Appropriate Sanitation Alternatives /f ;L A Technical and Economic Appraisal John M. Kalbermatten * DeAnne S. Julius * Charles G. Gunnerson H Kb World Bank Studies in Water Supply and Sanitation 1 Appropriate Sanitation Alternatives A Technical and Economic Appraisal WORLID BANK STUDIES IN WATER SUPPLY AND SANITATION 1 A i I~ ~~~~~~~~~~~~~~~~~ I Appropriate Sanitation Alternatives A Technical and Economic Appraisal John M. Kalbermatten, DeAnne S. Julius, and Charles G. Gunnerson Published for The World Bank The Johns Hopkins University Press Baltimore and London Copyright ©C 1982 by the International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America The Johns Hopkins University Press Baltimore, Maryland 21218, U.S.A. The views and interpretations in this book are the authors' and should not be attributed to the World Bank, to its affiliated organizations, or to any individual acting in their behalf. EDITOR James McEuen PRODUCTION Virginia deHaven Hitchcock FIGURES Pensri Kimpitak BOOK DESIGN Brian J. Svikhart COVER DESIGN George Parakamannil Library of Congress Cataloging in Publication Data Kalbermatten, John M. Appropriate sanitation alternatives. (World Bank studies in water supply and sanitation; 1) Bibliography: p. Includes index. 1. Underdeveloped areas-Sanitary engineering. l. Julius, DeAnne S., 1949- . II. Gunnerson, Charles G. III. Title. IV. Series. TD353.K26 628.1'09172'4 80-8963 ISBN 0-8018-2578-4 (v. 1: pbk.) Contents Foreword vii Part Two. Program Planning and Development Preface ix 6. Implementation of Appropriate Sa,iitation Acronyms and Abbreviations xii Technology 85 Obstacles 85 1. An Overview 3 Incentives 86 Water Supply and Sanitation in Developing Countries 3 7. Institutional Requirements 89 An Operational Definition of Appropriate Essential Components 89 Technology 5 Policy Implementation 90 Organizational Issues 90 Part One. Analysis of Field Study Results Division of Responsibihties 91 2. Technical and Environmental Assessment 11 8. Community Participation and Organization Household Sanitation Systems 11 93 Community Sanitation Systems 16 Objectives 93 Factors Affecting Choice of Scope 93 Technology 18 Implementation 94 Comparison of Technology 42 Linkage of the Institution and Treatment Alternatives 42 Community 95 Sullage Disposal 43 Resource Recovery 44 9. Project Development 97 Example of Management Schemes for Types of Sanitation Projects 97 Sewerage and Night Soil 46 Technology Selection 99 Future Resource Needs 47 Sanitation Sequences 101 Conclusion 47 Sample Staged Solutions 105 3. The Economic Comparison 50 10. A Concluding Note 108 Economic Costing in Theory 51 Special Problems of Sanitation Bibliography 110 Projects 52 Index 113 Field Results 54 Controlled Comparisons 59 Benefits from Reuse 61 Figures Financial Implications 63 1-1. Recommended Structure of Feasibility Conclusion 65 Studies for Sanitation Program Planning 6 4. Public Health Aspects 67 2-1. Generic Classification of Sanitation Water and Health 67 Systems 14 Excreted Infections 68 2-2. Special Plan of Two Low-income Urban Environmental Classification of Excreted Residential Neighborhoods 38 Infections 70 2-3. Typical Floor Plan of Low-income Rental Health Effects of Treatment and Units 40 Reclamation 73 2-4. Schematic of Relations between Levels of Conclusion 75 Water Service and Options for 5. Sociocultural Factors 77 Sanitation 41 Survey Results 78 2-5. Schematic of Beltsville Agricultural Behavioral Science and Sanitation Research Center (BARC) System for High- Project Design 79 rate Thermophilic Composting 46 Conclusion 82 3-1. Benefit-Cost Divergence over Time 52 v vi 4-1. Influence of Time and Temperature on 18. Known Geographical Distribution of Taenia Selected Pathogens in Night Soil and solium 36 Sludge 74 19. Known Geographical Distribution of Culex 5-1. The Sociocultural Dimension of Sanitation pipiens 37 Project Design: Contributions of Social 20. Sewerage and Night Soil Collection Areas in Science 80 the City of Kyoto, Japan 48 9-1. First-stage Algorithm for Selection of Sanitation Technology 100 Tables 9-2. Second-stage Algorithm for Selection of 2-1. Water Service Levels for Selected Study Sanitation Technology 102 .. 9-3. Third-stage Algorithm for Selection of Communities 12 ^ . . , ^,10 2-2. Sanitation Systems for Selected Study Sanitation Technotogy 03 Communities 13 9-4. Potential Sanitation Sequences 104 2-3. Descriptive Comparison of Sanitation 9-5. Sample Sanitation Sequences 105 Technologies 44 Maps 3-1. Summary of Total Annual Cost per Household (TACH) for Sanitation 1. Average January Temperature 19 Technologies 55 2. Average July Temperature 20 3-2. Selected Input Cost and Conversion Factors 3. Generalized Annual Global for Sanitation Technologies 56 Precipitation 21 3-3. Average Annual Investment and Recurrent 4. Temperature and Precipitation, Central and Cost per Household for Sanitation S-outh America 22Tehogis5 5. Temperature and Precipitation, Asia 23 Technologies 57 3-4. Percentage Investment and Recurrent Cost 6. Temperature and Precipitation, Africa 24 of Community Sanitation Svstems 57 7. Aridity Index for Asia 25 3-5. Average Annual On-site, Collection, and 8. Aridity Index for North and South Treatment Costs per Household 58 America 26 3-6 Annual Sewerage Costs per Household 59 9. Aridity Index for Africa and Australia 27 3-7. Annual Vacuum Cartage Costs per 10. Potential Productivity by Annual Carbon Household 59 Fixation 28 3-8. Comparative TACH5 of Sewerage and 11. Generalized Distribution of Soil Types and Vacuum-truck Cartage in Malacca, Processes 29 Malaysia 60 12. Trend of the Global Spread of Cholera, 3-9. Comparative TACHS of On-site Systems and 1 61-75 30 Sewerage in Gaborone, Botswana 61 13. Known Geographical Distribution of 3-10. Net Cost of Household Biogas Unit, Island Schistosoma haematobium and S. of Taiwan 62 japonicum 31 3-11. Financial Requirements for Investment and 14. Known Geographical Distribution of Recurrent Cost per Household 64 Schistosoma mansoni 32 15. Known Geographical Distribution of 4-1. Excreted Infections 71 Ancylostoma duodenale 33 4-2. Environmental Classification of Excreted 16. Known Geographical Distribution of Necator Infections 72 americanus 34 4-3. Potential Health Improvements 76 17. Known Geographical Distribution of Taenia 7-1. Institutional Responsibilities in saginata 35 Sanitation 92 Foreword DESPITE THE IMPRESSIVE LEVEL of economic growth fits were developed. Emphasis was also given to the the developing countries as a whole have achieved over effects of water service levels upon waste disposal the past quarter century, most of the people in these options and, where applicable, to opportunities for countries do not have a safe water supply or even recovering some of the costs by physically recycling rudimentary sanitation. Immediate investment costs the water and fertilizer components of the wastes. for providing these services at the standards which This is the first of a series of volumes that document prevail in developed countries are estimated at over the Bank's research findings. Based on case studies in US$800,000 million. Corresponding operating costs thirty-nine communities around the world, it presents are projected at another US$10,000 million per year. to planning officials and senior policy advisers a tech- These amounts vastly exceed the resources available nical and economic assessment of the many sanitation for the sector. To help address this problem, a two- options that are available and appropriate to condi- year research project to develop more appropriate tions in developing countries. Other volumes in the (that is, lower-cost) technologies for water supply and series include a planning and design manual for proj- waste disposal was undertaken by the World Bank in ect engineers, analysts, and technicians and a compila- 1976-78. Meanwhile, the member countries of the tion and synthesis of health and disease factors impor- United Nations have declared the 1980s to be the tant in sanitation system planning and implementa- International Drinking Water Supply and Sanitation tion. Their publication is particularly timely at the Decade, with the objective of satisfying for all popula- beginning of this decade. If the twin objectives of tions of the globe two of the most basic human economic growth and the eradication of absolute pov- needs-clean water and the sanitary disposal of hu- erty are to be met, the nations of the world must man wastes. ensure that everyone has access to safe water and The Bank's research revealed the technological, eco- adequate sanitation. It is to universal access to the nomic, environmental, and institutional interdepen- latter that this volume is dedicated. dence of water supply, sanitation, and health. Waste disposal technologies costing as little as one-tenth the WARREN C. BAUM amount of conventional sewerage were identified. Vice President, Central Projects Staff Means to ensure high health and environmental bene- The World Bank vii I I Preface OVER THE PAST DECADE, the focus of development tions entitled "World Bank Studies in Water Supply planners on economic growth has broadened to include and Sanitation," of which this appraisal is number 1. a parallel concern with the distribution of the benefits Other publications in this collection include: made possible by that growth. In his president's address John M. Kalbermatten and others, Appropriaie to the Board of Governors of the World Bank at its 1980 Sanitation Alternatives: A Planning and Design annual meeting, Robert S. McNamara reiterated that, Manual, World Bank Studies in Water Supply and to achieve the twin objectives of economic growth and Sanitation, no. 2 the eradication of absolute poverty, countries must do two basic things: assist the poor to increase their Richard G. Feachem and others, Sanitation and productivity and ensure their access to essential public Disease: Health Aspects of Exereta and Wastewater services. Management, World Bank Studies in Water Supply Among these essential public services are water and Sanitation, no. 3 supply and waste disposal. Few other services contrib- ute as much to an improvement in health and living Further publications in the collection will be issued as standards as does the provision of an adequate supply of ongoing research is completed. In addition, the Trans- safe water and the means for sanitary disposal of waste. portation, Water, and Telecommunications Depart- It has become apparent, however, that development ment (TWT) of the World Bank maintains a series of projects must be specifically designed to reach the reports- under the main title Appropriate Technology urban and rural poor if the poor are to be provided with for Water Supply and Sanitation-available from the services that they can afford and that meet their needs. Bank's Publications Unit. Subtitles of volumes in this In particular, sewerage-the conventional method of series are as follows: human waste disposal in the developed countries-re- vol. 1: Technical and Economic Options, by John M. quires massive investments of both foreign and local Kalbermatten, DeAnne S. Julius, and Charles capital that generally are not available in the develop- G. Gunnerson ing nations. Where sewerage systems have been built, they often have required charges to the consumers that . 1aS were beyond consumers' ability to pay. In acknowledg- tions ment of the limitations of traditional solutions, the vol. 2: A Planner's Guide, by John M. Kalbermat- World Bank in 1976 launched a two-year research ten, DeAnne S. Julius, Charles G. Gunnerson, project entitled "Appropriate Technology for Water and D. Duncan Mara Supply and Waste Disposal in Developing Countries." vol. 3: Health Aspects of Excreta and Sullage The objective of the project was to identify and evaluate Management-A State-of-the-Art Review, alternative sanitation technologies for their potential to by Richard G. Feachem, David J. Bradley, meet the needs and match the resources of project Hemda Garelick, and D. Duncan Mara beneficiaries. To accomplish this, the health, social, institutional, vol. 4: Low-Cost Technology Options for Sanita- as well as the technical and economic aspects of the tion-A State-of-the-Art Review and Anno- various technologies had to be considered. The overall tated Bibliography, by Witold Rybczynski, project has consequently generated a variety of subsid- Chongrak Polprasert, and Michael McGarry iary research by specialists in different disciplines, and [available, as a joint World Bank/Interna- its findings are being issued in a collection of publica- tional Development Research Centre publica- Ix x tion, from the IDRC, Ottawa, Ontario, Can- and many variations of each type were observed. adal Improved designs were prepared for several of these, vol. 5: Sociocultural Aspects of Water Supply and and for only one technology (bucket latrines) was it Excreta Disposal, by Mary Elmendorf and concluded that introduction of a system to new sites Patricia Buckles should be avoided. Two of the other technologies, vol. 6: Country Studies in Sanitation Alternatives, aquaprivies and communal facilities, were found to by Richard A. Kuhlthau (ed.) have limited applicability because of social factors. All vol. 7: Alternative Sanitation Technologies for Ur- of the remaining technologies (improved pit latrines, ban Areas in Africa, By Richard G. Feachem, pour-flush (PF) toilets, composting toilets, modified D. Duncan Mara, and Kenneth 0. Iwugo septic tanks, vault and cartage, small-bore sewerage, vol. 8: Seven Case Studies of Rural and Urban and conventional sewerage) can be recommended (sub- Fringe Areas in Latin America, by Mary ject to the physical conditions of the site and the social Elmendorf (ed.) preferences and economic resources of the beneficia- vol. 9: Low-Cost Design of Water Distribution Sys- ries) for adoption. Except in unusual circumstances, tems, by Donald T. Lauria, Peter J. Kolsky, both scattered, rural and densely populated, urban and Richard N. Middleton communities should find themselves with two or more vol. I ONight-soil Composting, by Hillel I. Shuval, technically feasible options, each with a range of design Charles G. Gunnerson, and DeAnne S. Julius alternatives. vol. 11 A Sanitation Field Manual, by John M. Another important technical contribution of this Kalbermatten, DeAnne S. Julius, and Charles research has been the design of "sanitation sequences": G. Gunnerson step-by-step improvements leading from one technol- vol. 12Eow-Cost Water Distribution-A Field Man- ogy to another and designed to minimize costs over the ual, by Charles Spangler. entire sequence. These enable a community initially to select one of the low-cost technologies in the knowledge The main purpose of this appraisal is to summarize that, as the community's socioeconomic status im- the technical, economic, health, and social findings of proves, the technology can be upgraded without the research, and to discuss the aspects of program wasting the initial investment. It is noteworthy that planning necessary to begin implementation of the none of the sanitation sequences has conventional findings. It is, therefore, directed primarily toward sewerage as its final improvement. In urban areas, the planning officials and advisors for sector policy in and final upgrading is generally to a low-volume flush toilet for developing countries. Although the focus is primar- connected to a vault that overflows into a small- ily on sanitation options (because water supply technol- diameter sewer. From the user's viewpoint, the only ogy is better known and understood), information on difference between this system and conventional levels of water service is included because water use is a sewerage is the size of the toilet tank in the bathroom. determining factor in waste disposal. Technical details Yet the cost savings, particularly if installation and designs are presented in Appropriate Sanitation proceeds in stages, are significant. Over a thirty-year Alternatives: A Planning and Design Manual, forth- period the total present value cost of this alternative coming as a complement to this volume. system is about one-half that of conventional sewerage. The study presents the results of two years of field Its environmental cost is also likely to be significantly studies undertaken by the World Bank in thirty-nine lower than that of sewerage because its eventual waste communities in fourteen countries of the developing discharge is less. As this example indicates, the critical world. A bibliographic search was initially conducted element in an economic solution to providing sanitation for information on nonconventional options, but only is the reduction of nonessential water use. (Toward this about 1 percent of the published technical literature on end, work is now underway to adapt water-saving wastewater related to technologies other than sewer- showerheads and other appliances for use in develop- age. Case studies carried out by local engineers, ing countries.) A corollary to this conclusion is that economists, and behavioral scientists thus formed the schemes for water supply, with or without sanitation, backbone of the research. should explicitly consider, in the design and cost The first and most important finding of the case comparison of alternative service levels, the require- studies was that there are many technologies between ment to dispose of the wastewater created. the unimproved pit privy and conventional sewerage In addition to these technical findings, the research that can be recommended for replication on a wide has produced a new and promising approach to the scale. In all, five types of household (on-site) systems problem of linking potential health benefits to improve- and four types of community systems were identified, ments in environmental sanitation. Rather than xi tackling yet again the intractable problem of quantify- portant for all technologies and never accounted for less ing health benefits, this research focused directly on the than 45 percent of total household costs on an annual transmission process of excreta-related disease and the basis; the latter was most important for sewerage and relation of that process to the various sanitation septic tank systems. Where the economic cost of water alternatives. The Ross Institute of Tropical Hygiene of is high, the payoff from designing systems with low the London School of Hygiene and Tropical Medicine requirements for flushing water is large. was contracted as part of this study to develop an The findings and recommendations of this appraisal environmental classification of excreta-related infec- are based on surveys of relevant literature (see tions that, together with a basic understanding of the Feachem and others, Sanitation and Disease, and factors important in disease transmission, would enable Rybczynski, Polprasert, and McGarry, Low-Cost the planner and engineer to maximize the health Technology Options), an evaluation of sociocultural benefits of whatever technology is chosen. The means of factors (see Elmendorf and Buckles, Sociocultural so doing include both the incorporation of specific Aspects of Water Supply), detailed field studies (see design features and the supplementation of "hardware" Kuhlthau, Country Studies, and Lauria, Kolsky, and with precisely directed educational campaigns for Middleton, Low-Cost Design), and the personal users. [In the process of developing the environmental observations, experience, and advice of colleagues in classification, Dr. Richard G. Feachem and others at the World Bank and other institutions. Because the list the Ross Institute conducted a comprehensive review of of contributors is so large, only a few can be mentioned. relevant literature and have produced a unique We wish to acknowledge, in particular, the support reference work on the subject (Sanitation and Disease, given to this project by Yves Rovani, director of the Part One) as well as an original and insightful analysis Energy, Water, and Telecommunications Department of the relation between sanitation and excreta-related at the time the research was done and currently director infections (Sanitation and Disease, Part Two and the of the Bank's Energy Department, and the valuable references therein).] review and direction provided by Kim Jaycox, chair- That alternative technologies exist and that they can man of the Project Steering Committee. Advice and be designed to maximize health benefits still leaves the expertise in particular areas were freely provided by planner with two important questions: what do the Jerry Warford and Harold Shipman, two of the early technologies cost, and what complementary inputs do supporters of the project, and by William Cosgrove, Art they require for successful implementation? The costs Bruestle, Fred Hotes, Johannes Linn, Ragnar Overby, collected from a statistically based, case-study ap- John Courtney, and Charles Weiss. In addition, David proach such as this one are bound to vary widely from Bradley and Richard Feachem of the Ross Institute of one community to another even for identical technolo- Tropical Hygiene, D. Duncan Mara of the University gies. When the problem is compounded by the fact that of Leeds, Gilbert and Anne White of the University of no two observed pit privies, for example, were exactly Colorado, and Mike McGarry of the International alike in their design or construction, it is obvious that Development Research Centre helped us considerably precise measures of cost sensitivities cannot be in shaping our approach to the health and social aspects obtained. Yet, by applying a consistent methodology to of the study and in developing the algorithm for all case-study calculations and by deriving economic technology selection. rather than financial costs, a broad cost comparison of Special thanks are due to the field consultants whose different technologies was possible. tireless efforts to obtain and evaluate information The nine technologies studied formed three distinct under diverse, and sometimes difficult, conditions cost groupings. Five of them cost less than $100 made possible our empirical analysis. Their individual annually per household (including both capital and contributions are acknowledged in the Appropriate recurrent costs); two technologies cost between $150 Technology publications for which they were responsi- and $200 annually per household; and two (septic tanks ble, but we would like to extend our particular thanks to and sewerage) cost more than $300 annually per Kenneth Iwugo, who was responsible for the case household. (All cost figures in the study refer to 1978 studies of Nigeria, Ghana, and Zambia; Ng Kin Seng, U.S. dollars.) The ratio of lowest to highest cost of the who did the work on Malaysia and the island of Taiwan; systems was 1:20. S. S. Soesanto for the excellent work of her team in The two most important influences on total house- Indonesia; Dong Min Kim for his Korean study; Mary hold costs were factors that have often been ignored in Elmendorf and Chuck Pineo for their work in Nicara- engineering analysis: on-site household costs (for ex- gua; Samir El Daher and Beshir Mohammed El Has- ample, internal plumbing) and the cost of flushing san, who undertook the Sudan study; Shohei Sata and water for water-carried systems. The former was im- Katsuyoshi Tomono of Nihon Suido Consultants for xii their important work on the Japanese cities; Raphael in coordinating the graphics and other aspects of this Rodriguez, who undertook the work on Colombia; publication with those of the other volumes in the and Mike Blackmore and his team in Botswana. In Studies of Water Supply and Sanitation series is greatly addition, Mei-Chan Lo and Robert T. C. Lee, both appreciated. of Taiwan, were instrumental in helping us evaluate Finally, we owe a special thanks to our spouses- the potential for wider replication of the interesting Nelly Kalbermatten, Ian Harvey, and Betty Gun- intermediate technologies for sanitation that were nerson-who endured the extra travel and long hours studied there. Harvey Ludwig and Saul Arlosoroff that this research project entailed. assisted us in the evaluation of various technologies. This book could not have been produced without JOHN M. KALBERMATTEN the dedication and cooperation of the secretarial staff: DEANNE S. JULIUS Margaret Koilpillai, Julia Ben Ezra, and Susan Pur- CHARLES G. GUNNERSON cell. David Dalmat and Sylvie Brebion's efficiency Acronyms and Abbreviations AIC Average incremental cost PF Pour-flush (toilet) BARC Beltsville Agricultural Research Center (U.S. PVC Polyvinyl chloride Department of Agriculture, Beltsville, ROEC Reed Odorless Earth Closet Maryland, U.S.A.) TACH Total annual cost per household BOD Biochemical oxygen demand VIDP Ventilated improved double-pit (latrine) BOD5 Five-day BOD (by the standard test) VIP Ventilated improved pit (latrine) DVC Double-vault composting (toilet) Appropriate Sanitation Alternatives A Technical and Economic Appraisal I I 1 An Overview A CONVENIENT SUPPLY of safe water and the sanitary ing countries have adequate sanitation services; that is, disposal of human wastes are essential ingredients of about 630 million out of 1.7 billion people.' Population a healthy, productive life. Water that is not safe for growth will add to this figure in the 1980s another 700 human consumption can spread disease; water that million people who will have to be provided with some is not conveniently located can reduce the productive means of sanitation if the goal of the International time and energy of the water carrier; and inadequate Drinking Water Supply and Sanitation Decade-ade- facilities for excreta disposal reduce the potential quate water supply and sanitation for all people-is to benefits of a safe water supply by transmitting path- be achieved. A similar number of people, about 2 ogens from infected to healthy persons. Over fifty billion, will require water supply by the same date. infections can be transferred from a diseased person Thus, approximately half a million people daily, from to a healthy one by various direct or indirect routes now until 1990, will need to be provided with water and involving excreta. sanitation services. One of the fundamental problems in meeting this Water Supply and Sanitation goal is the high cost of conventional sanitation services. in Developing Countries General estimates based on 1978 per capita costs indicate that up to $60 billion would be required to Coupled with malnutrition, these excreta-related provide water supply for everyone, and from $300 to diseases take a dredfutllnev$600 billion would be needed for sewerage.2 Per capita especially among children. For example, in one Middle investment costs for the latter range from $150 to $650, eastercilyaountry chald. Fof thexpldre,n bornalve die an amount totally beyond the ability of the beneficia- Eastern country half of the children born alive die t before reaching the age of five as a result of the ries to pay. combine effects of disease andmalnutrition. In con- In industrialized countries, the standard solution for combined effects of disease and malnutrition. In con- th trast, only 2 percent of the children born in the United sanitary disposal of human excreta is waterborne Kingdom die before reaching their fifth birthday. It is sewerage. Users and responsible officials have come to invariably the poor who suffer the most from the view the flush toilet as the absolutely essential part of absence of safe water and sanitation because they lack an adequate solution to the problem of excreta disposal. not only the means to provide for such facilities but also This technology, however, was designed to maximize the information on how to minimize the ill effects of the user convenience rather than health benefits. Conve- theaninorm aondon wtini the il. l reffets the nience may be an important objective in developed debilitating effects of endemic disease lower the pro- countries, but it has a lower priority in most developing deiting peffect of enderm disease lower the pro countries. In fact, conventional sewerage has been the such loss of productivity. result of a slow progress made over decades, even centuries, from the pit latrine to the flush toilet, and the Dimensions of the problem present standard of convenience has been achieved at substantial economic and environmental costs. To understand the magnitude of the problem, one The problem facing developing countries is a only need consult the data collected by the World familiar one: high expectations coupled with limited Health Organization (WHO) in preparation for the resources. Decisionmakers are asked to achieve the United Nations Water Conference (Mar del Plata, standards of convenience observed in industrialized Argentina, Spring 1977). These rough estimates show countries, but-given the backlog in service, the that only about one-third of the population in develop- massive size of sewerage investments, and the demands 3 4 OVERVIEW on financial resources by other sectors-they do not they have to press for sewerage because without it have the funds to realize this goal. Sewerage could be public health will not be secure. Few sewer systems in provided for a few, but at the expense of the vast developing countries are well maintained. Sewage majority of the population. As a consequence, many treatment works commonly discharge effluents in a developing countries have taken no steps at all toward condition little better (and in some cases worse) than improving sanitation. The very magnitude of the task the incoming sewage. In any case, current plant design has effectively discouraged action. concentrates on undoing the environmental problems At the present time the first priority of excreta waterborne collection has created rather than on health disposal programs in developing countries must be maintenance through pathogen removal. There is, human health; that is, the reduction and eventual therefore, little realistic basis for the commonly held elimination of the transmission of excreta-related dis- view that Western sanitation techniques are the appro- eases. This health objective can be fully achieved by priate solution for developing countries. Rather, re- nonconventional sanitation technologies that are much education of engineers to design for maximum health cheaper than sewerage. The goals for the International benefits, and to consider the whole range of available Drinking Water Supply and Sanitation Decade of the technologies, is essential. 1 980s intentionally do not specify sewerage, but call for Most municipalities in developing countries have the sanitary disposal of excreta-leaving the chosen difficulty in attracting and retaining well-trained staff, disposal method to the discretion of individual govern- and in consequence municipal services suffer. The ments. Similarly, objectives of the Decade include an potential for self-help in conventional sewerage is, adequate supply of safe water, but do not specify the however, minimal. The adoption of low-cost technolo- methods to be used to achieve the goal. The challenge of gies can capture the strong desire of most people to providing as many people as possible with the required improve their living conditions, and this motivation can facilities is to find techniques to achieve these objec- be put to good use. But this implies a change in the role tives with the resources available. of the municipality; it must become an active promoter and educator because experience makes it abundantly The constraints clear that technologies imposed on people without adequate consultation are likely to fail or go unused. The principal constraints to the successful provision of sanitation facilities in developing countries are lack A glimpse of a solution of funds, lack of knowledge about nonconventional sanitation technologies, and weak institutions with few Given these constraints, it is not surprising that levels trained personnel. There is no foreseeable way that of sanitation service in developing countries have re- waterborne waste disposal, with an average investment mained low. A major effort is needed to identify and cost of around $300 per person, can be made affordable develop alternative technologies for sanitation that are in countries in which annual per capita income averages appropriate to the conditions in developing countries less than that amount. In addition, and implicit in the and are designed to meet health requirements at a cost decision to provide sewerage, is a decision to provide a affordable to the user. Clearly, the solutions also must water connection to each house. About 40 percent of the reflect the communities' preferences. water from this connection will be used for no essential The identification and design of appropriate excreta purpose but to flush away wastes. Even in the unlikely disposal systems does not require the invention of new event of the developed countries' providing massive processes or devices. Rather, it calls initially for a grant funds towards the initial cost of installing sewer- review of the historical development of the present age in the developing world, the costs of operating and technology, a reexamination of the decisions leading to maintaining sewer systems and of satisfactorily dispos- sewerage, and the design of improvements to eliminate ing of the sewage would be prohibitive. Lower-cost problems that caused the abandonment of earlier, low- solutions have to be found for the majority of people. cost solutions. An examination of how sewerage came The lack of interest in sanitation technologies other about reveals three facts clearly. First, waste disposal than sewerage is in part because of the standardized went through many stages before sewerage. Second, education of most planners and engineers in developing existing systems were improved and new solutions countries. Engineers are trained in sophisticated (and invented whenever the old solutions were no longer intellectually stimulating) advanced technology that is, satisfactory. Third, improvements have been imple- in a sense, self-perpetuating: sewer systems lead to high mented over a long period of time and at substantial water consumption and the attendant problems of cost. Sewerage was not a grand design achieved in one source development and effluent disposal. Planners feel giant step but is the end result of progressively sophisti- OVERVIEW 5 cated solutions. It took industrialized countries over a proportions"4 or "induced bias."5 Schumacher's book hundred years to achieve their present status in a close served to bring some of the basic ideas into public view. matching of needs and the economic capacity to take There is no concise and universally correct definition care of them. With the benefit of hindsight, it should be of technological appropriateness; the concept is a rela- possible to correct not only some of the shortcomings of tive one, which can only be applied within a particular previous solutions, but also to develop a sanitation context. The standards for determining the appropri- system that can be further improved, step by step, to ateness of technology are related to the developmental meet both the user's requirements and economic capac- goals of the country making the choice and to the ity to pay for improvements. circumstances of the technology's use. Various rules of What is needed is a sequence of sanitation improve- thumb have been suggested that call for low capital ments, designed from the outset to provide maximum intensity, simplicity of operation, use of indigenous health benefits while minimizing costs over the long resources, and so forth, but they are of limited use in run. If sanitation facilities are to be used, each step of comparing diverse technologies, which would rank the sequence must consider consumers' preferences and differently according to each of these criteria, and they customs of personal hygiene. In fact, sequenced sanita- are not universally appropriate for all developing coun- tion is likely to be more successful than the immediate tries. installation of sewers has been because it allows the The operational definition used in this study is really user to progress as he sees fit, to whatever level of an abbreviated description of the process of determin- convenience he desires and at his own speed, in a ing which technology is appropriate in a particular reflection of his personal preferences and capacity to case. An appropriate technology is defined as a method install, operate, and pay for the facility. or technique that provides a socially and environmen- Fortunately, low-cost alternatives to sewerage exist tally acceptable level of service or quality of product and work well. When properly constructed and main- with full health benefits and at the least economic cost.6 tained, they provide all the health benefits of sewerage This "definition" immediately provokes questions. and have fewer adverse environmental effects. They How does one judge social or environmental acceptabil- are, in many cases, technologies that had been used for ity? What is the economic cost of a process? There is an many years in developed countries but were abandoned intuitive understanding of the words themselves, but rather than improved as those countries grew more their application is not straightforward. This study prosperous. They may not be applicable to parts of the looks in detail at the process of identifying appropriate dense, westernized, metropolitan centers of the devel- sanitation technology from the technical, economic, oping world, where sewerage may remain the most health, and social perspectives. The basic philosophy is appropriate technology, but they are ideally suited to that only those technologies that pass all these tests are rural areas, small towns, and metropolitan fringe areas, appropriate. The operational definition incorporates which closely resemble the environment for which they long-run benefits and costs by using life-cycle costing were originally developed. Their failures are usually and by paying particular attention to the technical attributable to poor design, inadequate education of potential for upgrading each alternative as the incomes users, or lack of maintenance-problems that plague and aspirations of the users grow over time. sewerage systems as well but can be overcome in developing countries if increased emphasis and atten- The selection process tion are given to improving health and sanitation. The process of selecting technology begins by identifying all the technological alternatives available for providing the good or service desired (in this case, An Operational Definition sanitation). Within that set of possibilities there will of Appropriate Technology usually be some technologies that can be readily excluded for technical, health, or social reasons. For A large body of literature has developed in recent example, septic tanks requiring large drainage fields years on the choice of appropriate technology, would be technically inappropriate for a site with high particularly in the manufacturing and agricultural population density. Similarly, a composting latrine sectors. The surge of interest in this topic dates from the would be socially inappropriate for people who have publication of E. F. Schumacher's book Small Is strong cultural objections to the sight or handling of Beautiful in 1973.3 Before 1973, the theory of techno- excreta. Some technologies may require institutional logical choice was written about mostly by economists support that is infeasible in the given social environ- and was concealed in technical jargon such as "factor ment. Once these exclusions have been made, the range Figure 1-1. Recommended Structure of Feasibility Studies for Sanitation Program Planning Sanitary Engineer and Economist Behavioral Scientist Community Public Health Specialist Examine physical and Collects Consults with communtAdvises on environimental coniditionis mareooncto collect informationprciean Stage 1 and establish community nacoecrmionmnd prefnces practices anc health profile onformaion an practices preferenzces I ndentify and cost Identifies economic Linsts sutoaland Stage tehnically and medicallv institutioially faible alternatives cosrit n iisfeasible, alternattves Prepares short li5t Identifies community's Stage 3 of feasible alternatives e contribution and level - Advises of feasible alternatives of affordability Prepare final design Agrees otn typical layouts Stage 4 and estimate unit cost and local community Advises of feasible alternatives \ / participation Sta e S \>| ~~~~~~~~~~Prepare.s financialt co'sling|J Stage offeasble alternative sy.stern.s\ Crommunity selects Stage 6 preferred alternative OVERVIEW 7 of technically and socially feasible alternatives that estimates for those technologies that have passed tech- provide full health benefits remains. For these technol- nical and social tests and selects the least-cost alterna- ogies, cost estimates are prepared that consider their tive for each. As the fourth step, the engineer prepares real tesource cost to the economy. As described in final designs and unit costs for these choices. At this chapter 3, this may involve adjustments in market point the social information collected in stage I should prices to counteract economic distortions or to reflect be used to determine, for example, the siting of the developmental goals such as the creation of employ- latrine on the plot, the size of the superstructure, the ment.' Least-cost solutions for each technology are materials to be used for the seat or slab, and other determined. On the basis of these economic costs and details whose technical and economic import may be discussions with government planners, financial costs low but which make a major difference in the way the are prepared for all least-cost solutions. Those alterna- technology will be accepted and used in the community. tives clearly outside the bounds of affordability for The final designs should also incorporate features consumers are excluded. Because the benefits of vari- necessary to maximize the health benefits expected ous sanitation technologies cannot be quantified, it is from each technology. Final designs are used in the impossible for the economist to do more than exclude fifth stage to determine financial costs (based on the various alternatives. A single least-cost alternative availability of national and municipal funding), includ- cannot usually be recommended because there is no ing how much the user will have to pay for construction way of quantitatively comparing benefits. The final and maintenance of each design. Any technology whose step in identifying appropriate sanitation technology total financial cost is more than 10-20 percent of user must rest with the eventual beneficiaries. Those alter- income probably should be excluded as financially natives that have survived technical, health, social, and unaffordable. The final step is for the behavioral economic tests are presented to the community with scientist to present and explain the alternatives and their corresponding financial price tags, and the users their costs to the community for final selection. must decide for which level of service they are willing to pay.8 Comparison with the traditional approach How the technical, health, social, and economic aspects of technological choice are actually coordinated The process above contrasts with that of the typical is shown in figure 1- 1, although the stages in the figure feasibility study. The conventional team conducting should not be interpreted too literally. A technology feasibility studies is heavily weighted with engineers. It may fail technically if the users' social preferences may contain a financial analyst, but rarely an econo- militate against its proper maintenance. The economic mist, and almost never a behavioral scientist. The cost of a system is heavily dependent upon social factors alternative technologies considered are usually only a such as labor productivity as well as upon technical small subset of the group discussed in this report, and in parameters. Because of these relations between the many cases the terms of reference of the study (written various boxes in figure 1-1, there must be a close by other engineers) limit the selection to waterborne working association among the different actors in the sewerage with several collection configurations and planning process. treatment alternatives. Thus the selection process de- For simplicity, it is assumed that separate individuals scribed in figure 1-1 is short-circuited and moves or groups are responsible for each part, although in directly through technical criteria to final design. The practice responsibilities may overlap. In stage 1 of conventional study team then prepares estimates of figure 1-1, each specialist collects the information financial costs and writes its report. necessary to make his respective test for exclusion. For The main problems of this customary process are the engineer, health specialist, and behavioral scientist, obvious from a comparison with the methods set forth this data collection will usually take place in the in this study. In the conventional procedure, the most community to be served. The economist will talk with appropriate technologies may never get considered. No both government and municipal officials to obtain the checks are made to ensure that the technical solution information necessary to calculate shadow rates and to designed and costed is socially acceptable. By excluding determine the availability of grant funds or other a meaningful economic comparison, the usual method means of subsidy. The engineer, health specialist, and makes no guarantee that the solution offered is the one behavioral scientist will then apply the information of least cost for the economy. The decisionmakers are they have collected to arrive at preliminary lists of presented at the end with a proposal that has not taken technically, medically, and socially feasible alterna- into account their own economic priorities or the ability tives. In the third stage the economist prepares cost to pay of their constituents, the ultimate beneficiaries.9 8 OVERVIEW The framework suggested in this report for the 5. C. Kennedy, "Induced Bias in Innovation," EconomicJournal, identification of appropriate technologies is probably vol. 74 (September 1964), pp. 541-47. more time intensive than that of traditional feasibility 6. A more rigorous definition would be the technology for which analysis. It also requires the recruitment of additional the net present value of the stream of health and environmental personnel. Thus, a clear case must be made for its benefits, subject to a constraint on social acceptability, is maximized. superiority in choosing appropriate technologies, and it The difficulty of quantifying health and environmental benefits, must be shown that the cost of choosing an inappropri- however, prevents such a definition from being operationally useful. ate technology is sufficiently high to warrant a more 7. An ideal analysis would go beyond economic costing to include costly selection process. The case studies of sanitation income-distributional factors by calculating social costs. Distribu- systems in thirty-nine communities, which form the tional weights, however, cannot be taken into account explicitly in basis of this report, lead us to believe that there is a very this analysis because benefit quantification is not possible. This is not high cost, both in wasted resources and in poorer a significant limitation because the major concern of this study is to community health, associated with the imposition of identify technologies that are specifically appropriate for the rural inappropriate sanitation technologies. Part One of this and urban poor. The study's case studies themselves were chosen to inappopriae saitatin tehnoloies. art ne ofth embody this concern. appraisal presents the detailed findings of the commu- nity studies. 8. Because the consumer is presented with financial rather than economic costs, it is important that economic cost ranking of the technologies be preserved in deriving financial costs. This may Notes to Chapter 1 preclude, for example, full construction grants for all technologies regardless of relative construction costs. I. "Billion" is equivalent to "thousand million." 9. Blame should not necessarily be placed on the consulting firms 2. All cost figures in this study are in 1978 U.S. dollars. See who prepare such conventional studies. Often they are guilty of no chapter 3 for their derivation. more than following current practice in a highly competitive field, 3. E. F. Schumacher, Small Is Beautiful (New York: Harper and and they must work within the constraints of their terms of reference. Row, 1973). A number of firms, in fact, are already implementing some of the 4. R. S. Eckaus, "The Factor Proportions Problem in Undevel- recommendations of this report and routinely use multidisciplinary oped Areas," American Economic Review, vol. 45 (September 1955), teams in their work. The obstacles to the choice and adoption of pp. 539-65. appropriate technologies are discussed in chapter 6. Part One Analysis of Field Study Results Technical and Environmental Assessment INVESTIGATIONS OF SANITATION SYSTEMS on sites in When the pit is about three-fourths full, the superstruc- thirty-nine communities around the world have pro- ture and squatting plate are removed and the pit is filled vided a wealth of practical design and operational data up with soil from a new pit dug nearby. upon which a technical assessment of various sanitation Most of the seven pit latrines evaluated in this study alternatives may be based. Although many variations were of the simple, unimproved variety and conse- of similar systems were observed, this chapter classifies quently had both odor and insect (flies and mosquitoes) all of the technologies studied into five types of house- problems. These undesirable features were almost com- hold systems and four types of community systems. The pletely absent in the ventilated, improved pit (vip) essential features of each are described, the technical latrine and the Reed Odorless Earth Closet (ROEC) requirements and environmental limitations are dis- observed in southern Africa. cussed, and an assessment is made of the potential for upgrading and widely replicating each. The cost and health implications of the technologies are presented in the following two chapters.I For a generic classification VIP latrines of the various sanitation systems, see figure 2-1. A summary of the data on population and service levels In a vIP latrine (figure 2-1, no. 5), the pit is slightly characteristic of some of the communities studied is displaced to make room for an external vent pipe. For shown in tables 2-1 and 2-2. maximum odor control the vent pipe should be at least 150 millimeters in diameter, painted black, and located on the sunny side of the latrine so that the air inside the Household Sanitation Systems pipe will heat up and create an updraft. (Indications are that painting the pipe black may increase the ventilat- Pit latrines, pour-flush (PF) toilets, composting toi- ing effect.) If the vent pipe is letting enough light into lets, aquaprivies, and septic tanks for use in individual the pit, and if the superstructure is fairly dark, flies will homes are the major types of household sanitation try to escape through the vent rather than back into the systems. The distinguishing feature of these, compared superstructure. Covering the vent pipe with a gauze with the community systems discussed in the next screen will prevent flies from escaping through that section, is that they require little or no investment in route and thus minimize the health hazard from these facilities outside individual homesites. insects.2 Where the user prefers a solid superstructure that cannot be moved, or where space is not available for moving a vIP latrine, a modification-a ventilated Pit latrines improved double-pit (VIDP) latrine-can be used. The VIDP latrine contains two pits, which are dug side by By far the most commonly observed technology side and are covered by the same superstructure. Use of around the world, particularly in rural areas, is the pit the two pits alternates, with the squatting plate being latrine (figure 2-1, no. 3). In its most elementary form, moved from the full to the empty pit as necessary. The a pit latrine has three components: the pit, a squatting full pit is emptied not less than twelve months after last plate (or seat and riser), and the superstructure. The pit use to be ready for renewed use when the second pit is is simply a hole in the ground into which excreta fall. full. 11 12 ANALYSIS OF FIELD STUDY RESULTS Table 2-1. Water Service Levels for Selected Study Communities Water service level Population Types of service, Liters per Cubic meters Community Total Per hectare Percent Percent capita daily per hectare daily East Asia 1 (rural) 56.000 5.2 13 (W) 87 (H) 271 1.4 2 (resort) 57,000 4.2 28 (W) 72 (H) 208 0.9 3 (suburban) 103,000 79.0 1 (W) 99 (H) 268 21.1 4 (urban) 1,426,000 24.0 3 (W) 97 (H) 340 8.2 5 (rural) 85,000 37.8 10 (0) 90 (H) 136 5.1 6 (urban) 175,000 26.9 80 (0) 20 (H) 108 2.9 7 (urban) 342,000 25.8 17 (0) 83 (H) 125 3.2 8 (rural) 285 n.a. 12 (W) 88 (H) 1(0 n.a. 9 (rural) 310 n.a. n.a. n.a. 40 n.a. 10 (urban) 141,000 4.2 23 (W) 77 (H) 143 0.6 Southeast Asia 11 (rural) 614 5.2 0 100 (H) n.a. n.a. 12 (urban) 2,000,000 407.0 0 100 (H) 41/67k 16.7/27.3b 13 (urban) 2,000,000 496.0 0 100 (H) 36/61h 17.9/30.2" 14 (urban) 5,000.000 9.0 0 100 (H) 45/60D 0. 40.5b 15 (urban) 5,000.000 582.0 0 100 (H) 31/62b 18.0/36.1I 16 (urban) 90,000 850.0 14 (S) 86 (H) 207 167.0' 17 (urban) 101,000 700.0 34 (S) 66 (H) 160 112.0' 18 (urban) 106,000 800.0 50 (S) 50 (H) 180 144.0c Latin America 19 (rural) 196 n.a. 0 100 (H) n.a. n.a. 20 (rural) 800 n.a. 0 100 (H) 53 n.a. 21 (suburban) 18,000 n.a. n.a. 50 (H) 152 n.a. 22 (urban) 535,000 n.a. n.a. 90 (H) 341 n.a.A 23 (rural) 12,000 30.0 0 100 (H) n.a. n.a.d East and West Africa 24 (urban) 20,000 n.a. n.a. n.a. 177' n.a. 25 (urban) 114,000 n.a. n.a. n.a. n.a. n.a. 26 (urban) 250,000 n.a. n.a. n.a. n.a. n.a. 27 (urban) 500,000 n.a. n.a. n.a. n.a. n.a. 28 (urban) 500,000 n.a. n.a. n.a. n.a. n.a. 29 (urban) 650,000 n.a. 7 (0) 93 (S) 20 n.a. North Africa 30 (rural) 4,000 11.0 n.a. 95 (H) 5(0 0.6 31 (rural) 24,000 12.0 0 100 (S) 60 0.7 32 (urban) 250,000 107.0 0 100 (H) 70 7.5 33 (urban) 350,000 73.0 n.a. 97 (H) 10( 7.3 n.a. Not available. a. Percentages of population served by house connection (H), standpipe (S), vendor (V), well (W), or other service (0). b. Paired values indicate liters per capita daily in dry/rainy season. c. High values are for areas of heavy year-round afternoon rains and runoff. d. Piped supplies available but not used because of water's disagreeable taste and color. e. In more detail: 12 liters per capita daily for standpipe; 74 liters per capita daily for low-cost housing; 122 liters per capita daily for medium-cost housing; 434 liters per capita daily for high-cost housing. ROECs breeding. The chute must therefore be cleaned regu- Another successful variation of the pit latrine is the larly with a long-handled brush or a small amount of ROEC (figure 2-1, no. 4). Its pit is completely displaced water. The advantages of the ROEC over the VIP latrine from the superstructure and connected to the squatting are that its pit can be larger and thus have a longer life plate by a curved chute. A vent pipe is provided, as in (because the superstructure is displaced), the users the VIP latrine, to minimize fly and odor nuisance. A (especially children) have no fear of falling into its pit, disadvantage of the ROEC, however, is that the chute is and it may be more acceptable in some societies easily fouled, thereby providing a possible site for fly because the excreta cannot be seen. TECHNICAL AND ENVIRONMENTAL ASSESSMENT 13 Table 2-2. Sanitation Systems for Selected Studv Communities Sanitation systems present (*) or utilized (percent) Household svstems On-site Communal Community Co systems (see table latpine Aqua- 2-] for latrinee privy Over- population Pit Contin- Aqua- Septic Off-site septic hung data) latrine ROEC VIP Batch uous PF privy tank Sewer Vault Bucket tank latrine East Asia 1 6 15 59 2 10 81 3 1 7 44 44 4 1 15 41 41 5 66 34 6 57 29 14 7 1 70 29 8.9 100 10 1 74 Southeast Asia 11 13 87 12 60 40 13 7 93 14 7 57 36 15 57 20 23 16 1 46 49 4 17 34 56 10 18 66 15 19 Latin America 19 30 2(0 60 21 50 15 29 22 27 73 23 35 14 9 Africa 24 * * * 25 * * 26 27 k 28 29 * * 31) 98 1 1 31 70 31) 32 84 1 1 33 30 70 Note: ROFC, Reed Odorless Earth Closet; vii', ventilated improved pit latrine; 'T, pour-flush toilet. Pit latrines are most suitable in low- and medium- the ground. In addition, if nearby groundwater is used density areas (up to about 300 persons per hectare) for drinking, pit latrines should not be placed within 10 where houses are single storied.3 It is customary to place meters or so of the well. If the soil is fissured, the the latrine 3 to 5 meters from the house. Where pollution from the latrine will be more extensive, and appropriate measures for odor and fly control are taken this distance may need to be increased. (as in VIP latrines and ROECs), the latrine may be Where these environmental limitations do not apply, placed adjacent to the house. In sandy soil the pits may or where the disadvantages of other systems outweigh need to be partially lined to prevent collapse, and where those of pit latrines, the VIP latrine and ROEC are the ground is rocky they may be difficult to dig. In areas suitable for replication. Their technical designs are that have a high water table or that are prone to good; they can easily be upgraded to PF toilets; their flooding, the latrine may need to be raised partly above costs are low; and their potential for health benefits is Figure 2-1. Generic Classification of Sanitatiotn Systems Sliln-^ v,tefll |()n->itc |r~~~~~~~~~~~~)II sitc or 0(11-mitv | DrV | Wt _~~~ ~ ~ ~~ We( We WE |DIrv I. (OcrOLtng lat rmle 8. 1 -r1turfInsil (,I) -arrui&c 'atiki- 14 1ou-ohenl ir,,ci,ernt-tIth1 . 17 (n d tnunv r i ecc Iri Vtlrtlt in)" -eLacnt tanlk 2 I rnch lamtrurt w-yv onakata vr cme r 1t. Vault. anu-al erromv:, tru-k. 3 Pit latimmt 9) qt latrine. .aqunprie . ommaki 15 1 mnt-vmmiuteeistcrI-tlttnh. nnt a rl 4. R-ed )dorl-e Inrtb ( thue wan tinaplivy, oA-:nka Ys . r 201 Buncket Lan,,- (1n: ) Int 1: laltlilC. p[rtne tIllrk. mimilt sew-I 21. M-hltaical bucket kltrine 5. Vetrtirtal itrtl-ra pit I 1 Sit lage-fitnelt h a. u r - I tw nolnlmt eisten tt-pt1]ttt wsoka,m scptic tannk soakai-iu a r Im Ittleti-u])i^g lninc 1a2t Slnltgc-ltla.sh scpti, l:mk. ew ke 7. (Conttnnmntn -enmllptoing lwyrinc -aka-y 1.1 ('tr^ed cistanl withlo I-w I1llc i11n,l IR 19 20 21 Q, Move'rient ttf liqmrids: *. irrnvemenr tl nstlices Sounrnt: hlic World 1.ank. Watrer SutpplI anntd Waste I)itmsn.el. Pn crt\ antd B.asic Needs Senies (Waishinileton. I) ( Septembier 190()). TECHNICAL AND ENVIRONMENTAL ASSESSMENT 15 high. When introduced with an appropriate educa- known as a "multrum." The composting pit, which is tional program for new users, they can be very effective immediately below the squatting plate, has a sloping at providing sanitation services affordable to the major- floor with inverted U- or V-shaped channels suspended ity of people in rural and urban fringe areas. above it to promote aerobic conditions in the chamber. Grass, ash, sawdust, or household refuse are added to PF toilets the pit to attain the necessary carbon-nitrogen ratios for composting to occur. Moisture must be carefully con- There are two types of PF toilets. The first is a simple trolled. The material slowly moves down the sloping modification of the unimproved pit latrine in which the floor and into a humus vault from which it must be squatting plate is made with a 25-millimeter water seal. removed regularly. Approximately 1-2 liters of water (or sullage) are If the temperature in the composting chamber is poured in by hand to flush the excreta into the pit.4 This raised by bacterial activity to above 60 C (degrees type of PF toilet is especially suitable wherever water is Celsius), all pathogens in the excreta will be destroyed. used for anal cleansing. The second type of PF toilet, In the units observed in southern Africa, however, the which was observed in Indonesia and Colombia, has a temperatures inside were only slightly above ambient. completely displaced soakaway pit (figure 2-1, no. 8) In addition, continuous composters were extremely that is connected to a PF bowl by a short length of 100- sensitive to the degree of user care: the humus had to be millimeter pipe. This type of PF toilet can be installed removed at the correct rate; organic matter had to be inside the house because it is free of odor and insect added in the correct quantities; and only a minimum of problems and its toilet fixture is displaced from the pit. water could be added. Even if all these conditions are When the pit is full, a new one is dug and the latrine is met, fresh excreta may occasionally slide into the connected to it. Alternatively, and especially in densely humus pile and limit the compost's potential for safe populated areas, a vault may replace the pit and be reuse. The conclusion of this study, therefore, is that emptied by vacuum cart (see figure 2-1, no. 18). The continuous composting toilets should not be recom- displaced PF toilet can therefore satisfy the aspiration mended for use in either the urban or the rural tropics. for an "inside" toilet at low cost. In addition, as water Double-vault composting (DVC) toilets (figure 2-1, use increases, the pit can be fitted with an outlet that no. 6) are the most common type of batch composting connects to a drainfield or small-bore sewer system. toilet. They have two adjacent vaults; one of these is This option is examined more fully in the discussion of used until it is about three-fourths full, at which point it sanitation sequencing (chapter 9, the section "Sanita- is filled with earth and sealed and the other vault is tion Sequences"). used. Ash and organic matter are added to the vault The environmental requirements for PF toilets are before it is sealed to absorb odors and moisture. The much the same as those for pit latrines. In addition, composting process is anaerobic and requires several however, 3-6 liters per capita daily of water are re- months, preferably a year, to make the compost patho- quired for flushing. Thus, in areas where water is gen-free and safe for use as a soil fertilizer or condi- carried from distant standpipes or surface sources, the tioner. pit latrine is probably a better choice until the commu- DVC toilets require some care by users to function nity's level of water service is improved. The simple properly and thus are harder to introduce than VIP technical design, low operational requirements, and latrines or PF toilets. They are unsuitable in areas where high potential for upgrading of PF toilets make them an organic waste matter or grass are not easily available or attractive technology for widespread replication in where the users do not want to handle or use the many areas of the world. Their most severe limitations composted humus. These factors tend to restrict their in practice are that users often do not use enough use to rural or periurban areas where users are likely to flushing water or that the toilets can become blocked by have gardens and access to grass for the composting solid materials used for anal cleansing. For these process. Even here, unless there is a strong tradition of reasons, an educational program for users should ac- reusing excreta in agriculture, DVC toilets have no company the introduction of these facilities into a new advantages-and in fact have major disadvantages- area. over the VIP latrine. Composting toilets Aquaprivies There are two basic types of composting toilets: The conventional aquaprivy consists of a squatting continuous and batch. The continuous composters (fig- plate above a small septic tank that discharges its ure 2-1, no. 7) are developed from a Swedish design effluent to an adjacent soakaway (figure 2-1, no. 9). 16 ANALYSIS OF FIELD STUDY RESULTS The squatting plate has an integral drop pipe that is solids settle to the bottom where they are digested submerged in the water of the tank to form a simple anaerobically just as in the aquaprivy. Although the water seal. As long as the water level in the tank is digestion is reasonably good-about 50 percent reduc- properly maintained, odor and insect nuisance are tion in biochemical oxygen demand (BOD)-enough avoided. In order to maintain the water level, the vault sludge accumulates so that the tank must be desludged must be watertight and the user must flush sufficient every one to five years. The effluent is usually disposed water into the tank to replace any losses to evaporation. of in subsurface drainfields. In impermeable soils either The tank normally requires desludging when it is about evapotranspiration beds or upflow filters can be used, two-thirds full, usually every two to three years. although there is little operational experience with In practice, maintenance of the water seal has either of these systems in developing countries. generally been a problem, either because users are Septic tank performance can be improved by various unaware of its importance to the system or because they modifications-for example, the use of three (rather dislike carrying water into the toilet. If the seal is not than two) compartments (figure 2-1, no. 10) or the maintained, there is intense odor release, and fly and addition of an anaerobic upflow filter. This latter mosquito problems abound. One African country modification requires further testing and evaluation banned the building of aquaprivies because of such before its widespread application can be recommended. problems. A variation on the conventional design, The former is a well-known modification that is par- called the self-topping or sullage aquaprivy (figure 2-1, ticularly useful for systems in which excreta and no. 11), was developed to overcome the problem of sullage are disposed of separately (as in PFlatrines). By losing the water seal. A sink is located either inside the this modification excreta can be emptied into the first latrine or immediately adjacent to it and is connected to compartment and sullage into the second, with the the tank so that sullage is regularly flushed into the effluent discharged from the third. This arrangement aquaprivy. Because this additional water necessitates a improves the settling efficiency of the wastes (including larger soakage pit, sullage aquaprivies cannot be used the separation and inactivation of pathogens), increases in urban areas where the soil is not suitable for the soil absorption of the effluent, and permits the soakaways or where the housing density or water table effluent's limited reuse. is too high to permit subsurface infiltration for effluent Septic tanks are suitable only for houses that have disposal. In such cases it is possible to connect the both a water connection (necessary for the cistern-flush aquaprivy tank to a small-bore sewer system, with toilet) and sufficient land with permeable soil for eventual treatment of the sullage in a series of waste effluent disposal. They are an important sanitation stabilization ponds. Desludging would still have to take option because they can provide a very high level of place every two to three years. service to those who can afford it in a given community, If properly maintained, the conventional aquaprivy without necessitating the commitment of community is a sound technical solution to excreta disposal. How- funds for the construction of a sewerage system. Thus, ever, it has no technical advantage over the PF latrine, as part of a sanitation package that can meet the needs which is easier to build and maintain and costs less. In of all the members in a given community, septic tanks addition, with its more sophisticated water seal the PF have a widespread potential for replication because, latrine can be located inside the house and is more with proper soil conditions, they permit satisfactory easily upgraded to a cistern-flush toilet (figure 2-1, no. excreta disposal even for users of cistern-flush toilets 15). The only comparative advantage of the aquaprivy (figure 2-1, no. 16). is that it is less easily blocked if solid cleansing materials are used or this material is thrown into the vault. Thus, except in cases in which users are unwilling Community Sanitation Systems to change such habits, the PF toilet should be preferred to the aquaprivy. Bucket latrines, vault toilets, communal toilets, and sewerage systems for communities are examined in this Septic tanks section. All require both off-site facilities and a perma- nent organizational structure with full-time employees The final household system to be discussed is the to operate successfully. septic tank. The conventional septic tank (figure 2-1, nos. 12, 13) is a rectangular chamber, separated into two compartments sited just below ground level, that receives both excreta and sullage. During the one to The traditional bucket latrine (figure 2-1, no. 20) three days of hydraulic retention time in the tank, the consists of a squatting plate and a metal bucket, which TECHNICAL AND ENVIRONMENTAL ASSESSMENT 17 is located in a small vault immediately below the Vaults are suitable for medium-rise buildings in which squatting plate. The bucket is periodically emptied by a excreta can be flushed down a vertical pipe into a night-soil laborer or "scavenger" into a larger collec- communal vault at, or below, ground level. From the tion bucket which, when full, is carried to a night-soil user's point of view, there is little difference between collection depot. From there the night soil is normally vault and PF toilets; either can be built inside the house taken by tanker to either a trenching ground for burial and no nuisance problems are likely. In addition, vault or to a night-soil treatment works. toilets require a minimal amount of water (3-6 liters per During the course of this study, bucket latrine capita daily) and are suitable for any type of soil and at systems were observed in four countries in Africa and very high population densities. They can easily be Southeast Asia. Problems of odor, insects, spillage, and upgraded into sewered PF toilets if at some stage it is generally unsanitary conditions at all collection and desired to improve facilities for sullage disposal. Their transfer points were ubiquitous. Although it is possible main disadvantage is one shared by all community to make several improvements to the normal bucket facilities: the need for an institutional capability to latrine system (for example, by providing facilities for organize the collection service and operate the treat- washing and disinfecting the buckets, covering collec- ment facilities. The vault toilet systems for which tion buckets with tightly fitting lids, or mechanizing the quantitative data were obtained in this study were system, as shown in figure 2-1, no. 21), it is still difficult found in East Asia, where municipal institutions were in practice to ensure that the system is operated well developed. Although the vault toilet system is satisfactorily in developing countries. Even an im- technically sound, its widespread replication in other proved bucket latrine system, therefore, is not one that parts of the world requires that it be subjected to can be recommended for new installations. Existing prototype testing on a scale large enough to involve bucket latrines should be improved as a short-term institutional development. measure and replaced by some other technology in the long term. In the high-density urban areas where bucket latrines are most often found, the most likely Communal sanitation facilities replacements for them include vault toilets and com- There are no unusual technical requirements for a munal facilities.Thrarnouuultcnclrqiensfra communal toilet. It may be a PF toilet, an aquaprivy, a low-volume cistern-flush toilet, or some other type. If Vault toilets shower, laundry, and clothesline facilities are not avail- Vault toilets (figure 2-1, nos. 18 and 19), which are able in the houses, they may be provided at the extensively used in East Asia, are similar to PF toilets, communal sanitation block. Such block facilities are except that the vault is sealed and emptied by a vacuum normally designed with a capacity for twenty-five to pump at regular intervals of two to six weeks. As with fifty persons per toilet compartment and thirty to fifty the PF toilet, the vault may be built immediately below persons per shower. The most frequent problems en- the squatting plate or displaced from it and connected countered in the communal facilities visited during this to it by a short length of pipe. In the latter case, the study were inadequate water supply (for PF toilets) and vault may be shared by adjacent houses with some poor maintenance. From a mechanical viewpoint, com- savings in construction and collection costs. munal facilities may be the only low-cost alternative for The vault itself need not be large. For example, for a providing sanitation to people living in very dense cities family of six and with the vault being emptied every two with no room for individual facilities. The social and weeks, the required vault volume is only 1.25 cubic institutional commitment to provide for their mainte- meters, and about 0.6 cubic meters of night soil must be nance, however, can be a serious constraint. removed each time the vault is emptied. The collection cart or truck is equipped with vacuum tubing, which may be as long as 100 meters to permit access to houses Sewerage distant from a road or path. Disposal of the collected Conventional sewerage (see figure 9-5, d, in chapter night soil is usually by trenching or treatment works 9) consists of a cistern-flush toilet connected to a (see the section "Treatment Alternatives," below). network of underground sewers, which transport sew- The vault toilet, emptied by either mechanically, age and sullage to a treatment or disposal facility. The electrically, or manually powered vacuum pump, is an cistern-flush toilet is a water-seal squatting plate or extremely flexible form of sanitation for urban areas. pedestal unit from which excreta are flushed away by Changes in urban land use are easily accommodated by 10-20 liters of water stored in an automatically refill- redefining the routes for collection tanker trucks. ing cistern connected to the household water supply. 18 ANALYSIS OF FIELD STUDY RESULTS Sanitary sewers are usually made from concrete, consider sewerage only for those areas in which it is asbestos cement, vitrified clay, or polyvinyl chloride clearly the most appropriate sanitation system for (Pvc) pipe. Sewers are designed for transport by social and economic, as well as technical, reasons. gravity of a maximum flow of up to four times the average daily flow, and they need to be laid with a steep enough slope to provide for a "self-cleaning" velocity of Factors Affecting Choice of Technology about 1 meter per second to avoid blockages. A conventional sewer system will require a 225-milli- Before the discussion proceeds to treatment, recla- meter pipe (the minimum recommended size) to be laid mation, and disposal alternatives, a summary of the at I in 90 slope, whereas a sewered PF system with a major technical and environmental factors that affect vault to settle solids needs only a 100-millimeter pipe the choice of sanitation technology may be useful. laid at a 1 in 200 gradient. Clearly, there is a considerable difference in excavation and pipe costs Physical environment between the conventional and small-bore sewers, which will grow larger as the ground becomes rockier. Be- A group of maps is provided in this chapter to cause small-bore sewers carry no solids, they also illustrate the range of constraints imposed by the global require fewer manholes than conventional sewers. distribution of environmental variables (maps 1 The main advantage of a conventional system is the through 11) and some of the common tropical sanita- high convenience to users it provides. The main techni- tion-related diseases (maps 12 through 19). cal constraints are its large water requirement, the Information on the natural physical environment of difficulty of the excavation in very dense areas or in an area will often permit the exclusion of certain those with poor ground conditions (rocky soil, high options. Winter temperatures (maps I and 2) affect the water table, and the like), the problem of laying sewers performance of waste treatment ponds, digesters, and in fairly straight lines through areas of "unplanned" biogas units because each decrease of about 10 C or housing without substantial demolition, the susceptibil- 18 F (degrees Fahrenheit) causes a decrease in bio- ity of the pipe and joint materials to corrosion in hot chemical reaction rates by one-half. The distribution of climates, and the blockage and extra maintenance precipitation (map 3) indicates the general levels of problems that may arise during the early years follow- flooding, runoff, water table, and plant growth. The ing construction of a sewer (when it is underused).5 A climate diagrams (maps 4 through 6) show details of further problem of conventional sewerage is the envi- temperature and precipitation for specific locations ronmental hazard created by point discharge of such considered in the present research. Horizontal scales on large volumes of wastewater. This problem is reduced the inset charts are in months (January to December in with (expensive) tertiary treatment plants, but devel- the northern hemisphere and July to June in the oped countries are now discovering that even elaborate southern hemisphere); in each case summer is in the treatment does not remove all of the environmental middle of the scale. Aridity index maps (maps 7 costs and consequences. through 9) show the ratios of potential evaporation to Over the past three decades numerous attempts have precipitation and indicate climatic zones, particularly been made to design and build sewer systems around those subject to desertification, in which the recovery of the world. The success rate has not been high. The water, fertilizer, and energy from wastes is most impor- majority of them have not gotten past the design stage tant. Soils and potential productivity are shown in maps because of the failure to take into account the financial 10 and 1 1; the former reflects long-term effects of constraints. Many of those built in developing countries climate, and the latter is a measure of land or aquatic have had very serious problems with consumer accep- plant growth. Soil and weather allow for higher produc- tance. Connection rates, even where mandated by law, tivity in the tropics, where rapid cycling of material have been very low. Of the eight sewer systems included through the biosphere is a major element in the efficien- in this study, three were operating close to capacity. cies of waste treatment ponds. One of those was in a Japanese city, and the other two Distributions of most of the diseases shown in maps were African systems built in the 1950s. Nevertheless, 12 through 19 indicate the environmental influence on many cities have a central commercial area with high- health in the tropics. The limits are based on reported rise buildings in which sewerage may be the most cases, and the absence of cases may be because of the feasible solution. The lesson seems to be that the absence of the disease itself or of specialists who can economies of scale in sewerage are illusory in areas recognize it.6 where consumer acceptance is not assured. In develop- In contrast to the regional or global environmental ing a sanitation package for a city, planners should influences, local changes in land use are often the (Text continues on page 39.) MAP I Average January Temperature (degrees Farenheit) ,ouree< -A. Natio.a ,k c and Jtterc Administratn IN oalder. C' .. )3 :o " . r o .Source: U.S. Nationael O)ccanic and Atmospheric Administration (im)^QA) I BOUI&r. Celo.), fromi U S. Department of Commerce rec(ords. 2- 20 MAP 3. Generalized Annual Global Precipitation (millimeters) 7 <250; 250-5(0; 50()},,O) m 1,000-2.000 >2,000. Source Same as for map I. MAP 4. Temperature and Precipitation, Central and South America (degrees Celsius; millimeters) 25,50 940 1 _ON 700N i X wk/ 9-4 -\7 200 50 N I1 I os 1034t 2 5200S 300N Oaaa *--30cS 2 00N Patz~~~~~cuaro 6Z 100N Mana2,1011Guat50 3 dry season; jf wet season; wet season, precipitation more than 100 millimeters. Numbers on inset charts indicate, in order, average annual temperature (°C) and precipitation (millimeters). Horizontal scales are in months-January through December in the northern hemisphere, July through June in the southern hemisphere-and showperiods of freezing (solid bars) and frost (hatched bars). Left-hand vertical scales ildicate temperatures above OC in 10' intervals; right-hapd vertical scales indicate 0-100 millimeters of precipitation in 20-millimeter intervals and 100 millimeters of precipitation in 100-millimeter intervals. Sources: Adapted from H. Walter and H. Lieth, Klimadiagram Weltatlas (Jena: VEB Gustav Fischer Verlag, 1960) and H. Walter, E. Harnickell, and D. Mueller-Dombois, Climate-diagra;n Maps (Berlin: Springer-Verlag, 1975). - 1960 Gustav Fischer Verlag and 1975 Springer-Verlag, respectively. Used by permission. MAP 5. Temperature and Precipitation, Asia (degrees Celsius; millimeters) 70| N 1 1 70_N 600N~~~~~~~~~~~~1,0 06O 50N NOO See map 4 for explanation of inset charts.Seoul ' Sources: 40o a 30'N 3O 200N 2N 1 O0N423ION 00la 1 00S~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 dry season; wet season; wet season, precipitation more than 100 millimeters. See map 4 for explanation of inset charts. Souirces: Same as for map 4. MAP 6. Temperature and Precipitation, Africa (degrees Celsius; millimeters) AL28,60 791- _ 30°N = ,~: 20,80 26 30°N 20°N I 20°N 10ON IO N 41 10°00 20, 60 835 00 1 00SI O 0°S k S 44_8- ..... - d1 .6 7_ .:--1: r_:: . ; 20,80 44 Frnitan /2 S 0°S~~~~~~~~~~~~~~~~~~~~1, 6 01: 300S30S r44 I I I1 1 1 9. dry season; wet season; wet season, precipitation more than 100 millimeters. See map 4 for explanation of inset charts. Sources: Same as for map 4. MAP 7. Aridity Index for Asia 70EN 70°N 600N 600N 500N 500N 400N 400N 300N 10300N 200N 200N 1 O0N 1 O0N f humid; E savannah or steppe. El desertification; E desert. Isopleths show ratios of potential evaporation to average precipitation. Ratios are calculated by dividing mean annual net radiation by product of mean annual precipitation and latent heat of vaporization. Sources: After Budyko-Lettau. Adapted from D. Henning, Atlas of Climate Aridity Indices (unpublished) and F. K. Hare, Climate and Desertification (Toronto: University of Toronto Press, 1976). - 1976 University of Toronto Press. Used by permission. MAP 8. Aridity Index for North and South America SOON 1 D0N 700N ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~20 30ON~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~0 500N ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~30 200N ~~~~~~~~~~~~~~~21400S IOON 5006 huid;i sdavannah or steppe; desertification; m desert. See also note to map 7. S'ource.s: Same as for map 7. MAP 9. Aridity Index for Africa and Australia b 0 z humid; 2 avannah or teppe; u deertificatio; S desert 200N 1 00S 1 00N 200S 00 N ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~: 300S 100OS 400S 50200S 005S 3005 F-humid; savannah or steppe; desertification; desert. See also note to map 7. Sources: Same as for map 7. MAP 10. Potential Productivity by Annual Carbon Fixation (grams of carbon fixed per sqiaic meter annually) >800; [ 400-800; [ 100-400; 0 (-1(0;0 oceans 0-200; J oceans >200, Potential productivity, a measurement of the amount of plant material produced aniually, is the response of plants (and the animals that eat them) to the combination of total annual sunlight, temperature, and precipitation (see maps 1-3). High values indicate rapid recycling of nutrients, potentially rapid growth of living matter, high year-round efficiency of sewage oxidation ponds, and favorable conditions for fish culture in ponds enriched with night soil or sewage. Low values indicate environmental constraints uponl these processes. Source: Adapted from 11. 1Lieth, "Modeling the Prilimary ProduCtivity of the World," in H. Lieth and R. H. Whittaker (eds.), Primary Productivity of the Biosphere (Berlin: Springer- Verlag, 1975). i 1975 Springer-Verlag. Used by permission. MAP 11. Generalized Distribution of Soil Types and Processes ~~X23o~2 ANTARCTIC CIRCLE \ _ > t .7 - - - ~ laterization; X salinization; Q calcification; a gleization; E3 podzolization. Both plants and soils are affected by climate, and regional variations occur not only in plant productivity (map 10) but also in soils and the processes governing their formation. Laterites are clayey soils with high contents of (red) iron oxide and aluminum oxide whose nutrients have been largely leached away by rains. Saline soils are created when evaporation from soil or plant surfaces exceeds water supply, causing dissolved salts to accumulate in the soil. Caliche is a layer, often discontinuous, of calcium and magnesium carbonates at about the depth to which water percolates. Gley is a blue- or olive-gray layer of clay under the surface of certain water-logged soils in cold climates. Podzols are leached, acidic soils formed in temperate-to-cold, moist climates under coniferous or mixed forest or heath. Combinations of soil types and potential productivity constitute constraints within which waste reclamation activities can function. Source: NOAA, Environmental Research Laboratories (Boulder, Colo.). MAP 12. Global Spread of Cholera, Pandemic El Tor Variety from Celebes to Africa, 1961 to 1975 Source: Adapted from Richard G. Feachem and others, Sanitation and Disease: Health Atspects of Excreta and Wastewater Management, World Bank Studies in Water Supply and Sanitation, no. 3 (Baltimore: Johns Hopkins University Press, forthcoming). MAP 13. Known Geographical Distribution of Schistosoma haematobium and S. japonicum S. haematobium - Africa S. japonicum - Southeast Asia Source: Same as for map 12. MAP 14. Known Geographical Distribution of Schistosoma mansoni Source: Same as for map 12. MAP 15. Known Geographical Distribution of Ancylostoma duodenale Source: Same as for map 12. MAP 16. Known Geographical Distribution of Necator americanus Source: Same as for map 12. MAP 17. Known Geographic Distribution of Taenia saginata (Beef Tapeworm) Source: Same as for map 12. MAP 18. Known Geographical Distribution of Taenia solium (Pork Tapeworm) Source: Same as for map 12. Source. Samie as for map 12. MAP 19. Known Geographical Distribution of Culex pipiens E northern and southern limits of range; B Bancroftian filariasis transmitted mainly by other mosquito species, E Bancroftian filariasis transmitted mainly by C. pipiens complex; * Residual Bancroftian filariasis transmission by C. pipiens complex. Culex pipiens is a complex of mosquito species and subspecies. The main tropical species, and the vector of Bancroftian filariasis in thosc tropical areas where the infection is transmitted hy Culex, is Culex quinquefasciatus (previously also known as Culex pipiens fatigans, C. p. quinquefasciatus, or C. fatigans). Other important species are C. p. pipiens, C. p. molestus (the probable vector of Bancroftian filariasis in Egypt), and C. p. pallens. Source: Same as for map 12. Figure 2-2. Special Plan of Two Low-income Urban Residential Neighborhoods X~~~ ~~~~~~~~~~~~~~~~~ L, -- :, t.: 4 0 50 100 I l I I I I I i East Asia Scale in meters IO) 100 l 1 l l l 1 West Africa Scale ini meters Source: The World Bank. TECHNICAL AND ENVIRONMENTAL ASSESSMENT 39 limiting factor, especially in urban areas. The crowding Housing density in single-story residential areas of two cities, East Asian In densely populated urban areas, VIP latrines, and West African, where average population densities PF toilely and urbanks VIP s ays may of 1,000 to 1,500 persons per hectare (10000 to ROECs, PFtoilets, and septic tanks with soakaways may 15o,000 per square kilometer) are found, is shownt be infeasible. Conventional sewerage is feasible if figure 2-2. space kisovauabl and mosofi is ' ou pie gradients are steep enough to provide self-cleansing byihouses2. Whaen istreetsble and highwyst are btis prin, velocities. Sewered PF systems are also feasible and can by houses. When streets and highways are brought in, beudfofltrgains.Vcm-ukcrae buildings suffer by being moved or truncated. The be used for flatter gradients. Vacuum-truck cartage buildings suffer.by being moved or truncated. The from vaults is a third possibility in dense areas. The addition of rental rooms to what was previously rela- choice among these possibilities is made essentially on tively spacious housing is shown in figure 2-3. The economic grounds, although sullage disposal facilities smaller of the two houses is occupied by 61 people, each and access for service vehicles are important for vault of whom has an average of about 5 square meters of toilets. living space (from figure 2-3) and no more than 9 It is not easy to define at which population density square meters of total space (equivalent to about 1, 100 on-site systems (such as vip latrines, ROECs, PF, and persons per hectare). Even under these conditions, DVC toilets) become infeasible. The figure is probably there is room for extended household latrines using most commonly around 250-300 persons per hectare buckets or vaults. By way of comparison, sewered for single-story homes and up to double that for two- communal latrines would occupy up to 3 percent of story houses. Pit latrines, however, have been found to total land area where population densities are about . . ' ' 1,000 persons per hectare and up to 10 percent if shower provide satisfactory service at much higher population ',000 persons per hectare and up to IO percent if shower densities. The essential point is to determine, in any and laundry facilities are provided (not including space for clothesyfaclinies) are provided (not including space given situation, whether or not there is space on the plot to provide two alternating pit sites that have a mini- smum lifetime of two years, or whether the pit could be Levels of water supply service easily emptied if space for alternating pit sites is not Hand-carried supplies from a public water hydrant available. For off-site systems (such as vaults and restrict feasible technologies to those not requiring wa- cartage), the limiting factor is normally the accessibil- ter, such as VIP latrines, ROECs, and DVC toilets. PF ity of the vault, not population density-so that, in very toilets may be feasible in a sociocultural environment crowded and irregularly laid-out areas, bucket latrines where anal cleansing practices already require the car- and communal facilities may be the only options. rying of water to the toilet. Even then, however, a sufficient amount of water may not be available for flushing. A system that requires water to transport Complementary investments excreta is clearly not feasible. The facilities mentioned Off-site night-soil or sewage treatment works are above can be converted to water-seal units, if desired, required for vault toilets, sewered PF toilets, and con- when the water supply service is improved by a yard or ventional sewerage systems. Sullage disposal facilities house connection. must be considered for all household systems and vault, Yard connections permit PF and vault toilets, but bucket, and community facilities. For those systems, not cistern-flush toilets. If sullage generation exceeds achieving disposal through reclamation (the reuse po- 50 liters per capita daily, sewered PF toilets also be- tential) must be thoroughly and realistically examined, come technically feasible. The choice among these ad- especially in areas where excreta reuse is not a tradi- ditional possibilities and vip latrines, ROECs, and DVC tional practice. For example, DVC latrines may be toilets, which are also still technically feasible, de- provided where there is a demand for reuse. Other pends on other factors such as soil conditions, housing technologies that require off-site treatment facilities density, and consumer preferences. have high potential for sludge or night-soil reuse. Household connections make cistern-flush toilets with conventional sewerage or septic tanks and soak- Potential construction by homeowners aways technically feasible. Sewered PF toilets are also for possible; but they have high capital costs, and alterna- Where financial constraints are severe, the potential tive improvements in sullage disposal may be economi- for "self-help" construction of the various technologies cally more attractive. Sanitation technologies are very should be considered. Self-help can provide the un- sensitive to water use. Figure 2-4 illustrates schemati- skilled labor and some (but not all) of the skilled labor cally how various levels of water use lead to different required for the installation of vip latrines, ROECs, and sanitation options. DVC and PF toilets. It requires organization and super- Figure 2-3. Typical Floor Plan of Low-income Rental Units I--- - - - - - BR12BRI3 -BR 13 F BR 12) (O + 5) Future (22) (3+ 2) (2+1) (6+ Future extension Bath extension _ …I_ _ _ _ _ _ BR I i6 t 5! / -~~~~~~~~~~~---2 ~~~~(I t 0} / BR 4 RI BR 11 2 BR 12 (6 + 5) (1(+3) Tree BR 11 R 4 0 BR 10 (0 + 2 Washing and drying of clothing, BR 6 BR IO family gathering, eating, B I+0 4+0 storage of household effects, (2 + 0) BR 9 playgrounds for children (I + 0)Kihe I BR 8 2 (family BR6 BR 5 BR4 ivorkshop BR7 (2 +0) (2 + 2) (0 ± 3)/ room) (2+ 1) 0~~ to Scale in meters Construction sequence Served by standpipe: BR bedroom. Note: Numbers in parentheses indicate numbers of adults and children in each bedroom. Source: The World Bank. Figure 2-4. Schematic of Relations between Levels of Water Service and Options for Sanitation Combincd with municipal scwagc (Japan) Fcrtiliter Carrv 5-2D Icd Raw 1TraditionalAnmlod Vendor 5-20) lcdAnmlfo = == On-site Agriculture Animal loud Agruculturc Fcrtilizcr 1-2 lcd excrctra Ltlid surfac Raw r Carry 5 -20 lcd dAnimal foiod Standpipc | Man lal 5 2(1 Icd Manu.d Agriculturc Vendor Horse-drawn 5-40 le(i f ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l till,f'ril'c Animal fooid CarrySltC -itE Agriculture Neighborhood Carry t I tildle i amenities Latrine R.lw_ Shower/la mndry -; Aminal food _ _ Mcchalnical - Agriculture C(Mpobt!2-- Animal food _ Agriculture 44 ~~~~~~~~~~~~~~~~~~~~~~~~~~~<4(1 lcd Pit pirivis/horc-holc latrincs . '_ _ Soi, I Groundwater Compo t ig toilets Solids Groundwater Supply Solids Biogas unit Methanc Energy Combincd LLq d L Yard spigot 3(t801 lcd Garbage Laid-surlace Groiundwater burial Cesspoolfscptic tank (with ncxretal Stock or gardpn waterig CcslpoLe/srppic Ian'is Sdads Groundily 5 40 Icd Ra lw ewilge Llndt dispos.;l Surfacc at{Cl Municipal sewer system Etflucnt | tIttundE tr rch.rg I rc lted cwagc Dlgest(onI C|ondae rehrgergjc I~ Sludgeewag Full phtmhtng >8(t Icd ~~~~~~~~~~~~~~~~~~~~~~~~Solids1- Full pitimbing >80 lcd~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Lalnd di,po,.d Icd Liters per capita dailv. 42 ANALYSIS OF FIELD STUDY RESULTS vision by the local authority, especially in urban areas. therefore, is to exclude certain technologies in a given The other technologies have less potential for self-help situation, rather than to select the best. The choice labor and, indeed, require experienced engineers and among the remaining technologies is often based princi- skilled builders for their design and construction. pally on considerations of cost, user preferences, and reliability. Algorithms to aid in the choice of technol- Hygienic habits of users ogy are described in chapter 9. The choice of anal cleansing materials, in particular, can affect the choice of technology. PF and cistern-flush Treatment Alternatives toilets cannot easily dispose of some anal cleansing materials (for example, mud balls, corncobs, stones, or The objectives of night-soil or sewage treatment are cement-bag paper) where a traditional practice of to eliminate pathogens so that human health will be disposing of such anal cleansing materials outside the protected and to oxidize organic matter so that odors, toilet does not exist. The practice of using water for anal nuisance, and environmental problems (such as algal cleansing may present problems for pit latrines in soil blooms or fish kills) will be eliminated. The first with limited permeability or for DVC toilets, the con- objective may be achieved by the separation of feces tents of which may become too wet for efficient com- from the community and the second by various combi- posting. nations of separation, sedimentation, digestion, and oxidation. Institutional constraints Details of conventional sewage treatment process- es-designed primarily to oxidize organic matter- are Institutional constraints often prevent the satisfac- dsrbdi tnadsntr niern et. tory operation of sanitation technologies even when the technologies are properly designed because adequate review here of the objectives and principles of the health maintenance (at the user or municipal levels or both) aspects of sewage treatment as they apply to developing often cannot be provided. Thus, educational programs countries, together with some principles of design of foftsens candinotibeuprovided.vThu,meducationalgpr ra selected technologies particularly suited to the treat- for users and institutional development should gener- g~ ally form an integral part of program planning for ment of night soil and sewage in developing countries, is sanitation. Changes, especially those in social attitudes, pprP can be accomplished only slowly and may require a planned series of incremental improvements over time. Conventional treatment processes The conventional treatment observed in developing countries during this research had the technical disad- Comparison of Technology vantages of extremely poor pathogen removal and frequent operation and maintenance problems from From the foregoing discussion of the factors affect- shortages of properly skilled personnel and imported ing the selection of technology, the technical suitability spare parts. of various technologies for application in a specific Effluent from a trickling filter plant with about five community can be determined. As a first step, compar- hours of retention will contain significant concentra- ative criteria must be defined. One possibility is to tions of viruses, bacteria, and protozoa and helminth compare the technologies in a matrix that displays ova and is thus unsuitable for unlimited reuse in performance according to the established criteria. Such agriculture. Activated sludge plant effluent with, say, a matrix, which can serve as a guide for nontechnical twelve hours of retention, is better than that from readers and a convenient summary for professionals, is trickling filters but will still be microbiologically con- given in table 2-3. Ranking technologies by means of taminated. subjective weighting produces a numerical comparison Batch digestion of sludge for 13 days at 50'C in a of spurious precision. Moreover, in any given commu- heated digester will remove pathogens and reduce nity there are always basic physical and cultural attri- volatile solids by some 50 percent. Digestion at 30 'C butes that-in conjunction with the existing level of for 28 days will remove protozoa and most entero- water supply service and the community's general viruses. Digestion for 120 days at ambient tempera- socioeconomic status-limit the choice of technologies tures will remove all pathogens except helminths. considerably, irrespective of the overall scores achieved Sludge drying for at least three months will be very in a comparison by numerical matrix of all possible effective against all pathogens except helminth ova (see technologies. The most useful function of a matrix, chapter 4). TECHNICAL AND ENVIRONMENTAL ASSESSMENT 43 Other methods of sewage treatment that are used in create shaded breeding sites. This can be prevented by industrial countries include oxidation ditches, aerated providing pond depths of at least I meter and concrete lagoons, sand filtration, chlorination, and land treat- slabs or stone riprap at the upper water level. ment; these are described in a number of standard works.' Aerobic composting Waste stabilization ponds Rapid stabilization and pathogen destruction is ensured by aerobic composting, in which raw night soil Waste stabilization ponds are large, shallow ponds in or sludge is mixed with straw or some other organic which organic wastes are decomposed by a combination matter or with previously composted night soil (or a of bacteria and algae. The waste fed into a stabilization combination of these) so as to provide a water content of pond system can be raw sewage, effluent from sewered 40-60 percent, a carbon to nitrogen ratio of 20-30: 1, PF toilets or aquaprivies, or diluted night soil. and bulk or workability of the mixture. This technology Waste stabilization ponds are an economical method has been applied in the United States (notably, at the of sewage treatment wherever land is available. Their U.S. Department of Agriculture's Beltsville Agricul- principal technical advantage in developing countries is tural Research Center [BARC], Beltsville, Maryland) that they remove excreted pathogens with much less to raw and digested sludge and to night soil. The least required maintenance than any other form of treat- expensive scheme is to form windrows of the night-soil ment. A pond system can be designed to ensure, with a mixture over loops of perforated irrigation drainage high degree of confidence, the elimination of all ex- pipe laid on the ground. Air is then drawn intermit- creted pathogens. This is normally not done in practice tently through the pile into the pipe by a '/3-horsepower because the additional benefits resulting from achiev- blower and expelled as exhaust through a small pile of ing zero survival, rather than very low survival, are less finished compost to reduce odors. Equipment require- than the associated incremental costs. There are three ments are limited to front-end loaders and blowers; types of ponds in common use: screens may be added if the bulking materials are to be separated and recycled. Temperatures in the pile are A high enough (even in winter) for a sufficient time to particularly for strong wastes-retention times are ensure complete pathogen destruction. The operation is one to five days and depths are 2-4 meters. * Facultative ponds, in which oxygen for further siPleadrlal (figur 2-). 'Facultativ ponds, in which oxygn fOther schemes for sludge or night-soil treatment biooxidation of the organic material is supplied include incineration, wet oxidation, and pyrolysis; they principally by photosynthetic algae-they have retention times of five to thirty days (sometimes are too expensive to be considered for general applica- more) and depths of 1-1.5 meters. * Maturation ponds for final settling and pathogen removal that receive facultative pond effluent and are responsible for the quality of the final efflu- Sullage Disposal ent-they have retention times of five to ten days and depths of 1-1.5 meters. The adoption of any of the sanitation technologies, Minimum designs incorporate a facultative pond and with the exception of septic tanks and conventional two or more maturation ponds; for strong wastes sewerage, requires that separate facilities be considered (BOD'-biochemical oxygen demand by the standard for sullage disposal. Sullage is defined as all domestic test-of >400 milligrams per liter), the use of anaero- wastewater other than toilet wastes; it includes laundry bic ponds as pretreatment units ahead of facultative and kitchen wastes as well as bathwater. It contains ponds will minimize the land requirements of the pond some excreted pathogens but, of course, considerably system. Well-designed pond systems in warm cli- fewer than toilet wastes. It also contains many organic mates-incorporating five ponds in series and having a compounds and approximately 40-60 percent of the minimum overall retention time of thirty days-pro- total household production of waste organics-that is, duce an effluent that will be essentially pathogen free some 20-30 grams of BOD per capita daily. This figure, and suitable for unlimited irrigation reuse. however, depends on water consumption. A family with Snail and mosquito breeding in stabilization ponds abundant water for personal and clothes washing and will occur only if poor maintenance allows vegetation to many water-using appliances will generate more sul- emerge from the pond bottom or to grow down the lage BOD than one that uses only small quantities of embankment into the pond, since these conditions water for drinking and cooking. 44 ANALYSIS OF FIELD STUDY RESULTS Table 2-3. Descriptive Comparison of Sanitation Technologies Sanitation Rural Urban Construction Operating Ease of Self-help technology application application cost cost construction potential vIps and ROECS Suitable Suitable in L/M-den- L L Very easy except in H sity areas. wet or rocky ground PF toilets Suitable Suitable in L/M-den- L L Easy H sity areas DVC (double vault com- Suitable Suitable in L/M-den- M L Very easy except in H posting) toilets sity areas wet or rocky ground Self-topping aquaprivy Suitable Suitable in L/M-den- M L Requires some H sity areas skilled labor Septic tank Suitable for rural Suitable in L/M-den- H H Requires some L institutions sity areas skilled labor Three-stage septic tanks Suitable Suitable in L/M-den- M L Requires some H sity areas skilled labor Vault toilets and cartage Not suitable Suitable M H Requires some H skilled labor (for vault con- struction) Sewered PF toilets, sep- Not suitable Suitable H M Requires skilled en- L tic tanks, aquaprivies gineer/builder Sewerage Not suitable Suitable H H Requires skilled en- L gineerlbuilder Note: L, low; M, medium; H, high; VH, very high. a. On- or off-site sullage disposal facilities are required for nonsewered technologies with water service levels in excess of 50 to 100 lcd, depending on population density. b. If groundwater is less than 1 meter below the surface, a plinth can be built. In developing countries sullage is a wastewater with provide attractive sites for mosquito breeding. Disposal approximately the same organic pollution potential as in covered drains or sewers is subject to the same exca- raw sewage in North America. Although its environ- vation and environmental problems as conventional mental impact may be moderate (see the Kyoto case sewerage. study, below), its health hazard will be many orders of magnitude less than that of sewage (see chapter 4). Resource Recovery Thus, an important factor to consider when choosing sullage disposal facilities is how much the community is Technologies for resource recovery in areas included willing to spend on environmental protection. within the present research provide for irrigation with There are basically four kinds of sullage disposal treatment plant effluent; garden watering with sullage; systems: casual tipping in the yard or garden, on-site crop fertilization by raw, digested, or anaerobically disposal in seepage pits, disposal in open drains (usually composted night soil or sludge; fish culture with raw stormwater drains), and disposal in covered drains or night soil; and methane production from municipal sewers. The first will be adequate where water use is sludge or night-soil digesters and from household bio- low and the soil and climatic conditions are such that gas units. the yard remains dry and puddles of water do not form. Readily available data on health aspects were Seepage pits can handle more water but also require collected, but were not sufficiently detailed to show appropriate soil conditions. If stormwater drains are either presence or absence of detrimental effects from used for sullage disposal, they should be designed with a the various resource-recovery practices. Technologies deep (rather than flat) center section so that the small characteristic of these and similar practices are pre- volumes of dry weather flow and sullage will not pond. sented in Appropriate Sanitation Alternatives: A In addition, drains must be carefully maintained to be Planning and Design Manual; their general features free of debris that could block the flow and thereby are summarized below. TECHNICAL AND ENVIRONMENTAL ASSESSMENT 45 Water Required soil Complementary off-site Reuse Health Institutional requirement conditions investment potential benefits requirements None Stable permeable soil; None L Good L groundwater at least 1 meter below surface' Water near toilet Stable permeable soil; None L Very good L groundwater at least 1 meter below surface' None None (can be built above None H Good L ground) Water near toilet Permeable soil; groundwater Treatment facilities for M Very good L at least 1 meter below sludge surfaceb Water piped to Permeable soil: groundwater Off-site treatment facilities M Very good L house and toilet at least 1 meter below for sludge surface' Water near toilet Permeable soiL groundwater Treatment facilities for M Very good L at least 1 meter below sludge surfaceb Water near toilet None (can be built above Treatment facilities for H Very good VH ground) night soil Water piped to None Sewers and treatment H Very good H house and toilet facilities Water piped to None Sewers and treatment H Very good H house and toilet facilities Agricultural reuse there are hazards of passive carriage of a range of Agricultural reuse is the most common form of pathogens and, in some parts of the world, of Clonor- excreta reuse. There are, however, health risks to chis sinensis (Chinese liver fluke) transmission as well. people and animals working in the fields where excreta Control measures include enriching ponds only with are reused and to those who consume the crops raised in settled sewage or stored night soil or sludge; placing the excreta-enriched soil. There are also problems associ- fish in clean water for several weeks prior to harvesting; ated with the chemical quality of the compost, sludge, clearing vegetation from pond banks to discourage the or sewage effluent-including concentration in crops of snail host of Clonorchis; promoting food hygiene in the heavy metals and potential damage to the soil structure handlmig and processing of fish; and discouraging the fro hig soiu cnetais. consumption of raw or undercooked fish. from high sodium concentrations. Pigs are fed raw excreta in a number of South and Yields of carp in fertilized ponds vary from 200 Southeast Asian, Central American, and West African kilograms per hectare yearly in rural, subsistence ponds locations. They provide direct and efficient conversion to 1,000 kilograms per hectare or more yearly in com- of wastes to protein, but the health risks are obvious, mercial ponds; yields of tilapia are 2,0003,000 kilo- and reliable epidemiological data are lacking. Thor- grams per hectare yearly in well-maintained ponds. ogcokngx re- s e is an Fish yields can be doubled by raising ducks, whose feces essential measure of effective pathogen control, provide additional nutrients, on the ponds. Ecological niches in the pond can be introduced; for example, the common carp (Cyprinus carpio) and the grass carp (Ctenopharyngodon idella) feed primarily on benthic Human excreta can be used for raising aquatic plants zooplankton and aquatic weeds, respectively. Up to and animals. The four main kinds of aquaculture are 7,000 kilograms per hectare yearly can be achieved if freshwater fish farming; marine culture of fish, shell- supplemental feeding with grass, other vegetation, rice fish or shrimp; production of algae; and the emergent bran, groundnut cake, or the like is provided and bot- production of aquatic plants. tom-feeding fish are added to the pond. Reliable data on freshwater fish farming are avail- The design of fishponds is essentially the same as that able from South and Southeast Asia. In this practice of waste stabilization ponds. Depths are usually > 1 46 ANALYSIS OF FIELD STUDY RESULTS Figure 2-5. Schematic of Beltsville Agricultural Research Center (BARC) System for High-rate Thermophilic Composting (meters) Aeration pipe is perlorated for even air distribution b E I',.--- V== a a o ii~~~ - - -- - -- - -- - - - As needed Plan 2,500 to 3,50 3.000 V 1/2 WHeight-i 2- + -~ Finished compost bed 7,500 Section a-a Section b-b Note: BARC (Beltsville. Maryland) is a facility of the U. S. Department of Agriculture. meter to prevent vegetation from emerging from the island of Taiwan, was completely self-sufficient in pond bottom; deep ponds (>2 meters) are disadvanta- cooking fuel with the added wastes from two pigs. geous because there is little oxygen, and hence few fish, Biogas may also be used for lighting, and large farms in the lower layers. What matters is the correct rate of and institutions are also suitable sites for biogas units. supply of nutrients; regular batch feeding on an empiri- cally determined basis is recommended. Example of Management Schemes Biogas production for Sewage and Night Soil Institutional and household biogas plants are opera- tive in China, India, Korea, the island of Taiwan, and An excellent comparison of well-designed, well-man- elsewhere and use diluted animal feces with or without aged systems for excreta and sullage disposal can be human excreta and with or without vegetable refuse. found in Kyoto, Japan, a city of 1.4 million. Here, The effluent slurry from these plants can be used in public health and aesthetic requirements are met by agriculture and fishponds. The dung from one cow or conventional sewerage for about 40 percent of the similar animal can produce around 500 liters of gas per population and by a vault and vacuum-truck system for day; it contains 50-70 percent methane and its calorific another 40 percent. Sullage from the latter is dis- value is around 4-5 kilocalories per liter. In contrast, charged to surface drainage facilities. After collection, human excreta yields only 30 liters of gas per person the 1.2 liters per capita daily of night soil undergo grit daily. The process is very sensitive to temperature, and removal, comminution, screening, and storage and are gas production is negligible below 15'C. released into sewers at off-peak hours for subsequent One family in Korea reported sufficient gas produc- activated sludge treatment and incineration. Trucks tion for cooking purposes for nine months of the year are thoroughly cleaned at the night-soil transfer station from human and household wastes. Another, on the after each trip. TECHNICAL AND ENVIRONMENTAL ASSESSMENT 47 The areas and water quality of streams in portions of pumps the city served by the two systems are shown in map 20. * Development of training materials, workshops, and Diffused discharges of sullage from unsewered areas do seminars for disseminating information on low- not affect concentrations of BOD and suspended solids cost water supply and sanitation systems and for in the streams; increases in these constituents cannot be training professionals and technicians in imple- distinguished from those due to urban runoff in sewered mentation areas. Moderate increases in stream BOD and solids * Development of a methodology to determine the downstream from the sewage treatment plants reflect most cost-effective mix of sewerage and low-cost both the excellent removals obtained by treatment and sanitation in urban areas the impact of point discharges to the streams. Health * Multidisciplinary evalution and pilot testing of data from the two areas reveal no differences. Costs of methods to convert waste materials into usable the two systems are presented in the following chapter. products. The most significant findings from this case study Much work has been done in this last area, but are that the less expensive vault and vacuum-truck usually only with a narrow (single-purpose) orienta- system can provide health protection equal to, and tion. A multidisciplinary approach-studying many protection of water quality in streams better than, that disposal and reuse possibilities and optimizing various of conventional sewerage. Both systems are providing simultaneous outputs-will result in more efficient and reliable service to areas of an historic, beautiful, and cost-effective solutions. modern city. Future Research Needs Conclusion Most of the technologies discussed in this chapter The review of existing technologies shows clearly have been applied successfully at specific sites and, in that low-cost sanitation methods exist that are either the case of on-site systems, on an individual basis. It is acceptable as now used (for example, PF and vault therefore necessary to design, implement, and monitor latrines) or capable of easy improvement (VIP latrines). pilot projects on a community scale to: Selection of the technique to be used depends on local conditions such as climate, soil permeability, social TsConfirm the replicability of technologies customs, and the like. All technologies identified pro- * fest the transfer and adaptation of technologies for vide the health benefits commonly associated with wa- different sociocultural environments t terborne sewerage, albeit at different standards of con- a Evaluate the ability of communities to organize venience. In short, the choice is not between different and operate communal systems such as the empty- levels of health and sanitary conditions but between ing of vaults and septic tanks service levels and affordability. Fortunately, this choice * Determine effects of sullage disposal and develop is made even easier by the fact that technologies can be methods of sullage disposal for various population upgraded over time to achieve higher standards of ser- densities vice that keep pace with the users' ability to pay for * Test the large-scale application of appliances with them (see chapter 9). low water use (for example, aerated spigots and The most important consideration for the selection of shower heads, overhead low-volume flush tanks) sanitation technology is existing and anticipated water and their effects on sanitation. consumption of the area studied. At low water use (say, Research is also needed in various areas to develop less than 50 liters per capita daily), sullage can usually technologies further, to measure their effects, or to find be disposed of on site without major problems. Above new, more efficient techniques. Among these areas for this level, on-site disposal becomes more difficult, and, further research are: at levels above 100 liters per capita daily, some sort of * On- and off-site sullage treatment and disposal drainage is probably required. Again, local conditions methods (infiltration, evaporation, anaerobic fil- such as population density and soil permeability will tration, oxidation ponds) also be determining factors. It is clear, however, that * Testing and monitoring of the performance of the lower the water consumption, the greater are the handpumps for water supply in rural areas and options for on-site disposal methods. For the same rea- development of a methodology for selection (simi- son (maximum number of options), sullage and excreta lar to that of the algorithms for sanitation) that should be disposed of separately. would reduce present high failure rates of hand- Selection of the technology suitable in any given case IBRD 16280 MAP 20. 135°45' CHINA f c Sewerage and Night-soil . i Collection Areas in Kyoto, Japan ( Sewage Treatment Plants N. KOREA A Night-soil Transfer Plants % * REP. OF * Water Sampling Points K6REA BOD Biochemical Oxygen Demand K o koo (parts per million) Toky ss Suspended Solids (parts per million) * a O -35°15' [III7 - Sewerage Servke Areas */35P15 acfic Ocean Night-soil Collection Areas ) Rivers City Boundary -* International Boundaries ( 0 5 10 KILOMETERS; C . ROD.1. 3 ppR 7 p SS 6.2pp S L9ppm isBOD 29.6 B O % BOD 12.0 ppm SS13.6 ppm SS ppm S53 p 96mpp :t ... .:~~- ) % A R9Opp*pp 130D 11.6ppm SS .&9ppm Bo il FUSHIMIP PLANTw-D32 ppmN .1ppmLANT IlODlI.Sppm SSJ3.6ppmf / SS 33. pmN ROD 11.8 ppm SS 27.7ppm This map has been prepared by the World Bank's staff exclusively for the convenience of the readers of the report to which it is attached. The denominations used and the boundaries shown on this map do not imply, on the part of the\ World Bank and its affiliates, any judgment on the legal status of any territoryaor any endorseement MR 195452 or aceptanee of such boundaries. IMARCH 1982 48 TECHNICAL AND ENVIRONMENTAL ASSESSMENT 49 requires a consideration of various factors in addition to (Baltimore: Johns Hopkins University Press, forthcoming). Much of purely technical ones. For an initial evaluation, a the technical description presented in this chapter is taken from this matrix comprising the various characteristics of companion work. available technologies is sometimes helpful. For final 2. As is discussed in chapter 4, a well-designed and -maintained pit tccio,hnologies, ise slometimes helpful.d For finalte latrine can provide the same level of health benefits in low-density selection, however, the algorithms described in chapter areas as a properly maintained sewerage system can in the inner city. 9 should be used. Both the matrix and algorithms 3. Latrines have been used satisfactorily at twice this suggested emphasize the need to take a comprehensive look at the density in areas where soil conditions or climatic factors are proposed solution-that is, the sociocultural aspects as especially favorable. well as the technical, financial, and economic consider- 4. Sullage is wastewater that does not contain excreta-for ations-to avoid the installation of a system that will example, laundry water and bathwater. not be used or will be quickly abandoned by the 5. Japanese experience has been that there is a lag time of five to community. ten years between commissioning of a sewerage system and voluntary connection to it by a significant number of households. 6. For further detail, see chapter 4 and Richard G. Feachem and others, Sanitation and Disease: Health Aspects of Exereta and Wastewater Management, World Bank Studies in Water Supply and Notes to Chapter 2 Sanitation, no. 3 (Baltimore: Johns Hopkins University Press, forthcoming), a companion to this study. 7. For example, Gordon M. Fair, John C. Geyer, and Daniel A. I. For engineering designs and detailed technical information on Okun, Water Purification and Wastewater Treatment and Disposal each technology, see also John M. Kalbermatten and others, (New York: John Wiley and Sons, 1968). Appropriate Sanitation Alternatives: A Planning and Design 8. Ibid. See also D. Duncan Mara, Sewage Treatment in Hot Manual, World Bank Studies in Water Supply and Sanitation, no. 2 Climates (New York: John Wiley and Sons, 1976). 3 The Economic Comparison COMPARATIVE COSTING lies at the heart of the analy- cost figures. sis of alternative sanitation technologies. The definition A scoring device that has been used on occasion for of technological "appropriateness" developed in chap- ranking alternatives with unquantifiable benefits is a ter I is based partly upon a systematic ranking of matrix that lists both cost and benefit components and feasible alternatives according to their economic costs. assigns values (or relative ranks) to each alternative Implicit in this definition is the search for a common based on an arbitrary scale (or the total number of denominator for the objective comparison of diverse alternatives). Varying degrees of complexity can be systems. That common denominator should reflect built into matrix ranking by weighting the criteria or by both the positive and negative consequences of a given using complex, summary variables of the values. Re- technology and also indicate its overall "score"-either gardless of the variations, however, the lack of objectiv- on an objective scale or relative to other alternatives. ity in the procedure remains a major disadvantage. It is One scoring measurement commonly used in project unlikely that two equally competent analysts would evaluation is the benefit-cost ratio.' It has the advan- arrive at the same value for various alternatives or even tage of providing a single, summary figure representing the same ranking across alternatives. There is no the net economic effect of a given project, which can objective basis for selecting one summary measure over then be readily compared with that of alternative another or any one set of weights for the cost and projects. The disadvantages of benefit-cost calculations benefit categories. Such a nonreplicable method is not are that they do not easily accommodate noneconomic only unscientific but can be misleading to the nontech- benefits and costs (particularly if these are unquantifi- nical reader, who may mistake its statistical complexi- able); they may give misleading results when applied to ties for objectivity. mutually exclusive projects; and they may not reflect In general, there is no completely satisfactory scoring macroeconomic goals such as the creation of employ- system for comparing alternatives with unquantifiable ment or the generation of savings and investment. benefits. Only in the case of mutually exclusive alterna- Fortunately, the last two problems can be remedied by tives with identical benefits can a cost-minimization variations of the basic calculations.2 The difficulties of rule be applied. In such cases the alternative with the measuring benefits for sanitation projects, however, lowest present value of cost, when discounted at the cannot be readily overcome. Indeed, in the case of water appropriate rate of interest, should be selected. For supply projects it has been concluded that the theoreti- given levels and qualities of service, the least-cost cal and empirical problems involved in quantifying alternative should be preferred. But, where there are incremental health benefits are so great as to make differences in the output or service, the least-cost serious attempts at the measurement of benefits inap- project often will not be the economically optimal one. propriate as part of project appraisals.3 Alternative sanitation systems provide a wide range There are also unquantifiable costs associated with of benefit levels. Although most properly selected alternative sanitation technologies. Although it is gen- systems can be designed to provide the potential for full erally possible to assess qualitatively the environmental health benefits (that is, to ensure pathogen destruc- consequences of installing a particular system, it is very tion), the convenience to users offered by an indoor difficult to quantify them since no "market" for such toilet with sewer connection is hard to match with a pit public goods exists. It is even more difficult to compare privy. Many benefits exist in the mind of the user, and consequences of installation with the environmental varying qualities of service result in varying benefit situation that would develop without the project's levels. For this reason, a least-cost comparison will not implementation, thus to determine net benefit or net provide sufficient information to select among sanita- 50 THE ECONOMIC COMPARISON 51 tion alternatives. Nonetheless, if properly applied, it community may incur costs that could have been will provide an objective, common denominator that postponed. By waiting too long, the community might reflects tradeoffs in cost corresponding to different face a rise in the per capita cost of the project (in real service standards. Once comparable cost data have terms) because of increases in population density, for been developed, the consumers or their community example, which could aggravate construction difficul- representatives can make their own determination of ties for some technologies. how much they are willing to pay to obtain various Once the relevant costs to be included have been standards of service. identified, the second costing principle concerns the Thus, the economic evaluation of alternative sanita- prices that should be used to value those costs. Since the tion technologies comprises three components: compa- objective of economic costing is to develop figures rable economic costing, the maximizing of health ben- reflecting the cost to a particular country of producing efits from each alternative through proper design, and a good or service, the economist is concerned that unit the involvement of users in making the final cost- prices represent the actual resource endowment of that benefit determination. This chapter deals with the first country. Thus, a country with abundant labor will have of these. Chapter 4 discusses the public health aspects relatively inexpensive labor costs because labor's alter- of sanitation alternatives, and chapter 5 develops native production possibilities are limited. Similarly, a methods of promoting the involvement of users in country with scarce water resources will have expensive choosing technology. water costs, in the economic sense, regardless of the regulated price charged to the consumer. Only by using prices that reflect actual resource scarcities can the Economic Costing in Theory economist ensure that the least-cost solution will make the best use of a country's resources. The primary intent of economic costing is to develop Because governments often have diverse goals that a price tag for a good or service that represents the may be only indirectly related to economic objectives, opportunity cost to the national economy of producing some market prices may bear little relation to real that good or service. Translated into practice, this economic costs. For this reason it is often necessary to intent can be summarized in three principles to be "shadow price" observed, or market, prices to arrive at followed in preparing cost estimates. meaningful costs for components of a sanitation tech- The first principle is that all costs to the economy, nology. Calculating these shadow rates, or conversion regardless of who incurs them, should be included. In factors, is a difficult task and requires intimate knowl- comparing costs of public goods such as water or edge of an economy's workings.4 The shadow rates used sanitation, too often only costs that the public utility in this report were obtained from World Bank econo- pays are considered in a cost comparison. The costs mists specializing in the countries concerned. borne by the household are often ignored. In analyzing One of the most important shadow values is the the financial implications of alternative technologies, opportunity cost of capital. This is defined as the such a comparison would be appropriate. For an eco- marginal productivity of additional investment in the nomic comparison (that is, for the determination of the best alternative use. It can also be thought of as the least-cost solution), however, it is necessary to include price (or yield) of capital. In countries where capital is all costs attributable to a given alternative-whether abundant, such as the industrialized countries of borne by the household, the utility, the national govern- Europe, one expects the yield on capital to be relatively ment, or other entities. low. This is because capital has already been employed The determination of which costs to include should in its most productive uses and is now being substituted rest on a comparison of the situation over time, with and for labor or other inputs in less and less profitable areas. without the project. This is not the same as a "before In many developing countries, however, capital is a and after" comparison. Rather than using the status scarce commodity and therefore has a relatively high quo for the "without" scenario, the analyst must esti- opportunity cost and should be used in those areas mate how the current situation would improve or where it produces a very high return. Therefore, a least- deteriorate over the project period were the project not cost comparison of alternatives that differ in their to be undertaken. In the case of sanitation systems for capital intensity should reflect the real cost of capital to urban fringe areas, for example, the costs of the economy rather than use capital's market price.5 groundwater pollution and the difficulty in siting new The third principle of economic costing is that latrines are likely to increase over time as population incremental rather than average historical costs should pressure mounts. There is likely to be an optimal time to be used. This principle rests upon the idea that sunk undertake a sanitation project. By acting too soon the costs (those already incurred) should be disregarded in 52 ANALYSIS OF FIELD STUDY RESULTS making decisions about future investments. Analysis of one used by ten people. For this reason, all costs the real resource cost of a given technology must value presented in this chapter are given in household as well the components of that technology at their actual as per capita units. replacement cost rather than at their historical price. In A further difficulty is that the per capita construc- the case of sanitation systems, this is particularly tion cost of a sewerage system will vary considerably as important in the evaluation of water costs. Because the design population varies. In addition, it would be cities develop their least expensive sources of water misleading to use the design population in deriving per first, it generally becomes more and more costly (even capita sewerage costs to compare with those of a pit excluding the effect of inflation) to produce and deliver latrine. In the case of sewerage, the benefits only reach an additional liter of water as the city's demand grows. a portion of the users during the early years. The If the analyst used the average cost of producing today's latrine's "design population," in contrast, is served water, the evaluation would seriously underestimate immediately upon completion of the facility. Any the cost of obtaining water in the future. The decision to technology that exhibits economies of scale will result install a water-carried sewerage system will increase in a diversion of the cost and benefit streams. With such the newly served population's water consumption by a facility, the investment costs are incurred at the around 50 to 70 percent.6 Thus, in calculating the costs beginning of its lifetime and the benefits (services) are of such an alternative, it is extremely important to value realized gradually over time. A schematic representa- properly the cost of the additional water required. tion of this diversion between cost and benefit streams is provided in figure 3-1. The construction period lasts until T,, when the new facility is commissioned and Special Problems of Sanitation Projects begins to produce benefits. As demand grows over time, more and more of the plant's capacity is utilized until, The application of these costing principles to sanita- at T2, the facility is fully used and provides full benefits tion projects is difficult for several reasons. The chief (that is, serves its design population). difficulty is the problem of finding a scaling variable Just as costs incurred in the future have a lower that allows comparison among diverse technologies present value than those incurred today, benefits re- that are designed to serve different numbers of people. ceived in the future are less valuable than those received On-site systems, such as pit latrines, are generally immediately. In the case of the derivation of per capita designed for a single family or household. The latrine's costs, this means that serving a person five years hence overall lifetime will depend on how many people use it. is not worth as much as serving the same person now. To The life of some components (such as a vent pipe), divide the cost of a sewerage system by its design however, may be independent of usage, so that the population would understate its real per capita cost annuitized per capita construction cost of a latrine used when compared with that of a system that is fully by six people will probably not be the same as that of utilized upon completion. Figure 3-1. Benefit-Cost Divergence over Time Total benefit Total cost I I I I II I I I I I I I I I t, t2 Time t, Time of sanitation facilitv's commissioning, t2 time of full use of facilitv s capacity THE ECONOMIC COMPARISON 53 A good method that has been used to overcome the sullage wastes. In sewerage, most septic tanks, and problem of rates of capacity utilization differing across some aquaprivy systems, sullage is disposed of along systems is the average incremental cost (AiC) ap- with excreta. In most of the on-site technologies, proach.7 The per capita AIC of a system is calculated by sullage disposal must be accomplished separately, dividing the sum of the present value of construction through stormwater drains or ground seepage. If storm- (C) and incremental operating and maintenance (0) water drains are present (or would be constructed costs by the sum of the present value of incremental anyway), then the incremental cost of disposing of persons served (N): sullage is very small because storm drains are usually designed to handle flood peaks.9 If sullage is left to soak t= T into the ground, nuisance and possible health risks may E (C1 + Od)/(1 + r)'-1 be created (depending on climate, soil conditions, and AIC, = = T groundwater tables).'" Alternatively, separate disposal E N,1(1 + r)'-' of sullage may be considered a positive benefit in areas t= 1 where the population recycles kitchen and bathwater to irrigate gardens or dampen dust. In such cases, the where r is the opportunity cost of capital and Tis the life removal of sullage through the introduction of a sewer- of the facility. All costs are in constant (noninflated) age system would produce a negative benefit. In a prices and have been appropriately shadow priced. For particular case it is not difficult for the analyst to decide a system that is fully utilized immediately, this calcula- how to treat costs of sullage removal when comparing tion reduces to the familiar calculation of annuitized different sanitation streams. For the purposes of this capital and incremental operating and maintenance study, however, and because a more general compari- costs divided by the design population. son is required, a consistent assumption has been In practice it is often easier to calculate the AIC on applied. Therefore, the costs in tables 3-1 through 3-11 the basis of a volume measure (for example, cubic include sullage disposal only if the sanitation system meters) rather than by person served. For the sewerage itself is designed to accommodate it. This is true of all costs in this study from the cities of Gaborone, Khar- the sewerage systems, all the septic tanks, and the toum, Malacca, Managua, and Ndola, the AIC per Ndola and Newbussa sewered aquaprivies. cubic meter was calculated first because year-by-year A final problem in preparing comparable cost figures projections of treated wastewater were available. These for sanitation systems is the method to be used in volumetric costs were then transformed into per capita gathering data. This study is statistically based, in and per household costs using per capita demand contrast to a synthetic framework that develops an figures. ideal model and tests the effects of varying assump- The AIC method is useful in deriving per capita costs tions. Both methods have their advantages and disad- that can be meaningfully compared with those of vantages. Because so little is known about the technol- systems with different rates of use. This is especially ogy or costs of nonconventional sanitation systems, it important in evaluating sanitation systems because of was decided that a broadly based study involving many the large variation in economies of scale (for example, systems in many different settings would provide the sewerage versus on-site systems or cartage). Whereas best comprehensive frame for designing particular economies of scale are often the engineer's best friend studies or, indeed, for selecting "typical" technologies (in the sense that he can overdesign "to be on the safe and settings to proceed from in developing a synthetic side" without incurring unduly large increases in cost), model. The major disadvantage of a statistical ap- they cause institutional and financial headaches when proach, however, is that it is very difficult to identify demand assumptions turn out to be optimistic or the the factors that cause increased or decreased costs city grows in a different direction from the one ex- because it is impossible to vary one factor at a time pected. Because of the inflexibility of large-scale sani- while holding all others constant. Cross-country com- tation systems once they are built, their financial parisons can be misleading unless one is familiar with feasibility, and even technical success, is extremely the background of the cases compared. For this reason, sensitive to the assumptions used in the design analysis. most technological comparisons are made within a In communities where there is no demand history on single country-whenever possible, within the same which to base forecasts, it is extremely risky to recom- community. mend a system with large economies of scale and with a In one case, that of Malacca, a synthetic study was correspondingly long design period.! carried out to test with more precision the cost differen- An additional problem in deriving comparable costs tial between sewerage and vacuum-truck cartage. In for sanitation systems is the differing treatment of Gaborone, an excellent comparison of the costs of on- 54 ANALYSIS OF FIELD STUDY RESULTS site systems was possible because of recent work carried average number of persons in a household. out there under a project of the International Develop- Because both investment and recurrent costs must ment Research Centre. It is expected that more cost- be included for a least-cost comparison, and because modeling exercises will be undertaken as routine different technologies have different lifetimes, the feasibility studies expand to include nonconventional TACH is an annuitized (or annual) figure. It should sanitation systems. not, however, be interpreted as an amount of money to be spent annually for a particular technology. To Field Results illustrate this, consider the case of the pour-flush (PF) toilet with a mean TACH of $18.7. This figure The costs discussed below have been disaggregated is derived as follows: in two ways, by function and by investment versus Cost recurrent costs. In disaggregating by function, the (U.S. dollars) categories used are on-site facilities, collection, treat- Mean investment cost 70.7 .' ' . Mean lifetime (years) 22.0 ment, and reuse. This distinction is made primarily Mean annuitized investment cost 13.2 because disaggregating by function allows a broad Mean annual maintenance cost 2.4 examination of the costs of repackaging components. Mean annual water cost 3.1 For example, many treatment alternatives can be TACH 18.7 linked with a variety of collection systems, on-site The figure for mean annuitized investment cost is the facilities, or both. In addition, disaggregation by func- tion is amenable to "value engineering" by its identifi- weighted average of three case studies with opportunity cation of the areas in which the greatest potential for costs of capital of 12, 20, and 20 percent and lifetimes of cost savings exists. It also provides the financial analyst twenty, twenty, and twenty-five years, respectively. with a rough guide for determining the proportion of Although the TACH is $18.7, the actual cash expendi- system costs that must be borne by the utility relative to ture (including water costs) is only $5.5 annually. the costs incurred directly by the household. The latter The TACHs obtained for ten technologies (arranged cost is a useful figure for estimating the willingness of by ascending mean TACH) are summarized in table 3-1. the consumer to pay utility rates; this willingness will be Several summary statistics are shown because of a wide based, in part, on the costs to the household of obtaining variation in the number of case studies and the range of the private facilities that will enable it to make use of costs." Contrary to expectation, the technologies do not the utility's service. divide cleanly into community and individual systems The second type of disaggregation is the separation when ranked according to cost. The most expensive of capital and recurrent costs. The difference between technological group (with a TACH greater than $300) technologies with high capital cost and high recurrent includes sewerage and Japanese and Taiwanese septic cost generally parallels that of capital-intensive versus tanks. The middle-range technologies (those with labor-intensive technologies. This is because the invest- TACHs between $150 and $200) are aquaprivies, ment costs of most systems are mainly in capital, and sewered aquaprivies, and Japanese cartage. The low- recurrent costs are mainly in labor. The distinction is cost technologies (those with TACHs less than $100) made here between investment and recurrent costs include both community systems (such as bucket -rather than between capital and labor-partly to latrines) and most individual systems. The divisions emphasize the main cause of the difference and partly between high-, medium-, and low-cost technologies are to stress the important institutional implications of fairly sharp, with large buffer areas available for managing a system with high recurrent costs. upgrading systems. The fact that variations on septic tanks and vacuum-truck cartage appear in two categories indicates the potential for installing a low- cost facility at an early stage of development and The single most useful figure for cost comparisons of improving its standard of service as development technologies is the total annual cost per household proceeds. (TACH), which includes both investment and recurrent Within the low-cost group of technologies there is a costs (properly adjusted to reflect real opportunity costs fairly large variety of systems, ranging from aqua- and averaged over time by the AIC method). The TACH privies and simple septic tanks to pit latrines and PF would, however, be misleading when applied to commu- toilets. Vacuum-truck cartage (non-Japanese) and nal facilities or cases where several households share bucket cartage, with TACHs in the $35 to $65 range, fall one toilet. In those instances, an adjusted TACH has in the middle of this group. The vacuum-truck cartage been calculated by scaling up per capita costs by the figures, however, are derived mostly from case studies THE ECONOMIC COMPARISON 55 Table 3-1. Summary of Total Annual Cost per Household (TA CH) for Sanitation Technologies (1978 U.S. dollars) Observations Technology (number) Mean Median Highest Lowest Low cost Pour-flush (PF) toilet 3 18.7 22.9 23.3 10.1 Pit latrine 7 28.5 26.0 56.2 7.6 Communal septic tank, 3 34.0 39.0 48.0 15.0 Vacuum-truck cartage 5 37.5 32.2 53.8 25.7 Low-cost septic tank 3 51.6 45.0 74.5 35.4 Composting toilet 3 55.0 56.2 74.6 34.3 Bucket cartage' 5 64.9 50.3 116.5 23.1 Medium cost Sewered aquaprivya 3 159.2 161.4 191.3 124.8 Aquaprivy 2 168.0 168.0 248.2 87.7 Japanese vacuum truck cartage 4 187.7 193.4 210.4 171.8 High cost Septic tank 4 369.2 370.0 390.3 306.0 Sewerage 8 400.3 362.1 641.3 142.2 a. Per capita costs were used and scaled up by the cross-country average of six persons per household to account for large differences in the number of users. in Korea and the island of Taiwan, which exhibit a groupings of technologies in high-, medium-, and low- degree of labor efficiency that might be difficult to cost systems indicates that the groupings are probably replicate in other parts of the world. Figures for bucket accurate, although the means may be 50 percent too cartage are mostly from Africa and represent poorly low or too high.'3 functioning systems that probably should not be The total variation is caused in part by differences in replicated without upgrading. Thus, the TACHs of the costs of basic inputs (such as labor) and in part by community systems in the low-cost group are likely to differences in the combination of inputs used (for understate their cost of construction and operation in example, labor-intensive versus capital-intensive treat- other countries. Of course, since all of the costs ment processes). To some extent, these two factors summarized in table 3-1 are derived from particular offset one another because a country with high capital case studies, none can be considered an accurate costs, for example, would be expected to choose a less representation of what it would cost to build a capital-intensive treatment process. For two of the particular system in a different country.'2 systems, vacuum-truck cartage and septic tanks, the difference in the combinations of inputs seems to be very important because the case studies' costs exhibited Cost comparison a bimodal distribution, which could be directly traced to differences in the technologies employed in different It is useful to consider the overall variation of cost in locations. In no two case studies is the exact design of a different countries before making an examination of system replicated; that is, no two pit privies are exactly the cost data for each technology. The magnitude of the alike. For most of the technologies, however, the varia- total variation is quite large, as is indicated in the last tion in costs parallels the general price levels of the two columns of table 3-1. In nearly all cases the range locations. (highest minus lowest) is nearly double the mean. In a A sampling of the most important input costs is statistical study of this type, such a wide variation is to shown in table 3-2. A wide range is exhibited in three of be expected and does not present a major problem the main inputs to sanitation systems: unskilled labor, because the figures are meant to be descriptive rather water, and land. The sensitivity of system costs to than predictive. Even within a single country, cross- changes in factor input prices is not easy to investigate community cost differences were apparent when a in a comparative study such as this. The final section of single technology was found both in urban and rural this chapter, however, presents general conclusions on areas. Further, the relatively wide margins between the cost sensitivities. 56 ANALYSIS OF FIELD STUDY RESULTS Table 3-2. Selected Input Costs and Conversion Factors for Sanitation Technologies (1978 U.S. dollars) Unskilled Water Land Conversion factors labor (per cubic (per hect- Capital Unskilled Foreign Location (daily), meter)' are)c (percent) labor exchange Botswana 1.80 0.38 100 10 0.7 1 00 Colombia 3.20 0.30 n.a. 12 0.3 1.00 Ghana 5.50 n.a. 65 12 0.8 1.75 Indonesia 1.80 O. lI d n.a. 20 1.0 1.00 Japan 10.50 0.85" n.a. 10 1.0 1.00 Korea 4.00 n.a. n.a. 14 0.8 1.12 Malaysia 3.40 0.35 200 12 1.0 1.00 Nicaragua 2.10 0.13 n.a. 20 0.5c 1.10 Nigeria 3.40 1.60 n.a. 12 0.8 1.15 Sudan 1.90 0139d 100 16 1.0 1.25 Taiwan 3.95 0.40d 800 12 0.9 1.00 Zambia 4.00 0.70 65 12 0.6 1.25 Mean 3.80 0.36 222 Median 3.40 0.39 100 Range 8.70 1.49 735 n.a. Not available. Source: World Bank country economists and field consultants reporting on case studies. For more detail, see Richard Kuhlthau (ed.), Appropriate Technology for Water Supply and Sanitation, vol. 6, Country Studies in Sanitation Alternatives (Washington. D.C.: World Bank, 1980). a. Market price in studied communities including benefit package where applicable. b. Average incremental cost (AIC). c. Estimated opportunity cost; collected in those cases where land costs were part of waste treatment. d. Average of several community studies. e. Unskilled rural labor only; for urban unskilled labor, conversion factor is 1. Investment and recurrent costs from a system with high recurrent cost (such as vacuum-truck collection) may be large enough to The distinction between investment and recurrent justify training (or importing) the necessary manage- costs is an important one for both financial and techni- ment skills. cal reasons. A city or community with very limited The breakdown for investments and recurrent costs present fiscal resources but with a good growth poten- for the technologies studied is presented in table 3-3. tial might find it impossible to raise the investment There is no consistent relation between the overall cost finance to build a system with large initial capital of systems and their percentage of investment or recur- requirements, but it could build and maintain another rent costs. The two high-cost and two of the medium- system (with the same TACH) whose recurrent costs cost technologies exhibit recurrent costs amounting to were relatively high. Conversely, a major city in a between 20 and 40 percent of TACH. The systems with developing country that has access to external sources the highest recurrent costs (as a percentage of the total) of funds might prefer to build an expensive system are in the low-cost group-namely, non-Japanese cart- initially with the help of grant or low-interest loan age and buckets, with 52 and 43 percent recurrent cost, capital and possibly reduce its need for recurrent respectively. The lowest percentage recurrent cost sys- funds.4 tems-pit privies, aquaprivies, and composting latrines From the technical viewpoint, high recurrent costs -are in the low- and medium-cost groups. generally stem from large operating and maintenance This somewhat surprising lack of correlation is in requirements. It may be unwise to opt for a system with part because of the nature of the figure for recurrent high recurrent costs in those developing countries in cost. Because economic-rather than strictly finan- which skilled labor is scarce or in which the manage- cial-costs are used in this study, a major item is ment necessary to coordinate large numbers of un- included in recurrent cost that typically does not appear skilled workers does not exist. An offsetting benefit to in engineering cost estimates: the water used to flush this problem is that the employment benefits arising some systems. Although water for flushing is absolutely THE ECONOMIC COMPARISON 57 necessary for the proper operation of many systems Table 3-4. Percentage Investment and Recurrent (from sewerage to PF toilets), its cost is often ignored in Cost of Community Sanitation Systems engineering or financial studies that focus only on those Percentage of total costs incurred by the utility. In order to see how the inclusion of the cost of water for flushing would affect Investment Recurrent the breakdown of investment versus recurrent costs, Technology cost cost separate calculations excluding water costs were made Sewerage 81 19 for those six systems that require water.'5 Sewered aquaprivy 84 16 The overall conclusion from table 3-3 is that nearly Japanese vacuum-truck cartage 68 32 all of the sanitation systems are relatively high in Other vacuum-truck cartage 48 52 Bucket cartage 57 43 investment, as opposed to recurrent, cost. Only in the Communal toilets 88 12 case of non-Japanese cartage do recurrent costs repre- sent more than half of TACH. In ten of the twelve Note: Percentages are calculated excluding costs of water used systems (with the two varieties of septic tank and in flushing. vacuum-truck cartage taken as separate systems), in- vestment costs account for more than 60 percent of In that sense, a distinction based on the relative impor- TACH. If costs of water for flushing are excluded from tance of investment and recurrent costs of different recurrent costs, as is done in table 3-4, only vacuum- systems becomes moot. Whereas sewerage and PF truck cartage and bucket systems show recurrent costs toilets both entail recurrent costs of about 30 percent of of more than 30 percent. their respective TACHs, the important point is that the There are several implications of this concentration investment cost (per household) of the former is more in investment costs. One is that there is probably scope than twenty times larger than that of the latter. for external financing regardless of which technology is Considering only the community systems covered in chosen by a particular city or community. High initial this study, the distinction between investment and costs almost invariably require some sort of financial recurrent costs becomes more relevant. The three mechanism to smooth payments so that they are more cartage systems are much more intensive in recurrent in line with benefits delivered to (and paid for by) the costs than are the water-carried systems (or even consumers. A second implication is that, where funding communal latrines) when water costs are excluded. The constraints are binding, the size of the initial invest- financial result of a system with high investment cost is ment requirement may be the most important determi- relatively high fixed costs that must be met regardless nant of technological choice. There is relatively little of how much service is provided (or how many new scope for substituting a system of higher recurrent cost. connections are made). This can put a real financial Table 3-3. Average Annual Investment and R'ecurrent Cost per Household for Sanitation Technologies (1978 U.S. dollars) Mean Investment Recurrent Percentage of total Technology TACH cost cost Investment Recurrent Low cost PF toilet 18.7 13.2 5.5 71 29 Pit latrine 28.5 28.4 0.1 100 Communal toilet 34.0 24.2 9.8 71 29 Vacuum-truck cartage 37.5 18.1 19.3 48 52 Low-cost septic tank 51.6 40.9 10.7 79 21 Composting toilet 55.0 50.9 4.8 92 8 Bucket cartagea 64.9 36.9 28.0 57 43 Medium cost Sewered aquaprivy' 159.2 124.6 34.6 78 22 Aquaprivy 168.0 161.7 6.3 96 4 Japanese vacuum-truck cartage 187.7 127.7 60.0 68 32 High cost Septic tank 369.2 227.3 141.9 62 38 Sewerage 400.3 269.9 130.4 67 33 Negligible. a. Per capita costs were used and scaled up by the cross-country average of six persons per household to account for large differences in the number of users. 58 ANALYSIS OF FIELD STUDY RESULTS burden on the utility or municipality during the early and aquaprivies was for the superstructures, which years of such a system. It also means that the financial were made of concrete blocks. If these experimental viability of the utility is extremely sensitive to the units had been built in the rural areas by villagers using accuracy of the demand forecast. With systems having local materials such as clay brick or straw matting, high recurrent costs (such as cartage), the response to their cost could have been reduced significantly. slow growth in demand is delayed investment in new The functional breakdown of costs for the twelve trucks and fewer new workers hired. With systems systems is given in table 3-5. Even among the six having high investment costs, however, there is little community systems, on-site costs account for at least 45 scope for reducing costs in response to reduced demand. percent of the total. Japanese and Taiwanese septic This is perhaps not a major worry in cities that already tanks have the highest on-site costs, over $300 per have sewerage in some areas, for example, and are household annually. The size of the costs incurred by ready to expand their system. Here the demand for the the household in total system costs shows the impor- service has already been tested. In cities of the develop- tance of finding ways for funding on-site facilities. The ing world, however, where no such service already very low connection rates of many sewerage systems in exists (and where the ability to pay for the necessary on- developing countries (even when connection is a legal site investment is limited), it is extremely risky to requirement) can probably be explained by the large choose a system that has high investment cost on the household expenditure involved. basis of hypothetical demand projections. In such a Two of the systems, sewerage and vacuum cartage, case, economies of scale are a two-edged sword. exhibit an interesting variety of cost patterns across the case studies. The sewerage costs for the eight cities covered are shown in table 3-6. There is a wide difference in on-site costs ranging from an average of The separation of TACH into its functional compo- about $300 per household annually for the Japanese nents is useful in determining where to direct the design systems to an average of just over $130 per household effort in an attempt to reduce costs. For most of the annually for the five systems in developing countries. individual systems, of course, all (or greater than 90 This variation is caused by the more elaborate internal percent) of the cost is on-site. Thus, an investigation of plumbing facilities that are found in middle-class Japa- the potential for cost reduction must concentrate on the nese homes and the high cost and relatively large on-site system components and the materials and meth- amount of flushing water required by the Japanese ods used to produce and install them. The case study of systems."6 The investment costs of collection for the Gaborone, Botswana, shows that between 40 and 60 sewerage systems do not fall into any clear groupings percent of the TACH for pit latrines, composting toilets, but vary with the terrain and population density of the Table 3-5. Average Annual On-site, Collection, and Treatment Costs per Household (1978 U.S. dollars) Mean Percentage of total Technology TACH On-site Collection Treatment On-site Collection Treatment Low cost PF toilet 18.7 18.7 100 Pit privy 28.5 28.5 100 Communal toilet 34.0 34.0 ... 100 Vacuum-truck cartage 37.5 16.8 14.0 6.6 45 37 18 Low-cost septic tank 51.6 51.6 . 100 Composting toilet 55.0 47.0 - 8.0 85 .. 15 Bucket cartagea 64.9 32.9 26.0 6.0 51 40 9 Medium cost Sewered aquaprivyv 159.2 89.8 39.2 30.2 56 25 19 Aquaprivy 168.0 168.0 100 Japanese vacuum-truck cartage 187.7 128.0 34.0 26.0 68 18 14 High cost Septic tank 369.2 332.3 25.6 11.3 90 7 3 Sewerage 400.3 201.6 82.8 115.9 50 21 29 * Negligible. a. Per capita costs were used and scaled up by the cross-country average of six persons per household to account for large differences in the number of users. THE ECONOMIC COMPARISON 59 Table 3-6. Annual Sewerage Costs per Household (1978 U.S. dollars) On-site Flushing Collection Treatment City Investment Recurrent water Investment Recurrent Investment Recutrrent Total Kyoto, Japan 166.1 41.3 126.4 88.9 12.1 147.1 59.4 641.3 Hannoh, Japan 146.0 45.5 112.8 58.5 13.3 96.0 89.4 561.5 Higashi Kurume, Japan 153.4 37.6 71.3 36.2 3.9 55.4 42.6 400.4 Khartoum, Sudan 89.2 32.3 174.3 8.5 255.0 12.8 572.1 Managua, Nicaragua 80.8 7.1 10.2 105.1 2.6 89.8 28.2 323.8 Ndola, Zambia 105.8 17.6 153.3 23.5 5.9 5.9 5.9 217.9 Malacca, Malaysiaa 98.9 10.3 34.2 56.9 25.1 8.2 9.3 242.9 Gaborone, Botswana 61.4 11.0 40.9 6.6 16.2 6.1 142.2 Negligible. a. Based on a sewerage master plan. cities. Recurrent costs of collection are uniformly low. the option of improving their individual facilities as In treatment costs there is an expected split between their income permits. those with conventional plants (activated sludge or trickling filter) and those with ponds. The average treatment costs of the former are $175 per household Controlled Comparisons annually, whereas those of the latter are less than $20 per household annually. As mentioned previously, the major disadvantage of The costs from the case studies involving vacuum- conducting comparisons of technology across locations truck cartage are shown in table 3-7. The major is that it is difficult to draw conclusions about how difference between the Japanese systems and the others particular technologies would compare in a given coun- is in the investment cost to the household. The collec- try or site. Analysis of a single location requires tion vehicles used in Japan are more expensive than controlled tests. Fortunately, the present study did those used elsewhere, and labor costs for vehicle opera- include selected cases in which various technologies tion and maintenance are much higher. These two were competing within a small geographic area. These factors, however, are far outweighed by the very large cases still do not yield the sort of precision that could be differential in household facility costs. This has impor- obtained with purely synthetic examples because even tant implications for the upgrading of cartage systems. adjacent areas are not identical. An offsetting advan- As long as the utility provides efficient and hygienic tage of using actual case data, however, is that these vacuum-truck collection, individual households have data do include the imperfections (for example, low use Table 3-7. Annual Vacuum Cartage Costs per Household (1978 U.S. dollars) Costs Population On-site Collection Treatment served On-site Collection Treatment City (thousands) Investment Recurrent Investment Recurrent Investment Recurrent Total Kyoto, Japan 1,462 92.5 18.5 6.1 35.3 15.3 4.1 171.8 Hannoh, Japan 56 113.3 22.7 7.5 43.2 18.7 5.0 210.4 Higashi Kurume, Japan 103 118.2 18.8 3.8 16.5 16.9 18.0 192.2 Tatayama, Japan 57 106.3 20.0 3.3 21.2 8.9 16.8 176.5 Keelong, Taiwan 342 9.6 2.2 1.9 12.2 3.9 2.4 32.2 Tainan, Taiwan 85 9.6 2.2 2.2 15.3 29.2 Pingtung, Taiwan 175 9.6 2.2 0.8 8.1 2.4 2.6 25.7 Chuncheon, Korea 141 20.9 6.6 0.1 5.7 5.0 8.1 46.4 Malacca, Malaysiaa 95 13.5 7.8 3.9 19.9 7.3 1.4 53.8 * Negligible. a. Based on a hypothetical system design. 60 ANALYSIS OF FIELD STUDY RESULTS of capacity or poor equipment maintenance) that might Malacca, a city of about 90,000 in Malaysia.'9 Cur- not be built into a model. Two controlled comparisons rently the city is served by a combination of bucket are discussed below: one between sewerage and cartage latrines, septic tanks, PF toilets, and privies that directly systems in the same cities and one between four on-site overhang the river. The wastes from kitchen and bath technologies and sewerage. are discharged to open, surface water drains. A sewer- age master plan was prepared for the city in 1968, Sewerage versus cartage but-because of lack of money, potential technical Three of the Japanese communities studied were problems stemming from a high water table, and served by both sewerage and vacuum-truck cartage. In community dissatisfaction with the proposed marine Kyoto and Higashi-Kurume, about 45 percent of the outfall-there has been no follow-up implementation of population are connected to public sewers and an equal the study's recommendations. A Malaysian engineer number enjoy vacuum-truck collection. In Hannoh, who was familiar with the local conditions and history n was asked to prepare an alternative master plan to serve nearly 60 percent are served by vacuum trucks, 15ppp percent by sewerage (with the rest using individual the city with vacuum-truck collection and to compare systems). The TACHs of the two systems in the three the costs with those of the sewerage study (adjusted for cities are as follows (TACHs are in U.S. dollars): inflation). The annual costs of waste disposal per household for TACH the two systems in Malacca are shown in table 3-8. Total annual sewerage costs are nearly five times KySto 641.3 171C 8 greater than cartage costs. No social weighting for Higashi-Kurume 400.4 192.2 employment benefits has been used in calculating these Hannoh 561.5 210.4 costs, but the cartage system would employ more than Average 534.4 191.5 twice as many people as the sewerage system, including In Kyoto, where sewerage is especially expensive 100 general laborers (compared with 14 for sewerage). (partly because of the high average incremental cost of water), cartage costs only about one-fourth as much per On-site systems household. In the other two cities, cartage costs are A controlled cost comparison of on-site systems is about half those of sewerage.'7 possible with the results of the study of Gaborone, There is a growing demand on the part of household- Botswana. The International Development Research ers in Kyoto for the sewer system to be extended to Centre sponsored an experimental latrine program in areas presently served by night-soil collection. Because Botswana to build and monitor a variety of on-site the tabulation above shows that sewerage costs nearly designs. The four designs that performed best were four times as much as cartage, it might seem that costed for inclusion in this study. While the costs for all people who can afford it value the increased conve- systems appear high relative to those for similar sys- nience of sewerage at least as much as the difference in tems in other countries, this should not affect the cost. But it is worth repeating that the costs developed relative comparison of technologies. The high costs are in this economic comparison reflect real resource costs to the economy, not financial costs actually charged to households. In Japan, as in many other locations, the Table 3-8. Comparative TA CHs of Sewerage construction costs of sewers and treatment facilities are and Vacuum-truck Cartage in Malacca, Malaysia heavily subsidized by the national government. In (1978 U.S. dollars) addition, the city of Kyoto provides municipal loans at TACH Sewerage Cartage no interest for the installation of a flush toilet and indoor plumbing. In addition, the sewerage authority in On-site Kyoto operates at a substantial loss (based on its Investment 98.9 13.5 sewerage revenues). In fiscal year 1976, subsidies from Recurrent 43.5i 7.8 Collection other city accounts represented 47 percent of the Investment 56.9 3.9 sewerage authority's total revenues. Thus, the financial Recurrent 25.1 19.9 cost of sewerage (and also of cartage) in Kyoto to the Treatment householder significantly understates its true economic Investment 8.2 7.3 cost.'8 Recurrent 9.3 1.4 A more detailed controlled comparison between Total 242.9 53.8 sewerage and vacuum-truck cartage was carried out for a. Includes S34.2 for water used in flushing. THE ECONOMIC COMPARISON 61 because of the pilot nature of the project, the difficulty Benefits from Reuse of obtaining even simple inputs (such as cement) locally, and some overdesign (particularly in the super- As discussed earlier in this chapter, it is assumed in structure) of the systems. this study that the major benefits of sanitation systems The results of the costing analysis for ventilated are related to health and convenience and therefore improved pit (vIP) latrines, aquaprivies, double-vault cannot be meaningfully quantified. Some of the tech- composting (DVC) latrines, and Reed Odorless Earth nologies studied, however, provide economic benefits in Closets (ROECs) in Gaborone are shown in table 3-9. To the form of fertilizer or biogas, to which a monetary enable a better comparison with the cost of sewerage in value can be assigned.2 One of the original aims of this Gaborone, the cost of substructures alone are also study, in fact, was to determine the scope for offsetting shown. On a household basis, the VIP and the ROEC are sanitation costs with reuse benefits. . by far the cheapest. The substructure cost of the DVC Unfortunately, it has been very difficult to locate in latrine is about 80 percent more. The aquaprivy cost developing countries working examples of human waste (substructure) is more than double that of the VIP or disposal systems with a sizable reuse component. A few ROEC, whereas sewerage costs are almost seven times of the sewage treatment systems produce small higher. amounts of methane from their digesters, and this is No analysis of the benefits derived from the compost used for heating. There is some demand from orchard eventually available from the DVC was possible because farmers in Korea for the night soil collected by vacuum all experimental units were recently constructed. De- truck, but the municipality makes no effort to set up a sign parameters suggest that the DVC would require delivery system or to charge a market-clearing price. emptying only at five-year intervals, so that the amount The composting latrines built in Botswana are too new of compost available (per household annually) is likely to yield useful data on reuse. All except one of the to be small. A firm analysis of the benefits from biogas units observed ran on animal, rather than composting cannot be undertaken until more opera- human, waste. In short, although there is much experi- tional experience is available. It is unlikely, however, mental and theoretical information on the economic that such benefits would affect the net cost ranking of potential of reuse technologies, there is a dearth of these alternatives. empirical data on actual experience.2" The conclusions from these three exercises in con- All of the significant reuse technologies found in this trolled costing are in line with those of the cross- study were located in the Far East. Biogas plants were country, comparative analyses of technologies. Sewer- found at the household level on the island of Taiwan age costs are at least twice and generally four to five and in Korea. Municipal systems involving reuse of times as large as those of well-run, vacuum-truck human excreta as an input into agriculture and aqua- cartage systems. On-site technologies can effect even culture were found on the island of Taiwan and, to a larger savings, particularly if superstructure costs can lesser extent, in Indonesia. In none of these cases was be kept low. the reuse element developed to its full potential through marketing analyses or optimal pricing strategies. The cases described below, therefore, should not be taken as examples of how much (or, more accurately, how little) reuse benefits can affect the economics of sanitation. Table 3-9. Comparative TACHs of On-site Systems and Sewerage in Gaborone, Botswana Biogas (1978 U.S. dollars) Annual There were about thirty family-size biogas units in substructure operation in one of the communities on the island of cost per Taiwan studied in 1977. Each biogas unit consisted of a Technology TACH household 6-foot diameter, excavated digester with an inverted Ventilated improved pit (vip) latrine 56.2 21.3 steel lid that floated up and down on a water seal. The Reed Odorless Earth Closet (ROEC) 56.2 21.3 methane generated was transported to the kitchen Double-vault composting (DVC) through a pressure hose connected to the outlet pipe at latrine 74.6 38.6 the top of the inverted lid. The digester was emptied Aquaprivy 87.7 50.6 twice a year, and the sludge was sold to neighboring Sewerage > 142.2a 142.2 farmers. All of the digesters studied ran on a mixture of a. Superstructure costs are included in the house construction. human and animal wastes. The usual input to a unit was 62 ANALYSIS OF FIELD STUDY RESULTS the night soil from five persons and the manure from capital can be made available or less expensive designs five pigs. This input loading produced sufficient gas for developed. The requirement for such large volumes of cooking purposes all year for a family of five, a animal waste is also likely to exclude the lowest income replacement for the 20-kilogram cylinder of liquid classes, who generally do not own animals (or land on petroleum gas formerly purchased each month. The net which to build the digester). Furthermore, that this cost of a typical biogas unit on the island of Taiwan is biogas unit is economically attractive on a household shown in table 3-10. scale does not imply that it would be advantageous on a This net cost does not represent the full cost of the community scale. Aside from the technical questions of total system of hog raising-excreta disposal-biogas the economies of scale of the digester itself, the collec- production. It has been developed within the more tion and transportation of human and animal wastes to limited framework in which the "with and without" a single point and the subsequent redistribution of gas cases are defined as with and without the latrine would involve large capital outlays and operating re- equipped with biogas digester. Thus, the cost of hog quirements that are avoided by a single-family unit feed and upkeep is not included, nor is the benefit from located in the courtyard of the house. the sale of the animals. For purposes of comparison A second case of household biogas production was with costs of other sanitation systems, this is the observed in Korea. There, the family claimed to be appropriate method to employ. For a complete project using only human excreta and kitchen wastes to stock appraisal, however, it would be inadequate. This point the unit. It produced sufficient gas to satisfy all of the was demonstrated on the island of Taiwan by the cooking needs during nine months of the year. Insuffi- marked decline in the number of household biogas units cient cost data were available, however, to permit a in the last five years because of the large increase in the calculation of net cost. price of animal feed, which has made hog raising uneconomic. Agricultural reuse In addition, although the net cost of this sanitation Reuse of composted night soil as fertilizer is also system is attractively low compared with that of the practiced at the household level in rural areas of Korea. other units considered above, its initial investment cost The large size of household pit privies in two rural is very high. In a subsistence economy, large investment villages studied was puzzling until it was discovered costs may present an insurmountable obstacle to the that the farmers deposited the animal wastes from adoption of a low-cost system unless subsidized loan nearby cattle pens into the pits and then allowed the entire quantity to "compost" over a six- to twelve- month period before spreading it on the vegetable fields Table 3-10. Net Cost of Household Biogas Unit, and orchards. Based on the Korean government's im- Island of Taiwan puted cost of such organic fertilizer, the composting (1978 U.S. dollars) operation yielded the farmer an annual net benefit of $37 on an annual cost of $34. These figures do not Item cost cost (years) include any cost for the farmer's time in digging out the latrine and transporting the compost to the field. Cost Nonetheless, they indicate the potential for agricul- Construction, 236.0 31.6 20 tural reuse at the household level in rural areas. Land (for 15 square At the community level, again, very few data are meters) 348.0 41.8 Infinite available. In Chuncheon, Korea, some of the night soil Annual desludgmng 16.6 16.6 collected by vacuum truck is sold to farmers (before TACH 90.0 treatment) for about $7.00 per truck (or $3.50 per cubic meter). At this rate, demand is sufficient to Benefit absorb about half of the night soil produced by the city. Biogas (12 cylinders of Because the demand is seasonal, however, the night-soil liqud petroleum 75.0 - treatment plant, which was designed for peak volumes, Sludge sales (2 carts) 11.0 _ operates inefficiently during the spring and summer months. There is very little net benefit, therefore, to the Net annual cost per city (about 3 percent of the cost on an annual basis) household 4.0 from the sale of untreated night soil, and the health - Not applicable. hazards to the farmer from handling it are probably a. Includes household latrine facilities. considerable. If a simple composting treatment plant THE ECONOMIC COMPARISON 63 had been built instead of the two-stage digester, a ment's equity goals and the degree of distortion in other market for treated night soil might have developed that prices in the economy. This correspondence could be would have minimized sanitation costs to the commu- accomplished with sewerage, for example, by setting a nity while providing a safe and valuable product to surcharge on the water bills of connected consumers nearby farmers. that is equal to the AIC of sewerage per cubic meter of water consumed.22 In the case of most on-site systems, the consumer would pay to construct the original Aquacultural reuse facility (either initially or through a loan at an interest The best case study involving community-scale reuse rate reflecting the opportunity cost of capital) and then was conducted in Tainan on the island of Taiwan. Both pay a periodic sum to cover the facility's operation and public and private night-soil collectors operate there, maintenance expenses (if any). In cases such as these, and untreated night soil is sold primarily to fish the financial cost would be identical to the economic farmers. The private collectors work only during the ten cost except for any taxes and shadow pricing of inputs months of the year in which there is a demand for night that must be purchased in the market.23 To the extent soil. The public system, of course, operates year-round that they account for a significant part of total eco- and is able to sell about 80 percent of its total collection. nomic costs, financial costs may be above or below The public system charges $0.65 per ton plus $0.57 per economic costs. kilometer for transportation costs, whereas the private In deriving financial costs in any particular case, it is collectors charge $7.00 per ton inclusive of transporta- necessary for the analyst to consult with officials of the tion (most trips were less than 10 kilometers). central and local government to determine their finan- No investment or operating costs were available on cial policies and noneconomic objectives. If the govern- the private collectors. The TACH of the public system ment places a high priority on satisfying the basic needs was $28.85, and the sale of night soil during the ten of all of its citizens, then it may be willing to subsidize months of the year yielded $1.28 on a household basis. part or all of the construction costs of a simple sanita- Because the private operators presumably earn a posi- tion system. The general policy of international lending tive income on their operations, the public system must agencies such as the World Bank is that, if the cost of either incur significantly higher costs or charge too the minimal sanitation facility necessary to permit little for its product, or both. adequate health is more than a small part of the household income of the lower income consumer (say, 5-10 percent), then the central or local government should attempt to subsidize its construction to make the Financial Implications facility affordable. If, however, some consumers wish to have better or more convenient facilities, they should The purpose of deriving economic costs is to make a pay the additional cost themselves. Similarly, if more meaningful, least-cost comparison among alternatives. affluent communities decide that, beyond meeting Such a comparison is extremely useful to the planner basic health needs, they wish to safeguard the cleanli- and policymaker. The consumer, however, is much ness of their rivers or general environment by building a more interested in financial costs-that is, what he will more expensive sanitation system, then they should pay be asked to pay for the system and how the payment will for that system either through direct charges to users or be spread over time. The difficulty in developing through general municipal revenues. Because the ma- financial costs is that they are entirely dependent upon jority of the poorest people in most countries live in policy variables that can change dramatically. rural areas, it is usually not appropriate to subsidize Whereas economic costs are based on the physical urban services from national tax revenues. conditions of the community (for example, its abun- Because financial costs are dependent upon policy dance or scarcity of labor, water, and so forth) and are decisions, it is not possible to present comparable therefore quite objective, financial costs are entirely financial costs of the various technologies in the same subject to interest-rate policy, loan maturities, central way that economic costs can be developed. It is, government subsidies, and the like. The financial cost however, possible to use the economic costs to derive of a sewerage system for a community can be zero if the total investment costs per household that will provide a central government has a policy of paying for such basis for the financial comparison of alternatives. The systems out of the general tax fund. other useful figure to be extracted from economic costs To promote the economically efficient allocation of is the annual recurrent cost (with water costs shown resources, financial costs should certainly reflect eco- separately), which will give an indication of periodic nomic costs as closely as possible, given the govern- financial requirements. 64 ANALYSIS OF FIELD STUDY RESULTS The financial requirements for the various low-, lute level is less relevant than their relation to other medium-, and high-cost systems examined are given in recurrent charges. In addition, it was not possible to use table 3-1 1. The first column shows the total investment market prices for water in the various communities cost (including on-site, collection, and treatment facili- because of different charging systems and, for some ties) divided by the number of households to be served. cases, lack of data. For the individual household systems (such as pit It is difficult to draw conclusions from the financial latrines), the investment cost is simply the total cost of requirements because of the reasons of noncomparabil- constructing the facility. For the community systems, it ity mentioned above. Some standard loan terms, how- is the total cost divided by the design population ever, were assumed in order to derive one possible set of (number of households). Thus, for those facilities that financial costs (shown in the fourth column of table 3- exhibit economies of scale, such as sewage treatment 11). Because the length of loans is generally related to plants, this figure will understate the real financial the life of facilities financed, loan periods of five, ten, requirements during most of the early years of opera- and twenty years were used for the low-, medium-, and tion. Note that because investment costs do not reveal high-cost systems, respectively. An 8 percent interest anything about the lifetime of facilities they should not rate (well below the opportunity cost of capital in most be used to make judgments about least-cost alterna- countries) was used as a representative interest charge tives. They are presented only to indicate an order of for loan funds to a utility. Financial affordability can be magnitude of the initial financial expenditure neces- roughly tested by comparing financial costs to house- sary for the various systems. hold income (see the table's fifth column). Average The monthly recurrent cost per household (second income per capita in the low-income countries (where column of the table) is the sum of recurrent costs for on- the bulk of the water and sanitation deficiencies exists) site, collection, and treatment facilities excluding costs was about $180 in 1978. With an average of six persons for water used in flushing, which are presented sepa- per household, this yields a monthly household income rately in the third column of table 3-11. Because water of $90. All of the medium- and high-cost systems have charges vary so much from one community to another monthly costs that amount to over 10 percent of income (both for economic and financial reasons), their abso- and, thus, are probably outside the range of affordabil- Table 3-11. Financial Requirements for Investment and Recurrent Cost per Household (1978 U.S. dollars) Percentage of income Total invest- Monthly recur- Monthly water H,ypothetical total of average low- ment cost rent cost cost monthly cost income householdb Technology (1) (2) (3) (4) (5) Low cost PF toilet 70.7 0.2 0.3 2.0 2 Pit latrine 123.0 - 2.6 3 Communal toilet, 355.2 0.3 0.6 8.3 9 Vacuum-truck cartage 107.3 1.6 3.8 4 Low-cost septic tank 204.5 0.4 0.5 5.2 6 Composting toilet 397.7 0.4 8.7 10 Bucket cartage' 192.2 2.3 5.0 6 Medium cost Sewered aquaprivy 570.4 2.0 0.9 10.0 11 Aquaprivy 1,100.4 0.3 0.2 14.2 16 Japanese vacuum-truck cartage 709.9 5.0 13.8 15 High cost Septic tank 1,645.0 5.9 5.9 25.8 29 Sewerage (design population) 1,478.6 5.1 5.7 23.4 26 Negligible. a. Assumes that investment cost is financed by loans at 8 percent over five years for the low-cost systems. ten years for the medium- cost systems. and twenty years for the high-cost systems. b. Assumes that average annual income is $180 per capita, with six persons in a household. c. Based on per capita costs scaled up to household costs to account for multiple household use in some of the case studies. THE ECONOMIC COMPARISON 65 ity without further subsidy. Sewerage and the Japanese access for rodents and scorpions during the night and and Taiwanese septic tanks have recurrent costs that embarrassed users (whose feet could be seen while they alone are over 10 percent of income, even if initial used the latrine, a matter of sensitivity in this culture). facilities could be provided free of charge by the Such unimportant details from the technical viewpoint government. For most of the other systems, affordabil- are often highly significant if health and aesthetic ity hinges on the arrangements that can be made to benefits (both of which generate a willingness to pay on subsidize investment costs froi.m other revenue sources. the part of the user) are to be fully realized. Such arrangements, however, are probably not replica- A final caution is appropriate for the interpretation ble on a wide scale. Furthermore, average figures of per of the costs developed by these case studies. In very few capita income should not be relied upon without a cases were the systems optimally designed. This point recognition of their limitations in countries in which has already been made with respect to the overdesigned much of the economy is nonmonetized. In addition, superstructure of the Botswana latrines and the reuse many developing countries in Africa and Asia have per components found in the East Asian countries. It is also capita incomes that are less than the average of the low- true of the sewered aquaprivies in Zambia (which fed income group used here, and, in all of the countries, into collectors of conventional size designed for a full more than half of the population earns less than the sewerage system) and most of the other cases. Never- country average. theless, the broad ranking of technologies, the patterns of cost sensitivity, and the method used to arrive at appropriate figures for a least-cost comparison are all Conclusion believed to have general applicability. It may be useful to summarize some broad conclu- sions from this review of cost data. Precise calculations Notes to Chapter 3 of the sensitivity of system costs to changes in particular parameters are impossible to generate within the 1. Variations of this calculation include the internal rate of return framework of an empirically based study such as this. and the net present value. For a discussion of the set of conditions Yet it is possible to discern areas of relatively greater under which each is appropriate, see Lyn Squire and Herman G. van and lesser importance. der Tak, Economic Analysis of Projects (Baltimore: Johns Hopkins The two most outstanding influences on total house- University Press, 1975), pp. 39-43. hold costs are factors that have often been ignored in 2. Ibid engineering analyses: on-site household costs and the 3. H. Shipman and others, "Measurement of the Health Benefits of Investments in Water Supply, " Public Utility Note 20 (Washing- costs of water used for flushing in water-carried sys- ton, D.C.: The World Bank, Transportation, Water, and Telecommu- tems. The former is important in all systems and, in the nications Department, January 1976; processed). cases studied, never accounted for less than 45 percent 4. See Squire and van der Tak, Economic Analysis of Projects, pp. of TACH. The latter is most important for sewerage and 99-132, for a description of the data requirements and methods of septic tank systems. When the economic cost of water is computing conversion factors. high, the payoff from designing systems with low 5. For example, in one Islamic country market interest rates are requirements of flushing water is large.24 set by law at 3 percent, whereas the opportunity cost of capital has been estimated at 16 percent. With such a wide discrepancy, it is very A further implication relates to those aspects of likely that the least-cost alternative using the market discount rate sanitation systems that do not significantly influence would be much more capital intensive than that selected by an costs but can make a big difference in benefits. Two economic least-cost analysis. components of individual systems-ventilation stacks 6. This percentage is based on data from developed countries, and water seals-aid greatly in reducing odors and fly which show that the water used to flush toilets is around 40 percent of breeding without adding significantly to system costs. total domestic water use (excluding garden watering). In one of the Latin American case studies, it was found 7. See R. J. Saunders, J. J. Warford, and P. C. Mann, Alternative Concepts of Marginal Cost for Public Utility Pricing: Problems of that people were very concerned about the color of the Application in the Water Supply Sector, World Bank Staff Working floors of their latrines. Although this preference is an Paper, no. 259 (Washington, D.C.: The World Bank, May 1977). aesthetic matter without technical importance, it may 8. This degree of risk can be explicitly built into the alternative make the difference between a facility that is kept clean selection process. Suppose technology A yields a net present value of and is regularly used and one that is not. In another case I 00 and technology B one of 90, given the demand forecast. There is a the latrine designers, in an effort to cut costs, had used 30 percent probability that the forecast is too high and a 10 percent pr.'ut shes o, z. T probability that it is too low. If it is too high, technology A's net precut sheets of zinc for the superstructure siding. ThIs present value drops to 30 because of its large unused capacity during meant, however, that the siding did not reach all the the early years, whereas technology B can be modified to cut costs so way to the floor-a design flaw that provided easy that its net present value falls only to 70. If demand is too low, A's net 66 ANALYSIS OF FIELD STUDY RESULTS present value falls to 90 and B's falls to 85. The weighted average, or because the opportunity cost of capital does not depend on the source expected value (E), of the net present value of the two technologies of the funds or the terms of a particular loan package. can be calculated as follows: 15. As would be expected, those systems requiring the most EA = 0.3(30) + 0.1(90) + 0.6(100) = 78.0 flushing water are most affected by the change. The recurrent cost A 037)+018)+069) 85.component of sewerage systems drops from 33 to 19 percent, whereas EB 0.3(70) + 0.1(85) + 0.6(90) 83.5. that of septic tanks falls from 36 to 24 percent. Given the uncertainty attached to the demand forecast, technology 16. Twenty liters per flush compared with 8-15 liters in the other B should be selected because the expected value of its net present countries. value is higher than that of technology A. If the demand forecast were 17. In all three cities the night soil from the cartage systems is certain, technology A would have the higher net present value. treated by dilution and transferred to the sewage treatment plant. It 9. The environmental cost of depositing sullage into nearby is likely that cheaper treatment methods could be used in cities watercourses must, of course, also be assesssed. The limited informa- without sewerage systems. tion available on the composition of sullage wastes suggests that its 18. See chapter 6 for a discussion of the effect of such policies on health hazard is low. This should also be assessed, however, for the the choice of appropriate technologies. site in question. 19. We are indebted to Mr. Ng Kin Seng for this comparison. A 10. The development of low-water-use appliances, such as show- summary of his report is included in Kuhlthau (ed.), Country ers, is a very promising means of realizing sanitation cost savings. Studies. Reducing the amount of sullage water to be disposed of not only saves 20. Other potential reclamation benefits include stock and garden water but also extends the range of applicability for on-site disposal watering with sullage and irrigation with sewage. systems.wtrn 21. The obvious exception to this statement is the experience of I1. All costs presented in this chapter are in 1978 prices and U.S. China, but scientific documentation of Chinese experience is rare, dollars. For price and foreign exchange conversion factors for each and it was not possible to include first-hand observation inthisstudy. country studied, see Richard Kuhlthau (ed.), Appropriate Technol- Many data are also available on biogas production in India, but most ogy for Water Supply and Sanitation, vol. 6, Country Studies in units use animal instead of human excreta. Appropriate Sanitation Alternatives (Washington, D.C.: The World 22. Suppose, for example, that the AIC of sewerage is S1.00 per Bank, 1980). cubic meter of sewage collected and treated. Because water rather 12. The anomalies introduced by aggregating across countries are than sewage is metered, this AIC must be related to the water illustrated by a comparison of the TACHs of the pit latrine and PF consumed. If, for a given city, sewage flows are 75 percent of water toilet in table 3-1. The mean cost of the former in the seven cases consumption, then the sewerage surcharge should be $0.75 per cubic studied was higher than that of the latter in its three case studies. Yet meter of water consumed. it is clear that, on any one site, a pit latrine would be cheaper than a PF 23. Note that the shadow price of capital may be reflected in the toilet because of the extra components and water required for the financial cost by using it as the interest rate at which money is loaned latter. to construct facilities. If market rates are lower, however, the 13. The rankings within the groups should not be taken too consumers will presumably borrow the money elsewhere and pay for seriously, however. For example, the mean TACH of sewered aqua- the new facility immediately. Shadow rates for labor and water (the privies is lower than that of nonsewered aquaprivies, but this is other important inputs in this analysis) cannot be incorporated into because of the very high cost of the Sudanese aquaprivy, which is financial costs if the consumer pays for them separately. nonsewered. 24. The savings would be even larger if the improved designs for 14. Note that this would not be an economically efficient solution facilities led to the redesign of water distribution networks. 4 Public Health Aspects IMPROVED COMMUNITY HEALTH is generally consid- developed an environmental classification of excreta- ered the major benefit of improved sanitation. As the related infections that, together with a basic under- discussion in the previous chapter has indicated, how- standing of the epidemiological factors important in ever, it has so far been impossible to determine precisely disease transmission, should enable the planner and how much improvement in health in a given community engineer to maximize the health benefits of whatever can be attributed directly or indirectly to a sanitation technology is chosen. The means of doing so include improvement. Even if a figure for the health improve- both the incorporation of specific features that inhibit ment could be agreed upon (for example, x fewer man- disease transmission in the design of sanitation facili- days of sickness annually), it is very difficult to assign a ties and the supplementation of "hardware" with care- meaningful economic value to it. Much of the illness fully directed educational campaigns. without the sanitation improvement would have been borne by children and others unemployed in the mone- tary sector. The noneconomic value to society of their Water and Health improved health may be equal to that of an employed adult, but the economist has no way of quantifying such Although the primary concern of the present study is a nonmarket value. Moreover, of those man-days of sanitation, the relation between water and health illness incurred by the employed population, some should be kept in mind.2 Water is important to health in (perhaps all) work is probably made up at no cost to two ways: contaminated water or insufficient amounts society during the days following absence because of of water for personal hygiene can be a direct cause of illness. To use an entire daily wage to value saved man- disease; and the disposal of sullage (wastewater or days of illness is almost certainly an overestimate. greywater, see chapter 2, the section "Sullage Dis- These inherent limitations of the health sciences in posal") can, theoretically, serve as a transmission quantifying the effects of environmental changes on vehicle for some kinds of disease. For these reasons, not community disease profiles, and of economics in quan- only poor water quality, but also too little and too much tifying benefits that have no market value, combine to water consumption, present problems. frustrate the measurement of health benefits. Available evidence indicates that most of the health Fortunately, the measurement of benefits is not the benefits from safe water are attainable at service levels primary objective of improved sanitation; achieving the of 30-40 liters per capita daily on site. These service benefits is. If funds are inadequate to build and main- levels will provide protection against the range of tain the elaborate sewerage systems known to provide water-related diseases and are adequate for the all these benefits, then it is essential to choose the personal hygiene that will lead (with health education) alternative technology that will maximize the health to a lowered incidence of diarrheal disease and skin and benefits achieved with the available funds. This effort eye infections. For the latter group, access to water is requires a more precise analysis of the relations be- more important than its microbiological or chemical tween disease and sanitation than has been attempted quality. In addition, concentrations of chemicals in in the past. Toward this end, consultants from the Ross drinking water in developing countries sometimes ex- Institute of Tropical Hygiene of the London School of ceed the published standards or guidelines, which were Hygiene and Tropical Medicine were contracted, as developed in industrial countries. For example, part of this study, to focus specifically on the transmis- groundwater in southern Africa containing several sion process of excreta-related diseases and to investi- hundred milligrams of nitrate per liter is used for gate the relation of the various sanitation technologies domestic supply, even though the concentration is an described in chapter 2 to this process.' They have order of magnitude greater than the 45 milligrams per 67 68 ANALYSIS OF FIELD STUDY RESULTS liter of the World Health Organization (WHO) stan- animals serve as a reservoir, however, are included. dard. This standard was developed in industrial coun- Second, excreta relate to human disease because tries to eliminate the risk of methemoglobinemia ("blue their disposal sometimes encourages the breeding of baby syndrome") in bottle-fed infants, but may be less insects. These insects may be a nuisance in themselves applicable in areas where infants are breastfed. (flies, cockroaches, mosquitoes); they may mechani- The fecal hazard of sullage has yet to be demon- cally transmit excreted pathogens either on their bodies strated. Crude estimates-based on data from the or in their intestinal tracts (cockroaches and flies); or United States and assuming a high value of 150 liters they may be vectors for pathogens that circulate in the per capita daily of sullage-indicate that per capita blood (mosquitoes). discharges of the indicators of bacterial pollution, fecal In considering the transmission of excreted infec- coliforms and fecal streptococci, in sullage are 106 and tions, the distinction between the state of being infected 105 bacteria per day, respectively.3 Corresponding per and the state of being diseased must be kept in mind. capita discharges in feces are approximately 100 for Very often the most important group of the population fecal coli and 109 for fecal streptococci, some four or involved in transmitting an infection shows little or no five orders of magnitude greater than those for sullage. sign of disease; conversely, individuals with advanced This means that, even though ratios of pathogens to states of disease may be of little or no importance in indicators may be higher for sick people than for transmission. A good example occurs in schistosomia- healthy ones, relative risks of infection from night soil sis, where as much as 80 percent of the total output of or sewage are four or five orders of magnitude greater. schistosome eggs in feces and urine reaching water This is consistent with the results of the inquiry into from a human population may be produced by children possible differences in health profiles between people five to fifteen years old. Many of these children will living in areas served by sewers and in adjacent areas show minimal signs of disease; conversely, middle-aged with night-soil collection and sullage discharge to people with terminal disease conditions may produce surface drains (reported in chapter 2). few or no viable eggs. Some concern has been expressed over a possible If an excreted infection is to spread, an infective dose contribution of sullage to increased populations of the of the relevant agent has to pass from the excreta of a Culex pipiens mosquito, which breeds in polluted water case, carrier, or reservoir of infection to the mouth or and is a vector of filariasis. The potential importance of some other portal of entry of a susceptible person. sullage to mosquito breeding is determined by environ- Spread will depend upon the numbers of pathogens mental factors in which low aridity (see maps 7-9, excreted, upon how these numbers change during the chapter 2) and local soil permeability would permit the particular transmission route or life cycle, and upon the water to remain on the surface long enough to permit dose required to infect a new individual. Infective dose mosquito breeding. Where there are extended periods is in turn related to the susceptibility of the new host. of relative drought (see maps 4-6), surface impound- Three critical factors govern the probability that, for a ments of sullage could contribute to extending periods given transmission route, the excreted pathogens from during which mosquitoes normally breed. one host will form an infective dose for another. These In sum, although disposal of large amounts of sullage are latency, persistence, and multiplication. Diagram- resulting from high water service levels may be pro- matically, the concepts can be represented thus: vided by sewerage in densely populated areas, in areas of lower water consumption or lower population density Latency the problem of sullage is one of lower priority. EXCRETED LOAD - Persistence - INFECTIVE DOSE Multiplication There is wide variation in the excreted load of Excreted Infections pathogens passed by an infected person. For instance, a person infected by a small number of nematode worms Excreta are related to human disease in two ways. may be passing a few eggs per gram of feces, whereas a First, the agents of many important infections escape cholera carrier may be excreting more than 106 Vibrio from the body in the excreta and thence eventually cholerae per gram, and a case may pass I0O3vibrios per reach others. These are the excreted infections. In some day. cases the reservoir of infection is almost entirely in Where large numbers of organisms are being passed animals other than man. These are not considered here in the feces they can give rise to high concentrations in because such infections cannot be controlled through sewage. Thus, even in England, where water use is changes in practices of human excreta disposal. A relatively high and salmonellosis relatively rare, raw number of infections for which both man and other sewage may contain 104 salmonellae per liter. At these PUBLIC HEALTH ASPECTS 69 concentrations, removal efficiencies of 99 percent in Among the helminths transmitted by excreta, all the conventional sewage treatment works will still leave 1 02 trematodes infecting man undergo multiplication in pathogenic organisms per liter in the effluent, and the aquatic snails. This introduces a prolonged latent implications of these organisms for health will depend period of a month or more while development is taking upon their ultimate disposal, their ability to survive or place in the snail, followed by an output of up to several multiply, and the infective dose required. thousand larvae into the environment for each egg that Latency is the interval between the excretion of a reached a snail. pathogen and its becoming infective to a new host. In principle, from a knowledge of the output of Some organisms-including all excreted viruses, bacte- pathogens in the excreta of those infected, the mean ria, and protozoa-have no latent period and are infective dose, and the extractive efficiency of the immediately infective when the excreta are passed. The excreta treatment process, simple calculation should requirements for the safe disposal of excreta containing enable one to assess risk. In practice, disease transmis- these agents are far more stringent than for those sion is much less predictable than this because of the helminthic infections in which there is a prolonged variable infective dose of most pathogens and the latent period. In particular, infections that have a uneven distribution of infection in the environment. considerable latent period are largely risk free in areas Whereas the minimal infective dose for some diseases where night soil is being carted by vacuum truck, may be a single organism, or very few, the doses whereas the others constitute a major health hazard in required in most bacterial infections are much higher. fresh night soil. Therefore, in the environmental classi- Data bearing on this are very hard to acquire, since they fication presented below the first two categories (in involve administering a known dose of a pathogen to a which no latency is observed) are separated from the volunteer. Information is scanty and is generally con- remaining categories (in which a definite latent period cerned with doses required to infect, say, half those occurs). exposed, rather than a minute proportion, at a single Persistence, or survival, of the pathogen in the exposure. The volunteers have usually been well-nour- environment is a measure of how quickly it dies after it ished adults from nonendemic areas. Such results have has been passed in the feces. It is the single property to be applied with great caution (if, indeed, they can be most indicative of the fecal hazard, in that a very applied at all) to malnourished children continuously persistent pathogen will create a risk throughout most exposed to infection. treatment processes and during the reuse of excreta. Host response is important in determining the result A pathogen that persists outside the body only for a of an individual's receiving a given dose of an infectious very short time needs to find a new susceptible host agent. In particular, acquired immunity and the rela- rapidly. Hence, transmission cannot follow a long route tion of age to pathology are important for predicting the through sewage works and the final effluent disposal effects of sanitation improvements. In general, the site back to man, but rather will occur within the family balance between exposure to infection and a host's by transfer from one member to another as a conse- response to it will determine the pattern of excreta- quence of poor personal hygiene. More persistent or- related disease. If transmission creating exposure to a ganisms can readily give rise to new cases of disease particular infection is low, then few people will have farther afield, and, as survival increases, so also must encountered the infection and most will be susceptible. concern for the ultimate disposal of the excreta. If a sudden increase in transmission of the disease Though it is easy to measure persistence or viability occurs, it will affect all age groups in epidemic form. of pathogenic organisms by laboratory methods, to Improvements in sanitation will have a significant interpret such results it is necessary to know how many effect under these circumstances by reducing the likeli- pathogens are being shed in the excreta (which is hood of an epidemic and, should one ever occur, its relatively easy to determine) and the infective doses for magnitude. man (which is extremely difficult to discover). By contrast, if transmission is very high the popula- Under some conditions, certain pathogens will multi- tion will be repeatedly exposed to an infection and first ply in the environment. Originally low numbers can be acquire it in childhood. Subsequent exposures may be multiplied to produce a potentially infective dose. without effect if long-lasting immunity is acquired Multiplication can take the form of reproduction by from the first attack. Alternatively, immunity may be bacteria in a favorable environment (for example, cumulative from a series of attacks. The infection will Salmonella on food) or of the multiplication by trema- always be present and is described as endemic. Under tode worms (the parasitic flatworms, including flukes) these conditions much transmission is ineffective be- in their molluscan intermediate hosts. cause of human acquired immunity, and reduced trans- 70 ANALYSIS OF FIELD STUDY RESULTS mission as a result of improved sanitation will only environmental classification presented below distin- delay the date of infection until later in life. Large guishes six categories of excreted pathogens (see also sanitary improvements will either render the infection table 4-2). rare or, if the disease was originally highly transmitted, make it an adult disease. Examples are typhoid, which Category I can be completely prevented in the community by cadequae ompletlympented in excrethe andofmsupplies, These are the infections that have a low infective dose adequate management of excreta and of water (s0)unpwoepahoeslriifctveimditl and poliomyelitis virus infection, which requires ex- (< IO') and whose pathogens are infective immediately treme hygienic precautions to prevent. In practice, on excretion. These infections are spread very easily improved sanitation increases the disease problem by from person to person wherever personal and domestic deferring infection to an age where its clinical course is hygiene are low. Therefore, it is likely that changes in more severe. excreta disposal technology will have little, if any, Consequences of a juvenile age-prevalence are that effect on the incidence of these infections if the changes not only do children suffer chiefly from the diseases, but are unaccompanied by sweeping changes in hygiene, also that they are the main sources of infection, so that which may well require major improvements in water the most important need for better community excret'a supply and housing, as well as major efforts in health disposal is among young children, the group perhaps education. The most important aspect of excreta dis- least inclined to use any facilities that may be available, posal for the control of these infections is the provision Some excreted diseases are infections exclusively or of a hygienic toilet (of any kind) in the home so that almost exclusively of man, but many involve other people have somewhere to deposit their excreta. What animals either as alternatives to man as host or as hosts subsequently happens to the excreta (that is, the means of other stages in the life cycle. In the case in which wild of transport, treatment, and reuse) is of less importance or domestic vertebrate animals act as alternative hosts, because most transmission will occur in the home. control of human excreta is not likely to achieve Although transmission can, and does, occur by complex complete prevention of the infection. Alternatively, if routes, most transmission is directly person to person, the infection under consideration needs an animal host and therefore the provision of hygienic toilets alone will for some intermediate stage, but also requires man, have a negligible effect. However, categories I and II then the control of human excreta can be very effective merge into each other and form a continuum (see in controlling the disease. Some excreted helminthic below). infections that have intermediate aquatic hosts fall into this category. These will be controlled if: excreta are Category 11 prevented from reaching the intermediate host; or the The infections in this category are all bacterial. They intermediate hosts are controlled; or people do not eat have medium or high infective doses (> I lO) and so are the intermediate host uncooked or do not have contact less likely than category I infections to be transmitted with the water in which the intermediate host lives by direct person-to-person contact. Their bacteria are (depending on the particular life cycle). persistent and can multiply, so that even the small numbers remaining a few weeks after excretion can, if they find a suitable substrate (such as food), multiply to Environmental Classification form an infective dose. Person-to-person routes are of Excreted Infections important, but so are other routes with longer environ- mental cycles, such as the contamination of water The list of human pathogens in excreta given in table sources or crops with fecal material. 4-1 is useful only insofar as it shows that the variety of The control measures listed under category I are pathogens is wide and that they are members of one of important-namely water supply, housing, health edu- four groups of organisms: viruses, bacteria, protozoa, cation, and the provision of hygienic latrines-but so and helminths. It is essentially a biological classifica- are waste treatment and reuse practices. Changes in tion. To the sanitation program planner it is interesting, excreta disposal and treatment practices alone may but not very helpful. An environmental classification reduce the incidence of some infections such as cholera that groups excreted pathogens according to common and typhoid but are unlikely to be as effective against transmission characteristics is much more helpful in enteroviral infections, salmonelloses (other than predicting the health effects of sanitation improve- typhoid), and infections from Shigella sonnei, Giardia ments and in understanding the health aspects of lamblia, Enterobius vermicularis, and enteropatho- excreta and sewage treatment and reuse processes. The genic Escherichia coli (these last pathogens are still PUBLIC HEALTH ASPECTS 71 Table 4-1. Excreted Infections Biological group and organism Diseasea Reservoire Viruses Coxsackievirus Various Man Echovirus Various Man Hepatitis A virus Infectious hepatitis Man Poliovirus Poliomyelitis Man Rotavirus Gastroenteritis in children ? Bacteria Campylobacter species Diarrhea in children Animals and man Pathogenic Escherichia coli Gastroenteritis Man Salmonella typhi Typhoid fever Man S. paratyphi Paratyphoid fever Man Other salmonellae Food poisoning Man and animals Shigella species Bacillary dysentery Man Vibrio cholerae Cholera Man Other vibrios Diarrhea Man Yersinia species Yersiniosis Animals and man Protozoa Balantidium coli Mild diarrhea Man and animals Entamoeba histolytica Amebic dysentery and liver abscess Man Giardia lamblia Diarrhea and malabsorption Man Helminths Ancylostoma duodenale Hookworm infection Man-soil-man Ascaris lumbricoides Ascariasis Man-soil-man Clonorchis sinensis Clonorchiasis Animal or man-*snail-fish---man Diphyllobothrium latum Diphyllobothriasis Animal or man-copepod--*fish-man Enterobius vermicularis Enterobiasis Man-*man Fasciola hepatica Fascioliasis Sheep-*snail-aquatic vegetation-man Fasciolopsis buski Fasciolopsiasis Pig or man->snail-aquatic vegetation-man Gastrodiscoides hominis Gastrodiscoidiasis Pig-*snail---aquatic vegetation-man Heterophyes species Heterophyiasis Dog or cat->snail-fish--*man Hymenolepis species Hymenolepiasis Man or rodent-man Metagonimus yokogawai Metagonimiasis Dog or cat--snail--*fish-->man Necator americanus Hookworm infection Man-.soil-man Opisthorchis felineus Opisthorchiasis Animal-*snail-*fish-man 0. viverrini Opisthorchiasis Animal--.snail---fish-man Paragonimus westermani Paragonimiasis Animal or man-snail--*crayfish--man Schistosoma haematobium Schistosomiasis Animal or man-*snail-man S. mansoni Schistosomiasis Man-*snail--->man S. japonicum Schistosomiasis Man-snail-man Strongyloides stercoralis Strongyloidiasis Man or dog(?)-man Taenia saginata Taeniasis Man--.cow-*man T. solium Taeniasis Man-*pig--man. or man--man Trichuris trichiura Trichuriasis Man-soil-man ? Uncertain. a. With all diseases listed, a symptomless human carrier state exists. b. For helminths, the transmission process is given. Source: Richard G. Feachem and others, Sanitation and Disease: Health Aspects of Excreta and Wastewater Management. World Bank Studies in Water Supply and Sanitation no. 3 (Baltimore: Johns Hopkins University Press, forthcoming). commonly transmitted within affluent communities in fragile will clearly tend to be spread in an intrafamilial industrialized countries). or other close pattern and depend for its control more on The criteria chosen to separate categories I and II are personal hygiene and less on sanitation. A low infective infective dose and "length" of the environmental cycle, dose, however, in an environmentally persistent organ- since the aim is to predict the efficacy of sanitation ism will lead to an infection very difficult to control improvements as a control measure. The reason they do either by sanitation or by personal and domestic hy- not form tidy groups is the variable persistence of the giene. Many viruses fall into this category and pose pathogens involved. The extreme category-I pathogen major problems of control. For them, induced immu- that has a low infective dose and is environmentally nity may be the best approach, as discussed above for 72 ANALYSIS OF FIELD STUDY RESULTS Table 4-2. Environmental Classification of Excreted Infections Category and epidemiological Environmental feature Disease transmission focus Major control measure 1. Nonlatent; low Amebiasis Personal Domestic water supply infective dose Balantidiasis Domestic Health education Enterobiasis Improved housing Enteroviral infection, Provision of toilets Giardiasis Hymenolepiasis Infectious hepatitis Rotaviral infection t. Nonlatent; medium Campylobacter Personal Domestic water supply or high infective infection Domestic Health education dose; moderately Cholera Water Improved housing persistent; able Pathogenic Crops Provision of toilets to multiply Escherichia Treatment of exereta coli infection before discharge or reuse Salmonellosis Shigellosis Typhoid Yersiniosis In. Latent and persistent; Ascariasis Yard Provision of toilets no intermediate host Hookworm infectionh Field Treatment of excreta Strongyloidiasis Crops before land application Trichuriasis iv. Latent and persistent; Taeniasis Yard Provision of toilets cow or pig as Field Treatment of excreta intermediate host Fodder before land application Cooking. meat inspection v. Latent and persistent; Clonorchiasis Water Provision of toilets aquatic intermediate Diphyllobothriasis Treatment of excreta host(s) Fascioliasis before discharge Fasciolopsiasis Control of animal Gastrodiscoidiasis reservoirs Heterophyiasis Cooking Metagonimiasis Paragonimiasis Schistosomiasis vt. Excreta-related Bancroftian filariasis Various fecally Identification and insect vectors (transmitted by C(ulex pipiens) contaminated sites in elimination of suitable and all infections in which insects breed insect breeding sites i-v for which flies and cockroaches can be vectors' Source: Feachem and others, Sanitation and Disease. a. Includes polio-, echo-, and coxsackieviral infections: poliomyelitis; viral meningitis; diarrheal, respiratory, and other diseases (see Feachem and others, chapter 1). b. Ancylostoma duodenale and Necator americanus. c. Culex pipiens is a complex of mosquito species and subspecies. The principal tropical species, and the vector of filariasis in those tropical areas where the infection is transmitted by Culex. is Culex quinquefasciatus (previously also known as Culex pipiens fatigans, C. p. quinquefasciatus, or C. fatigans). poliomyelitis. For category ii, sanitation improvements hygiene because the helminth eggs are not immediately reduce the efficacy of the longer cycles and thus have a infective to man. Domestic hygiene is relevant only greater overall benefit than for category I pathogens insofar as food preparation must be adequate to destroy (for which these longer cycles are of little significance). any infective stages present on food, and latrines must Category III be maintained in a tolerable state so that eggs do not remain on the surroundings for the days or weeks of This category contains the soil-transmitted hel- their latent period. If ova are not deposited on soil or minths. They are both latent and persistent. Their other suitable sites for their development, transmission transmission has little or nothing to do with personal will not occur. Therefore, any kind of latrine that PUBLIC HEALTH ASPECTS 73 contains or removes excreta and does not permit the Category VI contamination of the floor, yard, or fields will limit This category is reserved for excreted infections that transmission. Because persistence of ova is so long, are, or can be, spread by excreta-related insect vectors. however, it is not sufficient to stop fresh feces from The or tant ad by tese vectors reaching the yard or fields. Any fecal product that has The most important and ubiquitous of these vectors are not eenadeuatly reaed ustnotreah te sil, mosquitoes, flies, and cockroaches. Among the mosqui- not been adequately treated muse t rea o the toes there is one cosmopolitan group, Culex pipiens, Tberefore, in societies thatreustheirexcretaonthe that preferentially breeds in highly contaminated water land, effective treatment (for example, storage of ex- and is medically important as a vector of the worms that creta for at least a year) is vital prior to reuse. auS filarai Impothanttws aroupor ofith wormsotha cause filariasis. The other two groups, flies and cock- roaches, proliferate wherever feces are exposed. Both Category IV have been shown to carry large numbers and a wide This category contains only Taenia saginata and T. variety of excreted pathogens on their feet and in their solium, the beef and porktapeworms, respectively. Any intestinal tracts, but their importance in actually system that prevents untreated excreta from being spreading disease from person to person is, in fact, eaten by cattle and pigs will control transmission of controversial (though their nuisance value is great). these infections. Cattle are likely to be infected in fields Flies have also been implicated in the spread of eye treated with sewage sludge or effluent. They may also infections and skin lesions. eat feces deposited in cowsheds. Pigs are likely to The implicit control measure is to prevent access of become infected eating human feces deposited near the the insects to excreta, and this can be achieved by many home or in the pigsty. sanitation improvements of differing sophistication. In Therefore, the provision of toilets of any kind to general, the simpler the facility, the more care is needed which cattle and pigs do not have access, and the to maintain it insect-free. Cockroaches, flies, and Cu- treatment of all wastes prior to land application, are the lex mosquitoes have numerous breeding places other necessary control methods. It is also necessary to than those connected with excreta disposal and, thus, prevent birds, especially gulls, from feeding on trickling can never be controlled by sanitation improvements filters and sludge drying beds and subsequently depos- alone. iting the tapeworm ova in their droppings on pastures. Personal and domestic cleanliness are irrelevant as long Health Effects of Treatment as toilets are used. and Reclamation C.ategory V Category V As described above, some of the infections in cate- These are the water-based helminths, which need an gories Il-V require for their control proper treatment aquatic host or hosts to complete their life cycles. before disposal or reclamation. Waste treatment tech- Control is achieved by preventing untreated night soil nologies for developing countries depend upon the or sewage from reaching water in which the intermedi- level of water service and the kind of sanitation sys- ate hosts live. Thus, any land application system or any tem involved. The health aspects of three treatment dry composting system will reduce transmission. There options-stabilization ponds for waterborne wastes, are two complications. First, in all cases (except Schis- night-soil digestion with or without methane (biogas) tosoma mansoni and S. haematobium), animals are an recovery, and composting-may be evaluated accord- important reservoir of infection. Therefore, any control ing to the time-temperature relations that achieve the measures restricted to human excreta can have only a death of excreta-related pathogens. partial effect. Second, in the case of S. haematobium, it Minimal times and temperatures that will ensure is the disposal of urine that is of importance, and this is pathogen death are shown in figure 4-1. The most far more difficult to control than the disposal of feces. resistant pathogens are enteric viruses and Ascaris Because multiplication takes place in the intermediate eggs; by the time these are killed, all the others have hosts (except in the case of the fish tapeworm, Diphyl- died. The curve for Ascaris eggs is based upon a large lobothrium latum), one egg can give rise to many body of data; that for the viruses is less certain. In any infective larvae. A thousandfold multiplication is not event, the typical temperatures reached during aerobic uncommon. Therefore, effective transmission may be composting by the Beltsville Agricultural Research maintained at low contamination levels, and the re- Center (BARC) process described in chapter 2 are more quirements of adequate excreta disposal relative to the than enough to destroy all known pathogens.4 percentage of all feces reaching the toilet are very Figure 4-1 also indicates that anaerobic night-soil or exacting. sludge digestion-at the ambient or slightly raised 74 ANALYSIS OF FIELD STUDY RESULTS Figure 4-1. Influence of Time and Temperature on Selected Pathogens in Night Soil and Sludge 70 - 70 Enteric viruses 65 - 65 Shigella 60 Safety zone -60 55 Taenia \ ^< \5 55 55 50 - 5 Vibrio cholerae 45 -.--------45 40 Ascaris 40 Salmonella 35 _ 35 N:Z 30 _ 30 25 : Entamoeba 25 1 histolvtica 20 l l 20 0.1 1 10 100 I1,000 Ii0.000 1 day 1 week 1 month 1 year Time (hours) Note: The lines represent conservative upper boundaries for pathogen death-that is. estimates of the time-temperature combinations required for pathogen inactivation. A treatment process with time-temperature effects falling within the "safety zone" should be lethal to all excreted pathogens (with the possible exception of hepatitis A virus-not included in the enteric viruses in the figure-at short retention times). Indicated time-temperature requirements are at least: 1 hour at ¢621C. I day at ¢500C. and 1 week at a460C. Source: Richard G. Feachem and others, Sanitation and Disease: Health Aspects of Excreta and Wastewater Management, World Bank Studies in Water Supply and Sanitation, no. 3 (Baltimore: Johns Hopkins University Press, forthcoming). PUBLIC HEALTH ASPECTS 75 temperatures found in night-soil storage pits or vaults Other reports of possible infection due to aerosols from (say, up to 35°C in tropical areas) for detention periods spray irrigation in Israel, and to ridge and furrow of twenty to thirty days-will substantially reduce but irrigation of crops with poorly treated sewage in a not eliminate Ascaris eggs in the sludge.5 For digesters number of places, reaffirm the need for careful selec- heated to 45 or 50 °C, complete destruction will occur. tion and operation of waste treatment facilities that will Storage in a well-drained pit for one year will also adequately protect nonimmune human populations." suffice for an essentially complete kill; the same is true Where excreta are fed to fish or pigs-as in South and for excreta in a pit privy or a composting latrine. Southeast Asia, West Africa, and Central Ameri- If pathogens are not removed by prior treatment, ca-waste treatment should be complemented by care- they can survive on soil as follows: ful cooking of the meat. Methane from a household months Survival time biogas unit or a community digester has no hazard of Viruses <6 months, but generally <3 months infection, but the sludge or slurry will require the same Bacteria < 3 years, but generally <2 months treatment as that for night soil or sewage treatment Protozoa < 10 days, but generally <2 days plant sludge. Helminths < 7 years, but generally <2 years Risks of infection from eating foods grown with water or fertilizer from raw or treated sewage, sludge, Survival of excreted pathogens on crop surfaces may feces, or urine depend upon the kind of crop and be as follows:6 whether it is eaten raw, upon handling of the food Survival time before and after cooking, and upon the time-tempera- Viruses MM< 2 months, butgenerally 50 liters per subsequent se wered PF system is not capita daily) to enable a sewered pour-flush (PF) appropriate, the choice lies among the various on-site system to function satisfactorily, a sewered PF system excreta disposal technologies with appropriate facili- can be used provided that: it is cheaper than ties for the disposal of sullage (see the selection alternative systems with separate sullage disposal process for these in figure 9-2). facilities or the users, or the municipality are willing to If double-vault composting (DVC) toilets and three- pay the extra cost, and there is no overriding social stage septic tank systems cannot be used, the choice preference for night soil to be collected separately for lies between ventilated improved pit (VIP) latrines, PROJECT DEVELOPMENT 101 Reed Odorless Earth Closets (ROECs), PF toilets, vault socioeconomic status). Such feasible sequences, or toilets, and communal sanitation blocks as determined stages for upgrading, are summarized in figure 9-4. by the algorithm in figure 9-3. The advantages and disadvantages of these technologies and their applica- DVC toilets and three-stage septic tanks bility under different conditions have been discussed in chapter 2. These toilets, functioning well and with a continuing Once the most appropriate technology has been demand for compost or fertilizer, need no upgrading. selected by using the algorithm, several questions Upgrading of the water supply from hand-carried to should be asked as checks. These are: household service, increased housing density, or decreased demand for compost would, however, . Can the existing sanitation system (if any) be require modifications in these facilities. The toilets upgraded in any better way than that suggested could be easily modified to PF vault toilets or to vaults by the algorithm? with vacuum-truck collection. * Is the proposed technology socially acceptable? Is it compatible with cultural and religious require- ments? Can it be maintained by the user and, if VIP latrines and ROECs appropriate, by the municipality? Are municipal Many rural and suburban water and sanitation support services (for example, education and projects provide pit latrines and communal hand- inspection) required? Can they be made avail- pumps or public standpipes as the initial improvement. able? The pit latrine should be either a VIP latrine or ROEC. . Is the technology politically acceptable? The subsequent priority for improvement would most *Are the consumers willing to pay the full cost of likely be upgrading the water supply to yard taps (or the proposed technology? If not, are user subsidies household handpumps where applicable). Both the ViP (direct grants or "soft" loans) available? Is latrines and ROECs could then be upgraded to PF foreign exchange required? toilets. The conversion of a ROEC to a PF toilet is very . What is the expected upgrading sequence? What simple and inexpensive: a water-seal squatting plate or period of time is involved? Is it compatible with pedestal seat is installed in place of the ROEC chute, current housing and water development plans? and the existing displaced pit is used to receive the Are more costly technologies in the upgrading flush water. Depending on soil conditions it may be sequence affordable and desired now? necessary to enlarge the pit to provide more infiltra- . What facilities exist to produce the hardware tion area for the flush water. Alternatively, a second required for the technology? If lacking, can they pit or infiltration trench could be provided to receive be developed? Are the necessary raw materials the settled flush water from the original pit. locally available? Can self-help labor be used? A VIP latrine can also be converted to a PF toilet by Are training programs required? filling in the pit with soil and installing a water-seal * If the technology cannot dispose of sullage, can unit that is connected to a newly dug pit. Clearly, this adequate facilities for sullage disposal be in- is best done when the pit is close to the end of its life, stalled? Is the amount of sullage low enough (or and it is most advantageous when the superstructure could it be reduced) to avoid the need for sullage cannot easily be dismantled (for example, the super- disposal facilities? structure is constructed in concrete block or adobe brick). Sanitation Sequences PF toilets The selection of the technology best suited to effect When the water supply is upgraded to house initial improvements in sanitation for a particular area connections, it is possible to install a low-volume, has been discussed in the preceding section. This cistern-flush toilet. This is not essential and may not selection should, however, also reflect the future need be considered a priority by the users, to whom for improvements as the users' aspirations and upgrading of the water supply from a single yard tap socioeconomic status rise. The following subsections to multiple house connections usually first means examine the feasibility of upgrading sanitation in plumbing for kitchens and bathing areas. The main stages that take into account incremental improve- improvement required is better sullage disposal that ments in the level of water supply service (improve- does not have to be via sewers. One such conversion ments that are themselves, of course, measures of requires: Figure 9-2. Second-stage Algorithm for Selection of Sanitation Technology Start Is there art LtvvunetTd u5s Yes Is r-euse of liquidI prcefcrred Yes Is sutiffcicnt Water Ystre-ag Yes for coipost or st a ilizcva Are septic Three-sta bun-ms by houischold or oveeretae ofcmoseovilabt? frP:tanks affordable? septic tanks others'? xrt'olt No iNo No No ts sufficient=organic waste Yes Can doubic-vault Yes eAre DV toilet Ys K) ~ ~ ~ ~ ~ ~ ~~~~~~~~~~~~~ssfiintogii at0compostinig (u)vc) toilets Ar vOolt VO material or ash available? be expected to be affordable? toilets well mnaintaiined? |No No |No Arc ventilatedYe improved double- esIDP pit (vioP) latrincs latrines affordable? No No F Go to third-stagc algorithim Figure 9-3. Third-stage Algorithm for Selection of Sanitation Technology Start iiiorc than I mctcr Is soil sufficientlv clcansing materials ~~~~~~~Affc Pr tiletsiet s | ~~~~~Are Reed Oedorless |Yes ||Yes be raised? _ peferred over vip affoirdable? RF' No Areavu,riatrineseYe''H, >_ | No ANo b t ~~~~~~~~~~~ ~~~~ ~~~~~~~~~~~~~~~~~~~~~~~~~attordabj^ |t littrilie, jP No Commun.t sanitatiofn No facilities Is there sulihcient ,palce tor aI perm.lnent Yes Are V'IDP Ycs bVD doutlic-pit systern 0 latrines l atrne with a minipilum of I affordable'? air storage per tilultl' No Is thcrc either :a Yesc r al olt YeVal nmunicipid or pri- . Ar al eies sVul %atle s\stcni for affordabic' ? olodets cruipt\ing laltrines? No No Communal sanitulion facilities 104 PROGRAM PLANNING AND DEVELOPMENT Figure 9-4. Potential Sanitation Sequences Level of water service Satlitation techInolog,y Hand- Yard ratp or House carried household pump connection Composting toilets Double-vault Vaults Septic tank (Unlikelyl) Vault and vacuum truck (Unlikely) Improved pit latrines vip latrinc and VIDP latrine (Unlikely) ROECi (Unlikely) PF toilct +>O+<: Sewerage Small-bore se we red PF toilet Q Conventional sewerage or septic tank Q ) O lTechnically feasibic. * Feasible if sufficient pour-flush uxzater will he hand carried o Tcchnicallv infeasible. O Feasible if total wastewsater how exceeds 511 liters per capita daily. PROJECT DEVELOPMENT 105 * Construction of a small, single-chamber septic for ten years, after which either the next stage will be tank close to the existing PF pit and discharge of added or the existing facility will be replaced or the sullage directly into it (the tank should provide repaired. In addition, the schemes described can be twelve hours of retention time, subject to a varied substantially without adding greatly to the cost. minimum working volume of 0.5 cubic meters) For example, to a standard pit privy with a PF, a vault * Connection of the existing PF pit to the sullage could be added if housing density increases or the soil tank with pipe 100 millimeters in diameter (the becomes clogged. Similarly, a composting toilet that pit outlet "tee" junction should be located as near already has a watertight vault could be converted into the top of the pit as technically feasible) an aquaprivy or PF toilet with a vault. * Connection of the sullage tank to the street sewer As shown in figure 9-5, the initial sanitation facility (the invert of the tank outlet should be a nominal (a) would consist of a VIP latrine with a concrete 3 centimeters below that of the inlet from the PF squatting slab and concrete block superstructure. An pit to prevent sullage from flowing into the PF pit) actual facility of this kind in an East African city was or to the soakaway. used as the basis for the costs shown. Its unlined pit is If the existing PF pit has sufficient infiltration about 5.5 meters deep and I meter square, and the capacity, there will be little or no flow from the pit to normal filling time is ten years. Its initial construction the sullage tank. This does not matter-in fact, it will cost is $108, of which the superstructure accounts for improve the functioning of the soakaway. But as the infiltration capacity falls-and especially if cistern- In year 11 the community water system is upgraded from wells or standpipes to yard hydrants, and the dry fhe ptile are istalledted fo wi ease, and latrine is converted to a PF latrine (b) by digging a new the it wll at asa seled r seiseaed frst soakage pit near the superstructure and replacing the compartment of a two-stage septic tank. It is essential old squatting plate with a bowl and inverted siphon. that the sullage tank (the second compartment of the two-stage septic tank) is provided; otherwise, the The old pit is filled in prior to placement of the new small-bore flat sewers (or the soakaway) will become squatting plate. For costing purposes, it is assumed blocked. that the accumulated sludge would be removed from the new pit at five-year intervals and composted.3 The costs of trucks, land, and equipment for the compost- Vault toilets ing facility are therefore included in year 15, and the The system ofvalttiletsandacuumttrucks are replaced at five-year intervals thereafter. Thed system cofmvalt tilets andas. vecacuu trevaucs i The operating and maintenance costs incurred in years usedmos comonl in rba aras. ecase te vult 11-20 also included the cost of water for flushing the satisfactorily stores the excreta and PF water and has a PF latrine, which was calculated as 10 liters per capita water seal, no upgrading is necessary for excreta disposal. As the water supply service improves to a daily for six persons at $0.35 per cubic meter. house c c ,In year 21 the third stage of scheme I would begin, house~ ~ conetin hoee,swr.rohrsial when the water service is upgraded to house connec- arrangements for sullage disposal may be desirable. If tien and a larvoue of upgage te has tonbe sewers are installed, the vault toilet may be readily do converted to a sewered PF toilet by connecting the isposed of. At this point a new lined pit would be dug vault to the sewer system as described above. and the existing bowl and siphon would be connected to it. An overflow pipe would connect the pit to a newly constructed small-bore sewer system (c). This upgrading would permit the use of cistern-flush toilets Sample Staged Solutions if desired by the users. Annual collection of sludge would be required from the smaller vault, and a To demonstrate the feasibility of using a staged trickling filter plant would be constructed for treat- sanitation system, four possible schemes or variations ment of the effluent.' The combined flushing water are illustrated in figure 9-5, and comparative eco- and sullage flow from year 21 onwards is taken to be nomic costs are presented for each. The last scheme, 175 liters per capita daily. installation of a sewerage system with no preceding Comparative total economic costs, on a household stages, is obviously not a sequence but is included in basis, were prepared for this scheme and for the three the figure as a reference. Schemes 1-3 can be started variations-including the alternative of proceeding with any stage and terminated at any point, depending immediately with the construction of a sewerage on the desires of the users. For simplicity, it is assumed system (scheme 4).5 The total economic cost per that each stage within a scheme will remain in service household of the three stages of scheme I over a thirty- 106 PROGRAM PLANNING AND DEVELOPMENT Figure 9-5. Sample Sanritation Sequences (costs in 1978 U.S. dollars) Total present value Item Year I Year 10 Year 20 / Year 30 economic cost per household over 30-year period a .b r Scheme I Construction cost 108 65 9(5 354 a c Scheme 2 Construction cost 108 915 1 . lI Scheme 3 Construction cost 96(l 1.519 d Scheme 4 Construction cost 978 3.00i0 a VIP latrine. b PF toilet with soakawav. C PF toilet with small-bore sewer (with optional bowl and seat). d Conventional sewerage. PROJECT DEVELOPMENT 107 year period (in 1978 prices) is $354, which includes sewerage. This is not because conventional sewerage the salvage value of the sewerage system (assumed to systems should not be built (they are an excellent form have a forty-year life). The second scheme shown in of sanitation for those who can afford them and who figure 9-5 moves directly from the VIP latrine have plenty of water), but because they are not (installed in year 1) to small-bore sewers (in year 11). necessary to provide a high standard of sanitation. The The total cost per household over thirty years for sewered PF system (which can include a low-volume, scheme 2 is $1,1 1 1, or more than three times that of cistern-flush toilet for users' convenience) yields an the preceding three-stage alternative. The third equally high standard of service and has two major alternative (scheme 3) is simply the installation of a advantages over conventional sewerage: it is substan- small-bore sewerage system in year 1. This would have tially cheaper, and it can be reached by the staged a total cost of $1,519 per household over thirty years. improvement of several different sanitation technolo- The final alternative (scheme 4), for comparison gies. Thus, planners of sanitation programs can calculated in the same way and with data from the confidently select one of the low-cost technologies in same city as the small-bore sewerage system, is the the knowledge that, as socioeconomic status and immediate construction of a conventional sewerage sullage flows increase, it can be upgraded in a system. A construction period of five years is assumed, predetermined sequence of incremental improvements and the facility is assumed to be two-thirds utilized to an ultimate level of desired convenience. The upon completion and fully utilized ten years after important fact for concerned planners to remember is completion. Based on these assumptions, the total cost that sewers are required to dispose of sullage, not per household over thirty years is $3,000, which excreta, and that the elimination or reduction of includes the cost of water for flushing and all regular nonessential water use is thus the crucial element of an operating and maintenance costs (as do the costs of the economic solution to sanitation problems. This is other alternatives). It is nearly ten times as high as the particularly significant in developing countries, where cost of the three-stage scheme I and almost twice that the increasing competition for investment funds often of the one-stage sewered PF alternative shown as limits the amount of resources that can be allocated to scheme 3. the water and sanitation sector. Another alternative to these upgrading schemes would be to move from the VIP latrine to a vault toilet with vacuum-truck collection in year 11. Based on costs from such a system in a city on the island of Taiwan, the total cost per household over thirty years Notes to Chapter 9 would be $334. If in year 21 it was decided to convert from vacuum collection to a small-bore sewerage system (as described in the previous schemes), the 1. See Donald T. Lauria, Peter J. Kolsky, and Richard N. total cost would increase to $411 per household. These Middleton, Appropriate Technologyfor Water Supply and Sanita- costsare smmarzed i thefolloing abulaion:tion. vol. 9. Low-Cost Design of Water Distribution Systems costs are summarized in the following tabulation: (Washington, D.C.: The World Bank, 1980). Total economic cost 2. Sites-and-services projects generally provide streets, water per household supply, sanitation, and other basic infrastructure for an urban area; (U.S. dollars) loan funds are made available to potential residents who build their vip-vault collection 334 own houses. vip-vault collection-small-bore sewer 411 3. In some communities, sludge may be buried rather than VIP Pr-small-bore sewer 354 composted. vi|p-small-bore sewer 1 ,1 11 Small-bore sewer 1,519 4. This option is chosen for illustrative purposes because of Conventional sewerage 3,000 available cost data from the same East African city. 5. This is the present value (assuming an opportunity cost of As is shown in figure 9-5, none of the upgrading capital of 10 percent) of the thirty-year investment and mainte- sequences discussed above leads to conventional nance cost streams. 10 A Concluding Note IN A SPECIAL SESSION on November 10, 1980, the reexamination of technologies must, therefore, include United Nations General Assembly declared the 1980s a review of their development, and efforts must be the International Drinking Water Supply and Sanita- made to identify those traditional technologies which tion Decade. The objectives of the Decade-as are no longer, or only rarely, used in industrialized promulgated by the nations participating in the countries but which could be improved or adapted United Nations Water Conference in Mar del Plata, elsewhere to provide the health benefits of sanitary Argentina, in April 1977-are to provide an adequate disposal of waste at significantly reduced cost. supply of safe water and facilities for the sanitary This study has identified and evaluated traditional disposal of waste for all by 1990, if possible, or to technologies found in a variety of communities and reach such goals as governments consider feasible. At countries around the world. Costs and benefits have the General Assembly meeting, governments also been assessed and improvements suggested. A method submitted national plans for the Decade that indicated of sequential improvement of sanitation has been targets to be reached and described actions to be taken developed to permit a gradual increase in the level of during the Decade. convenience from the pit privy to a flush toilet in steps If the Decade is one of action, rather than words that keep pace with the users' ability to pay for them. -and preparatory activities since 1977 by both Incremental costs are low because each step in the governments and international organizations active in sequence makes use of previously built facilities. The the sector suggest that serious efforts will be made to study further examines institutional and engineering make the Decade a success-significant improve- aspects of sanitation systems and has provided detailed ments can be effected in the provision of water supply recommendations on how to evaluate sanitation needs, and sanitation services to the inhabitants of developing design and implement projects, and organize the countries. The success of the Decade will depend necessary institutional and community support. primarily on finding effective ways to bring service to In demonstrating the feasibility of using low-cost the poor. Women and children, in particular, suffer technologies appropriate to the conditions in large and most from a lack of these services because they can small communities of developing countries, this ap- least afford them, nor can they afford to protect praisal can play a significant role in the implementa- themselves from infection or to seek a cure for tion of the International Drinking Water Supply and sanitation-related disease. Sanitation Decade. Similarly, companion volumes The design of water supply and sanitation services addressing specific topics or reporting case studies affordable by the poor initially requires a reexamina- provide the planner, engineer, and community worker tion of existing technologies that emphasize conve- with detailed information on the design, implementa- nience rather than health benefits. This emphasis on tion, and implications for health of sanitation projects. convenience is not surprising, given that the greatest (See the list of publications in the World Bank series progress in developing sanitation systems in the past, a on water supply and sanitation given in the preface.) topic examined in this appraisal, occurred in Great Nevertheless, these publications are but a beginning in Britain and the United States at a time when the the process of designing appropriate solutions to the economies of these countries were most productive and pressing needs of the world's poor for water supply and their populations could afford to pay for "luxury" sanitation. The next steps are equally important: sanitation. Consequently, it also is not surprising to find that these systems are too expensive for universal . Economic planners and officials responsible for use in the developing world, where scarce resources the allocation of international funds must be in- must be carefully allocated to satisfy basic needs. The formed of the availability of appropriate technol- 108 A CONCLUDING NOTE 109 ogies that permit the provision of services to many traditional technologies, the adaptation of ad- more people for the same cost as conventional vanced technologies, and the development of technologies provide services to fewer users. health education techniques so that water supply * Engineers must learn to use these technologies and sanitation services may be extended at lowest and seek the participation of behavioral scientists possible cost. For example, upflow anaerobic and health educators to help in the design of filters may extend on-site disposal, or sullage projects fully responsive to the needs of the users water discharge to drains, to many areas hereto- and in the implementation of such projects with fore considered unsuitable for the on-site disposal the affected communities' participation. of wastewater. * Master plans for water supply and sanitation should provide service standards (and technolo- The list above is, of course, incomplete. But, given gies) that the different groups in the community imagination and the courage to examine and recom- can afford and that include the possibility of mend the unconventional, low-cost solutions to prob- adding future sequential improvements. Profes- lems of water supply and sanitation will be found. This sionals developing water plans and projects should study represents an initial step in this creative process. be recompensed for the work to be done rather It is the hope of the authors that it will stimulate than on the basis of the cost of their proposed others to join the effort and, thus, to ensure the solutions. International Drinking Water Supply and Sanitation * Work must continue on the improvement of Decade's success. Bibliography THE WORD "PROCESSED" describes works that are repro- Downing, Paul B. The Economics of Urban Sewage Dis- duced by mimeography, xerography, or by any means other posal. New York: Praeger, 1969. than conventional typesetting and printing; such works may Elmendorf, Mary, and Patricia K. Buckles. Appropriate not be cataloged or commonly available through libraries. Technology for Water Supply and Sanitation, vol. 5, American Society of Civil Engineers. Wastewater Treat- Sociocultural Aspects of Water Supply and Excreta ment Plant Design. Manual of Practice, no. 36. New Disposal. Washington, D.C.: The World Bank, 1980. York: American Society of Civil Engineers, 1977. Elmendorf, Mary (ed.). Appropriate Technology for Water Anders, W. "The Berlin Sewer Workers." Zeitschrift fuer Supply and Sanitation, vol. 8, Seven Case Studies of Hygiene (1954), pp. 341-71. Rural and Urban Fringe Areas in Latin America. Wash- Appleton, B. "Acid Test for Middle East Brain Drains." ington, D.C.: The World Bank, 1980. New Civil Engineer (February 19, 1976), pp. 23-29. Eckaus, R. S. "The Factor Proportions Problems in Under- .. . . ~~~~~~developed Areas." American Economic Review, vol. 45 Arlosoroff, Saul. "Tradition and Innovation in Water Use (Sepember 5. 539m654 and Reclamation." In Gunnerson and Kalbermatten (September 1955), pp. 539-65. (eds.), Appropriate Technology in Water Supply and Fair, Gordon M., John C. Geyer, and Daniel A. Okun. Waste Disposal, pp. 61-84. Water Purification and Wastewater: Treatment and Dis- Assar, M. Guide to Sanitation in Natural Disasters. Ge- posal. New York: John Wiley and Sons, 1968. neva: World Health Organization, 1971. Feachem, Richard G., and A. M. Cairncross. Small Excreta Disposal Systems. Ross Bulletin, no. 8. London: Ross Bardach, J. E., J. H. Ryther, and W. O. McLarney. Aqua- Institute Information and Advisory Service, Ross Institute culture: The Farming and Husbandry of Freshwater and of Tropical Hygiene, 1978. Marine Organisms. New York: John Wiley and Sons, ofe roicha Hg, D.D 1972. Feachem, Richard G., D. Duncan Mara, and Kenneth O. Iwugo. Appropriate Technology for Water Supply and Barnet, A., L. Pyle, and S. K. Subramanian. Biogas Tech- Sanitationrvol.a7, Aternative anitation echnologie nology in the Third World. Ottawa: International Devel- Sanitation, vol. 7, Alternative Sanitation Technologies opogyeint tesearThi Wonterld. Oa:n ailD - for Urban Areas in Africa. Washington, D.C.: The World opment Research Center, 1978. Bank, 1980. Blackmore, M., R. Boydell, N. Mbere, and P. Moselele. "Low-Cost Sanitation Research Project: Interim Reports m 1-3 and Final Report." Gaborone, Botswana: Ministry of Mara. Water, Wastes, and Health in Hot Climates. Local Government and Lands, 1976-78. Processed. Chichester, U.K.: John Wiley and Sons, 1977. Central Public Health Engineering Research Institute. Feachem, Richard G., David T. Bradley, Hemda Garelick, Health Status of Sewage Farm Workers. Technical Di- and D. Duncan Mara. Sanitation and Disease: Health gest, no. 17. Nagpur, India, 1971. Processed. Aspects of Excreta and Wastewater Management. World Bank Studies in Water Supply and Sanitation, no. 3. Clar, Sott,andothrs. Disase isk ofOccuatinal Baltimore, Md.: Johns Hopkins University Press, forth- Exposure to Sewage." Journal of the Environmental En- comn. ' gineering Division, American Society of Civil Engineers, commg. vol. 102, no. FE2 (1976), pp. 375--88. _ . Appropriate Technology for Water Supply and Cotteral, J. A., and D. P. Norris. "Septic Tank Systems." Sanitation, vol. 3, Health Aspects of Excreta and Sullage Journal of the Sanitary Engineering Division, American Management-A State-of-the-Art Review. Washington, Society of Civil Engineers, vol. 95, no. SA4 (1969), pp. D.C.: The World Bank, 1980. 715-46. Gloyna, E. F. Waste Stabilization Ponds. Monograph Se- "Directory of Composting Systems.'" Compost Science/Util- ries, no. 60. Geneva: World Health Organization, 1975. ization, vol. 20, no. 1 (1979), pp. 22-25. Gunnerson, Charles G., and John M. Kalbermatten. "Alter- 110 BIBLIOGRAPHY 111 native Approaches to Sanitation Technology." Paper pre- Sanitation, vol. 2, A Planner's Guide. Washington, D.C.: sented at the International Association for Water Pollu- The World Bank, 1980. tion Research conference, Stockholm, June 1978. Pro- _ . Appropriate Technology for Water Supply and cessed. Sanitation, vol. 11, A Sanitation Field Manual. Washing- __ (eds.). Appropriate Technology in Water Supply ton, D.C.: The World Bank, 1980. and Waste Disposal. New York: American Society of Kennedy, C. "Induced Bias in Innovation." Economic Jour- Civil Engineers, 1979. nal, vol. 74 (September 1964), pp. 541-47. . (eds.). Appropriate Technology in Resource Con- Kilama, W., E. K. Simbeye, and U. Winblad. "Alternative servation and Recovery. New York: American Society of Waste Disposal Methods, Tanzania, 1975-77: Final Re- Civil Engineers, October 1979. port." Dar es Salaam: Tanzania National Scientific Re- Hinghaugh, G. M. A. "Nightsoil Treatment (on Lagos search Council, 1978. Processed. Island)." The Consulting Engineer, vol. 37, no. 9 (1973), Kuhlthau, Richard A. (ed.). Appropriate Technology for pp. 47-48. Water Supply and Sanitation, vol. 6, Country Studies in Hirsheleifer, Jack, James C. DeHaven, and Jerome W. Sanitation Alternatives. Washington, D.C.: The World Milliman. Water Supply Economics, Technology, and Bank, 1980. Policy. Chicago: University of Chicago Press, 1960. Kumpf, E. H. W., K. Mass, and H. Straub. Mull- und Julius, DeAnne S. "The Economic Cost of Power Outages: Abfall-Beseitigung. Bielefeld: Erich Schmidt Verlag, Theory and Applications to Jamaica." Public Utility Re- 1964-80. search Papers. Washington, D.C.: The World Bank, En- Laak, R., K. Healey, and D. Hardistry. "Rational Basis for ergy, Water, and Telecommunications Department, 1976. Septic Tank Design." Groundwater, vol. 12 (1974), pp. Processed. 348-52. ____. "An Economic Perspective on Resource Recovery Lauria, Donald T., Peter J. Kolsky, and Richard N. Middle- and Reuse Projects." In Appropriate Technology for ton. Appropriate Technologyfor Water Supply and Sani- Resource Conservation and Recovery, Preprint 3787. tation, vol. 9, Design of Low-Cost Water Distribution New York: American Society of Civil Engineers, 1980. Systems. Washington, D.C.: The World Bank, 1980. -_____ . "An Economic Appraisal of Sanitation Alterna- Mara, D. Duncan. Sewage Treatment in Hot Climates. tives."Progress in Water Technology, vol. 11, no. I Chichester, U.K.: John Wiley and Sons, 1976. (1979), pp. 251-58. Marais, G. v. R. "Fecal Bacterial Kinetics in Stabilization ____. "Urban Waste as an Economic Good." In Pacey Ponds." Journal of the Environmental Engineering Divi- (ed.), Sanitation in Developing Countries, pp. 194-200. sion, American Society of Civil Engineers, vol. 100, no. Julius, DeAnne S., and J. B. Buky. "Assessment of the EEl (1974), pp. 119-93. Economic Contribution of Water Resources to National Metcalf and Eddy, Inc. Wastewater Engineering: Collec- Development." Proceedings of the 1979 Institution of tion, Treatment, and Disposal. 2d ed. New York: Civil Engineers Conference, "Water Resources-A McGraw-Hill, 1975. Changing Strategy." Reprinted in Natural Resources McGarry, Michael G. "Developing Country Sanitation," Forum, vol. 4. no. 2 (April 1980), pp. 212-19. Report to International Development Research Centre. Julius, DeAnne S., and Jeremy J. Warford. "The Economic Ottawa, February 1975. Evaluation and Financing of Sewerage Projects." In McGarry, Michael G., and J. Stainforth. Compost, Fertil- Guidelines and Standards. Washington, D.C.: The World izer, and Biogas Production from Human and Farm Bank, Transportation, Water, and Telecommunications Wastes in the People's Republic of China. Ottawa: Inter- Department, February 1977. Processed. national Development Research Centre, 1978. Kalbermatten, John M., and DeAnne S. Julius. "Intermedi- Mohanrao, G. J. "Waste Collection, Treatment, and Dis- ate Service Levels in Sanitation Systems." In Gunnerson posal in India." Indian Journal of Environmental Health, and Kalbermatten (eds.), Appropriate Technology in Wa- vol. 15, no. 3 (July 1973), pp. 222-35 (six table refer- ter Supply and Waste Disposal, pp. 123-45. ences). Kalbermatten, John M., DeAnne S. Julius, and Charles G. National Academy of Sciences. Making Aquatic Weeds Use- Gunnerson. Appropriate Technology for Water Supply ful: Some Perspectives for Developing Countries. Wash- and Sanitation, vol. 1, Technical and Economic Options. ington, D.C., 1976. Washington, D.C.: The World Bank, 1980. _ . Methane Generation from Human, Animal, and Kalbermatten, John M., DeAnne S. Julius, Charles G. Agricultural Wastes. Washington, D.C., 1977. Gunnerson, and D. Duncan Mara. Appropriate Sanita- National Environmental Engineering Research Institute. tion Alternatives: A Planning and Design Manual. World Night-soil Wheelbarrows. Technical Digest, no. 32. Nag- Bank Studies in Water Supply and Sanitation, no. 2. pur, India, 1972. Processed. Baltimore, Md.: Johns Hopkins University Press, forth- Okun, Daniel A., and G. Ponghis. Community Wastewater coming. Collection and Disposal. Geneva: World Health Organi- . Appropriate Technology for Water Supply and zation, 1975. 112 BIBLIOGRAPHY Pacey, A. (ed). Sanitation in Developing Countries. Chich- Shuval, H. 1. "Parasitic Disease and Wastewater Irriga- ester, U.K.: John Wiley and Sons, 1978. tion." In Pacey (ed.), Sanitation in Developing Countries, Pahren, H., and W. Jakubowski (eds.). Wastewater Aero- pp. 210-14. sols and Disease. Report no. USEPA-600/9-80-028. Wash- Shuval, H. I., Charles G. Gunnerson, and DeAnne S. Julius. ington, D.C.: U.S. Environmental Protection Agency, Appropriate Technology for Water Supply and Sanita- 1980. Available from National Technical Information tion, vol. 10, Night-soil Composting. Washington, D.C.: Service, Springfield, Va. 22161. The World Bank, 1980. Pearson, H. "Toxic Effects of Inorganic Salt Build-up on Spangler, Charles. Appropriate Technology for Water Sup- Microalgae in Zero- and Low-Flow Night-soil Treatment ply and Sanitation, vol. 12, Low-Cost Water Distribu- Ponds." Washington, D.C.: The World Bank, Transporta- tion-A Field Manual. Washington, D.C.: The World tion, Water, and Telecommunications Department, 1978. Bank, 1980. Processed. Squire, Lyn, and Herman G. van der Tak. Economic Analy- Pescod, M. B., and Daniel A. Okun (eds.). Water Supply sis of Projects. Baltimore, Md.: Johns Hopkins University and Wastewater Disposal in Developing Countries, Pro- Press, 1975. ceedings. Bangkok: Asian Institute of Technology, 1971. U.S. Environmental Protection Agency. 2 Water Quality Processed. Criteria. Washington, D.C., 1973. Pradt, L. A. "Some Recent Developments in Night-soil _ . Health Effects of Nitrates on Water. Report no. Treatment (in Japan)." Water Research, vol. 5 (1971), USEPA-600/1-77-30. Washington, D.C., 1977. Available pp. 507-21. from National Technical Information Service, Spring- Raman, V. Sanitation Facilities for Slums and Rural Ar- field, Va., 22161. eas. Nagpur, India: National Environmental Engineering U.S. Public Health Service. Manual of Septic Tank Prac- Research Institute, 1975. Processed. tice. USPHS Publication no. 526; revised 1967. Washing- Raman, V., and N. Chakludor. "Upflow Filters for Septic ton, D.C.: U.S. Government Printing Office, 1972. Tank Effluents." Journal of the Water Pollution Control _ . Sewage Stabilization Ponds in the Dakotas. Wash- Federation, vol. 44, no. 8, pp. 1552-60. ington, D.C.: U.S. Government Printing Office, 1957. Rybczynski, Witold, Chongrak Polprasert, and Michael G. Wagner, E. G., and J. N. Lanoix. Excreta Disposal for McGarry. Low-Cost Technology Options for Sanitation: Rural Areas and Small Communities. Monograph Series, A State-of-the-Art Review and Annotated Bibliography. no. 39. Geneva: World Health Organization, 1958. A joint International Development Research Centre/ Warford, Jeremy J. "The Multiple Objectives of Water World Bank publication. IDRC-102elAppropriate Tech- Rate Policy in Less Developed Countries." Water Supply nology for Water Supply and Sanitation, vol. 4. Ottawa: and Management (November 1977), pp. 335-42. IDRC, 1978. . "Water Rates in Developing Countries." Journal of Saunders, Robert J., and Jeremy J. Warford. Village Water the American Water Works Association (April 1979), pp. Supply. Baltimore, Md.: Johns Hopkins University Press, 199-203. 1976. Warford, Jeremy J., and DeAnne S. Julius. "Water Supply Saunders, Robert J., Jeremy J. Warford, and P. C. Mann. and Sewerage Service Standards in LDCs." Paper pre- Alternative Concepts of Marginal Costfor Public Utility sented at the American Society of Planning Officials Pricing: Problems of Application in the Water Supply Conference, San Diego, April 1977. Processed. Sector. World Bank Staff Working Paper, no. 259. Wash- White, A. U., and G. F. White. "Behavioral Factors in ington, D.C., 1977. Selection of Technologies." In Gunnerson and Kalbermat- Shaw, V. A. "A System for the Treatment of Night-soil and ten (eds.), Appropriate Technology in Water Supply and Conserving-tank Effluent in Stabilization Ponds." Public Waste Disposal. Health Johannesburg, vol. 63 (1963), pp. 17-22. In Afri- White, G. F., D. J. Bradley, and A. U. White. Drawers of kaans; English translation available as CSIR Reprint Water: Domestic Water Use in East Africa. Chicago: RW166 from the National Institute for Water Research, University of Chicago Press, 1972. P. 0. Box 395, Pretoria, South Africa. Winblad, U., W. Kilama, and K. Tortensson. Sanitation Shipman, H., and others. "Measurement of the Health without Water. Stockholm: Swedisfi International Devel- Benefits of Investment in Water Supply." Public Utility opment Authority, 1978. Report, no. 20. Washington, D. C.: The World Bank, Winneberger, J. H. Manual of Graywater Treatment Prac- Transportation, Water, and Telecommunications Depart- tice. Ann Arbor, Michigan: Ann Arbor Science Publish- ment, January 1976. Processed. ers, 1974. Index Page nsumbers in italics indicate maps, tables, or illustrations. Aesthetics, importance of, 65 Cartage systems, 44; annual vacuum costs Consumer preferences. See Community Affordability, 64, 99 per household for, 59; limiting factor views and preferences Algorithms, 99, 100, 102, 103 in, 39: sewerage vs. 60; upgrading, 59. Contamination of water, 95 Alternative sanitation technologies: ben- See also Vault and vacuum-truck sys- Continuous composting latrines and toi- efit levels of, 5(0-51 community pref- tem lets. 13, 15 erences for, 95; comparing, 50, 52, 53; Cattle, tapeworms and, 73 Conversion factors, price and foreign ex- costs of, 50. 51, 53, 54-61, 63; failures Children: deaths of (from disease and change, 66nll of. 5; identifying, 5; lack of design ex- malnutrition), 3; excreta disposal and, Cost-benefit ratio. See Benefit-cost ratios perience with. 86; matrix ranking of, 70; ROECs and, 12: schistosomiasis in. Costing: comparative, 50; economic, 51: 50; most common, 11: need for, 4; in 68; as sources of infection, 70 life-cycle. 5; principles of, 51-52. See urban settings, 97-98. See also Sanita- China, sanitation technology in. 75 also Costs tion program planning; Sanitation sys- Choice of technology: factors affecting. Costs: anomalies of, 66nl2; average in- tems 18, 39, 42, 44, 47, 50, initial investment cremental (AIC). 53: capital, 54: case Anal cleansing materials: for aquaprivies. and, 57: use of algorithms in, 49; use data for, 59; of conventional sanitation. 16; choice of technology and. 42; for PF of matrix of characteristics in, 49. See 97: conversion factors and input, 56; latrines, 15 also Alternative sanitary technologies; economies of scale and. 52. 53: field Ancvlostorma duodenale, 33 Appropriate technology: Sanitation studies and, 54-59: historical. 51: in- Appliances, low-water-use, 66nlO program planning cremental, 51; input and conversion. 5: Appropriate technology: consulting firms Cholera. 30 investment and recurrent. 56-57; per and choice of, 8n9; defined, 5: identi- Cistern-flush toilets. 99 capita. 52. 53: real resource, 52. 60: fication of, 4, 5. 7-8: incentives for im- Climate, 18 22, 23, 24 recurrent. 54, 56-57; subsidization of. plementing, 86-88: obstacles to imple- Clonorchis sinensis. 45 63: sullage wastes and, 53: total annual menting, 85-86. See also Alternative Cockroaches, 68, 72. 73 per household, 55: upgrading. 105, 117: sanitation technology: Choice of tech- Communal sanitation systems, 13, 17, 100; of various technologies. 55, 57-59: 64- nology investment costs of, 64 61: water, 52. 56-57. See also Economic Aquaprivies. 13. 14. 15-16, 61: self-top- Community involvement, 79, 81, 88. 91. costs: Financial costs: Sanitation pro- ping (or sullage), 16, 44 94-95; objectives of, 93; rural, 98; scope gram planning: TACH Aridity, 18, 25, 26, 27 of. 93-94; steps in, 96; urban. 97, 98 Crops: fertilization of, 44: survival of Ascaris eggs, survival of, 73, 74, 75 Community views and preferences, 4, 86. pathogens on, 75 Average incremental cost (AIC). See Costs, 88: costs and. 7; credibility gap and. 79: Culex pipiens, 37. 68 average incremental (AIC) design and, 65; determining, 94-95; on Cultural aspects of sanitation. See Com- latrine and toilet facilities, 78; on water munity views and preferences Bacteria, 71. See also Infections supply abundance and proximity, 78. BARC thermophilic composting system, 43, See also Incentives and motivation Data: gathering, 77. 87. 95; use of case, 46 Community water and sanitation systems, 59. See also Surveys Batch composting latrines and toilets. 13; 16-18; bucket latrines and. 14, 16-17; Decisionmaking, algorithms as guidcs for. DVC as most common type of, 15 investment and recurrent costs of, 57; 99 Beltsville Agricultural Research Center users' view about. 78; vault toilets and, Defecating practices, 78. 79 (BARC), 46 14, 17 Demand analvsis. 86 Benefit-cost ratio, 50: divergence over Comparative costing. See Costing Demand forecast, 65n8 time, 52 Complementary investments, 39 Design. importance of, 65, 77, 78. 79. 81 Biochemical oxygen demand (BOD), 43. Composting: aerobic, 43: anaerobic, 15; Desludging. 16 47 BARC system, 43, 46: carbon-nitrogen Diphyllobothrium latum, 73 Biogas, 44, 46. 61-62 ratio for, 15; health aspects of, 73, 75; Diseases, 18, 30, 31, 32, 33, 34. 35. 36. Birds, tapeworm transmission and, 73 latrines and toilets, 13. 14, 15: moisture 37 Botswana. 61 control and, 15; temperature and, 15; Double-vault composting (DVC) toilet 14, Bucket latrines, 13, 14, 16-17 thermophilic, 46 15. 44, 61, 100, 101 Conservation, 79. See also Reclamation Ducks and fish production, 45 Capital: opportunity cost of, 51; price of, and reuse 51: shadow price of, 66n23 Construction by homeowners, 39, 42, 87 Economic costs: financial costs vs., 8n8, Carbon-nitrogen ratios, composting and, Consulting firms and selection of appro- 86: per household for upgrading, 1(07; 15 priate technology, 8n9, 87, 97 purpose of deriving, 63. 113 114 INDEX Educational programs, 42; supplementing Land use, 18, 38, 39, 40 Prices: convcrsion factors and. 66n Il; real sanitation improvements. 76. See also Latrines, influence of floor color on use costs and. 51; used to value costs, 51 Training of, 65. See also name of specific latrine Productivity. potential, 28 Emergencies. arrangements for, 95 type Protozoa. 71 Engineers: awareness of range of sanita- Laundry facilities, 17 tion technology and, 85; reeducation of, Least-cost solution, 51 Questionnaires. See Surveys 4 Lending agencies, 63, 85, 87 Exchange visits, 95 Liver fluke, 45 Reclamation and reuse. 39. 79; agricul- Excreta disposal programs, 4. See also Loans, 64 tural. 44. 45. 62-63; aquacultural. 63; Sanitation projects Local customs, 5 benefits from. 61-63, 66n2(): costs and. Excreta disposal systems, 100. See also London School of Hvgiene and Tropical 61: health effects and. 73. 75. See also Sanitation systems Medicine, 67 Biogas Excreted infections, 4, 71, 72; environ- Low-income rental units, floor plans for, Reed Odorless Earth Closet (ROfC). 13. mental classification of, 70-73; spread 40 44, 61, 10(i. 101: advantage of. 12; dis- by insects, 73 Maintenance: continuous composters and advantage of. 12: odor and flv nui- importance of, 15; ensuring, 81; prob- sances and. 11. 12; upgrading. 13. 39. Feasibility studies 6, 7, 87 lems with, 42; social preferences and, 1l)1 Fertilization, 44 7 Research needs, future. 47 Filariasis, 68, 73 Malacca. 53, 60 Resource rccovery. See Reclamation and Filter plants, 42, 105 Management skills, cost and, 56 reuse Financial costs: difficulty in developing, Master plan, 97-98 Reuse. See Reclamation and reuse 63; economic costs vs.; 8n8. 86; for least- Methane production. See Biogas Ross Institutc of Tropical Hygiene. 67 cost solutions, 7; real resource costs vs., Methemoglobinemia, 68 60; user income and, 7; Models (construction), 81 Salmonella. 68-69 Fish culture using night soil, 44. 45, 75 Models (economic) 53 Sanitation agency, ingredients of, 89-911 Fishponds, design of, 45A46 Mosquitoes: breeding of (in waste stabi- Sanitation facilities: constraints on pro- Flatworms, See Trematodes lization ponds), 43; filafiasis and, 73 visfon of. 4: in neighborrna areas. 99: Flies, 68; pit latrines and, 11; as vectors sullage and, 68; as vectors of excreted for tha poor, 93 for rural arecas. 93 of excreted infections, 72, 73; viP la- infections, 73. Sanltathon 7Iprovmentsa effects of (on trines and, 11; water seals and, 65 Motivation. See Incentives and motiva- health)q 7n: importanc. of sequentialo Food: preparation of, 72, 75; risk of in- tion 5; sequences for. 1(11. 104. 105. See also fection from, 75 Municipal services in developing coun- Upgrading tries, 4 Sanitation program planning: educational Gaborone, 53-54, 61 programs and. 42; feasibility studies for. Necator americanus, 34 6; institutional responsibilities for. 92: Helminths, 71; intermediate aquatic hosts Night soil: disposal of. 17; facilities for statistical approach to. 53: traditional and, 70, 73; transmitted by excreta, 69; treatment of, 39; health aspects of approach to, 7- use of surveys and ques- transmitted by soil, 72 digestion techniques, 73, 74. 75; man- tionnaircs ini 77. 78 79. See also Choice Higashi-Kurume (Japan), 60 agement schemes for (examples), 46- of technolo_s,y ( osts Household sanitation systems, 11-13, 13, 47, 48; treatment alternatives. 42-43 Sanitation proicets: behavioral science and 14, 15-16; influences on costs of, 65; Odor: aquaprivies and, 16: bucket la- design of, 79-82; community partici- investments costs of, 64; users' views trines and. 17; pit latrines and, I I; vip pation in. 79. 81 SS. 90(: failure of. 77. on, 78. See also name of specific system latrines and, I l; water seals and. 65 82, 89: moNeTiment su.pport for. 89. 90: Housing density, 39 Off-site exereta disposal systems. 13, 14, incenti%cs and motivation for, 78-79: 39 institutional rvquirements for. 89-92: Immunity, 69-70, 71 On-site excreta disposal systems, 110( measuriiw hnefits of .5(, 67: m,onitor- Incentives and motivation, 78-79, 86-88 Overhung latrine, 13 ino. 82 or. Siniz tion for. 89-9s c pro- motion of, 81: ruirzil. 98: sociocultural Individual preferences. See Community Pathogens: in compost. 15, 73. in feces dimensions of desion of. 8(): survess and views and preferencesI Infections: bacterial, 70-73; control and sewerage. 68-69: host response to. questionnaircs in, 77; types of. 97-99: measures for, 70; insect transmission of, 69; hosts of. 7(0: latencv of. 69; most urban. 97 98. See albo SanitIation pro- 73;sures involvingexre, a 3,; 6ms 0 witlow o resistant. 73; persistence of. 69, re- gram planningo Sa nitation svstems infective dose, 70 moval of. 42. 43; spread of. 68. survival Sanitation scrvices. 8: constrainlts on (in Infrastructure, 98, 99 times of. 73. 74, 75 developing countries). 4; cost of con- Infraseructs:aquaprivies and. 16; bucketa Personal hvgiene, 88; effects of (on health). ventionl, 3: for enltire community. 97; Insects: aquaprivies and, 16; bucket la- 75 'orpanizationI for, S9-92. paymcnt- for, trines and. 17; excreta and, 68. See also orai Onfr899;pyetfr specific kind, of; iexcrta and, 68.Seealso Physical environment. 18, 39 87, 88n'7 911 solutioni to (in developing Institutiokna cofnstraints, 42Pigs. See Swine countries). 4. urban and rural. 97-99 International constraints, Researh Pit latrines, 13; as most common tech- Sanitation svstems, 42. 44; comparative International Development Research nology, I; population densitv and, 13. costs of 54-59: cost case data (specific Centre, 54; bibliography on sanitation 39; suitability of. 13; upgrading of. 13 locations) 59-63: economics of scale technologies, 85; experimental latrine Planning. See Sanitation program plan- and, 527. :g encric classification of. 14: program of (Botswana), 60-61 ning inflexihilits of laroe scale, 53; low-cost. International Drinking Water Supply and Ponds. See Waste stabilization ponds 47. 88: matjor benefits of. 5(). 61: per Sanitation Decade, 3, 4, 88 Population densitv. 13. 39 capita costs of sewerage. 52; per house- Interviews, 94 Pour-flush (PF) latrines and toilets, 13. 14. hold costs of sewcraee. 59; water supply Irrigaton, infection from, 75 15. 44. 10(), 106; upgrading. 39, 1()1, levels and options in 39. 41. Scc a/so Israel, 75 105 AlternatisNe sanitation tcchnoloeics: Precipitation (meteorological). 18. 21. 22 Sanitation program planning: Sanita- Korea, 61, 62 Preferences. See Community views and tion projects: name of sp)ecific ivpie of Kyoto (Japan), 46-47, 48, 60 preferences srsteom INDEX 115 Schistosoma haematobium, 31 Swine, 45, 75, 79; tapeworms and, 73 pit latrines, 13; from ROEC toilets. 13, Schistosoma japonicum, 31 Synthetic studies, 53 39. 101: samole solutions for staged. Schistosoma mansoni, 32 Io5, 106, 107; from vault toilets, 17, Schistosomiasis, 68, 73 TACH (Total annual cost per household): 101, 105; from viP toilets, 13, 39, 101; Self-help construction, 39, 42, 87 difference between actual cash expend- to water-seal units, 39; water supply Seng, Ng Kin, 66n19 ituresand, 54; investmentcostsand, 57. service and, 39 Septic tanks, 13, 14, 16, 44; with drainage 57; of on-site systems and sewerage in Urban sanitary facilities: demand for, 86; fields, 99; three-stage, 44, 100, 101 Gaborone, 61; of public night-soil col- sewers as, 18; urban growth predictions Sewerage, 17-18, 44, 48, 61; annual costs lection system in Taiwan, 63; recurrent and, 53 per household, 59; appropriateness of, costs and, 57, 57; separation of (into 107; bias in favor of, 85; cartage vs., functional components), 58-59; of sew- Vault latrines and toilets 13 14 17 39 60; high cost of, 97, 107; historical de- erage and vacuum-truck cartage in Ma- 17 , velopment of. 4; in industrialized coun- lacca, 60; for ten technologies, 55; use 44. 100 tries, 3; low-cost alternatives to, 5; of (in comparing technologies), 54 Vault and vacuum-truck system (exam- manage'ment schemes for (examples), Taenia saginata, 35, 71, 73 ple), 46-47 46-47, 48; treatment alternatives, 42- Taenia solium, 36, 71, 73 Ventilated mproved double-pit (VIDP) la- 43 Taiwan, 61 62, 63 trmes, 11 Sewered facilities (PF toilets, septic tanks, Tapeworms, 36, 36, 71, 72, 73 Ventilated improved pit (VIP) latrines, 11, aquaprivies), 44, 100; advantages over Tariffs, 87, 88n2, 90 3, 14, 44, , , , 5, up- conventional sewerage, 107 Technical support: arrangements for, 90, graded to PF toilets, 13, 105 Sewer systems, 13; conventional vs. small- 94; by periodic visits of technicians, 95 Ventilatcon sacks, 65 bore, 18; failure of, 18; maintenance of, Technological choice. See Alternative Virueo cholera, 68 4; materials for, 18; per capita construc- sanitation technologies; Appropriate Viruses, 71; control of infections caused tion costs of, 52; slope needed, 18; small technology; Choice of technology; San- by, 71-72; survival of enteric, 73, 74, bore, 105; treatment works for, 39 itation program planning Shadow values, 51 Technologies: cost comparisons and, 54- Showers, 17 61; demand curve for, 86; diffusion of, Waste disposal, waterborne, 4 Sludge: batch digestion of, 42; treatment 80-81; lack of awareness of range of, Waste stabilization ponds, 43; evaluation of, 42-43. See also Desludging; Sew- 85; least-cost solutions and, 7; low-cost, of health aspects of, 73, 75 erage 54-55, 107; modification for adoption Water: amount used to flush toilets, 65n6, Snails: eradication programs for (in China), of, 81-82; need for frequent reassess- 66nl6; health and, 67, 68; influence of 75; trematodes and, 69; in waste sta- ment of, 89; selection of, 5, 7, 80, 98, cost of, 56-57; significance of reducing bilization ponds, 43, 45 99-101. See also Alternative sanitation nonessential use of, 107; users' percep- Soils, 18, 29; survival of pathogens in, 75 technologies; Appropriate technology; tions about, 78; volume needed of safe, Spray irrigation, possible infection from, Choice of technology; name of specific 75 75 technology. 22, 23, 24 Waterborne waste disposal. See Sewerage Statistical studies, 53 Temperatures: in compost, 15; winter, 18, Water and sanitation agency, 89-90 Storm-water drains, 53 19. 20, 22 Water-saving appliances and septic tanks, Study communities, 12, 13 Training: for community workers, 99; for 99 Subsidization of costs, 63 staff, 90, 95. See also Educational pro- Water seal: in converting ROEC to PF toi- Sullage: costs and treatment of, 53; de- grams let, 101; in converting viP to PF toilet, fined, 49n4; disposal of, 43-44, 67, 101; Trematodes and snails, 69 101; maintenance of, 16; odors and, 65 as health hazard, 44, 67, 68; and sew- Water service levels, 39; sanitation op- ered PF system, 100, 105 United Nations Water Conference (Mar tions and, 41, 99, 101, 104; for selected Superstructures (latrine), moving of, 78 del Plata. Argentina. 1977), 3 study communities, 12; users' percep- Surveys, 80, 87; interviewers for making, Upgrading, 5, 47; from aquaprivy system, tions about. 78, 101 94; interviews to determine local atti- 16; from bucket latrine, 17; to cistern- Water supply projects and services, or- tudes, 94; methods, 77; questionnaires. flush toilet, 101; from DVC toilets, 39. ganization for, 89-92 79, 94; results of. 78-79 101; from PF toilets, 39, 101, 105; from Wells, distance of (from pit latrines), 13 The full range of World Bank publications, both free and for sale, is described in the Catalog of World Bank Publications; the continuing research program is outlined in World Bank Research Program: Ab- stracts of Current Studies. Both booklets are updated annually; the most recent edition of each is available without charge from the Pub- lications Unit, World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A. John. M. Kalbermatten is senior adviser for water and wastes and Charles G. Gunnerson is senior project officer in the Transportation and Water Department of the World Bank. DeAnne S. Julius is acting economic adviser for the Energy Department of the World Bank. A World Bank Publication The United Nations has designated the 1980s as the International Drinking Water Supply and Sanitation Decade. Its goal is to provide two of the most fundamental human needs-safe water and sanitary disposal of human wastes-to all people. To help usher in this important period of international research and coopera- tion, the World Bank is publishing two volumes on appropriate technology for water supply and waste disposal systems in developing countries. Since 1976, Bank staff and researchers from various countries have been analyzing the economic, environmental, health, and sociological effects of various technologies to identify the most appropriate systems for the needs and resources of different areas. The research has included field investigations in nineteen countries. Since the technology for supplying water is better understood, the emphasis in these volumes is on sanitation and waste reclamation technologies, their contributions to better health, and how they are affected by water service levels and the ability and willingness of communities to pay for the systems. This volume summarizes the technical, economic, environmental, health, and sociocultural findings of the World Bank's research program on appropriate sanitation alternatives and then discusses the aspects of program planning that are necessary to implement these findings. It is directed primarily toward planning officials and sector policy advisers for developing countries. The most important finding is that there are many different kinds of technology that can be safely and cheaply used on a wide scale. Sanitation sequences-step-by-step improvements in sanitation technology that can build on each other as a community can afford to upgrade its system-are also designed and costed. In addition, a new and promising approach is presented to the problem of linking potential benefits to health with improvements in environmental sanitation. The Johns Hopkins University Press Baltimore and London ISBN 0-8018-2578-4