groundwater QUALITY PROTECTION Groundwater Quality Protection a guide for water utilities, municipal authorities, and environment agencies Stephen Foster Ricardo Hirata Daniel Gomes Monica D'Elia Marta Paris Groundwater Management Advisory Team (GWeMATE) II I in associationwith the Global Water Partnership 1 co-sponsored by WHO-PAHO-CEPIS & UNESCO-ROSTIAC-PHI THEWORLD BANK Washington, D.C. Copyright02002 The International Bank for Reconstruction and Development /The World Bank 1818 H Street, NW Washington, DC 20433, USA Telephone: 202 473-1000 Facsimile: 202 477-6391 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved First printing September 2002 Second printing April 2007 2 3 4 5 0 9 0 8 0 7 This volume is a product of the staff of the International Bank for Reconstruction and Development I The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. TheInternational Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Center Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-656-8600; Inernet: www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Stephen Foster is Leader of the World Bank-Global Water Partnership Groundwater Management Advisory Team (GW-MATE),Visiting Profasor of Contaminant Hydrogeology in the University d London, Vice-Presidentof the International Association of Hydrogeologists and was formerly the World Health Organization's Groundwater Advisor for the Latin American-Caribbean Region and Divisional Direcmr of the British Geological Survey. R i d o Himta is Professor of Hydrogeology at the Uaivmidade de SPo Paulo-Brazil, having previously been a Post-Doctoral Research Fellow at the Unimrsity of Waterloo- Canada and a Young Professional d rhc WHOIPan-American Hrnkh Organization. Daniel Games is a Senior Consulrant of Waterloo Hychagtologic Inc-Canada, having previously been a Hydrogaalogist with CETESB-Braziland a Young Professional of the WHOIPan-American Health Organization. M-ca D'Elia and Marta Pans ace both Researchers and Lecturers in Geohydrology at the Universidad Nacional del Litoral-Facultad de Ingenieria y Cienciae Hidricas, Argentina. Left Cover Photo by Getty Images, photographer Jeremy Woodhouse Right Cover Photo courtesy of Ron GilinglStill Pictures Photos page 37 courtesy of Stephen Foster ISBN 0-8213-4951-1 Library of Congress Control Number: 2002728399, Library of Congress Call Number: TD426 .G753 2002 Contents Forewords vi Acknowledgments, Dedication vii 1. Why has this Guide been written? 2. Why do groundwater supplies merit protection? 3. What are the common causes of groundwater quality deterioration? 4. How do aquifers become polluted? 5. How can groundwater pollution hazard be assessed? 6. What does groundwater pollution protection involve? 7. Why distinguish between groundwater resource and supply protection? 8. Who should promote groundwater pollution protection? 9. What are the human and financial resource implications? - MethoddogkalApproaches to Groundwater Protection 13 B1 Mapping Aquifer PollutionVulnerability Principles Underlying the Vulnerability Approach Development of the Vulnerability Concept Need for an Absolute IntegratedVulnerability lndex Application of COD Vulnerability lndex Comparisonwith Other Methodologies Limitations of Vulnerability Mapping Procedural Issues in Vulnerability Mapping BZ Delineationof Croundwater Supply ProtectionAreas Basis for Definition of Perimeters of Areas (A) Total Source Capture Area (B) Microbiological Protection Area (C) Wellhead Operational Zone (D) Further Subdivision Croundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies Factors Controlling Shape of Zones Limitations to Supply ProtectionArea Concept (A) Common Problems with Suggested Solutions (B) Case of Karstic Limestone Aquifers (C) Case of Spring and Gallery Sources (D) Implementation in Urban Settings Methods for Definition of Protection Zone Perimeters (A) Analytical Versus Numerical Aquifer Models (B) 2-D Versus 3-D Aquifer Representation (C) Practical Considerations Dealing with Scientific Uncertainty agb Perimeter Adjustment and Map Production 63 Inventory of Subsurface ContaminantLoad Common Causes of Croundwater Pollution Basic Data Collection Procedures (A) Designing a Contaminant Load Inventory (B) Characteristics of Subsurface Contaminant Load (C) Practical Survey Considerations Classificationand Estimation of Subsurface Contaminant Load (A) Spatial and Temporal Occurrence (B) The POSH Method of Load Characterization Estimation of Subsurface Contaminant Load (A) Diffuse Sources of Pollution (B) Point Sources of Pollution Presentationof Results 64 Assessment and Control of Croundwater Pollution Hazards Evaluationof Aquifer Pollution Hazard (A) Recommended Approach (B) Distinction Between Hazard and Risk Evaluationof Groundwater Supply Pollution Hazard (A) Approach to Incorporation of Supply Capture Zones (B) Complementary Wellhead Sanitary Surveys Groundwater Quality Protection: n guide for tunter tttilities, rncrnicipnlauthorities, nnd environment agencies Strategiesfor Control of Groundwater Pollution (A) Preventing Future Pollution (B) Dealing with Existing Pollution Sources - (C) Approach to Historic Land Contamination (D) Selecting New Groundwater Supply Areas m Role and Approach to Groundwater Quality Monitoring 92 (A) Limitations of Production Well Sampling 92 (B) Systematic Monitoring for Groundwater Pollution Control 93 (C) Selection of Analytical Parameters 93 Mounting Groundwater Quality Protection Programs 95 (A) Institutional Requirements and Responsibilities 95 (B) Addressing Key Uncertainties and Challenges 96 (C) Creating a Consensus for Action - - - -- 98 . .. - - References Forewords T" T is is a much welcomed publication that provides clear his Guide has been produced in the belief that groundwater guidance to water-sector decision makers, planners, and pollution hazard assessment must become an essential part practioners on how to deal with the quality dimension of of environmental best practice for water supply utilities. Such groundwater resources management in the World Bank's client assessments should lead to a clearer appreciation of priority countries. It is very timely, since there is growing evidence of actions required of municipal authorities and environmental increasing pollution threats to groundwater and some well- regulators to protect groundwater, both in terms of avoiding documented cases of irreversible damage to important aquifers, future pollution and mitigating threats posed by existing following many years of widespread public policy neglect. activities. In the majority of cases the cost of these actions will be modest compared to that of developing new water supply The idea to undertake such a review came from Carl Bartone sources and linking them into existing water distribution and Abel Mejia Betancourt of the World Bank, following an networks. initial attempt to draw attention to the need for groundwater protection in the Latin American-Caribbean Region by the The situation in some Latin American countries has become WHO-PAHO Centrefor SanitaryEngineering8Environmental critical, in part because many of the aquifers providing many Science (CEPIS),who together with the UNESCO-IHPRegional municipal water supplies are experiencing serious overdraft Office for Latin American-Caribbean Region have provided andlor increasing pollution. Among the cities of the region that support for this new initiative. are highly dependent upon groundwater resources, are Recife in Brazil, Lima in Peru, numerous Mexican cities, and most of The publication has been prepared for a global target audience the Central American capitals. under the initiative of the World Bank's Groundwater Management Advisory Team (GW-MATE),which works in The Guide is thus particularly relevant for the World Bank's association with the Global Water Partnership, under the Latin American and Caribbean Region, urhere many countries coordination of the GW-MATE leader, Dr. Stephen Foster. It is have initiated major changes to modernize their institutional practically based in a review of the last decade's experience of and legal framework for water resources management, but groundwater protection in Latin America and of concomitant may not yet have considered groundwater at the same level advances in the European Union and North America. Following as surface water, because of lack of awareness and knowledge ' the approaches advocated will help make groundwater more of groundwater issues and policy options. A process of visible at the policy level and in civil society. specialist consultation informed the present work, and came out with the recommendation that the Gzride should focus on John Briscoe one technique for each component of groundwater pollution Country Director, Brazil hazard assessment in the interest of clarity and consistency for Latin America & Caribbean Region, The World Bank the average user. Abel Mejia Betancourt Sector Manager, Environment Latin America SC Caribbean Region, The World Bank Acknowledaments Four meetings in Latin America represented key steps in ' San Jose, Costa Rica: November 2001 undertaking the systematic assessment of relevant experience Maureen Ballesterosand YamilethAstorga (GWP-CATAC), in that region and in reviewing the substantive content of this Arcadio Choza (MARENA - Nicaragua), Jenny Reynolds Guide. The following are acknowledged for their support and (UNA-Costa Rica) and Jost Roberto Duarte (PRISMA-El input to the respective meetings: Salvador). Santa Fe, Avgentina: October 1999 the late Mario Fili (Universidad Nacional del Litoral); The production of the Guide was managed by Karin Kemper, Mario Hernindez (Universidad Nacional de La Plata); Coordinator of the Bank-Netherlands Water Partnership Program M6nica Blasarin (Universidad Nacional de Rio Cuarto); (BNWPP),with the assistance of Carla Vale. and Claudio Lexouw (Universidad Nacional del Sur), all from Argentina The authors would also like to acknowledge valuable Montevideo, Uruguay: Octobev 2000 Idiscussions with the following of their respective colleagues: Carlos Fernindez Jiuregui and Angilica Obes de Lussich ! Hictor Gardufio (GW-MATE),Brian Morris (BritishGeological (UNESCO);Alejandro Delleperreand Maria Theresa Roma Survey),Paul Martin (Waterloo Hydrogeologic Inc) and Ofelia (OSE-Uruguay) Tujchneider (Universidad Nacional del Litoral-Argentina). Lima,Peru: March 2001 Henry Salas and Pilar Ponce (WHO-PAHO-CEPIS),Maria The design and production of the publication was carried out, ConsueloVargas(INGEOMINAS-Colombia),HugoRodriguez on behalf of the World Bank Group, by Words and Publications (ICAyA-CostaRica),Julia Pacheco (CNA-YucatLn-MCxico) of Oxford, UK, with the support of Gill Tyson Graphics. and Juan Carlos Ruiz (SEDAPAL-Peru) Dedication The authors wish to dedicate this Guide to the memory o f Professor Mario Fili of the Universidad Nacional del Litoral-Facziltad de Ingenieria y Ciencias Hidricas, Santa Fe-Argentina, who died prematurely during the project. Mario was one of the leading groundwater specialists of Argentina and Latin America, azlthor of some 70 published technical papers and articles, a life-long professional friend of the first author and much-loved professor and colleague of two other atrthors o f this Guide. vii ati ion ale for Groundwater Protection _... . . . _ _ _. _ An Executive Overview for senior personnel of water service companies, municipal authorities, and environment agencies, answering anticipated questions about groundwater pollution threats andprotection needs, and providing essentialbackground and standardized approaches to adopt in compliance with their duty to safeguard the quality of water destined for public supply. Why has this Guide been written? 2 Why do groundwater supplies merit protection? ! What are the common causes of groundwater quality deterioration? 3 How do aquifers become polluted? 4 How can groundwater pollution hazard be assessed? 6 What does groundwater pollution protection involve? I Why distinguish between groundwater resource and supply protection? 5 Who should promote groundwater pollution protection? 10 What are the human and financial resource implications? 11 Rationale for Groundwater Protection 1. Why hasthis CU& beenwitten? At the broad scale, groundwater protection strategies (and their prerequisite pollution hazard assessment) have to be promoted by the water or environmental regulator (or that agency, department, or office of national, regional, or local government charged with performing this function). It is important, however, that attention is focused at the scale and level of detail of the assessment and protection of specific water supply sources. All too widely in the past, groundwater resources have, in effect, been abandoned to chance. Often those who depend on such resources for the provision of potable water supplies have taken no significant action to assure raw-water quality, nor have they made adequate efforts to assess potential pollution hazard. Groundwater pollution hazard assessments are needed to provide a clearer appreciation o f . the actions needed to protect groundwater quality against deterioration. If undertaken by water supply utility companies, it is hoped that, in turn, both preventive actions to avoid future pollution, and corrective actions to control the pollution threat posed by existing and past activities, will be realistically prioritized and efficiently implemented by the corresponding municipal authorities and environmental regulators. Why do groundwatw uipp#crmedt protactkm? Groundwater is a vital natural resource for the economic and secure provision of potable water supply in both urban and rural environments, and plays a fundamental (but often little appreciated) role in human well-being, as well as that of many aquatic ecosystems. Worldwide, aquifers (geological formations containing useable groundwater resources) are experiencing an increasing threat of pollution from urbanization, industrial development, agricultural activities, and mining enterprises. Thus proactive campaigns and practical actions to protect the (generally excellent) natural quality of groundwater are widely required, and can be justified on both broad environmental sustainability and narrower economic-benefit criteria. In the economic context, it is also important that water companies make assessments of the strategic value of their groundwater sources. This should be based on a realistic evaluation of their replacement value, including both the cost of developing the new supply source and Part A: Executive Overview Rationale for Grotrndwater Protection also (most significantly) the cost of connecting and operating increasingly distant sources into existing distribution networks. Special protection measures are (in fact) needed for all boreholes, wells, and springs (both public and private) whose function is to provide water to potable or equivalent standards. This would thus include those used as bottled mineral waters and for food and drink processing. For potable mains water supply, a high and stable raw water quality is a prerequisite, and one that is best met by protected groundwater sources. Recourse to treatment processes (beyond precautionary disinfection) to achieve this end should be regarded as a last-resort, in view of their technical complexity and financial cost, and the operational burden they impose. 3. What u c the cemon causes of groundwaterqiu&iy d&crforrtlon? There are various potential causes of quality deterioration in an aquifer and/or in a groundwater supply. These are classified by genesis and further explained in Table A.1. In this Guide we are primarily concerned with protection against aquifer pollution and wellhead contamination, but it is necessary to be aware that other processes can also be operative. I I Table A . l Classification of groundwaler cluality problems - f 4QUIFER inadequate protection of vulnerable aquifers pathogens, nitrate or ammonium, chloride, POLLUTION against manmade discharges and leachates sulphate, boron, arsenic, heavy metals, from urbanlindustrial activities and dissolved prganic carbon, aromatic and intensification of agricultural cultivation halogenated hydrocarbons, certain pesticide! WELLHEAD inadequate well design/construction allowing mainky pathogens /ICONTAMINATION u direct ingress of polluted surface water or Saline Intrusion ---- - saline (and sometimes polluted) groundwater mainly sodium chloride, but can also includc -'induced ro flow into freshwater a~uiferas persistent manmade contaminants *result of excessiveabstraction *w*+-~v4,r -- - - - r- Natura y W?@@"*' -I *.4>' >rep )it1 groundwater polhtion hazard assessment and quality protection, when "the interest is at mrvnicipal or provincial scale. This chapter discusses the evolution of the aquifer pollution vulnerability concept before recommending a methodological basis for uulnerability evaluation that can be used for mapping at that scale. Tho concept is also valid for wrlnerability appraisal a t more local levels within individual grotrndwater supply catrhm~nfn*pnc I Principles Underlvinothe VulnerabilityAmroach Groundwater recharge mechanisms and the natural contaminant attenuation capacity of subsoil profiles vary widely with near-surface geological conditions. Thus, instead of applying universal controls over potentially polluting Land uses and effluent discharges, it is more cost effective (and less prejudicial to economic development) to vary the type and level of control according to this attenuation capacity. This is the basic premise underlying the concept of aquifer pollution vulnerability and the need for vulnerability mapping. In view of the complexity of factors governing pollutant transport into aquifers in any given situation, it might at first s~ghtappear that: hydrogeological conditions are too complex to be encapsulated by mapped vulnerability zones it would be more logical to treat each polluting activity on individual merit and undertake an independent assessment of the pollution hazard it generates. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies However this type of approach: is unlikely to achieve universal coverage and avoid inconsistent decisions . requires large human resources and major financial investment for field investigations can present administrative problems where institutional responsibility is split. 1iDevelopmcnt --- - of the Vulnerability Concept rn - In hydrogCulogy the LE;IIIl vulnerability "egall LU be used intuitively from thc 1970s in France (Albinet and Margat, 1970) and more widely in the 1980s (Haertle, 1983; Aller and others, 1987; Foster and Hirata, 1988). While the implication was of relative susceptibility of aquifers to anthropogenic pollution, initially the term was used without any attempt at formal definition. The expression began to mean different things to different people. A useful and consistent definition would be to regard aquifer pollution vulnerability as those intrinsic characteristics of the strata separating the saturated aquifer from the land surface, which determine its sensitivity to being adversely affected by a surface-applied contaminant load (Foster,-1987).It would then be a function of: the accessibility of the saturated aquifer, in a hydraulic sense, to the penetration of pollutants the attenuation capacity of strata overlying the saturated zone resulting from the physiochemical retention or reaction of pollutants. In the same way, groundwater pollution hazard would then be defined as the probability that groundwater in the uppermost part of an aquifer will become contaminated to an unacceptable level by activities on the immediately overlying land surface (Foster and Hirata, 1988; Adams and Foster, 1992). Subsequently two major professional working groups reviewed and pronounced upon the applicability of the vulnerability concept and come out strongly in favor of its usefulness (NRC, 1993; IAHNrba and Zaporozec, 1994). It would have been desirable for them to have made a clearer statement on the use of the term, for example associating it specifically with the intrinsic characteristics of the strata (unsaturated zone or confining beds) separating the saturated aquifer from the land surface (Foster and Skinner, 1995).This would (most importantly) have related it directly with the potential impact of land-use decisions at the location concerned on the immediately underlying groundwater. Some, however, considered that a factor representing the natural mobility and persistence of pollutants in the saturated zone be included in vulnerability. This, however, does not appear to view vulnerability mapping from the most useful perspective, namely that of providing a framework for planning and controlling activities at the land surface. Part B: Technical Guide Methodologrcal Approaches to Groundwater Protectioh TWOfundamental questions that arise In relation to aquifer pollution vulnerability are whether it is possible: to present a single integrated vulnerability index, or be obliged to work with specific vulnerability to individual contaminants and to pollution scenarios to provide an absolute indicator of integrated pollution vulnerability, or be restricted to much less useful relative vulnerability indices. Subsurface water flow and contaminant transport are intricate processes. In reality, the interaction between components of aquifer pollution vulnerability and subsurface contaminant load, which determine the groundwater pollution hazard, can be complex (Figure 1.1). In particular, the degree of contaminant attenuation can vary significantly with the type of pollutant and polluting process in any given situation. Thus a "general (integrated)vulnerability to a universal contaminant in a typical pollution scenario" has no strict validity in rigorous terms (Foster and Hirata, 1988). Scientifically, it is more consistent to evaluate vulnerability to pollution by each pollutant, or failing this by each class of pollutant (nutrients, pathogens, microorganics, heavy metals, etc.) individually, or by each group of polluting activities (unsewered sanitation, agricultural cultivation, industrial effluent disposal, etc.) separarely. For this reason (Andersen and Gosk, 1987)suggested that vulnerability mapping would be better carried out for individual contaminant groups in specific pollution scenarios. However, the implication would be an atlas of maps for any given area, which would be difficult to use in most applications, except perhaps the evaluation and control of diffuse agricultural pollution (Carter and others, 1987; Sokol and others, 1993; Loague, 1994). Moreover, there will not normally be adequate technical data and/or sufficient human resources to achieve this ideal. In consequence, a less refined and more generalized system of aquifer vulnerability mapping is required. The way forward for most practical purposes is to produce an integrated vulnerability map, provided the terms being used are clearly defined and the limitations clearly spelled out (Foster and Hirata, 1988). Such health warnings have been elegantly expressed in the recent U.S. review (NRC, 1993) in the form of three laws of groundwater vulnerability: all groundwater is to some degree vulnerable to pollution uncertainty is inherent in all pollution vulnerability assessments in the more complex systems of vulnerability assessment, there is risk that the obvious may be obscured and the subtle indistinguishable. An absolute index of aquifer pollution vulnerability is far mote useful (than relative indications) for all practical applications in land-use planning and effluent discharge control. An absolute integrated index can be developed provided each class of vulnerability is clearly and consistently defined (Table 1.1).In this way it is possible to Groundwater Quality Protection: a guide for water utilities, mtrnicipal nctthorities, nnd environment ngencies Figure 1.1 Interactions between components of subsurface contaminant load and aquifer pollutionvulnerability determining aquifer pollution hazard SUBSURFACE AQUIFER CONTAMINANT POLLUTION LOAD VULNERABILITY I ?eFigure 3.3 see Figure1.2 I ulnerabilil I J PollutionPokntlal Ran,., , Part 0:Technical Guide Methodological Approaches to Grortndruater Protection Extreme vulnuabIe to most water pollutants wifh rapid impact in many pollurion scenarios High vulnerable to many pollutants (except those strongly absorbed or readily transformed) in many pollution scenarios Moderate vulnerable to some pollutants but only when continuously discharged or leached Low only vulnerable to conservative pollutants in the long term when continuousjy and widely discharged or leached Negligible confining beds present with no significant vertical groundwater flow (leakage) I overcome most (if not all) the common objections to the use of an absolute integrated vulnerability index as a framework for groundwater pollution hazard assessment and protection policy formulation. ,n~lication COD Vulnerrrbilitv Index of The GOD method of aquifer pollution vulnerability assessment has had wide trials in Latin America and the Caribbean during the 1990s (Table 1.2), and because of its simplicity of concept and application, it is the preferred method described in this Guide. Two basic factors are considered to determine aquifer pollution vulnerability: the level of hydraulic inaccessibility of the saturated zone of the aquifer the contaminant attenuation capacity of the strata overlying the saturated aquifer; however they are not directly measurable and depend in turn on combinations of other parameters (Table 1.3).Since data relating to many of these parameters are not generally available, simplification of the list is unavoidable if a practical scheme of aquifer pollution vulnerability mapping is to be developed. Based on such considerations, the GOD vulnerability index (Foster, 1987; Foster and Hirata, 1988)characterizes aquifer pollution vulnerability on the basis of the following (generally available or readily determined) parameters: Groundwater hydraulic confinement, in the aquifer under consideration. Overlying strata (vadose zone or confining beds), in terms of lithological character and degree of consolidation that determine their contaminant attenuation capacity Depth to groundwater table, or to groundwater strike in confined aquifers. Groundwater Quality Protection:a guide for water utilities, nzunicipol nuthorities, and environment agencies Part B: Technical Guide Methodological Approaches to GroundwaterProtection -. Box 1.1 IVulnerabil ;em mfined aq 2rs-field c--a from Leh, Mexico !neaaquifer of low pollution vulnerability can be sen'ously :-- -=tea m me long rrr by persistent contami~nts(suchas cr,,oride, nitrate, and certain synthetic organic compounds), it ,yatecontinuously I discharged on the overlying ground surface. This possibility must always be taken into account wk hazard to watevtyells abstracting from such aquifers. a m Lebn (Guanajuato) is one of the fastest-growing cities 7I in Mtxico and one of the most important leather- (A) Attenuation of chromium in soils of wastewater irrigation area nanufacturing and shoe-making centers in Latin Americ 1 The city is located in an arid upland tectonic valley fillec total Cr in soil (mglkg) by a mixture of alluvial, volcanic, and lacustrine deposits, o 100 which form a thick complex multi-aquifer system. 0 A substantial proportion of the municipal water supply 0.2 is derived from downstream wellfields, which tap a serni- :onfined aquifer from below a 100-meter depth. One of h e wellfields is situated where municipal wastewater has been used over various decades for agricultural irrigation The inefficient irrigation characteristic of wastewater reuse :esults in a substantial (and continuous) recharge of the .ocal groundwater system. Thus groundwater levels have .?ere remained within 10 meters of the land surface, desp~te - :he fact that in neighboring areas they have been in steady I ong-term decline at rates of 1-3 meters per year (rnla). long-termwastewater floor offormer I irrigation field wastewater lagoon The wastewater historically included an important I- component of industrial effluent with very high d i s d ~ L chromium, organic carbon and overall salinity. l%&d -- field investigations in the mid-1990s by the C o w (6) Variation of groundwater quality with depth Nacional del Agua-Gerencia de Aguas Subterraneas'. beneath wastewater irrigation and the Servicio de Agua Potable de Leon have m n *. .A that most elements of the contaminnnt load fin&dhg "-'JRCE OF TYPICAL PUBLIC SUPPLY : ---Logenic microbes and heavy metals) are r a p i i i 1--: 4PLE SHALLOW WELL BOREHOLES - -. luated in the subsoil profile (Figure A). Verf lit* ' intake deprh <30 m 200-300 rn . h e s the semi-confined aquifer (Stuart and Milnq; -- , ~ 1997),whose po[lution vulnerability under the GO^ EC (pS/cm) 3400 1000' . . m v d classify in the low range. C1 ( m g ) 599 203 ; 4 - However, persistent contarninants-nota5 y sal~nityas H C 0 3 (mg/l) 751 239. indicated by C1concentrations (Figure B)-do penetrate 13.5 6.0 'nto the semi-confined aquifer and are threatening the NO3(mdl) . --: I qualiry and security of municipal water supplies in this 227 44 area (Stuart and Milne, 1997). Groundwater Quality Protection:n guide for water tctilities, municipal authorities, and environment agencreb ---- --- - - . - COMPONENT OF HYDROGEOLOGICALDATA VULNERABILITY ideally required normally available Hydraulickcemihdity degree of aquifer confinement type of groundwater confinement I I depth t o groundwater table or depth to groundwater table or top of groundwater mike confined aquifer unsaturated zone moisturecohtenr vertical hydraulic conductivity of strata io vadose zone or confining beds Attenuation Capzcity gr;ain and fissure size distributionof grade of mnmlidation/fissllring strata in vadose zone or confining beds these stnta I mineralogy of strata in vadose zone or lithological character of these mata conlining beds Further consideration reveals that these parameters embrace, if only in a qualitative sense, the majority of those in the original list (Table 1.3). w LLI The empirical methodology proposed for the estimation of aquifer pollution vulnerability LL s ! (Foster and Hirata, 1988) involved a number of discrete stages: Q first, identification of the type of groundwater confinement, with consequent indexing of this parameter on scale-0-1 isecond,specificationofthestrataoverlyingtheaquifersaturatedzoneinterms of (a) grade of consolidation (and thus likely presence or absence of fissure permeability) and (b) type of lithology (and thus indirectly dynamic-effective- porosity, matrix permeability, and unsaturated zone moisture content or specific I retention); this leads to a second score on a scale 0.4-1.0 ' third, estimation of the depth to groundwater table (of unconfined aquifers) or I depth of first major groundwater strike (for confined aquifers), with consequent ranking on the scale 0.6-1.0. The final integrated aquifer vulnerability index is the product of component indices for these parameters (Figure 1.2).It should be noted that this figure has been modified slightly from the original version (Foster and Hirata, 1988) in light of experiences in its application during the 1990s. The modifications include: somewhat reduced weighting to the "depth to groundwater" factor some simplification of the geolog~caldescr~ptorsas regards "potentially fractured rocks of ~ntermed~ateintrinsic vulnerability" clarification of the "groundwater confinement" factor as regards semi-confined aquifers. Part 0: Technical Guide Methodologicnl Approaches to Growndtunter Protection Figure 1.2 COD system for evaluation of aquifer pollution vulnerability GROWWATER CONFINEMENT I) -- OVERLYINGSWA ff~uWwilMbn1 sandmna , i ~ ~ ~ a CONSOLOA~D dkgraaof I) + shales wbnk tuffs cwm (p~musrocks) cwuolld.donofvadow -M--) I I I - 1 ipmeyhic 1 writ &R&S + CONSOLlDATED 4 rmabons and der volcanic karst Limestones vokanicr lavas - (&,,* wkj 1 I! I I I I I DEPTHTOGROUNDWATER (u- I) OR STRIKE (confined) NECLlClBLE LOW MODERATE HIGH EXTREME I It should also be noted that, where a variable sequence of deposits is present, the predominant or limiting lithology should be selected for the purpose of specification of the overlying strata. In the GOD scheme, a descriptive subdivision of geological deposits (involving grain- size and mineral characteristics) could have been used and might appear easier to apply. However, a genetic classification better reflects factors important in the pollution vulnerability context (such as depositional structure), and thus a hybrid system (compatible with those used for many geological maps) is adopted. Almost all of the sediments in the classification (Figure 1.2) are transported geological deposits. However, two other types of deposits are retained because of their widespread distribution--deep residual soils (such as the laterites of the tropical belt) and desert calcretes (an in-situ deposit). Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies In the context of the classification of overlying strata, there was concern that too much consideration might inadvertently be placed on dynamic porosity (and thus merely on recharge time lag rather than contaminant attenuation). Vulnerability would then (incorrectly) become more a measure of when (as opposed to if and which) pollutants reach the aquifer. Thus greatest emphasis was put upon the likelihood of well-developed fracturing being present, since this may promote preferential flow even in porous strata such as some sandstones and limestones (Figure 1.3). The possibility of such flow is considered the most critical factor increasing vulnerability and reducing contaminant attenuation, given that hydraulic (fluid) surcharging is associated with many pollution scenarios. The original GOD vulnerability scheme did not include explicit consideration of soils in an agricultural sense. However, most of the processes causing pollutant attenuation and/or elimination in the subsurface occur at much higher rates in the biologically active soil zone, as a result of its higher organic matter, larger clay mineral content and very much larger bacterial populations. A possible modification to the method (GODS) incorporates a soil leaching susceptibility index (based on a soil classification according to soil texture and organic content), as a fourth step capable of reducing overall ranking in some areas of high hydrogeological vulnerability. Within urban areas the soil is often removed during construction or the subsurface pollutant load is applied below its base in excavations (such as pits, trenches, or lagoons), thus the soil zone should be assumed absent and the uncorrected hydrogeological vulnerability used. Figure 1.3 Development and consequences of preferential flow in the vadose zone CONTAMINANT SOLUBLE DENSE IMMISCIBLE WATERBORNE LOAD ON MOBILE IONS COMPOUNDS COLLOIDAL PARTICLES , LAND (chloride, nitrate) (DNAPLs, creosote) (bacteria, virus) 8 SATURATED ZONE (AQUIFER) Part 0: Technical Guide Methodological Approaches to Groundtoater Protection A number of other schemes of aquifer pollution vulnerability assessment have been presented in the literature, and these can be classified into three main groups according to the approach adopted (Vrba and Zaporozec, 1995): Hydrogeological Settings: these base vulnerability assessment in qualitative terms on the general characteristics of the setting using thematic maps (eg. Albinet and Margat, 1970) Analogue Models: these utilize mathematical expressions for key parameters (such as average vadose zone transit time) as an indicator of vulnerability index (EC/ Fried approach in Monkhouse, 1983) Parametric Systems: these use selected parameters as indicative of vulnerability and 1 assign their range of values and interactions to generate some form of relative or absolute vulnerability index (examples of this approach include Haertle, 1983 and DRASTIC of Aller and others, 1987, in addition to the GOD methology described in this Guide).A further method of note in this category is EPIK, which is specifically designed for karst limestone aquifers only and usefully discussed by Doerfliger and Zwahlen, 1998; Gogu and Dassargues, 2000; Daly and others, 2001. Among these the best known is the DRASTIC methodology. It attempts to quantify 1 relative vulnerability by the summation of weighted indices for seven hydrogeological variables (Table 1.4). The weighting for each variable is given in parentheses, but changes (especially for parameters S and T) if vulnerability to diffuse agricultural pollution alone is under consideration. The method has been the subject of various evaluations (Holden and others, 1992; Bates and others, 1993; Kalinski and others, 1994; Rosen, 1994).All of these evaluations revealed both various benefits and numerous shortcomings of this methodology. On balance, it is considered that the method tends to generate a vulnerability index whose significance is rather obscure. This is a consequence of the interaction of too many weighted parameters, some of which are not independent but quite strongly correlated. The fact that similar indices can be obtained by a very different combination of circumstances may lead to dangers in decision making. I I I Depd, to youdwstat (XS) Topographic aspect (XI) Natural Recharge rams (X4) Impact (effect) af vadose zone (XS) Aquifer media (X3) Hydraulic Conductivity (X3) Soil media (X2) Groundwater Quality Protection: a guide for water utilities, municipal authorities, nnd environment ngencies luifer pollution vulnerability map a sdl-cover factor ir e Cauca Valley, Colombia - h- \ k.! la ome Latin American workers have :on to the bUUmethod of ~qui' utton vulnerability estimtrc which adds a factor in respect of the attenuation capacrry of the soil cover, based on texture atone. In general terms it is considered valid to include a "soil factor," although not in areas where there is risk that the soil profile has been rernweb ur disturbed and not in cases where the contaminant load is applied below the base of the soil. Moreover, if a soil factor is to I - be itzcluded it is prefn.61 to base it wpon soil thickness, together with those properties which most djrectiy influence in-ritu I r &nitrification and pesticide attenuation organic content), 5 - The Cauca Vatley has the largest groundwater storage 2haracteristiCso! the soil, which range from very fine resources of Colombia, and its aquifers currently :predominantly clayey) to very coarse (gravelly),in arrg support an abstraction of around 1000 Mm3/a, which where this is more than 0.5 thi k '0.:- is of fundamental importance to the valley's economic development and provides the municipal water supply A map of the values of this soil-cover factor was for various towns including Palmira, Buga, and pans c produced, which was then overlaid on the GOD aquifer Cali. The valley is a major tectonic feature with a largc vulnerability index map. In areas where the soil cover thickness of mixed valley-fill deposits in which alluvial was well preserved and of substantial thickness, the value fan and lacustrine deposits predominate. 3f the GOD index was correspondingly reduced (Paez, ..ith the aim of providing a tool for land-use planning' .- - protect these resources, the pollution vulnerability o f t & Environment Agency of England & Wales also );, aquifers was mapped by the local water resource agenc. ~ d ac soil factor in their aquifer vulnerabitity (the Corporaci6n del Valle de Cauca) using the GOD mapping. This is based on a set of soil properdcs method. A modification was introduced (as first proposed determining leaching susceptibility, but its effect is limit by the Pontificia Universidad de Chile-Dpto de Ingenieria to potentially reducing the mapped vulnuahiliv h e : i Hidraulica y Ambiental) incorporating an S factor in m d areas, and it is not considered operative in urban respect of the contaminant attenuation capacity of the ir-where soil profile disturbance due to engineering soil cover. The modified methodology (known as GODS) -tion is widepread (Foster, 1997' 1 invc ' s assigning values of S according to the textural - - - I F m - -. 1 1 - L COD Index Value (0-1.0) 1 r I I I I c-*l Sor rmnsh&idn shrinking coarsea n d thin/ a,, s'y 8gravel absent --a , lYCE 05 0.6 0.8 0 9 1 0 Part B: Technical Guide Methodological Approaches to Groundwater Protection More specifically it should also be noted that: the method underestimates the vulnerability of fractured (compared to unconsolidated) aquifers including a parameter reflecting contaminant mobility in the saturated zone is an unnecessary complication (for reasons stated earlier). A number of hydrogeological conditions present problems for aquifer pollution vulnerability assessment and mapping: the occurrence of (permanent or intermittent) losing streams, because of uncertainties in evaluating the hydrological condition, in defining the quality of the watercourse and in appraising streambed attenuation capacity (it is, however, essential to indicate potentially influent sections of streams crossing unconfined aquifers) excessive aquifer exploitation for water supply purposes, which can vary the depth to groundwater table and even the degree of aquifer confinement, but given the scheme of indexation proposed, such effects will only occasionally be significant over-consolidated (and therefore potentially fractured) clays, for which there are usually significant uncertainties about the magnitude of any preferential flow component. Aquifer vulnerability maps are only suitable for assessing the groundwater pollution hazard associated with those contaminant discharges that occur at the land surface and in the aqueous phase. Strictly speaking they should not be used for assessing the hazard from: contaminants discharged deeper in the subsurface (asmay be the case in leakage of large underground storage tanks, solid-waste landfill leachate, effluent discharges to quarries, and mine shafts, etc.) spillages of heavy immiscible synthetic organic pollutants (DNAPLs). Both are likely to result in high groundwater pollution hazard regardless of aquifer vulnerability. The only consideration in such circumstances will be the intensity and probable duration of the load. The technical validity of the aquifer pollution vulnerability index and map can be maintained, if it is made clear that these types of contaminant load are excluded from consideration by the proposed methodology and that such practices need to be specifically controlled irrespective of field conditions. Another condition that needs a special procedure is the existence of naturally poor- quality (normally saline) groundwater at shallow depth. This requires specific mapping since such aquifers will not generally merit special protection, even in cases of high anthropogenic pollution vulnerability. Groundwater Quality Protection: aguide for water utilities, municipalauthorities, and enurronmentage~tcies Figure 1.4 Interpretation of the pollution vulnerability of semi-confined aquifers 'Progiem:using the GOD method, the 0factor represents the lithology of confining beds or d ~ zone e unsaturatedzones, but for semi-confined aquifers this is difficult to determine allow aquih Soldm:consider the thinnest part of the L aquitard and calculate the 0 factor as a weighted 3value aquitard of different materials (vadose zone, shallow aquifer, and aquitard) P r o m using the COD method, the D factor is the distance between the land surface and the water table or water strike, but for a semi-confined 1 aquifer what is the correct value? - m n : use the depth to the aquifer (A+%) %bbt~poor quality shallow aquifer covering the semi-confined aquifer that requires protection r hhtb:considertheshallowaqu~ferasa potential contaminant source and thus use the characteristics of the aquitard only for the 0 and I-' D factors Pmknx hydraulic inversion caused by groundwater extractionfrom deep aquifer $&#ion: use C factor appropriate to new I r hydraulic condition and treat deep aquifers as now semi-confinedor even covered Part B: Technical Guide Methodological Approaches to Groundwater Protection r ~ r o c e d u r aIssues in Vulnerability Mappin'! l 1 The generation of the map of GOD aquifer vulnerability indices follows the procedures indicated in Figure 1.5. Such a process can be carried out manually for a series of points on a grid basis and contoured, but is increasingly generated by GIs (geographical information system) technology. In the majority of instances, hydrogeological maps andlor groundwater resource reports will be available, and generally these will contain adequate basic data to undertake the evaluation procedure proposed. However, it will often be necessary to supplement this information by the direct study of geological maps and waterwell drilling records, and sometimes by limited field inspection. (A) Approach to Layered Aquifers One of the most frequent difficulties encountered in aquifer pollution vulnerability mapping is the presence of layering of strata of widely different water-transmitting properties. Stratification is a fundamental characteristic of both sedimentary and volcanic geological formations, and such formations include almost all major, and many minor, aquifers. Problems may result when the layering occurs both: above the regional groundwater table, giving rise to perched aquifers or covered unconfined aquifers (where weighted average or limiting values of the relevant properties need to be considered), and # , 1 \ I Figure 1.5 Generation of aquifer pollution vulnerability map using the COD system unconfinedaquifer 7 semi-confined aquifer f i u v i o - 2 ll~vialgrave1 sands and silts extreme low moderate high Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies , @. . bcf~w& kgiorul pdGalar &Me, cavring semi-confinement of aquifers at depth (for which a consistent decision needs to be clearly made and stated on which aquifer is represented by vulnerability mapping, and the attenuation capacity of the overlying strata assessed accordingly). The approach to classification detailed in Figure 1.4 should then be followed for vulnerability estimation, and a record made (by suitable ornament) where an overlying (more vulnerable) local aquifer is also present. (B)Necessary Level of Simplification It must be stressed that aquifer pollution vulnerability maps are designed to provide a general framework within which to base groundwater protection policy. The two, however, are distinct in both concept and function. The former should represent a simplified (but factual) representation of the best available scientific data on the hydrogeological environment, no more or no less. This general framework is not intended to eliminate the necessity to consider in detail the design of actual potentially polluting activities before reaching policy decisions. Aquifer vulnerability maps are aimed only at giving a first general indication of the potential groundwater pollution hazard to allow regulators, planners, and developers to make better informed judgements on proposed new developments and on priorities in groundwater pollution control and quality monitoring. They are based on the best available information at the time of production and will require periodic updating. In concept and in practice they involve much simplification of naturally complex geological variations and hydrogeological processes. Site-specific questions need to be answered by site-specific investigations, but the same philosophical and methodological approach to the assessment of groundwater pollution hazard is normally possible. The data required for the assessment of aquifer pollution vulnerability-and for that matter inventories of subsurface contaminant loads-should (wherever possible) be developed on a suitable GIs platform, to facilitate interaction, update, and presentation. Separate colors can be used for major lithological divisions of the strata overlying the saturated zone, with different densities of color for each subdivision of depth to groundwater. MethodologicalApproaches to Groundwater Protection Delineation of Groundwater Supply Protection Areas Groundwater supply protection areas (called wellhead ptotoction tones in the United States) should be delineated to provide special vigilance against pollution for water sources destined for public (mains) water supply. Cmside~ationmust also be given to sources developed for other potentially smitivc uses, and especially of bottled natural mineral waters, which do not receive any form of disinfection. 1 Basis for Definition of Perimeters of Arear The concept of groundwater supply protection is long established, being part of legal codes in some European countries for many decades. However, increasing hydrogeological knowledge and changes in the nature of threats to groundwater quality mean that the concept has had to evolve significantly and requires consolidation (US- EPA, 1994; NRA, 1995; EA, 1998). A key factor influencingthe hazard posed by a land-use activity to a groundwater supply (well, borehole, or spring) is its proximity. More specifically, the pollution threat depends on: whether the activity is located within the (subsurface)capture area of that supply (Figure 2.1) the horizontal groundwater flow time in the saturated aquifer from the location of the activity to the point of abstraction of the supply. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and envrronment agencies - Figure 2.1 Distinction between area of capture and zone of influence I of a production waterwell a) vertical profile - k- qjp - - I 1- groundwater divide I pumping land surface I , b) plan view - groundwater groundwater divide flow direction Supply protection areas (SPAS)-also known as source protection zones (SPZs)-have to defend against: 'a contaminants that decay with time, where subsurface residence time is the best measure of protection nondegradable contaminants, where flowpath-dependent dilution must be provided. Both are necessary for comprehensive protection. Contaminant dilution resulting from the advection and dispersion mechanisms associated with groundwater flow is usually the dominant attenuation process, but degradation (breakdown) is also likely to occur Part B: Technical Guide Methodological Approaches to Groundwater Protection for some contaminants (and various other processes such as adsorption and precipitation for others). In order to eliminate completely the risk of unacceptable pollution of a supply source, all potentially polluting activities would have to be prohibited (or fully controlled) within its entire recharge capture area. This will often be untenable or uneconomic, however, due to socio-economic pressure for development. Thus, some division of the recharge capture zone is required, so that the most stringent land-use restrictions will only be applied in areas closer to the source. This subdivision could be based on a variety of criteria (including: horizontal distance, horizontal flow time, proportion of recharge area, saturated zone dilution, and/or attenuation capacity), but for general application it is considered that a combination of (horizontal) flow time and flow distance criteria are the most appropriate. Special protection of a proportion of the recharge capture area might (under certain circumstances) be considered the preferred solution to alleviate diffuse agricultural pollution, but even here the question arises of which part it is best to protect. A series of generally concentric land-surface zones around the groundwater source can be defined, through knowledge of (and assumptions about) local hydrogeological conditions and the characteristics of the groundwater supply source itself. The three most important of these zones (Figure 2.2) are described below (Adams and Foster, 1992; Foster and Skinner, 1995). In the interests of supply protection, the zones will need to be subjected to increasing levels of control over land-use activities, which will tend to vary with local conditions and needs. (A) Total Source Capture Area The outermost protection zone that can be defined for an individual source is its recharge capture (or catchment) area. This is the perimeter within which all aquifer recharge (whether derived from precipitation or surface watercourses) will be captured in the water supply under consideration. This area should not be confused with the area of hydraulic interference caused by a pumping borehole, which is larger on the down-gradient side (Figure 2.1). Recharge capture areas are significant not only for quality protection but also in resource management terms, and in situations of intensive groundwater exploitation they might also be used as areas of resource conservation (or reserve) for potable supply. The total capture zone is determined in area by water balance considerations and in geometry by groundwater flowpaths. It is the zone providing the protected long-term yield. Thus, if the groundwater flow system is assumed (as is normally the case) to be in steady-state, its area will be determined by reference to the long-term average groundwater recharge rate. However, it should be recognized that in extended drought (when groundwater recharge is lower than average), the actual capture area will be Groundwater Quality Protection:a guide for water utilities, municipal authorities, and environment agencies larger than that protected. Moreover, in areas where the aquifer is confined beneath impermeable strata, the capture area will be located distant from the actual site of groundwater abstraction (Figure 2.2b). The protected yield is usually taken as the authorized (licensed) annual abstraction, but may be less than this where the licensed quantity is in practice: unobtainable, since it exceeds the hydraulic capacity of the borehole installation unsustainable, since it exceeds the available groundwater resource unreasonable, because it greatly exceeds actual abstraction. In such situations the protected yield is better based on recent abstraction rates, together with any reasonably forecast increase. (B) Microbiological Protection Area Preventing ingestion of groundwater contaminated with pathogenic bacteria, viruses, and parasites is of paramount importance. These pathogens enter shallow aquifers from some septic tanks soakaways, latrines, contaminated drainage or surface watercourses, and various other routes. Inadequately constructed wells are particularly prone to this type of contamination. However, in all but the most vulnerable formations, contamination via the aquifer route is prevented by the natural attenuation capacity of the vadose zone or the semi-confining beds. An inner protection zone based on the distance equivalent to a specified average horizontal flow time in the saturated aquifer has been widely adopted to protect against activities potentially discharging pathogenic viruses, bacteria, and parasites (Foster and Skinner, 1995), such as (for example) the spreading of wastewater and slurries on ' farmland. The actual flow time selected in different countries and at various times in the past, however, has varied significantly (from 10 to 400 days). 1 Published data (Lewis and others, 1982) suggests that the horizontal travel distance of pathogens in the saturated zone is governed principally by groundwater flow velocity. In all reported contamination incidents resulting in waterborne-disease outbreaks, the horizontal separation between the groundwater supply and the proven source of pathogenic pollution was (at maximum) the distance travelled by groundwater in 20 days in the corresponding aquifer flow regime. This was despite the fact that hardy pathogens are known to be capable of surviving in the subsurface for 400 days or more. Thus the 50-day isochron was contirmed a reasonable basis with which to define the zone (Figure-2.2), and this conforms with existing practice in many countries. This protection perimeter is perhaps the most important of all in terms of public health significance, and since it is usually small in size, implementation and enforcement are more readily achieved. Experience has shown that in fissure-flow aquifers (which are often very heterogeneous in hydraulic properties), it is prudent to establish a limiting criterion of SO-m radius. Part 8: Technical GuideMethodological Approaches to Groundwater Protection Figure 2.2 Idealized scheme of groundwater capture areas and transit- time perimeters around a waterwell and springhead a) unconfined aquifer watirwe~~ \ A I 1 I 200rnJ 10 days 10 y 00 h A MICROBIOLOGICAL b) locally confined aquifer 50m precautionary ,atemell ("0 50-doyzone) limit of confining - -t I I 500 days 10 years00 c) unconfined spring sourc -- iO days ys 10 years Groundwater Quality Protection: n guide for water utilities, municipal authorities, and environment agencies Moreover, even if aquifers are covered or confined beneath thick low permeability strata, a SO-meter-radius zone is also recommended as a precautionary measure (Figure-2.2b), in recognition of the uncertainties of vertical flow and to protect against subsurface engineering construction, which could compromise source protection. (C)Wellhead Operational Zone The innermost protection perimeter is that of the wellhead operational zone, which comprises a small area of land around the supply source itself. It is highly preferable for this area to be under ownership and control of the groundwater abstractor. In this zone no activities should be permitted that are not related to water abstraction itself, and even these activities need to be carefully assessed and controlled (Figure 2.3) to avoid the possibility of pollutants reaching the source either directly or via adjacent disturbed ground. All parts of the zone used for well maintenance activities should have a concrete floor to prevent infiltration of oils and chemicals used in pump maintenance. Fencing is also standard practice to prevent invasion by animals and vandalism. Specification of the dimension of this area is necessarily rather arbitrary and dependent to some degree on the nature of local geological formations, but a radius of at least 20-meters is highly desirable (Figure 2.2a). Detailed inspections of sanitary integrity, however, should be conducted over a larger area of 200 meters or more radius. (D)Further Subdivision It may be found useful to subdivide the total source capture area further, to allow gradational land-use controls beyond the microbiological protection zone. This can be done on the basis of a horizontal flow isochron of 500 days, for example (Figure 2.2a), to provide attenuation of slowly degrading contaminants. The selection of the time-of- travel is somewhat arbitrary. In reality such a perimeter is most significant in terms of providing time for remedial action to control the spread of persistent pollutants (at least in cases where a polluting incident is immediately recognized and notified) and is thus sometimes called the source inner-defensive zone. Furthermore, a horizontal flow isochron of 10 years or more (Figure 2.2a) is sometimes substituted for the perimeter of the total capture area in high-storage aquifer systems with complex boundary conditions andlor abstraction regimes, where the former will be of less complex shape and subject to less scientific uncertainty. wctors Controllinq Shape ot Lones Most protection zone delineation has to assume that steady-state groundwater flow conditions effectively exist. On this basis the factors controlling the actual shape of the various zones to be delineated are summarized in Table 2.1. Part B: Technical Guide methodological Approaches to Grocindwater Protection Figure 2.3 Actual examples of wellhead completion for major public water supply boreholes I a) well-designed, drained, and maintained wellhead operational zone in rural wooded area b) inadequatelysized and protected wellhead operational zone threatened by agricultural irrigation with urban wastewater Groundwater Quality Protection: a guide for tuater utilities, municipnl atrthorities, and environment agencies - Box 2.1 Operationof a long ing oundwater source p ection zone policy in Barbados This case study revears tne oeneftrs of early introduction or grouna er srdpply protection areas, nature of the aquifer /low regime and the pollution hazards are nor yet completely undmtood. Supplementary acrrvrzs can always be taken to subseqweutly reinforce existingprovisions. . I The Caribbean island of Barbados is very heavily urbanization with in-situ sanittr@n around the capital, dependent upon groundwater for its public water supply, Bridgetown, and leakage from commercial and dorncstic abstracting some 1 1 5 ' ~ from 17production wells in ~ d oil storage installations. 1 . a highly permeable karstic limestone aquifer of exnemt pollution vulnerability. 3 However, aaaitlc :am have subsequently merged (Chilton and othcra, 1990)such as: The potential impact of urban development and thc - the replacement of traditional extensive sugar-cane Z great strategic importance of groundwater supplies led cultivation with much more intensive horticultural 8 the Barbados government to establish special protection cropping involvillg much higher fertilizer and pesticide g areas around all of its public-supply wells about 30 applications years ago. The perimeters of these protection areas are - illegal disposal of industrial solid waste disposal by e defined on the basis of average groundwater travel times fly tipping in abandoned small timestone quarries and 2 to the wells, and the range of restrictions imposed is effluent disposal down disused wells. 3 1 summarized in the table below. These for the most *shave now been introduced to control and to V) have been successful in conserving water supply qualit) itor such activities. At the tune of introducing the policy, the main hazards- to eroundwater was perceived to be the spread of rrincipal features of development control zones I Zone Definition of Maximum Depth of Domestic Industrial I Outer Boundary Wastewater Soakaway Pits Controls Controls , 1 300-day none no new housing; no new travel time allowed no changes to existing industrial wastewater disposal development m 2 600-day 6.5 m septic tank with separate soakaway all liquid 1 travel time pits, for toilet effluent and other domestic wastewater, no storm industrial waste runoff to sewage soakaway pits, no to disposal new fuel tanks specified by . Water Authority - 3 5-6 year 13m as above for domesticwastewater, fuel with maximum travel time . depths as for I - 4 other areas no limit no restrictions on domestic wastewater domestic waste disposal, fuel tanks approved subject to leakproof design 'r . - \ Part B: Technical Guide MethodologicnlApproaches to Groundwater Protection I Table 2.1 Factorsdetermining the shape and extension of - groundwater supply protection areas* PROTECTIONAREA CONTROLLING FACTORS L Uveralt Location and Shape aquifer recharge and Row regime (recharge areahoundaries, natural discharge areas, hydraulic condition of streams*', aquifer boundaries, aquifer confinement, aquifer I hydraulic gradients) presence of other pumping wells/boreholcs** Area of Supply Capture Zone protectednicensed annual abstraction rate annual groundwater recharge rate(s)** Perimeter of Inner Flow-Time- aquifer transmtssivity dstribution Based Zones (SO-day and 500- aquifer dynamic flow thickness*** day isodron) aquifer (effectjve) dynamic porosity* * excludes manmade changes in groundwater regme due to urban construc~onand mining activities " these factors are generally time vuianr in nature and will provoke transient changes in the form of capture zones and isochrons, but average (or in s o m instances worst case) values arc taken in steady-stateformulations "termed dynamic in view of the fact that in heterogeneous (and especially fissured) aquifers, only a pan of the roraL thicknessandlor porosity (and m somecases only a minor part) may be mvolved in the Haw regime ro the groundwater supply sourceconcerned Microbiological protection zones are generally of fairly simple geometry, tending to be ellipsoidal or circular in form reflecting the cone of pumping depression around an abstraction borehole. For fissured aquifers the areal extent of these zones is very sensitive to the values taken for effective aquifer thickness and dynamic porosity (Figure-2.4), while their shape is sensitive to aquifer hydraulic conductivity. The key factors determining the geometry of overall source capture zones are the aquifer recharge regime and boundary conditions (Adams and Foster, 1992); their shape can vary from very simple to highly complex. More complex shapes may be the result of variable groundwaterlriver interactions, the interference effects from other groundwater abstractions andlor lateral variations in hydraulic properties. Long narrow protection zones will be delineated where the supply source is located at large distance from aquifer boundaries andlor where the abstraction rate is small, the hydraulic gradient is steep and the aquifer transmissivity is high. Groundwater Quality Protection: a guide for water utllittes, municipal authorities, and environment agencies Figure 2.4 Sensitivity of 50-day transit-time perimeter to interpretation of fissured aquifer properties 50-day isochron (axial length40m) total groundwater Source capture area thickness (rn) effective .-: 0.40 porosity l~imitationsto SUDVIY Protection Area Conce~t The supply protection area (>IJA) concept is a simple and powerful one, which 1s readily understood by land-use planners and others who need to make the often difficult public decisions generated by groundwater protection policies. However, many technical challenges can be posed by those who demand either greater protection or less restriction, and the test of any concept is whether it deals fairly with these competing criticisms, in the context of the circumstances it has to address (Foster and Skinner, 1995). SPASare most easily defined and implemented for major municipal wells and wellfields in relatively uniform aquifers that are not excessively exploited, but it is a valuable and instructive exercise to attempt to define them regardless of local conditions and constraints. (A) Common Problems with Suggested Solutions There are a number of hydrogeological situations where the concept encounters significant complications: the most serious limitation arises when aquifers are subject to heavy seasonally variable pumping for agricultural irrigation or industrial cooling, since interference Part 0: Technical Guide Methodological Approaches to Groundwater Protection Figure 2.5 Effect of various types of hydraulic interference and boundaries on the shape and stability of groundwater supply capture areas (a) effect of intermittent abstraction when irrigation I \ V k 4 wells pumping public water-supply irrigation wells borehole (seasonalpumping) (continuous pumping) (b) effect of effluent river (c) effect of influent river uence via river I I public water-supply borehole I public water-supply borehole total groundwater supply capture area between pumping wells produces excessively complex and unstable protection ,I zones (Figure 2.5a); recourse to overall resource protection via aquifer vulnerability criteria may then be the only feasible approach for aquifers whose long-term abstraction considerably exceeds their long-term recharge, a condition of continuously falling groundwater levels and inherently unstable SPASarises the presence of surface watercourses gaining intermittently or irregularly from natural aquifer discharge can produce similar complications (Figure 2.Sb) where losing surface watercourses are present within the capture zone to a supply source, any potentially polluting activity in the surface watex catchment upstream of the recharge capture area could affect groundwater quality (Figure 2.Sc), although it will usually be impractical to include this catchment in the source protection area special problems arise, especially with rhe definition of recharge capture areas, in situations where the groundwater divide is at a great distance and/or the regional hydraulic gradient is very low, and it will often be necessary to adopt a cut-off isochron (of 10 years) Groundwater Quality Protection: n guide for water ~ltilities,municipal azrthorities, and environment ngencies the presence of multi-layered aquifers, where vertical hydraulic gradients may develop inducing vertical leakage between aquifer units; each multi-layered aquifer situation will need to be examined on a site-by-site basis and some simplifying assumptions on hydraulic behavior made where the annual variation of the source capture area is very large (asin low-storage aquifers), the maximum (rather than average) area might be more appropriate, and local modifications may thus be required small groundwater supplies (with yields of less than 0.5 MVd ) because in some situations their capture areas will be very narrow and of unstable locus. Some may regard the 50-day travel-time criterion as excessively conservative because it takes no account of the large time-lag during percolation down the vadose zone, but in reality this needs to be balanced against the following factors: the possibility of rapid preferential flow through fissures, which can significantly reduce the retardation normally associated with vadose zone transport the isochron is calculated using mean saturated flow velocities, derived from average local aquifer properties and hydraulic gradients, and in fissure-flow aquifers a proportion of the water will travel much more rapidly than the average some contaminants may enter the ground with significant hydraulic loading (via drainage soakaways) and others (such as dense immiscible organic solvents) may have physical properties that favor more rapid penetration into the ground than water there is significant scientific evidence that some more environmentally hardy pathogens (such as Cryptosporidium oocysts) can survive much longer than 50 days in the subsurface (Morris and Foster, 2000). (B)Case of Karstic Limestone Aquifers Flow patterns in karstic limestone aquifers are extremely irregular due to the presence of dissolution features (such as caves, channels, and sinks), which short-circuit the more diffuse flowpaths through the fractured media as a whole. Contaminants moving through such a system can travel at much higher velocities than those calculated by average values of the aquifer hydraulic properties on an "equivalent porous media" approach. This simplification can be valid if the scale of analysis (and modelling) is regional, and if known major dissolution cavities associated with faults, or other structural features, are included, but in other cases the assumption can be misleading. Where karstic features are present, they should be systematically mapped through field reconnaissance, aerial photograph interpretation, and (possibly) geophysical survey, at least in the vicinity of the springs or wells to be protected. Knowledge gained through local hydrogeological investigation (especially using artificial tracer tests andlor environmental isotopes) and speleological inspection should be also used on a site-by-site basis for protection area delineation, rather than using average aquifer properties and hydraulic gradients for the calculation. It must be accepted that major departures from Part B: TechnicalGuide Methodological Approaches to Groundwater Protection I Box 2.2 F --a Delineatiol moundwater SL ly protecti zor land-use planning in Esperanza, Argentina I The delineation or grounuwuter capture gether with the mapping of aquifer polrurrvn vumt.rrrr,r Ian essential component of water sotlrce "'ection and planning at the rnunicipd level. The town of Esperanza (Sante Fe Prov~nce)meets its WHPA semi-analytical method using groundwater travel water demand entirely from groundwater. Locally, the times up to 5 years, as a basis for recommending graduated semi-confined aquifer is intensively exploited not only to measures of aquifer pollution control and land-use restriction I r meet these demands, but also for agricultural irrigation (Paris and others, 1999). and for a neighboring industrial center. h w -- a The implementation of g r o u n d w ~source protection r The town's groundwater sources compri: areas, however, ISnot a straightforward task, and it mayk- - a wellfield in a rural setting, where nan t be strongly resisted by those industries for which severe ly regulations or restrictions exist use constraints or total relocation are proposed (as a result of - a number of individual wells within the urban area, their character). Such actions can prove difficult to achieve which has incomplete sanitary infrastructure and in view of their socioeconomic repercussions. Because of various ~ndustrialpremises and these considerat~onsand with the object of facilitating proved levels of groundwater source protection, the I -his situation, couplea u n an aqulrcr p o ~ ~ v I rernative strategy of relocating groundwater abstraction to .wulnerabilityrated asmouerate by the GOD LL151h~d~I~~ new wellfield outside the area of urban influence has been baggested the existence of a significant groundwater yloposed. The perimeters of protection for the proposed ' lollutionhazard and the need for the introduction.Qf wellfield would thtn be delineated, with legal provision and otectlon measujes including land-use plannin technical regulations being introduced to guarantee their effectiveness. A groundwater monitoring network would For this purpose a range of possible protection perimeters also be established for che early detection and remechation of were delineated for the 20 municipal Ils, employing the any potential prob - urban area-! :. .. . ., - , - - 1 e location of 5-year travel protection perimeters for Esperanza wellfields industrial - premises - - Groundwater Quality Protection: a guide for woter utilities, municipal authorities, and environment agencies Figure 2.6 Adaptation of microbiologicalprotectionperimetersfor the case of karstic limestone aquifers - clav-covered area ,m. - SO-day isochrol using average aquifer iydraulic propert~es additional 15-rn buffer zones normal zone geometry should be expected (Daly and Warren, 1998) and that known surface solution features at large distances from the supply source, and the surface water catchment draining to them, will also warrant special protection (Figure 2.6). 1 (C)Case of Spring and Gallery Sources In some places groundwater abstraction takes place from springs, that is from points of natural discharge at the surhce. Springs present special problems for protection area delineation in that the abstraction is governed by natural groundwater flow driven by gravity. The size of the capture area is thus dependent on the total flow to the spring, rather than the proportion of the flow actually abstracted. Springflow may be intermittent, reducing drastically or even drying-up in the dry season as the water table falls. Springs often occur at the junction of geological discontinuities, such as lithology changes, faults or barriers, the nature and extent of which may be at best only partially understood. Moreover, there may also be considerable uncertainty on the actual location of springs, given the presence of infiltration galleries and pipe systems. Inevitably for all these cases, rather approximate, essentially empirical, and somewhat conservative assumptions have to be made in the delineation of protection perimeters (Figure 2.2). The delineation of protection zones around well sources can also be complicated by the presence of galleries (or adits), which distort the flow-field by providing preferential pathways for water movement; empirical adjustment is normally the method used to Part 0: Technical Guide Methodological Approaches to Groundwater Protection deal with this problem, although numerical modelling may also be an aid where sufficient data are available. (D)Implementation in Urban Settings The concept of groundwater supply capture areas and flow zones is equally valid in all environments, but substantial problems often occur in both their delineation through hydrogeological analysis and their implementation as protection perimeters in the urban environment. This results from the complexity of aquifer recharge processes in urban areas, the frequently large number of abstraction wells for widely differing water uses and the fact that most of the SPASdefined will already be occupied by industrial andlor residential development. Nevertheless, the zones delineated will serve to prioritize groundwater quality monitoring, inspection of industrial premises and groundwater pollution mitigation measures (such as changes in industrial effluent handling or chemical storage and introduction of mains sewer coverage in areas of high aquifer pollution vulnerability). r ,Methodsfor Definition of ProtectionZone Perimeters The delineation of perimeters of source protection zones can be undertaken using a wide variety of methods (Table 2.2),ranging from the oversimplistic to extremely elaborate. Historically, arbitrary fixed-radius circular zones and highly simplified, elliptical shapes have been used. However, due to the obvious lack of a sound scientific foundation, it rri w was often difficult to implement them on the ground, because of their questionable reliability and general lack of defensibility. Table 2.2 Assessment of methods of delineation of groundwater supply protection areas t - t METHOD OF DELINEATION COST lowest 1,- Arbitrary Fixed/Calculated Radius , Simplified VariabIe Shapes Analytical Hydrogeological Models Hydrogeological Mapping Numerical Groundwater Flow Models (with particle tracking ', routines for flowpath definition) , , highest most-. II Groundwater Quality Protection: n guide for toatel-utilities, mrrnicipnl artthorities, and environment ngencies Emphasis will thus be put here on two methodological options: simple, but scientifically based, analytical formula, tools, and models more systematic aquifer numerical modelling but the choice between them will depend more on hydrogeological data availability than any other consideration. In both cases it is essential to reconcile the zones defined with local hydrogeological conditions, as depicted by hydrogeological maps. The delineation process is highly dependent upon the reliability of the conceptual model adopted to describe the aquifer system and on the amount and accuracy of data available. However, the geometry of the protection zone defined will also be influenced by the method used for its delineation. It must be remembered that the delineation of protection perimeters, like the groundwater regime it operates on, is a dynamic system. No zone is immutable, because groundwater conditions may physically change or because new hydrogeological data may come to light that enable the aquifer to be more accurately represented. Equally, while accepting that many groundwater flow systems show complex behavior in detail (especially very close to wells), such local complexities are less critical at the scale of protection zone delineation. And in most situations, existing simulation techniques applied to sound aquifer conceptual models provide acceptable results. In general terms the reliability of source protection areas decreases with increasing time of groundwater travel in the aquifer. For example, the SO-day flow-time perimeter usually shows little variation between different methods of delineation, but the 10-year flow-time perimeter can vary by many ha's or even km2 with great divergence of shape. Recent developments have made groundwater models more widely available, more user-friendly and with improved visual outputs. Several public domain codes, such as the analytical model WHPA can now be downloaded from websites. And user-friendly interfaces such as FLOWPATH or Visual MODFLOW are now available for widely tested numerical flow models, such as MODFLOW, incorporating particle tracking techniques such as MODPATH (Livingstoneand others, 1985).As a result, hydrogeologists worldwide have easier access to sophisticated, yet easy to use, modelling techniques (Table 2.3). ' (A) Analytical versus Numerical Aquifer Models Analytical tools and models apply relatively simple analytical formula to simulate groundwater flow, normally in two dimensions. Models such as WHPA are easy to use, require little information, and many codes are available free on websites. However, analytical models are essentially limited to various assumptions (such as homogeneous aquifer properties and thickness, infinite aquifer extent, etc.) that prevent their use in more complex field conditions. They are, however, a good option for areas with limited hydrogeological data and relatively uniform aquifer systems. Part 8: Technical Guide methodological Approaches to Groundwater Protection Table 2.3 Usefulwebsite addresses on numericalgroundwater mdetlina for source wotctiotl nternational Association of http://www.iah.org/weblinks.htrn#softw Hydrogeologists http://www.mines.edu/igwmd ational Groundwater Association http://www.ngwa.org/ A Center for Subsurface http://www.epa.gov/ada/csmos.html SGS Water Resources Applications http://water.usgs.gov/software/ Numerical models are technically superior in that they can accommodate complex variations in aquifer geometry, properties, and recharge patterns, thus giving results closer to reality. However, they do require more data and are more time- consuming. Numerical aquifer modelling is recommended for areas where reasonable hydrogeological data are available and hydrogeological conditions cannot be readily simplified to the point required for the utilization of analytical modelling codes. Furthermore, numerical models can be readily used to evaluate the effects of uncertainties on the shape and size of protection zones and as predictive tools to assess future abstraction scenarios and hydrological system impacts. Such models may be based on finite difference or finite element codes. Finite difference methods use variable-spaced rectangular grids for system discretization, and are easy to understand, computationally stable and widely used, but may encounter difficulties in adjusting to complex geological boundaries. Finite element codes use triangular or prismatic elements that adapt well to irregular geology, but localized mass balance problems may occur. Where possible numerical aquifer models, employing a particle-tracking routine, are preferred. In these the movement of groundwater toward a source during pumping can be tracked numerically in small time-steps. Particle tracking produces flowlines emanating from the source in different directions, and the total capture perimeter under steady-state flow conditions is determined by the extent of the pathlines at infinite time and must continue to a point of zero flow velocity or the edge of the area under study. Particle tracking techniques form the basis for protection zone delineation, since most particle tracking codes are able to undertake velocity calculations within the flow-field, Groundwater Quality Protection: n guide for runter utilities, municipnl authorities, and environment agencies permitting isochron definition. It should be noted, however, that only advective (nondispersive) flow is simulated by particle tracking codes. (B) 2-D versus 3-D Aquifer Representation In order to apply numerical models to represent actual aquifer systems several simplifications are made. One of the most common is the transformation of a complex three-dimensional system to a simplified two-dimensional model, since in most cases there are not enough hydrogeological data (in terms of aquifer vertical permeability values and hydraulic head variations) to characterize and calibrate the vertical groundwater flow components. Given this and the fact that most aquifers are relatively Figure 2.7 Comparison between total capture area of idealized wells with shallow and deep intake in an unconfined aquifer showing the theoretical influence of vertical flow (a) shallow well in unconfined aquifer plan projection of capture zone = recharge area (b) unconfined deep well Part 8: Technical GuideMethodological Approaches to Groundwater Protection thin compared to their aerial extension, two-dimensional models are usually adequate and much more commonly used. However, in cases where vertical fluxes are important, two-dimensional flow modelling may overestimate the dimensions of capture zones, and therefore produce larger protection areas (Figure 2.7).Thus three-dimensional flow models are, in the future, likely to be increasingly used for complex aquifer systems if sufficient data are available. (C)Practical Considerations There are a number of distinct steps in the process of protection zone delineation. The most important stage in the whole process is probably data acquisition. Information is required not only on aquifer properties, but also on well construction, source operational regime, groundwater levels, recharge processes, and rates, and the aquifer interaction with surface watercourses. N o source protection zones can be delineated in isolation, and all require the consideration of the groundwater unit involved, at least to a radius of 5 km and more normally 10 km. When the basic data have been compiled, all available information should be synthesized into a conceptual model with the objective of providing a clear statement of the groundwater setting. This can then be used either as the basis for analytical zone definition or to guide the numerical modeller in setting up a simulation of the actual groundwater conditions. The choice of delineation technique will be a function of: the degree of understanding of the groundwater setting involved the operational importance of the groundwater supply concerned the human and financial resources available. Integrated GIS and databases provide a useful means of organizing the data within a single system, and provide the visualization powers to cross-check for inconsistencies and to model geographically distributed data. [ Dealingwith Scientlfk Uncertainty A numerical aquifer model can only be as good as its input data and the conceptual understanding of the groundwater flow regime. The size, shape, and location of source protection areas is largely controlled by hydrogeological parameters, which are often inadequately quantified. It follows that confidence in the predicted zones will be limited by uncertainty in the parameters involved. Models have to be calibrated by comparing model outputs to observed aquifer head conditions. A sensitivity analysis should be performed, in which key input parameters are systematically varied within reasonable ranges, and the effects of such variations on capture zone and flow time perimeters established. Themost rigorousapproach tosensitivityanalysisisto useaMonte Carlo (statistically based) approach, to define the maximum protection perimeter, which is the envelope of all Groundwater Quality Protection: n guide for wnter utilities, mcrnicipal authorities, and environment agencies Figure 2.8 Practical approach to incorporation of hydrogeological uncertainty into delineation of groundwater source protection areas -best estimate of total source capture area Izone of confidence (in all predicted capture areas) I zone of uncertainty (remaining area falling in at least one predicted capture area) --- perimeter of microbiological protection area (50day isochron) aquifer numerical model boundary U pumping well credible zones. By itself this approach is only likely to be acceptable in public policy terms where protection of groundwater is of overriding importance. In most circumstances, however, there are balances of interest to be struck that do not accept a zero-risk approach. The question of uncertainty must not be dismissed, however, because ~tis important that stakeholders understand the basis on which protection zones are defined. The numerical groundwater model used will be based on the best estimate of parameter values, and the best-fit protection zones defined are the only ones to meet the groundwater balance criterion. However, any model must inevitably be open to uncertainties, because it is physically impossible to verify in the field all the parameters represented by the simulation. The most critical variables affecting protection zone geometry are aquifer recharge rate, hydraulic conductivity, and effective porosity (Table 2.1). Best estimate and credible limit values for each of these variables can be I determined from available data and all combinations that achieve acceptable hydraulic ' head distributions are used to compile an envelope for each protection areas. From this , envelope the following can be defined (Figure 2.8): Zone of Confidence: defined by the overlap of all plausible combinations Zone of Uncertainty: the outer envelope formed by the boundaries of all plausible combinations. The parameters usually varied to allow the construction of the two zones are aquifer recharge and hydraulic conductivity. Acceptable ranges of these two parameters are Part 6:Technical Guide Methodological Approaches to Gro~indwaterProtection established by varying them systematically around the best estimate value, running the model, and noting the bounds within which the calibration targets are satisfied. Sensitivity runs, using parameter values from within the acceptable range, are subsequently carried out to compile the above zones. In a typical, well-calibrated model, recharge and hydraulic conductivity multipliers to the best estimate in the range 0.8-1.2 and 0.5-5.0, respectively, are applied universally across the model. An additional set of model runs using multipliers for effective porosity normally in the range 0.5-1.5 are carried out; the resulting travel-time zones are invariably more uncertain than the source capture area, because of the influence of this additional uncertain parameter. New automated parameter estimation programs (such as MODFLOW-P or PEST) are becoming an integral part of conducting systematic parameter uncertainty analysis. These inverse-model routines use complex algorithms to estimate the best input parameters for matching observed heads and fluxes. Professional judgement is essential in using such codes, however, since no hydrogeologically based interpretation is performed by them. Overall parameter uncertainty should be a major consideration when delineating groundwater capture zones, and the identification of those areas that are definitely (or possibly) contributing to a given supply source is an important tool in the definition of groundwater protection strategies. However, it must be noted that the methodology described above does not take account of errors arising from the use of inappropriate conceptual and/or numerical aquifer models, and expert judgement in this regard remains critical to overall zone modelling and uncertainty assessment. erimeter Adjustment and Map Production Once groundwater source protection zones have been delineated, the results should be inspected to assess whether adjustments are needed. Empirical adjustments are often required to provide protection zones that are both robust and credible in application. The output from the delineation process has to be translated into final source protection area maps, which can be superimposed on aquifer vulnerability maps for the purpose of groundwater supply pollution hazard assessment. This stage involves a sequence of modifications to the computed outputs, which experience has shown is probably best carried out with CAD software. The general sequence is as follows: final checks that the zones meet the minimum criteria in the definitions adjustment of boundaries to deal with problems of scale, and where possible, to make model boundaries conform with actual field property boundaries map production and reproduction, at scales in the range 1-25,000 to 100,000. When drawing protection zone boundaries, actual hydrogeological features should be used rather than model boundaries wherever possible. A sound general convention is to Groundwater Quality Protection:a guide for zuater utilities, municipal authorities, and environment agenciec draw and label actual boundaries where these are known and indicate model boundaries where they are indistinct, with suitable labelling to make this clear to the map user. 8 , 1 A further degree of judgement is often required when dealing with confining layers; I where there is a proven, substantial confining layer around a source, the microbiological protection zone is limited to a radius of SO-meters. However, where there are known or planned major manmade subsurface structures (such as road tunnels or mine access shafts) the full SO-day zone should be shown. Where a low permeability confining layer ( or cover occurs around the source, its extension is identified on protection zone maps using hatched shading, to indicate some uncertainty especially if it was not taken as an area of zero rainfall recharge in the numerical modelling. I Protection zones with long thin tails may arise due to pumping interference from other boreholes andlor from the imprecision of computer-model zone delineation. Wherever such features arise, they should be truncated at a minimum radius of SO-meters. This is MethodologicalApproaches to Groundwater Protection Inventory of Subsurface IContaminant Load In m y program of groundwater quality protection, knowledge of potential sources of contamination is critical becuuse it is t k e that generate the mission of contaminants into the subsurface environment. This cbapter presents a systematic approach to the szrrvey of subsurface contaminant load. (CommonCauses of GroundwaterPollution General review of known incidents of groundwater pollution leads to the following important observations, which are of relevance despite the fact that most published work refers to the more industrialized countries and may not be fully representative of those in the earlier stages of economic development: a large number of anthropogenic activities are potentially capable of generating a significant contaminant load, although only a few types of activity are generally responsible for the majority of serious cases of groundwater pollution (Table-3.1) the intensity of aquifer pollution is not normally a direct function of the size of the potentially polluting activity on the overlying land surface; in many instances smaller industrial activities (such as mechanical workshops) can cause a major impact on groundwater quality. These are widely distributed, often use appreciable quantities of toxic substances, sometimes operate outside formal commercial registers or are clandestine, and thus not subject to normal environmental and public health controls more sophisticated, large-scale industries generally exert more control and monitoring over the handling and disposal of chemicals and effluents, to avoid off-site problems due to inadequate effluent disposal or accidental spillages of stored chemicals Groundwater Quality Protection:a guide for water ~~trlities,municipal a~rthorities,and environment agencies Table 3.1 Summary of activities potentially generating a subsurface contaminant load -n LU CHARACTER OF POLLUTIONLOAD u 3 distribution main types of hydraulic soil zone TYPE OF ACTIVITY category pollutant surcharge bypass I b-- (+ indicates increasingimportance) rban Development P nsewered sanitation u/r P-D n f o c + eaking sewers (a) u P-L o f n t + sewage oxidation lagoons (a) u/r P o i n t ++ sewage land discharge (a) ulr P-D n s a f t + sewage to losing river (a) L I / ~ P-L o 0 f t ++ leaching refuse landfiwtips (a) u/r P a s h r fuel storage tanks u/r P-D t highway drainage soakaways u/r P-D S t + I Indusrrial Production leaking tanks/pipcIines (b) ti P-D t h accidental spillage3 u P-D t h process waeerteffluent lagoons u P t o h s effluent land discharge u P-D t o h s effluent3 to lasing river u P-L t o b s leaching residue tips ulr P o h s t soakaway drainage d r P t h aerial fallout u/r D S t I Agricultural Productiso (c) a) crup cdtivatiun -with agrochemicals n t -with irrigation n t s + -with sludge/slurry n t s o 1- with wastewater irrigation n t o s f + lvestock rearindcrop processing P ++ I - effluent lagoons r f o n t - r P-D n r o f t I T effluent land discharge effluent to losing river r P-L o n f t ++ Mineral Extraction hydraulic disturbance rlu P-D s h drainage water discharge r/u P-D h r process water/sludge lagoons r/u P h s leaching residue tips du P s h (a) u n inclrdc industrial componemr n nwrient compounds t ruxic micro-organisms + (b) can also occur in nonindustrial areas f fecal pathogens increasingsignificance (c) intdcation presents main pdlution &k o overall organic load idr urbadrural a saliniry PlUD point/line/diffuse h heavy metals - Part B: Technical GuideMethodologtcal Approaches to Groundwater Protection because of unstable economic conditions, it is relatively commonplace for small industrial enterprises to open and close over short time periods, which complicates the identification and control of potentially polluting activities and may leave a legacy of contaminated land the quantity of potentially polluting substances used in industry does not bear a direct relationship with their occurrence as groundwater contaminants, and it is the subsurface mobility and persistence of contaminant species that is the key factor (Table-3.2) Table 3 3 Most eommon types of prwndwuter cantaminantfound during wrveys in industrial nations (Duijvenboden, 1981) Ponution Source Types af Contaminant Coal Gas Works aromatic hydrocarbons (BTEXgrorrp) phenols, cyanide I Waste Tips and variable, often ammonium, chlorinated Sanitary Landfills hydrocarbons, heavy metals, atkylbenzene, domestic/industrial pesticides, etc. Metal Industries chlorinated hydrocarbons, heavy metals Hydrocarbon aromatic hydrocarbons (BTEXgroup), Storage and Handling lead Chemical Plants wide range of halogenated and aromatic hydrocarbons, phenols, alkylbenzenc, etc. Paint Factories aromatic hydrocarbons (BTEX group), chlorinated hydrocarbons - b) USA: 546 monitoring sites on priority aquifers (Ref. ASTM, 1995) Types of Contarninant Frequency of Occurrcoce (%) trich!oroerhy!ene lead toluene benzene polychlorinated hiphenyts chloroform tetrachleroethylcne phenols arsenic chromium - Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies relatively small amounts of more t o x ~ cand persistent chemical compounds are capable of generating large groundwater contamination plumes, particularly in aquifer systems characterized by high groundwater flow velocities the nature of the polluting activity (particularly in terms of contaminant type and intensity) can, in some cases, exert an overriding influence on the groundwater qual~tyImpact regardless of aquifer vulnerab~lity. It is therefore possible to conclude that certain sorts of anthropogenic activity, which tend to be associated with specific contaminant types, represent the greatest threat to aquifers. Thus a systematic inventory and classification of potential contaminant sources is a key step in programs of groundwater pollution hazard assessment and quality protection. [Bask Data Collectfon Prucdurtr (A) Designing a Contaminant Load Inventory Drawing up an inventory of potentially polluting sources includes systematic identification, siting, and characterization of all such sources, together with obtaining information on their historical evolution where appropriate and feasible. Such information will serve as a foundation for the assessment of which activities have the greatest potential for generating a potentially hazardous subsurface contaminant load. There is a common basis for all studies of this type, but local socio-economic conditions will also exert a significant influence on the approach that can and should be adopted. The inventory of potentially polluting activities (Figure-3.1) can be divided into three stages (Zaporozec, 2001): Id Figure 3.1 Development of an inventory of potential sources of subsurface contaminant load Identification of identification of Data S o w s Conrideration ol Determination of (1) lnventory Design lnvent~ryObjectivesand and Assessment ot Financial Rso~rrcesand Scopeof Inventory and . Area Characteristics Available Data Available Personnel Selection of Methods i 1 4 (2) lnventory Organization of Inventory Inventory d Contamination Verification of Data and Assessment of Needsfor lmplementatioll Team and Preparation of Source and Existing PceUmlnary C!&slf'~ationand Additional Date and M a p and Proformas Contamination Ranking of Sources Completion of Field Survey .. - (3) Evaluation Survey Organization and PolhrtimContFol Evaluation of Data bring a#Swrccs and FindInventory 1 Recornnre~lon~ Part B: Technical Guide Methodological Appronches to Groundtuater Protection inventory design, which includes the identification of information sources, the available financial budget, the level of technical personnel required, and the basic survey method inventory implementation, which includes the organization of the survey, the preparation of survey proformas, and the actual process of data acquisition suwq evaluation, which includes the analysis of data generated, including verification of its consistency and reliability, the classification of polluting activities, and the construction of a database that can output information in map or GIs form. The identification of information sources is particularly important to the work. In many instances most of the relevant data are held by provincial/rnunicipal government organizations and by the private sector. Previous studies for other purposes can be valuable sources of summary information, as can telephone directories (Yellow Pages) and listings of industrial boards and associations. Archive aerial photographs and satellite images are a valuable basis for the generation of land-use maps, including historic changes. It is essential that the approach to identification of potential pollution sources be fairly conservative, because it would be wrong to discard or downgrade activities just because available information was insufficient. There is a range (Figure 3.2) of inventory approaches (US-EPA, 1991): from exclusively desk-top evaluation of secondary data sources to basic field reconnaissance, in which teams survey selected areas to verify the existence of potential contamination sources. Figure 3.2 Approaches to data collection for surveys of potential groundwater pollution sources Agency Files 1 DOOF-to-Door '11Pu~iishednformatim & Archives Field Seafches I Maps, Air Photos, I Satenite Images + CLASSIFICATIONAND RANKING OF POTENTIAL GROUNDWATERCONTAMINRTIONSOURCE5 Groundwater Quality Protection: a guide for water utilities, municipnl arrthorities, and environment agencies The type of inventory and the level of detail required has to be a function of the ultimate objective of the work program, the size of the area under study, the range of industrial activities present, the availability of existing data, the financial budget provision, and the technical personnel available. The process of inventory ought to be undertaken on the basis of clearly defined, measurable and reproducible criteria, such that it is capable of generating a reasonably homogeneous dataset. For this reason it is preferable to base the design survey proformas and data-entry systems on a list of standardized questions and answers. As far as possible, some cross-checking of the consistency of information should be included. (B) Characteristics of Subsurface Contaminant Load From a theoretical viewpoint the subsurface contaminant load generated by a given anthropogenic activity (Figure 3.3) has four fundamental and semi-independent characteristics (Foster and Hirata, 1988): the class of contaminant involved, defined by its probable persistence in the subsurface environment and its retardation coefficient relative to groundwater flow the intensity of contamination, defined by the probable contaminant concentration in the effluent or leachate, relative to the corresponding WHO guideline value for drinking water, and the proportion of aquifer recharge involved in the polluting process the mode of contaminant discharge to the subsurface, defined by the hydraulic load (surcharge) associated with contaminant discharge and the depth below land surface at which the contaminated effluent or leachate enters is discharged or generated the dtrration of application of the contaminant load, defined by the probability of contaminant discharge to the subsurface (either intentionally, incidentally, or accidentally) and the period during which the contaminant load will be applied. (C)Practical Survey Considerations Ideally, information on each of the above characteristics for all significant potentially polluting activities is required. It would be even better if it were possible to estimate the actual concentrations and volumes of pollutant discharge to the subsurface. However, as a result of the great complexity, frequently high density, and considerable diversity of potential pollution sources, this ideal is not achievable in practice. Nevertheless, the ideal data requirements (Figure-3.3) should not be ignored because they constitute the rational basis for a detailed study of subsurface contamination load, including effluent inspection and sampling and leachate monitoring, where detailed follow-up is justified (Foster and Hirata, 1988). More generally, all techniques of contaminant inventory and classification are subject to significant imperfections and limitations. Nevertheless, because of the impossibility of controlling all polluting activities, it is essential that a method be found that is capable of identifying those that Part B: Technical Guide Methodological Approaches to Groundwater Protection Figure3.3 Characterization of componentsof subsurface contaminant load (increasingscale ofpotential impact is indicated by the darker shading) a) class of contaminant CONTAMINANT DEGRADATION slow 1 - 3 i d -- for aerobic alkalinesystems, but with changes for: Eh or pH falling b) intensityof contamination RELATIVE POLLUTIONCONCENTRATION(to WHO Guideline Value) 1oO 1o3 1o6 0.010h 0.1% 1.O% 10% u r k n unseweredsanitation 100% c) mode of contaminant disposition HYDRAULICLOAD 10 100 1000 10 000 (mm/a) I I 0.01 0.1 1 10 100 1000 mm/d /-I ;rnd:;ii , I !i E continued . . . Groundwater Quality Protection: n guide for water utilities, municipal nuthorities, and environment agencies d) duration of contaminant load I 0 50% 100% PROBABILITY OF LOAD present the greatest likelihood of generating a serious subsurface contaminant load, so that priorities for control can be established. Because of the frequent complexity in detail of land occupation and use, and related potentially polluting activities, clearly defined data collection criteria are required and special attention needs to be paid to the following: adjusting the scale of data representation to the available time and budget; it should be noted that general groundwater pollution hazard reconnaissance usually requires surveys at a scale of around 1:100,000 to superimpose on maps of aquifer pollution vulnerability, whereas more detailed scales 1:10,000-50,000 will be required for assessment and control of the pollution hazard to specific waterwells and springs @ ensuring that the outputs of survey work, in terms of the different origins of potential contaminant load, are at a compatible level of detail, with the aim of facilitating a balanced overall analysis of the area under study avoiding the indiscriminate mix of information of widely varying survey data, because this can lead to serious interpretation errors, and when this is not possible, to clearly record the limitations of the datasets in this respect taking a staged approach to the development of the register of potentially polluting sources, eliminating those with /OW probability of generating a significant subsurface contaminant load, before proceeding to more derailed work. eD (A) Spatial and Temporal Occurrence There are various published methods of assessing the pollution potential of anthropogenic activities, although few are directed to rating their potential to generate Part B: Technical Guide Methodological Approaches to Groundwater Protection iBox 3.1 Evaluationof t- surl !contaminant load generated by agricukural cultivation in S60 Paulo State, Brazil Diffuse sources oy suusuryace conzamznanr road are difficult to monitor directly for a number of practical reasons. Nevertheless, reasonable estimates of potential leaching losses can be made indirectlv oiven reliable datn nn ugrochemical usage, cultivation regime, and soil types. Slo Paulo State in Brazil, with an area of some hy a team fror ;bP, CETESB,DAEE, and & I W A 250,000-square kilomtttn and a population of .-' The lollowing data were avaihble and compiled: the million, is divided into some 560 municipal authurr~~rs. cul&vation type, the amount of various a p c h a i c a h -7 Groundwater resources play a major role in meeting, a@td by crop, the properti- of there agrocbernicalr, its urban, industrial, and irrigation water demand. - thtsuithrwteristicsintermsoftextureandorganic 4: ( Agricultural activity occupies 83 percent of the land content, and the rainfall regimetirrigation application In - j' with the cultivation of sugarcane, coffee, citrus, and tams of timing/volumc oi infiltration. . II -r - maize dominant. - - . Using these data, thc potenrial or nltrate cac ing was In 1990 this agricultura activity use some 2.59 estimated an the basis of the continuity of crop cover and tons of fertilizers (with phosphate applications being the generation and application of soil ttiiaatc compared especially high) and some 0.07 million tons of with plant rsquirmwnts. The pesticide-leach& b r d (by active ingredient), making it the most intensive was estimated m rhe b& of the types of compound agricultural area in Brazil. Additionally, the majority of used, their adsorption potential according to pertition I soils are acidic and some 1.10 million tons of lime 1 year +- coefficient, and soil organic nrbon content (Hiramand I are applied for soil conditioning and to others, 1995).With data on a more detailed nah r ing. gber resolution assessment wnu e possible. I For the purpose of measuring Statistical summary of assessments of potential hazard, the use of agrochemicals for 1 intensity of subsurface contaminant load I 0 r I I A3ssessed in terms of its potential to generate a s u b ~ u r f a c e ~ - ~ a elevated lmoderate reduced I contaminant loacl +'-roughsoil leaching. This was done Class of Principal Main Crops Treated Agrochemical Types tYPe area (ha) pesticide metam~dophos cotton 325,300 monocrophos soya 459,300 vamidoton beans 452,630 acephare herbicide dalapon soya 459,300 simazine sugar cane 1,752,700 atrazine benrazon 2,4-D nitrate N fertilizers sugar cane 1,752,700 citrus 769,000 pasture d a PESTICIDES HERBICIDES Groundwater Quality Protection: a guide for water utllitres, municipnl authorities, and environment agencies a subsurface contaminant load; more emphasis is generally put on their river or air pollution hazard (Foster and Hirata, 1988; Johansson and Hirata, 2001). The classification of potentially polluting activities by their spatial distribut~onprovides a direct and visual impression of the type of groundwater contamination threat they pose and the approach to control measures that is likely to be required: diffuse polhtion sources do not generate clearly defined groundwater pollution plumes, but they normally impact a much larger area (and thus volume) of aquifer point pollution sources normally cause clearly defined and more concentrated plumes, which makes their identification (and in some cases control) easier; however, when point-source pollution activities are small and multiple, in the end they come to represent an essentially diffuse source, as regards identification and control. Another important consideration is whether the generation of a subsurface contaminant load is an inevitable or integral part of the design of an anthropogenic activity (for example as is the case with septic tanks) or whether the load is generated incidentally or accidentally (Foster and others, 1993). Another useful way of classifying polluting activities is on the basis of their historical perspective, which also exerts a major influence on the approach to their control: past (or inherited) sources of contamination, where the polluting process or the entire activity ceased some years (or even decades) before the time of survey but there is still a hazard of generating a subsurface contaminant load by the leaching of contaminated land existing sources of contamination, which continue to be active in the area under survey potential future sources of contamination, relating to activities at the planning stage. (B)The POSH Method of Load Characterization It is necessary to take into consideration these various forms of classification during the survey of potential sources of subsurface contaminant load. However, for the type of simplif~edinventory proposed for the purposes of this Guide, it is convenient to characterize the potential sources of subsurface contaminant load on the basis of two characteristics: the likelihood of the presence of contaminants, which are known or expected to be persistent and mobile in the subsurface the existence of an associated hydraulic load (surcharge) capable of generating advective transport of contaminants into aquifer systems. Such information is not always readily available, and it is generally necessary to make the following further simplifying assumptions: associating the likelihood of the presence of a groundwater-polluting substance, with the type of anthropogenic activity (Tables-3.1 and 3.2) estimating the probable hydraulic surcharge on the basis of water use in the activity concerned. Part B: Technical Guide Methodological App~onchesto Gro~tndwaterProtection SUBSURFACE CON IP I'U JKLL LOAD POTENTlAl n-situ sanitation gricultural practicer Elevated mains sewer coverage less than ~ntensivecash crops and most 25 percent and population monocultures on well-drained soils in density above 100 personslha humid climates or with low-efficiency irrigation, intensive grazing on heavily fertilized meadows Moderate intermediate between above and below Reduced mains sewer coverage more traditional crop mtatiotu, extensive than 75 percent and pasture land, eco-farming systems, population density below high-efficiency irrigated cropping in 50 personska arid areas Thus the approach to assessment of potentially polluting activities used in this Guide- the so-called POSH method-is based on two readily estimated characteristics: the Pollutant Origin and its Surcharge Hydraulically. The POSH method generates three qualitative levels of "potential to generate a subsurface contaminant load": reduced, moderate, and elevated (Tables-3.3 and 3.4). (A) Diffuse Sources of Pollution Urban Residential Areas without Mains Sewerage In most towns and cities of the developing world, rapid urban population growth has resulted in large areas that are dependent upon in-situ systems (such as latrines, septic tanks and cesspits) for their sanitation (Lewis and others, 1982).Such systems function by liquid effluent percolation to the ground, and in permeable soil profiles, this results in aquifer recharge. As regards the solid fraction, it should be periodically removed and disposed off site, but in many cases it remains in the ground and is progressively leached by infiltrating rainfall and other fluids. The types of contaminant commonly associated with in-situ sanitation are the nitrogen compounds (initially in the form of ammonium but normally oxidized to nitrate), microbiological contaminants (pathogenic bacteria, viruses, and protozoa), and in some cases community synthetic organic chemicals. Among these contaminants, nitrates will always be mobile and often be stable (and thus persistent), given that in most groundwater systems, oxidizing conditions normally prevail. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencres w I \ hazard in Rio Cuarto, Argentina 1 I P The evaluation of aquifer pollution uulnerabilityprovides a framework within which to design and zmplement sur af subsurface contaminant load, and to use the results for assessrng groundwater pollution hazard, designing focused roundluater sampling cnmpoigns, and throsrgh these, prioritizing remedial actians. 'The town of Rio Cuarro (Cordoba),Argentina has a With the aim of confirming the population of some 140,000 who are dependent u p hazard assessment and of establishing a strategy fa,.,. - groundwater for all heir water supply requirements the problem that it presented, a detailed .. umanaging About 75 pcrcen~have access LO mains water suppl) groundwater quality stuh was undertaken in w o and the mains sewerage system has around 50 Tdistricts (Quintitas Golf and Villa Dalcar), neither of I percent coverage, with the remainder utilizing directly 'which yet have mains sewerage. Some 60 percent of I 1 abstracted well-water and in-siruwastewater disposal the samples analyzed proved to be unfit for human 1 respectively. lu consumption as a rsuIt of the elevated fecal colifonll counts, and in some cases both nitrate and chloride B- The town is underlaid by a largely unconfined were elevated in relation to background levels (Blasarin aquifer formed in very heterogenous Quaternarj -and others, 1999). 1 sediments, and its groundwater is of good naturrf luality appropriate for human consumption. TI-:-& The coexistence of domestic water supply wells ;OD methodology suggests that the aquifer pollution . -and in-situ sanitation facilities in areas of high rulnerability, however, ranger from moderare to high. e a q u i * r pollution wlnerabiliry was declared to be a iuperimposing the results of a systematic sanit~rion public balth risk, and priorities were, accordingly, ,urvey, it was predicted that the aquifer pollution -mended for the expansion of the mains water hazard varies spatially from very low to ext&dy m l p network and tb provement in the design of * . hrgh (Biasarin and others, 1993). *u sanitation 1 ;. * 1 QuintitasGolf (1 10 personsiha) Villa Dalcar (80 persons/ha) I AQUIFER (high rulni?roWry~ Part 0:Technical Guide Methodological Approaches to Groundwatev Protection Figure 3.4 Estimation of nitrogen load in groundwater recharge of areas with in-situ sanitation a) variation with Iand u b) variationwith I and f v 140 - - -- o 140 u 0 50 100 150 200 250 300 OO-~ 50 100 150 200 250 300 - p population density (personslha) -I ---I (mmla)for f = O.Su = 50 I/d/cap I (mmla), f = 0 . 5 ~ 250 I/d/cap (mmla)for f = 0 . 2 ~= 50 Ildlcap = - - a + WHO gu~del~nevalues for potable water: F,-5 maximum recommended Note: Variation with population density, natural rate of rainfall infiltration (I in mm/a), and the nonconsumptive portion of total water use (u in I/d/cap) is shown; f being the proportion of excreted nitrogen leached to groundwater. The presence of in-situ sanitation (together commonly with high rates of water mains leakage) often results in heavy hydraulic surcharging and high rates of aquifer recharge in urban areas, despite the general tendency for the land surface to be irnpermeabilized and rainfall infiltration to be reduced (Foster and others, 1998). Overall rates of urban recharge in developing nations are believed widely to exceed 500 mrn/a. In districts where mains sewerage cover is limited or absent, and where urban population densities exceed 100 personslha, there exists an elevated potential subsurface contaminant load (Figure-3.4), especially where in-situ sanitation units are improperly operatEd and maintained. However, in predominantly residential areas with extensive coverage of mains sewerage, this potential is reduced, despite the probable existence of leakage from mains sewerage systems (which only threatens groundwater quality locally). In many urban and periurban areas it is commonplace to find small manufacturing and service industries (including motor vehicle workshops, petrol filling stations, etc.), that often handle toxic chemicals (such as chlorinated solvents, aromatic hydrocarbons, etc.). In this case it is important to identify any areas where such activities may be discharging effluents directly and untreated to the ground (rather than to other means of disposal or recycling). Groundwater Quality Protection: a guide for water utilities, municrpal authorities, and environment agencies Data on population density (Table 3.3), together with the proportion of the urban area with mains sewerage cover, are generally available from municipal authorities. Moreover, in many instances municipal authorities or water service utilities have reliable information on which industries are connected to the sewerage system. However, in some cases it may be necessary to survey in the field, through direct inspection on a block-by-block basis. Agricultuval Soil Cultivation The agricultural cultivation of soils exerts a major influence on the quality of groundwater recharge, and also with irrigated agriculture the actual overall recharge rates (Foster and Chilton, 1998; Foster and others, 2000). Some agricultural soil cultivation practices cause serious diffuse contamination, principally by nutrients (mainly nitrates) and sometimes by certain pesticides. This is especially true in areas with relatively thin, freely draining soils (Foster and others, 1982; Vrba and Romijn, 1986; Foster and others, 1995;Barbash amd Resek, 1996).However, the other major plant nutrients (potassium, phosphate) tend to be strongly retained in most soils and not heavily leached to groundwater. It is of relevance here to note that a major U.S. national evaluation of the occurrence 01 pesticide compounds in groundwater (20 major catchments during 1992-96) showed: pesticidepresence in 48 percent of the 3,000 samples collected (Kolpin and others, 2000), but in the majority of cases at concentrations below WHO potable quality guidelines that in the phreatic aquifers of the maize and soya bean cultivation tracts of the mid-western states, 27-pesticide compounds were detected, and of the 6 most widely detected, no fewer that 5 were herbicide metabolites (partial breakdown products) the presence of alachlor derivatives was especially significant, since the parent compound was not detected, implying breakdown in the soil to a more mobile and persistent derivative pesticide contamination was widely found in urban areas, as a result of excessive application to private gardens, recreational facilities, sports grounds, and other areas. The types of agricultural activity that generate the most serious difiuse contamination of groundwater are those related to extensive areas of monoculture. More traditional crop rotations, extensive pasture land, and ecological farming systems normally present less probability of a subsurface contaminant load. Agriculture involving the cultivation of perennial crops also normally has much lower leaching losses than where seasonal cropping is practiced, because there is less disturbance and aeration of the soil and also a more continuous plant demand for nutrients. However, when perennial crops have to be renewed and the soil plowed, there can be major release and leaching of nutrients. There normally exists some correlation between the quantity of fertilizers and pesticides applied, and their leaching rates from soils into groundwater. Nevertheless, only a proportion of agrochernicals applied are leached, and since leaching results from a complex interaction between: Figure 3.5 Estimationof potential contaminant load in groundwater recharge from cultivated land a) nitrate (as N03-N) b) pesticide compounds quantity leached from cultural soil (kg/ha/a) (application rate leaching index*) hitiattancontinuity plowing rwsuetcy'm -- ."LA nitrate leaching index* pesticide leaching index* cultivation type soil properties rainfall and irrigation regime management of soil and agrochemical applications, it is difficult to provide simple methods for the estimation of leaching rates. Moreover, only a small proportion of the nitrate leached from soils is normally derived directly from the application of fertilizers in the preceding growing season. However, fertilization levels influence the level of soil organic nitrogen; from this level nitrate is released proportionally by oxidatjon, especially at certain times of the year and following plowing or irrigation. Values of leaching losses obtained from the literature indicate that up to 75 percent of the total N applied can be oxidized and leached to groundwater (although values of 50 percent are more common). In the case of pesticides, leaching losses rarely reach 5 percent of total active ingredient applied Groundwater Quality Protection: a guide for water utilities, rn~dnicipalauthorities, nnd enuironment agencies and more normally are less that 1 percent (Foster and Hirata, 1988). The factors that determine the rates of soil leaching from cultivated soils within this range are summarized in Figure-3.5 (Foster and others, 1991). Given the difficulty in making precise estimates of leaching losses, the classification of agricultural land in terms of its potential to generate subsurface contaminant load must begin by mapping the distribution of the more important crops, together with inventory of their fertilizer and pesticide applications. With these data it will usually be possible to classify the cultivated land area on the basis of likelihood that the farming activity will potentially generate a low, moderate, or elevated subsurface contaminant load. In some instances the total amounts of agrochemicals applied to a given crop are not known with certainty. In this case reasonable approximations can often be made through consultation with agricultural extension staff on recommended application rates, assuming that farmers are making correct use of the product concerned. If this type of approach is used, it is necessary to bear in mind that farmers commonly opt for specific products according to their local market availability and commercial publicity. If it is not possible to obtain the above information, then a further simplification can be used, based on a classification (Table 3.3) of: -0 probable levels of fertilizer andlor types of pesticide use the hydraulic load on the soil as a result of the rainfall and/or irrigation regime. Another frequent difficulty is the lack of reliable up-to-date information on the distribution of agricultural crop types, even where the total area planted to a given crop in any given year is known at municipality or county level. Moreover, in developing economies there are often rapid changes in agricultural land use. Often land-use maps are outdated and it is necessary to use more recent aerial photographs for such information if available. Satellite images can also be used, despite the fact that their resolution does not generally allow a close differentiat~onof crop types, but they have the advantage of being up-to-date and offering the possibility of studying trends in land-use change. One other aspect has to be considered, especially in the more arid climates, and this is agricultural irrigation with wastewater. Wastewaters invariably contain nutrients and salts in excess of crop requirements, and thus leads to significant leaching losses from agricultural soils. There also exists the risk of infiltration of pathogenic micro-organisms and trace synthetic organic compounds as a result of wastewater irrigation. Additionally, it must be kept in mind that the risk of pesticide leaching to groundwater from agricultural practices is not limited to their use at field level, since storage and use in livestocl< rearing can also lead to groundwater contamination, especially where such compounds are inadequately stored andlor handled. Part 0: Technical Guide Methodological Approaches to Groundwater Protection (B) Point Sources of Pollution Industrial Activities Industrial activities are capable of generating serious soil pollution and major (I contaminant loads on the subsurface, as a result of the volume, concentration, and range I, of chemical products and residues that they handle. In general terms, any industrial I:; activity is capable of generating a subsurface contaminant load as a result of the I emission of liquid effluents, the inadequate disposal of solid wastes (Pankow and others, 1984; Bernardes and others, 1991), and unwanted materials, together with accidents I1 involving leaks of hazardous chemical products (Sax, 1984). Compounds frequently 1 detected in groundwater contamination plumes related to industrial activities usually I show a close relationship with those used in the industrial activity, which in turn are :I directly related to the type of industry concerned (Table-3.5). The handling and discharge of liquid effluents is one aspect of industrial activity that merits detailed attention in relation to groundwater contamination. In industries located close to surface watercourses, direct discharge of liquid industrial effluents is often ) 0 n C Table 3.4 Classificationand ranking of point pallutlon sources under the POSH system V, s POTENTIAL FOR POLLUTIONSOURCE V) 5 SUBSURFACE F CONTAMINANT solid waste wastewater miscellaneous mining and oil F;t LOAD GENERATION disposal slres- lagoons urban exploration n Elevated industrial type 3 type 3 list, any all induseial type oilfield 0 waste, waste of activity handling 3, any effluent operations, 1 ;? unknown origin ,300 kgld of (except residential rnetaniferous hazardous sewage) if area mining chemicals >5 ha f Moderate rainfall >500mm/a type 2 list residential sewage gas filling stations, some mining/ ? with residential/ if area >5 ha, transportation quarrying g industrial type 11 u other cases not r o w s with regular of inert agroindustrial above or below traffic of hazardous materials .- wastes, all other chemicals cases Reduced rainfall <500mm/a type 1list residential, mixed cemeteries with residential/ urban, agro- industrial type 1/ industrial, and agroindustrial nonmetalliferous wastes mining wastewater if area <1ha * c~nraminatcdland fm abanctoned indmtrier should have same ranking ar inhsrry itself List 1 Industries: woodworking, food and beverage manufacnrrers, sugar and alcohol distilleries, non-metnllic rnnteriaI processing Lisa 2 Industries: rubber factories, paper and pulp mlk, textile factories, fertilizer manufacturers,electrical factories, detergent and soap manufacturers List 3 Industries: engineeringworkshops, oillgas refineries,chcmlca~/phmaceuticaWp~astic/pesticidcmanufacturers, leather tanneries, electronic factories, metal processing Groundwater Quality Protection: a guide for water utilities, municipal authorities, and enuironment agencipc VDUSTRIAL 'YPE Iron and Steel Metal Processing . Mechanical Engineering Nonferrous Metals e.. t / Nonmetallic Mmerals Petrol and Gas Refineries Plastic Products Rubber Products Organic Chemicals Inorganic Chemicals Pharmaceutical Woodwork Pulp and Paper Soap and Detergents Textile Mills Leather Tanning Food and Beverages Pesticides ........ . Fertilizers - ( Sugar and Alcobol 2 4 ** 0.. 11 '1 Therrno-Electric Power *** · · · 0.0 · orno 0. Lectric and Electronic 5-8 * · · · 0.0 0. 0.0 3 - -- 1 0. moderate probability of troublesome concentrations in process fluids andlor effluents 00. high 'Ow Source: Abstracted from BNA, 1975; DMAE, 1981; Hackman, 1978; Luin and Starkenburg, 1978; Nemerow, 1963 and 1971; Mazurek, 1979; US- EPA, 1977 and 1980, and WHO, 1982 and other minor unpublished reports. Part 8: Technical Guide Methodological Approaches to Groundwater Protection practiced, and in other situations the disposal of effluents through soil infiltration is sometimes used. Other than in cases where the industry concerned undertakes systematic effluent treatment, such practices will always present a direct or indirect hazard to groundwater quality. Moreover, where effluent storage and treatment is undertaken in unlined lagoons, these also represent a significant groundwater pollution hazard. The POSH classification of industrial activities in relation to their potential for generation of a subsurface contaminant load is based on (Table-3.4): the type of industry involved, because this controls the likelihood of certain serious groundwater contaminants being used the probable hydraulic szrrcharge associated with the industrial activity, estimated by the volume of water utilized. In terms of the type of industry, great emphasis needs to be put on the likelihood of utilizing appreciable quantities (say more than 100-kilograms per day) of toxic or dangerous substances, such as hydrocarbons, synthetic organic solvents, heavy metals, etc. (Hirata and others, 1991, 1997).In all such cases the index of subsurface contamination potential should be elevated, since factors like chemical handling and effluent treatment cannot be considered a result of the general difficulty in obtaining reliable data. Effluent Lagoons Effluent lagoons are widely used in many parts of the world for the storage, treatment, evaporation, sedimentation, and oxidation of liquid effluents of industrial origin, urban wastewaters, and mining effluents. Such lagoons are generally relatively shallow (less than 5-meters deep), but their retention time can vary widely from 1-100-days. Following the POSH classifications, the subsurface contamination potential of these installations depends on two factors: the likelihood of serious groundwater pollutants being present in the effluent, which is primarily a function of their industrial origin the rate of percolation from the lagoon into the subsoil, which is primarily a function of lagoon construction and maintenance (whether base and walls are fully impermea bilized). In a process of rapid assessment, it is difficult to obtain reliable estimates of the total volume of effluents entering and leaving the system. But studies of unlined lagoons (still the most popular form of construction in the developing world) show that infiltration rates are often equivalent to 10-20 milligrams per day (Miller and Scalf, 1974; Geake and others, 1987). However, while it is not easy to make full hydraulic balances for lagoons, it is possible to estimate whether they are generating significant recharge to underlying aquifers on the basis of their areal extension and hydrogeological location. In the majority of cases, it is not possible to obtain data on the quality of liquid effluents, but the likelihood of serious groundwater contaminants being present can be judged Groundwater Quality Protection: a guide for water utilities, municipal authorities, nnd environment agencies from the type of industrial or mining activity involved (Table-3.5). It must be borne in mind that many less mobile contaminants will be retained in sediments forming the lagoon bed; this is especially true of pathogen~cmicroorganisms and heavy metals. Lagoons receiving urban wastewater generally have a heavy load of organic material and pathogenic microorganisms, together with high concentrations of nutrients and sometimes salts. If the assoc~atedsewerage system serves nonresidential areas, it is likely to contain the effluents of small-scale industries (such as mechanical workshops, dry cleaning shops, printing works, etc.), and in such cases wastewater could contain synthetic organic solvents and disinfectants. The POSH classification approach to the assessment of the relative potential of wastewater lagoons to generate subsurface contaminant loads is given in Table-3.4, which uses easily obtained data on: the type of activity generating the wastewater and effluents involved . the area occupied by the lagoon(s). Solid WasteDisposal ' The inadequate disposal of solid waste is responsible for a significant number of cases ' of groundwater pollution (US-EPA, 1980; Gillham and Cherry, 1989). This is more prevalent in regions of humid climate where substantial volumes of leachate are generated from many sanitary landfills and waste tips, but also occurs in more arid climates where leachates will generally be more concentrated. The subsurface contaminant load generated h m a waste tip or sanitary landfill is a funccion of two factors: 1). the probability of the existence of groundwater contaminants in the solid waste I the generation of a hydraulic surcharge sukficient to leach such contaminants. The type of contaminants present is principally related to the origin of the waste and to (bio)chemical reactions that occur within the waste itself and in the underlying vadose zone (Nicholson and others, 1983). Evaluation of the actual qualiry of leachates requires a detailed monitoring program, but can a h be estimated in general terms on the basis of waste origin (urban residential, industrial, or mining) and the construction and age of the disposal facility. Calculation of the hydraulic surcharge necessitates a month!y hydraulic balance for the landfill, together with knowledge of the level of impermeabilization of its surface and bnx, even allowing for the fact that some ieachate will be gcners~dfrom the waste materials themselves. A classificarion of the relative potential ro generate a - - . subsurface contaminant load can be obtained by the interaction (Table 3.4)of: A, the origin of ttrewaste, which indicam the kkdy presence of grounduamcontaminants the probable hydraulic surcharge estimated frum the rainfall at the waste di~poralsire. .- &. A in some caws &origin ofthe mild waste is uncertain,as a r e s u ) t ~ over the types of residues received. In this case, it is a wise precaution to clasnify the solid waste disposal activity as generating a potentiany elevated subsurface contaminant bad, I regardless of the precipitation regime. Such a precawionaq approech ir not considmtd Part B: Technical Guide Methodological Approaches to Groundwater Protection excessive because small volumes of toxic substances (such as synthetic organic compounds) can cause major groundwater quality deterioration (Mackey and Cherry, 1996). Gas Stations Gas stations are responsible for a large number of cases of groundwater contamination (Fetter, 1988),although individual incidents are not major. Such installations are widely distributed and handle major volumes of potentially polluting hydrocarbons stored in underground tanks that do not allow visual inspection for leaks. The main sources of soil and groundwater pollution are corroded tanks, and there is a strong correlation between the incidence and size of leaks and the age of installed tanks (Kostecki and Calabrese, 1989; Cheremisinoff, 1992). There is a high probability that tanks more than 20 years old are seriously corroded and subject to substantial leaks unless they receive regular maintenance. Moreover, pipe work between tanks and delivery systems can become ruptured due to the traffic of heavy vehicles or due to initial poor q~~ality installation. Most gas stations measure hydrocarbon fuel levels at the beginning and end of every working day as a matter of routine, normally through electric level-measuring systems. These figures are compared to the volumes sold, as measured by discharge gauges. However, such measurements do not necessarily give a clear idea of subsurface leakage from tanks, because they are not especially sensitive, and relatively small losses can cause significant groundwater contamination plumes as a result of the high toxicity of the substances concerned. Regular standardized tests of tank integrity are a far better measure of the likely losses of hydrocarbon fuels. Losses due to tank corrosion can be significantly reduced if higher design, construction, operation, and maintenance standards are applied. In particular the use of steel or plastic tanks reinforced with glass fibers or double-walled tanks offer much greater security against leakage, and cathodic protection greatly reduces corrosion. Taking into account the small areas generally affected and the strong natural attenuation of hydrocarbon compounds, the presence of gas stations and storage facilities with underground storage tanks should be interpreted as a subsurface contaminant load source of moderate intensity, unless high design standards and regular maintenance are evident. An additional hazard will exist where gas stations are combined with auto repair shops that use large quantities of synthetic organic solvents and hydrocarbon lubricants, because these may be discharged to the soil without controls. Mining Activities and Hydrocarbon Exploitation Mining and hydrocarbon exploitation activities can cause important impacts on groundwater quality as a result of: hydraulic modifications to groundwater flow systems, either directly or indirectly, as a result of the construction and operation of both open-cast and subsurface excavations Groundwater Quality Protection: a guide for water utilities, municipalauthorities, and environment agencres + increase in the pollution vulnerability of aquifers, as a result of the physical removal of parts of the vadose zone or confining beds that provided natural protection disposal of mine drainage waters or saline hydrocarbon reservoir fluids, by land spreading, discharge to surfacewater courses, or in evaporation lagoons subject to percolation infiltration of leachate from mine spoil heaps 1. disposal of solid wastes and liquid effluents in abandoned mine excavations 1. operation of subsurface mines or oil wells when they are located immediately below important water supply aquifers imobilizationofheavymetalsandothercompoundsduetochangesingroundwater flow regime in mined areas and associated changes in hydrochemical conditions. As a result of the great complexity of these activities and the hydraulic changes they provoke, it is necessary to analyze them on an individual basis to assess their potential impact on groundwater quality. Thus no rapid assessment method can be recommended. However, at the preliminary evaluation level, it is possible to differentiate three principal groups of extractive industries, each of which have significantly different requirements in terms of evaluating the groundwater pollution hazard that they pose: quarrying of inert materials, such as those used for civil engineering construction where the principal concern is assessing the changes that mining activity may have caused to pollution vulnerability of underlying aquifers and their groundwater flow system + mining of metals and other potentially reactive deposits, where more attention needs to be paid to the handling of mining spoils, which in many cases can contain potential groundwater contaminants (such as heavy metals and arsenic), and the disposal of mine drainage waters that can be highly contaminating if not properly handled hydrocarbon fuel exploitation, where large volumes of saline formation water and other fluids are extracted during well drilling and operation, and-depending on their handling and disposal--can represent a major hazard for shallow aquifers in the areas concerned. Contaminated Land All major urban and mining areas have experienced historic changes in land use, and the closure of industrial and mining enterprises is a common occurrence especially in developing economies. The land abandoned by such enterprises can have high levels of contamination and can generate a significant subsurface contaminant load through leaching by excess rainfall. The existence of contaminated land not only poses a threat to underlying groundwater systems, but is also a health and environment hazard to those now using the land concerned. However, this latter topic is outside the scope of the current Guide. Part B: Technical Guide Methodological Approaches to Groundwater Protection Changes in land ownership and/or use can result in difficulties in obtaining detailed information on earlier activities and likely types1 levels of contamination arising. Old maps and aerial photographs are an important source of information in this respect, and the information they provide can sometimes be substantiated from local government archives. The classification and evaluation of contaminated land in terms of its likelihood to generate a subsurface contaminant load to underlying aquifers requires that the historical use be established. From the type of industrial or mining activity it is possible to predict in general terms the probable occurrence and type of land contamination likely to be present. In some instances whole districts have been dedicated historically to a given type of industrial activity, and in this situation it is probably simpler to deal with the entire land area rather than attempt to work on a site-by-site basis. The issue of responsibility for any remaining groundwater pollution risk will also arise. This may be difficult to resolve where the associated contamination could have occurred at any moment during a long time interval, perhaps before the existence of legislation to control discharges to the soil. Polluted Surfnce Watercottrses A relatively common situation is the presence of contaminated (permanent or intermittent) surface watercourses crossing an area under study for groundwater pollution hazard assessment. Such watercourses will often present a major contamination hazard to underlying groundwater, and generate a significant subsurface contaminant load. Two main factors will determine the potential for groundwater contamination: whether the surface watercourse exhibits a loosing (influent) or gaining (effluent) behavior with respect to the underlying aquifer; the main hazard arises in relation to the former condition, but it should be noted that groundwater pumping for water supply purposes can reverse the watercourse condition from effluent to influent the quality of water infiltrating through the bed of surface watercourses can be greatly improved as a result of the natural pollutant attenuation during this process; however, more mobile and persistent contaminants are unlikely to be removed and will form the most important components of the associated subsurface contaminant load. It is not easy to establish reliably the rate and quality of water infiltrating from surface watercourses without detailed investigation and sampling. But from a general knowledge of the types of contamination present and the hydrogeological setting, it should normally be feasible to establish the relative severity of the subsurface contaminant load. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies Figure 3.6 Legend for mapping of subsurface contaminant load CONTAMINANT-GENERATING CARTOGRAPHIC ACTIVITY REPRESENTATION reduced moderate elevated Diffuse Sources urban residential area agricultural land use Point Sources industrial activity effluent lagoon solid waste disposal polluted surface watercourse transportation routes Transportation Routes Accidents involving the transport of hazardous substances occur intermittently, and the handling and disposal of any such substances following these accidents is capable of causing a significant subsurface contaminant load and threatening groundwater quality in some aquifers. A similar situation occurs at major transportation terminals where these substances are regularly handled and sometimes accidentally discharged. It is necessary to locate the major terminals and important routes, and consider the probability of them generating a subsurface contaminant load. This is by no means straightforward, but there may be statistics available on the occurrence of accidents and the frequency of transport of substances posing major hazards to groundwater, together with the types of emergency procedure normally adopted. In general terms these locations must be treated as potential sources of a contaminant load of moderate intensity, unless it is clear that there are special provisions within routine operational procedures to reduce the incidence of spillages and to avoid groundwater contamination should they occur. Cemeteries The burial of human remains and (in some cases animal corpses) is a relatively I common practice in many cultures around the world. The question is thus sometimes asked as to whether cemeteries represent significant potential sources of groundwater contamination. Generally, this type of practice generates only a relatively small microbiological contaminant load over a restricted area, and this will be further reduced if special waterproofing of tombs andlor corrosion-resistant coffins are used. The same may not be true when large numbers of animal corpses have to be disposed of rapidly Part 6: Technical Guide Methodological Approaches to Groundwater Protection following a disease outbreak, since rapidly excavated pits might be used without special precaution or evaluation. The POSH method for the inventory of subsurface contaminant ioad permits an assessment of potential pollution sources into three levels: reduced, moderate, and elevated. The approach to classifying contaminant loads (and from them to groundwater pollution hazard assessment) presented here is very useful in relation to the prioritization of groundwater quality monitoring programs and of environmental inspection of field installations. The data on potential point sources of pollution can readily be represented on maps of the same scale as those used for mapping aquifer pollution vulnerability and delineating groundwater supply protection areas. This will allow ready consideration of the interaction of the data they contain and facilitate the assessment of aquifer or source contamination hazard (see Technical Guide Part B4), but it is important that each activity is also identified by a code and registered in a database. For disperse and multi-point sources, it is generally more practical to define the land areas occupied and thus generate a potential subsurface contaminant load map, using different shading to represent the relative load intensity. A convenient legend for all such maps is presented in Figure 3.6 (Foster and Hirata, 1988).It is possible that more detailed mapping scales will be required in densely populated urban situations with a wide range of industrial and other activity. In developing nations, land use by anthropogenic activities shows relatively rapid change, and this complicates the production of subsurface contaminant load maps. However, major advances in computing and improved facilities for color printing will increasingly make it possible for subsurface contaminant load maps to be regularly updated and printed. CIS systems are very useful in this respect, since they also allow the electronic correlation and rapid manipulation of spatial data, as well as the generation of colored images and analog maps of different attributes. Another great advantage of holding the relevant information in digital databases and maps is that they can be made available via a website and accessed by all land and water stakeholders. This introduction to the POSH method and classification is intended to provide general orientation for the user, but it is important that it is adapted to local realities and requirements of a given groundwater pollution hazard assessment project. MethodologicalApproaches to Groundwater Protection B4 Assessment and Control of Croundwater Pollution Hazards 1 Grogndwater pollution hazard can be depned as tbe prooaorlrty tnar an aquifer will experience negative impacts from a given anthropogenic activity to such a level that its groundwater would become umcceptable for human I consumption, according to the WHO guideline values for potable water quality. This chapter deals with ik assessment and control on a practical and prioritized basis. I (A) Recommended Approach The aquifer pollution hazard at any given location (Figure 4.1) can be determined by considering the interaction between: the subsurface contaminant load that is, will be, or might be applied on the subsoil as a result of human activities the vulnerability of the aquifer to pollution, which depends upon the natural characteristics of the strata that separate it from the land surface. In practical terms, hazard assessment thus involves consideration of this interaction (Foster, 1987)through superimposition of the outputs from the subsurface contaminant load inventory (as described in Chapter 3) on the aquifer pollution vulnerability map (as specified in Chapter 1). The most serious concern will arise where activities capable of generating an elevated contaminant load are present, or are projected, in an area of high or extreme aquifer vulnerability. Figure 4.1 Conceptual scheme for groundwater resource hazard assessment AQUIFER POLLUTIONVULNERABILITY hydraulic inaccessibility The assessment of aquifer pollution hazards is an essential prerequisite for groundwater resource protection, since it identifies those human activities that have the highest probability of negative impacts on the aquifer and thus indicates prioritization for the necessary control and mitigation measures. (B) Distinction between Hazard and Risk The use of the term "groundwater pollution hazard" in this publication has exactly the same meaning as the term "groundwater pollution risk" in Foster and Hirata (1988). The change in terminology is necessary to conform with that now used for other areas of risk assessment to human or animal health and ecosystems, where risk is now defined as the product of "hazard times scale of impact." The scope of the current Guide is t I restricted (in this terminology) to assessing groundwater pollution hazards and does ! not consider potential impacts on the human population or the aquatic ecosystems 1 dependent upon the aquifer, nor for that matter the economic value of aquifer resources. r~valuatlonof Groundwater Supply Pollution Hazard (A) Approach t o Incorporation of Supply Capture Zones The hazard concept can be extended beyond evaluation of aquifers as a whole to specific supply sources, through projection of groundwater capture zones (as delineated in Part B: Technical Guide Methodological Approaches to Groundwater Protection Chapter 2) onto aquifer pollution vulnerability maps (Figure 4.2) (Hirata and Rebou~as, 1999), prior to superimposing the outputs from the subsurface contaminant load inventory. If activities having potential to generate an elevated subsurface pollution load occur in an area of high aquifer vulnerability which is also within a groundwater supply capture zone, there will be a serious hazard of causing significant pollution of the water supply source. For complex or unstable groundwater flow regimes, the delineation of capture zones (protectionperimeters) can be fraught with problems and only limited application is feasible. In such situations aquifer pollution vulnerability mapping will have to assume the primary role in assessing groundwater pollution hazards to individual water supply sources while accepting the substantial uncertainty over the precise extension of their capture areas. (B)Complementary Wellhead Sanitary Surveys As a complement to the above methodology, it is strongly recommended that systematic wellhead sanitary surveys are also carried out. A standardized procedure for such surveys, leading to an assessment of microbiological pollution hazard for groundwater supplies, has been developed (Lloyd and Helmer, 1991). The survey is normally restricted to an area of 200-500 rn radius (Figure 2.2), and involves scoring a series of factors through direct visual inspection and using regular monitoring of fecal coliform counts in the groundwater supply for confirmation (Table 4.1). This approach can also be readily applied in the case of domestic supplies using tubewells or dug-wells equipped with hand-pumps or using gravity-fed springs, whose abstraction rates are very small and make the delineation of capture zones impracticable. t r a t m n i e r fnr rnntrnl n f C . r n ~ a n d r a ~ a t oDnllmmkinm r Aquifer pollution vulnerability should be conceived interactively with the contaminant load that is (will be, or might be) applied on the subsurface environment as a result of human activity, thereby causing a groundwater pollution hazard. Since contaminant load can be controlled, groundwater protection policy should focus on achieving such control as is necessary in relation to the aquifer vulnerability (or, in other words, to the natural pollution attenuation capacity of the overlying strata). (A) Preventing Future Pollution Where land-use planning is normally undertaken, for example in relation to the expansion of an urban area or to the relocation of an industrial area, aquifer pollution vulnerability maps are a valuable tool to reduce the risk of creating future groundwater pollution hazards. They identify the areas most vulnerable to groundwater pollution, such that the location of potentially hazardous activities can be avoided or prohibited. If the area concerned already has important groundwater supplies, source protection zones (perimeters)for these sources should be established as part of the planning process, with the Groundwater Quality Protection: a guide for water utilities, mrrnicipal ntlthorities, and environment crgencies Figure 4.2 Summary of overall approach to groundwater quality protection Part B:Technical Guide Methodological Approaches to Groundwater Protection I FACTORS IN SANITARY SURVEY SCORE (present = 1 absent = 0) Environmental Hazards (off-site) llocalcaves,sinkholes, orabandoned boreholesusedfordrainage lfissuresinstrataoverlayingwater-bearingformations d a t i v e n e a r b y sewers, pit latrines, cesspools, or septic tanks score of 5-6 .nearby agricultural wastes discharged or spilled indicates high (and 7-8 very Construction Hazards (on-site) high) potential l well-casing leakingornot penetratedorsealedtosufficientdepth pollution l well-casing notextendedabovegroundorfloor of pump room hazard leaks in system under vacuum a welIhead pump, suction pipes, or valve boxes vulnerable to flooding FC RAW WATER COUNTS CONFIRMED POLLUTIONRISK (mpn or cfu/IOOml) none low intermediate-to-high high very high Source: Modified from Lloyd and Helmer, 1991 aquifer pollution vulnerability map being used to guide the levels of control of potentially polluting activity required (Table 4.2). Such an approach ought to be applied flexibly with each case analyzed specifically on its merits, taking into account the likely future level of water demand on the aquifer and the cost of alternative sources of water supply. In the case of new potentially polluting activities of large scale and potential impact, the requirement for an Environmental Impact Assessment (EIA)as part of the authorization process is now an accepted technical andlor legal practice in many countries. Experience has shown that this mechanism ensures better consideration of environmental impacts (including those on groundwater quality) at the planning phase, facilitating a more effective approach to environmental protection. EIAs focus (Figure4.3)on the definition and analysis of problems, conflicts, and limitations related to project implementation, including the impact on neighboring activities, the local population, and the adjacent environment (UNEP, 1988), and in certain instances may lead to project relocation at a more acceptable location. The EIA is an integral part of the feasibility study for the I Table4.2 A c c w W m d r af rocrrmmnpotantlitUypaHutkrg activities md krstalktlorir acodng to J& WI~COfor zones poundwatwpratactim 3TENTIALLY POLLUTING ACTIVITY (A) BY AQUIFER VULNERABILITY I REQUIRING CONTROL MEASURES high - medium low 7 - - Septic Tank, Cesspits and Latrines: individual properties commuml properties, public gasoline station Solid Wasre Dirposal Facilities municipal domestic consmtion/inert industrial hazardous industrial (classI) industrial (classII and III) cemetery kinerator Mineral and Oil EKtraction construction material (inert) others, includingpetroleum and gas fuel lines hdustrial Premises , 1 type IJ and 111 Military Facilities JnfikrationLagoons municipallcoolingwater industrial effluent hnkawap Drainage truading roof A major road PN minor road PA amenity areas A parking lots PA industrial sites PN" airport/railway station PN Effluent Land Application food industry PA all other industries DN sewage effluent PA sewage sludge PA farmyard slurry A Intensive Livestock Rearing effluent lagoon I 1'A farmyard and feedlot drainage Agricultural Areas with pesticide with uncontrolled use of fertilizers pesticide storage t Part 8: Technical Guide Methodological Approaches to Groundwater Protection POTENTIALLY POLLUTINGACTlWTY (B) BY SOURCE PROTECTIONAREA REQUIRING CONTROL MEASURES 1 Septic Tanks, Cesspirs and L a t r ~ e s individual properties N N A A comm~rnalproperties, puMic N N PA A gasoline station N N PN PA Solid Waste Disposal Facilities municipal domestic N N N PN cmstrucuon/incrt N N PA PA industrial hazardous N N N N indusvial (class I) N N N PN industrial (class I1 and lii) N N N N cemtcry N N PN A incinerator N N N PN Mineral Extraction cansuuction material (inert) N N PN PA others, including petroleum and gas N N N N fuel lines N N N PN Industrial Premises tYPe I type I1 and it[ Military Facilities N N N 1 Infiltration Lagoons rnunicipal/cooling water indusrrial effluent Soakaway Drainage buifding roof major road minor road amenity areas parking lots industrial iites aitportlrailway station N = unacceptable in virtually all cases; PN = probably unacceptable, except in some cases subject to detailed investigation and special design; PA = probably acceptable subject to specific investigation and design; A = acceptable subject to standard design I = operational zone; I1 = microbiological zone; I11 = intermediate zone; 1V = entire capture area. Source: Modified from Foster and others, 1993;Hirata, 1993. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies Figure 4.3 Typical project implementationcycle with anticipated interventionof an Environmental Impact Assessment detailed evaluation (if significantImpacts] identificationof mitigation measuresand considerationof cost-benefitanalysis sound~ng,evaluation, and strategyfor controlmeasul 3L [-I DESIGN 1 AND MONITORING EVALUATION implementationof wntrol measum mJ project concerned and groundwater considerations must assume particular importance where certain types of industrial production, major landfills for solid waste disposal, mining enterprises, large-scale intensive irrigated agriculture, etc., are involved. There are various distinct approaches to undertaking an EIA (Weitzenfeld, 1990),but the need to identify the capacity of the surrounding land to attenuate potential contaminant loads and the identification of groundwater supplies that might be impacted are critical, because many activities (by design or by accident) lead to effluent discharge to the soil. Thus the aquifer pollution vulnerability map and delineation of water supply source flow-time and capture areas are both key inputs, and fit into the classical EIA evaluation scheme of (potential pollution) source-pathway-receptor (Figure 4.4). Trying to eliminate the possibility of effluent discharge can be very costly and sometimes unnecessary. Thus one of the best ways to obtain economic advantage and reduce environmental pollution hazard is to ensure that the proposed land use is fully compatible with its capacity to attenuate possible contaminants. (B)Dealing with Existing Pollution Sources The most frequent need will be to prioritize groundwater pollution control measures in areas where a range of potentially polluting activities are already in existence. Both in urban and rural settings it will first be necessary to establish which among these activities poses the more serious hazard to groundwater quality. The same three Part 8: Technical Guide Methodological Approaches to Groundwater Protection Figure 4.4 Conceptual EIA evaluation scheme of (potential pollution) source-pathway-receptor I components (aquifer vulnerability mapping, delineation of water supply protection areas, and inventory of subsurface contaminant load) form the fundamental basis for such an assessment (Figure 4.5). Table 4.3 should help in the selection of those activities that need significant artention, I I according to their location by aquifer vulnerability class and their position with respect Figure 4.5 Priority groundwater pollution control action-levelsbased I on aquifer vulnerability, source protection areas, and potential I: contaminant load 1 low I medium ( high -1 ACTION-LEVEL 1 = high 2 = intermediate 3 = low 'Numbersofzonestareasreducedtosimplifypresentation. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies I Table 4.3 Examples ofmethodsfsr controlof pdential sources of groundwatercontamination SOURCE OF POLLUTION POSSIBLE RESTRICTIONS ALTERNATIVES Fertilizers and Pesticides nutrient and pesticide management to none mecr crop needs; c ~ u aofl rate and rimlng of applicatiw; bans on use of sclc~tedpesticides; rtgnlaticm of disposal of umd c o n n i n w In Sku Sanitation (latrines, &ow septic tanks if water urc h g h mains sewerage cesspits, septic tanks) apply septic rank design wandards Underground Storage double lining install above ground Tenks/Pipelines Ieak detection Solid Waste Disposal domestic impermeabilization of both base and domestic and industrial surface leachate collection and remote diqosal rtcyclingltrcatrnent monitor impact I Effluent Lagoons I agricultural impermeabilization of base none municipal impermeabilization of base treatment plant industrial monitor impact remote disposd Cemeteries impermeabilization of tombs cremarorir superficial drainage Wastewater Injection Wds invcrtigarion and monitor treatment apply strict design standards remote disposal I Mine Drainage and Wastctip operational control treatment monitor impact (pH correction) Source: Modified from Foster and others, 1993; Zaporozec and Miller, 2000 to source protection zones. In many cases it should be possible to reduce or eliminate subsurface contaminant load with modified design. For example, in-situ sanitation might be replaced by mains sewerage, effluent evaporation/percolation lagoons could be replaced by closed effluent treatment processes, and even a traditional cemetery might be replaced by a crematorium. LI It must be recognized, however, that controls on polluting activities aimed at reducing future subsurface contaminant load will not eliminate contaminants that are already in the subsurface as a result of past practices. For example, the installation of mains I sewerage in an urban district will radically reduce the existing subsurface contaminant load from in-situ sanitation, but various tons of contaminants deposited in the subsoil Part B: Technical Guide Methodological Approaches to Groundwater Protection over previous decades may still be capable of liberating a significant contaminant load to an underlying aquifer. In some instances and at certain locations, it may be possible to accept a potentially polluting activity without any alteration to its existing design, subject to the implementation of an offensive campaign of groundwater quality monitoring. This would require the installation of a monitoring network (capable of detecting any incipient groundwater contamination and of giving "early warning" of the need to take remedial action) in the immediate proximity of the activity concerned (Section 4.4B). (C)Approach to Historic Land Contamination Significant tracts of urban land and more isolated rural sites that have experienced extended periods of occupancy by certain types of industrial, mining, or military activity often exhibit serious contamination, even where the corresponding activity was shut down years previously. This contaminated land can generate a serious pollution load to groundwater under certain circumstances. In such cases it is necessary to evaluate the risk in terms of probability of impacts on humans, animals, and plants, resulting from contact with and/or ingestion of the contaminated land andlor groundwater. This type of risk assessment, which is normally used to guide the decision on priorities for remedial or clean-up measures, is not dealt with in detail here and those requiring further detail are referred to ASTM (1995).Such risk assessments often use the following criteria (Busmaster and Lear, 1991): where there is 95 percent probability of health impacts on a 1-in-10,000 basis, then immediate remediation works are essential where the corresponding value is between 1-in-10,000 and 1-in-1,000,000, more detailed cost-benefit studies and uncertainty evaluation are recommended below the latter level no action is generally taken. (D)Selecting New Groundwater Supply Areas The selection of areas in which to site new municipal groundwater supply sources should involve the same procedure as recommended above for assessing the pollution hazard to existing groundwater supplies. In situations where such an assessment identifies anthropogenic activities capable of generating an elevated subsurface contaminant load and/or the aquifer pollution vulnerability is high or extreme over most of the designated groundwater supply capture area, this assessment should be followed by a technical and economic appraisal to establish whether: it will be possible to control adequately all relevant potential pollution sources it would be advisable to look for other sites for the new groundwater supply sources. Groundwater Quality Protection: n guide for water utilities,mzrnicipnl authorities, and environmentagencies f 1 - L idwater pollu n hazard assessment in the Cagapava area of Brazil ' U I ne zatton of GIS (Ceo; arcal l n f o , ~ ~ ~ tSystem) techniqmcs i o n city of Cagapava for data management is especially \ appropriate in the work of Paraiba do Sul river \ groundwater pollution hazard wessment and control. The: facilituteefficient data storage, up- dates, manipulation, and integration. Moreover, they allow the flexibk k'/ - Dresentation of results, for both environment sector professionaIs and stakeholders, itr a variety of interactrveand paper outputs. The town of Ca~apava(Sao I Paulo) in Brazil is highly clay lens iependent upon groundwater I lwpermeability alluvial aquifer ources. The alluvial aquifer bedrock .~nderexploitation consists of i and gravel deposits with ~nterbeddedclay horizons reaching in total a thickncs 200-250 m. Its groundwate~ is mainly unconfined, exceut Aquifer Pollution Vulnerability ocnlly where it become mi- extreme high moderate -31 significant financial losses as plow a result of a number of cases '3 of aquifer ccmtamination, 4riversand which manifested the need - I ' ' 9- streams for a syrtcrnatic approach to \ roads groundwater pollution assessment and a rational waterwells svategy for prioritizing n r; in pollution control measures. 'I The mapping of aquifel I pollution vulnerability by the GOD method was one of the Part B: Technical GuideMethodological Approaches to Groundtunter Protection I first its groundwatt xotection program. A CIS was used to put into a database the spatial variation of the GroundwaterSupply factors entering into the GOU Capture Zones I . methodology (Martin and others, 1998). 1 1 The next step was to delineate the protection perimeters and thus capture zones) of 1 t t h e principal municipal water w I supply boreholes corresponding to 10 and 50 years saturated zone travel time. This was - m e using a numerical 3- 4 D groundwater flow model \ r generating a CIS-compatible ! output to facilitate thei geographical superimposition I on t wlnerab map. 1) 4 t a 11 r A survey ana ~nventoryor \i potential pollution sources I Potential Industrial (mainly industrial premises en Contaminant Load -' ,gas stations) was then carried 0 reduced out. Application of the POSHt 1 moderate approach to assessment led l elevated o their ranking as elevated, - or reduced potential rivers and streams L a significan subsurface contaminant roads load. These results were also waterwells incorporated in the CIS to A industrialfacility highlight locations for priority II action or special vigilance in the interests of protecting the existing sources of potable water supply. I Croundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies [ Role and Au~roachto Groundwater Qualitv Monitorina m An additional and essential component of groundwater protection programs is aquifer water level and quality monitoring (Figure 4.2). This is needed to: understand the baseline natural quality of the groundwater system collect new data on the aquifer system to improve its conceptual and numerical modelling provide verification of groundwater pollution hazard assessments confirm the effectiveness of groundwater quality protection measures This monitoring need is distinct from that required for direct analytical surveillance of the quality of water (from waterwells and springs) destined for public supply. The representativity and reliability of aquifer groundwater quality monitoring is very much a function of the type and number of monitoring installations in place. The cost of borehole drilling as such often exercises a severe constraint on the number of monitoring installations (except in situations of a shallow water table) and exerts a strong pressure to make recourse to production wells for aquifer monitoring. (A) Limitations of Production Well Sampling Most production wells have their groundwater intake over a large depth range, so as to maximize their yield-drawdown performance. They thus tend to pump a "cocktail of groundwater" of widely different 4 origin, in terms of recharge area and date (inmany casesmixing groundwaterwith residence times ranging over decades, centuries, or even millenia) .+ hydrogeochemical evolution, in terms of modification through aquifer-water interaction and natural contaminant attenuation. This will inevitably exert a serious limitation on the extent to which such monitoring data can be interpreted and extrapolated in many types of aquifer system (Foster and Gomes, 1989). Moreover, production well sampling is usually undertaken via a wellhead tap during routine operation of a high-capacity pumping plant. Thus another factor complicating the interpretation of this type of groundwater quality data is possible physiochemical modification of groundwater samples (compared to the in-situ condition) due to such processes as: air entry from borehole pumps (or other sampling devices) causing oxidation, and precipitation-dissolved metal ions and other constituents sensitive to changes in Eh .( volatilization,causinglossofunstablecompoundssuchaspetroleumhydrocarbons and synthetic organic solvents Idepressurization,causinglossofdissolvedgasessuchasC02andmodifyingpH. Such limitations are, all too often, not taken into account when interpreting the data provided by routine water quality surveillance in production waterwells for 92 Part B: Technical Guide Methodological Approaches to Groundwater Protection groundwater resource management and protection purposes. Fuller technical details of these limitations, and approaches to reducing sampling bias, can be found in Foster and Gomes (1989). (B) Systematic Monitoring for Groundwater Pollution Control Purpose-drilled, intelligently sited, and carefully constructed monitoring boreholes (or piezometers) are the most accurate means of obtaining groundwater samples representative of in-situ conditions in an aquifer system. These comprise small-diameter boreholes (50 millimeters or even less) with short screen lengths (2-5 meters), completed with relatively inert materials (stainless steel, teflon, or pvc). Appropriate drilling and installation procedures (including a bentonite seal to prevent cross-contamination via the borehole annulus) are required, but these are usually available in most countries (Foster and Gomes, 1989). Three distinct strategies can be adopted in systematic monitoring for groundwater pollution protection (Figure 4.6): Offensive Monitoring of Potential Pollution Sources. The objective is to provide early detection of incipient aquifer contamination by known sources of potential pollution, with monitoring immediarely down hydraulic gradient, and analytical parameters chosen specifically, with respect to the pollution source. This approach is expensive and thus has to be highly selective, primarily targeting the more hazardous pollution sources located within groundwater supply capture zones in aquifers of high pollution vulnerability. Defensive Monitoring for Groundwater Supply Sources. The objective is to provide warning of pollution plumes threatening potable wellfields or individual waterwells and springs, through the installation of a monitoring network up hydraulic gradient, that is capable of detecting approaching polluted groundwater in time for further investigation and remedial action to be taken. A thorough understanding of the local groundwater flow system and contaminant transport pathways is required, (especially in relation to selection of the depths of monitoring borehole intakes), to avoid the possibility of by-pass of the defensive monitoring network. Evaluation Monitoring for Sites of Known Aquifer Contamination. A similar approach to that described under offensive monitoring should be adopted: most importantly to confirm the effectiveness of natural contaminant attenuation processes, where these are considered to be the most economic or only feasible way to manage aquifer pollution to confirm the effectiveness of remedial engineering measures taken to clean up or contain aquifer contamination, where these have been judged technically and economically feasible. (C) Selection of Analytical Parameters There is also pressing need to improve the selection of analytical parameters determined for groundwater samples. Routine monitoring of groundwater supply sources is widely limited to EC, pH, FC counts, and free CI (if used for supply disinfection). Although Groundwater Quality Protection: n guide for water utilities, mrtnicipal authorities, and enuironment agencies I Figure 4.6 Schematic summary of groundwater quality monitoring strategies a) offensive detection monitoring for aquifer protection J I source b .1 I b) defensive detection monitoring for water supply protection 7 I c) evaluation monitoring of existing aquifer pollution incidents - 0 h atural groundwater - f& > jiJ a *!., + m' 0 + Part 8: Technical Guide Methodologicnl Approaches to Groundwater Protection these parameters give an indication of water purity, they provide very little information in relation to the presence or absence of the more frequent types of groundwater contamination. For example, if the waterwell was located in the vicinity of an industrial estate (including metal processing activity) it is essential to include monitoring for chlorinated industrial solvents and the heavy metals themselves, since the above monitoring schedule is unlikely to suggest their presence. The selection of monitoring parameters must be undertaken in the light of the groundwater pollution hazard assessment (Table A.2 in the Overview.). The frequency of sampling in groundwater monitoring networks also has to be defined. Other than in aquifers of extreme or high pollution vulnerability, it will not normally be necessary to monitor aquifer groundwater quality more frequently than at three-month intervals. Aounting Groundwater Quality ProtectionPrograms (A)Institutional Requirements and Responsibilities In general terms, the water resource or environment regulator (or that agency, department, or office of national, regional, or local government charged with performing this function) is normally empowered to protect groundwater quality. In principle they are thus best placed to mount groundwater quality protection programs including: the establishment of land-surface zoning based on groundwater protection requirements the implementation of appropriate groundwater protection measures although in practice they often lack the institutional resources and political commitment to act comprehensively or effectively. It is critical that attention focuses down to the scale and level of detail necessary for the assessment and protection of specific water supply sources. To this end it is essential that water service companies become intimately involved. Moreover, given their responsibility to conform to codes of sound engineering practice, there would appear to be an obligation on water service companies themselves to take the lead in promoting or undertaking pollution hazard assessments for all their groundwater supply sources. The procedures presented for groundwater pollution hazard assessment are the logical precursor to a program of protection measures. As such they provide a sound basis for forceful representations to be made to the local water resource and/ or environment regulator for action on gro~~ndwaterprotection measures where needed. Even if no adequate pollution control legislation or agency exists, it will normally be possible to put pressure on the local government or municipal authority to take protective action under decree in the greater interest of the local population. Groundwater Quality Protection: a guide for water utilities, rnunrcipal authorities, and environment agencles (B)Addressing Key Uncertainties and Challenges Significant scientific uncertainties are likely to be present in many groundwater pollution hazard assessments, notably those related to: the subsurface attenuation capacity for certain synthetic organic contaminants I. the likelihood and scale of preferential vadose-zone flow in some geological strata the rates of water leakage and contaminant transport in some confining aquitards the groundwater flow regimes around waterwells in complex heterogenous aquifers, which can lead to large error bands in the definition of protection requirements. The complication that this presents needs to be recognized (Reichard and others, 1990)and approached in an explicit and systematic way. In many instances it will be necessary in this context to obtain clear evidence of actual or incipient aquifer contamination through groundwater monitoring before it is possible to justify the cost of the necessary pollution control measures. If the groundwater pollution hazard is confirmed it will then be necessary to appraise the risks that it presents and to define appropriate actions. In general, technical, and administrative terms, such actions could include: negotiation (and possible subsidy) of modifications to the design and operation of polluting activities, through the introduction of improved technology to reduce or eliminate subsurface contaminant load, with appropriate monitoring or remediation of existing groundwater contamination at the site transfer of the polluting activity to another (hydrogeologically less vulnerable) location, (in some cases with payment of compensation), with appropriate monitoring or remediation of existing groundwater contamination at the site relocation of groundwater supply sources to a new area of low pollution hazard, with the concomitant introduction of appropriate land-use development controls. It should also be borne in mind that for some aquifers, or parts of aquifer systems, it will not be realistic to implement pollution protection, since their natural characteristics are such that poor quality groundwater is widely present. It will often be appropriate to designate such areas for the preferential location of industries or activities that have high probability of generating a heavy subsurface contaminant load. But in such cases it is important to evaluate carefully whether: the local groundwater may sometimes be used for small-scale domestic supply effluent infiltration could cause changes in groundwater flow direction that might threaten areas of better quality groundwater the construction of new waterwells or wellfields in adjacent areas could change the groundwater flow direction so as to be threatened by the neighboring groundwater contamination. It also has to be recognized that shallow groundwater in urban areas is often likely to be significantly contaminated. Nevertheless, an integrated ancl coordinated approach including various of the following actions will often be beneficial in helping to protect Part B: Technical Guide MethodologicalApproaches to Groundwater Protection Box 4.2 m I ) Groundw r source pollution hazard evaluation and management around Managua, Nica~ T ystematic groundwater resource hazard evaluation, agricultural, and it is considereo that the frequent 1 rncluding aquifer vulnerability mapping and subsu ' of mobile pesticides (such as the carbamate insectir~ucs, "- I contaminant load survey with a clear polic- '- poses the major pollution threat, and control over all stakeholders, has been carried out to pl agricultural activity will be needed in the interests of municipal wellfields. municipal water supply. Groundwater is of the utmost importance domestic, I industrial, and agricultural water supply in the region C and is extracted from deep municipal and private rollution assessment mapping for Managua groundwater system boreholes in a major volcanic aquifer system located south of Lake Managua. There is little soil development Subsurface Contaminant Load on the most recent lava flows, and this area is classified reduced moderate elevated as highly vulnerable, despite the relatively deep water- industrial sites rn table (more than 25 m bgl). The main existing wellfield gas stations · abstracts some 195 MVd and is located in the urban landfill sites A fringe east of lManagua City, but a new wellfield of 70-Mlld at a more rural location some 10 km south of /Aqulter Pollution Vulnerability the city is under investigation and development. '7 = .$ Ihe capture zone of the ex g weIlfield i enec moderate by a range of industries incluuingtanneries, , u l r b a t high Managua Lake workshops, and textile manufacturers in the Zona Franca industrial area, as well as fuel and chemical storage at the international airport and a number o developing periurban towns with in-situ sanitation estimated (Scharp, 1994; Scharp and others, 1997, MARENA and municipal wellfield A - 1 1 - 7 7 KTH, 2000). There are also several small air strips in the area, which were historically used for storage, loading, and aerial spraying of agricultural land. In the past 30 years there was intensive cotton cultivation using many highly persistent pesticides, such as toxaphene and DDT. The predicted flow zone to the new wellfield is classified as having moderate vulnerability, but there are areas of high vulnerability due to the absence of soil cover, which has been removed through erosion. While there are a number of porential point sources of contamination from industry, gas stations, and waste disposal sites, only one industrial site with underground storage t a n h has been classified as having high potential contaminan* -+ .., load. The caDture area is more predominantlv n L Groundwater Quality Protection: a guide for water utilities, municipal authorities, and environment agencies potable groundwater supplies: prioritizing mains sewerage extension to areas of high aquifer pollution vulnerability, where aquifers are used at any scale for potable water supply improving the location and quality of wastewater discharge from mains sewerage systems, after consideration of the potential impacts on periurban and downstream municipal wellfields and other groundwater users restricting the density of new residential development served by conventional in- situ sanitation units constraining industrial effluent discharge to the ground through permits and charges, thereby stimulating effluent recycling, minimization, and treatment enforcing special handling requirements for persistent toxic chemicals and effluents at any industrial site located in areas of high aquifer pollution vulnerability directkg the location of landfill solid-waste disposal facilities to areas of low aquifer pollution vulnerability. There are also some further significant obstacles to the implementation of groundwater protection measures including: controlling diffuse agricultural practices, especially where this implies changes in crop or farm type as opposed to refining management of existing cropping practices and animal husbandry dealing technically and financially with the legacy of historic land and water contamination, especially in longer-standing industrialized areas lack of clarity over legal responsibility for serious (current and historic) groundwater pollution related to such questions as the timing of pollution incidents or episodes in relation to the introduction of legal codes, and whether the pollution occurred intentionally, knowingly, incidentally, or accidentally from the activity concerned resistance to land surface zoning for groundwater protection because of alleged reduction in land values (or property blight) resulting from implied lost opportunity or increased cost for industrial development or agricultural productivity. (C)Creating a Consensus for Action The control of groundwater pollution hazard requires taking technical action to achieve the reductions in subsurface contaminant load defined as priority from the preceding analysis. These actions have to be promoted within the social and economic framework of the area concerned, thus full stakeholder participation in the pollution hazard assessment and in the formulation of control measures will be essential for success. Every effort should be made to make groundwater pollution hazard assessments transparent and available to civil society in general. A systematic socioeconomic assessment of the potential barriers to implementing groundwater protection measures (KTH and IMARENA, 2000) will often provide key tactical information with which to frame and prioritize the action plan. Part B: Technical Guide Methodologica[ Appronches to Grotrnd~unterProtection The procedures for groundwater pollution hazard assessment presented in this text constitute an effective vehicle for initiating the involvement of relevant stakeholders (especially water-user interests, but also potential groundwater polluters). This is (in part) because they facilitate communication through synthesis and simplification of hydrogeological conditions, while in essence still remaining scientifically based. In more general terms, land surface zoning through maps combining aquifer pollution vulnerability classes and groundwater supply capture areas (protection perimeters) can be readily used for the elaboration of acceptability matrices for various types of potentially polluting activity. Both are extremely valuable for: raising stakeholder awareness of groundwater pollution hazards offering a credible and defensible groundwater input to land-use planning procedures promoting public understanding of groundwater protection needs. References (- in blue d 4 r m Adams, B. and S. S. D. Foster. 1992. "Land-surface zoning for activities and prospectives." Water Science and Technology groundwater protection." Journal of Institution of Water and 24(11): 271-281. Environmental Management 6: 312-320. BNA (U.S. Bureau of National Affairs). 1975. Water Pollution Albinet, M. and J. Margat. 1970. "Cartographie de la vulnerabilite Control. BNA Policy and Practice Series. Washington D.C. a la pollution des nappes d'eau souterraine." Bulletin BRGM 2nd Burmaster, D. and J. Learh. 1991. "It's time to make risk Series 3(4):13-22. Orleans, France. assessment a science." Ground Water Monitoring and Aller, L., T. Bennett, J. H. Lehr, R. J. Petty, and G. Hackett. 1987. Remediation 11(3):5-15. DRASTIC: A standardized system for evaluating groundwater Carter, A. D., R. C. Palmer, and R. A. Monkhouse. 1987. pollution potential using hydrogeologic settings. Environmental "Mapping the vulnerability of groundwater to pollution from Protection Agency Report 60012-87-035. Washington, D.C. agricultural practice particularly with respect to nitrate." TNO Andersen, L. J. and J. Gosk. 1987. "Applicability of vulnerability Committee for Hydrological Research: Proceedings and Information maps." TNO Committee for Hydrological Research: Proceedings 38: 382-390. Delft, The Netherlands. and Information 38: 321-332. Delft, The Netherlands. Cherernisinoff, P. 1992. A guide to underground storage tanks ASTM (American Society for Testing and Material). 1995. Standard evaluation, site assessment and remediation. New Jersey: Prentice- guide for risk-based corrective action applied to petroleum release Hall. site. ASTM Designation E1739-95. Washington, D.C. Ch~ltonP. k&A wan,&.&.;$ Qd ,1990. *A Barbash, J. E. and E. A. Resek. 1996. Pesticides in ground water: ~roundwak- rtsk W ~ a t esupply r distribution, trends, and governing factors. Chelsea, Michigan: Ann sources In Barbados." Amencan Warer Resources Assoc~ar~on Arbor Press. fnternat~onalCozilerenceon Trop~calHydrology and Caribbean Water Resources 279-289. San Juan de Puerto Rlco. Bates, L. E., C. Barber, J. Ross, and J. Verhoeven. 1993. "Vulnerability of groundwater to pollution: evaluation of the Daly, D., A. Dassargues, D. Drew, S. Dunne, N. Goldschneider, N. DRASTIC system in the Peel river catchment, northern New South Neale, I. C. Popescu, and F. Zwahlen. 2001. "Main concepts of the Wales, Australia." Proceedings of Australian National University European approach for karst groundwater resource assessment and Conference: Aquifers at Risk. Canberra, Australia. mapping." IAH Hydrogeology]oumal 10: 340-345. Blarasin. bl.. C. Eric. C. Frigerio, and 5. Bettera. 1993. Daly, D. and W. P.Warren. 1998. "Mapping groundwater "Detenninacirjn del riesgo de contamination del acuifero libre pot vulnerability: the Irish perspective." Geological Society Special siste~nasde silneamiento in situ: ciudad de Rio Cuarto." Dpro. Rio Ptiblication 130: 179-190. Cuarto, COrdoha, Argentina. Puhlicacion Espccirtl de 1~Asocirlcion DMAE (Departamento Municipal de Agua e Esgotos). 1981. Argentina de G'eolr>giaAplicadrt rl Icr Ingenieria 1: 114-1 3 1. "Equivalentes populacionais de residuos liquidos industriais da regiio Rlarasin, M., A. Cabrera, M. Villegas. C. Frigerio, and 5. Bettera. metropolitans de Porto Alegre." Informe 27. Porto Alegre, Brazil. 1999. Groundwarer contamination from septic tanks system in Doerfliger, N., and F. Zwahlen. 1998. Practical guide to groundwater two ~~ei~hborhoodsin Rio Cuarto City, Cordoba. Argentina. uulnerability mapping In karstic regions. Berne, Switzerland:A Swiss Intenradorrol Assucicrtion of Hydrogs01agists-lnteff~ti~n~/ Agency for Environment, Forest & Landscape Publication. Cnntrib~rtionsto Hvdrogeology 21: 31-38. A (Environment Agency). 1998. Policy and practice for tho Bernardes, Jr. C., R. Hirata, J. Mendes, and R. Cleary. 1991. protection of groundwater. London: HMSO. "Remedial accion for an industrial open dump-proposed Groundwater Quality Protection: a ptide for water utilities, mzcnicipal authorities, and enuzronment agencies Fetter, C. 1988. Applied hydrogeology. New York: Macmillan lnternntionnl Association of Hydrological Sciences 142: Publishing Company. 1073-1085. Foster, S. S. D. 1985. Groundwarer pollution protection in Hirata, R. 1993. "0s recursos hidricos subterrineos e as novas developing countries. IAH Intl. Contrl Hydrogeology. International exiggncias ambientais." Reursta do Inst~tutoGeologic0 de Sdo Association of Hydrogeologists-lnternat~onalContributions to Paolo 14(1):39-62. Hydrogeology 6: 167-200. Hirata, R.,C.Bastos, and^. k&. 1997. "Mapa de -1987."Fundamentalconceptsinaqu~fervulnerability vulnerabilidade das Bguassubterrbeas no Estado de Slo Paulo." pollution risk and protection strategy." Proceedings of International Instintto GeoI6gic0, Companhia de Saneamento Ambicntal, Conference: Vulnerability of Soil and Groundwater to Pollutants. Departamento de Aguas e Energia Elkica. S o Paulo, Brasil. 2 vol. Noord\vijk, The Netherlands. Hirata, R., C. Bastos, G. Rocha, D. Comes, and M. Iritani. 1991. Foster, S. S. D., B. Adams, M. Morales, and S. Tenjo. "Groundwarer pollution risk and vulnerability map of the SHo Paulo 1993. "Groundwater protection strategies: a guide towards State - Brasil."WaterScienre and Technology 24(21): 1.59-169. implementation." UK ODA, CPR, WHOIPAHO-HPE Technical Hirata, R. and A. Rebou~as.1999. "La proteccion de 10s recursos Manual. Lima, Peru. 88pp. hidricos subterraneos: una vision integrada, basada en perimetros Foster, S. S. D., P. J. Chilton, and M. E. Stuart. 1991. "Mechanisms de prorecci6n de pozos y vulnerabilidad de acuiferos." Boletin of groundwater pollution by pesticides." journal of Institution of Geologic0 Minero de Espaiia 110 (4):423-436. Water and Environmental Management 5: 186-193. Hirata, R.,G.S.Rodrigues, L. C.Paraiba, and C. C. Buschinelti. Foster, S. S. D., A. C. Cripps, and A. K. Smith Carington. 1982. 1995.Groudwuter contamination risk from agriczdtrrralactivity in Nitrate leaching to groundwater. Philosophrcal Transactions of G o Paulo State (Braril). BGS (British Geological Survey)Technical Royal Society of London 296: 477-489. Report WD/95/26: 93-101. Nottingham. U.K.. Foster, S. S. D. and D. C. Gomes. 1989. "Groundwarer quality Holden, L.R., J. A. Graham, R. W. Whitmore, W. J. Alexander, mon~toring:an appraisal of practices and costs." WHO-PAHOI R. W. Pratt, S. K. Liddle, and L. L. Piper. 1992. "Results of the HPE-CEPIS Technical Manual. Lima, Peru. national alachlor waterwell survey." Environmental Scienceand Technology 26: 935-943. Foster, S. S. D. and R. Hirata. 1988. "Groundwater pollution risk assessment: a methodology using available data." WHO-PAHOI Johansson, P.0, and R. Hirata. 2001. Rating of groundwater con- HPE-CEPISTechnical Manual. Lima, Peru. taminanon sources. In: Zaporozec, A., ed., Groundwater contamim- tion inventory: A methodological guideline. Paris, France: UNESCO. Foster, S. S. D., A. R. Lawrence, and B. L. Morris. 1998. "Groundwater in urban development: assessing management needs Kalinski, R. J., W. E. Kelly, I. Bogardi, R. L. Ehrman, and P. D. and formulating policy strategies." World Bank Technical Paper Yarnarnoto. 1994. "Correlation between DRASTIC vulnerabilities 390. Washington, D.C. and incidents of VOC contamination of municipal wells in Nebraska." Ground Water 32(1):31-34. Foster, S. S. D. and A. C. Skinner. 1995. "Groundwater protection: the science and practice of land surface zoning." international Kolpin, D. W., J. E Barbash., and R. J. Gilliom. 2000. "Pesticides Association of Hydrological Sciences 225: 471-482. in ground water of the United States, 1992-1996." Ground Water 38(6):858-865. Gillham, R. and J. Cherry. 1989. Refuse disposal sites and their long-term behavior. Dusseldorf, Germany: ENVITEC. Kostecki, P. T. and E. Calabrese. 1989. Petroleum contaminated soil: remediation techniques, environmental fate, and risk Hackman, E. E. 1978. "Toxic organic chemicals: destruction and assessment. Vol I. Mich.: Lewis Publishers. waste treatment." Pollution Technology Revrew 40. KTH and MARENA. 2000. "Identificacion de barreras a la Haertle, A. 1983. "Method of working and employment of EDP protecci6n sostenible del agua subterranea. Ministerio del Ambiente during the preparation of groundwater vulnerability maps." y Recursos Naturales Inforrne. Managua, Nicaragua. Groundwater Quality Protection: a guide for water utilities, municipal authorities, and envrronment agencies Lewis, W. J., S. S. D. Foster, and B. Drasar. 1982. The risk of Nicholson, R., J. Cherry, and E. Readon. 1983. "Migration groundwater pollution by on-site sanitation in developing countries. of contaminants in groundwater at a landfill: A case study, 6 . WHO-PAHOMPE-CEPIS Technical manual. Lima, Peru. Hydrogeochemistry."]ozrrnal of Hydrology 63(1/2): 131-176. Lloyd, B, and R. Helmet. 1991. Surveillance of drinking water NRA (National Rivers Authority). 1995. Guide to groundwater quality in rural areas. WHO-UNEP Publication. Longman Scientific protection zones in England and Wales. London: HMSO. and Techn~cal.London, U.K. NRC (National Research Council). 1993. Groundwater Loague, K. 1994. "Regional scale groundwater vulnerability vulnerability assessment: contamination potential cinder conditions estimates: impact of reducing data uncertainties for assessments in of uncertainty. Washington, D.C.: National Academy Press. Hawaii." Ground Water 32: 605-616. Luin, A. B. Van and W. Van Starkenburg. 1985. "Behaviour of contaminants in groundwater." Water Scienceand Technology 17: 843-853. Pankow, J., R. Johnson, J. Houck, S. Brillante, and W. Bryan. Mackey, D. and J. Cherry. 1996. "Groundwater contamination: 1984. "Migration of chlorophenolic compounds at the chemical pump and treat remediat~on."Environmental Scienceand waste disposal site at Alkali Lake, Oregon. 1,Site description and Technology 23: 630-636. groundwater flow." Ground Water 22(5): 593-601. Reichard, E., C. Cranor, R. Raucher, and G. Zapponi. 1990. "Groundwater contaminant risk assessment: A guide to understanding and managing uncertainties." Internationnl Association of Hydrological Sciences. Wallingford, U.K. Mazurek, J. 1979. "Summary of modified Le Grand method." Rosen, L. 1994. "A study of the DRASTIC methodology with National Center for Ground Water Research Report. Norman, emphasis on Swedish conditions." Ground Water 32(2):278-285. Oklahoma. Sax, N. 1984. Dangerous properties of industrial materials, sixth Miller, D. and M. Scalf. 1974. "New priorities for groundwater edition. New York: Van Nostrand Reinhold. quality protection." Ground Water 12: 335-347. Monkhouse, R. A. 1983. "Vulnerability of aquifers and groundwater quality in the United Kingdom." Institute of Geological Sciences Report. Nottingharn, U.K. Morris, B. L. and S. S. D. Foster. 2000. "Cryptosporidiurn contamination hazard assessment and risk management for British groundwater sources." Water Science and Technology 41 (7): 67-77. e.,'C'.'fe%urLit, Nemerow, N. L. 1963. Theories and practices of industrial waste ~okoi' K. P.Scbulz, and W. Weinzierl. 1993. treatment. Reading, Mass.: Addison-Wesley. "Mapping procedures for assessing groundwater vulnerability to nitrates and pesticides in Application of Geographic Information -1971.Liquidwasteofindustry:theoriesandpracticesof System in Hydrology and Water Resources Management." industrial waste treatment. Reading, Mass.: Addison-Wesley. Internatzonal Association of Hydrological Scrences211: 80-92. Groundwater Quality Protection: a guide for water utilities, mctnicipal acrthorities, and enuironment agencies Stuart, M. and C. Milne. 1997. "Groundwater quality implications -.1994.Handbookongroundwaterandwellheadprotection. of wastewater irrigation in Lcbn, Mexico." Proceedings of Env~ronmencalProtection Agency, Washington, D.C. International Association of Hydr~~eolog~sts Congress:Problems, Vrba, J. and E. Rornijn. (eds.) 1986. "Impact of agricultural Processes and Management. Nortingham 1: 193-190. activities on groundwater." lnternatzonal Association of Thomann, R., j.Lobos, H. Sallas, and j.Dos Santos. 1987. Hydrogeologists-I)?ternationalContribution to Hydrogeology 5. "Manual de evaluacidn y manejo de sustancias toxicas en aguas Vrba, J. and A. Zaporozec. 1994. "Guidebook on mapping superficiales (3).Evaluation preliminar del problems." WHO/ groundwater vulnerability." International Association of PAHO-CEPIS Technical Report. Lima, Peru. Hydrogeologists-international Contributions to Hydrogeology 16. UNEP (Un~tedNations Environmental Programme). 1988. Weitzenfeld, H. 1990. "Manual brisico de evaluaci6n del impact0 en "Environmental impact assessment: basic proceedings for developing el ambiente y la salud de proyectos de desarrollo." ECO-OPSIOMS. countries." Internal Guidebook. Geneva, Switzerland. Metepec, Mexico. 198 pp. EPA (Environmental Protection Agency). 1977. Procedures mantra1 WHO (World Health Organization). 1982. "Rapid assessment for groundwater monitoring of solid waste disposal facilities. of sources of air water and land pollution." World Health Report EPA-530-SW-61. Washington, D.C. Organization Offset Publication 62. 113pp. -.1980a.Treatabilitymanual(2)industrialdescription.Report Zaporozec, A. and j. Miller. 2000. Groundwater polltrtion. Paris, EPA-600-8-80-042B. Washington, D.C. France: UNESCO. ---. 1980b. Procedures manual for groundwater monitoring at Zaporozec, A. 2001. Contaminant source Inventory. In: Zaporozec, solid waste disposal facilities: Report SW-611. Washington, D.C. A. (ed.) Groundcuater contamination inventory A methodological -.1991. "Guide for conduction contaminant source inventories gzrideline. Paris, Prance: UNESCO. for public drinking water supplies: technical assistance document." Environmental Protection Agency, Office of Water, Washington, D.C. ,ww-,-w . . -- . aplqw&i,- , I , ,.I , ... -..r .-w,mu.~w.:~ .H6w e ~ 'ww~mw, ~M & & & , & ~ m: a - , ' .. -..., ' # . ? i ~ . ~ ~ ' 9 #!w:sMmwf t 1 0 1 ~ ~ 4Crs,qw(4i, , a *p.l*.irum*ausjwr.llCrr~ ~ ~ d ~ m ~ ~ ~ w m *. . - ~ ~ ~ mr ,: ~~ w. ~~ ~ww ~* ~. ~~ . ~' ~ ~ ~ J I m.-'CilL!g# . , f#q.am#m@q . - I , '