Abstract This field pilot study evaluates the feasibility of applying sulfate to a light non‐aqueous phase liquid (LNAPL) body for achieving enhanced source zone depletion (ESZD) at a refinery site. Two rounds of gypsum land application (GLA) delivered sulfate to an 8000 ft 2 area encompassing LNAPL‐containing monitoring wells MW‐2 and MW‐3, while MW‐1 served as a natural source zone depletion (NSZD) control. Over the 1044‐day study period, LNAPL and groundwater samples were collected to assess changes in LNAPL composition and key groundwater geochemical indicators, respectively. Principal component analysis revealed distinct and consistent compositional shifts in LNAPL under sulfate‐enhanced biodegradation conditions. LNAPL depletion rate for ESZD was 3‐fold faster than for NSZD. Notably, BTEX compounds also exhibited three‐ to sixfold enhanced depletion, while normal‐, iso‐, and cyclo‐alkanes also showed significant increases in depletion rates. Groundwater from GLA‐influenced wells demonstrated elevated sulfide, Fe(II), methane, and enriched δ 34 S‐sulfate, supporting active sulfate‐enhanced biodegradation. These results demonstrate that sulfate delivery can accelerate depletion of LNAPL and of its individual constituents in source zones, offering a practical and cost‐effective strategy for long‐term management at sites with persistent, unrecoverable LNAPL. Continued monitoring and adaptive sulfate delivery are recommended to further optimize BTEX attenuation in groundwater.
Diisopropyl ether (DIPE) is a branched ether used as a solvent, extractant, and gasoline additive. While aerobic biodegradation of DIPE has been documented, attenuation under anaerobic conditions (typical of groundwater impacted by petroleum hydrocarbons and fuel oxygenates) remains poorly constrained. This study investigated DIPE attenuation under anaerobic conditions using bench-scale microcosms constructed with site-specific groundwater and sediments collected from four locations within a DIPE-impacted aquifer in Southern California. Microcosms were assembled and incubated anoxically, with sterile controls included to quantify abiotic losses from sorption and volatilization. DIPE concentrations were monitored for approximately 120 d. After an initial equilibration period attributable to sorption, sustained DIPE decreases were observed in microcosms from three of four locations. The most pronounced and reproducible attenuation occurred in microcosms from one location (MW-C), where repeated DIPE respiking resulted in continued removal, indicating sustained attenuation capacity. First-order attenuation rate constants ranged from 0.0186 to 0.0266 d-1 prior to respiking and from 0.0076 to 0.0266 d-1 following respiking (estimated half-lives of 26-91 d). In contrast, sterile controls exhibited stable DIPE concentrations after sorption equilibration, supporting a biological contribution to attenuation. To evaluate microbial community shifts associated with DIPE exposure, 16S rRNA gene amplicon sequencing was performed on the MW-C microcosms. Communities shifted from taxa associated with oxic groundwater (e.g., Pseudomonas) toward enrichment of anaerobic and facultative anaerobic groups (e.g., Sulfuricurvum) during incubation. These results provide evidence that DIPE attenuation can occur under anaerobic conditions relevant to groundwater systems and support consideration of natural attenuation for DIPE-impacted plumes where aerobic conditions are limited.
Successfully attributing per- and polyfluoroalkyl substances (PFAS) contamination to specific sources may require employing advanced analytical techniques and interpreting complex multivariate datasets. This paper, the second in a series on PFAS forensics, presents some of the analytical and data analysis tools practitioners use to identify source-specific signatures and differentiate potential contamination sources. Advanced analytical approaches (i.e., chemical forensics) include indicator compound identification, isomer analysis, Total Oxidizable Precursors (TOP) Assay, and high-resolution mass spectrometry. Each technique can reveal different aspects of PFAS source characteristics, with the forensic value increasing as a combination of methods is applied. Data analysis frameworks (i.e., data forensics), including exploratory data analysis, proportional ratio analysis, visualization techniques, and multivariate statistical methods such as principal component analysis, hierarchical cluster analysis, and receptor modeling, can reveal forensic insights from these analytical results. Effective PFAS forensics requires an understanding of what analytical methods can measure and how to interpret resulting data within the context of site history, transformation pathways, and fate and transport mechanisms. The integration of advanced analytical techniques with systematic data interpretation often provides a foundation for defensible source attribution to inform site management decisions.
The long-term and frequent sampling of groundwater monitoring networks is common practice for monitoring the concentrations of constituents of potential concern (CoPC) over space and time. Sampling can be a costly and time-intensive process that also incurs health and safety risks. A redundancy-based ranking of candidate sampling locations can provide empirical evidence (after statistical modeling) on sites that are more influential for concentration estimation across the network, thus potentially reducing operational costs. Feedback for the open-source groundwater monitoring data analysis software GWSDAT highlighted the lack of dedicated statistical tools for such analyses. Using cross-validation (CV) to estimate the redundancy of candidate locations (by evaluating how the removal of each location affects the concentration estimates after model fitting) is computationally inefficient given large monitoring networks, due to the need for repeated model fitting and the large number of potential well combinations. In this paper, a novel, computationally efficient approach is proposed 'well redundancy analysis', which closely approximates CV-based results and only requires one model fitting to acquire an initial redundancy ranking of candidate sampling locations. The approach is based on Cook's distance (CD), an influence diagnostic for regression analysis, which is commonly used to quantify the influence of individual data points on regression models. The proposed approach uses the median CD value (computed after fitting a spline-based spatiotemporal model to historic observations at each candidate location) to rank the locations. The median CD values of the candidate locations are inversely proportional to their redundancy for the model. Using synthetic and real groundwater CoPC concentration monitoring data, this paper demonstrates that the proposed approach produces redundancy rankings that closely approximate the CV-based results, while providing computational advantages. The impact of this study is highlighted by the implementation of the proposed approach in the well redundancy analysis feature of GWSDAT.
Injectable colloidal activated carbon (CAC) has been commercially used to remediate per and polyfluoroalkyl substances (PFAS) in groundwater since 2016. The basis of the technology is enhanced PFAS retention within the aquifer from engineered increases in aquifer-matrix sorptivity ("PFAS enhanced retention"), with re-equilibration from the dissolved to the sorbed phase reducing contaminant mobility and exposure potential. Emplacement of 0.1 to 0.2% CAC can increase PFAS retention in aquifer solids by orders of magnitude relative to natural sorption alone. We synthesize a decade of publications and field experience to define practical design and performance expectations for CAC-based in situ PFAS remediation. Governing principles and key features of CAC reagents are outlined, showing how transport and emplacement behavior, enhanced sorptivity, and competitive PFAS/co-contaminant sorption control treatment longevity and spatial performance. Field data from 37 CAC barriers show PFAS concentration reductions of up to five orders of magnitude, with many post-treatment concentrations at or below reporting limits, while reactive-transport modeling under representative loading and competitive-sorption conditions indicates that well-designed CAC barriers can sustain protective performance over multidecadal to century-scale times. We argue that retention-in-place via CAC can achieve comparable risk reduction to active removal at lower lifecycle cost and liability, and outline conditions where integration with PFAS monitored retention may be advantageous. Practitioner experience highlights that accommodating matrix mass, competitive interactions, and flux delineation in design, together with optimized monitoring well placement, is critical to achieving and interpreting CAC performance.
Quantifying contaminant mass flux and plume mass discharge is essential for flux-based management of per- and polyfluoroalkyl substances (PFAS), yet few monitoring technologies provide the spatial resolution, repeatability, and cost efficiency needed for long-term application. Many PFAS plumes originating from discrete source areas exhibit high width-to-thickness ratios driven by recharge-dominated transport, weak vertical dispersion, and hydraulic conductivity anisotropy, limiting the effectiveness of conventional vertical transects for control-plane discharge assessment. This study evaluated the Vertebrae (TM) Well System (VWS), a segmented, nested horizontal multi-screen technology, for quantifying groundwater flux and perfluorooctane sulfonate (PFOS) mass discharge under relatively homogeneous aquifer hydraulic conditions at the Grayling Army Airfield (GAAF) in northern Michigan, USA. Two 500-feet VWS arrays were installed downgradient of a former aqueous film-forming foam source area, oriented perpendicular to groundwater flow. Vertical placement was informed by prior high-resolution characterization. Groundwater flux was quantified using multiple independent methods, including hydraulic profiling, tracer testing, active distributed temperature sensing, and grain-size-based permeability estimation, and integrated with PFOS concentrations across eight performance monitoring (PM) events. Geostatistical interpolation was used to support transect-scale flux and discharge integration in an unconfined sandy aquifer. Results indicate consistent spatial patterns and reproducibility across methods, with cross-method agreement within expected field-scale variability, supporting the interpretation that segmented horizontal wells can capture subtle lateral flux variability within the groundwater table fluctuation zone and quantify plume-scale PFOS mass discharge with high spatial and temporal resolution. Horizontal multi-screen systems provided repeatable, spatially resolved estimates of groundwater flux and PFOS mass discharge when screen placement targeted dominant concentration and groundwater flow horizons. These findings highlight the utility of segmented horizontal wells for flux-based remediation evaluation, regulatory decision-making, and long-term performance monitoring at PFAS-impacted sites.
Environmental forensics employs a variety of analytical tools and techniques to evaluate contamination sources and assess the scope and impact of contamination in environmentally relevant media. The key questions for an environmental forensics analysis are where, when and how contamination occurred. This paper, the first in a series on per- and polyfluoroalkyl substances (PFAS) forensics, presents some considerations for understanding and identifying source-specific PFAS signatures. Because different PFAS applications utilize distinct properties of these compounds, the composition of PFAS, the potential for their release and their fate and transport in the environment will vary based on industry and consumer products. Information critical to understanding and identifying PFAS sources include: (1) source type(s) (i.e., primary vs. secondary), (2) the manufacturing process, (3) formulations used, (4) how formulations may have changed over time, (5) how PFAS or PFAS formulations were used or applied, (6) relative magnitude of known or suspected releases and (7) potential precursor transformation products. This information will help develop a source-specific signature that can then be compared to documented contamination. In addition, the widespread usage of PFAS across numerous products, materials, and industries has potentially created a large, non-attributable source of PFAS (i.e., anthropogenic background). As a result, an understanding of baseline or "background" PFAS levels in our environment may be needed in an environmental forensics analysis. The combined information of potential source-specific signatures and anthropogenic background are essential for making informed and defensible decisions on where, when and how PFAS contamination occurred.
Identifying groundwater discharge locations is critical for understanding and monitoring groundwater contributions to streams in terms of water quantity and quality. Streams and rivers are under increasing strain from factors, including increased urbanization, agricultural land use, groundwater extraction, and climate change, which can alter groundwater-surface water exchange and increase contaminant loadings to surface water. Recommendations are needed to weigh the advantages, disadvantages, and applicable spatial scales of available discharge detection techniques. The current study evaluated the efficacy of employing a combination of manual point measurement techniques (visual analysis, infrared (IR) imaging, and streambed sediment temperature probing) to identify gaining stream reaches within a mid-sized watershed (similar to 80 km(2)) in southern Ontario. A reconnaissance survey was conducted where access was possible more than similar to 20 km (20%) of the total watercourse network, including the main channel and several tributaries. Two study sites identified as discharge hotspots based on the reconnaissance survey were selected for verification and detailed site characterization. Darcy's law flux estimates based on piezometer measurements and streambed flux estimates derived from multidepth vertical temperature profiling confirmed net groundwater discharge conditions at both sites. Differential stream gauging revealed net gaining conditions when applied at one site. Results suggest that a reconnaissance approach combining visual, handheld infrared, and streambed sediment temperature observations is a highly effective method for identifying locations of discharge hotspots and is recommended for other similar shallow streams (similar to <= 1 m depth). The approach is particularly useful along streams where overhanging vegetation impedes aerial IR imagery and/or discharge is largely submerged.
Natural source zone depletion (NSZD) is the process whereby immiscible hydrocarbon mass in the subsurface is removed by natural degradation processes. Hydrocarbon degradation produces methane, which exothermically oxidizes to carbon dioxide and results in elevated subsurface temperatures. Temperature gradient measurements in the unsaturated zone above the methane oxidation zone are an established means to calculate NSZD rates, and the governing equations and calculation methods are readily available. In contrast, downward heat flux beneath the zone of methane oxidation is excluded from temperature gradient NSZD rate estimates when there is difficulty measuring the vertically compressed downward temperature gradient in the unsaturated zone. Available NSZD guidance documents acknowledge the value of temperature measurements for assessing downward heat flux below the methane oxidation zone; however, methods to quantitatively calculate this contribution to NSZD rate are not readily available in instances where the temperature profile data are insufficient. In this contribution, we provide a method to calculate the NSZD rate attributable to downward heat flux into the saturated zone using groundwater temperature measurements in the saturated zone and provide a freely available R and python script to perform the rate calculation.
Characterizing the concentration and mass flux spatial distributions at sites with complex hydrogeologic conditions is critical to evaluate the fate and transport of contaminants, as well as to design effective remediation programs. PFAS (per- and polyfluoroalkyl substances)-contaminated sites are even more complex due to the large number of compounds and their diversity in physicochemical properties. To aid in this effort, a high-resolution passive profiler (HRPP), previously validated for measuring sediment porewater concentrations of chlorinated solvents, groundwater velocity, and geochemical parameters at high resolution (<= 20 cm) was adapted to evaluate PFAS. The study location was a shallow (<5 ft. to water) aquifer at a site with historical AFFF (aqueous film-forming foam) contamination. A discrete depth groundwater sampling system was used to obtain groundwater samples from specified depths (taken within 5 ft. of the HRPP installation locations) to compare to depth specific concentrations produced from the HRPP. Continuous cores were obtained from each HRPP installation location and HRPP were direct pushed into the borehole created during soil coring. Soil cores were field logged for soil texture. Five sets of HRPP strings ranging from 4 to 25 ft. below-ground surface (BGS) were deployed and left to equilibrate for 28 d. HRPP and DPT (direct push technology) Geoprobe (R) Screen Point Sampling System (SP16)-enabled discrete well samples were analyzed for concentrations of anions (Cl-, Br-, and SO42-) and subjected to PFAS target and suspect screening. The porewater concentration distributions of anions and targeted PFAS produced by the HRPP and SP16 methods were highly similar. Significant correlations were found at all sites with an r(2) of 0.76 for 23 targeted species quantified in both sample types (P << 0.01, n = 1386). The HRPPs and paired SP16 targeted PFAS concentrations matched with 60% within a factor of 2 and 88% within a factor of 5. Some concentration distribution features captured by the HRPP were not captured by the SP16 samples due to a lack of resolution. Importantly, steep concentration gradients at the capillary fringe and other features around low permeability zones also were not accurately quantified by the SP16 but were captured by the HRPP. Anionic, cationic, and zwitterionic suspect screening species were correlated in paired depth HRPP and SP16 samples (P << 0.01; n = 256 for anions, n = 258 for zwitterions, n = 188 for cations), although with anions the SP16 concentrations were generally higher. Overall, 29% of paired anion concentrations were within a factor of 2 and 85% within a factor of 5. These values were 38% and 89% for zwitterions, and 42% and 78% for cations, respectively. This work highlights the ability of the HRPP to produce accurate, high-resolution concentration profiles of groundwater using direct drive devices, a major advantage in highly heterogenous systems and low permeability media.
An overall assessment of groundwater quality and soils under paddy fields is vital for the sound management of groundwater resources. This study aims to assess human health risks from groundwater contamination in areas of continuous rice cultivation, by analyzing 72 groundwater from Someh Sara, Gilan, Iran. Groundwater and soil samples from beneath paddy fields were analyzed for major ions and heavy metals (HMs), nitrate (NO3), and phosphorus (P). Visual MINTEQ was used as a chemical equilibrium modeling software. The low NO3 (ranged from 0.01 to 1.88 mg/L) in groundwater was attributed to the anaerobic conditions. The P ranged from 0.001 to 0.089 mg/L. While Cu (ranged from 36 to 62.5 mu g/L) and Ni (ranged from 13.5 to 77 mu g/L) (only 2.7% of samples) concentrations did not exceed safe drinking water thresholds, Cd (ranged from 11.5 to 22 mu g/L) concentrations exceeded the acceptable limit in all groundwater samples. The results of HM exposure underline that all aged groups, especially infants and children are very vulnerable to noncarcinogenic health risks, spotlight the need for careful attention to reducing Cd concentrations in groundwater considered for drinking.
Groundwater in the upper Arkansas River corridor has been contaminated by elevated concentrations of dissolved solids in Arkansas River water. Most of the dissolved solids originate from the soils and bedrock in eastern Colorado, where river water is diverted for irrigation and storage systems, and evapotranspiration significantly increases the salinity of return flow to the river. Dissolved solids contents in low flows of the Arkansas River can sometimes exceed 4000 mg/L at the Colorado-Kansas border. To assess how sulfate and uranium in the river have contaminated the aquifer system in the corridor, which includes several municipal wellfields, a groundwater flow and transport model was developed. Due to its largely conservative characteristics, sulfate was directly simulated by the model; uranium, however, was estimated based on empirical relations between sulfate and uranium concentrations in groundwater samples. The comparison between the simulated and observed sulfate concentrations demonstrated a high level of agreement, affirming the accuracy of the model for assessing various water management scenarios. Our results indicate that river contaminants enter the aquifer through streambed infiltration and ditch diversion of river water for irrigation. Ditch irrigation and lateral groundwater movement have caused significant spreading of contaminants in the aquifer on both the north and south sides of the river. Any contamination treatment of the river water will likely take centuries to make a significant impact on the drinking water supply wells in the area due to the large amounts of contaminants already in the subsurface and the slow rate of lateral groundwater flow. Site-specific remediation measures in the immediate vicinity of wellfields will be needed to reduce contaminant concentrations in municipal supply wells.
Groundwater near a former fire training area (FTA) discharges to a nearby small perennial stream and is impacted by per- and polyfluoroalkyl substances (PFAS), including (generally listed in order from highest to lowest concentrations): perfluorooctane sulfonate (PFOS), perfluorooctanoic acid (PFOA), perfluorohexanesulfonic acid (PFHxS), perfluorohexanoic acid (PFHxA), perfluorobutanesulfonic acid (PFBS), and perfluorobutanoic acid (PFBA). Upstream of the FTA, total PFAS (i.e., the sum of PFAS measured by United States Environmental Protection Agency [USEPA] Method 537.1, or Sigma 24PFAS) concentrations in the stream are about 10,000 ng/L resulting from various other PFAS sources, however, after passing by a 400-feet long reach adjacent to the FTA, in-stream Sigma 24PFAS concentration increase by an average of about 15,000 ng/L as a result of the discharge of impacted groundwater (the average Sigma 24PFAS mass discharge from groundwater is estimated at about 5.0 g/d. Reactive core matting (RCM) consisting of adsorptive media (FLUORO-SORB (R) 200) woven between two layers of permeable geo-fabric was installed along a targeted reach of the stream to passively remove PFAS from discharging groundwater. Nine quarterly performance monitoring events, which included measurements of streamflow, and porewater and surface water concentrations, were conducted over a 2.4-year performance period. The RCM was highly successful in reducing Sigma 24PFAS mass discharging from groundwater to the stream, removing an estimated 8.2 kg of Sigma 24PFAS, which represents nearly all PFAS mass discharging from groundwater. The stream treatment values (calculated from the increase in concentration over the reach after RCM installation compared to baseline) were relatively high and consistent, all averaging over 80%. However, some PFAS (notably PFBA, PFBS, PFHxA, and PFHxS) show a slight decline in stream treatment values for the last two monitoring events, possibly indicating treatment efficiency may have begun to decline for these specific PFAS. This work highlights both the practical implementability and efficacy of this technology, and therefore, RCMs are recommended for consideration at other similar PFAS-impacted sites.
Accurately quantifying per- and polyfluoroalkyl substances (PFAS) mass discharge from the vadose zone to groundwater is critical for effective site characterization, risk assessment, and remediation planning. However, few standardized approaches exist for translating site-specific PFAS data into actionable mass discharge estimates. This paper presents a practical modeling framework that uses HYDRUS, a widely validated vadose-zone flow and transport code, to simulate PFAS leaching and estimate mass discharge to groundwater. The framework outlines step-by-step procedures for model selection, domain design, parameterization, and boundary condition specification, with emphasis on key PFAS processes such as tension-driven flow, nonlinear kinetic solid-phase sorption, and air-water interfacial partitioning. Guidance is provided on incorporating site-specific data-such as soil hydraulic properties, PFAS concentrations in biosolids or soils, and climate-driven infiltration dynamics-while addressing uncertainty through sensitivity analyses. Application of the framework demonstrates how HYDRUS can bridge laboratory, field, and regulatory needs by offering defensible, mechanistic predictions of PFAS flux to groundwater. The approach is adaptable across diverse site conditions, from agricultural fields impacted by biosolids land application to fire-training areas affected by aqueous film-forming foams (AFFF). By providing a transparent and reproducible methodology, this framework supports practitioners and regulators in improving conceptual site models, prioritizing monitoring strategies, and evaluating remediation performance. Ultimately, the framework advances the integration of vadose-zone modeling into groundwater management and decision-making for PFAS-impacted sites.
