J. Environ. Monit., 1999, 1 55N Focus Doing what comes naturally With high-tech remediation technologies proving too expensive for all but the most seriously polluted sites, scientists and regulators are looking to the restorative powers of Nature to clean-up contamination. EVective implementation of so-called ‘natural attenuation’ techniques relies crucially on high quality environmental analysis.Pollution of groundwater is a worldwide problem and remediation entails high environmental and economic costs. In many countries groundwaters are an essential part of drinking water supplies and their protection has long been an integral part of environmental policy.1 Groundwater protection practices are complicated, however, drawing on aspects of geology, soil chemistry, hydrogeology and ecology.While a wide variety of remediation technologies have been developed, particularly in the United States, the costs of remediation for contaminated land and associated groundwater remain high. In the absence of a true understanding about the fate of contaminants in the environment, and indeed their true toxicity, some argue that remediation Box 1: How it works Natural attenuation makes use of natural processes to contain or treat pollution. Natural attenuation may reduce the potential risk posed by site contaminants in three ways:4 (1) Transformation of contaminants to a less toxic form through destructive processes such as biodegradation or abiotic transformations.(2) Reduction of contaminant concentrations through dilution and/or dispersion, whereby potential exposure levels are reduced.(3) Reduction of contaminant mobility or bioavailability through (ad)sorption onto the soil or rock matrix. Biodegradation processes are of special importance to natural attenuation because they can transform toxic contaminants into non-toxic by-products.5 In many subsurface environments, microorganisms (yeast, fungi or bacteria) breakdown pollutants (primarily organics) under conditions that can be either aerobic (with oxygen) or anaerobic (without oxygen).The mechanisms are complex and still not completely understood. Typically, aquifer redox conditions vary within the groundwater plume. Near the source, conditions are highly reducing: oxygen and nitrate are depleted and concentrations of dissolved iron and manganese are high. In this area the primary reaction mechanisms are methanogenesis and sulfate utilisation.Downgradient from the source, where contaminant concentrations are lower, groundwater is again enriched with nitrate and dissolved oxygen. The occurrence of specific non-oxygen electron acceptor reaction zones is dependent upon the pool of electron acceptors available in the aquifer, and the nature of the electron donor available to the microorganisms from the contaminant release.The position of each reaction zone and the points of transition from one dominant electron acceptor area to another is dependent on the dissolution rate of the contaminant from the source, the utilisation rate of contaminants under specific electron acceptor conditions, and the rate of groundwater migration below the site.goals for contaminated sites are unnecessarily stringent, and hence the costs of clean-up unnecessarily high.2 As a result, large numbers of contaminated sites lie undeveloped because estimated remedial costs exceed the land value. Over recent years attention has focused on the assimilative capacity of the environment as an acceptable remediation option.In natural attenuation, also known as intrinsic remediation, bioattenuation or intrinsic bioremediation, contaminants are contained and site remediation achieved through naturally occurring processes.3 Natural attenuation processes include a variety of physical, chemical or biological processes that act without human intervention to reduce the mass, toxicity, mobility, volume or concentration of contaminants. These insitu processes include biodegradation; dispersion; dilution; sorption; volatilisation; and chemical or biological stabilisation, or destruction of contaminants (see Box 1).Status of natural attenuation The attenuation and degradation of contaminants in soil and groundwater has been studied for many years.As early as 1978 in the UK, the Cooperative Programme of Research on the Behaviour of Hazardous Wastes in Landfills demonstrated that natural processes occurred in a wide range of geological conditions.2 More recently, research in Denmark and elsewhere has shown that concentrations of polynuclear aromatic hydrocarbons (PAHs), phenols and volatile organic compounds such as benzene to be significantly reduced, many to below analytical detection limits, in groundwater only 50 m away from contamination sources.2 While scientific research is being pursued worldwide, the majority of the practical experience comes from the United States.Under a recent policy directive, monitored natural attenuation (MNA) is recognised by the EPA as a legitimate method of remediation for soil and groundwater that can be evaluated and compared to other remediation processes.4 The EPA defines MNA as ‘...the use of natural attenuation processes within the context of a carefully controlled and monitored site clean-up approach56N J.Environ. Monit., 1999, 1 Focus that will reduce contaminant concentrations to levels that are protective of human health and environment within a reasonable time frame’.4 The Agency prefers those processes that degrade contaminants and expects MNA to be most appropriate to situations where plumes are stable.EPA recognition of natural attenuation extends to sites regulated under the Comprehensive Environmental Response, Compensation, and Liability Act (the CERCLA or ‘Superfund’); the Resource Conservation and Recovery Act (RCRA); and the underground storage tank (UST) regulations.Potential advantages of MNA over engineered remediation schemes include that it generates little or no remediation wastes and causes minimal disturbance to existing infrastructure.2,4 The techniques can be applied to all or part of a given site, and in conjunction with or as a follow up to active remedial measures.The potential for cross-media transfer of contaminants and the risk of human exposure to contaminated media are both reduced. And overall remediation costs should be lower than for active remediation. One of the principal limitations is the long timescales that may be required to achieve remediation objectives.2,4 Consequently, monitoring and analysis too may have to be sustained over a long period, and site characterisation can be more complex and costly than for active remediation.Another limitation is that the eVectiveness of MNA may be aVected by changes (natural or engineered) in geochemical or hydrogeological conditions. Thus, for example, changes in drainage or surface levels during the redevelopment of a site can result in renewed mobility of previously stabilised contaminants.In some cases the toxicity of transformation products may exceed that of the parent compound. Pollution targets Intrinsic bioremediation has been demonstrated for many common groundwater contaminants across a wide variety of field conditions.2 Fuel components such as benzene, toluene, ethylbenzene and xylenes (BTEX) have been shown in many circumstances to be remediated eVectively. Over recent years attention has focused particularly on the fuel additive methyl tertiarybutyl ether (MTBE), which has been found to migrate large distances in groundwater.The most recent research suggests that, unlike BTEX, MTBE is biodegraded only very slowly, making biodegradation less likely as a viable remediation option on its own.6 An alternative might be to supply oxygen enhancers to encourage the aerobic conditions (see Box 2).Evidence for the anaerobic biodegradation of aromatics, such as phenol, also looks promising but is insuYciently understood for all redox regimes. Current work focuses in particular on investigating the role of factors such as mineral oxidants, sulfate- and nitrate-reducing conditions, and methanogenesis in the degradation of the phenolic organics.6 Chlorinated solvents are one of the most prevalent groups of groundwater contaminants worldwide, and pose serious threats to water resources.These compounds are more dense than water and referred to as DNAPLs (dense nonaqueous phase liquids). The microbial degradation of chlorinated solvents is complex, and can proceed via reductive, oxidative and cometabolic degradation pathways.4,5,7 Under some redox conditions, chlorinated solvents can serve as electron acceptors, where the microorganisms use them as a sink for electrons.Under other redox conditions, these compounds can serve as electron donors in microbial metabolism, and in other conditions various cometabolic processes are employed.In general, the complete biodegradation of chlorinated solvents is facilitated by initial reducing conditions followed by oxidising conditions as contaminants are transported along the groundwater flowpaths. For sites containing chlorinated alkenes, such as perchloroethene (PCE) and trichloroethylene (TCE), careful analysis is needed since the daughter products of the initial reductive dechlorination, for example dichloroethylenes (DCEs) and vinyl chloride (VC), are themselves pollutants.These can be further reduced or directly oxidised. Under certain circumstances, the concentrations and/or toxicity of inorganic species may also be eVectively reduced by MNA.8 Sorption and redox reactions are the dominant mechanisms responsible for reducing the mobility, toxicity or bioavailability of inorganic contaminants.Metal speciation depends primarily on the ambient biogeochemical conditions of the soil and groundwater: pH, redox state (electron availability), alkalinity, and the presence of chelating (e.g. EDTA, natural organic acids) or solid-forming (e.g. phosphate) ligands are critically important. Box 2: Giving Nature a helping hand Although natural attenuation is, by definition, a passive approach, in theory it should be possible to assist natural processes through techniques that alter the redox conditions.US technology company Regenesis Inc is one of the few suppliers oVering commercial solutions to enhance natural attenuation. Its proprietary ORC product is an oxygen release compound that the company claims enhances aerobic bioremediation.11 ORC is a patented formulation of magnesium peroxide that time releases oxygen to improve the bioremediation regime for soil and groundwater.The oxygen-rich environment encourages the degradation action of naturally occurring aerobic microbes. Similarly, HRC is a hydrogen release compound that the company claims oVers a passive, low-cost solution for the in-situ treatment of chlorinated aliphatic hydrocarbons (CAHs) through anaerobic bioremediation.While these enhancement technologies are still under development, initial evaluations by the EPA’s Federal Remediation Technologies Roundtable and the New Jersey Department of Environmental Protection suggest they can be highly cost-eVective. For commercial product details see: www.regenesis.com and for FRTR evaluation report (ref: EPA Document 542-R-98–015) see: www.ftfr.orgJ. Environ.Monit., 1999, 1 57N Focus Putting MNA to work The increasing recognition of MNA by regulatory agencies does not constitute a change in clean-up goals. The burden of proof remains on the proponent, rather than the regulator, and the decision to implement MNA as a remediation option is site-specific. EPA guidance stresses five key principles as the basis for an implementation framework:4 (1) Demonstrable eYcacy: MNA is not a default or ‘do nothing’ option.Although the approach itself is essentially passive, MNA is an active choice and the decision to use it needs to be based on detailed, site-specific risk assessment. (2) Reasonable timeframe: MNA is only appropriate where it can be demonstrated to achieve remedial objectives within a reasonable timeframe.‘Reasonable’ is a site-specific decision but should not be excessive compared to other remedies. (3) Remediation of sources: Source control measures should be evaluated at all sites, and are especially important in the case of MNA since uncontrolled sources may risk overriding natural attenuation mechanisms. (4) Performance monitoring: Extensive monitoring is required to gauge the eVectiveness of the technique.In view of the longer timeframes for clean-up, this is especially important for MNA and should be maintained for as long as contamination levels remain above clean-up goals. (5) Contingency remedies: Stakeholders should agree an alternative clean-up technology or approach as a back-up in the event that MNA fails to perform as anticipated.This is especially important where MNA is selected primarily on the basis of predictive analysis. Monitoring strategies Performance monitoring is of even greater importance for MNA than for other remediation options because of the potentially long timeframes, the potential for contaminant migration, and other uncertainties. Three kinds of monitoring are employed within the MNA approach:9 (1) Site characterisation—to describe the disposition of the contamination and forecast its future behaviour.(2) Validation monitoring—to determine whether the predictions of site characterisation are accurate. (3) L ong-term monitoring—to ensure that the behaviour of contaminant plumes does not change.The frequency of all monitoring should be adequate to detect changes in site conditions, and in particular to determine the rates of attenuation for individual pollutants and how the rates are changing over time. Although redox chemistry is an important indicator it gives only a snapshot picture: it may not reflect the historical behaviour of the contaminants, nor will it necessarily predict their future behaviour.9 EVective site characterisation monitoring should consider multiple lines of evidence, specifically: the distribution of daughter products (a record of past conditions); redox conditions (situation as presently observed); and the hydrological framework (predictor of future conditions).The crucial step between site characterisation and subsequent validation and long-term monitoring is the conceptual model.A robust model requires determination of the nature and extent of the contamination in three dimensions. The site processes mobilising the contaminants and factors influencing the migration pathways should also be known, as should changes in contaminant location and concentration over time.Since many of these aspects are not even theoretically understood for many pollutants, the conceptual model is often the weak link in the MNA process. Modelling results are only as good as the input data. Site-specific data should be used to predict the fate and transport of solutes, given the controlling physical, chemical and biological processes. Several commercial solute fate and transport models are available, including some tailored to particular applications or pollutants.Utah State University, for example, has developed the Natural Attenuation Decision Support System (NADSS), application software for evaluating the application of MNA to underground storage tank sites.10 In general, however, better models are needed. Once a conceptual model has been accepted, a period of monitoring is required to verify that the forecast is adequate. The frequency of validation monitoring should be related to factors such as: the natural variability in contaminant concentrations; the distance and time between the source and the target location; and the reduction in contaminant concentrations required to meet the acceptance criteria.9 If validation monitoring confirms that natural attenuation will meet the acceptance criteria, a programme of long-term monitoring should be implemented.Here the sampling interval should be related to the contaminant’s expected time of travel along the flow path from one monitoring well to the next. The monitoring should continue until remediation objectives have been achieved, and longer if necessary to prove that the site no longer poses a threat to human health.4 The research agenda Progress on the use of natural attenuation is being held back by lack of knowledge of the fate and transport of large-scale pollution plumes in both saturated and unsaturated zones.1 A good understanding of the potential attenuation of particular pollutants is needed for diVering soil types and underlying geological strata in order to establish the impact of diVuse pollutants.Much current research focuses on gaining a better understanding of these processes for key pollutants within diVerent aquifers. The role of mineral oxidants (such as iron and manganese), fermentation (transformation without any external electron acceptors), and nitrate- and sulfate-reducing environments are all important areas of investigation.6 Work on the fate of leachate leakages and landfill gas migration will be crucial for the better regulation and remediation of landfill sites.1 The degradation rate is a critical component for conceptual models and is used to determine the risk of migration to potential receptors or across the site boundary.4,7,9 Techniques such as isotopic fractionation are being investigated as determinants of degradation rates.6 These should lead to simple tools that allow early judgements regarding the applicability and risks of MNA for a given site.In the US, MNA research and development continues to be supported through EPA’s OYce of Solid Waste and Emergency Response, with58N J.Environ. Monit., 1999, 1 demonstrations at several military and/or Superfund sites. In the EU, natural attenuation is supported under the key action on Sustainable Management and Quality of Water within the new Fifth Framework Programme for R&D. In the UK, the Environment Agency is providing funding of around £1 m over three years for investigation of MNA within its R&D programme, led by the Agency’s new National Centre for Contaminated Land and Groundwater.This includes support for the Network on Natural Attenuation in Groundwater (NNAGS) initiative and for establishing links with researchers and agencies overseas. MNA in perspective Much progress has been made in recent years in understanding the underlying processes of natural attenuation and in assessing its viability as a remediation option.As a result, MNA is being seen as increasingly attractive by consultants, responsible parties and regulators. Policy-making needs to reflect this improved technical understanding, for example by ensuring that public (or for that matter private) money is not spent on expensive remediation schemes that are unnecessary or unlikely to succeed.5 In some cases engineered schemes may actually inhibit natural attenuation mechanisms.Groundwater pump and treat systems and air sparging systems, for example, can increase oxygen concentrations in a contaminant plume. This introduction of dissolved oxygen into an anaerobic system may inhibit and Underground Storage T ank Sites, Directive 9200.4–17, Environmental Protection Agency, OYce of Solid Waste and Emergency Response, 1997.Available at www.epa.gov/oswer 5 Natural attenuation can be option for chlorinated solvents, Todd H. Wiedemeier, Michael J. Pound, Soil and Groundwater Cleanup Magazine. See on-line at www.sgcleanup.com 6 Unpublished research by various authors presented at the Network on Natural Attenuation in Groundwater (NNAGS) Conference, 21–22 June 1999, SheYeld, UK.Details from the Groundwater Protection and Restoration Group, University of SheYeld. See www.shef.ac.uk/~gprg/ 7 Principles and Practices of Natural Attenuation of Chlorinated Solvents, Remediation Technologies Development Forum, 1997. Details at www.rtdf.org. 8 Natural Attenuation of Metals and Radionuclides—An overview of the Sandia/DOE approach, R.D. Waters, P. V. Brady. D. J. Borns, Sandia National Laboratories, paper presented at Waste Management 98. Available at www.sandia.gov/ eesector/gs/gc/snap.html 9 Seminar Series on Monitored Natural Attenuation for Groundwater, Environmental Protection Agency, OYce of Solid Waste and Emergency Response, 1997. Available at www.epa.gov/oswer 10 For description of NADSS see project website at http://kemb.uwrl.usu.edu/ api/intro.html 11 For descriptions of Regenesis’s products see company website: www.regenesis.com 12 Setting Rational L imits on Natural Attenuation, John R.Odermatt, Soil and Groundwater Cleanup Magazine. See on-line at www.sgcleanup.com Mike Sharpe reductive dechlorination and remobilise a formerly stable plume. The acceptability of MNA as a remediation option will necessarily depend on the regulator’s perception of cost-eVectiveness.12 Project proponents will have to demonstrate that MNA will be a reasonably eVective means of protecting human health and the environment, taking accounting of the relative risks for this and alternative technologies for the application concerned. The level of confidence required in these risk assessments will be much greater at high risk sites than at low risk ones: without such data, MNA should necessarily be screened out as a remediation option. Nevertheless, experience in the US suggests that for many sites MNA oVers a viable component within a broader remediation strategy. Notes 1 Natural Attenuation of Petroleum Hydrocarbons and Chlorinated Solvents in Groundwater, R&D Technical Report P305, Environment Agency, Bristol, UK. Available from WRC (Tel: +44 (0)1793 865138). 2 Intrinsic bioremediation: an economic option for cleaning up contaminated land, Barry Ellis, Kyle Gorder, in Chemistry and Industry, March 1997, Society for Chemical Industry. 3 A Citizen’s Guide to Natural Attenuation, EPA 542-F-96–015, Environmental Protection Agency, OYce of Solid Waste and Emergency Response, 1996. Available from the hazardous waste information service: http://clu-in.org 4 Use of Monitored Natural Attenuation at Superfund, RCRA Corrective Action, Focus