VZJ sign up for citetrack
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 13 May 2005
Published in Vadose Zone J 4:337-344 (2005)
DOI: 10.2136/vzj2004.0100
© 2005 Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
Related Collections
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Ground Water Quality
Right arrow Capillary Fringe Processes
Right arrow Organic Compounds

SPECIAL SECTION: ZNS'03 VADOSE ZONE RESEARCH

Volatile Organic Compounds in the Saturated–Unsaturated Interface Region of a Contaminated Phreatic Aquifer

Daniel Ronena,*, Ellen R. Graberb and Yael Laorc

a Research Dep., Israel Hydrological Service, P.O. Box 20365, Tel Aviv 61203, Israel, and Ben-Gurion Univ. of the Negev, The Jacob Blaustein Institute for Desert Research, Dep. of Environmental Hydrology & Microbiology, Sde Boker Campus 84990, Israel
b Institute of Soil, Water and Environmental Sciences, The Volcani Center, Agricultural Research Organization, P.O. Box 6, Bet-Dagan, 50250, Israel
c Institute of Soil, Water and Environmental Sciences, Newe-Ya'ar Research Center, Agricultural Research Organization, P.O. Box 1021, Ramat Yishay, 30095, Israel
* Corresponding author (danronen{at}bgu.ac.il)

Received 24 June 2004.



   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Three volatile organic compound (VOC) field profiles were obtained during a period of 13 mo with a passive multilayer sampler (MLS) from a monitoring well located in the VOC-contaminated sandy phreatic Coastal Plain aquifer of Israel in the Tel Aviv area. The profiles presented here are unique in that they span both the saturated and unsaturated zones, through the saturated–unsaturated interface region (SUIR), and represent VOC concentrations from a single borehole. In groundwater just below the water table, the major contaminant, trichloroethylene (TCE), was present in concentrations up to 260000 µg/L water, and in the unsaturated zone just above the water table, in concentrations up to 124000 µg/L air. Other contaminants detected in high concentrations (as high as several thousands of µg/L) included tetrachloroethylene (PCE), cis-1,2-dichloroethylene (cis-1,2-DCE) and 1,1-dichloroethylene (1,1-DCE). In the three profiles, TCE and PCE concentrations were greatest at the water table and decreased with increasing distance from the water table both into the saturated and unsaturated zones. Temporal variations in maximal TCE vapor concentrations ranging from 44000 to 124000 µg/L air were also observed between profiles. The passive diffusion sampling characteristic of the MLS makes it possible to obtain unmixed vertical samples such that, for example, differences as great as 24000 µg TCE/L air can be measured in consecutive samples located only 12 cm apart in the unsaturated zone. The vertical detail is unique compared with other field sampling methods. Vertical detail is of utmost importance in interface regions, such as the SUIR, where water content in both the unsaturated and saturated zones varies significantly with depth, time, and space.

Abbreviations: {theta}, vertical water content • DCE, dichloroethylene • DO, dissolved oxygen • EC, electrical conductivity • KH, Henry's constant • MLS, passive multilayer sampler • MQL, minimum quantification level • p, pressure • PCE, tetrachloroethylene • PVC, polyvinyl chloride • SUIR, saturated–unsaturated interface region • TCE, trichloroethylene • VOC, volatile organic compound


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
IN THE UNSATURATED ZONE of an aquifer, VOCs (e.g., TCE, PCE, benzene) may be transported as vapors in all directions by diffusion and density driven flow, creating widespread gas plumes (Marrin and Thompson, 1987; Schwille, 1988; Mendoza and McAlary, 1990; Pouisen and Kueper, 1992; McCarthy and Johnson, 1993; Amali et al., 1996; Conant et al., 1996). Since vapor transport is independent of groundwater flow direction and velocity, the vapor VOC contamination front may precede the aqueous VOC front, creating secondary contamination sources in the unsaturated zone (Conant et al., 1996; Graber et al., 2002).

The VOC vapors may also intrude into overlying buildings and other structures, creating both acute and chronic hazards (Interstate Technology and Regulatory Council, 2003). The hazards associated with vapor phase contaminants and their role as a source for groundwater contamination depend in large part on site specific conditions affecting vapor spread. These conditions include porous media permeability, which can vary as a result of spatial changes in water content. To assess VOC fluxes, and hence the risks of vapor intrusion into buildings or the role of vapors in groundwater and unsaturated zone contamination, comparative investigations of VOC distributions in groundwater and in the vapor phase along vertical profiles are essential. Such investigations need to also consider, with particular emphasis, temporal changes in VOC vapor concentrations along vertical profiles in the SUIR (McCarthy and Johnson, 1993; Jellali et al., 2003).

The SUIR is an important interface connecting the vadose and saturated zones. It is very heterogeneous and of transient and spatially variable character in terms of water content, gas-filled porosity, and bacterial activity (Ronen et al., 1997, 2000; Affek et al., 1998). The SUIR was defined as the region composed of the vertical juxtaposition of two zones: (i) the capillary fringe, the zone of vertical water content ({theta}) with a water phase under negative pressure (head; p < 0), bounded from below by the imaginary surface where p = 0 (the water table) and from above by the minimum height at each point above this surface at which {theta} is equal to the background residual water content, and (ii) the zone below the water table defined by the imaginary plane where saturation is 100% and p > 0 to the plane where p = 0 (Ronen et al., 2000). In their study of the sandy phreatic Coastal Plain aquifer of Israel, Ronen et al. (2000) showed that the average thickness of the SUIR was about 3 m, and calculated that residence time of recharge in this zone could exceed 5 yr.

Existing active and passive field sampling methods (e.g., Pankow et al., 1984; Conant et al., 1996; Hewitt, 1999; Hers et al., 2000; Jellali et al., 2003) are not adequate for exploring VOC dynamics in the SUIR where water content is highly variable with depth such that the vertical distance between samples should be only a few centimeters. We suggest that such small scale variability can be explored using a MLS, as recently demonstrated for TCE in a laboratory study by Laor et al. (2003). In the present study, we present results demonstrating a field application of the MLS, where temporal variations in detailed vapor and groundwater profiles of TCE, PCE, cis-1,2-DCE and 1,1-DCE were observed at the water table region of a deep ({approx}18 m) phreatic VOC-contaminated aquifer.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Monitoring Well
The VOC profiles were obtained in a monitoring well (Fig. 1) located in the Tel Aviv area in the Coastal Plain aquifer of Israel. Bulk groundwater in this area (about 200 km2) is extensively contaminated by VOCs (Graber et al., 2002), with concentrations in water production wells as high as 650 µg/L TCE, 92 µg/L PCE, 35 µg/L cis-1,2-DCE, and 56 µg/L 1,1-DCE. The average thickness of the unsaturated zone of Coastal Plain aquifer of Israel is about 30 m, and the maximum thickness of the aquifer is about 180 m. The monitoring well is situated at a former industrial site located at a highly contaminated portion of the aquifer. The well was drilled with a spiral drill without addition of drilling fluids and was designed for studying the water table region and shallow groundwater, with a continuous 10-cm-i.d. polyvinyl chloride (PVC) screen installed from 13 m below the water table to 3 m above it.



View larger version (60K):
[in this window]
[in a new window]
  		Fig. 1. Schematic representation of the monitoring well and sedimentary sequence. The inset shows a photograph of two dialysis cells (150 mL each) of the multilayer sampler (MLS) used in the study. PVC = polyvinyl chloride, wt = water table.

 
Passive Multilayer Sampler
Profiles were obtained using a MLS. The MLS has been described in detail elsewhere (Ronen et al., 1987; Laor et al., 2003). Briefly, it consists of individual stainless steel cells (150 mL each) that can be attached in a modular fashion (Fig. 1). Each cell is isolated from the adjacent one by a seal, and the distance between the edges of two consecutive cells is 6 cm. For obtaining a groundwater profile or a profile of gases from the unsaturated zone, the cells are filled with distilled water and are closed at both ends with a permeable membrane (0.2 µm, Versapor, Pall Gelman Sciences, Inc., East Hills, NY). The sampler is lowered into an observation well and then retrieved after a given time interval (days, weeks, months). The chemical composition of the water in each dialysis cell is then determined. The dialysis cells achieve equilibrium with dissolved VOCs in water in about 6 d (for benzene, toluene, ethylbenzene, and xylene; Kaplan et al., 1991) and with gas phase VOCs in about 48 to 72 h (for TCE and naphthalene, respectively; Laor et al., 2003). After equilibration, concentrations of VOCs in the water of cells situated below the water table are taken to directly reflect aqueous concentrations in the aquifer adjacent to the cells, while for the unsaturated zone, aqueous concentrations measured in the dialysis cells are used to calculate concentrations in the gas phase according to Henry's law. Gas phase VOC values were calculated using the following nondimensional Henry's constant (KH) at 20°C: TCE = 0.40 (average of values reported in Munz and Roberts, 1987; Culver et al., 1991; and Schwartzenbach et al., 1993); PCE = 0.533, cis-1,2-DCE = 0.140, and 1,1-DCE = 0.975 (Staudinger and Roberts, 2001). Groundwater temperature in the Coastal Plain aquifer has been shown to be fairly constant (around 22°C; Ronen et al., 1988). Therefore, temperature corrections for KH were neglected.

In Laor et al. (2003), the performance of the MLS under conditions of soil heterogeneity was specifically tested in laboratory experiments. It was demonstrated that the MLS accurately reproduces TCE unsaturated zone concentrations in stratified sediments consisting of a thin layer (3 cm) of sandy loam (moisture content 16.5%) sandwiched between two layers of coarse air-dry sand (moisture content < 0.1%; average thickness 40 cm).

MLS Installation and Retrieval
Three profiles were obtained during the course of this work. In the first profile (hereafter referred to as MLS-1), five MLS sections were deployed in the well to achieve a maximum sampling interval of about 11 m in the saturated zone and about 1 m above the water table. Each section, composed of a sequence of cells, was connected to the adjacent one by a rope. In the second profile (MLS-2), all cells were connected together, forming a continuous 4.6-m-long sampler. In the third profile (MLS-3), four MLS sections were connected with a stainless steel wire for a total length of about 5 m. For all profiles, the MLS was lowered into the well with PVC-coated weights attached to the lower end of the sampler.

Immediately on retrieval of the MLS, the cells were closed with PVC caps until each cell was sampled by filling duplicate precleaned and acidified 40-mL VOC vials by pouring from the dialysis cells in a randomized order. Altogether, sampling lasted no more than 1 h. In our previous study (Laor et al., 2003), we showed that maximum losses of TCE from a capped dialysis cell, during a 1 h time interval, did not exceed 10%. Electrical conductivity (EC) and O2 were measured in the water remaining in each cell. Water samples for VOC analysis were stored on ice and packed for same-day express air shipping to the analytical laboratory in a refrigerated container guaranteed to maintain temperature between 2 to 8°C. The samples were received after 4 d by the analytical laboratory in good condition (4°C) and were analyzed for VOC content within 1 wk of arrival.

The three profiles were obtained over a period of 13 mo, during which time the depth to the water table decreased by 0.9 m. Passive Multilayer Samplers 1, 2, and 3 were inside the monitoring well for 122, 82, and 40 d, respectively. The retrieval dates were 11 Nov. 2000 (MLS-1; Fig. 2) , 27 Feb. 2001 (MLS-2), and 24 Dec. 2001 (MLS-3). As the dialysis cell methodology is based on dynamic equilibrium and not accumulation, measured VOC concentrations represent environmental concentrations during the several days preceding the retrieval of the MLS, and not time-averaged concentrations. The exact position of the water table in relation to all MLS profiles is evident by the EC of the water in each dialysis cell on retrieval. For example, in MLS-1, above the water table EC ≤ 0.125 mS/cm, and below the water table EC ≥ 1.34 mS/cm.



View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2. Rainfall before and during the study period. Arrows pointing downward and upward denote introduction and extraction of the multilayer sampler (MLS), respectively. The dashed lines denote the time period each MLS was inside the monitoring well.

 
Analytical Techniques
The VOCs were extracted from water samples by automated purge and trap and analyzed by gas chromatography and mass spectrometry according to EPA SW-846, Method 8260. As a result of extremely high TCE concentrations, samples had to be diluted as much as 5000-fold before analysis, significantly decreasing analytical sensitivity for other VOC components, and often resulting in excessively high minimum quantification levels (MQLs; as high as 2000 µg/L in many cases). Spike recoveries met method requirements. The EC was measured in the field with a portable EC meter with instrument reproducibility of ±1%. Dissolved oxygen (DO) was also measured in the field with a portable DO meter with a method reproducibility of about ± 0.5 mg/L.

Cl, NO3– and SO4= were measured within 48 h of sampling using a Lachat autoanalyzer (colormetric technique). Analytical precision was about 5 to 10% for the different components.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
The TCE concentrations in MLS-1 were very high, ranging from 260000 µg/L just below the water table to about 20000 µg/L at 11 m below the water table (Fig. 3a) . Just above the water table, TCE vapors that equilibrated with distilled water in the dialysis cells gave even greater concentrations in the cell water, 310000 µg/L, corresponding to 124000 µg/L air. The TCE concentrations decrease with increasing distance from the water table in both the unsaturated and saturated zones.



View larger version (31K):
[in this window]
[in a new window]
  		Fig. 3. Profiles of volatile organic compounds (VOCs) obtained by multilayer sampler 1 (MLS-1): (a) trichloroethylene (TCE); (b) cis-1,2-dichloroethylene (cis-1,2-DCE), and (c) 1,1-Dichloroethylene (1,1-DCE). The lower x axis denotes VOC concentration in water; the upper x axis denotes VOC concentration in air, calculated using Henry's constant. Closed square symbols denote concentrations in the saturated zone, open squares denote concentrations in the unsaturated zone; wt = water table.

 
Profiles for cis-1,2-DCE and 1,1-DCE are shown in Fig. 3b and 3c. In most cases, reported concentrations for these components were close to the MQLs or below. Despite the analytical difficulties, it is seen clearly that cis-1,2-DCE and 1,1-DCE are present both below and above the water table in significant concentrations. In groundwater, the concentrations of cis-1,2-DCE and 1,1-DCE are up to about 3100 and 2800 µg/L, respectively. Tetrachloroethylene was detected in two samples in the unsaturated zone and in one sample at the water table at concentrations of about 3000 µg/L (not shown), but most of the samples were below MQL, presumably because of the analytical difficulties.

The VOC profiles obtained for MLS-2, 106 d after MLS-1, are shown in Fig. 4 . It is seen that TCE concentrations in MLS-2 were still very high, up to 88000 µg/L near the water table, and 48000 µg/L air at 78 cm above the water table. Below the water table, a sharp decrease in aqueous TCE concentration from >88000 µg/L to 40000 µg/L was observed during an interval of little more than 1 m. The shape of the PCE profile above and below the water table was very similar to that of TCE (Fig. 4b), although concentrations of PCE were lower by about two orders of magnitude. In comparison, the trends in MLS-2 profiles of cis-1,2-DCE and 1,1-DCE (Fig. 4c, 4d) below the water table sharply contrast those of PCE and TCE (Fig. 4a, 4b). Both cis-1,2-DCE and 1,1-DCE are seen to increase in concentration from about 250 µg/L near the water table to >700 µg/L at a depth of 1.5 m below the water table. In the unsaturated zone, cis-1,2-DCE was detected in three of the six cells, and 1,1-DCE was detected in only one of the six cells (presumably because of inadequately high MQLs).



View larger version (32K):
[in this window]
[in a new window]
  		Fig. 4. Temporal variability in volatile organic compound (VOC) concentrations: (a) trichloroethylene (TCE); (b) tetrachloroethylene (PCE); (c) cis-1,2-dichloroethylene (cis-1,2-DCE); and (d) 1,1-dichloroethylene (1,1-DCE). The lower x axis denotes VOC concentrations in water; the upper x axis denotes VOC concentrations in air, calculated using Henry's constant. Closed symbols denote concentrations in the saturated zone; open symbols denote concentrations in the unsaturated zone; wt = water table.

 
In MLS-3, obtained >1 yr after MLS-1, it is seen that both the very high VOC concentrations and the concentration trends are maintained (Fig. 4). The TCE concentrations in the unsaturated zone range from 76000 µg/L air near the water table to 22000 µg/L air 4 m above the water table (Fig. 4a). Below the water table, TCE concentrations sharply decrease from 190000 to 47000 µg/L within the uppermost 56 cm below the water table. The PCE concentration trends mimic those of TCE, with differences in concentration of about two orders of magnitude (Fig. 4b). The PCE ranges from about 700 µg/L air at 4 m above the water table to 2000 µg/L air just above the water table. Below the water table, PCE decreases from 3500 to 500 µg/L water within the uppermost 56 cm. Neither cis-1,2-DCE nor 1,1-DCE were detected in the unsaturated zone (Fig. 4c,d), while below the water table, both compounds are present in concentrations close to their respective MQLs.

Despite low sensitivity (see Analytical Techniques), the parallel (PCE and TCE; cis-1,2-DCE and 1,1-DCE) and divergent (TCE and cis-1,2-DCE) trends can be seen clearly in Fig. 5 , where concentrations of the different components (both above and below the water table) are plotted against each other in x–y scatter plots.



View larger version (19K):
[in this window]
[in a new window]
  		Fig. 5. Relation between volatile organic compounds (VOCs) (below and above the water table) in multilayer sampler 2 (MLS-2): (a) TCE (trichloroethylene) vs. PCE (tetrachloroethylene); (b) 1,1-DCE (1,1-dichloroethylene) vs. cis-1,2-DCE (cis-1,2-dichloroethylene); (c) TCE vs. cis-1,2-DCE.

 
Several observations can be made concerning temporal variations in TCE concentrations and concentration trends measured during the course of 13 mo (Fig. 4): (i) The highest concentration detected of TCE is about 20% of the pure TCE solubility at 25°C in distilled water (1349 mg/L; Laor et al., 2003), and the highest gas phase concentration is about 30% of the concentration of a saturated TCE vapor phase (540000 µg/L air; Laor et al., 2003). For all three profiles, TCE concentrations near the water table are highest, and decrease both above and below the water table. Such concentrations and gradients are suggestive that there is a source of TCE at or near the water table (Rivett 1995; Pankow and Cherry, 1996). Alternatively, the gradients may reflect density driven flow of TCE vapors from a source located higher in the unsaturated zone (Falta et al., 1989; Conant et al., 1996). (ii) The TCE concentrations at the SUIR vary maximally by a factor of three between the three sampling events. In Fig. 2 it can be seen that there was no appreciable rainfall (aquifer recharge) before retrieval of MLS-1, with the highest TCE concentration (Fig. 4a), while before retrieval of MLS-2 (lowest TCE concentration) there was 516 mm of rainfall (data courtesy of the Israel Meteorological Service). Rainfall before retrieval of MLS-3 was intermediate in value (310 mm), as was TCE concentration. Although the recharge process has not been studied in detail at the SUIR of this study area, it is reasonable to assume that the variations in TCE profiles are related to the recharge mechanism and the variable quantities of water in the SUIR during drainage from the unsaturated to the saturated zone (Ronen et al., 2000). (iii) In the groundwater, it is generally seen that cis-1,2-DCE and 1,1-DCE concentrations increase with increasing depth below the water table (Fig. 4c, 4d), in contrast to concentration trends of TCE and PCE (Fig. 4a, 4b). From these trends it appears there is a net flux of cis-1,2-DCE and 1,1-DCE into the gas phase of the unsaturated zone. As cis-1,2-DCE and 1,1-DCE are known to be degradation products of PCE and TCE, the profiles may reflect natural microbially mediated transformation of PCE and TCE into less chlorinated degradation by-products. It should be noted, however, that we have no direct proof of microbial degradation of TCE or PCE in the study area. Both cis-1,2-DCE and 1,1-DCE are known to be formed by a reductive dehalogenation mechanism under anaerobic conditions (Bradley, 2000; Ferguson and Pietari, 2000; Bourg et al., 1992). Water quality parameters measured in MLS-1 demonstrate oxic conditions (dissolved oxygen between 2 to 3 mg/L, NO3– around 100 mg/L, and SO4= between 100–300 mg/L; Graber et al., 2002) and therefore are not in accordance with a reductive dehalogenation mechanism. Moreover, as it is known that cis-1,2-DCE is the dominant compound among the three DCE isomers (cis-1,2, trans-1,2 and 1,1-DCE) formed during in situ reductive dehalogenation of parent TCE or PCE (Vancheeswaran et al., 1999; Hunkeler et al., 1999), we would anticipate much higher cis-1,2-DCE than 1,1-DCE concentrations from this mechanism. Such differences were not detected in the field profiles.

Interpreting the source of cis-1,2-DCE and 1,1-DCE (either as parent compound or degradation by-product) is complicated in our study area, as the history of chemical use is incomplete, and as the study area may be affected not only by the major former industry at the site but also by other adjacent industries. Reports of chemical usage by the facility at the site do not include either of these dichlorinated ethylene compounds (cis-1,2-DCE and 1,1-DCE). Furthermore, VOC plumes in both the saturated and unsaturated zones from multiple sources can overlap each other and complicate the system significantly. It is conceivable that the two compounds were transported together to the sampling site from another location, as the aqueous solubility and n-octanol-water partition coefficients Kow of these two isomers are of the same order of magnitude (water solubility of 5087 mg/L and 2490, log Kow of 1.86 and 1.48, for 1,2-DCE and 1,1-DCE respectively; Schwarzenbach et al., 2002).

(iv) The profiles of VOCs at the SUIR are quite variable during the 13-mo study interval, and concentrations decrease and increase again with time (Fig. 4). It is difficult to correlate these transient conditions to changes of a VOC source since the industrial area has been dismantled in mid-1997. We suggest that these changes are related to the active nature of the SUIR.

A detailed examination of the SUIR was not conducted at the study site. However, from data obtained at the SUIR of the same aquifer in two regions located 10 km (Ronen et al., 1997) and 25 km (Ronen et al., 2000) north of the study area, and from the apparent response of TCE to rainfall reported here, we can surmise that the transient patterns of the VOC profiles (Fig. 4) are associated with the complex and transient water content properties of the SUIR. We suggest that these patterns may be the result of (i) different rates of gas–water exchange of VOCs, impacted by variable water contents (Ronen et al., 1997, 2000); (ii) VOC sorption to sedimentary organic matter and mineral phases (Graber and Mingelgrin, 1994; Yaron et al., 1998; Borisover and Graber, 2002); (iii) enhanced bacterial activity in microenvironments within the SUIR (Ronen et al., 1987; Affek et al., 1998); and (iv) advection and diffusion related transport mechanisms within the SUIR (Ronen et al., 1997; Silliman et al., 2002). The variability in the concentration of VOCs in the SUIR defines flux direction between the unsaturated and saturated zones. In MLS-1, for example, the concentration of TCE in the unsaturated zone just above the water table is higher than that detected below it, suggesting TCE influx into the saturated zone, while in MLS-3, TCE aqueous and gas phases are in equilibrium (Fig. 4a).

All of these mechanisms, and the almost stagnant conditions expected to prevail at the SUIR (Ronen et al., 1986), may lead to the creation of adjacent parcels of dissolved VOCs and gas phase VOCs, with sharp interfaces between them. The profiles obtained by the MLS, as presented here, are snapshots through these parcels, created in a very dynamic system.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
The vertical detail between samples obtained by the MLS (distance between samples = 6 cm for 150-mL cells) is unique compared with other field sampling methods (e.g., Pankow et al., 1984; Kreamer et al., 1988; Imbrigiotta et al., 1995; Smith et al., 1996; Conant et al., 1996; Hers et al., 2000; Jellali et al., 2003). The profiles presented here are distinctive in that they span both the saturated and unsaturated zones, through the SUIR, in a single borehole. The passive diffusion sampling property of the MLS allows obtaining unmixed vertical samples such that, for example, differences as great as 24000 µg TCE/L air can be measured in consecutive samples located only 12 cm apart in the unsaturated zone. In contrast, samples obtained by active pumping methods may represent mixed parcels of air that do not necessarily originate from the immediate vicinity of the probe due to permeability heterogeneities in the unsaturated zone. Since MLS samples are obtained by diffusion, the concentration of VOCs in cells located below the water table and in the unsaturated zone will change dynamically according to changes in the respective compartments of the aquifer. Notwithstanding the period of time the MLS is located in the borehole, the profile obtained depicts the VOC concentration in the immediate vicinity of the sampler during the previous several days. One negative aspect of this sampling technique is that we cannot determine, for example, if the changes observed between the first and second profiles (MLS-1 and MLS-2) reflect changes that developed in the aquifer 80 d or only several days before obtaining MLS-2. Decreasing the time span between profiles may reduce uncertainty.


   ACKNOWLEDGMENTS
 
Funding for this research was provided in main by the Israel Water Commission and in part by the Volcani Center, Agricultural Research Organization, Israel. Technical assistance by Leonid Kutsishin is gratefully acknowledged.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
GeoRef
Right arrow GeoRef Citation
Agricola
Right arrow Articles by Ronen, D.
Right arrow Articles by Laor, Y.
Related Collections
Right arrow Vadose Zone Processes and Chemical Transport
Right arrow Ground Water Quality
Right arrow Capillary Fringe Processes
Right arrow Organic Compounds

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Plant Registrations Soil Science Society of America Journal
Journal of Natural Resources
and Life Sciences Education		 Journal of
Environmental Quality