Has the Burden and Distribution of PCBs and PBDEs Changed in

ARTICLE pubs.acs.org/est

Has the Burden and Distribution of PCBs and PBDEs Changed in European Background Soils between 1998 and 2008? Implications for Sources and Processes Jasmin K. Schuster,†,* Rosalinda Gioia,† Claudia Moeckel,†,‡ Tripti Agarwal,§ Thomas D. Bucheli,§ Knut Breivik,||,^ Eiliv Steinnes,3 and Kevin C. Jones†,* †

Centre for Chemicals Management, Lancaster Environment Centre, Lancaster University, Lancaster, LA1 4YQ, United Kingdom, Centre For Ecology & Hydrology, Lancaster Environment Centre, Lancaster, LA1 4AP, United Kingdom § Agroscope Reckenholz-T€anikon Research Station ART, Reckenholzstrasse 191, 8046 Z€urich, Switzerland Norwegian Institute for Air Research (NILU), P.O. Box 100, NO-2027 Kjeller, Norway ^ Department of Chemistry, University of Oslo, P.O. Box 1033, NO-0315 Oslo, Norway 3 Department of Chemistry, Norwegian University of Science and Technology, NO-7491, Trondheim, Norway

)

‡

bS Supporting Information ABSTRACT: Background soils were collected from 70 locations on a latitudinal transect in the United Kingdom and Norway in 2008, ten years after they had first been sampled in 1998. The soils were analyzed for polychlorinated biphenyls (PCBs), polybrominated diphenyl ethers (PBDEs), and organochlorine pesticides (OCs), to see whether there had been any change in the loadings or distributions of these persistent organic pollutants (POPs). The same transect has also been used to sample air between the mid-1990s and the present, so the air and soil spatial and temporal trends provide information on air-soil transfers, source-receptor relationships, long-range atmospheric transport (LRAT), and recycling phenomena. Comparisons of the 2008 and 1998 data sets show a general decline for PBDEs in surface soil, and a smaller averaged net decline of PCBs. Changes between the years were observed for total POP concentrations in soil and also for correlations with site and sample characteristics assumed to affect those concentrations. POP concentrations were correlated to distance and strength of possible sources, a relationship that became weaker in the 2008 data. Fractionation, a commonly discussed process for the global cycling of POPs was also lost in the 2008 data. As in 1998, soil organic matter content continues to have a strong influence on the loadings of POPs in surface soils, but changes in the PCB loads were noted. These factors indicate an approach to air
surface soil equilibrium and a lessening of the influence of primary sources on POP concentrations in soil between 1998 and 2008.

’ INTRODUCTION Background soils play an important role in the global fate and distribution of POPs.1,2 They can be significant environmental reservoirs, sinks or sources for these chemicals,3 receiving inputs of POPs via atmospheric deposition4 and potentially exchanging them with the atmosphere.5 The distribution of POPs in background surface soils is a complex function of proximity to source regions, the LRAT potential of the POP in question, soil properties, climatic conditions, land use/cover, and processes of air-surface exchange. As the emissions and atmospheric loadings of POPs change over time, the role of soils as sinks or secondary sources back to atmosphere can potentially change over time,5 while their storage capacity, within-soil transport, degradation, and processing of POPs is strongly influenced by the soil organic matter (SOM) content and environmental conditions.1,2,6
8 r 2011 American Chemical Society

We have previously reported on the distribution of POPs in European background soils on a latitudinal transect from southern United Kingdom to Northern Norway8
11 and on the atmospheric distributions of POPs in space and time on this same transect.12,13 The data can be interpreted in terms of sourcereceptor, LRAT, and recycling phenomena and provide evidence of the following: ongoing primary source controls on the atmospheric burdens of many “legacy POPs” (e.g., PCBs; PBDEs),12,13 fractionation in the mixture of POPs in air and soils with distance from sources,1,2,4,14
16 and a strong association/correlation with SOM.1,2,7,8 However, key questions remain over the contribution Received: March 22, 2011 Accepted: July 16, 2011 Revised: June 23, 2011 Published: August 05, 2011 7291

dx.doi.org/10.1021/es200961p | Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology of re-emissions from soils to the current atmospheric burden of POPs, how the burden and distribution in soils is changing over time (whether a recyclable pool of POPs approaches air-SOM equilibrium) and the role of biodegradation in soils in removing POPs from the global pool.6,17,18 The aim of this project is to collect further data to contribute to the effort to answer these questions. POPs are often reported to have soil half-lives around ∼5
10 years based on laboratory studies and the values are often used in models. The original soil transect was taken in 1998. We therefore sampled the sites again 10 years later and analyzed the soils to gain information about the possible recycling of POPs.

’ MATERIALS AND METHODS The difficulties of observing changes in soil cores due to its heterogeneity and slow rates of change are well-known. Every effort was therefore made to take and analyze the 2008 samples in exactly the same way as the 1998 ones. Surface soil samples were collected along the original transect in summer 2008 (see Figure SI-1 of the Supporting Information, SI), using the same procedures as in 1998.8 The UK sample sites were revisited in 2008 following detailed maps (Ordnance Survey, 1:50 000) and notes made in 1998. In Norway, the experienced field guide who led the first sampling campaign relocated the same sites. GPS data were taken at all sampling sites and compared to references from the 1998 campaign. The litter and vegetation layer was removed and then a bulb planter was used to extract 10 soil cores each of 6 cm diameter, from 0 to 5 cm and 5
10 cm depths. In total, 70 samples were collected. In brief, all analytical methods were the same as applied to the 1998 soils, except for the BC determination and the addition of the chiral PCB analysis. Chirality provides evidence for biodegradation, as opposed to abiotic loss processes.19,20 The fieldmoist soil cores were mixed, before the equivalent of 3
55 g soil dry weight (dw) was separated for chemical analysis. The samples were Soxhlet extracted in dichloromethane; cleanup included alumina/silica columns, acid digestion, and gel permeation columns. The samples were transferred into dodecane containing PCB 30, [13C12] PCB 141, [13C12] PCB 208, PBDE 69, and PBDE 181 as internal standards and analyzed by gaschromatography
mass-spectrometry (GC
MS) with an EI+ source operating in selected ion monitoring mode (SIM) for PCBs and PBDE. The congeners reported here are the tetra-CBs (4CB) 44, 49, 52, 70, 74, the penta-CBs (5CB) 87, 90/101, 95, 99, 105, 110, 118, the hexa-CBs (6CB) 138, 141, 149, 151, 153/ 132, 156, 157, 158, 167, the hepta-CBs (7CB) 170, 174, 180, 183, 187, and the octa-CBs (8CB) 194, 203 as well as the PBDEs 47, 99, 100, 153, 154. We also report data for HCB, which was analyzed in the 1998 samples8 and other OCs (R-chlordane, γchlordane, p,p0 -DDD, o,p0 -DDE, p,p0 -DDE, o,p0 -DDT, p,p0 DDT) which were analyzed but not reported. Chiral analysis for PCBs 95, 132, 149, and 174 was performed on the 2008 soils and stored extracts of 2002
2004, 2004
2006 and 2006
2008 air samples from the transect13,14 applying GC
GC
MS
MS at the Agroscope ART. The instrument consisted of two Varian CP3800 gas chromatographs with electron capture detection and a triple quadrupole mass spectrometer (Varian 1200) in EImode. Achiral separations were performed on a HT-8 capillary column and chiral separations on a Chirasil Dex column. Details of the method and calculation of enantiomeric fractions (EF) are described by Bucheli and Br€andli.19 The soils were characterized

ARTICLE

by determining bulk density, organic matter (SOM), and black carbon (BC) content. SOM was determined by loss-on-ignition at 450 C1 and BC by a modified version of the chemo-thermal oxidation method adapted for application in soils.21 Solvent blanks were analyzed in parallel with the samples. All samples were blank corrected. The limits of detection (LOD) were determined as the sum of the average blank value plus three times the standard deviation of the blanks (see Table SI-2 of the SI). All samples were spiked with recovery solution containing a [13C12] -PCB mix (28, 52, 101, 138, 153, 180) and a PBDE-mix (51, 128, 190). PCB recoveries ranged from 84
123% (average 101%) and PBDEs from 72
133% (average 100%). Definition of Sampling Site and Soil Characteristics as the Basis for Data Analysis. Details of sampling site and soil characteristics can be found in Figure SI-3 of the SI and information about the statistical tools applied for the data analysis in Figure SI-6 of the SI. Of the samples originally collected in 1998, the data of 41 soil sampling sites in the UK and Norway (17 and 24 sites, respectively) was published by Meijer et al.1 At 25 of those sites, soil samples could be collected at both grassland (GL) and woodland (WL) sites in close proximity. In 2008, 44 of the 1998 soil sampling sites were revisited (19 in the UK and 25 in Norway), of which 18 sites contributed both GL and WL soil samples. The sampling sites were chosen in remote/rural areas covering a transect from 50.6
70.4N and 6.2W
27.9E. The mean winter and summer temperatures range from
12
5.3 C and 10.4
16.0 C, respectively. The sampling sites are further differentiated by the type of vegetation. GL encompasses all sampling sites without trees or shrubs, mainly meadows, pastures, heath, and peat bog. Vegetation at WL sites is forest with shrubs and either deciduous (D), coniferous (C), or mixed tree types. In the discussion and data analysis, the whole data set is considered and the subsets GL, WL, D, and C. The sampling sites were further characterized by their proximity to populated areas and the population density. Details for the estimation of the relative population density (RPD) and site specific RPD values can be found in Figure SI4 and Table SI-3.a of the SI. Figure SI-3.b-c of the SI summarizes the SOM and BC information. SOM in 1998 and 2008 ranged from 3
97% and 2
99% (g SOM per g dw soil), respectively, and was not significantly different between the years (p > 0.4). BC values in 1998 and 2008 ranged from 0.05
0.31% and 0.03
0.95% respectively and were significantly different (p < 1.2  10
6) between the two sampling periods. However, this probably reflects a slight difference in methodology; sample acidification was performed after the removal of nonpyrogenic OM for the 2008 samples21 and before for the 1998 samples.11,22,23 For comparison, BC content in Swiss soils sampled during the last 25 years and analyzed with the method described in Agarwal and Bucheli24 remained constant. Due to current soot emissions and legislation, a significant rise in BC is not likely between the sampling periods. In both 1998 and 2008, the average SOM content of GL sites was lower than for WL. SOM was also significantly different between C and D sites (SOM average C > D) in both years. BC was significantly different (p < 0.04) between GL (0.03
0.6%) and WL (0.1
0.9%) sites in 2008 (though not in 1998). The discussion and analysis of the POPs soil data presented here is based on the following premise: The soil sampling sites of 1998 were all traced within a perimeter in which there are no differences in possible POP sources or vegetation. It was 7292

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology

ARTICLE

Table 1. Average Soil Concentrations and Ranges for the Data Sets 1998 and 2008 for ∑31PCB, ∑5PBDE and HCBa average ( standard deviation minimum
maximum [pg/g SOM] ∑5PBDE

HCB 1120 ( 1250

all 2008

6450 ( 5450

590 ( 750

n = 70

210
27 100

12
3660

37
7310

Norway

7760 ( 6440

300 ( 400

1390 ( 1030

n = 40

210
27 100

12
1580

230
4580

UK

4700 ( 3050

940 ( 910

780 ( 1440

n = 30 GL

450
10 100 4800 ( 4860

100
3660 490 ( 770

37
7310 990 ( 1460

n = 30

210
22 800

16
3660

64
7310

WL

7690 ( 5590

660 ( 730

1210 ( 1100

n = 40

590
27 100

12
2850

37
4580

All 1998

9190 ( 8900

2740 ( 3210

1400 ( 1000

n = 74(48)

720
39 700

140
19 900

90
5500

Norway

8470 ( 8750

1260 ( 780

1600 ( 1200

n = 40(21) UK

720
39 700 10 200 ( 9160

300
2940 3730 ( 3810

91
5500 1100 ( 770

n = 34(27)

2400
39 600

140
19 900

250
4200

GL

6180 ( 7420

4030 ( 5130

730 ( 340

n = 31(15) WL n = 43(33) a

∑31PCB

980
39 700

140
19 900

250
1400

11 300 ( 9330

2210 ( 1830

1900 ( 1100

720
39 600

300
9140

90
5500

n = samples analysed in data set (in brackets for PBDEs if different).

established in previous publications that due to their physical chemical properties, POPs are mainly associated with the SOM of a heterogeneous soil sample. The soil sampling data here is mostly reported as mass unit per unit SOM (if not stated otherwise) to compensate for differences in SOM fraction in the soil samples between different years and sites.

’ RESULTS AND DISCUSSIONS Summary of the POP results. Table 1 and Figure 1 summarize the 2008 and 1998 ΣPCB, ΣPBDE and HCB data in pg/g SOM (Table SI-5.a of the SI summarizes the data for 4CB, 5CB, 6CB, 7CB, 8CB, ΣPCB, ΣPBDE, and OCs in pg/g SOM and pg/g dw). To establish if there were changes for the POP concentrations in soil between 1998 and 2008 two different methods were applied. To observe general trends, the whole data sets (and subsets) were compared utilizing the unpaired and paired t test. To observe specific trends for local changes in POP concentrations the data of the individual sites was compared separately (allowing for an empirical analytical uncertainty of 13%). Details for the statistical approaches used in the discussion can be found in Figure SI-6.a of the SI. PCBs. ∑31PCB concentrations in 1998 ranged over 2 orders of magnitude, from 720 to 39 700 pg/g SOM for all soil samples (Table 1). Despite much higher usage and emissions of PCBs in the UK than in Norway,25 the soil concentrations for the two countries were not significantly different.8 While the lighter

congeners (4CB, 5CBs, 6CBs) were found in higher concentrations in WL than GL sites (p < 0.01), the data sets were not significantly different for the heavier PCBs nor was there a notable difference between the C and D data. In 2008 ∑31PCB concentrations again ranged widely, from 210 to 27 100 pg/g SOM. The concentrations in the subsets were significantly different with Norwegian soils > UK soils (p < 0.01) and WL > GL (p = 0.04) (but again no difference between C and D). The ∑31PCB data from 1998 and 2008 analyzed with the unpaired and paired t test showed significant differences for the sampling years for the whole data set as well as all subsets with 2008 < 1998 concentrations. By analyzing the data for the homologue groups the results vary. There was little difference between the sampling years for the Norwegian sites, whereas there was for the UK sites. The data for the 5CBs were not different between sampling years for the whole data set or the subsets. For all homologue groups, there was no difference for the GL data in 1998 and 2008, while for the WL sites (except 5CB) the data were significantly lower in 2008 but no difference was observed between the WL subsets C and D. (see Table SI-6.b of the SI). The individual assessment of the samples of 1998 and 2008 (n = 68) showed that the ∑31PCB concentration for 56% of the samples was significantly lower in 2008, 25% were higher and 19% of the sites showed no difference. On average, ∑31PCB concentrations in 2008 were 61 ( 84% of the 1998 values (or 42 ( 21% for the 56% of the data with 2008 < 1998). Assuming first order kinetics the average half-life found for ∑31PCB for all samples would be 10 ( 183 years (the samples that show nearly no change between the sampling years or are actually higher in 2008 cause a high standard deviation). If only those samples with a reasonable decline (i.e., the 56% of the data) are considered, then the average half-life is found to be 9.8 ( 6.2 years. This is close to the average half-life for atmospheric PCB data along the same transect reported with 8.4 ( 3.2 years.13 PBDEs. In 1998, concentrations for ∑5PBDE9 ranged from 140 to 19 900 pg/g SOM (Table 1). The Norwegian and UK sites were significantly different with UK > Norwegian data (while there were no differences noted for GL/WL and C/D). In 2008, ∑5PBDEs concentrations ranged from 12 to 3660 pg/g SOM. The concentrations in the UK were again higher than in Norway. The t test results showed that there was a significant difference for ∑5PBDE and individual congeners between the sampling years with 1998 > 2008 concentrations. As for the PCBs, only the GL data showed no difference between the years. Comparing the PBDE samples individually (n = 48) shows that values in 2008 were significantly lower at 79% of all sites, not different at 6% and higher at 15%. The 2008 data for all sites were on average 66 ( 102% of the 1998 concentrations (23 ( 15% for the 79% of the data with 2008 < 1998). The half-life for ∑5PBDE is 0.7 ( 21 years as an average at all sites and 4.6 ( 3.7 years considering only the sites with lower concentrations in 2008. The 50% decline rate of atmospheric concentrations on the same transect was 2.2 ( 0.4 years.13 HCB and Other OCs. In 1998, HCB concentrations ranged from 90 to 5500 pg/g SOM. Among the group of DDT and its derivatives p,p0 -DDT was the most abundant (330
170 000 pg/g SOM) followed by p,p0 -DDE (220
71 800 pg/g SOM). R-Chlordane was found to be more abundant than γ-chlordane. In general, there were no differences between the soil subsets noted. HCB concentration in 2008 ranged from 37 to 7300 pg/g SOM. The trends among the DDT derivatives and chlordanes had not shifted between 1998 and 2008 (p,p0 -DDT 40
80 000 7293

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology

ARTICLE

Figure 1. Soil concentrations for the data sets 1998 and 2008 for ∑31PCB, ∑5PBDE and HCB (on log scale) for GL and WL sites (Figure SI-5.b. of the SI presents concentrations for the whole data, UK/Norway and C/D).

Table 2. Overview for the Trends for ∑31PCB, ∑5PBDE, and HCB between Data Sets and Subsets 1998 and 2008a data for ∑31PCB 1998

data for ∑5PBDE 2008

1998

UK < Nor

UK > Nor

PCB 98 < PCB 08 UK ∼ Nor

2008

UK > Nor

GL ∼ WL

2008 HCB 98 ∼ HCB 08

UK ∼ Nor

UK ∼ Nor UK 98 ∼ UK 08

UK 98 > UK 08

Nor 98 > Nor 08 # GL < WL #

Nor 98 > Nor 08 GL ∼ WL

GL < WL

Nor 98 ∼ Nor 08 GL ∼ WL

GL 98 > GL 08 #

GL 98 > GL 08

GL 98 ∼ GL 08

WL 98 > WL 08

WL 98 > WL 08

WL 98 > WL 08

C∼D

a

1998

PBDE 98 < PBDE 08

UK 98 > UK 08 GL ∼ WL #

data for HCB

C∼D

C∼D

C∼D

C∼D

C∼D

C 98 > C 08 #

C 98 > C 08

C 98 >C 08

D 98 > D 08 #

D 98 > D 08

D 98 ∼ D 08

∼indicates not different in the t-Test, # indicates different trends for the homologues/congeners which can be found in Figure SI-6.b of the SI.

pg/g SOM, p,p0 -DDE 17
6700 pg/g SOM). The t tests of the 1998 and 2008 data showed that most data sets were not significantly different (with the exception of HCB WL data). The statistical information for the different POPs and the data subsets can be found in the SI-6.b. The overall trends observed are summarized in Table 2 and were rather different between the contaminant groups. While for OCs there were almost no changes between years, the PBDEs show the strongest decline, except in the GL sites. A possible reason why PBDEs have the

greatest measurable decline is that they are more rapidly degraded in soils than other compound classes.5,26 Another reason could be that their observed decline in the atmosphere along the transect is faster than for PCBs (possibly due to a quicker reduction of emissions).13 The trends for PCBs are less obvious and differ between subsets and homologues groups. It was previously noted that a general trend for the POP concentrations in both sampling years was WL > GL. Previous studies have highlighted the role of the “forest filter effect” in enhancing 7294

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology

Figure 2. Idealized plots of log PCB against log SOM for the homologue groups for 1998 and 2008 highlighting the deviation for the slopes in 1998.

deposition rates, with resulting elevated loadings in forest soils.3,27
29 Several factors could contribute to changing burdens in surface soils
reduced inputs (lowering air concentrations and deposition) and losses via biodegradation, SOM formation and burial, leaching and re-emission to atmosphere. The following discussion examines those trends to identify processes leading to changes (or the lack thereof) in POP concentrations in soils during the 10 years. Observations on Factors Influencing the Concentrations and Trends. SOM Content. It was previously discussed by Meijer et al.8 that SOM exerted a strong influence on POP loadings in the 1998 samples. This varied between homologue groups: when log POP concentrations are plotted against log SOM steeper slopes with less scatter were generally seen for lighter, more volatile compounds than the heavier ones. This leads to the hypothesis that their distribution reflects repeated air-surface exchange (“hopping”) and that they approach an “air-soil equilibrium”.5 Interesting issues to investigate here were therefore whether the slopes or the strengths of the regressions were different or similar in 2008 from 1998. If they are similar, then this would imply that little recycling/re-emission of the bulk of the burden in the 0
5 cm soil layer has taken place (or that there were several counter-acting processes). If they were different, then this could infer that there was a redistribution of a substantial proportion of previously deposited POP in the intervening 10 years. Steeper slopes can be interpreted as an indication that the SOM has a loading capacity which was not fully saturated in 1998. If slopes become less steep over time, then this may indicate POP revolatilisation or leaching to deeper layers (coupled with decreasing atmospheric concentrations to replenish the SOM reservoir). It has been reported that with the aging of soil POPs and SOM form a “nonextractable residue”.30 Therefore, the POPs in SOM might only be available for a certain time as a “recyclable pool” that undergoes exchange with the surface “skin” of the soil and plant OM pool before it becomes incorporated into subsurface soil layers.3 Table SI-7.a of the SI contains all log POP/log SOM slopes and statistical parameters for the 1998 and 2008 data sets for

ARTICLE

individual congeners. There were no differences for PCBs between the slopes of GL soil in 1998 and 2008 (0.7 ( 0.2) (see Figure SI-7.c of the SI) which is in accordance with the earlier observation that there were no significant differences in the POP concentrations between GL 1998 and 2008. For the WL sites the 2008 PCB slopes were 1.0 ( 0.2 over the whole SOM range, whereas the 1998 slopes gave a deviation of the correlation curve close to 40
60% SOM (Figure 2 and Figure SI-7.b of the SI). Slopes calculated for SOM < 60% only are 1.3 ( 0.2 and similar to the ones observed in 2008. Including values for SOM > 60% for the 1998 data results in fewer slopes with significant linear correlation. The resulting slopes correlate with the octanolair partitioning coefficient with decreasing slopes for increasing SOM. (This was not observed in GL 1998 due to SOM < 63%.) This implicates a slow saturation of soils with SOM > 60% during the 10 years for heavier PCBs and the approach of an air-soil equilibrium. Soils with high SOM will take longer to reach air-SOM equilibrium. However, these data suggest that by 2008 even soils with very high SOM are now approaching equilibrium within the atmosphere, whereas a decade earlier they were not. This will have implications for the past and future trends in the global PCB cycle. For PBDEs, there was no significant difference between the slopes in 1998 and 2008. No significant correlation was observed between POP concentrations and black carbon in both years. Fractionation with Latitude. Air concentrations have been monitored along the transect at 11 sites with passive samplers since 1994.8,13
15,31 Atmospheric PCB concentrations on this transect have always demonstrated fractionation with latitude (and temperature). With increasing latitude (decreasing temperature) the fractions of the lighter PCB congeners increase, while the heavier ones decrease. This did not change between the sampling years from 1994
2008 even though PCB concentrations declined in general. This is considered evidence for the continuing (though declining) influence of ongoing primary sources on ambient and background levels of PCBs in the atmosphere.13 Fractionation was also observed for PCBs in the 1998 soils data on this transect, while the 2008 results showed no evidence of fractionation of PCBs (Figure SI-8 of the SI) or PBDEs in 2008. This loss of fractionation between 1998 and 2008 is evidence of “weathering”—selective removal of congeners which make up the soil burden—and of the declining influence of atmospheric inputs. Influence of Proximity to Areas of Population/Population Density. In a primary source controlled world, PCB and PBDE air concentrations will be highest close to emission sources.25 von Waldow et al.16 and Schuster et al.12 recently investigated the influence of proximity to populated areas on the air concentrations on the transect. PCB air concentrations were correlated to population density. Information on the approach used to explore the link between soil concentrations in 1998 and 2008 with population is given in Figure SI-4 of the SI, together with the full results. In summary, the findings were as follows: (i) better correlations were obtained if the WL and GL soils or Norway and UK soils were assessed separately, because land use and national usage/emissions differences are confounding factors; (ii) correlation between soil loadings and population were clear in the 1998 data, but weaker/nonexistent for the subsets of the 2008 data (see Figure SI-4.c,d of the SI); (iii) in 1998 population correlated with soil 4CB, 5CB, 6CB, 7CB, and 8CB homologues (not with 3CBs), but only with the heavier 6CB, 7CB, and 8CB homologues in 2008. These observations are therefore further evidence that primary sources within the study region had a 7295

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology

ARTICLE

Figure 3. Enantiomer fraction (EFs) for PCB 95 and PCB 149 for 2008 soil samples and 2006
2008 air data.13

stronger influence on PCB distributions in soils in 1998 than in 2008 and further evidence for net weathering/loss of PCBs from the surface soils between 1998 and 2008. The changes of correlation between population density and POP concentration in soil between 1998 and 2008 are greater for lighter than heavier homologues. Similar observations were noted for PAHs by Br€andli, et al.32 In the 1998 survey,8 lighter congeners were more strongly correlated to SOM (which is evidence of recycling and re-emission), while heavier congeners are “stickier” and tend to stay closer to sources. Transport to Deeper Layers. In 2008 and 1998, 4 WL and 4 GL locations were sampled from 0 to 10 cm and split into surface (0
5 cm) and subsurface (5
10 cm) layers. In 1998, higher amounts of PCBs were generally found in the top layers. In 2008, there were more sites with higher PCB loadings in the lower layer in GL, while they were still higher in the top layer for WL (ongoing forest filter effect). These observations could be interpreted as a stronger weathering process with less atmospheric input in GL compared to WL or of greater amounts of mobile SOM in GL, but it has to be noted that the number of deep soil cores analyzed limit the strength of these observations.3 Biodegradation. Direct information on the rates of biodegradation that apply to POPs in the environment (and soils in particular) is difficult to obtain due to their persistence in general and the lack of control/information on environmental variables. However, chiral signatures provide clear evidence that biodegradation is occurring in field soils.17,20,33 If enantiomer mixtures are racemic, then this is evidence of fresh inputs and/or lack of biodegradation (or equal rates of biodegradation for both enantiomers), while the dominance of the + or
enantiomer provides evidence that (selective) biodegradation has occurred. The enantiomers of PCB congeners 95, 132, 149, and 174 were analyzed and enantiomeric fractions (EFs) determined for all the 2008 soil samples. Enantiomers are distinguished by elution order (E1/E2) or optical activity ((). Values ranged from 0.32
0.65 for PCB 95 (E1), 0.45
0.70 for PCB 132 (+), 0.30
0.55 for PCB 149 (+) and 0.23
0.57 for PCB 174 (+) (the EF quantification precision for the analytical method reported by Bucheli and Br€andli19 ranges from 0.8
2.2%). EFdata for 2008 in soil and atmospheric samples13 are shown in Figure 3. EF-values were considered significantly different from racemic for values > three times the standard deviation of EFs of

racemic mixtures. On this basis, 87%, 78%, 88%, and 50% of the soils were nonracemic for congeners PCB 95, 132, 149, and 174, respectively. Clearly, PCBs—both lighter and heavier congeners—are being degraded in these background soils. This strongly suggests that biodegradation plays a (potentially important) part in weathering and removal of PCBs and other POPs from these background soils. The extent of enantioselectivity (both with absolute numbers and with racemic deviations) observed in the soils was tested against land use, SOM and BC content, latitude, average temperature, relative population density and total PCB concentrations. None of these factors were linked to a preference for (+)- or (
)-enantiomers or with the degree of enantioselectivity. The comparison of the chiral signature in atmospheric samples along the transect gives a clear indication for the question about the dominating influence of primary versus secondary sources. Extracts of air samples taken on the transect in 2002
2004, 2004
2006, and 2006
2008 were analyzed for chiral PCBs. As Figure 3 shows, these were not significantly different from racemic. For atmospheric PCB concentrations diffuse primary sources are still dominant34 here; any re-emissions from these surface soils are at this time a minor component of the atmospheric burden in remote areas of this transect.

’ ASSOCIATED CONTENT

bS

Supporting Information. Details about sampling sites, fractionation, log SOM slopes, POP concentrations, and relative population density. This material is available free of charge via the Internet at http://pubs.acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected].

’ ACKNOWLEDGMENT We thank colleagues involved in the soils sample collection and the Norwegian Research Council and the UK Department of Environment, Food and Rural Affairs for financial support. 7296

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297

Environmental Science & Technology

’ REFERENCES (1) Meijer, S. N.; Ockenden, W. A.; Sweetman, A.; Breivik, K.; Grimalt, J. O.; Jones, K. C. Global distribution and budget of PCBs and HCB in background surface soils: Implications or sources and environmental processes. Environ. Sci. Technol. 2003, 37 (4), 667–672. (2) Wania, F.; Su, Y. S. Quantifying the global fractionation of polychlorinated biphenyls. Ambio 2004, 33 (3), 161–168. (3) Moeckel, C.; Nizzetto, L.; Di Guardo, A.; Steinnes, E.; Freppaz, M.; Filippa, G.; Camporini, P.; Benner, J.; Jones, K. C. Persistent organic pollutants in boreal and montane soil profiles: Distribution, evidence of processes and implications for global cycling. Environ. Sci. Technol. 2008, 42 (22), 8374–8380. (4) Harrad, S. J.; Sewart, A. P.; Alcock, R.; Boumphrey, R.; Burnett, V.; Duartedavidson, R.; Halsall, C.; Sanders, G.; Waterhouse, K.; Wild, S. R.; Jones, K. C. Polychlorinated-biphenyls (PCBs) in the British environment - Sinks, sources and temporal trends. Environ. Pollut. 1994, 85 (2), 131–146. (5) Li, Y. F.; Harner, T.; Liu, L. Y.; Zhang, Z.; Ren, N. Q.; Jia, H. L.; Ma, J. M.; Sverko, E. Polychlorinated biphenyls in global air and surface soil: Distributions, air-soil exchange, and fractionation effect. Environ. Sci. Technol. 2009, 44 (8), 2784–2790. (6) Ayris, S.; Harrad, S. The fate and persistence of polychlorinated biphenyls in soil. J. Environ. Monit. 1999, 1 (4), 395–401. (7) Dalla Valle, M.; Dachs, J.; Sweetman, A. J.; Jones, K. C. Maximum reservoir capacity of vegetation for persistent organic pollutants: Implications for global cycling. Global Biogeochem. Cycles 2004, 18, 4. (8) Meijer, S. N.; Steinnes, E.; Ockenden, W. A.; Jones, K. C. Influence of environmental variables on the spatial distribution of PCBs in Norwegian and UK soils: Implications for global cycling. Environ. Sci. Technol. 2002, 36 (10), 2146–2153. (9) Hassanin, A.; Breivik, K.; Meijer, S. N.; Steinnes, E.; Thomas, G. O.; Jones, K. C. PBDEs in European background soils: Levels and factors controlling their distribution. Environ. Sci. Technol. 2004, 38 (3), 738–745. (10) Hassanin, A.; Lee, R. G. M.; Steinnes, E.; Jones, K. C. PCDD/Fs in Norwegian and UK soils: Implications for sources and environmental cycling. Environ. Sci. Technol. 2005, 39 (13), 4784–4792. (11) Nam, J. J.; Thomas, G. O.; Jaward, F. M.; Steinnes, E.; Gustafsson, O.; Jones, K. C. PAHs in background soils from Western Europe: Influence of atmospheric deposition and soil organic matter. Chemosphere 2008, 70 (9), 1596–1602. (12) Schuster, J. K.; Gioia, R.; Sweetman, A. J.; Jones, K. C. Temporal trends and controlling factors for polychlorinated biphenyls in the UK atmosphere (1991
2008). Environ. Sci. Technol. 2010, 44 (21), 8068–8074. (13) Schuster, J. K.; Gioia, R.; Breivik, K.; Steinnes, E.; Scheringer, M.; Jones, K. C. Trends in European background air reflect reductions in primary emissions of PCBs and PBDEs. Environ. Sci. Technol. 2010, 44 (17), 6760–6766. (14) Gioia, R.; Steinnes, E.; Thomas, G. O.; Mejier, S. N.; Jones, K. C. Persistent organic pollutants in European background air: derivation of temporal and latitudinal trends. J. Environ. Monit. 2006, 8 (7), 700–710. (15) Meijer, S. N.; Ockenden, W. A.; Steinnes, E.; Corrigan, B. P.; Jones, K. C. Spatial and temporal trends of POPs in Norwegian and UK background air: Implications for global cycling. Environ. Sci. Technol. 2003, 37 (3), 454–461. (16) von Waldow, H.; MacLeod, M.; Jones, K.; Scheringer, M.; Hungerbuhler, K. Remoteness from emission sources explains the fractionation pattern of polychlorinated biphenyls in the Northern hemisphere. Environ. Sci. Technol. 2010, 44 (16), 6183–6188. (17) Borja, J.; Taleon, D. M.; Auresenia, J.; Gallardo, S. Polychlorinated biphenyls and their biodegradation. Process Biochem. 2005, 40 (6), 1999–2013. (18) Sinkkonen, S.; Paasivirta, J. Degradation half-life times of PCDDs, PCDFs and PCBs for environmental fate modeling. Chemosphere 2000, 40 (9
11), 943–949.

ARTICLE

(19) Bucheli, T. D.; Brandli, R. C. Two-dimensional gas chromatography coupled to triple quadrupole mass spectrometry for the unambiguous determination of atropisomeric polychlorinated biphenyls in environmental samples. J. Chromatogr. A 2006, 1110 (1
2), 156–164. (20) Koblickova, M.; Ducek, L.; Jarkovsky, J.; Hofman, J.; Bucheli, T. D.; Klanova, J. Can physicochemical and microbial soil properties explain enantiomeric shifts of chiral organochlorines? Environ Sci. Technol. 2008, 42 (16), 5978–5984. (21) Agarwal, T.; Bucheli, T. D. Adaptation, validation and application of the chemo-thermal oxidation method to quantify black carbon in soils. Environ. Pollut. 2011, 159 (2), 532–538. (22) Gustafsson, O.; Gschwend, P. M., Soot as a strong partition medium for polycyclic aromatic hydrocarbons in aquatic systems. In Molecular Markers in Environmental Geochemistry, Eganhouse, R. P., Ed. American Chemical Society: Washington DC, 1997; Vol. 671, pp 365
381. (23) Nam, J. J.; Gustafsson, O.; Kurt-Karakus, P.; Breivik, K.; Steinnes, E.; Jones, K. C. Relationships between organic matter, black carbon and persistent organic pollutants in European background soils: Implications for sources and environmental fate. Environ. Pollut. 2008, 156 (3), 809–817. (24) Agarwal, T.; Bucheli, T. D. Is black carbon a better predictor of polycyclic aromatic hydrocarbon distribution in soils than total organic carbon? Environ Pollut. 2011, 159 (1), 64–70. (25) Breivik, K.; Sweetman, A.; Pacyna, J. M.; Jones, K. C. Towards a global historical emission inventory for selected PCB congeners—A mass balance approach 1. Global production and consumption. Sci. Total Environ. 2002, 290 (1
3), 181–198. (26) Wania, F.; Dugani, C. B. Assessing the long-range transport potential of polybrominated diphenyl ethers: A comparison of four multimedia models. Environ. Toxicol. Chem. 2003, 22 (6), 1252–1261. (27) Horstmann, M.; McLachlan, M. S. Atmospheric deposition of semivolatile organic compounds to two forest canopies. Atmos. Environ. 1998, 32 (10), 1799–1809. (28) Maddalena, L. R.; McKone, E. T.; Riley, J. W. Is There a Forest Filter Effect for Organic Pollutants? Springer: Berlin, Allemagne, 2003; Vol. 17, p 4. (29) Nizzetto, L.; Cassani, C.; Di Guardo, A. Deposition of PCBs in mountains: The forest filter effect of different forest ecosystem types. Ecotox. Environ. Safe. 2006, 63 (1), 75–83. (30) Doick, K. J.; Klingelmann, E.; Burauel, P.; Jones, K. C.; Semple, K. T. Long-term fate of polychlorinated biphenyls and polycyclic aromatic hydrocarbons in an agricultural soil. Environ. Sci. Technol. 2005, 39 (10), 3663–3670. (31) Jaward, F. M.; Meijer, S. N.; Steinnes, E.; Thomas, G. O.; Jones, K. C. Further studies on the latitudinal and temporal trends of persistent organic pollutants in Norwegian and UK background air. Environ. Sci. Technol. 2004, 38 (9), 2523–2530. (32) Br€andli, R. C.; Bucheli, T. D.; Ammann, S.; Desaules, A.; Keller, A.; Blum, F.; Stahel, W. A. Critical evaluation of PAH source apportionment tools using data from the Swiss soil monitoring network. J. Environ. Monit. 2008, 10 (11), 1278–1286. (33) Harner, T.; Wiberg, K.; Norstrom, R. Enantiomer fractions are preferred to enantiomer ratios for describing chiral signatures in environmental analysis. Environ. Sci. Technol. 1999, 34 (1), 218–220. (34) Robson, M.; Harrad, S. Chiral PCB signatures in air and soil: Implications for atmospheric source apportionment. Environ. Sci. Technol. 2004, 38 (6), 1662–1666.

7297

dx.doi.org/10.1021/es200961p |Environ. Sci. Technol. 2011, 45, 7291–7297