Journal of Hydrology (2008) 360, 265 280

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jhydrol

Groundwater-derived contaminant uxes along a channelized Coastal Plain stream

Danita M. LaSage a, Alan E. Fryar Linnea J. Heraty d
a

b,*

, Abhijit Mukherjee c, Neil C. Sturchio d,

Kentucky Department of Natural Resources, Division of Mine Permits, 2 Hudson Hollow Complex, Frankfort, KY 40601, USA Department of Earth and Environmental Sciences, University of Kentucky, 101 Slone Building, Lexington, KY 40506-0053, USA c Bureau of Economic Geology, Jackson School of Geosciences, University of Texas at Austin, University Station, Box X, Austin, TX 78713-8924, USA d Department of Earth and Environmental Sciences, University of Illinois at Chicago, SES (m/c 186), 845 W. Taylor Street, Chicago, IL 60607-7059, USA
b

KEYWORDS Trichloroethene; Technetium; Groundwater; Stream; Coastal Plain; Kentucky

Summary Recent studies in various settings across eastern North America have examined the movement of volatile organic compound (VOC) plumes from groundwater to streams, but few studies have addressed focused discharge of such plumes in unlithied sediments. From 1999 through 2002, we monitored concentrations of trichloroethene (TCE) and the non-volatile co-contaminant technetium-99 (99Tc) along Little Bayou Creek, a rst-order perennial stream in the Coastal Plain of western Kentucky. Spring ow contributed TCE and 99Tc to the creek, and TCE concentrations tended to vary with 99Tc in springs. Contaminant concentrations in stream water uctuated seasonally, but not always synchronously with stream ow. However, contaminant inuxes varied seasonally with stream 2 ow and were dominated by a few springs. Concentrations of O2, NO 3 , and SO4 , values 37 of d ClDOCl in groundwater, and the lack of less-chlorinated ethenes in groundwater and stream water indicated that anaerobic biodegradation of TCE was unlikely. Losses of TCE along Little Bayou Creek resulted mainly from volatilization, in contrast to streams receiving diffuse contaminated discharge, where intrinsic bioremediation of VOCs appears to be prevalent. 2008 Elsevier B.V. All rights reserved.

Introduction
* Corresponding author. Tel.: +1 859 257 4392; fax: +1 859 323 1938. E-mail address: [email protected] (A.E. Fryar).

During the past two decades, awareness of the movement of groundwater contaminants to surface water has gradually

0022-1694/$ - see front matter 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jhydrol.2008.07.026

266 increased (Winter et al., 1998). In particular, Ward (1996), Stanley and Jones (2000) have noted the need to assess the impacts of groundwater-derived contaminants on aquatic ecosystems. Stanley and Jones (2000, p. 414) remarked, Even though the study of groundwater contaminant movement is well established, investigations of this same topic for near-stream environments and for the transmission of contaminants between groundwater and streams (in either direction) are rare. An emerging body of work has addressed the location of zones where plumes of volatile organic compounds (VOCs) discharge to streams. These compounds, which include fuel hydrocarbons and chlorinated solvents, are of concern because they are common groundwater contaminants; they can impact human health at aqueous concentrations as low as micrograms per liter; and some of them are resistant to degradation (National Research Council, 1984; Patrick et al., 1987; Zogorski et al., 2006). VOC plumes discharge to streams in a variety of geologic settings in eastern North America, including clastic sediments in the Atlantic and Gulf Coastal Plains (Vroblesky et al., 1991; Bradley and Chapelle, 1998; Lorah and Olsen, 1999; Fryar et al., 2000), glaciogenic sediments in New England and Canada (Kim et al., 1995; Kim and Hemond, 1998; Lyford et al., 1999; Church et al., 2002; Conant et al., 2004; Chapman et al., 2007), fractured crystalline rocks in the Piedmont and northern Appalachians (Vroblesky et al., 1996; Church et al., 2002), and karstied carbonates in the Interior Low Plateaus (Williams and Farmer, 2003; Vesper and White, 2006). Groundwater discharge at these sites ranges from diffuse in unconsolidated sediments to focused along preferential pathways such as fractures and conduits. Where diffuse discharge occurs, geochemical conditions may promote intrinsic biodegradation of compounds such as chlorinated ethenes along the groundwater-stream interface (Lorah and Olsen, 1999; Conant et al., 2004; Chapman et al., 2007). Within the literature on interactions of VOC plumes with streams, an underappreciated scenario is that of focused discharge through unlithied sediments. As noted by Lyford et al. (1999) and Conant et al. (2004), springs can develop along heterogeneities such as sand stringers or fractures in clay, thereby enabling plumes to short-circuit nergrained zones where biodegradation or sorption might be promoted. Herein, we present a case study of focused discharge of the chlorinated solvent trichloroethene (TCE) and the radionuclide technetium-99 (99Tc) to a stream in the northern Gulf Coastal Plain of the USA. This paper builds upon studies of TCE fate in groundwater (Sturchio et al., 1998), TCE and 99Tc discharge and in-stream attenuation (Fryar et al., 2000; Mukherjee et al., 2005), and groundwater discharge along the stream (Fryar et al., 2000; LaSage et al., 2008). The current study examines indicators of natural attenuation in stream water and groundwater, seasonal variability in contaminant uxes, and the utility of 99Tc as a conservative co-contaminant against which TCE attenuation can be compared, as suggested by Fryar et al. (2000).

D.M. LaSage et al. the stream has been channelized around ash ponds at the Tennessee Valley Authoritys Shawnee Plant, a coal-red generating station, and connected to Bayou Creek, a second-order perennial tributary of the Ohio River. Baseow in both creeks is sustained by process water from outfalls at the Paducah Gaseous Diffusion Plant (PGDP) and by groundwater discharge in the Ohio River oodplain (Fryar et al., 2000; LaSage et al., 2008). The main aquifer in the study area is the informally-named regional gravel aquifer (RGA), which consists of chert pebbles and cobbles in a sandy matrix (the Miocene-Pleistocene-age Mounds Gravel) (Clausen et al., 1992; LaSage et al., 2008). The RGA is underlain by the Porters Creek Clay (Paleocene) and the sandy to clayey McNairy Formation (Upper Cretaceous), and is locally conned by silt and clay of the overlying Metropolis Formation (Pleistocene) and by loess. Groundwater is probably recharged by leakage beneath pipes and lagoons at PGDP as well as by meteoric inltration (Jacobs EM Team, 1999). Flow in the RGA is north-northeastward toward the Ohio River, although temporary reversals in hydraulic gradient occur when the river is at ood stage (Clausen et al., 1992; Fryar et al., 2000). LaSage et al. (2008) provide additional detail on watershed hydrology, climate, and the geologic setting. PGDP has enriched uranium since 1952. From 1953 until 1993, TCE was used to decontaminate equipment and waste material at PGDP (Clausen et al., 1997). From 1953 to 1976, the feed plant at PGDP received reprocessed uranium oxide (UO2) and converted it to UF6 in a multi-step process, which resulted in the introduction of over 600 kg of 99Tc to the plant (Smith, 1984). Since 1988, offsite movement of groundwater contaminated with TCE and 99Tc has forced the abandonment of several private wells (CH2M Hill, 1992) and the provision of an alternate water supply to affected residents. In May 1994, the site was added to the National Priorities List (USEPA, 2007). Contaminant plumes extend laterally several kilometers within the RGA (Fig. 1). Several sources of TCE contamination exist on the site, including a residual pool of 795 m3 of dense, nonaqueous phase liquids either at the base of the RGA or within the Metropolis Formation (Sturchio et al., 1998). The plumes are thought to contain 3200 to 4700 kg of dissolved TCE, with concentrations ranging from near the detection limit (1 lg/L) in the distal portions to near the solubility limit (1100 mg/L) at or close to the source. 99Tc concentrations in the plumes at PGDP range from near the detection limit (25 pCi/L) at the far edges to 40,000 pCi/L near the source (ATSDR, 1997). Remediation of contaminated groundwater beyond the PGDP property boundary has entailed hydraulic control (via pumping wells) and treatment of pumped water. Coupled ow and transport models (Jacobs EM Team, 1999) indicated that, based on the current remediation strategy, off-site TCE concentrations in the northwest plume would continue to exceed the Safe Drinking Water Act maximum contaminant level (MCL; 5 lg/L) after 30 yr, while 99Tc concentrations would fall below the derived MCL (900 pCi/L) after 15 yr. Consequently, natural attenuation of TCE is of interest. Because of its widespread industrial use during the 20th century (Bedient et al., 1999), TCE is one of the most prevalent chlorinated VOCs detected in groundwater in North America (Bagley and Gossett, 1990; Stroo et al., 2003). Pro-

Background
Little Bayou Creek is a rst-order perennial stream in McCracken County, Kentucky (Fig. 1). Along its lower 2.5 km,

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

EB1

267

(b)
Ohio River

WB1 WB1.5
WB2/WB3

EB4 MS1 MS2 RR LBC-6 LBC-5

LB1
Y

ash pond LBC-1

Metro polis Lake

LBC-4

Bayou Creek

northwest plume LB6

Y

LBC-2 LBC-3

offsite 99Tc plume LBC-7 LBC-8 northeast plume

PGDP

(a)

IL MO KY
study area

Little Bayou Creek

stream monitoring site spring monitoring well 0

500 1000 m

TN
Figure 1 (a) Inset showing regional location of study area (modied from Fryar et al. (2007)). IL = Illinois; MO = Missouri; KY = Kentucky; TN = Tennessee. (b) Study-area map with monitoring locations and groundwater contaminant plumes. Insets (not to scale) along Little Bayou Creek show relative positions of the stream channel, springs, sites LBC-5 and -6, and wells LB1Y and LB6Y of Fryar et al. (2000). Stippled areas indicate TCE concentrations in groundwater P5 lg/L (MCL); hachured areas indicate 99Tc concentrations in groundwater P25 pCi/L (approximately 2 the detection limit). Modied from Mukherjee et al. (2005).

cesses that naturally remove dissolved TCE include volatilization (Rathbun, 1998); sorption to sediments, particularly when the fraction of organic carbon (foc) exceeds 0.001 (Fetter, 1999); and bacterially-mediated biodegradation, which is the only one of these processes that actually destroys contaminant mass. Intrinsic biodegradation of TCE typically entails anaerobic reduction, either cometabolic

degradation (Norris, 1994) or direct dehalogenation (halorespiration) (Middeldorp et al., 1999). Under sulfate-reducing conditions, TCE can be cometabolically reduced to one of the dichloroethene (DCE) isomers (cis-1,2-DCE [cDCE], trans-1,2-DCE, or 1,1-DCE), of which cDCE is the most prevalent (DeWeerd et al., 1998). In turn, the DCE isomers can be reduced under methanogenic conditions to vinyl

268 chloride (VC) and thence to ethene (Lorah and Olsen, 1999); cDCE can be oxidized in the presence of natural organic matter or CH4 (Bradley and Chapelle, 1998; Bradley et al., 1998a); or VC can be anaerobically oxidized in the presence of Fe(III) (Bradley et al., 1998b). Unlike TCE, 99Tc (the most abundant Tc isotope) is rarely encountered in environmental studies. 99Tc decays by beta emission with a half-life of 2.1 105 yr (Lloyd and Macaskie, 1996). 99Tc typically enters the environment via efuent at uranium enrichment facilities using reprocessed fuel (Blaylock et al., 1984), often through equipment decontamination (Till, 1984; Liang et al., 1996). Besides PGDP, sites in the USA where 99Tc occurs in groundwater have been documented by Beals and Hayes (1995), Ryan and Kipp (1997), Beasley et al. (1998), and the US Government Accountability Ofce (GAO (2005, 2006)). At all these sites except one (Beasley et al., 1998), 99Tc discharge to streams has been either documented or suspected. In oxygenated environments, Tc is rapidly hydrolyzed to TcO 4 , which is relatively mobile (Wildung et al., 1979). Under metal- or sulfatereducing conditions, TcO 4 can be reduced to a relatively insoluble (oxyhydr)oxide or sulde (Lloyd and Macaskie, 1996; Lloyd et al., 1998; Lloyd et al., 2000; Wildung et al., 2000).

D.M. LaSage et al. loids were preserved with 6 N HNO3, while TOC was preserved with 6 N H2SO4. VOC samples were collected in 40mL amber glass vials, which were closed with screw caps containing TeonTM-lined septa while excluding bubbles. Two to ve drops of 6 N HCl were added to each vial as a preservative. Samples for 99Tc were collected in 1-L PE bottles and preserved with 10 mL 6 N HCl or HNO3 except for 2000 and 2001 samples, which were not acidied, following the recommendation of laboratory staff. In May 2001, ve springs were sampled for analysis of the Cl stable isotope ratio (37Cl/35Cl) of dissolved organic Cl (d37ClDOCl) using preevacuated 350-mL Schlenk tubes with glass valve barrels and inert O-rings (Holt et al., 1997). Samples for VOCs and 37 Cl analyses were chilled during transport (except for an overnight shipment in May 2000) and stored at 4 C prior to analyses. During July 2000, we collected porewater from streambed sediments at three sites between LBC-5 and -4 to obtain vertical proles of VOC concentrations. We used multichambered acrylic samplers (dialyzers, or peepers) based on the design by Adams (1994). N2 gas was bubbled into boiling deionized water to lower DO concentrations to <1 lg/L. The sampler chambers were then lled with the deaerated, deionized water, a 0.45-lm polysulfonate membrane was placed over the lled chambers, and an acrylic cover with openings aligned with the chambers was placed over the membrane. Following Webster et al. (1998), the peepers were inserted into the stream bed and allowed to equilibrate for 3 weeks. Peeper 1 was inserted into sandy sediment proximal to channel spring MS2. Peeper 2 was in clayey sand adjacent to bank spring WB1.5. Peeper 3 was emplaced at the bend between WB1.5 and MS2, near several mineralized joints. Samples were collected at vertical intervals of 14 cm (typically 2 cm) beginning at 6 cm below the bed. Each sample was withdrawn from the chamber using a disposable syringe, injected into a weighed 8-mL glass vial containing a weighed amount of pentane equal to 4 mL, and mixed gently for 1 min. Samples were chilled during transport and stored at 4 C. Except as noted, pH was measured in the eld using a portable meter and electrode calibrated against pH 4 and 7 buffers, and DO was measured using the azide modication of the Winkler method with a digital titrator (Hach Company, 1997). Bicarbonate alkalinity was measured by digital titration with 1.600 N H2SO4 to a pH endpoint of 4.5. Other anions were analyzed by ion chromatography, and metals and metalloids were analyzed by inductively coupled plasma spectrometry, at Severn Trent Laboratories (STL, formerly Quanterra; Earth City, MO, USA) or the Kentucky Geological Survey. TOC was analyzed by oxidation on a TOC analyzer at STL. VOCs were analyzed at STL by gas chromatograph (GC)-mass spectrometer (MS) according to USEPA method SW 846/8260B. 99Tc was analyzed by liquid-scintillation counting at Teledyne Brown Engineering (Northbrook, IL, USA) for 1999 samples and at STL for other samples. Detected values for VOCs and 99Tc are considered to be those above laboratory reporting limits and, except as noted, 99Tc results refer to unltered samples. d37ClDOCl was analyzed using a gas-source isotope-ratio MS at the University of Illinois at Chicago. VOCs in porewater were analyzed using a GC with a 63Ni electron-capture detector in the Department of Plant and Soil Science at the University of

Methods
From June 1999 through May 2001, at 3- to 4-month intervals, we sampled stream water at as many as eight sites (LBC-1 through -8 (Fig. 1)) where LaSage et al. (2008) gauged stream ow. In addition, we sampled as many as eight springs (six along the stream banks and two in the channel (Fig. 1)) studied by LaSage et al. (2008) along the upper 400 m of the channelized reach of Little Bayou Creek. Typically, sampling began at the farthest downstream point and proceeded upstream, and stream samples were collected during one day of a given trip, while spring samples were collected on a second day. In January, June, August, and October 2002, we again sampled stream water at sites LBC-1 through -4. For comparison, we report analyses of stream and groundwater samples collected quarterly during 1997 at sites LB1 (750 m downstream of LBC-1) and LB6 (10 m downstream of LBC-4) of Fryar et al. (2000). Stream and spring samples were collected by immersing containers except at channel springs, where a seepage meter made from 4-in. nominal (10.2-cm actual inside diameter) PVC pipe was used (Fryar et al., 2000). At sites LB1 and LB6, monitoring wells were sampled using a submersible pump in January 1997 and a peristaltic pump in May, August, and October 1997. Springs, wells and a subset of streamwater sites were monitored for dissolved O2 (DO). All 1997 samples and some from January and August 2000 were analyzed for pH, metals, metalloids, and major anions; the 2000 subset was also analyzed for total organic carbon (TOC). All sites were monitored for VOCs and 99Tc except site LB1 in January 1997. For metals, metalloids, anions, and TOC, for VOCs and 99Tc during 1997, and for selected 99 Tc analyses thereafter, samples were passed through disposable 0.45-lm lters to obtain estimates of dissolved concentrations. DO samples were collected in 300-mL BOD bottles. Anion, metal, metalloid, and TOC samples were collected in polyethylene (PE) bottles. Metals and metal-

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream Kentucky. Further details on d37ClDOCl and porewater analyses are given by Holt et al. (1997) and LaSage (2004), respectively.

269

Results and discussion

Stream water and groundwater chemistry
Stream water shows two distinct spatial groups in terms of major solutes (Table 1). Samples from LBC-4/LB6 and farther upstream tended to have similar concentrations of Ca and Na on a milliequivalent basis, with no dominant anion. In contrast, LBC-1 and LB1 had CaSO4 or CaNaSO4 waters, with SO2 4 concentrations as high as 98.6 mg/L (Table 1). The downstream change in chemistry probably resulted primarily from seepage of ash-pond leachate into Little Bayou Creek, as inferred by Fryar et al. (2000) and LaSage et al. (2008). Flue gas desulfurization waste, which is generated at the Shawnee Plant (Hower et al., 1996), contains CaSO4 and is commonly co-disposed with y ash (US Federal Highway Administration [FHWA], 1998). DO concentrations in stream water (Table 1) approached equilibrium with atmospheric O2 (8 mg/L [Langmuir, 1997]). Fe and Mn concentrations, which were <1 mg/L (Table 1), may have been an artifact of metal (oxyhydr)oxide colloids passing through the 0.45-lm lter (Grout et al., 1999). As in the case of stream water, groundwater samples fall into two groups based on solute chemistry. Concentrations of Na, Ca, Mg, Cl, and HCO 3 were markedly higher, and SO2 concentrations were lower, for the springs and well 4 LB6Y than for well LB1Y (Table 2). Both wells are located on the west bank of Little Bayou Creek opposite the Shawnee Plant ash ponds. Depths from the streambed to the top of the 0.76-m-long well screens are 4.4 m for LB6Y and 2.6 m for LB1Y. The springs and LB6Y are located within the northwest plume, whereas LB1Y is cross-gradient to the plume (Fig. 1). In general, groundwater samples were more acidic (pH 5.326.79, median 5.83 (Table 2)) than surface water (pH 6.447.24, median 6.72 (Table 1)). TOC values for springs, which were near or below the detection limit of 1.0 mg/L (Table 2), were also lower than for stream water (Table 1). DO concentrations in springs, including values from sampling rounds not shown in Table 2, ranged from 2.92 to 5.10 mg/L and were 3.3 to 4.5 mg/L for 45 of 47 samples. These values are similar to those for wells LB6Y and LB1Y (Table 2) and overlap the DO range reported for six upgradient wells (4.47.9 mg/L (Clausen et al., 1997; Sturchio et al., 1998)), which are located in the northwest plume at depths as great as 29 m below land surface. Nitrate concentrations for springs and wells ranged from 6.99 to 11.1 mg/L (median 8.9 mg/L) (Table 1). Together, DO and NO 3 concentrations indicate aerobic conditions in groundwater (Clausen et al., 1997; Sturchio et al., 1998; Wiedemeier et al., 1998). Like for surface water, low but detectable Fe and Mn concentrations in groundwater could be associated with colloidal phases not excluded by ltration (Puls and Powell, 1992).

Contaminants in stream water and groundwater

TCE was the predominant VOC in stream water. Other VOCs (cDCE, CS2, 2-butanone, methylene chloride, and toluene)

were detected occasionally, usually at concentrations <10 lg/L. Some of these trace constituents (2-butanone, methylene chloride, and toluene) were also detected in blanks, which suggests laboratory contamination. TCE concentrations were below the detection limit (15 lg/L) at sites LBC-8 through -5 and peaked, except in January 2000, at site LBC-4 (Table 3). All of the springs except for RR were located between LBC-5 and -4. The maximum TCE concentration in stream water was 59 lg/L in June 2002. As noted by Fryar et al. (2000), TCE concentrations tended to decrease with distance downstream of LBC-4 (LB6). However, we do not make quantitative comparisons between contaminant data of Fryar et al. (2000) and our 19992002 data because the former values were collected on ltered samples. Except in June 1999, 99Tc concentrations in stream water at sites LBC-8 through -5 were near or below the detection limit, which was typically <10 pCi/L (Table 3). We do not know why 99Tc concentrations at sites LBC-7 through -5 were elevated during the rst round of sampling. As observed for TCE, 99Tc concentrations increased markedly from LBC-5 to -4, but in contrast to TCE, 99Tc concentrations did not systematically decrease with distance downstream of LBC-4. For seven of 12 sampling rounds between 1999 and 2002, the highest 99Tc concentration occurred at LBC-3, 300 m downstream of LBC-4. The overall maximum value in stream water (94.8 pCi/L) occurred in June 1999 at LBC-3. However, except for January 2002 (when the concentration was higher at LBC-3) and August 2002 (when the concentration was higher at LBC-4), the difference in concentrations between LBC-4 and -3 was within counting error. Contaminant concentrations at stream sampling points varied seasonally (Fig. 2ab). For both TCE and 99Tc, the highest concentrations occurred during June sampling events (1999 and 2002), while the lowest concentrations occurred in January (2000 and 2002). Stream ow also uctuated seasonally (LaSage et al., 2008), with maximum values in May or June and minimum values in autumn or winter. However, TCE concentrations increased from May to August 2000, and 99Tc concentrations increased from May through October 2000, whereas stream ow decreased from May through October 2000. Flow rates from the PGDP outfall sustaining Little Bayou Creek baseow (K010) did not vary markedly between May and October 2000, averaging 21.5 5.9 L/s (US Enrichment Corporation, unpublished data). Between August and October 2002, TCE and 99Tc concentrations decreased while stream ow rates increased. Divergent behaviors of contaminant concentrations and stream ow suggest that a seasonal decrease in overall groundwater discharge can coincide with an increase in the proportion of discharge from more contaminated parts of the RGA, and vice versa. Clausen et al. (1995) noted uctuations of contaminant concentrations and hydraulic head in PGDP monitoring well R-17, upgradient of Little Bayou Creek, indicating seasonal wobbling in plume orientations. Peaks in TCE and 99Tc concentrations were out of phase with hydraulic-head peaks, which is consistent with our observations. As in stream water, TCE was the predominant VOC in spring samples. TCE concentrations in springs ranged from 25 lg/L (EB1, May 2001) to 440 lg/L (MS2, May 2001) (Table

270

Table 1

Temperature, pH, and solute concentrations for stream monitoring sites along Little Bayou Creek T (C) pH Ca (mg/L) 34.1 40.8 41.0 37.6 18.0 21.5 28.0 17.8 26.6 18.7 21.9 17.7 21.7 Na (mg/L) 15.6 23.4 19.3 22.6 17.8 30.1 34.5 31.9 29.7 19.7 23.7 20.0 39.9 Mg (mg/L) 3.95 5.54 5.23 5.18 4.66 6.34 8.43 7.18 6.04 4.88 6.04 4.64 6.99 K (mg/L) 3.16 3.38 3.47 3.26 2.40 2.57 4.02 2.87 3.75 4.31 1.99 4.06 4.07 Fe (mg/L) 0.083 0.131 0.154 0.106 0.0287 0.075 0.034 0.159 0.421 0.131 0.440 0.499 0.130 Mn (mg/L) 0.529 0.563 0.452 0.368 0.153 0.080 <0.002 <0.002 Si (mg/L) 6.01 5.54 5.30 4.79 5.10 5.15 4.23 2.88 3.44 1.98 3.34 1.83 2.28 Cl (mg/L) 16.7 22.8 18.9 21.5 16.0 30.0 36.0 30.2 25.3 18.4 28.4 18.9 27.1 HCO 3 (mg/L)
SO2 4 (mg/L)

NO 3 (mg/L) 3.8 4.5 4.9 3.3 2.2 4.2 2.9 1.6 6.2 1.2 2.1 <0.089 2.5

DO (mg/L) 7.57b 7.74 6.96 6.96

TOC (mg/L)

LB1S January 1997 May 1997 August 1997 October 1997 LB6S January 1997 May 1997 August 1997 October 1997 LBC-1 January 2000 LBC-4 January 2000 August 2000 LBC-6 January 2000 LBC-8 January 2000

9.3 16.0 21.0 20.4 5.1 18.2 22.7 21.4 1.3 3.2 26.4 2.4 2.1

6.55 7.07a 6.44 6.70 6.80 7.07a 6.78 6.74 6.52 7.24

72 66 54

77.3 85.2 84.2 98.6 22.5 29.7 77.3 54.2 66.8 33.7 35.0 43.0 64.0

85 74 60 57 49 67 39 48

9.36 7.62 8.28 2.6 6.0 1.7 4.0 12.68 2.8

6.54 6.47

Blank = no data. a pH measured at start of alkalinity titration because of meter malfunction in eld. b DO measured using electrode.

D.M. LaSage et al.

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

Table 2

Temperature, pH, and solute concentrations for wells and springs along Little Bayou Creek T (C) pH Ca (mg/L) 13.2 12.0 10.9 9.68 21.7 19.4 17.9 17.4 24.3 25.1 26.8 26.9 24.1 24.6 20.9 23.9 21.5 Na (mg/L) 9.88 10.3 9.78 9.80 31.2 29.4 29.4 30.4 31.6 31.5 31.2 30.7 32.5 31.5 30.5 34.0 30.3 Mg (mg/L) 3.18 3.05 3.14 2.80 7.57 6.94 7.04 6.79 8.92 9.22 9.44 9.79 8.38 8.65 6.64 7.76 7.36 K (mg/L) <1.21 <1.21 1.72 1.36 <2.33 <1.21 2.36 2.02 2.03 <5.00 1.85 2.95 <5.00 <5.00 2.48 <5.00 2.60 Fe (mg/L) <0.006 0.049 <0.006 <0.006 0.239 <0.006 <0.006 0.030 0.121 0.0199 0.101 0.111 0.162 0.0234 <0.100 0.0280 0.0641 Mn (mg/L) <0.002 <0.002 <0.002 <0.002 2.05 0.130 0.233 0.551 Si (mg/L) 7.60 8.27 7.62 7.65 7.58 7.06 6.77 7.19 6.65 6.45 6.53 6.58 7.09 6.85 6.92 7.15 9.37 Cl (mg/L) 12.7 11.8 12.1 12.3 32.5 34.6 32.4 34.3 33.5 36.1 37.8 38.4 35.0 36.0 24.3 29.5 29.5 HCO 3 (mg/L)
SO2 4 (mg/L)

NO 3 (mg/L) 9.3 8.0 8.9 8.6 6.99 8.8 9.2 8.4 9.3 10.2 11.1 10.6 10.2 10.2 7.1 8.4 7.5

DO (mg/L) 3.64b 3.24 3.58 3.42 4.11b 3.78 4.20 3.94 3.90 3.80 3.82 3.50 3.94 3.72 2.92 3.50 3.90

TOC (mg/L)

LB1Y January 1997 May 1997 August 1997 October 1997 LB6Y January 1997 May 1997 August 1997 October 1997 EB1 January 2000 August 2000 EB4 January 2000 August 2000 WB1 January 2000 August 2000 WB2 January 2000 WB3 August 2000 RR January 2000

14.7 14.7 15.5 16.2 9.8 14.8 16.3 17.0 13.4 15.0 11.8 15.5 12.7 15.5 12.1 14.9 10.6

5.50 5.71a 5.32 5.61 6.02 6.28a 5.83 6.20 5.97 6.79 5.65 6.08 5.77 6.13 5.35 6.29 5.75

28 28 21

26.2 19.7 17.7 17.5 10.4 13.2 10.0 10.8 12.9 9.90 10.3 10.1 10.6 10.0 15.4 14.6 13.5

98 101 95 103 112 107 119

<1.0 <1.0 <1.0 <1.0 1.2 <1.0 <1.0 <1.0 <1.0

114 94 122 91

Blank = no data. a pH measured at start of alkalinity titration because of meter malfunction in eld. b DO measured using electrode.

271

272
Table 3 TCE and
99

D.M. LaSage et al.

Tc concentrations in stream water LBC-2 29 85.3 8.3 23 28.3 6.0 6.7 15.6 2.9 15.0 2.8 17 26.3 4.5 LBC-3 44 94.8 8.8 27 61.2 7.0 7.8 13.7 2.8 10.1 2.6 22 27.8 4.1 23 31.7 4.4 3.7 32.8 4.4 4.3 14.7 3.5 9.9 43.7 6.9a 8.2 23.7 4.2a 23 73.4 7.7 11 42.1 5.8 5.4 20.9 2.7 9.6 31.6 4.2 6.9 14.7 3.1 21 48.1 5.7 12 24.5 3.2 37 70.1 7.2 23a 44.6 5.1 10 32.7 3.9 15 38.0 4.7 9.3 17.0 2.8 26 56.0 6.3 16 28.6 3.5 45b 74.8 7.9 28 44.5 6.2 13 21.9 2.9 LBC-4 52 86.7 8.3 36 51.5 6.6 7.5 15.2 2.8 <13 15 25 24.6 3.8 30 30.7 4.3 17a 34.2 4.6 12 14.5 2.7 33 56.6 6.5a 18 21.9 3.0a 63b 82.9 8.5 37 58.7 6.9 18 23.6 3.0 LBC-5 4J 37.9 6.0 3J <6.0 1.3 6.4 J 2.2 5.6 J 2.2 1.6 7.1 J 4.0 2.4 <1.7 2.2 1.2 5.7 J 2.3 0.81 J 4.6 J 1.9 2.6 8.1 J 2.4 LBC-6 <5 23.9 5.1 <5 <5.6 0.52J 3.5 J 2.2 <4.2 2.7 1.1 6.3 J 3.1 0.73 J <2.5 3.8 <1.0 5.1 J 2.3 0.29 J 4.7 J 2.0 1.2 6.3 J 2.3 <1.0 <2.5 1.8 <1.0 <4.2 4.3a <1.0 4.4 J 2.2 <1.0 <2.6 2.3 <1.0 4.3 J 2.4 <1.0 <2.4 2.5 LBC-7 <5 91.8 9.0 <5 <6.6 <1.0 <0.5 2.1 6.8 J 4.0a <1.0 <3.9 2.7 LBC-8 3J <6.3 <5 <7.3 <1.0 <2.7 2.0 <2.3 2.1

LBC-1 June 1999 TCE (lg/L) 99 Tc (pCi/L) Sep. 1999 TCE (lg/L) 99 Tc (pCi/L) January 2000 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) May 2000 TCE (lg/L) 99 Tc (pCi/L) August 2000 TCE (lg/L) 99 Tc (pCi/L) October 2000 TCE (lg/L) 99 Tc (pCi/L) January 2001 TCE (lg/L) 99 Tc (pCi/L) May 2001 TCE (lg/L) 99 Tc (pCi/L) January 2002 TCE (lg/L) 99 Tc (pCi/L) June 2002 TCE (lg/L) 99 Tc (pCi/L) August 2002 TCE (lg/L) 99 Tc (pCi/L) October 2002 TCE (lg/L) 99 Tc (pCi/L) 16 68.6 7.5 10 40.4 5.9 3.5 11.5 4.8a 11.4 2.5 10 26.4 6.3

Blank = no data; = 99Tc counting error; J = value below laboratory reporting limit; < = value below detection limit; For January 2000 99Tc values, (f) = ltered sample. a Average of duplicate analyses (on a single sample or on split samples). b Analysis on diluted sample (original analysis out of calibration range).

4). Concentrations tended to be highest within 200 m of the head of the channelized reach (in the channel spring MS2, west-bank springs WB2 and WB3, and the tributary spring RR). Data are not available for all sampling rounds for all locations because (1) the channel springs (MS1 and MS2) were not always visible; (2) RR did not always ow vigorously enough to be sampled; and (3) ow from WB2 progressively declined between June 1999 and May 2000, while WB3 had emerged 5 m upstream of WB2 by May 2000 (LaSage et al., 2008). TCE concentrations were lowest in the farthest downstream springs (EB1 and EB4) along the east bank of the creek. We occasionally detected cDCE, CS2, 2-butanone, methylene chloride, toluene, and CCl4. Concentrations of cDCE were <16 lg/L, with the highest values at WB2, WB3, and MS2.

The relative spatial distribution of 99Tc concentrations in springs was generally similar to that of TCE. There is a strong linear relationship between concentrations of the contaminants (TCE = [1.199Tc] + 8.9; r2 = 0.86) (Fig. 3), as observed by Clausen et al. (1992) for PGDP monitoring wells. 99Tc ranged from 28.8 pCi/L in EB1 (August 2000) to 461 pCi/L in MS2 (May 2001) (Table 4). Except for September 1999, concentrations were greatest in springs along the uppermost 200 m of the channelized reach. In contrast to stream-water sampling sites, springs did not display consistent seasonality with respect to TCE or 99 Tc concentrations, although some uctuations occurred. 99 Tc concentrations in WB3, WB1, EB4, and EB1 decreased from May to August 2000, increased in October 2000, decreased in January 2001, and increased in May 2001. In con-

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

273

a
70

LBC-1 LBC-2

LBC-3 LBC-4

LBC-5 LBC-6

Concentration (g/L)

60 50 40 30 20 10 0 12/1/1999 12/1/2000 12/1/2001 12/1/2002 6/1/2000 6/1/1999 9/1/1999 3/1/2000 9/1/2000 3/1/2001 6/1/2001 9/1/2001 3/1/2002 6/1/2002 9/1/2002

Date

b
100

LBC-1 LBC-2

LBC-3 LBC-4

LBC-5 LBC-6

80 60 40 20 0

12/1/1999

12/1/2000

12/1/2001

12/1/2002

lower groundwater mix before discharging from RR, as suggested by LaSage et al. (2008) based on temperature data. Analyses of samples from the peepers indicate that TCE concentrations in stream-bed porewater are higher than in stream water, and maximum concentrations in porewater are higher than in groundwater from adjoining springs. No peaks of other VOCs were observed on the chromatograms. TCE concentrations ranged from 149 to 1691 lg/L in peeper 1, 40505 lg/L in peeper 2, and 16747 lg/L in peeper 3 (Fig. 4). Discounting obvious outliers (at 9 and 20 cm depth for peeper 1 and 21 cm depth for peeper 3), concentrations tended to decrease as depth decreased at all three sites. At the two downstream sites (peepers 2 and 3), these concentration decreases were more pronounced and appear to reect dilution of upwelling groundwater by stream water in the hyporheic zone, consistent with the conceptual description of Vroblesky (2002). Concentrations in peeper 1 were generally greater than measured in MS2 and WB3, although differences in the timing of sampling and analytical procedures preclude an exact comparison between peeper and spring samples. Concentrations of redox-sensitive species (in particular, 2 O2, NO 3 , and SO4 ), TCE concentrations, low to undetectable levels of cDCE (or other putative TCE daughter products) in springs, and the lack of VOCs other than TCE in porewater all indicate that intrinsic biodegradation of TCE via reductive dehalogenation was unlikely at our monitored sites (Wiedemeier et al., 1998). Likewise, d37ClDOCl data do not indicate reductive biodegradation of TCE. We report stable isotope data in using the conventional d notation, d37 ClDOCl; Rsample =Rstandard 1 1000 1 where R is the Cl/ Cl ratio and the reference material is standard mean ocean chloride (SMOC). The range of d37ClDOCl for the ve springs we sampled (RR, WB3, WB1, EB4, and EB1) was +0.3 to +1.3, with an analytical uncertainty of 0.2. This range falls within the range for ve northwest plume wells sampled by Sturchio et al. (1998) (+0.2 to +1.6). Because virtually all of the dissolved organic Cl in the sampled springs and wells occurred as TCE, the isotopic results in effect represent d37Cl of dissolved TCE. A plot of d37ClDOCl vs. TCE concentrations (Fig. 5) shows an overall decrease in TCE concentrations between wells and springs (i.e., along the ow path within the northwest plume) without a corresponding change in d37ClDOCl. Fractionation of 37Cl has been documented for volatilization of TCE (a = 0.99836 at 22 C (Poulson and Drever, 1999) to 0.99818 at 24 C (Huang et al., 1999)) and anaerobic reductive dehalogenation of TCE (a = 0.9940.995 at 30 C (Numata et al., 2002)), where a Rproduct =Rparent 2
37 35

Concentration (pCi/L)

6/1/2000

6/1/1999

9/1/1999

3/1/2000

9/1/2000

3/1/2001

6/1/2001

9/1/2001

3/1/2002

6/1/2002

Date

Figure 2 Concentrations of (a) TCE and (b) Tc in stream water. Values for monitoring locations LBC-7 and -8 are not shown because of a paucity of data. Values below detection limit are shown as 0.

99

trast, TCE concentrations in WB3 during the same period were relatively invariant (420430 lg/L). During June 1999 and May 2000, 99Tc concentrations in channel spring MS1 were comparable to values in west-bank spring WB1, while in September 1999, the 99Tc concentration in MS1 was similar to that in east-bank spring EB4. No simultaneous shift in TCE values was evident at those sites. However, both TCE and 99Tc concentrations in channel spring MS2 increased with the shift in discharge from WB2 to WB3. Of particular interest are springs WB1 and EB4, which are located opposite each other. WB1 exhibited consistently higher TCE and 99Tc concentrations than EB4, which might suggest that the east-bank springs discharge a mixture of more-contaminated groundwater and less-contaminated surface water. However, similarities in major solute concentrations for the two springs, and dissimilarities relative to stream water concentrations, suggest that both WB1 and EB4 are composed of groundwater with little or no stream water. An alternative explanation is that contaminated groundwater owing west to east beneath the stream mixes with uncontaminated groundwater entering the spring along the east bank before discharging into the stream. Likewise, the similarity in solute concentrations between RR and other springs suggests that deeper and shal-

9/1/2002

The apparent lack of fractionation in our case suggests dilution with water containing less (or no) TCE, which is consistent with other evidence of groundwater mixing in the springs.

Contaminant uxes to and from the stream

Analyses of stream water and groundwater indicate that geochemical conditions were not conducive to reductive

274

Table 4

TCE and

99

Tc concentrations in springs EB1 EB4 51 60.3 7.5 WB1 120 131.6 9.9 WB1.5 MS1 180 132.2 11.2 MS2 280b 257.4 13.6 WB2 305ab 249.2 13.5a 130.7 10.2a 325a 314.9 15.2 158.7 10.9 270b 221 22 183 19 130 84.6 9.3 400b 325 33 430b 271 28 271 28 420 267 27 381 38 420b 450 45 379 38 420 249 25a 276 28 450b 461 46 420b 395 40a 388 39 160b 155 16 WB3 RR 170 173.3 11.9

June 1999 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) Sep. 1999 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) January 2000 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) May 2000 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) August 2000 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) October 2000 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) January 2001 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L) May 2001 TCE (lg/L) 99 Tc (pCi/L) 99 Tc(f) (pCi/L)

32 42.4 6.3

38 45.6 6.6

72 81.2 7.8

110 135.5 10.1

170 81.7 7.9

250 107.9 9.1

170 71.6 7.5

40 33.8 4.4a 35.1 4.5 33 32.0 4.7

65 56.4 6.3 63.8 7.2 57 45.9 4.9

110 100 11 91.5 9.9 100 80.1 9.0

160 118 12 101 11 190 127 13

32 28.8 3.8

50 34.9 4.7

97 74.5 8.4

95 72.0 8.1

35a 47.7 5.9

53b 82.7 8.9

76b 116 12

120 116 12

37 34.6 4.4

42b 49.1 5.6

110 93 10

110 76.7 8.3

25b 41.9 5.3

34b 54.0 6.3

87b 112 12

80b 89.1 9.5

190b 229 23

Blank = no data; = 99Tc counting error. For 99Tc values, (f) = ltered sample. a Average of duplicate analyses (on a single sample or on split samples). b Analysis on diluted sample (original analysis out of calibration range).

D.M. LaSage et al.

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

500 400

275

300 200 100 0 0 100 200

99

6/99 9/99 1/00 5/00

8/00 10/00 1/01 5/01

300 Tc (pCi/L)

400

500

Figure 3 Variation of TCE with 99Tc concentrations in springs for different sampling rounds.

0 5 10

Peeper 1 Peeper 2 Peeper 3

et al., 1989; Gu and Dowlen, 1996). 99Tc was undetectable (<3.11 pCi/g) in four sediment samples collected between LBC-5 and -4 by LaSage (2004). For stream water collected in January 2000, 99Tc concentrations were slightly less in ltered than in unltered samples from LBC-2, -3, and -4; essentially equal in both samples from LBC-1; and below detection limit in both sets of samples from LBC-5, -6, -7, and -8 (Table 1). For spring WB2, concentrations were lower in ltered than in unltered samples in June 1999, September 1999, and January 2000 (Table 2). However, for splits collected subsequently from WB3, ltered values were higher in August 2000 and January 2001, equal in May 2000, and lower in October 2000 and May 2001 (Table 2). Taken together, water and sediment analyses indicate that reductive dehalogenation of TCE was unlikely in the discharge zone, and sorption of TCE and 99Tc was unlikely in the discharge zone or downstream sediments. Consequently, given the linear relationship between TCE and 99 Tc concentrations in groundwater, we can use 99Tc to estimate inuxes of TCE, corrected for volatilization, along the reach between LBC-5 and -4 where most of the springs occur. We dene ux as F Cd Q d Cu Q u 3

Depth (cm)

TCE (g/L)

15 20 25 30 35 0 200 400 600 800 1000

TCE (g/L)

Figure 4 TCE concentrations as a function of depth in porewater samples. Note that the highest concentration (1691 lg/L at 20 cm depth in peeper 1) is off-scale and thus not shown.

biodegradation of TCE during focused discharge. Additional data suggest that TCE sorption to stream-bed sediment was minimal. The foc of seven bed-sediment samples collected from three sites between LBC-5 and -4 averaged 0.000152 lg/lg (LaSage, 2004), which is below the limit for preferential sorption of TCE to organic matter (Fetter, 1999). These values are slightly lower than foc values measured for 12 subsoil samples from the Metropolis Formation and possibly loess at PGDP (range 0.00020.001 lg/lg; mean 0.0005 lg/lg) (CH2M Hill, 1992). Scha fer et al. (2000) reported that 10104 mg/kg TCE may sorb onto surfaces of minerals, particularly certain clays, but water outcompetes TCE for sorption sites on kaolinite particles. Using X-ray diffraction, LaSage (2004) determined the mineralogy of the ne (<0.075-mm) fraction of stream-bed sediments, which constituted 10% of the sample mass, to be 38% kaolinite and 37% to 46% illite, with the remaining percentages unresolved. Similarly, 99Tc sorption to stream-bed sediment appears to have been minimal. Using the PHREEQC code (Parkhurst, 1995) with solute data from January and August 2000, speciation modeling by LaSage (2004) indicated that Tc occurred in groundwater and stream water as TcO 4 , which is unlikely to sorb to sediment (Blaylock et al., 1984; Meyer

where C is concentration, Q is the stream-ow rate in L/s (from LaSage et al., 2008), and d and u refer to downstream and upstream monitoring sites, respectively. Contaminant inuxes between LBC-5 and -4 show a strong seasonality, consistent with stream ow (Fig. 6). Maximum values (1910 pCi/s for 99Tc inuxes and 1460 lg/ s for net inuxes of TCE [uncorrected for volatilization]) occurred in June 1999. 99Tc inuxes and net inuxes of TCE decreased from June 1999 to January 2000, rebounded in May 2000, decreased in October 2000, then rebounded in May 2001. Inux values are not shown for January 2001 because gauged Q at LBC-4 was less than at LBC-5, which results in physically nonsensical ux calculations. As shown by net Q (stream inows) between LBC-5 and -4 (Fig. 6), contaminant inuxes track seasonal uctuations in stream ow (LaSage et al., 2008) more closely than uctuations in contaminant concentrations (Fig. 2ab). Most of the

2.5
LBC springs
37 Cl DOCl (per mil SMOC)

2.0

NW plume NE plume

1.5

1.0

0.5

0.0 1 10 100 1000 10000 100000

TCE (g/L)

Figure 5 d37ClDOCl as a function of TCE concentrations for springs sampled in May 2001 and for northwest and northeast plume wells sampled by Sturchio et al. (1998). Arrow shows that compositions of springs can be explained by dilution of northwest plume water.

276
Tc, g/s TCE)
2500 2000 1500 1000 500 0 6/1/1999 9/1/1999 3/1/2000 6/1/2000 9/1/2000 3/1/2001 12/1/1999 12/1/2000 6/1/2001
20 15 10 5 0 -5

D.M. LaSage et al.

99Tc flux 5-4

NetQ 5-4 (L/s)

GW 99Tc flux Net TCE flux 5-4 Max TCE flux 5-4 GW TCE flux Net Q 5-4

Influx (pCi/s

99

Date

Figure 6 Net stream ow between LBC-5 and -4, in-stream contaminant uxes between LBC-5 and -4, and cumulative contaminant uxes for springs (GW) having both discharge and concentration data. Maximum TCE ux represents net TCE ux between LBC-5 and -4 corrected for volatilization.

spring discharge measurements were made between June 1999 and May 2000 (LaSage et al., 2008). Even though not all springs were gauged, and not all gauged springs were sampled, those that were both gauged and sampled (i.e., those for which point uxes could be calculated) comprised a substantial proportion of contaminant inuxes. These measured point uxes contributed 18% (in January 2000) to 72% (in May 2000) of the 99Tc inuxes, and 40% (in June 1999) to 93% (in May 2000) of the net TCE inuxes. In turn, springs WB2 and WB3, which had both the greatest ow rates and the greatest contaminant concentrations, represented 54% (in January 2000) to 94% (in May 2000) of measured 99Tc point uxes, and 55% (in January 2000) to 95% (in May 2000) of measured TCE point uxes. Because the linear regression of TCE vs. 99Tc for all spring samples failed statistical tests of normality and constant variance, we used linear regressions for individual sampling events, which passed those tests (except for May 2001 data, which failed the constant variance test), to correct net TCE inuxes for volatilization. The corrected inuxes suggest that losses of TCE owing to volatilization between discharge points and LBC-4 could have been as much as 890 lg/s (in June 1999) (Fig. 6). Interestingly, the proportion of mass loss by volatilization did not show the same sort of seasonality: the maximum proportional loss of 49% occurred in January 2000,

when volatilization overall should have been at a minimum because of low temperatures. However, combined errors in stream gauging, concentration measurements, and linear regressions make the TCE loss calculations more qualitatively than quantitatively signicant. Downstream of LBC-4, the attenuation of TCE relative to 99 Tc can also be explained by volatilization. Using Br, rhodamine WT, and NO 3 as slug tracers, Mukherjee et al. (2005) determined that sorption of TCE and reduction of TCE and 99 Tc were negligible or absent between LBC-4 and -3. Propane tracer tests conrmed that volatilization could account for TCE attenuation (Mukherjee et al., 2005). Efuxes between LBC-4 and -1 were positive for all sampling rounds for TCE but uctuated around zero for 99Tc (Fig. 7). Net Q values between LBC-4 and -1 track TCE efuxes more closely than 99Tc efuxes (Fig. 7). However, because discharge increases with both stream velocity and depth, whereas volatilization increases with stream velocity but decreases with depth (Rathbun, 1998), volatilization should not simply track discharge. Seasonal uctuations in volatilization are likely to be partly a function of temperature. Because volatilization decreases with decreasing aqueous concentrations (Rathbun, 1998), TCE could still be detected at lg/L levels at LB1 by Fryar et al. (2000), even after dilution by diffuse seepage downstream of LBC-1. Ultimately,

2000

30 25 20 15 10 5 0 -5 -10 6/1/1999 9/1/1999 3/1/2000 6/1/2000 9/1/2000 3/1/2001 6/1/2001 9/1/2001 3/1/2002 6/1/2002 12/1/1999 12/1/2000 12/1/2001 9/1/2002 12/1/2002

Tc, g/s TCE)

1500 1000 500 0 -500 -1000 -1500 -2000

NetQ 4-1 (L/s)

99Tc flux TCE flux Net Q 4-1

Efflux (pCi/s

99

Date

Figure 7

Net stream ow,

99

Tc uxes, and net TCE uxes in stream water between LBC-4 and -1.

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

277

TCE volatilization

strea

m flo

w
downwelling upwelling

diffuse groundwater discharge? focused groundwater discharge (spring)

b
TCE volatilization ash pond seepage (diffuse) w m flo strea

Contaminant discharge along Little Bayou Creek appears to have more in common with the sites in glaciated terrain than with the other Coastal Plain sites noted. As in springs along the Pine and Royal rivers, chlorinated ethene concentrations in springs along Little Bayou Creek were as high as several hundred micrograms per liter. TCE concentrations were tens of micrograms per liter in Little Bayou Creek, as in the Royal River and a small stream at the Connecticut site. Apart from the Royal River site, our site differs from those proled above in that discharge from the northwest plume is primarily focused through springs.

regional groundwater (plume) underflow

Summary and conclusions

We have documented that groundwater discharge is responsible for TCE and 99Tc loading to Little Bayou Creek, a rstorder perennial stream in the Gulf Coastal Plain of western Kentucky. Both stream water and groundwater were oxic, and TCE was the predominant VOC detected. TCE and 99Tc concentrations in stream water typically were undetectable upstream of the springs, near the head of the channelized reach of the stream. TCE concentrations peaked immediately downstream of the springs, then progressively decreased with distance farther downstream, whereas 99Tc usually peaked within 300 m downstream of the last spring, but did not systematically decrease with additional distance. Contaminant concentrations downstream of the springs appeared to uctuate seasonally, but those uctuations were not always in phase with stream-ow uctuations, which suggests that discharge originates from different parts of the aquifer at different times. 99Tc concentrations varied linearly with TCE concentrations for springs. Maximum concentrations of TCE (>400 lg/L) and 99 Tc (>400 pCi/L) occurred along the west bank near the head of the channelized reach. Mixing with less-contaminated groundwater appeared to dilute contaminant concentrations in springs, while mixing with downwelling stream water appeared to dilute TCE (and presumably 99Tc, which was not analyzed) in stream-bed porewater, as conceptualized in Fig. 8a. TCE and 99Tc concentrations in springs uctuated over sampling intervals of 3 to 4 months, but no seasonal trends were evident. Putative daughter products of TCE dehalogenation were detected in stream water, springs, and porewater only at low concentrations or not at all. The lack of these compounds, TCE concentrations, concentrations of redox-sensitive solutes, and the apparent lack of isotopic fractionation of ClDOCl in springs relative to up-gradient groundwater indicate that anaerobic biodegradation of TCE was insignicant in focused discharge zones. Likewise, reduction of 99Tc and sorption of TCE and 99Tc appear to have been minimal in focused discharge zones and stream sediments. Consequently, 99 Tc (which is non-volatile, and which varies with TCE) can be used to correct the net inux of TCE for in-stream volatilization. The 99Tc inux, the uncorrected net inux of TCE, and the corrected TCE inux uctuated seasonally with stream ow and were dominated by a few discharge points. Differential attenuation of TCE relative to 99Tc downstream of the springs is consistent with further volatilization. Dilution by diffuse seepage from the ash ponds and perhaps the RGA (LaSage et al., 2008) lowered concentrations of both

regional groundwater (plume) discharge (focused and diffuse[?]) regional groundwater discharge (diffuse)

Figure 8 Schematic diagram (not to scale) showing processes that affect contaminant concentrations (a) along the reach of Little Bayou Creek where springs discharge (between LBC-6 and -4) and (b) along the entire channelized reach of Little Bayou Creek (from LBC-6 to the conuence with Bayou Creek).

dilution by mixing with Bayou Creek and the Ohio River should make TCE and 99Tc in stream water undetectable.

Comparison with other sites

Prior studies of chlorinated ethene plumes discharging from unconsolidated aquifers to streams have shown a range of behaviors. At two sites in the Atlantic Coastal Plain, chlorinated ethenes were completely biodegraded during diffuse discharge through the stream bed. Oxidation of cDCE occurred under aerobic conditions (DO  2 mg/L in shallow bed sediments) at Cecil Field, Florida (Bradley and Chapelle, 1998). Reductive dechlorination of TCE occurred along the West Branch of Canal Creek, Maryland, as initially aerobic groundwater (DO > 1 mg/L) discharged through Fe(III)reducing, sulfate-reducing, and methanogenic zones within 5 m of land surface (Lorah and Olsen, 1999). In contrast, at three sites on glaciogenic sediments, pervasive biodegradation was not observed. At an unidentied site in Connecticut, TCE was biodegraded as groundwater evolved from aerobic to methanogenic along the owpath, but the primary mechanism of TCE loss from groundwater was discharge to a pond and tributary streams (Chapman et al., 2007). Conant et al. (2004) found that tetrachloroethene (PCE) discharged locally via geological windows through. . .semi-conning deposits (p. 263) to the Pine River in Ontario. Degradation of PCE to TCE and less chlorinated daughter compounds occurred in less permeable stream-bed sediments, where redox conditions evolved from nitrate-reducing to methanogenic (Conant, 2001). Lyford et al. (1999) delineated TCE discharge along the Royal River in Maine where erosion had breached a conning unit, but did not report water quality analyses for parameters other than TCE.

278 TCE and 99Tc in stream water, as conceptualized in Fig. 8b. Because we did not analyze sediment chemistry except for four 99Tc analyses or porewater chemistry other than chloroethenes in the peepers, we cannot rule out the possibility of reduction or sorption of contaminants in ner-grained sediments along Little Bayou Creek. Our ndings expand prior ndings on discharge of VOC plumes to streams elsewhere in eastern North America. Where focused discharge limits the residence time of contaminated groundwater in the riparian or hyporheic zone, reduction or sorption can be limited, and contaminant removal from the aqueous phase may depend substantially on volatilization downstream of the discharge zone. Moreover, because of the dynamic nature of spring discharge in unlithied sediments (LaSage et al., 2008), migration of discharge points where streams ow within such sediments may pose challenges for long-term monitoring.

D.M. LaSage et al.

Beasley, T.M., Dixon, P.R., Mann, L.J., 1998. 99Tc, 236U, and 237Np in the Snake River Plain aquifer at the Idaho National Engineering and Environmental Laboratory, Idaho Falls, Idaho. Environmental Science & Technology 32, 38753881. Bedient, P.B., Rifai, H.S., Newell, C.J., 1999. Ground Water Contamination, Transport and Remediation, second ed. Prentice-Hall, Upper Saddle River, NJ, 604p. Blaylock, B.G., Frank, M.L., Hoffman, F.O., DeAngelis, D.L., 1984. Behaviour of technetium in freshwater environments. In: Desmet, G., Myttenaere, C. (Eds.), Proceedings of a Seminar on the Behaviour of Technetium in the Environment. Elsevier Applied Science Publishers, San Diego, CA, pp. 7989. Bradley, P.M., Chapelle, F.H., 1998. Effect of contaminant concentration on aerobic microbial mineralization of DCE and VC in stream-bed sediments. Environmental Science & Technology 32, 553557. Bradley, P.M., Chapelle, F.H., Lovley, D.R., 1998a. Humic acids as electron acceptors for anaerobic microbial oxidation of vinyl chloride and dichloroethene. Applied and Environmental Microbiology 64, 31023105. Bradley, P.M., Chapelle, F.H., Wilson, J.T., 1998b. Field and laboratory evidence for intrinsic biodegradation of vinyl chloride contamination in a Fe(III)-reducing aquifer. Journal of Contaminant Hydrology 31, 111127. CH2M Hill, 1992. Results of the site investigation, Phase II, at the Paducah Gaseous Diffusion Plant, Paducah, Kentucky. Report KY/SUB/13B-97777C P-03/1991/1, prepared for the US Department of Energy under subcontract to DOE contract DE-AC0576OR00001. Chapman, S.W., Parker, B.L., Cherry, J.A., Aravena, A., Hunkeler, D., 2007. Groundwater-surface water interaction and its role on TCE groundwater plume attenuation. Journal of Contaminant Hydrology 91, 203232. Church, P.E., Vroblesky, D.A., Lyford, F.P., Willey, R.E. (Eds.), 2002. Guidance on use of passive-vapor-diffusion samplers to detect volatile organic compounds in ground-water discharge areas, and example applications in New England. US Geological Survey Water-Resources Investigations Report 02-4186, 81p. Clausen, J.L., Douthitt, J.W., Davis, K.R., Phillips, B.E., 1992. Report of the Paducah Gaseous Diffusion Plant groundwater investigation phase III. Report KY/E-150, prepared by Martin Marietta Energy Systems, Inc. for the US Department of Energy, 147p. Clausen, J.L., Sturchio, N.C., Heraty, L.J., Huang, L., Abrajano, T., 1997. Evaluation of natural attenuation processes for trichloroethylene and technetium-99 in the northeast and northwest plumes at the Paducah Gaseous Diffusion Plant, Paducah, Kentucky. Report KY/EM-113, prepared by Martin Marietta Energy Systems, Inc. for the US Department of Energy, 65p. Clausen, J.L., Zutman, J.L., Pickering, D.A., Farrow, N.D., 1995. Final report on drive-point proling of the northwest plume and analysis of selected data. Document KY/ER-66, prepared by Martin Marietta Energy Systems, Inc. for the US Department of Energy, 173p. Conant Jr., B., 2001. A PCE plume discharging to a river: investigations of ux, geochemistry, and biodegradation in the streambedPh.D. Dissertation. University of Waterloo, Canada, 543p. Conant Jr., B., Cherry, J.A., Gillham, R.W., 2004. A PCE groundwater plume discharging to a river: inuence of the streambed and near-river zone on contaminant distributions. Journal of Contaminant Hydrology 73, 249279. DeWeerd, K.A., Flanagan, W.P., Brennan, M.J., Principe, J.M., Spivack, J.L., 1998. Biodegradation of trichloroethylene and dichloromethane in contaminated soil and groundwater. Bioremediation Journal 2 (1), 2942. Fetter, C.W., 1999. Contaminant Hydrogeology, second ed. Prentice Hall, Upper Saddle River, NJ, 500p.

Acknowledgements
The research described herein was funded by the US Department of Energy (DOE) via the Agreement-in-Principle with the Commonwealth of Kentucky (through a contract from the Kentucky Division of Waste Management to the University of Kentucky (UK)) and via the Kentucky Research Consortium for Energy and Environment. Field work during 1997 was supported by US Geological Survey grant 1434-HQ-96-6R-02672 through the Kentucky Water Resources Research Institute. However, the contents of this paper do not necessarily reect views and policies of the US Government or the Commonwealth of Kentucky. Supplemental funding was provided by grants from the Geological Society of America Southeastern Section, the UK Graduate School, and the UK Department of Earth and Environmental Sciences Brown-McFarlan Fund to LaSage and Mukherjee. We thank DOE for providing data and access to PGDP and the Tennessee Valley Authority and the Kentucky Department of Fish and Wildlife Resources for providing access to their properties. Brian Begley, Gaye Brewer, David Butler, Lee Gatterdam, Andrea Hougham, Michael LaSage, Janet Miller, Christy Suh, Karen Thompson, and Eric Wallin helped with eld work. Brian Begley and two anonymous reviewers provided helpful comments on drafts of this paper.

References
Adams, D.D., 1994. Sediment pore water sampling. In: Mudroch, A., MacKnight, S.D. (Eds.), Handbook of Techniques for Aquatic Sediments Sampling, second ed. Lewis Publishers, Boca Raton, FL, pp. 171202. ATSDR (Agency for Toxic Substances and Disease Registry), 1997. Toxicological Prole for Trichloroethylene. <http://www.atsdr.cdc.gov/toxproles/tp19.html> (accessed June 2003). Bagley, D.M., Gossett, J.M., 1990. Tetrachloroethene transformation to trichlorothene and cis-1,2-dichloroethene by sulfatereducing enrichment cultures. Applied and Environmental Microbiology 56, 25112516. Beals, D.M., Hayes, D.W., 1995. Technetium-99, iodine-129 and tritium in the waters of the Savannah River Site. The Science of the Total Environment 173/174, 101115.

Groundwater-derived contaminant uxes alonga channelized Coastal Plain stream

FHWA, 1998. User guidelines for waste and byproduct materials in pavement construction. Publication FHWA-RD-97-148, US Department of Transportation, Federal Highway Administration. Fryar, A.E., Thompson, K.E., Hendricks, S.P., White, D.W., 2007. Groundwater ow and reservoir management in a tributary watershed along Kentucky Lake. Journal of the Kentucky Academy of Science 68 (1), 1123. Fryar, A.E., Wallin, E.J., Brown, D.L., 2000. Spatial and temporal variability in seepage between a contaminated aquifer and tributaries to the Ohio River. Ground Water Monitoring & Remediation 20, 129146. GAO (US Government Accountability Ofce), 2005. Department of Energy: preliminary information on the potential for Columbia River contamination from the Hanford Site. Report GAO-06-77R, 33p. GAO, 2006. Nuclear waste: DOEs efforts to protect the Columbia River from contamination could be further strengthened. Report GAO-06-1018, 28p. Grout, H., Wiesner, M.R., Bottero, J.-Y., 1999. Analysis of colloidal phases in urban stormwater runoff. Environmental Science & Technology 33, 831839. Gu, B., Dowlen, K.E., 1996. An investigation of groundwater organics, soil minerals, and activated carbon on the complexation, adsorption, and separation of technetium-99. Oak Ridge National Laboratory, Environmental Sciences Division Publication No. 4502. Report ORNL/TM-13154, prepared for the US Department of Energy under contract DE-AC05-96OR22464, 62p. Hach Company, 1997. Hach Water Analysis Handbook. Hach Company, Loveland, CO, 1309p. Holt, B.D., Sturchio, N.C., Abrajano, T.A., Heraty, L.J., 1997. Conversion of chlorinated organic compounds to carbon dioxide and methyl chloride for isotopic analysis of carbon and chlorine. Analytical Chemistry 69, 27272733. Hower, J.C., Robertson, J.D., Thomas, G.A., Wong, A.S., Schram, W.H., Graham, U.M., Rathbone, R.F., Robl, T.L., 1996. Characterization of y ash from Kentucky power plants. Fuel 75, 403 411. Huang, L., Sturchio, N.C., Abrajano Jr., T., Heraty, L.J., Holt, B.D., 1999. Carbon and chlorine isotope fractionation of chlorinated aliphatic hydrocarbons by evaporation. Organic Geochemistry 30, 777785. Jacobs EM Team, 1999. Transport modeling results for the northeast plume interim remedial action and the northwest plume at the Paducah Gaseous Diffusion Plant, Paducah, Kentucky. Report DOE/OR/07-1803&D1, prepared for the US Department of Energy by Jacobs Environmental Management, Kevil, KY, 100p. Kim, H., Hemond, H.F., 1998. Natural discharge of volatile organic compounds from contaminated aquifers to surface waters. Journal of Environmental Engineering 124, 744751. Kim, H., Hemond, H.F., Krumholz, L.R., Cohen, B.A., 1995. Insitu biodegradation of toluene in a contaminated stream. 1. Field studies. Environmental Science & Technology 29, 108 116. Langmuir, D., 1997. Aqueous Environmental Geochemistry. Prentice-Hall, Upper Saddle River, NJ, 600p. LaSage, D.M., 2004. Natural attenuation along a rst-order stream recharged by contaminated ground water. Ph.D. Thesis, University of Kentucky, USA, 238p. LaSage, D.M., Sexton, J.L., Mukherjee, A., Fryar, A.E., Greb, S.F., 2008. Groundwater discharge along a channelized Coastal Plain stream. Journal of Hydrology 360 (1-4), 252264. Liang, L., Gu, B., Yin, X., 1996. Removal of technetium-99 from contaminated groundwater with sorbents and reductive materials. Separations Technology 6, 111122. Lloyd, J.R., Macaskie, L.E., 1996. A novel PhosphorImager-based technique for monitoring the microbial reduction of technetium. Applied and Environmental Microbiology 62, 578582.

279

Lloyd, J.R., Nolting, H.-F., Sole , V.A., Bosecker, K., Macaskie, L.E., 1998. Technetium reduction and precipitation by sulfate-reducing bacteria. Geomicrobiology Journal 15, 4558. Lloyd, J.R., Sole , V.A., Van Praagh, J.R., Lovley, D.R., 2000. Direct and Fe(II)-mediated reduction of technetium by Fe(III)-reducing bacteria. Applied and Environmental Microbiology 66, 3743 3749. Lorah, M.M., Olsen, L.D., 1999. Natural attenuation of chlorinated volatile organic compounds in a freshwater tidal wetland: eld evidence of anaerobic biodegradation. Water Resources Research 35, 38113827. Lyford, F.P., Flight, L.E., Stone, J.R., Clifford, S., 1999. Distribution of trichloroethylene and geologic controls on contaminant pathways near the Royal River, McKin Superfund Site area, Gray, Maine. US Geological Survey Water-Resources Investigations Report 99-4125, 20p. Meyer, R.E., Arnold, W.D., Case, F.I., OKelley, G.D., 1989. Thermodynamics of technetium related to nuclear waste disposal: solubilities of Tc(IV) oxides and the electrode potential of the Tc(VII)/Tc(IV)-oxide couple. Division of Engineering, Ofce of Nuclear Regulatory Research, US Nuclear Regulatory Agency Report NUREG/CR-5235. Oak Ridge National Laboratory report ORNL-6503, 30p. Middeldorp, P.J.M., Luijten, M.L.G., van de Pas, B.A., van Eekert, M.H.A., Kengen, S.W.M., Schraa, G., Stams, A.J.M., 1999. Anaerobic microbial reductive dehalogenation of chlorinated ethenes. Bioremediation Journal 3 (3), 151169. Mukherjee, A., Fryar, A.E., LaSage, D.M., 2005. Using tracer tests to assess natural attenuation of contaminants along a channelized Coastal Plain stream. Environmental & Engineering Geoscience 11 (4), 371382. National Research Council, 1984. Groundwater Contamination. Studies in Geophysics, National Academy Press, Washington, DC, 179p. Norris, R.D. (Ed.), 1994. Handbook of Bioremediation. Lewis Publishers, Ann Arbor, MI, p. 257p. Numata, M., Nakamura, N., Koshikawa, H., Terashima, Y., 2002. Chlorine isotope fractionation during reductive dechlorination of chlorinated ethenes by anaerobic bacteria. Environmental Science & Technology 36, 43894394. Parkhurst, D.L., 1995. Users guide to PHREEQC a computer program for speciation, reaction-path, advective-transport, and inverse geochemical calculations. US Geological Survey WaterResources Investigations Report 95-4227, 143p. Patrick, R., Ford, E., Quarles, J., 1987. Groundwater Contamination in the United States, second ed. University of Pennsylvania Press, Philadelphia, PA, 513p. Poulson, S.R., Drever, J.I., 1999. Stable isotope (C, Cl, and H) fractionation during vaporization of trichloroethylene. Environmental Science & Technology 33, 36893694. Puls, R.W., Powell, R.M., 1992. Acquisition of representative ground water quality samples for metals. Ground Water Monitoring & Remediation 12, 167176. Rathbun, R.E., 1998. Transport, behavior, and fate of volatile organic compounds in streams. US Geological Survey Professional Paper 1589, 151p. Ryan, B.J., Kipp, K.L., Jr., 1997. Ground-water ow and contaminant transport at a radioactive-materials processing site, Wood River Junction, Rhode Island. US Geological Survey Professional Paper 1571, 89p. Scha fer, L., Yu, C., Newton, S.Q., Norman, M.A., Miller, D.M., Teppern, B.J., 2000. Molecular dynamics simulations of the sorption of organic compounds at the clay mineral/aqueous solution interface. European Congress on Computational Methods in Applied Sciences and Engineering, Barcelona, 1114 September 2000. Smith, R.F., 1984. Historical impact of reactor tails on the Paducah cascade. Report prepared by the Paducah Gaseous Diffusion

280
Plant, Paducah, Kentucky, for the US Department of Energy under contract No. W-7405-Eng-26. Stanley, E.H., Jones, J.B., 2000. Surface-subsurface interactions: past, present, and future. In: Jones, J.B., Mulholland, P.J. (Eds.), Streams and Ground Waters. Academic Press, San Diego, CA, pp. 405417. Stroo, H.F., Unger, M., Ward, C.G., Kavanaugh, M.C., Vogel, C., Leeson, A., Marqusee, J.A., Smith, B.P., 2003. Remediating chlorinated solvent source zones. Environmental Science & Technology 37, 224A230A. Sturchio, N.C., Clausen, J.L., Heraty, L.J., Huang, L., Holt, B.D., Abrajano Jr., T.A., 1998. Chlorine isotope investigation of natural attenuation of trichloroethene in an aerobic aquifer. Environmental Science & Technology 32, 30373042. Till, J.E., 1984. Source terms for technetium-99 from nuclear fuel cycle facilities. In: Desmet, G., Myttenaere, C. (Eds.), Proceedings of a Seminar on the Behaviour of Technetium in the Environment. Elsevier Applied Science Publishers, San Diego, CA, pp. 120. USEPA (US Environmental Protection Agency), 2007. Kentucky NPL/ NPL caliber site summaries, US DOE Gaseous Diffusion Plant. <http://www.epa.gov/region4/waste/npl/nplky/paducaky.htm> (accessed August 2007). Vesper, D.J., White, W.B., 2006. Comparative storm response of contaminants in a carbonate aquifer, Fort Campbell, KentuckyTennessee. In: Harmon, R.S., Wicks, C. (Eds.), Perspectives on Karst Geomorphology, Hydrology, and Geochemistry A Tribute Volume to Derek C. Ford and William B. White. Geological Society of America (GSA) Special Paper 404, GSA, Boulder, CO, pp. 267274. Vroblesky, D.A., 2002. Part 1. Guidance on the use of passive-vapordiffusion samplers. In: Church, P.E., Vroblesky, D.A., Lyford, F.P., Willey, R.E. (Eds.), Guidance on Use of Passive-VaporDiffusion Samplers to Detect Volatile Organic Compounds in Ground-Water Discharge Areas, and Example Applications in New England. US Geological Survey Water-Resources Investigations Report 02-4186, pp. 518. Vroblesky, D.A., Lorah, M.M., Trimble, S.P., 1991. Mapping zones of contaminated ground-water discharge using creek-bottom-sediment vapor samplers, Aberdeen Proving Ground, Maryland. Ground Water 29 (1), 712.

D.M. LaSage et al.

Vroblesky, D.A., Rhodes, L.C., Robertson, J.F., Harrigan, J.A., 1996. Locating VOC contamination in a fractured-rock aquifer at the ground-water/surface-water interface using passive vapor collectors. Ground Water 34 (2), 223230. Ward, C.H., 1996. Introductory talk: where are we now? Moving to a risk-based approach. In: Symposium on Natural Attenuation of Chlorinated Organics in Ground Water. US Environmental Protection Agency report EPA/540/R-96/509, pp. 13. Webster, J.T., Teasdale, P.R., Grigg, N.J., 1998. Theoretical and experimental analysis of peeper equilibration dynamics. Environmental Science & Technology 32, 17271733. Wiedemeier, T.H., Swanson, M.A., Moutoux, D.E., Gordon, E.K., Wilson, J.T., Wilson, B.H., Kampbell, D.H., Haas, P.E., Miller, R.N., Hansen, J.E., Chapelle, F.H., 1998. Technical protocol for evaluating natural attenuation of chlorinated solvents in ground water. US Environmental Protection Agency report EPA/600/R98/128. Wildung, R.E., McFadden, K.M., Garland, T.R., 1979. Technetium sources and behavior in the environment. Journal of Environmental Quality 8 (2), 156161. Wildung, R.E., Gorby, Y.A., Krupka, K.M., Hess, N.J., Li, S.W., Plymale, A.E., McKinley, J.P., Fredrickson, J.K., 2000. Effect of electron donor and solution chemistry on products of dissimilatory reduction of technetium by Shewanella putrefaciens. Applied and Environmental Microbiology 66, 2451 2460. Williams, S.D., Farmer, J.J., 2003. Volatile organic compound data from three karst springs in middle Tennessee, February 2000 to May 2001. US Geological Survey Open-File Report 03-355, 69p. Winter, T.C., Harvey, J.W., Franke, O.L., Alley, W.M., 1998. Ground water and surface water: a single resource. US Geological Survey Circular 1139, 79p. Zogorski, J.S., Carter, J.M., Ivahnenko, T., Lapham, W.W., Moran, M.J., Rowe, B.L., Squillace, P.J., Toccalino, P.L., 2006. The quality of our Nations waters volatile organic compounds in the Nations ground water and drinking-water supply wells. US Geological Survey Circular 1292, 102p.