts ofthe Greater Vancouver area, the distribution water is aggressive toward A.C. pipe,tending to soften it to a point of failure. Most of the A.C. pipe in the District hasalready been replaced, and all new pipe installations are ductile iron.From a health perspective, the major concern has to do with excessive coliformbacteria counts in the water. Probably much of this is due to a loss of chlorine residualbetween the source at the Coquitlam Reservoir and the points of distribution, ratherthan with the distribution system itself. Concerns regarding high metal concentrations(ie. iron or lead) or asbestos cement are not pronounced at this point in time.2547.8. DiscussionsAs the information gathered on the District’s system is quite general, a number ofimportant observations can be made. While a much more exhaustive investigation wouldbe required to assess the problems in detail, it is possible to highlight the potential forgeneral improvements.Information on the water distribution system is limited, though the District hastaken definite steps to remedy this through the recent hiring of an engineeringtechnologist, as recommended in a report by a management consultant to help alleviatethe heavy workload put on the development engineer. The technologist is responsiblefor drafting, records, design, dealing with the works yard, and assistance on capital works.It has been noted that when people come in to inquire about the services to aparticular parcel, on occasion there has been no records and someone must physically goout in the field and inspect (Cross 1991). The recent acquisition of the CAD and GISequipment is promising as the District eventually plans to interconnect the legals,infrastructure, and tax information. To date information transfer is going well, but itshould be noted that approximately 10 percent of the system information is not on paperbut is carried by the senior public works superintendent, who is soon to retire. Acomputerized maintenance management system in the public works yard is being set upto track timesheets, unit costs, repair costs, etc. and eventually may include watermainrepairs. However, the community is basically on its own when it comes to inputting andorganizing the information. This is one area where especially the smallest communitiescould use assistance and guidance in wading through the new equipment and programsavailable and in setting up a system which is useful from a operations and managementstandpoint.255The higher than average population growth rate in the District means that moreand more people will be connecting to the existing system, increasing the need for areliable system capable of providing both adequate fire and consumptive flows. The1981 report on system expansion used assumed “C” values for pipe roughness and as suchmay not have reflected accurately areas with reduced capacities due to deterioratinghydraulic characteristics within the pipes. Pipe flow tests utilizing the existing hydrantswould be useful.Due to the relatively low density of most of the District and the low propertyvalues, liability due to flooding caused by main breaks is limited. However, liabilityconcerns regarding fire protection do exist and caution should be exercised respectingthe proposed extension of fire protection out to the rural area, which may be prone toinadequate flows during peak irrigation periods.The high unaccounted water volume is a concern, though it can be explained byhigh consumptive uses and the fact not all consumers are served by sewer systems whichmeasure the return flows. A water audit to confirm leakage and account for alldistributed water would be invaluable in assuring that the system is sound.Unfortunately, unless the District plans on metering all users of the system, long termday to day monitoring will be difficult.Based on pipe sizes, the problems of limited capacity investigated in the 1981report (CBA 1981) are quite easily seen. A number of lines are undersized based oncurrent municipal standards for fire flow. Of course, the rural component is aimed atproviding flows to meet the domestic/commercial/agricultural needs rather than fireflows. The municipality is proceeding to replace sections of the system restricted bysmall diameter pipe.256While health concerns regarding the distribution system per se have been relativelynon-existent, the conditions which brought on elevated levels of coliform bacteria canlead to increased bacteriological growth within the distribution system. While theproblem will likely be solved by increasing the residual chlorine in the system, regularprograms of cleaning and lining mains can eliminate the tubercles where growth mayoccur while at the same time increasing the hydraulic capacity of the mains. Thecleaning and lining program which was set up should be carried through.The water prices charged to the residents of Pitt Meadows are among the lowest inthe country. Based on the rate structures and consumption rates for the District, thetypical residential and commercial users are paying between $0.055 and $0.09 per m3.Compared to the national average of about 0.47 per m3, water in Pitt Meadows is verycheap. Undoubtedly, this low price contributes significantly to the high consumptionrates observed in the community.Considering the GVRD wholesale water rate is around $0.05 per m3, it is easy tosee that there is little built into the District’s water rates to provide for the eventualreplacement of the system. This is typical of most communities in Canada. Based onunit prices provided by MacLaren (1983) and CBA (1981) and adjusting for inflationusing the ENR index, the replacement costs for the Pitt Meadows distribution systemwould be in the order of $25 million, and the services in the order of $16 million. UsingMacLaren’s methodology, with the average age of the system 31 years and assuming a100 year replacement cycle, the replacement would be completed in 69 years. Based onthe domestic users paying their share of this, the average annual water bill would have toincrease by about $143 from the current level of $33. However, MacLaren (1983) notesa combined repair/renovation program could be carried out at about 60 percent of the257replacement program. In Pitt Meadows, this would cost $90 per household per year.This is very close to MacLaren’s figure of $25 per capita per year in 1981 after adjustingfor an average household size of 2.82 in Pitt Meadows and a 25 percent increase in costsdue to inflation based on the ENR index. However, in the case of Pitt Meadows thistranslates into more than a tripling of the average water bill, from the current $3 permonth to $10. Such an increase would increase the current rate per m3 from about $0.07to $0.22, which is still less than half of the national average.In any case, such increases would be very difficult for the public to swallow,especially after years of extremely cheap water. It would take a lot of education and areal political will to carry out such measures. In addition, to properly carry out anyrevamping of the rate schedule, it would require metering of the other half of theDistrict; at $250 per meter, new meters for each households in the community would costin the order of $850,000. The Provincial government could play a key role in financingsome of the short term requirements based on its current cost sharing formulas.However, the trend of senior government seems to be, and perhaps well should be, tomove out of local water system funding for system replacement. Pitt Meadows shouldconsider a plan for eventual self-sufficiency in water system management and also moveaway from supplementing general revenues with revenues from the water system. Ofcourse, it will be difficult for a small district to do on its own. Such measures should bepromoted by the Province and the GVWD with input and cooperation from the othermember communities in the GVWD.2587.9. Summary RecommendationsIn summary, more information is required to allow a conclusive assessment of thecondition of the Pitt Meadows water distribution system. Information is being gathered,but needs to be amalgamated from reports to be useful. Although the community ismaking great strides to improve its management systems, guidance on the information tobe input and standards of assessment are required. In the community, averageconsumption rates are very high and the water rates are both very low and structured insuch a way as they do not promote conservation. Both water metering and inclusion of acharge for capital replacement in the water rates should be considered. Of course anysuch action will require substantial public education and input. A water audit should becarried out by the District to reconcile the high unaccounted for water. The rapidgrowth of the community means the existing system will be called on to serve more andmore people, therefor measures should be taken to assure it is reliable and will serve thecommunity well in years to come. Finally, there should be more information sharing andthe promotion of ideas among the communities both in the Greater Vancouver area, aswell as in the Province of B.C. and the nation as a whole.A number of recommendations may be made to assist the municipality inimproving both the efficiency and effectiveness of the system:1) add meters to the system and increase the water rates to both reduce consumptionand increase revenues for eventual capital replacement; such a program should becarried out in a rational and phased manner with input from the public;2) communicate with other communities of similar size and/or growth characteristics aswell as larger centres (ie. Vancouver) to compare consumption, water rates, andrehabilitation programs;2593) make upgrading of the existing system a priority to accommodate future growth;4) implement a replacement and rehabilitation strategy; this will be based on carryingthrough the program already in place as well as continuing with therecommendations on the distribution upgrading report of 1981; less emphasis will beon property loss due to flooding as development densities are relatively low;5) consideration should be given to the additional liability concerns and potential costsshould fire flow and fire fighting service be extended to the rural area;6) concentrating on the growth of the system and enhancing its capacity should remaina priority, with some consideration given to the reliability of the system in the urbanHighlands area;7) the District should consider a number of options to reduce its high unaccounted forwater:a) water audits to assist in locating leakage and lost water;b) metering to both help reduce consumption and to assist in the auditcalculation;8) consider increasing water rates to include eventual capital replacement;9) upgrade small diameter iron and galvanized pipes as recommended in the 1981report to help enhance economic growth potential;10) consider the need for eventually replacing the high percentage of pipes in the systemwhich may be over 40 years old; collect installation date and repair/replacementdata on the system if available;11) consider in the urban area increasing land use densities, filling vacant lots, and infillif feasible as the 106 persons per kilometer of pipe served is quite low, although it isreasonably good considering the large percentage of rural users;26012) continue with the development of the p.c. database and management systems andattempt to get all records on the system from the hardcopy paper records;13) introduce preliminary descriptive analyses to help pinpoint system breakage orleakage problem areas;14) carry on the planned cleaning and lining program to help reduce potential healthand flow problems;15) maintain current staffing levels, but in the short-term hire on part-time help to assistin the data entry, should budgets allow;16) revamp the funding and financing mechanism to separate the water system fundingfrom general revenues and prepare for the real possibility of reduced funding levelsfrom senior levels of government.261CHAPTER 8: CONCLUSIONS AND RECOMMENDATIONS8.0. OverviewApplied to the Canadian context, the developments in pipe technology over thepast century will be major determinants in the deterioration of piped systems over thenext 25 to 50 years. Larger, older cities with significant lengths of old mains, may notnecessarily be in a significantly worse state than their newer counterparts. Small townsand cities which have developed systems since the 1950s using largely small diameterpipes and new, less durable materials, may develop comparable problems over relativelyshort periods of time. Thus, while the reduction in material and installation costsassociated with modern materials and designs may have significantly accelerated theexpansion of water systems over the past 30 years, it may also have accelerated the needfor rehabilitation and replacement.8.1. ApplicationThe basis of this thesis has been to exhibit through the study of water distributionsystems the basis upon which a rational national policy on infrastructure rehabilitationand restructuring can be developed. Many of the principles outlined can be applied toroads, sewers, water treatment, sewage treatment, and a host of other systems whichmake up the physical fabric of our nation. Water distribution systems were specificallyinvestigated to illustrate the shortcomings of current infrastructure management andfinancing which is evident to some degree in all the physical systems in Canada.While they are in decline, our water distribution systems, and indeed most of ourother large scale infrastructure systems, are by no means on the verge of imminentcatastrophe. However, the small leaks and cracks in the systems are slowly showing and262should they remain unchecked, there will be significant health, economic, and fiscalconsequences down the road. It is prudent to begin planning for the impending fate ofour infrastructure systems now, and a logical place to begin is with organizing theinformation required to make useful decisions and with improving both the technical andfinancial systems which will be required to carry out the rehabilitation.As with water distribution systems, all infrastructure management informationmust be detailed enough and in a readily usable form to make useful decisions. Policiesshould not only aim at rationally replacing, rehabilitating, or renovating systemcomponents which are wearing out, but should recognize and improve methods oftracking system condition, improving performance, minimizing risk, collecting revenues,maintaining equity, and ultimately serving the end user better. Such policies will reachacross a variety of administrative and legislative levels, and as such should be developedon a national level while remaining flexible enough to be adopted and administered atthe provincial and local level.8.2. Recommendations and SolutionsThe historical development of Canadian infrastructure systems is not insignificantwhen focusing on today’s solutions. Economic and health concerns were at the heart ofthe development of these early systems and advances in system development wasparalleled by unprecedented increases in the standard of living in our nation. Thehistorical development of piped water distribution systems is also significant for study inthat much of the materials originally installed 75 to 100 years ago are still in service andwill soon require replacement.While most systems have developed under public control and management, the263involvement of the private sector in future systems management cannot be overlooked,especially with the ability of the private sector to raise capital for rehabilitationprograms. Strong legislation and regulation with respect to servicing and water qualitycan assure standards are maintained.There is a definite need to consider improvements to fire flow considerations inwater systems, especially in small communities where many do not meet the currentinsurance industry standards and liability concerns are growing. The courts in Canadaare placing even more responsibility and liability for the short-comings of existingsystems directly on the shoulders of the public utilities. While legislation may beintroduced to reduce the communities exposure to such liability, the fact of the matterremains that inadequate systems can prove to be very costly in terms of damages, lostindustry, and waning public support.The solutions will involve specific strategies aimed at specific problems. A jointtask force involving the business, design, and construction communities as well as utilitymanagers, universities, and all levels of government should be set up to effect the properstrategies, better involve the public, and continue the research first undertaken by theFCM. Huge sums of money distributed in a broad fashion is neither appropriate norrealistic considering today’s fiscal atmosphere. The biggest problems are oftenconcentrated in a relatively small number of communities. In addition, the solutionswillvary widely based on community size, system age, regional considerations, and abilitiestopay. Large cities such as Toronto, Montreal, and Winnipeg have the resources toeffectively tackle and identify their problems whereas small communities such as PittMeadows must consider means of not only identifying solutions, but tracing theproblems. Non-structural solutions such as water rate restructuring, demand264management, water metering, and infilling can be especially effective in the smallcommunities by must be introduced to the consumers in a rational and straight forwardmanner.Current research by government, universities, and the private sector shouldconcentrate on developing simple, effective, and economical infrastructure managementsystems for personal computers. Guidance manuals based on the Canadian context mustbe developed to assist local operators and governments in planning for the secure futureof these lifeline systems. University programs in engineering and planning must begin torecognize the importance of maintaining the billions of dollars already invested in theground. Research should also aim at improving pipe monitoring and replacementtechniques for relatively small diameter 6” (150 mm) and 8” (200 mm) sized cast andductile iron pipes which make up the majority of the pipes soon to require replacement.Research should also consider solving the predominant problems faced by Canadiansystems: failures due to frost and temperature effects, poor quality construction andmaterial failures due to corrosion and age. Reduced strength in new plastic pipes andthe reduced corrosion resistance of thinner walled ductile iron pipes will dictate moreintensive long-term monitoring of the performance of these newer pipe materials.Continued investigation into the mechanisms of structural failure will also provevaluable. Benefits can come from reducing the high breakage rates of small diameter150 mm to 200 mm mains and realizing that the nature of breaks is spatial in nature,with most breaks recurring in areas of previous break history. Such developments canhelp to channel replacement resources to the appropriate problem areas. Researchstrategies which also include more global factors, such as liability issues, can alsoimprove overall management of systems. As O’Day (1984) suggests, areas of high risk of265leaks but low risk of damages can best be served by leakage control while areas of highrisk of leaks and high probability of damages should be considered for outrightreplacement. These examples of concentrating on the specifics of the problems andengaging the appropriate technologies to the appropriate problems will only help toeffectively solve the current crisis.Long term structural and non-structural improvements must be made andimplemented with the understanding from the public bodies and the consumers served.Trenchless technologies will help to reduce the conflicts with public transportationsystems, but it must be realized such technologies also reduce the exposure of the publicto the day to day needs and operations of the water distribution sector. Education andadvertising aimed at the public must be encouraged, especially in overcoming the“conservation conundrum” where reduced demand through conservation will inevitablymean having to increase water rates to maintain revenues. Without continued fundingfrom senior levels of government, increases in water rates over the long term couldaverage as much as 41 percent over current levels.Infrastructure funding as outlined by the Federal government’s new $2 billionprogram is long overdue and very much required. But as evident in the public scepticismexhibited during the program’s promotion throughout the election, there needs to bemuch more interaction and improved communication between public utilities and thepeople they serve. Ultimately, large funding programs, whether they come from federallyor provincially collected taxes or locally collected water rates, must be administered toeffectively solve specific problems rather than unconditionally distributed to where theycould be inappropriately used to prop up general revenue funds or other discretionaryuses at the hands of local officials. The recent infrastructure program announced by the266federal government should base funding upon sound plans for the future by the localmunicipalities. Infrastructure assessments which rationally assess the system deficienciesand aim at effectively rectifying the problems must be promoted and could even befunded under the new program. These systems must be based on generally accepted andadopted standards of system condition and service. Investigations similar to the oneoutlined within Pitt Meadows must be carried out in even more detail in all communitiesacross Canada. Communities which have been maintaining systems properly over theyears should also be permitted to participate and receive an equitable share of thefunding, to better develop their management techniques and technologies, or to assist insharing their information and techniques with other communities.Such information exchanges must be promoted. Federal and provincial agenciescould assist in the continued collection and analysis of condition assessments through thedevelopment of standardized, user-friendly condition assessment programs, breakagereports, and decision costing programs. Guidelines for effective combinationreplacement/renovation programs should be developed and implemented over the nexttwo to three years, perhaps in coordination with the new infrastructure program.As a nation, we must recognize that the problems and solutions will vary acrossthe country. The smaller, older communities of Atlantic Canada are in dire need ofimmediate upgrading of their deteriorating water systems. In Western Canada, wherefunds have typically been more available, especially in B.C. and Alberta, systems arerelatively younger and in better condition. The program implemented must have asufficiently long time frame which can address both the immediate and the future needsof these systems.As with so many considerations in our nation, equity among regions must be267maintained to enhance the national character, standards, and condition of all systems.This will mean appropriating of funds by the federal government based on need willhave to be properly balanced by equity concerns among all regions. In any case, theimportance of water infrastructure and indeed all infrastructure must be promoted for itsinherent benefits, not just as excuses for “make-work” employment exercises. As such,research and policies for roads, sewers, storm systems, treatment facilities, and bridgesshould also be promoted at the national levels. Studies similar in purpose and content tothis thesis, focusing on the unique character of Canadian systems, must also be promotedin the other component systems of our physical infrastructure. The standard of livingand the fabric of our nation did not merely happen by accident and should not beallowed to diminish by accident. Rather than being taken for granted, infrastructuresystems must be actively promoted and supported. All too often we forget the real needfor effective, efficient, and safe infrastructure systems until a water main breaks - and weare thirsty.268BIBLIOGRAPHY269BIBLIOGRAPHYAdams, B.J. and G.W. Heinke; 1987; “Canada’s urban infrastructure: rehabilitation needsand approaches”; Canadian Journal of Civil Engineering; No. 14; 1987; pp. 700-707.Alberta Municipal Affairs; 1943-1988; “Annual Report: Sunimary Municipal Statistics”;Department of Municipal Affairs; Government of Alberta; Edmonton, Alberta; 1943,1946, 1950, 1951, 1956, 1958, 1961, 1966, 1971, 1981, 1986, 1988.Alberta Environment; 1988; Standards and Guidelines for Municipal Water Supply.Wastewater. and Storm Drainage Facilities; Alberta Environment; Edmonton; March,1988.Aliman, R.J. and J.D. Dilay; 1983; “Policies for Pipeline Safety Near Urban Areas”;Proceedings on the Conference on Pipelines in Adverse Environment II; PipelineDivision of the ASCE; San Diego, Calif.; Nov. 14-16; 1983; pp. 212 - 237.Amborski, D. and E. Slack; 1987; “Federal, Provincial, Municipal Co-operation: Past,Present, and Future”; The Proceedings of the First Canadian Conference on UrbanInfrastructure; Sodanell Canada Inc.; Edmonton, Alberta; February 5-6, 1987; pp. 37 -52.Anderson, J.D.; 1979; “The Municipal Government Reform Movement in WesternCanada, 1880-1920”; Usable Urban Past: Planning and Politics in the Modern Canadian.ity; edited by F.J. Artibise and G.A. Stelter; MacMillan Company of Canada Limited inassociation with the Institute of Canadian Studies; Carleton University; Ottawa; 1979.Anderson, L.; 1988; “Water Supply”; Building Canada: A History of Public Works;Norman R. Ball, Senior Editor; Canadian Public Works Association; University ofToronto Press; Toronto; 1988.Andreaou, S. and D. Marks; 1987; “Maintenance Decisions for Deteriorating WaterPipelines”; Journal of Pipelines; Elsevier Science Publishers B.V., Amsterdam, TheNetherlands; Vol. 7; 1987; pp. 21-31.Arnold, G.E.; 1960; “Experience with main breaks in four large cities”; Journal of theAmerican Water Works Association; 53(8); 1960; pp. 1041 - 1044.Artibise, Alan F.J.; 1981; Town and City: Aspects of Western Canadian UrbanDevelopment; Canadian Plains Research Centre; University of Regina; Regina, Sask.;1981.ASCE; 1989; Reliability Analysis of Water Distribution Systems; Task Committee onRisk and Reliability Analysis of Water Distribution Systems of the Committee onProbabilistic Approaches to Hydraulics of the Hydraulics Division of the AmericanSociety of Civil Engineers; Edited by Larry W. Mays; American Society of Civil270Engineers; New York, New York; 1989; 532 p.AWWA; 1974; “A Study of the Problem of Asbestos in Water”; AWWA Journal Part 2;American Water Works Association; Denver, Colorado; 1974.AWWA; 1983; Manual of water supply practices: Water rates; American Water WorksAssociation; Ml; Denver, Colorado; 1983.AWWA; 1986a; Guidance Manual: Water Main Evaluation for Rehabilitation /Replacement; Report prepared by D. Kelly O’Day; American Water Works Association;Denver, Colorado; 1986; 182 p.AWWA; 1986b; “Are Asbestos Fibres in Drinking Water Harmful?”; AWWA Journal, 78,No. 111; American Water Works Association; Denver, Colorado; November, 1986; pp.18-26.AWWA; 1988; Bacterial Regrowth in Distribution Systems; Report prepared by W.G.Characklis; American Water Works Association; Denver, Colorado; January, 1988; 331 p.AWWA; 1989; Economics of Internal Corrosion Control; American Water WorksAssociation; Denver, Colorado; October, 1989.B.C. Municipal Affairs; 1980; Residential Services and Site Planning Standards; Ministryof Municipal Affairs; Province of British Columbia; Victoria, B.C.; 1980.B.C. Municipal Affairs; 195 1-1988; Municipal Statistics; Ministry of Municipal Affairs;Province of British Columbia; Victoria, B.C.; 1951, 1952, 1953, 1955, 1956, 1958, 1966,1971, 1975, 1976, 1981, 1982, 1985, 1986, 1988.Baldwin, 1988; Building Canada: A History of Public Works; Norman R. Ball, SeniorEditor; Canadian Public Works Association; University of Toronto Press; Toronto; 1988.Barclay, D.S.; 1984; “Retrofitting apartment buildings to reduce costs and water demand”;Canadian Water Resources Journal; 9(3); 1984; pp. 45-47.Birchan, P.D. and W.K. Bond; 1984; “The Impacts of Land Use on CMHC MunicipalInfrastructure Assistance, 1961 to 1981”; Working Paper No. 32; Lands Directorate;Environment Canada; March, 1984.Bloomfield, E., G. Bloomfield, and P. McKaskel; 1983; Urban Growth and LocalServices: The Development of Ontario Municipalities to 1981; Occasional Papers inGeography No. 3; Dept. of Geography; University of Guelph; Guelph, Ont.; September,1983.Bradley, J.; 1987; “Political and Financial Challenges of Infrastructure Renewal;JProceedings of the First Canadian Conference on Urban Infrastructure; Sodanell CanadaInc.; Edmonton, Alberta; February 5-6, 1987; pp. 31 - 35.271Bratton, George N.; 1990; Personal Interview; Water Works Engineer; City ofVancouver, B.C.; May 31, 1990.Bratton, G.N., G.P. Duppong, and D.D. Green; 1986; “Planning of WatermainReplacement Using Time-To-Failure Analysis and Risk Assessment Techniques”; APaper presented at the Conference of the Pacific Northwest Section of the AmericanWater Works Association; Portland, Oregon; 1986.Brooks, D.B., and R. Peters; 1988; “Water: The potential for demand management inCanada”; Discussion Paper; Science Council of Canada; Ottawa; 1988; as cited in Tate1990.Buelow, R.W., J.R. Millette, E.F. McFarren and J.M. Symons; 1980; “The Behavior ofAsbestos-Cement Pipe Under Various Water Quality Conditions: A Progress Report”;AWWA Journal; Research and Technology; February, 1980; pp. 91-102.Bury, K.; 1975; Statistical Models in Applied Science; John Wiley; New York; 1975.Cain, L.P.; 1976; “Water and Sanitation Services in Vancouver: An HistoricalPerspective”; B.C. Studies; University of British Columbia Press; Vancouver, B.C.; No.30; Summer 1976; pp. 27-43.Canada World Almanac; 1990; The Canadian World Almanac and Book of Facts; GlobalPress; Canada Publishing Corporation; Toronto, Ontario; 1990; 691 p.Canadian Geotechnical Society; 1985; Canadian Foundation Engineering Manual; 2ndEdition; Canadian Geothechnical Society; BiTech Publishers Ltd.; Vancouver, B.C.;1985; 456 p.Caproco; 1985; Underground Corrosion of Water Pipes in Canadian Cities Case: TheCity of Calgary; Final Report; Prepared by Caproco Corrosion Prevention Ltd.;Edmonton, Alberta; Published by Minister of Supply and Services Canada, Ottawa; 1985.Careless, J.M.S.; 1978; The Rise of Canadian Cities in Canada before 1914; CanadianHistorical Association; No. 32; Love Printing Services Ltd.; Ottawa; 1978.CBA; 1981; Feasibility Study for Upgrading and Extending Water Supply in PittMeadows and Pitt Polder; Report No. 811; B.C. Ministry of Environment prepared byCBA Engineering Ltd.; Vancouver, B.C.; February, 1981.City of Winnipeg; 1991; “DRAFT”: Guidelines for Usage of Trenchless Methods for theRehabilitation or Renewal of Sewer Infrastructures; City of Winnipeg, Works andOperations Division; Winnipeg, Manitoba; May, 1991; 10 p.Chambers R.E. and T.J. McGrath; 1981; “Structural Design of Buried Plastic Pipe”;Proceedings of the First International Conference on Underground Plastic Pipe; PipelineDivision of the American Society of Civil Engineers; New Orleans, Louisiana; March27230 - April 1, 1981; PP. 10 - 25.Clark, R.M., C.L. Stafford, and J.A. Goodrich; 1982; “Water Distribution Systems: ASpatial and Cost Evaluation”; Journal of the Water Resources Planning and ManagementDivision; American Society of Civil Engineers; Vol. 108; No. WR3; October, 1982.Clarke, N.W.B.; 1968; Buried Pipelines: A Manual of Structural Design and Installation;MacLaren and Sons Ltd.; London; 1968.Conimins, B.T.; 1979; Asbestos in Drinking Water; Water Research Centre TechnicalReport TR100; Medmenham Laboratory; Buckinghamshire, England; 1979.Cotruvo, J.; 1984; “Contaminants in Drinking Water”; National Conference on CriticalIssues in Drinking Water Oualitv; Federation of Associations on the CanadianEnvironment; Ottawa; Feb. 6 and 7; 1984.Cross, Greg; 1991; Telephone Interview; Engineering Technologist; The Corporation ofthe District of Pitt Meadows, B.C.; July 24, 1991.Curtis, W.H.; 1987; “State of Urban Infrastructure in Canada”; The Proceedings of theFirst Canadian Conference on Urban Infrastructure; Sodanell Canada Inc.; Edmonton,Alberta; February 5-6, 1987; pp. 1-16.Curtis, W.H.; 1991a; Personal Interview; City Engineer, City of Vancouver; June 13,1991.Curtis, W.H.; 1991b; “Report to the Task Force on Municipal Infrastructure FromChairman of the Technical Committee on Infrastructure: Status of MunicipalInfrastructure Renewal”; unpublished report; March 5, 1991.Davey, T.; 1987; “Why Canadians can’t afford cheap water”; Water and PollutionControl; Vol. 125; No. 5; October, 1987; p. 5.Davis, M.J., B.L. Haerndon, E.P. Shea, and M.K. Snyder; 1979; “Occurrence, EconomicImplications and Health Effects Associated with Aggressive Waters in Public WaterSupply Systems”; Final Report, Midwest Research Institute Project No. 4552-L; KansasCity, Missouri; 1979.deCamp, L. Sprague; 1990; The Ancient Engineers; Dorset Press; New York; 1990.Denis, Leo G.; 1912; Water-Works of Canada; Commission of Conservation; MurtimerCo.; Ottawa; 1912.Denis, Leo G.; 1916; Water Works and Sewerage Systems; Commission of Conservation;Ottawa; 1916; p.5.DoE and NWC; 1977; Sewers and Water Mains - a National Assessment; DoE and273NWC Standing Technical Committee Report No. 4; NWC; London, England; 1977; 34p.; as cited in Edwards and Cox 1982.Doherty, D.C., N. Vardin, and C.S. Kitchen; 1987; “Upkeep of Municipal Services: TheToronto System”; The Proceedings of the First Canadian Conference on UrbanInfrastructure; Sodanell Canada Inc.; Edmonton, Alberta; February 5-6, 1987; pp. 229-254.Dunlop; 1983; Rebuilding America’s Vital Public Facifities; Labour Management Group;Harvard University; 1983; as cited in FCM 1984.Edwards, G. and G.C. Cox; 1982; “The renewal of urban water and sewerage systems”;Planning and the Civil Engineer; proceedings of a joint conference of the AmericanSociety of Civil Engineers, the Canadian Society for Civil Engineering and the Institutionof Civil Engineers, held in Torquay, September 20-25, 1981; Thomas Telford Limited;London; 1982.Environment Canada; 1975; Canada Water Year Book; Environment Canada; publishedby Information Canada; Ottawa; 1975.Environment Canada; 1983; Canada Water Year Book: Water and the economy;Environment Canada; published by the Minister of Supply and Services Canada; Ottawa;1983.Environment Canada; 1986; Stopping Acid Rain; Government of Canada, Departmentof the Environment; Ottawa; 1986; 5 p.Environment Canada; 1987; Federal Water Policy; Government of Canada, Minister ofthe Environment; Ottawa; 1987; p. 21.Environment Canada; 1990a; A Primer on Water: Ouestions and Answers; EnvironmentCanada; Ottawa; 1990.Environment Canada; 1990b; Federal Water Policy: A Progress Report;Interdepartmental Committee on Water; Government of Canada, Minister of theEnvironment; Ottawa; March, 1990; p. 32-35.ERCB; 1983; An Analysis of Pipeline Performance in Alberta; Energy ResourcesConservation Board; Calgary, Alberta; 1983.ESE; 1991; “Corrosion control in Calgary’s water distribution system”; EnvironmentalScience and Engineering; July, 1991; pp. 44-46.FACE; 1975; National Inventory of Municipal Waterworks and Wastewater Systems inCanada - 1975; Federation of Associations on the Canadian Environment andparticipating Federal and Provincial agencies; published by the Minister of Supply andServices Canada; Ottawa; 1975.274FACE; 1978; National Inventory of Municipal Waterworks and Wastewater Systems inCanada - 1977; Federation of Associations on the Canadian Environment andparticipating Federal and Provincial agencies; published by the Minister of Supply andServices Canada; Ottawa; 1978.FACE; 1981; National Inventory of Municipal Waterworks and Wastewater Systems inCanada - 1981; Federation of Associations on the Canadian Environment andparticipating Federal and Provincial agencies; published by the Minister of Supply andServices Canada; Ottawa; 1981.FACE; 1987; National Inventory of Municipal Waterworks and Wastewater Systems inCanada - 1986; Federation of Associations on the Canadian Environment andparticipating Federal and Provincial agencies; published by the Minister of Supply andServices Canada; Ottawa; 1987.FCM; 1984; Canada’s urban Infrastructure: Physical Condition and Funding Adequacy; areport prepared for the Federation of Canadian Municipalities Task Force on MunicipalInfrastructure by the Technical Committee on Canada’s Urban Infrastructure; Ottawa;June, 1984.FCM; 1989; Defining Minimum Maintenance Standards for Infrastructure FeasibilityEvaluation; a report prepared for the Federation of Canadian Municipalities Task Forceon Municipal Infrastructure by the Technical Committee on Canada’s UrbanInfrastructure; Ottawa; September, 1989.FCM; 1991a; “Impact of Infrastructure Improvement on the Environment, Public Health,and Safety”; unpublished report of the Federation of Canadian Municipalities TechnicalCommittee on Canada’s Urban Infrastructure; Ottawa; May 28, 1991.FCM; 1991b; “Minutes of the Infrastructure Technical Meeting - Ottawa May 27, 1991”;unpublished minutes of the Federation of Canadian Municipalities Technical Committeeon Canada’s Urban Infrastructure; Ottawa; May 27, 1991.Fedotoff, R., Frutchey, R. and Thomson, J; 1990; “Pipe Rehab: Old Notions Challenged”;WATER / Engineering & Management; April, 1990; pp. 24-27.Flack, E.J.; 1981; “Residential water conservation”; Journal of Water Resources PlanningManagement Division of the ASCE; ASCE; Vol. 107; No. 1; 1981; p. 85-95.FUS; 1981; Water Supply for Public Fire Protection: A Guide to RecommendedPractice; Fire Underwriters Survey; Insurance Bureau of Canada; Toronto, Ontario;1981; 24 p.Graf, F.L.; 1989; “Locating Pipe and Piping Leaks Using Ground Penetrating Radar”;Proc. of the First International Conference on Underground Infrastructure Research;Nov. 15-17, 1988; AWWA; Denver; published 1989; pp. 61-72.275Grigg, N.S.; 1986; “Water Infrastructure: Research Needs and Education Needs”; WaterForum ‘86; World Water Issues in Evolution; Proceedings of the Conference Sponsoredby the Hydraulics Division of the American Society of Civil Engineers; Vol. 2; LongBeach, California; Aug. 4-6, 1986.Grigg, N.S.; 1988; Infrastructure Engineering and Management; John Wiley and Sons,Inc.; New York, New York; 1988; 380 p.Grima, A.P.; 1972; Residential demand for water: alternative choices for management;University of Toronto Press; Toronto, Ontario; 1972; as cited in Tate 1989.Grover, B. and D. Zussman; 1985; “Safeguarding Canadian Waters”; Inquiry on FederalWater Policy Research Paper No. 4; Environment Canada; Ottawa; March, 1985.Grover, A.; 1990; “Speech to the Canadian Conference on Urban Infrastructure”;Proceedings of the Second Canadian Conference on Urban Infrastructure; February 13and 14, 1989; Sodanell Canada Inc.; Edmonton, Alberta; 1990.GVRD, 1990; Greater Vancouver Key Facts: A Statistical Profile of Greater Vancouver.Canada; GVRD Development Services; Burnaby, B.C.; 1990; 82 p.Hall, R.; 1984; “The Quality of Existing Water Systems and Public Perception”; NationalConference on Critical Issues in Drinking Water Quality; Federation of Associations onthe Canadian Environment; Ottawa, Ontario; February 6 & 7, 1984; pp. 46-49.Haney, P.E. and Hagar, P.E.; 1985; “Dual water system design”; AWWA Seminarproceedings - Dual water systems; No. 20189; Conference of the American Water WorksAssociation; Denver, Colorado; June 23, 1985; as cited in Tate 1990.Hanke, S.H.; 1978; “A method for integrating engineering and economic planning”;Journal of the American Water Works Association; AWWA; Vol. 70; No. 9; pp. 487-491;as cited in Tate 1990.Health and Welfare Canada; 1989; Guidelines for Canadian Drinking Water Ouality;Fourth Edition; Minister of National Health and Welfare; published by the Minister ofSupply and Services Canada; Ottawa; 1989; 25 p.Heinke, G.W. and E.J. Bowering; 1984; “Literature Review of municipal Infrastructure”;Appendix “A” of Canada’s urban Infrastructure: Physical Condition and FundingAdequacy; a report prepared for the Federation of Canadian Municipalities Task Forceon Municipal Infrastructure by the Technical Committee on Canada’s UrbanInfrastructure; Ottawa; June, 1984.Hennigar, G.W.; 1984; “Water Leakage Control and Sonic Detection”; Canadian WaterResources Journal; Vol. 9; No. 3; 1984; pp.48-50.Hickman, R.; 1984; “Hazards in Drinking Water”; National Conference on Critical Issues276in Drinking Water Oualitv; Federation of Associations on the Canadian Environment;Ottawa, Ontario; February 6 & 7, 1984; PP. 91-98.Howe, C.W. and F.P. Linaweaver, Jr.; 1967; “The impacts of price on residential waterdemand and its relation to system design and price structure”; Water Resour. Res.; 3(1);pp. 13-32; as cited in Tate 1989.Hudson, W.R. and Haas R.; 1978; Pavement Management Systems; McGraw - Hill BookCompany; New York; 1978.Hunsinger RB., D.B. Jobb, K.J. Roberts, P.J. Roberts, O.R. Lindow and J.F. Wainright;1989; Asbestos in Ontario Drinking Water Supplies; Water Resources Branch andLaboratory Services Branch; Ontario Ministry of Environment; Queen’s Printers forOntario; 1989.IBC; 1989; Facts: Facts of the General Insurance Industry in Canada; Insurance Bureauof Canada; Toronto; 1989.IUCN; 1980; “World Conservation Strategy”; International Union for Conservation ofNature and Natural Resources; Gland, Switzerland; 1980; p. 1; as cited in Tate 1990.James, B. and Nq, W.; 1991; “Corrosion control in Calgary’s water distribution system”;Environmental Science & Engineering; July, 1991; pp. 44-46.Kettler A.J. and I.C. Goulter; 1985; “An analysis of pipe breakage in urban waterdistribution networks”; Canadian Journal of Civil Engineering; 1985; pp. 286 - 293.Lalani, Aniin; 1990; Personal Interview; Water and Sewer Engineer; Municipality ofSurrey, B.C.; June 21, 1990.Loudon, R.M.; 1986; “Municipal experiences in pricing and water conservation”; A paperdelivered at the Canadian Public Works Conference and Equipment Show; Ottawa,Ontario; May 11-14, 1986; as cited in Tate 1990.Lowry, Jim; 1991; Personal Interview; Director of Engineering; The Corporation of theDistrict of Pitt Meadows, B.C.; September 4, 1991.MacLaren, J.W.; 1983; Report to the Ministry of the Environment of the Province ofOntario on a Water Distribution and Sewerage Rehabilitation Program for OntarioMunicipalities; prepared by James W. MacLaren Consulting Engineer; Ontario Mimistryof the Emvironment; Toronto, Ontario; 1983.MacLaren, J.W.; 1985; “Municipal Waterworks and Wastewater Systems”; Inqui onFederal Water Policy Research Paper No. 3; Environment Canada; Ottawa; January,1985.MacLaren, J.W.; 1987; “Underground Piping Systems”; Proceedings of the First Canadian277Conference on Urban Infrastructure; Sodanell Canada Inc.; Edmonton, Alberta;February 5-6, 1987; PP. 137-152.MacLaren, James W.; 1991; Telephone Interview; Technical Advisor, Ontario WaterServices Secretariat; Toronto, Ontario; May 24, 1991.MacLean, John; 1991; Telephone Interview; District of North Vancouver; DesignEngineer/Supervisor; July 15, 1991.Macleans; 1990; “Danger in the Water”; Macleans Magazine; Maclean Hunter CanadianPublishing; Toronto, Ontario; Vol. 103; No. 3; January, 1990; pp. 30-41.Marmion, J.B.; 1988; “Health Effects of Drinking Water Contaminants”; Journal of theNew England Water Works Association; Vol. 102; No. 2; March 1988; pp. 6-12.Marks, D.H.; 1987; “Maintenance Management in Urban Water DistributionInfrastructure”; Third Canadian Seminar on systems Theory for the Civil Engineer”; Ecol.Polytech.; Montreal, Quebec; June 10, 1987.Marks, D.H., S. Andreou, and R. Clark; 1986; “Guidelines for Maintenance of AgingWater Pipes”; Water Forum ‘86: World Water Issues in Solution; 1986; pp. 1225 - 1231.Mavinic, Donald 5.; 1990; Personal Interview; Professor of Civil Engineering; Universityof British Columbia; May 9, 1990.Mays, L.W.; 1989; “Introduction”; Reliability Analysis of Water Distribution Systems; Bythe Task committee on Risk and reliability Analysis of Water Distribution Systems of theCommittee on Probabilistic Approaches to Hydraulics of the Hydraulics Division of theAmerican Society of Civil Engineers; American Society of Civil Engineers; New York,New York; 1989; pp. 1-10.McClelland, N.I.; 1981; “Monitoring for Toxicological Safety”; International Conferenceon Underground Plastic Pipe; Sponsored by the Pipeline Division of the AmericanSociety of Civil Engineers; New Orleans, Louisiana; Mar. 20 - Apr. 1, 1981; pp. 401-419.McColl, S.R.; 1985; “Risk Assessment and Standard-Setting for Control of ChemicalHazards in Canadian Drinking Water”; A paper prepared for the Risk ManagementWorkshop; Inquiry on Federal Water Policy Research Paper #19; EnvironmentalProtection Service; Environment Canada; Hull, P.Q.; June, 1985.McIntyre, C.E. and J.C. Elstad; 1987; “A Rehabilitation Program for Water and SewerServices in Ontario”; Proceedings of the First Canadian Conference on UrbanInfrastructure; Sodanell Canada Inc.; Edmonton, Alberta; February 5-6, 1987; pp. 279-301.Millette, L. and D.S. Mavinic; 1988; “Effect of pH adjustment on the internal corrosionrate of residential cast-iron and copper water distribution pipes”; Canadian Journal of278Civil Engineering; Vol. 15; Feb. 1988; PP. 79-90.Montreal; 1979-1984; Annual Report: Ville de Montreal; Montreal; 1979, 1980, 1981,1982, 1983, 1984.Morris, R.E.; 1967;” Principal Causes and Remedies of Water Mian Breaks”; Journal ofthe American Water Works Association; 59(7); 1967; pp. 782 - 798.Morrison, A.; 1984; “Is Canada’s Drinking Water Safe?”; National Conference on CriticalIssues in Drinking Water Quality; Federation of Associations on the CanadianEnvironment; Ottawa; Feb. 6 and 7, 1984.Moruzzi, L.; 1987; “Reasons for Pipe Damage”; Water Supply; Vol. 5; No. 3/4; 1987; pp.SS16-1 to SS16-8.Moser, A.P.; 1990; Buried Pipe Design; McGraw Hill, Inc.; New York, New York; 1990;219 p.MOT; 1973; Estimating the Depth of Pavement Frost and Thaw Penetrations; Ministryof Transportation, Canadian Air Transportation Administration, ConstructionEngineering and Architectural Branch, Engineering Design Division; Report No. CBED6-266; Ottawa; February 1973; 26 p.MU; 1961; “Statistical Summary of Waterworks Systems in Canada”; Municipal Utilities;J.P. Russell, Editor; Monetary Times Publications Ltd.; Toronto; February, 1961.MU; 1951; “Statistical Summary of Waterworks Systems in Canada”; Municipal Utilities;J.P. Russell, Editor; Monetary Times Printing Co. of Canada Ltd.; Toronto; Vol. 89; No.12; December, 1951.MUD; 1990; “Municipal (Water) Use Database (MUD)”; File 89MUD.WK1; WaterPlanning and Management Branch, Inland Waters Directorate, Environment Canada;Ottawa, Ontario; April 5, 1990.Nebesar, B. and G.W. Riley; 1984; Asbestos/Cement Pipe Corrosion Part 1 - Historical.Technological. Economic and Statistical Background; Canadian Centre for Mineral andEnergy Technology; Energy, Mines and Resources Canada; Minister of Supply andServices Canada; Ottawa; October 1984.Nova Scotia Municipal Affairs; 1965-1988; Annual Report of Municipal Statistics;Department of Municipal Affairs; Government of Nova Scotia; Halifax, Nova Scotia;1965, 1971, 1977, 1986, 1988.O’Day, K.D.; 1983; “Analyzing Infrastructure Condition - A Practical Approach”; CivilEngineering; American Society of Civil Engineers; April, 1983; pp. 39-42.O’Day, K.D.; 1984; “Aging Water Supply Systems: Repair or Replace?”; Infrastructure:279Maintenance and Repair of Public Works; Annals of the new York Academy of Science;New York, New York; December 5, 1984; pp. 24 1-258.Ontario Municipal Statistics; 1934-1967; Annual Report of Municipal Statistics;Department of Municipal Affairs; Government of Ontario; Toronto, Ontario; 1934, 1941,1942, 1951, 196 1,1967.Parker, M.E. and E.G. Peattie; 1984; Pipe Line Corrosion and Cathodic Protection; GulfPublishing Company; Houston, Texas; 1984; 166 p.Philips, M.R., F. Wu, D. Maudie, and T. Field; 1991; “Management of Urban Drainage:Challenges and Trends”; Proceedings of the Second Canadian Conference on UrbanInfrastructure; Edmonton, Alberta, Feb. 13 and 14, 1989; Sodanell Canada Inc.;Edmonton, Alberta; 1991.Postel, S.; 1985; “Conserving water: The untapped alternative”; Paper 67; WorldwatchInstitute; Washington, D.C.; 1985; as cited in Tate 1990.Price, R.E. and S.M. Wolf; 1989; “Non-Destructive Evaluations of Pipelines”; Proc. of theFirst International Conference on Underground Infrastructure Research; Nov. 15-17,1988; AWWA; Denver; published 1989; pp. 73-104.Robinson, J.E.; 1980; “Demand modification as a supply alternative: A case study of theRegional Municipality of Waterloo, Ontario, Canada”; Department of EnvironmentalStudies; University of Waterloo, Waterloo, Ontario; Unpublished ms as cited in Tate1990; 1980.Robinson, Bill; 1991; Municipality of Burnaby; Public Works Superintendent; TelephoneInterview; June 26, 1991.Rodi, S.; 1987; “Infrastructure Management: A Geographic Information SystemApplication”; Proceedings of the First Canadian Conference on Urban Infrastructure;Sodanell Canada Inc.; Edmonton, Alberta; February 5-6, 1987; pp. 99 - 106.Ryder, R.A.; 1989; “Corrosion of Underground Services”; Proc. of the First InternationalConference on Underground Infrastructure Research; Nov. 15-17, 1988; AWWA;Denver; published 1989; pp. 147-158.SCC; 1989; “Laurentide Motels v. Beauport (City)”; Supreme Court of Canada Decision;L’Heureux-Dube Judge; 1 R.C.S.; April 20, 1989.Shamir, U. and D.D. Howard; 1979; “An Analytic Approach to Scheduling PipeReplacement”; Journal of the American Water Works Association; AWWA; May 1979;pp. 248-258.Shilts, W.W.; 1981; Sensitivity of Bedrock to Acid Precipitation: Modification by GlacialProcesses; Geological Survey of Canada; Government of Canada; Ottawa; Paper 81-14;2801981; 7 p.Smith, H.W.; 1989; Corrosion Management in Water Supply Systems; Van NostrandReinhold; New York, New York; 1989; 125 p.Smith, J.Grove; 1918; Fire Waste in Canada; Commission of Conservation; Ottawa; 1918;p. 216-217.Somers, E.; 1984; “The Risk from Drinking Water”; National Conference on CriticalIssues in Drinking Water Oualitv; Federation of Associations on the CanadianEnvironment; Ottawa; Feb. 6 and 7, 1984.Stein, D., K. Mollers and R. Bielecki; 1989; Microtunnelling: Installation and Renewal ofNonman - Size Supply and Sewage Lines by the Trenchless Construction Method; Ernst& Sohn Verlag fur Architektur und technische Wissenschaften; Berlin, Germany; 1989;352 p.Stephenson, D.; 1989; Pipeline Design for Water Engineers (Third Revised and UpdatedEdition; Elsevier Science Publishing Company Inc.; New York, New York; 1989; 263 p.Surrey; 1978; Waterworks Manual; District of Surrey, B.C.; Surrey EngineeringDepartment; March 1978.Tate, D.M.; 1990; Water Demand Management in Canada: A State-of-the-Art Review;Inland Waters Directorate, Water Planning and Management Branch, EnvironmentCanada; Ottawa; Social Sciences Series No. 23; 1990; 52 p.Tate, D.M. and D.M. Lacelle 1987; Municipal Water Use in Canada; Water Planningand Management Branch; Inland Waters/Lands Directorate; Environment Canada;Ottawa; Social Sciences Series No. 20; 1987.Tate, D.M.; 1989; Municipal Water Rates in Canada. 1986 - Current Practices andPrices; Inland Waters Directorate, Water Planning and Management Branch,Environment Canada; Ottawa; Social Sciences Series No. 21; 1989; 16 p.Toft, P. and M.E. Meek; 1983; “Asbestos in Drinking Water: A Canadian View”;Environmental Health Perspectives; Vol. 53., pp. 177-180; March, 1983.Toronto; 1971; Annual Report; City of Toronto; 1971.Vagt, G.O.; 1980; “Asbestos”; Canadian Minerals Yearbook 1980; Energy, Mines andResources Canada; published by the Minister of Supply and Service Canada; Ottawa;1980. VVardin, N.; 1991; “Keele Street Microtunnelling Project - Toronto, Ontario”;Unpublished paper by Nicholas Vardin, P.Eng., City Engineer and Commissioner ofPublic Works and the Environment, City of Toronto; Toronto, Ontario; May 23, 1991.281Vernooy, B. and Jordan, S.; 1983; “An Inside Look at the Pipeline”; Proceedings on theConference on Pipelines in Adverse Environment II; Pipeline Division of the ASCE; SanDiego, Calif.; Nov. 14-16, 1983; pp. 397-414.Versamie, D.; 1987; “Causes of Water Main Damage”; Water Supply; Vol. 5; No. 3/4;1987; pp. SS16-8 to SS16-12.Walski, T.M.; 1982; “Economic Analysis of Rehabilitation of Water Mains”; Journal ofthe Water Resources Planning and Management Division; American Society of CivilEngineers; 108(WR3); October 1982; pp. 296 308.Wareham, D.G. and E.A. McBean; 1985; “Assessment of Municipal Practices ofCommunities with Respect to Fireflow Guidelines”; Canadian Water Resources Journal;Vol. 10; No. 3; 1985; pp. 14-26.Wilging, R.C.; 1981; “Plastics in Piping”; Proceedings of the International Conference onUnderground Plastic Pipe; Sponsored by the Pipeline Division of the American Societyof Civil Engineers; New Orleans, Louisiana; March 30 - April 1, 1981; pp. 1-9.Williams, H.G.; 1987; “Legislation Needed to Rehabilitate Our Water and SewerInfrastructure”; Infrastructure: Maintenance and Repair of Public Works; Annals of theNew York Academy of Sciences; v. 431; New York; 1984; pp. 82 - 85.Winnipeg; 1971-1986; Annual Financial Report; City of Winnipeg; 1971, 1976, 1981,1986.Wilging, R.C.; 1981; “Plastics in Piping”; Proceedings of the International Conference onUnderground Plastic Pipe; Sponsored by the Pipeline Division of the American Societyof Civil Engineers; New Orleans, Louisiana; March 30 - April 1, 1981; pp. 1-9.Wilging, R.C.; 1981; “Plastics in Piping”; Proceedings of the International Conference onUnderground Plastic Pipe; Sponsored by the Pipeline Division of the American Societyof Civil Engineers; New Orleans, Louisiana; March 30 - April 1, 1981; pp. 1-9.Winnipeg; 1931-1966; Report of the Commissioner of Finance; City of Winnipeg; 1931,1932, 1936, 1941, 1946, 1951, 1956, 1961, 1966.Wood, D.; 1987;”Urban Infrastructure: Municipal Liability”; Proceedings of the FirstCanadian Conference on Urban Infrastructure; Sodanell Canada Inc.; Edmonton,Alberta; February 5-6, 1987; pp. 185-197.Young, O.C. and J.J. Trott; 1984; Buried Rigid Pipes: Structural Design of Pipelines;Elsevier Applied Science Publishers Ltd.; Essex, England; 1984; 234 p.282APPENDIX A283APPENDIX A: DATA SOURCES FOR THE NATIONAL INVENTORYA.1. Database InformationThe lack of a comprehensive inventory of Canada’s underground plant has beennoted by a variety of sources (Tupper 1981, FCM 1984, MacLaren 1985) and to date noagency keeps such records on a national scale.Although a variety of information exists within individual municipalities in someform or another, no agency has yet to compile a comprehensive data base, although the1984 FCM survey was a first step (FCM 1984). In England a comprehensive survey wascarried out in 1977, from which courses of action could be taken (DoE and NWC 1977).Historical data from provincial municipal statistics reports, trade journals, annualreports of various cities, and past system surveys is used to estimate the amount of pipebeneath Canada’s villages, towns, and cities (see Table A.1).Table Al: Data Sources Used in Estimating the Canadian Water PiDes InventorySOURCE EXTENT COMMUNITIES DATA INCLUDED YEARSDenis 1912 National alt Population; Pipe length, type, size range 1912Municipal Utilities 1951 National pop. > 1,000 Population; Pipe length, type, size range 1951Municipal Utilities 1961 National pop. > 1,000 Population; Pipe length, type, size range 1961Provincial Municipal B.C. all Population; Pipe Length 1951-1988Statistics Reports Alberta all Population; Pipe Length 1943-1988(Years Vary) Ontario all Population; Pipe Length 1934-1967Nova Scotia all Population; Pipe Length 1965-1988Annual Reports Winnipeg n/a Pipe Length 1932-1986(Years Vary) Montreal n/a Pipe Length 1979-1984Toronto n/a Pipe Length 1971FCM 1984 (survey data) National limited Pipe Length 1984FACE 1975 National all Population Served; Number of Systems 1975FACE 1978 National all Population Served; Number of Systems 1977FACE 1987 National all Population Served; Number of Systems 1986284Unfortunately none of the sources are ideal, neither individually nor when combined.There remains major gaps in the data, but what does exist is still very useful in formingan initial picture of what for now cannot be seen below our streets.One of the earliest comprehensive surveys of Canada’s piped water systems wascarried out by the Federal Government’s Commission of Conservation in 1912 (Denis1912). On a community by community basis, this listing provides information on pipelengths by the material type and the minimum to maximum pipe diameter range in asystem. The survey also includes information on the population served, the ownership,the nature of the supply source (ie lake or river), the operating pressure, theconsumption, the number and capacity of storage reservoirs, as well as a count of thevalves, services, and hydrants on the each community’s system.Two surveys of a similar format were carried out by Municipal Utilities in 1951and 1961, but the sample was limited to communities of at least 1,000 in population.Later surveys carried out by the publication did not include information specific to thedistribution system such as pipe length and type.Additional information which is useful in the estimation of pipe length has beenextracted from various provincial municipal statistics publications which are typically putout by the provincial ministry or department responsible for municipal affairs. Althoughthese publications are primarily summaries of the fiscal situation of individualmunicipalities, some provinces include information on the length of public services suchas roads, sewers, and water mains. Unfortunately, the format of each publication andthe information gathered varies both from year to year and from province to province285(see Table A.1). Both historical and current statistics for British Columbia, Alberta, andNova Scotia were obtained from these records, while the Ontario reports were onlyavailable up to 1967 and thus proved to be much less useful. Provinces such asSaskatchewan, Manitoba, Quebec, New Brunswick, Newfoundland, and Prince EdwardIsland do not publish such statistics in a readily available publication.The annual report of larger cities Winnipeg and Montreal are used to obtainrecent estimates of system length. Unfortunately, only large centres publish and widelydistribute such reports, thus limiting their usefulness as an accessible source.Pipe length data from the FCM survey in 1984 has been obtained for anadditional 14 municipalities in Canada and estimated for 1986. The municipalities areall in the large or medium sized category and are all located outside of the threeprovinces of B.C., Alberta, and Nova Scotia. A few of the municipalities included areToronto, North York, York, Sault Ste. Marie, London, Quebec City, Fredericton, andRegina.The final source which is used in the 1986 estimate of pipe length in Canada isthe National Inventory of Municipal Waterworks and Wastewater Systems in Canada1986 (FACE 1986) which provides the most comprehensive and current information onthe number of communities in Canada with water distribution systems and thecorresponding populations served. Two similar documents (FACE 1975, FACE 1978)which were published earlier are not used extensively in this report.286A.2. MethodologyThe data is stored in a spreadsheet program, having been transferred from thevarious publications by the author. Heinke and Bowering (1984) estimate that in 1981there were 100,000 km of water main in Canada based on a served population of20,275,786 (FACE 1981) and an average density of 200 persons per kilometer ofwatermain, a value extracted from an Ontario Department of the Environment report(MacLaren 1983).A.2.1. Missing DataNationally, the data available gives a very accurate total system length for 1912, areasonably accurate length for both 1951 and 1961, and a reasonably good approximationof the length in 1986. The 1912 data (Denis 1912) is the only comprehensive account ofCanadian systems (see Table 3.2) but there is still some missing data. The 1951 and1961 data are more incomplete with respect to missing data, both within the surveyedcommunities and for communities less than 1,000 in population. For these three datasets, where population data is missing, appropriate census information is used. Wheresystem length data is missing in a surveyed community, the length is approximated byapplying the average population density per unit of water main from communities ofsimilar size (ie in the same population interval) to the population of the community.For the 1951 and 1961 data sets, the population in the communities under 1,000in population is approximated by multiplying the number of communities by an assumed287average population of 500; the pipe length is approximated by dividing the resultingpopulation by the average population density per unit length of main for communitiesunder 2500 in population for each respective year.Table A.2: Canadian Water Main Inventory Data1912 1951 1961 1986Pop. Pipe Pop. Pipe Pop. Pipe Pop. PipeSystems with Complete Information 334 335 609 609 770 770 2887 658Systems with Missing Information 11 10 103 103 118 118- 2229Systems with Missing Information -- 753 753 990 990--(pop. < 1000)Total Systems 346 346 1465 1465 1878 1878 2887 2887Source benis 1912 MU 1951 MU 1961 Ab. Mun. Stats. 1986B.C. Mun. Stats. 1986N.S. Mun. Stats. 1986FACE 1987Misc. Annual ReportsFor the 1986 estimate of main length in Canada, population densities per unitlength of pipe are calculated for the same population intervals as are used in theNational Inventory of Municipal Waterworks and Wastewater Systems in Canada 1986(FACE 1987) using five main sources: the 1986 Alberta Municipal Statistics (AlbertaMunicipal Affairs 1986), the 1986 B.C. Municipal Statistics (B.C. Municipal Affairs1986), the 1986 Nova Scotia Municipal Statistics, the FCM (1984) survey data, andannual reports of Winnipeg and Montreal (Winnipeg 1984, Montreal 1984). The tenpopulation intervals for which average population per unit length of pipe are calculated288include:1) greater than 1,000,0002) 500,001-1,000,0003) 250,001-500,0004)100,001-250,0005) 30,001-100,0006) 10,001-30,0007) 5,001-10,0008) 2,501-5,0009)1,001-2,50010) less than or equal to 1,000For each interval, an average population density per unit length of pipe is calculated andapplied to the “population served” information provided by the national inventory (FACE1987) to get total provincial and national values of main length.The five sources represent 8,672,603 persons served, or 40 percent of the 21.4million Canadians served by water distribution systems in Canada (FACE 1987). Withsuch a significant sample size, the data is assumed to be representative of the remainingsystems in Canada.To calculate the most recent estimate of both the current total length of systemsin Canada and the current population served, the national values estimated for 1986 areextrapolated linearly to 1991 using average growth values from 1986 to 1988 which areavailable for B.C. and Alberta, average growth values from 1977 to 1986 which areavailable for Nova Scotia, and some judgement on the part of the author for the otherprovinces.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0094724/manifest

Comment

Related Items