Mobility control: How injected surfactants and biostimulants may be forced into low‐permeability units

Recovering dense nonaqueous‐phase liquid (DNAPL) remains one of the most difficult problems facing the remediation industry. Still, the most common method of recovering DNAPL is to physically remove the contaminants using common technologies such as total fluids recovery pumps, vacuum systems, and “pump‐and‐treat.” Increased DNAPL removal can be attained using surfactants to mobilize and/or solubilize the pollutants. However, very little is understood of the methods developed by petroleum engineers beginning in the 1960s to overcome by‐passed, low‐permeability zones in heterogeneous oil reservoirs. By injecting or causing the formation of viscous fluids in the subsurface, petroleum engineers caused increased in‐situ pressures that forced fluid flow into low permeability units as well as the higher permeability thief zones. Polymer flooding involves injecting a viscous aqueous polymer solution into the contaminated aquifer. Foam flooding involves injecting surfactant to decontaminate the high‐permeability zones and then periodic pulses of air to cause a temporary viscous foam to form in the high‐permeable zones after all DNAPL is removed. Later surfactant pulses are directed by the foam into unswept low‐permeable units. These methods have been applied to DNAPL removal using surfactants but they can also be applied to the injection of bio‐amendments into low‐permeability zones still requiring continued remediation. Here we discuss the principles of mobility control as practiced in an alluvial aquifer contaminated with chlorinated solvent and coal tar DNAPLs as well as some field results. © 2003 Wiley Periodicals, Inc.     

An assessment of hydraulic containment to control vertical DNAPL migration

Because of increased awareness and knowledge about their behavior, dense nonaqueous phase liquids (DNAPLs) are being detected at an increasing frequency at hazardous and solid waste land disposal units. Remedial systems at sites containing DNAPLs need to be designed to address the specific problems presented by DNAPLs. Because of their physical properties, DNAPLs migrate downward and are difficult to remove using conventional recovery methods. Groundwater pumping schemes can be designed to hydraulically contain this vertical migration. The purpose of this article is to present an approach for evaluating the potential to halt the vertical migration of DNAPLs using hydraulic control. Detailed groundwater flow modeling of a group of waste basins indicates that groundwater pumping in low-permeability sands can impose the upward hydraulic gradients required to stop downward DNAPL movement. However, a recovery well system located around the perimeter of the waste basins will not impose the required gradients over a sufficiently large area to effectively contain DNAPLs under the basins because the distribution of vertical gradients that could stop DNAPLs extends only about fifty to sixty feet from the wells. Additional modeling indicates that horizontal recovery wells located directly under the basins can contain the vertical migration of DNAPLs.     

Nanotechnology for site remediation

As a remediation tool, nanotechnology holds promise for cleaning up hazardous waste sites cost‐effectively and addressing challenging site conditions, such as the presence of dense nonaqueous phase liquids (DNAPLs). Some nanoparticles, such as nanoscale zero‐valent iron (nZVI) are already in use in full‐scale projects with encouraging success. Ongoing research at the bench and pilot scale is investigating particles such as self‐assembled monolayers on mesoporous supports (SAMMS™), dendrimers, carbon nanotubes, and metalloporphyrinogens to determine how to apply their unique chemical and physical properties for full‐scale remediation. There are many unanswered questions regarding nanotechnology. Further research is needed to understand the fate and transport of free nanoparticles in the environment, whether they are persistent, and whether they have toxicological effects on biological systems. In October 2008, the U.S. Environmental Protection Agency's Office of Superfund Remediation and Technology Innovation (OSRTI) prepared a fact sheet entitled “Nanotechnology for Site Remediation,” and an accompanying list of contaminated sites where nanotechnology has been tested. The fact sheet contains information that may assist site project managers in understanding the potential applications of this group of technologies. This article provides a synopsis of the US EPA fact sheet, available at http://clu‐in.org/542F08009 , and includes background information on nanotechnology; its use in site remediation; issues related to fate, transport, and toxicity; and a discussion of performance and cost data for field tests. The site list is available at http://clu‐in.org/products/nanozvi . © 2008 Wiley Periodicals, Inc.     

Recent Developments in Cleanup Technologies

Although the reaction mechanics are somewhat mysterious, the use of iron for in situ groundwater treatment has recently gained considerable attention and respect in the remediation industry. The basic scientific principles of both applications of iron have been known for over a century; however, both were nearly unheard of as remediation technologies five years ago. Both technologies have a strong potential for widespread use. They are commercially available, have been proven in field studies, are less expensive than traditional pump and treat technologies, and, in many types of groundwater systems, may be able to meet difficult-to-achieve groundwater treatment standards. As these technologies continue to undergo development, there could be considerably more aggressive applications used to treat ground-water containing high concentrations of chlorinated organics and DNAPLs.     

Biogeochemical gradients as a framework for understanding waste‐site evolution

The migration of biogeochemical gradients is a useful framework for understanding the evolution of biogeochemical conditions in groundwater at waste sites contaminated with metals and radionuclides. This understanding is critical to selecting sustainable remedies and evaluating sites for monitored natural attenuation, because most attenuation mechanisms are sensitive to geochemical conditions, such as pH and redox potential. Knowledge of how gradients in these parameters evolve provides insights into the behavior of contaminants with time and guides characterization, remedy selection, and monitoring efforts. An example is a seepage basin site at the Savannah River Site in South Carolina where low‐level acidic waste has seeped into groundwater. The remediation of this site relies, in part, on restoring the natural pH of the aquifer by injecting alkaline solutions. The remediation will continue until the pH upflow of the treatment zone increases to an acceptable value. The time required to achieve this objective depends on the time it takes the trailing pH gradient, the gradient separating the plume from influxing natural groundwater, to reach the treatment zone. Predictions of this length of time will strongly influence long‐term remedial decisions. © 2008 Wiley Periodicals, Inc.     

Coupling Flow Analysis With Geochemical and Microbiological Analyses to Assess the Bioremediation Feasibility of a Contaminated Aquifer

An integrated approach combining classic and molecular microbiological methods, “in vitro” bioremediation assays and groundwater numerical modeling, has been established to identify optimized solutions for remediating aquifers contaminated with organic pollutants. Bacteria have been isolated from an aquifer contaminated with toluene and methyl tert‐butyl ether (MTBE), selected for their growth with contaminants as a sole carbon source and identified through 16S rDNA partial sequencing. Successive biodegradation laboratory tests have been performed to determine which chemical conditions were more appropriate for the isolated bacteria to more efficiently oxidize toluene and MTBE. A groundwater model was created using FEFLOW code first to determine the movement of the plume front and second to simulate the impact of the biodegradation processes along the groundwater flow directions based on the bioremediation rates obtained in the laboratory. The results show that this innovative and interdisciplinary model can be used to assist in developing monitoring and remediation plans for cleaning up complex contaminated groundwater sites. This approach successfully combines the identification of the optimum biogeochemical conditions for bacterial biodegradation to occur with the predictability of the development of the process over time, ensuring decisive support in the management of contaminated sites. ©2016 Wiley Periodicals, Inc.     

Partitioning tracer tests as a remediation metric: Case study at naval amphibious base little creek,Virginia Beach,Virginia

The partitioning tracer test (PTT) is a characterization tool that can be used to quantify the porespace saturation (S_N) and spatial distribution of dense nonaqueous phase liquids (DNAPLs) in the subsurface. Because the method essentially eliminates data interpolation errors by directly measuring a relatively large subsurface volume, it offers significant promise as a remediation metric for DNAPL‐zone remediation efforts. This article presents, in detail, the design and results of field PTTs conducted before and after a DNAPL‐zone treatment at the Naval Amphibious Base Little Creek, Virginia Beach, Virginia. The results from different tracers yield a relatively large range in S_N estimates, indicating notable uncertainty and presenting significant challenges for meaningful interpretation. Several potential interpretation methods are presented, resulting in an estimated DNAPL removal range of 15 to 109 L. While this range is large, it is consistent with the DNAPL removal (～30 L) determined from analysis of effluent concentration measurements collected during the remediation efforts. At this site, the initial and final S_N values are low, and the relatively inconsistent performance of the various tracers indicates that these levels are near the lower practical quantification limit for these PTTs; however, the effective lower quantification limit for these tests is unknown. Generally, an understanding of lower quantification limits is particularly important for interpretation of post‐remediation PTTs because S_N values are likely to be low (due to remediation efforts) and the S_N estimated from the PTT may be used to predict long‐term dissolved plume behavior and assess associated environmental risk. Partitioning tracer test quantification limits are test‐specific, as they are dependent on a variety of factors including analytical uncertainty, tracer breakthrough characteristics, and tracer data integration techniques. The results of this case study indicate that methods for estimating lower quantification limits for field PTTs require further development. © 2004 Wiley Periodicals, Inc.     

Climate-influenced hydrobiogeochemistry and groundwater remedy design: A review

The process of designing a remedy for contaminated groundwater historically has not commonly included climate-future, hydrologic, and biogeochemical aquifer characteristics. From experience, the remedy design process also has not consistently nor directly integrated or projected future hydrologic and biogeochemical effects of the human-induced or developed environment—aka the anthropogenic influence—on potential remedy performance. The apparent practice of (1) not regularly assessing anthro-influenced hydrological (termed here as anthrohydrology) or biogeochemical characteristics (collectively hydrobiogeochemistry) of a site and (2) rarely accounting for future climatic shifts as design factors in remedy design may be due, in part, to the general practice-level view that groundwater remediation systems (whether in situ or ex situ) have seldom been anticipated to last more than a few years (or one or two decades at the most). Second, methods to reliably and quantitatively estimate site-specific, climate-future shifts in groundwater conditions using global and/or regional climate models and the resultant impacts on contaminant plume characteristics have not been readily available. The authors here suggest that while the concept of remedy design resilience and durability, within an envelope of climate change and anthropogenic influence, has been discussed in some technical circles as a component of “sustainable remediation,” we have found that direct application of these technical concepts in quantifiable terms remains rare. By incorporating the potential influence of future hydrobiogeochemical scenarios into remedy design, however, the design process could account for reasonable climate-induced influence on the groundwater system for a given site. These scenarios could then be applied within the remedy selection process to assess performance durability under potentially changing hydrologic, biological, and chemical conditions.     

A method to compare groundwater cleanup technologies

DuPont has developed a method to compare, on a consistent economic basis, in situ remediation technologies. The methodology employs a template site with a perchloroethylene plume 1000 ft long by 400 ft wide, and incorporates various aquifer thicknesses and depths. Variables considered in the methodology include duration of the remediation; estimated engineering and flow/transport modeling costs; equipment costs; and operation, maintenance, and monitoring costs. In this article, substrate-enhanced anaerobic bioremediation, intrinsic bioremediation, in situ permeable reactive barriers, and pump-and-treat systems are evalutated. Cost metrics include present cost, cost per pound of contaminant removed, and cost per 1000 gals treated, using a discounted cash-flow analysis. Costs of the remedial alternatives increase starting from intrinsic bioremediation, to substrate-enhanced anaerobic bioremediation, to a biological substrate-enhanced anaerobic barrier, to in situ permeable reactive barriers, to pump-and-treat systems with air stripping and carbon adsorption.     

A Preliminary Risk Assessment Protocol for Renegade Nanoparticles Deployed During Nanoremediation

The NanoRem European research project aims to support and develop the appropriate use of nanotechnology for contaminated land remediation by facilitating practical, economic, and exploitable nanotechnology for in situ remediation. This can only be achieved in parallel with a comprehensive understanding of the environmental risk‐benefit balance for the use of the nanoparticles (NPs) being investigated. While the NanoRem NPs could have a significant toxicity this is likely to be less potent than NPs currently being released into the environment, such as those from a variety of antibacterial products. The NanoRem NPs are likely to interact with the aquifer matrix, each other, and groundwater chemistry to rapidly cease to be mobile and are unlikely to penetrate into the aquifer more than a few meters from the point of injection. In terms of the source‐pathway‐receptor paradigm, the NanoRem NPs are cautiously presumed to represent a hazard (i.e., source). At least one receptor, in the form of not yet polluted groundwater, is present at all the NanoRem case study sites. While there are considerable uncertainties particularly with regards to the transport of NanoRem NPs, the ability of NPs to penetrate far into the formation is likely to be very limited. The relatively short travel distances reported in the literature for a variety of NP types and geological conditions suggest that the pathways are at best very limited in extent. Overall, this means that in many cases the level of risk renegade NPs can pose to the environment or human health is at most minimal. A qualitative protocol developed for the NanoRem field trials can demonstrate that injecting NanoRem NPs into contaminated groundwater poses a minimal level of risk due to the reduced pathway. ©2016 Wiley Periodicals, Inc.     

Insights from years of performance that are shaping injection‐based remediation systems

Research and field experience from the past 15 years has allowed remediation professionals to purposefully design injection‐based remediation systems with a high potential for success. Industry professionals can now claim a number of achievements that were unthinkable just a few years ago: (1) we have demonstrated that maximum contaminant levels (MCLs) can be achieved for multiple contaminants; (2) we have successfully targeted dense nonaqueous‐phase liquid (DNAPL) source zones; (3) we have expanded our understanding of injection hydraulics to treat large plumes; and (4) we have collected sufficient data on rates of treatment to be more predictive regarding outcomes. The next decade will continue to evolve the design and execution of these types of systems for application to more complex problems. At this point on the timeline, questions regarding the mechanisms of treatment have largely been addressed, allowing a shift in focus to operational enhancements. Specific operational insights arising from the body of work to date that arguably will continue to shape and influence the design and execution of injection‐based remediation systems include: (1) the fact that delivery does not always equal distribution, (2) treatment optimization requires aquifer tuning, and (3) life‐cycle costs can be reduced with remedy‐optimized investigation. The number of examples that support these concepts and their ramifications to future technology refinement is already increasing, demonstrating how the refinements that can be made around these areas of focus will enhance our ability to effectively tackle larger and more complicated plumes, and do so with maximum efficiency. © 2011 Wiley Periodicals, Inc.     

Comparing PFAS to other groundwater contaminants: Implications for remediation

Established groundwater contaminants such as chlorinated solvents and hydrocarbons have impacted groundwater at hundreds of thousands of sites around the United States and have been responsible for multibillion dollar remediation expenditures. An important question is whether groundwater remediation for the emerging contaminant class comprised of per‐ and polyfluoroalkyl substances (PFAS) will be a smaller, similar, or a larger‐scale problem than the established groundwater contaminants. A two‐pronged approach was used to evaluate this question in this paper. First, nine quantitative scale‐of‐remediation metrics were used to compare PFAS to four established contaminants: chlorinated solvents, benzene, 1,4‐dioxane, and methyl tert‐butyl ether. These metrics reflected the prevalence of the contaminants in the U.S., attenuation potential, remediation difficulty, and research intensity. Second, several key challenges identified with PFAS remediation were evaluated to see similar situations (qualitative analogs) that have been addressed by the remediation field in the past. The results of the analysis show that four out of nine of the evaluated quantitative metrics (production, number of potential sites, detection frequency, required destruction/removal efficiency) indicate that the scale of PFAS groundwater remediation may be smaller compared to the current scale of remediation for conventional groundwater contaminants. One attenuation metric, median plume length, suggests that overall PFAS remediation could pose a greater challenge compared to hydrocarbon sites, but only slightly larger than chlorinated volatile organic compounds sites. The second attenuation metric, hydrophobic sorption, was not definitive regarding the potential scale of PFAS remediation. The final three metrics (regulatory criteria, in‐situ remediation capability, and research intensity) all indicate that PFAS remediation might end up being a larger scale problem than the established contaminants. An assessment of the evolution of groundwater remediation capabilities for established contaminants identified five qualitative analogs for key PFAS groundwater remediation issues: (a) low‐level detection analytical capabilities; (b) methods to assess the risk of complex chemical mixtures; (c) nonaqueous phase dissolution as an analog for partitioning, precursors, and back diffusion at PFAS sites; (d) predictions of long plume lengths for emerging contaminants; and (e) monitored natural attenuation protocols for other non‐degrading groundwater contaminants. Overall the evaluation of these five analogs provided some comfort that, while remediating the potential universe of PFAS sites will be extremely challenging, the groundwater community has relevant past experience that may prove useful. The quantitative metrics and the qualitative analogs suggest a different combination of remediation approaches may be needed to deal with PFAS sites and may include source control, natural attenuation, in‐situ sequestration, containment, and point‐of‐use treatment. However, as with many chlorinated solvent sites, while complete restoration of PFAS sites may be uncommon, it should be possible to prevent excessive exposure of PFAS to human and ecological receptors.     

Geochemical evaluation of metals in groundwater at long‐term monitoring sites and active remediation sites

At many sites, long‐term monitoring (LTM) programs include metals as chemicals of concern, although they may not be site‐related contaminants and their detected concentrations may be natural. At other sites, active remediation of organic contaminants in groundwater results in changes to local geochemical conditions that affect metal concentrations. Metals should be carefully considered at both types of sites, even if they are not primary contaminants of concern. Geochemical evaluation can be performed at LTM sites to determine if the monitored metals reflect naturally high background and, hence, can be removed from the analytical program. Geochemical evaluation can also be performed pre‐ and post‐treatment at active remediation sites to document the effects of organics remediation on metals and identify the processes controlling metal concentrations. Examples from both types of sites are presented in this article. © 2008 Wiley Periodicals, Inc.     

In-situ bioremediation of hanford groundwater

The U.S. Department of Energy has generated liquid wastes containing radioactive and hazardous chemicals throughout the more than forty years of operation at its Hanford site in Washington State. Many of the waste components, including nitrate and carbon tetrachloride (CCl₄), have been detected in the Hanford groundwater. In-situ bioremediation of CCl₄ and nitrate is being considered to clean the aquifer. Preliminary estimates indicate that this technology should cost significantly less than ex-situ bioremediation and about the same as air stripping/granular activated carbon. In-situ bioremediation has the advantage of providing ultimate destruction of the contaminant and requires significantly less remediation time. Currently, a test site is under development. A computer-aided design tool is being used to design optimal remediation conditions by linking subsurface transport predictions, site characterization data, and microbial growth and contaminant destruction kinetics.     

PFAS Experts Symposium: Statements on regulatory policy,chemistry and analytics,toxicology, transport/fate,and remediation for per‐ and polyfluoroalkyl substances (PFAS) contamination issues

Sixty leading members of the scientific, engineering, regulatory, and legal communities assembled for the PFAS Experts Symposium in Arlington, Virginia on May 20 and 21, 2019 to discuss issues related to per‐ and polyfluoroalkyl substances (PFAS) based on the quickly evolving developments of PFAS regulations, chemistry and analytics, transport and fate concepts, toxicology, and remediation technologies.  The Symposium created a venue for experts with various specialized skills to provide opinions and trade perspectives on existing and new approaches to PFAS assessment and remediation in light of lessons learned managing other contaminants encountered over the past four decades. The following summarizes several consensus points developed as an outcome of the Symposium:

• Regulatory and policy issues: The response by many states and the US Environmental Protection Agency (USEPA) to media exposure and public pressure related to PFAS contamination is to relatively quickly initiate programs to regulate PFAS sites. This includes the USEPA establishing relatively low lifetime health advisory levels for PFAS in drinking water and even more stringent guidance and standards in several states. In addition, if PFAS are designated as hazardous substances at the federal level, as proposed by several Congressional bills, there could be wide‐reaching effects including listing of new Superfund sites solely for PFAS, application of stringent state standards, additional characterization and remediation at existing sites, reopening of closed sites, and cost renegotiation among PRPs.
• Chemistry and analytics: PFAS analysis is confounded by the lack of regulatory‐approved methods for most PFAS in water and all PFAS in solid media and air, interference with current water‐based analytical methods if samples contain high levels of suspended solids, and sample collection and analytical interference due to the presence of PFAS in common consumer products, sampling equipment, and laboratory materials.
• Toxicology and risk: Uncertainties remain related to human health and ecological effects for most PFAS; however, regulatory standards and guidance are being established incorporating safety factors that result in part per trillion (ppt) cleanup objectives. Given the thousands of PFAS that may be present in the environment, a more appropriate paradigm may be to develop toxicity criteria for groups of PFAS rather than individual PFAS.
• Transport and fate: The recalcitrance of many perfluoroalkyl compounds and the capability of some fluorotelomers to transform into perfluoroalkyl compounds complicate conceptual site models at many PFAS sites, particularly those involving complex mixtures, such as firefighting foams. Research is warranted to better understand the physicochemical properties and corresponding transport and fate of most PFAS, of branched and linear isomers of the same compounds, and of the interactions of PFAS with other co‐contaminants such as nonaqueous phase liquids. Many PFAS exhibit complex transport mechanisms, particularly at the air/water interface, and it is uncertain whether traditional transport principles apply to the ppt levels important to PFAS projects. Existing analytical methods are sufficient when combined with the many advances in site characterization techniques to move rapidly forward at selected sites to develop and test process‐based conceptual site models.
• Existing remediation technologies and research: Current technologies largely focus on separation (sorption, ion exchange, or sequestration). Due to diversity in PFAS properties, effective treatment will likely require treatment trains. Monitored natural attenuation will not likely involve destructive reactions, but be driven by processes such as matrix diffusion, sorption, dispersion, and dilution.

The consensus message from the Symposium participants is that PFAS present far more complex challenges to the environmental community than prior contaminants. This is because, in contrast to chlorinated solvents, PFAS are severely complicated by their mobility, persistence, toxicological uncertainties, and technical obstacles to remediation—all under the backdrop of stringent regulatory and policy developments that vary by state and will be further driven by USEPA. Concern was expressed about the time, expense, and complexity required to remediate PFAS sites and whether the challenges of PFAS warrant alternative approaches to site cleanups, including the notion that adaptive management and technical impracticability waivers may be warranted at sites with expansive PFAS plumes. A paradigm shift towards receptor protection rather than broad scale groundwater/aquifer remediation may be appropriate.     

How much heterogeneity? Flow versus area from a big data perspective

Over the past 10 years, there has been an increased recognition that matrix diffusion processes are a significant factor controlling the success of groundwater remediation. New field techniques and modeling tools have, consequently, been developed to understand how contaminants diffuse into and then out of low‐permeability (“low‐k”) zones and assess the resulting impact on groundwater quality. Matrix diffusion, in turn, is driven by one key factor: geologic heterogeneity. The importance of heterogeneity is being emphasized in the groundwater field by general rules of thumb such as “90% of the mass flux occurs in 10%‐20% of the cross‐sectional area” and conceptual models that show most of the groundwater flow occurs through the aquifer's “mobile porosity” which just a small fraction of commonly used effective porosity values (between 0.02 and 0.10 for mobile porosity vs. 0.25 for effective porosity). For this study, 141 boring logs from 43 groundwater remediation sites were evaluated to develop an empirically based estimate of the groundwater flow versus aquifer cross‐sectional area to confirm or reject the general flow versus area rules of thumb. This study indicated that at these 43 sites, an average of 30% of the cross‐sectional area carried 90% of the groundwater flow. Our flow‐only analysis does provide moderate (but not confirmatory) support for the “mobile porosity” concept with an estimated representative mobile porosity value of about 0.11 at the 43 sites.     

Yardsticks to Integrate Risk Assessment,Risk Management,and Groundwater Remediation

Remediation of contaminated sites has focused largely on restoration of groundwater aquifers. Often the stated remedial goal is to achieve conditions allowing unrestricted use and unrestricted exposure. Such total groundwater cleanup has occurred at some sites, but is the exception rather than the rule. At the same time, significant effort occurs to perform risk assessments for potential exposure to contaminants in groundwater at sites, both before and after remediation. The logical synergy between risk assessment and remediation is for risk management to seek opportunities for optimal use of groundwater based upon realistic expectations of cleanup technologies and the relevant acceptable residual (postremediation) levels of contaminants. This article explores an approach to improve this synergistic relationship between risk assessment, risk management, and remediation for groundwater cleanups. ©2015 Wiley Periodicals, Inc.     

Enhanced anaerobic bioremediation in a DNAPL residual source zone: Test Area North case study

An Interstate Technology and Regulatory Council (ITRC) forum was recently held that focused on case studies in which bioremediation of dense nonaqueous‐phase liquids (DNAPLs) was performed. This first case study, the Test Area North (TAN) site of the Idaho National Engineering and Environmental Laboratory, involves a trichloroethene (TCE) residual source area in a deep, fractured basalt aquifer that has been undergoing enhanced bioremediation since January 1999. Complete dechlorination from TCE to ethene was documented within nine months of operation, and sodium lactate injections were shown to enhance TCE mass transfer from the residual source. Since that time, optimization of injection strategies has maintained efficient dechlorination while demonstrating accelerated cleanup at a lower cost by changing to a whey powder amendment that solubilizes DNAPL. © 2006 Wiley Periodicals, Inc.     

Numerical simulation of surfactant-enhanced remediation of heterogeneous aquifer contaminated with nonaqueous phase liquids

A three-dimensional, multiphase, multicomponent compositional simulator was employed to simulate nonaqueous phase liquid (NAPL) migration during surfactant-enhanced aquifer remediation (SEAR) in spatially correlated heterogeneous fields. Aquifer heterogeneity was accounted for by considering the permeability to be a spatially random variable, and a geostatistical method was used to generate random distributions of the permeability. Spatial distributions of saturations in the NAPL and temporal changes of organic recovery, effluent concentrations of organics and surfactant, and pressure drop at the injection well for heterogeneous aquifers were compared with those in a homogeneous aquifer to examine the effects of different levels of heterogeneity. Variations in permeability fields have a pronounced effect on the organic recovery efficiency due to the long-term persistence of nonaqueous phase liquid and additional dispersion. Permeability heterogeneity also leads to the tailing off of effluent organic concentrations and significant loss in injectivity over the remediation period. For a small slug of surfactant, surfactant-enhanced remediation resulted in a relatively small improvement in the recovery of NAPL, especially in highly heterogeneous aquifers. Migration of high-concentration organic plumes to other layers by crossflow was also found to have a significant influence on SEAR behavior.