Numerical simulations of radon as an in situ partitioning tracer for quantifying NAPL contamination using push-pull tests

Presented here is a reanalysis of results previously presented by [Davis, B.M., Istok, J.D., Semprini, L., 2002. Push-pull partitioning tracer tests using radon-222 to quantify non-aqueous phase liquid contamination. J. Contam. Hydrol. 58, 129-146] of push-pull tests using radon as a naturally occurring partitioning tracer for evaluating NAPL contamination. In a push-pull test where radon-free water and bromide are injected, the presence of NAPL is manifested in greater dispersion of the radon breakthrough curve (BTC) relative to the bromide BTC during the extraction phase as a result of radon partitioning into the NAPL. Laboratory push-pull tests in a dense or DNAPL-contaminated physical aquifer model (PAM) indicated that the previously used modeling approach resulted in an overestimation of the DNAPL (trichloroethene) saturation (S(n)). The numerical simulations presented here investigated the influence of (1) initial radon concentrations, which vary as a function of S(n), and (2) heterogeneity in S(n) distribution within the radius of influence of the push-pull test. The simulations showed that these factors influence radon BTCs and resulting estimates of S(n). A revised method of interpreting radon BTCs is presented here, which takes into account initial radon concentrations and uses non-normalized radon BTCs. This revised method produces greater radon BTC sensitivity at small values of S(n) and was used to re-analyze the results from the PAM push-pull tests reported by Davis et al. The re-analysis resulted in a more accurate estimate of S(n) (1.8%) compared with the previously estimated value (7.4%). The revised method was then applied to results from a push-pull test conducted in a light or LNAPL-contaminated aquifer at a field site, resulting in a more accurate estimate of S(n) (4.1%) compared with a previously estimated value (13.6%). The revised method improves upon the efficacy of the radon push-pull test to estimate NAPL saturations. A limitation of the revised method is that 'background' radon concentrations from a non-contaminated well in the NAPL-contaminated aquifer are needed to accurately estimate NAPL saturation. The method has potential as a means of monitoring the progress of NAPL remediation.     

Analysis of a gas-phase partitioning tracer test conducted in an unsaturated fractured-clay formation

The gas-phase partitioning tracer method was used to estimate non-aqueous phase liquid (NAPL), water, and air saturations in the vadose zone at a chlorinated-solvent contaminated field site in Tucson, AZ. The tracer test was conducted in a fractured-clay system that is the confining layer for the underlying regional aquifer. Three suites of three tracers were injected into wells located 14, 24, and 24 m from a single, central extraction well. The tracers comprised noble gases (traditionally thought to be nonsorbing), alkanes (primarily water partitioning), perfluorides (primarily NAPL partitioning), and halons (both NAPL and water partitioning). Observations of vacuum response were consistent with flow in a fractured system. The halon tracers exhibited the greatest amount of retardation, and helium and the perfluoride tracers the least. The alkane tracers were unexpectedly more retarded than the perfluoride tracers, indicating low NAPL saturations and high water saturations. An NAPL saturation of 0.01, water saturation of 0.215, and gas saturation of 0.775 was estimated based on analysis of the suite of tracers comprising helium, perfluoromethylcyclohexane and dibromodifluoromethane, which was considered to be the most robust set. The estimated saturations compare reasonably well to independently determined values.     

Diffusive partitioning tracer test for the quantification of nonaqueous phase liquid (NAPL) in the vadose zone: Performance evaluation for heterogeneous NAPL distribution

A partitioning tracer test based on gas-phase diffusion in the vadose zone yields estimates of the residual nonaqueous phase liquid (NAPL) saturation. The present paper investigates this technique further by studying diffusive tracer breakthrough curves in the vadose zone for a heterogeneous NAPL distribution. Tracer experiments were performed in a lysimeter with a horizontal layer of artificial kerosene embedded in unsaturated sand. Tracer disappearance curves at the injection point and tracer breakthrough curves at some distance from the injection point were measured inside and outside of the NAPL layer. A numerical code was used to generate independent model predictions based on the physicochemical sand, NAPL, and tracer properties. The measured and modeled tracer breakthrough curves were in good agreement confirming the validity of important modeling assumptions such as negligible sorption of chlorofluorocarbon (CFC) tracers to the uncontaminated sand and their fast reversible partitioning between the soil air and the NAPL phase. Subsequently, the model was used to investigate different configurations of NAPL contamination. The experimental and model results show that the tracer disappearance curves of a single-well diffusive partitioning tracer test (DPTT) are dominated by the near-field presence of NAPL around the tip of the soil gas probe. In contrast, breakthrough curves of inter-well tracer tests reflect the NAPL saturation in between the probes, although there is no unique interpretation of the tracer signals if the NAPL distribution is heterogeneous. Numerical modeling is useful for the planning of a DPTT application. Simulations suggest that several cubic meters of soil can be investigated with a single inter-well partitioning tracer test of 24-hour duration by placing the injection point in the center of the investigated soil volume and probes at up to 1 m distance for the monitoring of gaseous tracers.     

Radon as a naturally occurring tracer for the assessment of residual NAPL contamination of aquifers

The noble gas radon has a strong affinity to non-aqueous phase-liquids (NAPLs). That property makes it applicable as naturally occurring partitioning tracer for assessing residual NAPL contamination of aquifers. In a NAPL contaminated aquifer, radon dissolved in the groundwater partitions preferably into the NAPL. The magnitude of the resulting radon deficit in the groundwater depends on the NAPL-specific radon partition coefficient and on the NAPL saturation of the pore space. Hence, if the partition coefficient is known, the NAPL saturation is attainable by determination of the radon deficit. After a concise discussion of theoretical aspects regarding radon partitioning into NAPL, related experimental data and results of a field investigation are presented. Aim of the laboratory experiments was the determination of radon partition coefficients of multi-component NAPLs of environmental concern. The on-site activities were carried out in order to confirm the applicability of the "radon method" under field conditions.     

Performance assessment of NAPL remediation in heterogeneous alluvium

Over the last few years, more than 40 partitioning interwell tracer tests (PITTs) have been conducted at many different sites to measure nonaqueous phase liquid (NAPL) saturations in the subsurface. While the main goal of these PITTs was to estimate the NAPL volume in the subsurface, some were specifically conducted to assess the performance of remedial actions involving NAPL removal. In this paper, we present a quantitative approach to assess the performance of remedial actions to recover NAPL that can be used to assess any NAPL removal technology. It combines the use of PITTs (to estimate the NAPL volume in the swept pore volume between injection and extraction wells of a test area) with the use of several cores to determine the vertical NAPL distribution in the subsurface. We illustrate the effectiveness of such an approach by assessing the performance of a surfactant/foam flood conducted at Hill Air Force Base, UT, to remove a TCE-rich NAPL from alluvium with permeability contrasts as high as one order of magnitude. In addition, we compare the NAPL volumes determined by the PITTs with volumes estimated through geostatistical interpolation of aquifer sediment core data collected with a vertical frequency of 5-10 cm and a lateral borehole spacing of 0.15 m. We demonstrate the use of several innovations including the explicit estimation of not only the errors associated with NAPL volumes and saturations derived from PITTs but also the heterogeneity of the aquifer sediments based upon permeability estimates. Most importantly, we demonstrate the reliability of the     

Tracer test for the measurement of gas diffusion and non-aqueous phase liquid (NAPL) saturation in soil

During soil bioremediation, the diffusion of oxygen into the soil is an important prerequisite for aerobic biodegradation, and the decrease of petroleum products is the ultimate goal. Both processes need to be monitored. The aim of this work was to develop a gas tracer test that yields information on both, gas diffusion and residual saturation with non-aqueous phase liquids (NAPLs) in unsaturated soil heaps. One conservative tracer (methane) and 4 partitioning gas tracers (diethylether, methyl tert-butyl ether, chloroform and n-heptane) were injected as vapors into laboratory columns filled with unsaturated sand with increasing NAPL saturation. Breakthrough curves of gaseous compounds were measured at two points and compared to analytical solutions of an analytical diffusive-reactive transport equation. By fitting of methane data, robust results for effective diffusivity (tortuosity) were obtained. NAPL saturation was most accurately measured by the moderately water soluble tracers (ethers and chloroform). The hydrophobic tracer n-heptane did not partition into water-immersed NAPL. An easy and accurate way to assess air-NAPL partitioning constants from gas chromatography retention times is furthermore reported. It is concluded that gas tracer tests have the potential for measuring two important properties in soil bioremediation systems easily and quickly.     

Design of aquifer remediation systems: (1) describing hydraulic structure and NAPL architecture using tracers

Aquifer heterogeneity (structure) and NAPL distribution (architecture) are described based on tracer data. An inverse modelling approach that estimates the hydraulic structure and NAPL architecture based on a Lagrangian stochastic model where the hydraulic structure is described by one or more populations of lognormally distributed travel times and the NAPL architecture is selected from eight possible assumed distributions. Optimization of the model parameters for each tested realization is based on the minimization of the sum of the square residuals between the log of measured tracer data and model predictions for the same temporal observation. For a given NAPL architecture the error is reduced with each added population. Model selection was based on a fitness which penalized models for increasing complexity. The technique is demonstrated under a range of hydrologic and contaminant settings using data from three small field-scale tracer tests: the first implementation at an LNAPL site using a line-drive flow pattern, the second at a DNAPL site with an inverted five-spot flow pattern, and the third at the same DNAPL site using a vertical circulation flow pattern. The Lagrangian model was capable of accurately duplicating experimentally derived tracer breakthrough curves, with a correlation coefficient of 0.97 or better. Furthermore, the model estimate of the NAPL volume is similar to the estimates based on moment analysis of field data.     

Design of aquifer remediation systems: (1) Describing hydraulic structure and NAPL architecture using tracers

Aquifer heterogeneity (structure) and NAPL distribution (architecture) are described based on tracer data. An inverse modelling approach that estimates the hydraulic structure and NAPL architecture based on a Lagrangian stochastic model where the hydraulic structure is described by one or more populations of lognormally distributed travel times and the NAPL architecture is selected from eight possible assumed distributions. Optimization of the model parameters for each tested realization is based on the minimization of the sum of the square residuals between the log of measured tracer data and model predictions for the same temporal observation. For a given NAPL architecture the error is reduced with each added population. Model selection was based on a fitness which penalized models for increasing complexity. The technique is demonstrated under a range of hydrologic and contaminant settings using data from three small field-scale tracer tests: the first implementation at an LNAPL site using a line-drive flow pattern, the second at a DNAPL site with an inverted five-spot flow pattern, and the third at the same DNAPL site using a vertical circulation flow pattern. The Lagrangian model was capable of accurately duplicating experimentally derived tracer breakthrough curves, with a correlation coefficient of 0.97 or better. Furthermore, the model estimate of the NAPL volume is similar to the estimates based on moment analysis of field data.     

Two-dimensional laboratory simulation of LNAPL infiltration and redistribution in the vadose zone

A quantitative two-dimensional laboratory experiment was conducted to investigate the immiscible flow of a light non-aqueous phase liquid (LNAPL) in the vadose zone. An image analysis technique was used to determine the two-dimensional saturation distribution of LNAPL, water and air during LNAPL infiltration and redistribution. Vertical water saturation variations were also continuously monitored with miniature resistivity probes. LNAPL and water pressures were measured using hydrophobic and hydrophilic tensiometers. This study is limited to homogeneous geological conditions, but the unique experimental methods developed will be used to examine more complex systems. The pressure measurements and the quantification of the saturation distribution of all the fluids in the entire flow domain under transient conditions provide quantitative data essential for testing the predictive capability of numerical models. The data are used to examine the adequacy of the constitutive pressure-saturation relations that are used in multiphase flow models. The results indicate that refinement of these commonly used hydraulic relations is needed for accurate model prediction. It is noted in particular that, in three-fluid phase systems, models should account for the existence of a residual NAPL saturation occurring after NAPL drainage. This is of notable importance because residual NAPL can act as a non negligible persistent source of contamination.     

Impact of edible oil injection on the permeability of aquifer sands

Recent laboratory and field studies have shown that food-grade edible oils can be injected into the subsurface for installation of in-situ permeable reactive barriers. However to be effective, the oil must be distributed out away from the oil injection points without excessive permeability loss. In this work, we examine the distribution of soybean oil in representative aquifer sediments as non-aqueous phase liquid oil (NAPL oil) or as an oil-in-water emulsion. Laboratory columns packed with sands or clayey sands were flushed with either NAPL oil or a soybean emulsion followed by plain water, while monitoring permeability loss and the final oil residual saturation. NAPL oil can be injected into coarse-grained sands. However NAPL injection into finer grained sediments requires high injection pressures which may not be feasible at some sites. In addition, NAPL injection results in high oil residual saturations and moderate permeability losses. In contrast, properly prepared emulsions can be distributed through sands with varying clay content without excessive pressure buildup, low oil retention and very low to moderate permeability loss. For effective transport, the emulsion must be stable, the oil droplets must be significantly smaller than the mean pore size of the sediment and the oil droplets should have a low to moderate tendency to stick to each other and the aquifer sediments. In our work, oil retention and associated permeability loss increased with sediment clay content and with the ratio of droplet size to pore size. For sandy sediments, the permeability loss is modest (0-40% loss) and is proportional to the oil residual saturation.     

Kinetic limitations on tracer partitioning in ganglia dominated source zones

Quantification of the relationship between dense nonaqueous phase liquid (DNAPL) source strength, source longevity and spatial distribution is increasingly recognized as important for effective remedial design. Partitioning tracers are one tool that may permit interrogation of DNAPL architecture. Tracer data are commonly analyzed under the assumption of linear, equilibrium partitioning, although the appropriateness of these assumptions has not been fully explored. Here we focus on elucidating the nonlinear and nonequilibrium partitioning behavior of three selected alcohol tracers - 1-pentanol, 1-hexanol and 2-octanol in a series of batch and column experiments. Liquid-liquid equilibria for systems comprising water, TCE and the selected alcohol illustrate the nonlinear distribution of alcohol between the aqueous and organic phases. Complete quantification of these equilibria facilitates delineation of the limits of applicability of the linear partitioning assumption, and assessment of potential inaccuracies associated with measurement of partition coefficients at a single concentration. Column experiments were conducted under conditions of non-equilibrium to evaluate the kinetics of the reversible absorption of the selected tracers in a sandy medium containing a uniform entrapped saturation of TCE-DNAPL. Experimental tracer breakthrough data were used, in conjunction with mathematical models and batch measurements, to evaluate alternative hypotheses for observed deviations from linear equilibrium partitioning behavior. Analyses suggest that, although all tracers accumulate at the TCE-DNAPL/aqueous interface, surface accumulation does not influence transport at concentrations typically employed for tracer tests. Moreover, results reveal that the kinetics of the reversible absorption process are well described using existing mass transfer correlations originally developed to model aqueous boundary layer resistance for pure-component NAPL dissolution.     

Controlled release,blind tests of DNAPL characterization using partitioning tracers

The partitioning tracer technique for dense nonaqueous phase liquid (DNAPL) characterization was evaluated in an isolated test cell, in which controlled releases of perchloroethylene (PCE) had occurred. Four partitioning tracer tests were conducted, two using an inverted, double five-spot pumping pattern, and two using vertical circulation wells. Two of the four tests were conducted prior to remedial activities, and two were conducted after. Each test was conducted as a "blind test" where researchers conducting the partitioning tracer tests had no knowledge of the volume, method of release, nor resulting spatial distribution of DNAPL. Multiple partitioning tracers were used in each test, and the DNAPL volume estimates varied significantly within each test based on the different partitioning tracers. The tracers with large partitioning coefficients generally predicted a smaller volume of PCE than that expected based on the actual release volume. However, these predictions were made for low DNAPL saturations (average saturation was approximately 0.003), under conditions near the limits of the method's application. Furthermore, there were several factors that may have hindered prediction accuracy, including tracer degradation and remedial fluid interference.     

Investigation of surfactant-enhanced mass removal and flux reduction in 3D correlated permeability fields using magnetic resonance imaging

Magnetic resonance imaging (MRI) was used to visualize the NAPL source zone architecture before and after surfactant-enhanced NAPL dissolution in three-dimensional (3D) heterogeneously packed flowcells characterized by different longitudinal correlation lengths: 2.1 cm (aquifer 1) and 1.1 cm (aquifer 2). Surfactant flowpaths were determined by imaging the breakthrough of a paramagnetic tracer (MnCl(2)) analyzed by the method of moments. In both experimental aquifers, preferential flow occurred in high permeability materials with low NAPL saturations, and NAPL was preferentially removed from the top of the aquifers with low saturation. Alternate flushing with water and two surfactant pulses (5-6 pore volumes each) resulted in approximately 63% of NAPL mass removal from both aquifers. However, overall reduction in mass flux (Mass Flux 1) exiting the flowcell was lower in aquifer 2 (68%) than in aquifer 1 (81%), and local effluent concentrations were found to increase by as high as 120 times at local sampling ports from aquifer 2 after surfactant flushing. 3D MRI images of NAPL revealed that NAPL migrated downward and created additional NAPL source zones in previously uncontaminated areas at the bottom of the aquifers. The additional NAPL source zones were created in the direction transverse to flow in aquifer 2, which explains the higher mass flux relative to aquifer 1. Analysis using a total trapping number indicates that mobilization of NAPL trapped in the two coarsest sand fractions is possible when saturation is below 0.5 and 0.4, respectively. Results from this study highlight the potential impacts of porous media heterogeneity and NAPL source zone architecture on advanced in-situ flushing technologies.     

Fluorescent dye imaging of the volume sampled by single well forced-gradient tracer tests evaluated in a laboratory-scale aquifer physical model

This study presents a new method to visualise forced-gradient tracer tests in 2-D using a laboratory-scale aquifer physical model. Experiments were designed to investigate the volume of aquifer sampled in vertical dipole flow tracer tests (DFTT) and push-pull tests (PPT), using a miniature monitoring well and straddle packer arrangement equipped with solute injection and recovery chambers. These tests have previously been used to estimate bulk aquifer hydraulic and transport properties for the evaluation of natural attenuation and other remediation approaches. Experiments were performed in a silica glass bead-filled box, using a fluorescent tracer (fluorescein) to deduce conservative solute transport paths. Digital images of fluorescein transport were captured under ultraviolet light and processed to analyse tracer plume geometry and obtain point-concentration breakthrough histories. Inorganic anion mixtures were also used to obtain conventional tracer breakthrough histories. Concentration data from the conservative tracer breakthrough curves was compared with the digital images and a well characterised numerical model. The results show that the peak tracer breakthrough response in dipole flow tracer tests samples a zone of aquifer close to the well screen, while the sampling volume of push-pull tests is limited by the length of the straddle packers used. The effective sampling volume of these single well forced-gradient tests in isotropic conditions can be estimated with simple equations. The experimental approach offers the opportunity to evaluate under controlled conditions the theoretical basis, design and performance of DFTTs and PPTs in porous media in relation to measured flow and transport properties.     

A constitutive model for air-NAPL-water flow in the vadose zone accounting for immobile, non-occluded (residual) NAPL in strongly water-wet porous media

A hysteretic constitutive model describing relations among relative permeabilities, saturations, and pressures in fluid systems consisting of air, nonaqueous-phase liquid (NAPL), and water is modified to account for NAPL that is postulated to be immobile in small pores and pore wedges and as films or lenses on water surfaces. A direct outcome of the model is prediction of the NAPL saturation that remains in the vadose zone after long drainage periods (residual NAPL). Using the modified model, water and NAPL (free, entrapped by water, and residual) saturations can be predicted from the capillary pressures and the water and total-liquid saturation-path histories. Relations between relative permeabilities and saturations are modified to account for the residual NAPL by adjusting the limits of integration in the integral expression used for predicting the NAPL relative permeability. When all of the NAPL is either residual or entrapped (i.e., no free NAPL), then the NAPL relative permeability will be zero. We model residual NAPL using concepts similar to those used to model residual water. As an initial test of the constitutive model, we compare predictions to published measurements of residual NAPL. Furthermore, we present results using the modified constitutive theory for a scenario involving NAPL imbibition and drainage.     

Persistence of LNAPL sources: relationship between risk reduction and LNAPL recovery

Light nonaqueous phase liquids (LNAPLs), such as fuels, are the source of much soil and groundwater contamination. Though the mobility of LNAPLs is limited in many environments, dissolved-phase components, such as benzene, can produce groundwater plumes that are more mobile than the LNAPL source. In such a setting, it is commonly assumed that recovery of the LNAPL will result in a reduction in risk associated with the dissolved phase. This paper synthesizes several existing multiphase and chemical transport solutions into a single linked methodology that predicts concentrations of soluble constituents within and downgradient of LNAPL source zones from dissolution of those constituents into groundwater flowing through and below LNAPL sources. This approach has been applied to a variety of LNAPL spill conditions. For biodegradable compounds, these analyses show that the period of time where the dissolved-phase plume is expanding is very small compared to the duration of most LNAPL sources, and that the downgradient extent is generally less than about 100 m for BTEX compounds. Therefore, the risk to receptors, as measured by the maximum downgradient extent of dissolved-phase plume or the maximum concentration of these compounds at a downgradient receptor, is generally unrelated to the longevity of the LNAPL sources. The maximum downgradient extent of the dissolved-phase plume is determined almost entirely by the groundwater velocity and the biodegradation rate. These analyses further demonstrate that recovery of LNAPL by hydraulic methods is often ineffective at reducing risk. Except in coarse-grained soils or intermediate soils with significant LNAPL saturations, free-product recovery approaches do not result in significant reductions in the longevity of downgradient dissolved-phase contamination. Further, for biodegradable constituents, remediation does not result in a near-term decrease in the downgradient extent of contamination. Cleanup methods that act to change the composition of the LNAPL source are more effective at reducing the downgradient concentrations, particularly for fine-grained soils or when LNAPL saturations are low.     

Characterization of pore scale NAPL morphology in homogeneous sands as a function of grain size and NAPL dissolution

In this study, we investigate pore scale morphology of nonaqueous phase liquids (NAPLs) trapped in different pore sizes using tracer techniques. Specific interfacial area and saturation of NAPL trapped in homogeneous sands were measured using the interfacial and partitioning tracer techniques. The observed NAPL-water interfacial areas increased in a log-linear fashion with decreasing sand grain size, but showed no clear trend with residual NAPL saturation formed in the various grain sizes. The measured values were used to calculate the NAPL morphology index, which characterizes the spatial NAPL distribution within the pore space. The NAPL morphology indices, increased exponentially with decreasing grain size, indicating that the NAPL becomes smaller, but more blobs. For a fixed grain size, the specific interfacial area and saturation of the NAPL were measured following changes caused by dissolution using alcohol. The observed interfacial areas showed a decrease linearly as a function of the NAPL saturation.     

The effect of entrapped nonaqueous phase liquids on tracer transport in heterogeneous porous media: laboratory experiments at the intermediate scale

This work considers the applicability of conservative tracers for detecting high-saturation nonaqueous-phase liquid (NAPL) entrapment in heterogeneous systems. For this purpose, a series of experiments and simulations was performed using a two-dimensional heterogeneous system (10x1.2 m), which represents an intermediate scale between laboratory and field scales. Tracer tests performed prior to injecting the NAPL provide the baseline response of the heterogeneous porous medium. Two NAPL spill experiments were performed and the entrapped-NAPL saturation distribution measured in detail using a gamma-ray attenuation system. Tracer tests following each of the NAPL spills produced breakthrough curves (BTCs) reflecting the impact of entrapped NAPL on conservative transport. To evaluate significance, the impact of NAPL entrapment on the conservative-tracer breakthrough curves was compared to simulated breakthrough curve variability for different realizations of the heterogeneous distribution. Analysis of the results reveals that the NAPL entrapment has a significant impact on the temporal moments of conservative-tracer breakthrough curves.     

Three-dimensional visualization and quantification of non-aqueous phase liquid volumes in natural porous media using a medical X-ray Computed Tomography scanner

This study demonstrates the capabilities of a typical medical X-ray Computed Tomography (CT) scanner to non-destructively quantify non-aqueous phase liquid (NAPL) volumes, saturation levels, and three-dimensional spatial distributions in packed soil columns. Columns packed with homogeneous sand, heterogeneous sand, or natural soil, were saturated with water and injected with known quantities of gasoline or tetrachloroethene and scanned. A methodology based on image subtraction was implemented for computing soil porosity and NAPL volumes in each 0.35 mm x 0.35 mm x 1 mm voxel of the columns. Elimination of sample positioning errors and instrument drift artifacts was essential for obtaining reliable estimates of above parameters. The CT data-derived total NAPL volume was in agreement with the measured NAPL volumes injected into the columns. CT data-derived NAPL volume is subject to a 2.6% error for PCE and a 15.5% error for gasoline, at average NAPL saturations as low as 5%, and is mainly due to instrument noise. Non-uniform distributions of NAPL due to preferential flow, and accumulation of NAPL above finer-grained layers could be observed from the data on 3-D distributions of NAPL volume fractions.