SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Monitoring of a well-defined septic system groundwater plume and groundwater discharging to two urban streams located in southern Ontario, Canada, provided evidence of natural attenuation of background low level (ng/L) perchlorate (ClO4) under denitrifying conditions in the field. The septic system site at Long Point contains ClO4 from a mix of waste water, atmospheric deposition, and periodic use of fireworks, while the nitrate plume indicates active denitrification. Plume nitrate (NO3-N) concentrations of up to 103 mg/L declined with depth and downgradient of the tile bed due to denitrification and anammox activity, and the plume was almost completely denitrified beyond 35 m from the tile bed. The ClO4 natural attenuation occurs at the site only when NO3-N concentrations are <0.3 mg/L, after which ClO4 concentrations decline abruptly from 187 ± 202 to 11 ± 15 ng/L. A similar pattern between NO3-N and ClO4 was found in groundwater discharging to the two urban streams. These findings suggest that natural attenuation (i.e., biodegradation) of ClO4 may be commonplace in denitrified aquifers with appropriate electron donors present, and thus, should be considered as a remediation option for ClO4 contaminated groundwater.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Trace levels of perchlorate (ClO4) are widespread in groundwater in some regions of the United States and Canada (Parker et al. 2008; Bickerton et al. in preparation). Perchlorate can interfere with human thyroid function, and as result the U.S. Environmental Protection Agency (U.S. EPA) has introduced a daily recommended maximum safe dose of 0.7 µg/kg of body weight, which translates to a drinking water guideline in the range of 6 to 24.5 µg/L, depending on factors such as body weight and the fraction of daily dose that is obtained from drinking water (Renner 2005; U.S. EPA 2005). Perchlorate is mobile and persistent in aerobic groundwater environments (Motzer 2001), and concentrations approaching and exceeding the above range are known to occur at some industrial sites related to the manufacture of explosives and rocket propellants (Motzer 2001; Hatzinger 2005; Sturchio et al. 2012). Elevated concentrations also occur in a number of areas not directly related to point source contamination, including areas impacted by agricultural fertilizers containing trace quantities of ClO4 (Motzer 2001; Böhlke et al. 2005), and areas where ClO4 may occur naturally in association with evaporate deposits (Orris et al. 2003) or from evapotranspirative enrichment of atmospheric deposition (Böhlke et al. 2005; Plummer et al. 2006; Rajagopalan et al. 2006; Munster et al. 2009; Jackson et al. 2010). Increased use of irrigation in semiarid regions may mobilize atmospheric ClO4 that has accumulated in the vadose zone over long periods (Rajagopalan et al. 2006).

In a 1997 survey of 428 water supply wells in California, 27% had ClO4 above 4 µg/L (Logan 2001), and some of these have been treated to reduce ClO4 concentrations (U.S. EPA 2005). In a survey of groundwater in Canada (Bickerton et al. in preparation) 9% of groundwater samples tested from nonmilitary areas had concentrations >1 µg/L, with a maximum of 45.3 µg/L. All nonmilitary groundwater concentrations greater than Health Canada's Drinking Water Guidance value of 6 µg/L were associated with fireworks use. Meanwhile, trace levels of ClO4 occurred widely. Thus, there is considerable interest in understanding the fate of ClO4 in groundwater, including the degree to which natural attenuation (instead of active treatment) might play a role in mitigating its relatively widespread occurrence in groundwater in some regions.

Perchlorate is readily degraded in waste water treatment systems by reduction to chloride (Cl) (Hatzinger 2005). Degradation is usually observed under conditions that also promote denitrification (Herman and Frankenberger 1999; Hunter 2002; Coates and Achenbach 2004; Hatzinger 2005), and a number of bacteria types that have been identified can metabolize both nitrate (NO3) and ClO4 (Coates et al. 1999; Herman and Frankenberger 1999; Logan 2001). Thus, ClO4 treatment systems operating at industrial sites often incorporate reactors where denitification is stimulated by addition of electron donors such as ethanol, methanol, lactate, and acetate (Hatzinger 2005). Similarly, successful in situ treatment has been achieved in subsurface bio-barriers and bioreactors by injecting or incorporating a variety of electron donor materials, such as methanol, molasses, edible oils and carbonaceous solids, all of which are capable of stimulating denitrification and promoting ClO4 degradation (Logan 2001; Hunter 2002; Hatzinger 2005; U.S. EPA 2005; Robertson et al. 2009).

The ubiquitous occurrence of ClO4 degrading bacteria in the environment has been inferred based on surficial soil and sediment samples (Coates et al. 1999). Meanwhile, microcosm studies (Tipton et al. 2003; Tan et al. 2004; Simon and Weber 2006; Wilkin et al. 2007; Lieberman et al. 2010), and stream bed profiling at sites near McGregor, Texas (Tan et al. 2005), have shown that ClO4 degradation can occur in unamended natural sediments. It has also been inferred that natural attenuation occurs in the anoxic bottom waters of California's Salton Sea because ClO4 is absent from the sea but is elevated (about 4 µg/L) in the inlet streams (Holdren et al. 2008). All of these studies targeted surface waters and their associated stream and lakebed sediments which were relatively rich in organic carbon (OC). There apparently are no conclusive examples of field scale natural attenuation of ClO4 in groundwater systems (Lieberman et al. 2010).

This science gap blocks any opportunity to consider unamended groundwater environments (i.e., not augmented with additional electron donor supplies) for mitigation of ClO4 contamination by natural attenuation. In this study, our objective was to assess ClO4 behavior along a groundwater flowpath (Long Point site) and groundwater discharging into two urban streams (Tuck and Shoreacres Creeks), where it could be clearly shown that denitrification was occurring. The results indicate that ClO4 natural attenuation may be commonplace in denitrified groundwater and these sites provide new insight into the potential for natural attenuation of ClO4 in groundwater environments.

Long Point Site

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

The Long Point campground (42°35′N, 80°25′E), located on the north shore of Lake Erie, has 256 overnight campsites and is open seasonally from mid-May until mid-October. Sewage from a single washroom facility is treated on-site, in a conventional septic system. The site was selected because it has a large septic system plume where elevated NO3-N values of up to 103 mg/L were present in the shallow groundwater, but then concentrations declined to <0.1 mg/L at locations farther along the plume, primarily as a result of denitrification (Aravena and Robertson 1998; Robertson et al. 2012; Robertson et al. 2013). Tile Bed 2 has a groundwater plume that flows southward toward the Lake Erie shoreline (Figure S1, Supporting Information) within an approximately 5-m-thick unconfined calcareous sand aquifer that is underlain by clayey silt (Figure 1). Previously, a multilevel groundwater monitoring network with 90 monitoring points aligned along the plume centreline was used to establish that the waste water constituents, Na+ and the artificial sweetener acesulfame, were distinctly elevated in the plume throughout its entire 200 m mapped length (Figure S2), and that dispersive dilution with background groundwater was relatively minor (Van Stempvoort et al. 2011b; Robertson et al. 2013). Below the tile bed and up to about 17 m downgradient, the plume occupies almost the entire thickness of the aquifer and based on seasonal breakthrough of elevated electrical conductivity (EC) values and previous NaBr tracer tests, this zone represents about 1 year of wastewater loading (Robertson et al. 2012). Farther downgradient, the plume occupies only the bottom 1 to 2 m of the aquifer and represents waste water that is up to about 15 years old based on tritium-helium age dating (Robertson et al. 2013). Except for within about 0.5 m of the water table, groundwater throughout the site, both in and outside of the wastewater plume, is sub-oxic (dissolved oxygen [DO] <1 mg/L, Robertson et al. 2012).

image

Figure 1. Long Point septic system groundwater plume, centreline transect (October 27, 2010); (a) NO3-N and (b) ClO4 distribution. Fine dashed line is the core zone of the wastewater plume identified by Na+ >10 mg/L and acesulfame (artificial sweetener) >8 µg/L (Figure S2). Note: all groundwater is sub-oxic (DO < 1 mg/L) except within 0.5 m of the water table. Dots represent monitoring well screened intervals (5 cm length). Numerical data values are presented in Figure S3.

Download figure to PowerPoint

Stream Sampling Sites

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Perchlorate and NO3 were also analyzed in a larger study assessing the influence of groundwater on urban stream water quality in Canada (Bickerton et al. in preparation; Van Stempvoort et al. 2011a). In this study, detailed transects of groundwater chemistry were obtained using drive-point mini-profilers (Roy and Bickerton 2012), along 100 to 1000 m long reaches of eight gaining, first-order, urban streams in Canada, including Tuck and Shoreacres Creeks. These two streams are located close to the Long Point site, in the City of Burlington, ON (43°20′N, 80°10′E), and there are no known point sources of industrial or sewage ClO4 contamination in these watersheds. Thus, these stream transect results provide an opportunity for comparison with the ClO4-NO3 relationship observed at Long Point. Tuck Creek is 2 to 5 m in width and is located in a watershed that has largely commercial land use, whereas Shoreacres Creek is 2 to 4 m in width and has land use that is both commercial and residential. The surrounding area is a till plain where the water table is inferred to be on the order of about 1 to 10 m in depth. Seeps present along these stream reaches during the time of sampling contained anthropogenic contaminants (e.g., Cl from road salt, etc.), indicating that the shallow groundwater originated from nearby terrestrial sources (Roy and Bickerton 2012), and that the streams were under groundwater discharge conditions. Riparian zones, approximately 5 to 30 m in width and naturally vegetated with grasses and trees, were generally present on both sides of the streams. Although groundwater samples from both stream sites were predominately aerobic (DO > 1 mg/L), a number of sub-oxic zones (DO < 1 mg/L) were also sampled. Additional site information is available from the previous studies (Van Stempvoort et al. 2011a; Roy and Bickerton 2012).

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Sample Collection

In this study, approximately 200 samples from the Long Point site were analyzed for ClO4, NO3 and other water quality parameters (EC, DO, pH, Cl, Na+, artificial sweeteners, etc.). This included initial screening of the proximal plume zone during September 2008 to October 2009 (n = 65), complete sampling of the full 200 m length of the plume centreline in October 2010 (n = 95), and time series sampling at the water table below the tile bed (well 122) during campground start up, April to July 2011 (n = 31). In addition, samples of the septic tank effluent were collected during 2008 to 2011 (n = 13), and a single sample of campground tap water was collected in 2010.

Twenty one groundwater samples were collected from an approximately 200-m-long reach of Tuck Creek and 19 groundwater samples were collected from a similar length of Shoreacres Creek, during June 2009. Groundwater was collected with a drive-point mini-profiler connected to a peristaltic pump, from a depth of between 0.25 and 1.0 m below the streambed, at a spacing of about 10 to 15 m along the streams (Roy and Bickerton 2010). Sampling commenced once field parameters (DO, pH, and EC), measured with portable meters, stabilized.

Perchlorate was analyzed at Canada Centre for Inland Waters, Burlington, ON, using the same ion chromatography technique (Dionex 2500 system, Dionex, Sunnyvale, California) coupled with tandem mass spectrometry (AB Sciex QTRAP 5500 triple-quadupole, AB Sciex, Concord, Ontario, Canada) described by Van Stempvoort et al. (2011a). This technique provided a detection limit of 2 ng/L for ClO4. Sample collection procedures and laboratory analytical methods are further described in Appendix S1.

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Long Point Nitrate

Figures 1a and S3a show the presence of highly elevated NO3-N of up to 103 mg/L in the proximal plume zone in October 2010, which is the result of oxidation of the wastewater NH4+ in the vadose zone below the tile bed. However, at the base of the aquifer and beyond 35 m from the tile bed, NO3-N is attenuated to ≤0.1 mg/L at most locations in the plume core (Figures 1a and S3a). This is consistent with previous studies at this site which showed that NO3 is attenuated by denitrification that utilizes trace quantities of OC and reduced sulfur (S) from pyrite present in the aquifer sediments, as electron donors (Aravena and Robertson 1998). Dissolved OC present in the plume (1 to 7 mg/L, Aravena and Robertson 1998) was insufficient to account for the amount of nitrate attenuation observed. In addition to denitrification, some NO3 is attenuated with NH4+ at this site, by anaerobic ammonium oxidation (anammox, Robertson et al. 2012); however, denitrification is likely dominant over anammox in most plume zones, particularly beyond 17 m from the tile bed where NH4+ is absent but NO3 attenuation continues (Robertson et al. 2013).

In this study, NO3-N concentrations in many of the background wells, both upgradient of the tile bed and overlying the plume, were also low (≤0.1 mg/L, Figure 1a). These wells were also sub-oxic (DO < 1 mg/L at depths greater than 0.5 m below the water table) and thus were likely similarly affected by denitrification, utilizing the same electron donors (aquifer OC and pyrite) as the plume water. Nitrate concentrations measured in the proximal zone in 2008 (Figure S4) were similar to the 2010 values as shown in Figures 1a and S3a.

Long Point Perchlorate

The septic tank effluent had relatively low ClO4 concentrations (2 to 76 ng/L, n = 13) which were generally lower than the value measured in the single sample of tap water (64 ng/L). Figures 1b and S3b show generally higher ClO4 of 13 to 802 ng/L present in the shallow proximal plume zone. Elevated values of up to 392 ng/L were also found in some of the groundwater that was sampled from above the septic plume in the area immediately downgradient of the tile bed (Figures 1b and S3b). Additionally, elevated ClO4 of 309 ± 172 ng/L was found in seven time-series samples taken at the water table below the tile bed (wells 120, 121, and 122) in April and May 2011, just prior to opening of the campground, and these samples also had very low Cl (5.6 ± 2.4 mg/L). The low Cl values indicated that these samples originated not from the wastewater, but from natural precipitation recharge that had occurred during the November to May nonuse period. These findings indicate that the high ClO4 concentrations in the plume are likely augmented by elevated values present in natural precipitation recharge, rather than being entirely derived from the high Cl wastewater. Perchlorate concentrations measured in the proximal zone of the plume in 2008 (Figure S4) were similar to the 2010 values shown in Figure 1b. These elevated concentrations point to an additional source, or sources, of ClO4 at the study site.

In a recent nation-wide survey of atmospheric deposition (wet only) at 26 monitoring stations in the United States (Rajagopalan et al. 2009), a relatively low mean ClO4 value of 14 ng/L was measured. However, in another study, which measured both wet and dry deposition at six sites on Long Island, New York (Munster et al. 2009), a much higher mean ClO4 value of 210 ng/L was obtained, with a distinct peak associated with periods of fireworks use (up to 2780 ng/L). These concentrations are similar to the concentrations observed in the shallow groundwater at Long Point and we note that fireworks use also occurs periodically at Long Point. In fact, spent fireworks casings have been found lying directly on the tile bed and this grassed area could be a focused area of more intensive fireworks use. Similar high values could be present near the water table in other areas, but the monitoring network generally does not capture this shallow groundwater elsewhere at the site, except in the two upgradient wells where denitrification appears complete even in the shallowest points. Thus, we conclude that atmospheric deposition, with some contribution from fireworks residuals, are likely the main sources of the ClO4 found in shallow groundwater at Long Point.

Despite the persistent occurrence of elevated ClO4 in the groundwater near the septic tile bed, both in and overlying the plume, ClO4 was virtually absent (<30 ng/L) from the distal portion of the plume (Figures 1b and S3). To assist in understanding ClO4 behavior within the plume, normalized concentrations of the two wastewater indicators, Na+ and the artificial sweetener, acesulfame, which are relatively mobile and conservative within the plume (Van Stempvoort et al. 2011b; Robertson et al. 2013), were compared to normalized NO3-N and ClO4 values (Figure 2). For this comparison, initial values (C0) were calculated as the mean values in the four proximal monitoring wells located immediately below the tile bed (wells 120, 121, 122, and 123; total of 20 monitoring points, Figure 1). Sodium and acesulfame declined along the plume in a relatively uniform manner as a result of hydrodynamic dispersion and possibly other factors, but both still remain at approximately 25 to 50% of the initial values 200 m downgradient. In contrast, both NO3-N and ClO4 declined abruptly beyond 35 m from the tile bed and, except for a single monitoring point from well 7 (50 m distance) that is elevated in both NO3-N and ClO4, NO3-N declined to  <1% of the initial value, and ClO4 declined to <10% of the initial value. This suggests that ClO4 and NO3 experienced similar fates within the waste water plume, despite their different sources.

image

Figure 2. Normalized groundwater concentrations along the plume core zone and in the upgradient wells, October 27, 2010: (a) Na+, (b) acesulfame, (c) NO3-N, and (d) ClO4. All values downgradient of the tile bed are from the plume core zone (Na+ >10 mg/L and/or acesulfame >8 µg/L). Initial values (C0) are the mean October 2010 values of the monitoring points located directly below the tile bed (wells 120, 121, 122, 123, n = 20). Values greater than two times the initial value are plotted as C/C0 = 2. Note, 2008 ClO4 values are shown twice (Figure 2d); original values sampled in 2008 (Figure S3) and same values projected 46 m downgradient (and adjusted −14% for dispersive dilution) to reflect 2 years of groundwater flow during the period 2008 to 2010.

Download figure to PowerPoint

Normalized concentrations of the 2008 ClO4 samples, which were all collected within 17 m of the tile bed (Figure S3), are also shown in Figure 2d. Perchlorate concentrations measured in subtile wells 120, 121, 122, and 123 in 2008 (117 ± 69 ng/L, n = 22) remained similar to values measured in the same wells in 2010 (119 ± 113 ng/L, n = 20). The 2008 values are also “projected” 46 m farther downgradient in Figure 2d, based on the time interval between September 2008 and October 2010 sampling events (2.1 years) and the horizontal groundwater velocity in the tile bed area indicated from a previous tracer test (22 m/year, Robertson et al. 2012). This projection assumes that ClO4 is highly mobile, because it is normally considered unaffected by sorption processes in sandy soils (Motzer 2001). Also, to account for the likelihood that the 2008 ClO4 values have been diluted by hydrodynamic dispersion during this 2-year period, the projected values in Figure 2d have been decreased by14% from their original concentrations, based on similar attenuation observed for Na+ (decrease of 0.31% per meter of flowpath distance, Figure 2a). The projected 2008 ClO4 values indicate a plume 40 to 70 m downgradient; however, this zone is largely devoid of ClO4 (and NO3) except for the single high value from well 7 (Figure 1b). The loss of ClO4 is attributed to natural attenuation processes. Thus, regardless of the source of the elevated ClO4 concentrations, similar elevated values were measured in the proximal plume zone during two sampling snapshots 2 years apart, and the low concentrations of ClO4 observed in the distal, nitrate-depleted portion of the septic plume are compelling evidence that natural degradation of ClO4 is occurring in groundwater at this site.

Although Figure 1 points to a strong relationship between NO3 and ClO4, the scatter plot of NO3-N vs. ClO4 concentrations, in both the plume water and the background groundwater (Figure 3), shows little evidence of declining ClO4 concentrations as NO3-N concentrations decrease, until nitrate is almost entirely removed. Considering the data shown in Figure 3 (both plume and background samples), ClO4 concentrations in groundwater with NO3-N of 0.3 to 1 mg/L (72 ± 124 ng/L, n = 9) and 1 to 10 mg/L (196 ± 138 ng/L, n = 11) are not significantly different (p > 0.1) than groundwater with  >10 mg/L NO3-N (187 ± 202 ng/L, n = 107) even through isotopic evidence shows that most plume groundwater with NO3-N < 10 to 20 mg/L has been highly affected by denitrification at this site (Aravena and Robertson 1998; Li 2010). Only when NO3-N is below 0.3 mg/L are ClO4 concentrations significantly lower (11 ± 15 ng/L, n = 56, p < 0.01). This behavior indicates that ClO4 is generally not co-metabolized with NO3 during denitrification at this site and that the bacteria consortium present only degrades ClO4 after the preferred electron acceptor, NO3, has been mostly consumed.

image

Figure 3. Scatter plot of ClO4 vs. NO3-N concentrations at the Long Point site during 2008 to 2011, showing both plume values (Na+ >10 mg/L and/or acesulfame >8 µg/L) and background values, upgradient of the tile bed and overlying the plume. ClO4 values of 2 ng/L are detection limit values.

Download figure to PowerPoint

Stream Sites: Nitrate and Perchlorate

Figure 4a shows the relationship of ClO4, NO3, and DO concentrations in the 21 groundwater samples collected from Tuck Creek. All of the aerobic samples (DO > 1 mg/L) had consistently elevated ClO4 in the 100 to 300 ng/L range, and NO3 was present at 2 to 5 mg/L (0.5 to 1 mg/L as N). Most of the aerobic samples from the Shoreacres Creek site had similarly elevated ClO4 of 100 to 300 ng/L (Figure 4b). These ClO4 concentrations were similar to the atmospheric deposition values measured on Long Island, New York (mean of 210 ng/L, Munster et al. 2009), which is also an urbanized area. This evidence suggests that the relatively uniform ClO4 concentrations observed in the aerobic samples from both Tuck and Shoreacres Creek likely represent background values from atmospheric deposition.

image

Figure 4. Scatter plots of ClO4 vs. NO3 and DO concentrations in groundwater discharging to two gaining, first-order urban streams in southern Ontario, Canada: (a) Tuck Creek and (b) Shoreacres Creek. Samples collected using a drive-point sampler inserted from 0.25 to1.0 m depth, below, or alongside, the stream bed. Detection limit for DO is approximately 0.3 mg/L. Nitrate concentrations are presented as NO3 rather than as NO3-N for convenient axis scaling with DO.

Download figure to PowerPoint

In contrast to the aerobic samples which had consistently high ClO4, most sub-oxic samples (DO < 0.5 mg/L) from both Tuck Creek and Shoreacres Creek had much lower ClO4 concentrations (<20 ng/L), and were accompanied by NO3 values that were below detection (<0.1 mg/L, Figure 4). This suggests that these sub-oxic samples were likely affected by denitrification, which was possibility enhanced at some locations by increased OC availability in the stream riparian zones. Thus, the ClO4-NO3 relationship observed at both of these stream sites was similar to the relationship observed at Long Point.

Implications

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Detailed spatial monitoring at Long Point suggests that NO3-N inhibits ClO4 reduction at concentrations above about 0.3 mg/L at this site. This is somewhat below the inhibition level indicated in a previous in situ, bioreactor study (1 to 2 mg/L, Robertson et al. 2009), and other studies have indicated that ClO4 and NO3 can be degraded simultaneously (Herman and Frankenberger 1999; Hunter 2002; Logan and LaPoint 2002; Xueyuan et al. 2007; Choi and Silverstein 2008). However, most previous studies used an added electron donor, usually labile OC, H2 gas, or Fe0, to stimulate denitrification and promote ClO4 degradation. Although several microcosm studies have demonstrated ClO4 and NO3 degradation together in unamended sediments, these were mostly organic-rich creek bed sediments (4 to 12% OC, Tan et al. 2004), lake bottom sediments (0.18 to 3.5wt% OC, Wilkin et al. 2007), and loam (0.96% OC, Tipton et al. 2003), and consequently the bacterial populations may have been less discriminatory in their choice of electron acceptor at these sites. The microcosm studies of Tan et al. (2004) did, however, indicate that ClO4 degradation was delayed until NO3-N was below 1 to 2 mg/L. In contrast, at Long Point, the primary electron donor supply is natural OC and biogenic pyrite present in the aquifer sands, but only at relatively low concentrations (0.02wt% S and 0.15wt% OC, Aravena and Robertson 1998). This electron donor limitation may cause the bacterial population to be more selective in its choice of electron acceptor, preferring NO3 over ClO4. Microcosm studies (Herman and Frankenberger 1999) have also indicated that NO3 inhibition can be more pronounced when ClO4 concentrations are relatively low, as they are at Long Point. Microbiological studies (Xu et al. 2004) and recent biofilm modeling studies (Tang et al. 2012) have also suggested that very low NO3-N concentrations (3 to 15 µg/L) may actually stimulate the degradation of ClO4 because of the beneficial effect of perchlorate reducing bacteria using both electron acceptors, whereas at higher NO3-N concentrations, increased competition effects favor degradation of NO3. The spatial relationship of NO3 and ClO4 at Long Point where ClO4 declines abruptly very close to the denitrification front (Figures 1 and 2), is consistent with these microbiological and modeling results.

Nonetheless, the abrupt and complete attenuation of ClO4 in denitrified zones of the Long Point plume shows that natural attenuation has the potential to play an important role when considering remediation options for ClO4 contaminated groundwater. The ClO4- NO3 -DO relationships observed at both of the stream sampling sites, where ClO4 concentrations were much lower in samples that were likely denitrified, suggests that ClO4 degradation could be commonplace in groundwater environments if the electron donors present naturally in the system are sufficient for both denitrification and either subsequent or concurrent reduction of ClO4. Naturally denitrified groundwater is widespread in many groundwater flow systems worldwide (Korom 1992; Rivett et al. 2008), and the principle electron donors at most of these sites are OC and reduced sulfur (pyrite) occurring naturally in the sediments at trace levels, the same as at Long Point. At contaminated industrial sites, careful examination of the redox zones that naturally occur within plumes, along their flow paths in groundwater, might allow costly active treatment systems to be avoided or minimized in some cases. Further studies should be undertaken to confirm that the higher ClO4 concentrations present at industrial sites also degrade in naturally denitrified groundwater and to assess if the presence of NO3 inhibits ClO4 degradation to the same degree. It is possible that further research will demonstrate a relatively low risk for ClO4 contamination of water supply wells in denitrified groundwater zones.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information

Pam Collins provided assistance with laboratory analyses and Robin Barnes, Melissa Hollingham, John Voralek provided field assistance at the stream sites. Access to the Long Point site was kindly provided by Rhonda Card. In-depth comments kindly provided by three anonymous reviewers contributed greatly to the manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information
  • Aravena, R., and W.D. Robertson. 1998. Use of multiple isotope tracers to evaluate denitrification in ground water: Study of nitrate from a large-flux septic system plume. Ground Water 36: 975982.
  • Böhlke, J.K., N.C. Sturchio, B. Gu, J. Horita, G.M. Brown, W.A. Jackson, J. Batista, and P.B. Hatzinger. 2005. Perchlorate isotope forensics. Analytical Chemistry 77: 78387842.
  • Choi, H., and J. Silverstein. 2008. Inhibition of perchlorate reduction by nitrate in a fixed biofilm reactor. Journal of Hazardous Materials 159: 440445.
  • Coates, J.D., and L.A. Achenbach. 2004. Microbial perchlorate reduction: rocket-fuelled metabolism. Nature Reviews in Microbiology 2: 569580.
  • Coates, J.D., U. Michaelidou, R.A. Bruce, S.M. O'Connor, J.N. Crespi, and L.A. Achenbach. 1999. Ubiquity and diversity of dissimilatory (per)chlorate-reducing bacteria. Applied Environmental Microbiology 65: 52345241.
  • Hatzinger, P.B. 2005. Perchlorate biodegradation for water treatment. Environmental Science and Technology 39: 239A247A.
  • Herman, D.C., and W.T. Frankenberger Jr. 1999. Bacterial reduction of perchlorate and nitrate in water. Journal of Environmental Quality 28: 10181024.
  • Holdren, G.C., K. Kelly, and P. Weghorst. 2008. Evaluation of potential impacts of perchlorate in the Colorado River on the Salton Sea, California. Hydrobiologia 604: 173179.
  • Hunter, W.J. 2002. Bioremediation of chlorate and perchlorate contaminated water using permeable barriers containing vegetable oil. Current Microbiology 45: 287292.
  • Jackson, W.A., J.K. Böhlke, B. Gu, P.B. Hatzinger, and N.C. Sturchio. 2010. Isotopic composition and origin of indigenous natural perchlorate and co-occurring nitrate in the southwestern United States. Environmental Science and Technology 44: 48694876.
  • Li, L. 2010. Use of N2O and its isotopic composition to investigate nitrogen processes in groundwater. M.Sc. thesis, University of Waterloo, Waterloo, ON, Canada.
  • Lieberman, M.T., S.L. Knox, J. Keener, and R.C. Borden. 2010. Evaluation of potential for monitored natural attenuation of perchlorate in groundwater. Final report, ESTCP Project ER-200428, U.S. Department of Defense.
  • Logan, B.E. 2001. Assessing the outlook for perchlorate and remediation. Environmental Science and Technology 35: 482487.
  • Logan, B.E., and D. LaPoint. 2002. Treatment of perchlorate- and nitrate-contaminated groundwater in a autotrophic, gas phase, packed-bed bioreactor. Water Research 36: 36473653.
  • Korom, S.F. 1992. Natural denitrification in the saturated zone: A review. Water Resources Research 28: 16571668.
  • Motzer, W.E. 2001. Perchlorate: Problems, detection and solutions. Environmental Forensics 2: 301311.
  • Munster, J., G.N. Hanson, W.A. Jackson, and S. Rajagopalan. 2009. The fallout from fireworks: Perchlorate in total deposition. Water Air and Soil Pollution 198: 149153.
  • Orris, G.J., G.J. Harvey, D.T. Tsui, and J.E. Elridge. 2003. Preliminary analyses for perchlorate in selected natural materials and their derivative products. Open-File Report 03-314. Reston, Virginia: U.S. Geological Survey.
  • Parker, D.P., A.L. Seyfferth, and B.K. Reese. 2008. Perchlorate in groundwater: A synoptic survey of “pristine” sites in the coterminous United States. Environmental Science and Technology 42: 14651471.
  • Plummer, L.N., J.K. Bohlke, and M.W. Doughten. 2006. Perchlorate in Pleistocene and Holocene groundwater in North-Central New Mexico. Environmental Science and Technology 40: 17571763.
  • Rajagopalan, S., T.A. Anderson, S. Cox, G. Harvey, Q. Cheng, and W.A. Jackson. 2009. Perchlorate in wet deposition across North America. Environmental Science and Technology 43: 616622.
  • Rajagopalan, S., T.A. Anderson, L. Fahlquist, K.A. Rainwater, M. Ridley, and W.A. Jackson. 2006. Widespread presence of naturally occurring perchlorate in high plains of Texas and New Mexico. Environmental Science and Technology 40: 31563162.
  • Renner, R. 2005. Perchlorate report doesn't dispel controversy. Environmental Science and Technology 39: 96A97A.
  • Rivett, M.O., S.R. Buss, P. Morgan, J.W.N. Smith, and C.D. Bemment. 2008. Nitrate attenuation in groundwater: A review of biogeochemical processes. Water Research 42: 42154232.
  • Robertson, W.D., D.R. Van Stempvoort, D.K. Solomon, J. Homewood, S.J. Brown, J. Spoelstra, and S.J. Schiff. 2013. Persistence of artificial sweeteners in a 15-year-old septic system plume. Journal of Hydrology 477, no. 1: 4354.
  • Robertson, W.D., T. Moore, J. Spoelstra, L. Li, R. Elgood, I.D. Clark, S.L. Schiff, R. Aravena, and J.D. Neufeld. 2012. Natural attenuation of septic system nitrogen by anammox. Ground Water 50: 541553.
  • Robertson, W.D., C.J. Ptacek, and S.J. Brown. 2009. Rates of nitrate and perchlorate removal in a 5-year-old wood particle reactor treating agricultural drainage. Ground Water Monitoring and Remediation 29: 8794.
  • Roy, J.W., and G. Bickerton. 2012. Toxic groundwater contaminants: An overlooked contributor to urban stream syndrome? Environmental Science and Technology 46: 729736.
  • Roy, J.W., and G. Bickerton. 2010. Proactive screening approach for detecting groundwater contaminants along urban streams at the reach-scale. Environmental Science and Technology 44: 60886094.
  • Simon, R., and E.J. Weber. 2006. Reduction of perchlorate in river sediment. Environmental Toxicology and Chemistry 25: 899903.
  • Sturchio, N.C., J.R. Hoaglund III, R.J. Marroquin, A.D. Beloso Jr., L.J. Heraty, S.E. Bortz, and T.L. Patterson. 2012. Isotopic mapping of groundwater perchlorate plumes. Ground Water 50: 94102.
  • Tan, K., T.A. Anderson, and W.A. Jackson. 2005. Temporal and spatial variation of perchlorate in streambed sediments: Results from in-situ dialysis samplers. Environmental Pollution 136: 283291.
  • Tan, K., T.A. Anderson, and W.A. Jackson. 2004. Degradation kinetics of perchlorate in sediments and soils. Water Air and Soil Pollution 151: 245259.
  • Tang, Y., H. Zhao, A.K. Marcus, R. Krajmalnik-Brown, and B.E. Rittmann. 2012. A steady-state biofilm model for simultaneous reduction of nitrate and perchlorate, Part 2: Parameter optimization and results and discussion. Environmental Science and Technology. DOI:10.1021/203130r46:1608-1615.
  • Tipton, D.K., D.E. Rolston, and K.M. Scow. 2003. Transport and degradation of perchlorate in soils. Journal of Environmental Quality 32: 4046.
  • U.S. EPA (U.S. Environmental Protection Agency). 2005. Perchlorate treatment technology update. Technical Report, EPA 542-R-05-015, May, 2005.
  • Van Stempvoort, D.R., S.J. Brown, J.W. Roy, and G. Bickerton. 2011a. Artificial sweeteners as potential tracers in groundwater in urban environments. Journal of Hydrology 401: 126133.
  • Van Stempvoort, D.R., W.D. Robertson, and S.J. Brown. 2011b. Artificial sweeteners in a large septic plume. Ground Water Monitoring and Remediation 31: 95102.
  • Wilkin, R.T., D.D. Fine, and N.G. Burnett. 2007. Perchlorate behaviour in a municipal lake following fireworks displays. Environmental Science and Technology 41: 39663971.
  • Xu, J., J.J. Trimble, L. Steinberg, and B.E. Logan. 2004. Chlorate and nitrate reduction pathways are separately induced in the perchlorate-respiring bacterium Deschlorosoma sp. KJ and the chlorate-respiring bacterium Pseudomonas sp. PDA. Water Research 38: 673680.
  • Xueyuan, Y., C. Amrhein, M.A. Deshusses, and M.R. Matsumoto. 2007. Perchlorate reduction by autotrophic bacteria attached to zerovalent iron in a flow-through reactor. Environmental Science and Technology 41: 990997.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Long Point Site
  5. Stream Sampling Sites
  6. Methods
  7. Results and Discussion
  8. Implications
  9. Acknowledgments
  10. References
  11. Supporting Information
FilenameFormatSizeDescription
gwat12031-sup-0001-Appendixs1.docWord document31KAppendix S1. Sample collection and analytical methods.
gwat12031-sup-0002-figures1.docWord document584KFigure S1. Long Point septic system site showing location of the tile bed, multilevel monitoring wells, and contours of maximum electrical conductivity (EC μS/cm) measured in the monitoring wells in October 2010 (from Robertson et al. 2013).
gwat12031-sup-0003-figures2.docWord document1087KFigure S2. Distribution of indicator parameters along the plume centreline, October 2010: (a) Cl, (b) Na+, (c) electrical conductivity (EC), and (d) artificial sweetener acesulfame. Dashed line indicates plume core zone indicated by Na+ >10 mg/L. Note changes in distance scale (adapted from Robertson et al. 2013).
gwat12031-sup-0004-figures3.docWord document649KFigure S3. Long Point septic system groundwater plume, centreline transect (October 27, 2010); (a) NO3-N and (b) ClO4 distribution. Fine dashed line is the core zone of the wastewater plume identified by Na+ >10 mg/L and acesulfame (artificial sweetener) >8 µg/L (Figure S2). NO3-N values of 0.1 mg/L and ClO4 values of 2 ng/L are detection limit values.
gwat12031-sup-0005-figures4.docWord document355KFigure S4. Plume centerline, September 11, 2008: (a) NO3-N and (b) ClO4 distribution. Dashed line indicates plume core zone indicated by Na+ >10 mg/L.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.