PAH Fluxes to Siskiwit Revisted: Trends in Fluxes and Sources of

Apr 12, 2013 - Trends in concentrations and radiocarbon content of pyrogenic PAHs and perylene ... Environmental Science & Technology 2018 Article ASA...
0 downloads 0 Views 948KB Size
Download PDF
Article pubs.acs.org/est

PAH Fluxes to Siskiwit Revisted: Trends in Fluxes and Sources of Pyrogenic PAH and Perylene Constrained via Radiocarbon Analysis G. F. Slater,*,† A. A. Benson,† C. Marvin,‡ and D. Muir‡ †

School of Geography and Earth Sciences, McMaster Unviersity, General Sciences Building room 306, 1280 Main Street West, Hamilton, Ontario, L8S 4K1, Canada ‡ Aquatic Contaminants Research Division, Environment Canada, Burlington, Ontario, Canada ABSTRACT: Trends in concentrations and radiocarbon content of pyrogenic PAHs and perylene were determined 20 years after a previous study by Mcveety and Hites (1988). Pyrogenic PAH fluxes to sediments were observed to continue to decrease over the period from 1980 to 2000 at this remote site in contrast to observations in more urban areas. Radiocarbon analysis of pyrogenic PAHs showed a 50% decrease in the proportion of pyrogenic PAH derived from fossil fuel combustion over the past 50 years, consistent with decreasing emissions from regional coal-fired powergenerating plants. Fluxes of pyrogenic PAHs related to biomass burning were consistent over this same period and found to exceed fossil fuel sources in the most recent samples. Fluxes of biomass-derived pyrogenic PAHs were similar in magnitude to total pyrogenic PAH fluxes in early 1900, suggesting that these fluxes may represent wildfire inputs. Not only did perylene concentrations in these sediments increase with depth as previously observed but also concentrations from the same sedimentary layers analyzed 20 years previously showed large increases in perylene concentrations. Radiocarbon analysis of perylene indicated that 70−85% of perylene observed in the deeper sediments could be explained by production from total organic carbon.

■

INTRODUCTION Polycyclic aromatic hydrocarbon (PAHs) are persistent organic pollutants (POP) consisting of two or more fused aromatic rings that are ubiquitous in environmental compartments.1 Several PAHs, notably benzo(a)pyrene, have been linked to health-related concerns.2 The primary source of PAH to the environment is pyrogenic release as a result of incomplete combustion of organic material including fossil fuels (e.g., coal, petroleum) and modern biomass (e.g., wood)1,3 The majority of these PAHs are deposited close to the combustion source; however, a fraction of the particulate can be transported over large distances.1,4 Determination of trends in PAH concentrations in lake sediments can be used to assess regional trends in PAH sources and atmospheric concentrations.5−11 In the 1980s and early 1990s studies of sedimentary profiles of pyrogenic PAHs generally observed deposition peaking in the mid to late 1950s and decreasing afterward (e.g., refs 8 and 12). It was assumed that this decline would continue; however, more recent studies, including several in the Great Lakes region, observed that in many cases PAH fluxes have either stabilized or increased since ca. 1980.5−7,9,10,13 In urban environments large increases in PAH loadings have been related to stormwater runoff carrying inputs derived from parking lot sealcoat.14,15 In more remote regions, increased inputs have been related to diesel consumption5,16 and/or changes in combustion inputs.6,17 Concurrently, atmospheric studies of PAH concentration have generally found decreasing © XXXX American Chemical Society

trends in PAH concentrations in the Great Lakes region during this time period18,19 but have noted that variability in these trends make generalizations to a regional area difficult.19 In many studies perylene has been observed to have a distinct behavior from pyrogenic PAHs associated with atmospheric deposition. Profiles of increasing perylene concentrations with depth, often to concentrations far in excess of the pyrogenic PAHs, have led to an ongoing discussion concerning sources of perylene.20−27 The distinct behavior of perylene in a number of sedimentary environments has led to the hypothesis that the observed perylene in these profiles is a result of in situ production by microbial communities living under anoxic conditions utilizing the more recalcitrant fractions of total organic carbon.20,25−28 Several studies have noted a first-order kinetic model is consistent with perylene profiles.20,28 However, other studies have related perylene inputs to terrestrial runoff of fungiderived perylene21,23,24 or to a combination of terrestrial inputs and in situ diagenesis.29 At this time the precursor of perylene production has not been identified. Compound-specific radiocarbon analysis has been used in a number of studies to resolve PAH sources to sedimentary and Received: January 19, 2013 Revised: April 7, 2013 Accepted: April 12, 2013

A

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 1. Location of Siskiwit Lake, Isle Royale, and regional potential sources for combustion-related PAHs including coal-fired power plants in Atikokan and Thunder Bay. Wind roses illustrate prevailing winds in the region for Atikokan, Thunder Bay, and Duluth, MN, for 1970−2000.

atmospheric environments.30−36 These studies have capitalized on the difference in radiocarbon content between fossil fuel sources of PAHs which contain no significant radiocarbon (Δ14C = −1000‰) and biomass burning sources which contain modern levels of radiocarbon (Δ14C = ca. 50‰).37 Using an isotopic mass balance approach, the relative contributions of these sources to PAHs present in an atmospheric or sedimentary sample can be apportioned. This approach has also been applied to gain insight into the origins of perylene in samples. Reddy et al. (2002) showed that perylene was the most modern PAH in several reference samples, indicating inputs from modern biomass as well as fossil sources. Conversely, Madalakis et al. (2004) showed that the Δ14C of perylene in surface sediments from Stockholm, Sweden, had values from −916‰ to −812‰, indicating that perylene is predominantly derived from fossil fuel combustion at this site. The goal of this study was to revisit the Siskiwit Lake site studied by McVeety and Hites (1988) and determine trends in PAH concentrations and molecular level radiocarbon content to address the following three questions. (1) Did decreases in atmospheric deposition of pyrogenic PAHs continue beyond 1984, or did trends stay constant or increase as observed in other studies? (2) What was the relative importance of fossil carbon versus biomass sources of pyrogenic PAHs from peak deposition in 1950 until recent times? (3) What is the extent and source of perylene production since the previous study?

loadings are derived from atmospheric deposition of particulate to the surface of the lake. A series of piston cores (6.7 and 10 cm diameter) were collected from the deepest section of the lake. Cores were sliced directly after coring into 1 cm intervals, placed into plastic Ziploc bags, and kept cool and dark until they could be frozen upon return to the laboratory where they were stored at −20 °C until further processing. Sediment was dried at 50 °C in solvent rinsed and precombusted aluminum pans. Dried sediment was ground with an agate mortar and pestle and stored in the dark in precombusted 500 mL glass vials. Quantification of PAH. PAH analysis was performed using modified EPA methods 8100, 3051A, and 3630C. In short, 50 g of sediment was spiked with 5 μg/mL ppm surrogate standards (9,10-dihydrophenanthrene and m-terphenyl) then microwave extracted (MarsX) in 1:1 hexane/acetone at 115 °C for 15 min. Extracts were allowed to cool to approximately 30 °C and then filtered with hexane and dichloromethane (DCM) through glass fiber filter paper (Whatman) to remove solid sediment portion. Filtered extracts were treated with activated copper pellets to remove elemental sulfur, reduced in volume to approximately 1 mL, and fractionated on fully activated silica gel. A total of four fractions were collected: F1, hexane (aliphatic hydrocarbons); F2, 2:1 hexane/DCM (aromatic hydrocarbons); F3, DCM; F4, methanol. All fractions were archived except for the F2 containing the aromatic hydrocarbons. The F2 fraction was reduced in volume and solvent exchanged into 100% hexane, then transferred to 1 mL amber vials, and spiked with an internal standard (o-terphenyl). PAH quantification was performed on a GC/MS (Agilent 6890 GC, Agilent 5973n MS) equipped with a split/splitless injector kept in splitless mode and a 0.25 mm i.d. 30 m DBXLB column (Jackson and White). The oven was programmed to 65 °C and ramped at 4 °C/min to 320 °C and held for 10 min. The following 16 EPA priority PAHs were quantified: naphthalene, acenapthylene, acenapthene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene, benz(a)anthracene,

■

EXPERIMENTAL SECTION Study Area and Sediment Sampling. Siskiwit Lake is located on Isle Royale, MI (Figure 1). The lake is surrounded by the protected land of Isle Royal National Park, and Isle Royale itself is located 50 km offshore of Thunder Bay and is entirely surrounded by Lake Superior. Thus, there are no significant local point sources of PAH deposition to Siskiwit Lake. Consistent with the previous assumption by McVeety and Hites (1988) it is reasonable to assume that all pyrogenic PAH B

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 2. Total pyrogenic PAH flux (ng cm−2 year−1) versus calendar year for this study (closed circles), and the McVeety and Hites (1988) comparison study (open squares).

Mandalakis et al. (2004):38 Agilent GC equipped with an autosampler, a Gerstel CIS-3 cooled injection system, and a Gerstel preparative trapping device. Ten percent of effluent passes to the flame ionization detector (FID) and the remaining 90% is collected by a series of U-shaped tube traps that were cooled to 20 °C. Individual perylene peaks or groups of pyrogenic PAHs that were well separated on the DB-5MS capillary column (60 m × 0.54 mm i.d. × 0.25 μm film thickness) using a GC temperature program of 80 °C (1 min), 10 °C/min to 200 °C, 4 °C/min to 300 °C (15 min), and 10 °C/min to 320 °C10 were collected. Pyrogenic PAHs were collected into three cryogenic traps over the course of each run. Perlyene was collected individually into a fourth cryogenic trap. After collection, PAHs were eluted using DCM. Pyrogenic PAHs were combined into one sample, and perylene was eluted separately. Samples were then submitted for 14C analysis as the NOSAMs facility. Concurrently, samples of total organic carbon (TOC) were dried and submitted for radiocarbon analysis at the NOSAMS facility. 14C analysis at NOSAMS was by standard combustion methodologies. Briefly, samples were transferred to quartz ampules, and solvents were evaporated. PAHs were combusted to CO2 by closed tube combustion at 850 °C with cupric oxide. Evolved CO2 was cryogenically purified and graphitized for 14C analysis by AMS.39 Sedimentation Rates and Sediment Dating. 210Pb analysis was performed by Environment Canada’s Aquatic Contaminants Research Division (Burlington, Ontario) on a separate core taken from the site on the same date as cores used for PAH analysis. Briefly, sediment was dried and then acid treated and prepared using a 210Pb enrichment method modified from Eakins and Morrison (1978).16 The final 210Pb was plated on highly polished Ag discs, and the activity of each sample was measured with an alpha counting device in disintegrations/min (dpm). Focusing factors were calculated by integrating the total amount of excess 210Pb over the core and dividing by the expected amount based on atmospheric 210 Pb flux at the surface.

chrysene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, benzo(g,h,i)perylene, dibenz(a,h)anthracene, and ideno(1,2,3-cd)pyrene. Perylene was also quantified. All PAHs were quantified in scan mode with 5-point calibration curves generated for responses relative to the internal standard oterphenyl (R2 > 0.999). Quality Control. In order to enable subsequent isotopic analysis, addition of internal standards and surrogate standards had to be limited. Surrogate standard (9,10-dihydrophenanthrene and m-terphenyl) recoveries were always greater than 80%. Accuracy of the PAH method was determined using three samples of standard reference material EC-1 (Environment Canada, lacustrine sediment). Results were all within 16% of the reference values. Precision (average relative standard deviation < 6%) was measured by triplicate injections of standards (2 and 8 μg/mL) as well as triplicate injections of two samples during each instrumental sequence. The limit of quantification for parent PAH compounds was 1 ppm in solution or 19.4 ng/g dw. PAH extraction took place in batches, procedural blanks (n = 6) were run with each batch, and there were no detectable PAH concentrations in any of the blanks. PAH Radiocarbon Analysis. Molecular level radiocarbon analysis was applied to samples from the upper portion of the PAH deposition profile covering the period from the peak of PAH deposition in 1955 to the most recent sediments. While the combination of multiple cores inherently affects the precision of the depth profile as discussed in relation to Figure 2 below, in order to obtain sufficient masses of PAHs for radiocarbon analysis samples of 250−400 g of sediments were collected by sectioning and combining up to 40 cores of 10 cm diameter. Notwithstanding the number of cores collected, insufficient PAHs were collected in the upper sediments, necessitating combination of 0−2 and 2−4 cm sections for 14C analysis. Unfortunately the 2−4 cm sample was lost during sample processing. Sediments were extracted by microwave extraction using the same methods as concentration analysis. PAHs were purified using scaled up Cu and silica gel methods. Once PAHs had been extracted and purified, PAHs for radiocarbon analysis were collected by preparative capillary gas chromatography (PCGC) using an approach based on C

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 3. Fluxes of representative individual PAHs (ng cm−2 year−1) versus calendar year. All pyrogenic PAHs measured followed this trend.

Figure 4. Concentrations of perylene (ng g−1) versus calendar year for this study and McVeety and Hites (1988), illustrating increases in perylene concentrations over the intervening 20 year period.

■

RESULTS Pb210 Dating. Both the constant initial concentration (CIC) and the constant rate of supply (CRS) models were applied to the 210Pb data,40 and there was good agreement between the models. The CIC model was used for calculation of ages for this core as it was the closest to matching the dating from McVeety and Hites (1988) and the method used in their study. Furthermore, focusing factors calculated from 210Pb fluxes for this study were identical to those reported by McVeety and Hites (1988). We note that the CRS model indicated increases in sedimentation rate between the most recent sediments and those deeper in the core, a fact also observed by Baker and Hites (2000).41 As we did not have this data for the previous study, we applied the CIC model that allowed the most direct comparison to the previous results of McVeety and Hites (1988). Pyrogenic PAH Fluxes. The total (Σ)PAH flux (exclusive of Perylene) versus year of deposition as determined by 210Pb dating is shown in Figure 2. The pattern in ΣPAH fluxes in this

study followed that observed by McVeety and Hites (1988) but had lower peak fluxes and was somewhat broader in profile. These differences are likely related to the fact that this study combined a large number of cores to obtain sufficient PAH mass for radiocarbon analysis, as well as the fact that a separate core was used for dating. The inevitable variations between the up to 40 cores collected due to varying compaction during sampling and/or sedimentation rates likely resulting in some blurring of the profile in this study. However, the patterns of PAH fluxes were retained. The ΣPAH curve can be divided into two sections: from 1911 to 1955 fluxes increased by 278% from 6.4 ± 0.32 ng cm−2 year−1 to a maximum in 1955 of 24.2 ± 1.2 ng cm−2 year−1. This increase occurred at a constant linear rate of 0.43 ng cm−2 year−2 (n = 6, r2 = 0.990). After the maximum in 1955, PAH fluxes steadily declined to 8.5 ± 0.42 ng cm−2 year−1 in 2001 at a constant linear rate of 0.33 ng cm−2 year−2 (n = 6, r2 = 0.963). Representative plots of single PAH phenanthrene, pyrene, fluoranthene, and benzo(g,h,i)perylene fluxes are shown in D

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 3. Phenanthrene, pyrene, fluoranthene, and benzo(g,h,i)perylene fluxes increased by 120%, 283%, 220% and 363% respectively, during the period from 1911 to 1955. After 1955, all individual PAHs observed a continuous decline in flux. As expected, the trends in these plots are similar to the total PAH curve. With the exception of perylene, all individual PAHs exhibit similar trends over the length of the core. Perylene. Perylene concentrations in the uppermost sample were 194 ng g−1 in 2001 and increased by an order of magnitude to 2000 ng g−1 at 1955 (Figure 4). This increase occurs at a linear rate of 26 ng g−1 year−1 (n = 9, r2 = 0.96). After this point perylene concentrations become somewhat variable; however, they remain at ca. 2000 ng g−1 with the exception of an anomalous low value at the bottom of the core. Radiocarbon Results. Δ14C of pyrogenic PAHs, perylene, and total organic carbon (TOC) are shown in Figure 5 and

previously been correlated with the amount and type of fuel used for energy production in the United States.7,9,10 Recent Declining PAH Trends. The trends in pyrogenic PAH deposition seen for the 20 years after 1983 in this study demonstrate that long-range PAH deposition to Siskiwit Lake has continued to decrease in more recent sediments (Figure 2), and current deposition is trending toward levels observed in the early stages of industrialization as recorded at the bottom of the profiles. The most likely source of PAHs deposition to Siskiwit lake, based on general trends in wind directions in the region illustrated in Figure 1 (Environment Canada 2009), are regional sources such as two coal power generation plants in Thunder Bay and Atikokan Ontario (Figure 1), traffic-related combustion on regional highways, or regional wildfires. As suggested in previous studies, the continuing decreases in PAH deposition to Siskiwit Lake are consistent with continuing reductions in emissions from power plants and vehicle emissions.1,5,7,10,12,13,42 Trends in Δ14C of Pyrogenic PAHs. Results from radiocarbon analysis of pyrogenic PAH in Siskiwit sediments corroborate a decreasing trend in deposition of PAHs derived from fossil fuel combustion and indicate an increase in the relative proportion of PAHs derived from biomass burning. Δ14C of pyrogenic PAHs in 1950 was −783‰. Assuming a maximum Δ14C value for biomass produced in 1950 of 225‰ as per Xu et al. (2012) to account for the impact of atmospheric nuclear weapons testing, a 14C mass balance as per eq 1 indicates that PAH inputs were 80% derived from fossil fuel sources at this time (Δ14Cfossil fuel assumed to be −1000‰) (Table 2).

Figure 5. Δ14C of pyrogenic PAH (diamonds), perylene (squares), and TOC (circles) versus depth for Siskiwit sediments. Dashed vertical line represents the approximate Δ14C of the modern atmosphere. Deepest sample (7−8 cm depth interval) is equivalent to 1955; uppermost sample (0−2 cm interval) is equivalent to 2001.

Δ14 CPAH = (Δ14 Cfossil fuel )(ffossil fuel ) + (Δ14 C biomass)(1 − ffossil fuel )

(1)

With decreasing depth in the core and assuming a decrease in biomass Δ14C to100‰ due to decreases in atmospheric Δ14C levels, this proportion systematically decreases to a calculated 44% contribution from fossil sources in the most recent sample in the year 2000. This estimate is not sensitive to the biomass Δ14C used to represent recent production at any given time point. Using a Δ14C of 50‰ decreases the proportion of fossil fuel inputs by 2% or less, and using a Δ14C of 225‰ for all samples increases the proportion of fossil fuel contribution by 6% or less. Thus, radiocarbon analysis provides evidence of a ca. 50% decrease in the contribution of fossil fuel combustion to pyrogenic PAH deposition to Siskiwit Lake over the 50 year period from 1950 to 2000. Using the proportions of fossil- and biomass-derived PAHs from the radiocarbon data and total pyrogenic PAH flues observed in Siskiwit sediments, the fluxes of PAHs related to each source were estimated (Table 2). The flux of PAHs related to fossil fuel combustion decreased as expected from ca. 17 ng cm−2 yr−1 in 1950 to ca. 3.8 ng cm−2 yr−1 in 2000. Concurrently, it is noteworthy that fluxes apportioned to biomass burning have remained at 6 ± 1.3 ng cm−2 yr−1 over the same time period. The consistency in the flux of biomass-

Table 1. Pyrogenic PAHs trended in Δ14C from −783‰ in 1950 to −388‰ in the uppermost sample. Concurrently, perylene Δ14C values ranged from −296‰ to −199‰ over the core with no systematic pattern. TOC Δ14C ranged from −12‰ at the top of the core to −115‰ at the deepest sample analyzed corresponding to 1950. As noted, the sample from 2 to 4 cm was lost during analysis, resulting in the gap in the sample set at this depth.

■

DISCUSSION Pyrogenic PAH Fluxes. The PAH flux profile from this study agrees well with those observed by McVeety and Hites (1988) at the same site for the period between 1920 and 1980. Furthermore, the timing of the maximum PAH deposition and the overall shape of the profile are also consistent with profiles from the northeastern United States from the Pettaquamscut River in Rhode Island,7 Lake Michigan,13 and L227 in the Experimental Lakes Area.11 This PAH deposition profile has

Table 1. Total PAH Concentrations, Pyrogenic PAH Flux, and Δ14C of PAH and TOC depth (cm)

equivalent year

pyro PAH ng g−1

pyro PAH ng cm−2 year−1

perylene ng g−1

Δ14C pyro PAH

Δ14C perylene

Δ14C TOC

−1 −4.5 −6.5 −7.5

2000 1978 1960 1950

590 999 1185 1037

9 19 22 22

331 1379 1633 1690

−388 −583 −690 −783

−289 −199 −196 −296

−12.7 −96.9 −115.6 −98.3

E

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Table 2. Calculated Pyrogenic PAH Fluxes and Fraction of Perylene derived from TOC depth (cm)

fraction fossil derived PAH (fossil vs atm)

PAH flux fossil source (ng cm−2 year−1)

PAH flux biomass sources (ng cm−2 year−1)

biomass PAH flux (−100‰) (ng cm−2 year−1)

biomass PAH flux (−300‰) (ng cm−2 year−1)

fraction perylene TOC source (%)

perylene mass TOC souce (ng g−1)

−1 −4.5 −6.5 −7.5

0.44 0.62 0.72 0.79

3.8 11.5 15.5 17.4

5.2 7.5 6.5 4.6

6 9 8 5

8 11 10 7

26 79 86 71

87 1088 1405 1202

first-order reaction rate constants. To ensure consistency the data from Gschwend et al. (1983) was replotted and the same reaction rate constant (0.012 year−1) and precursor concentration estimate (4.9 nmol g−1) was determined. Application of this program to the data in the current study yielded an estimated rate constant of 0.043 year−1 with an estimated precursor concentration of 7.2 nmol g−1, while the data from McVeety and Hites (1998) yielded a rate constant estimate 0.048 year−1 but an estimated precursor concentration of 3.2 nmol g−1. This similarity in rate constants from these two data sets supports the concept that this reaction has been occurring at a fairly constant rate in these sediments. The notable difference in precursor concentration estimates is a result of the fact that the fitting program back calculates the precursor concentration based on observed perylene concentrations assuming 100% conversion. Thus, it is not surprising that the more recent estimate estimates a higher precursor concentration. The rate constant estimated for Siskiwit Lake was higher than that observed by Gschwend et al. (1983) (0.012 year−1) and much higher than the estimate in recent work by Fan et al. (2011), who observed a fitted rate constant of 0.0024 year−1.20 The cause of these observed variations in production rates is not known, but in all cases production has been consistent with a first-order process. Perylene Δ14C Results. Δ14C values for perylene support the in situ production of the majority of the perylene from concurrent TOC. Perylene Δ14C levels fluctuate from −296 to −199‰ over the core, with the uppermost sample (Δ14C = −289‰) being within 100‰ of the pyrogenic PAH value (−388‰) at this depth and all perylene Δ14C values being depleted with respect to the Δ14C TOC values by 100−200‰. Applying an isotopic mass balance as per eq 1, the proportion of contribution from pyrogenic PAH deposition versus in situ production from TOC was assessed (Table 2). Using PAH and TOC Δ14C as end members, the proportion of perylene related to pyrogenic PAH deposition was estimated to decrease from 75% at the top of the core to 15−30% in the lower samples. This is equivalent to a total estimated in situ production of 87 ng g−1 of perylene observed in the upper sample to 1000−1400 ng g−1 of perylene observed in the deeper samples (Table 2). In order to assess whether the residual perylene was deposited as a pyrogenic PAH, the perylene present in excess of these estimates was converted to a flux. Results showed a generally decreasing trend with perylene representing 20−40% of pyrogenic PAHs deposited. It is difficult to assess the specific contribution of perylene that would be expected in this region, and there is limitation to the resolution of the radiocarbonbased estimate. Nevertheless, this component is small compared to the total mass of perylene, and these results demonstrate that the majority of perylene (70−85% in the deeper samples) can be explained by production from TOC. While this does not clarify the precise precursor molecule, it

derived PAHs to Siskiwit Lake and the fact that it is comparable to the pyrogenic PAH fluxes observed in the early 1900s by both our study and that of McVeety and Hites (1988) are indicative that this is in fact a background level of deposition related to regional wildfires. If this is indeed the case, an interesting corollary is the implication that there has been no significant trend in wildfire occurrence over the past 100 years in the region recorded in these sediments. If wildfires are indeed the source of this component of the PAH fluxes to Siskiwit Lake, it should be recognized that wildfires may not only burn recently fixed carbon but also release older carbon stored in peat or soil organic matter which would have older Δ14C values. Using a net biomass source Δ14C of −300‰, slightly more enriched than the most enriched pyrogenic PAH value of −388‰ observed in the upper sediments (Table 2), fossil PAH fluxes estimated by mass balance still comprised 70% (ca. 15 ng cm−2 yr−1) of the PAH flux in 1950 and decreased to ca. 13% (1 ng cm−2 yr−1) in the most recent sediments. Concurrently, biomass contributions to fluxes increased to 9 ± 2 ng cm−2 yr−1 over the time period. This extreme case estimate, which results in 90% of current PAH deposition being related to biomass burning in 2000, is considered unlikely. A mass balance using the predominant Δ14C of TOC in the core of −100‰ resulted in fluxes related to biomass burning that were the same as the original estimate within 1 standard deviation. These results demonstrate that the trend in contributions from fossil fuel combustion is robust. Further this sensitivity analysis indicates that while fluxes of biomass-derived PAHs could be as high as 11 ng/cm2/year in the most unlikely scenario, they are more likely in the range of 6−7 ng cm−2 yr−1. Perylene. The fact that the core from this study was obtained from the same location as a core reported on by McVeety and Hites (1988) but ∼20 years later provides the opportunity to directly assess the potential in situ production of perylene. Perylene production in these sediments over a period of 20 years is directly demonstrated by the offset between the curve from this study and the curve from McVeety and Hites (1988) (Figure 4). A 158−1300% increase in perylene concentrations was observed for sediments corresponding to those sampled by McVeety and Hites (1988) with the increase being greatest at the top of the core and diminishing with depth in the core. Over the 20 years between sampling events, perylene production, as determined by the difference in perylene concentration between the two cores, ranged from 850 to 1000 ng g−1 in sediments equivalent to the top of the core taken by McVeety and Hites (1980 and 1972, respectively) to 450 to 475 in the lowest corresponding samples (1920 and 1911, respectively). The shape of these curves and diminishing concentration increases with depth suggest that the reaction may be concentration dependent and thus could be described by simple kinetic models as per Gschwend et al. (1983). An iterative fitting program similar to that used by Gschwend et al. (1983) was used to determine F

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

(14) Mahler, B. J.; Van Metre, P. C.; Crane, J. L.; Watts, A. W.; Scoggins, M.; Williams, E. S. Coal-Tar-Based Pavement Sealcoat and PAHs: Implications for the Environment, Human Health, and Stormwater Management. Environ. Sci. Technol. 2012, 46 (6), 3039− 3045. (15) Van Metre, P. C.; Mahler, B. J. Contribution of PAHs from coaltar pavement sealcoat and other sources to 40 US lakes. Sci. Total Environ. 2010, 409 (2), 334−344. (16) Eakins, J. D.; Morrison, R. T. A new proceedure for determination of lead-210 in lake and marine samples. Int. J. Appl. Radiat. Isotopes 1978, 29, 531−536. (17) Kaluarachchi, J. J.; Cvetkovic, V.; Berglund, S. Stochastic analysis of oxygen- and nitrate-based biodegradation of hydrocarbons in aquifers. J. Contam. Hydrol. 2000, 41, 335−365. (18) Sun, P.; Blanchard, P.; Brice, K. A.; Hites, R. A. Trends in polycyclic aromatic hydrocarbon concentrations in the Great Lakes atmosphere. Environ. Sci. Technol. 2006, 40 (20), 6221−6227. (19) Venier, M.; Hites, R. A. Time Trend Analysis of Atmospheric POPs Concentrations in the Great Lakes Region Since 1990. Environ. Sci. Technol. 2010, 44 (21), 8050−8055. (20) Fan, C. W.; Shiue, J.; Wu, C. Y. Perylene dominance in sediments from a subtropical high mountain lake. Org. Geochem. 2011, 42 (1), 116−119. (21) Grice, K.; Lu, H.; Atahan, P.; Asif, M.; Hallmann, C.; Greenwood, P.; Maslen, E.; Tulipani, S.; Williford, K.; Dodson, J. New insights into the origin of perylene in geological samples. Geochim. Cosmochim. Acta 2009, 73 (21), 6531−6543. (22) Gschwend, P. M.; Hites, R. A. Fluxes of Polycyclic AromaticHydrocarbons to Marine and Lacustrine Sediments in the Northeastern United-States. Geochim. Cosmochim. Acta 1981, 45 (12), 2359− 2367. (23) Itoh, N.; Hanari, N. Possible precursor of perylene in sediments of Lake Biwa elucidated by stable carbon isotope composition. Geochem. J. 2010, 44 (3), 161−166. (24) Itoh, N.; Tamamura, S.; Kumagai, M. Distributions of polycyclic aromatic hydrocarbons in a sediment core from the north basin of Lake Biwa, Japan. Org. Geochem. 2010, 41 (8), 845−852. (25) Silliman, J. E.; Meyers, P. A.; Eadie, B. J.; Klump, J. V. A hypothesis for the origin of perylene based on its low abundance in sediments of Green Bay, Wisconsin. Chem. Geol. 2001, 177 (3−4), 309−322. (26) Silliman, J. E.; Meyers, P. A.; Ostrom, P. H.; Ostrom, N. E.; Eadie, B. J. Insights into the origin of perylene from isotopic analyses of sediments from Saanich Inlet, British Columbia. Org. Geochem. 2000, 31 (11), 1133−1142. (27) Aizenshtat, Z. Perylene and its geochemical significance. Geochim. Cosmochim. Acta 1973, 37, 559−567. (28) Gschwend, P. M.; Chen, P. H.; Hites, R. A. On the Formation of Perylene in Recent Sediments - Kinetic-Models. Geochim. Cosmochim. Acta 1983, 47 (12), 2115−2119. (29) Bakhtiari, A. R.; Zakaria, M. P.; Yaziz, M. I.; Lajis, M. N. H.; Bi, X. H.; Rahim, M. C. A. Vertical distribution and source identification of polycyclic aromatic hydrocarbons in anoxic sediment cores of Chini Lake, Malaysia: Perylene as indicator of land plant-derived hydrocarbons. Appl. Geochem. 2009, 24 (9), 1777−1787. (30) Kanke, H.; Uchida, M.; Okuda, T.; Yoneda, M.; Takada, H.; Shibata, Y.; Morita, M. Compound-specific radiocarbon analysis of polycyclic aromatic hydrocarbons (PAHs) in sediments from an urban reservoir. Nucl. Instrum. Methods Phys. Res. B 2004, 223−224, 545− 554. (31) Kumata, H.; Uchida, M.; Sakuma, E.; Uchida, T.; Fujiwara, K.; Tsuzuki, M.; Yoneda, M.; Shibata, Y. Compound class specific C-14 analysis of polycyclic aromatic hydrocarbons associated with PM10 and PM1.1 aerosols from residential areas of suburban Tokyo. Environ. Sci. Technol. 2006, 40 (11), 3474−3480. (32) Mandalakis, M.; Gustafsson, O.; Alsberg, T.; Egeback, A. L.; Reddy, C. M.; Xu, L.; Klanova, J.; Holoubek, I.; Stephanou, E. G. Contribution of biomass burning to atmospheric polycyclic aromatic

further supports that in situ production of perylene in anoxic sediments is being derived from TOC.

■

AUTHOR INFORMATION

Corresponding Author

*Phone 905-525-9140 (ext 26388); fax 905-546-0463; e-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS J. Kirby provided technical support at McMaster University. We thank Fan Yang (EC Burlington) for sediment core dating and Dan Walsh and Bruce Grey (Research Support Services, EC Burlington) for field logistical support and sediment sampling. We are thankful for funding to support this work from the Natural Science and Engineering Research Council of Canada (grants awarded to G.F.S.), an Ontario ERA award (to G.F.S.), and funding from Environment Canada, Great Lakes Action Plan.

■

REFERENCES

(1) Lima, A. L. C.; Farrington, J. W.; Reddy, C. M. Combustionderived polycyclic aromatic hydrocarbons in the environment - A review. Environ. Forensics 2005, 6 (2), 109−131. (2) Sauvain, J. J.; Duc, T. V.; Guillemin, M. Exposure to carcinogenic polycyclic aromatic compounds and health risk assessment for dieselexhaust exposed workers. Int. Arch. Occup. Environ. Health 2003, 76 (6), 443−455. (3) Tan, Y. L.; Heit, M. Biogenic and Abiogenic Polynuclear Aromatic-Hydrocarbons in Sediments from 2 Remote Adirondack Lakes. Geochim. Cosmochim. Acta 1981, 45 (11), 2267−2279. (4) Hites, R. A.; Laflamme, R. E.; Farrington, J. W. Sedimentary Polycyclic Aromatic-Hydrocarbons - Historical Record. Science 1977, 198 (4319), 829−831. (5) Elmquist, M.; Zencak, Z.; Gustafsson, O. A 700 year sediment record of black carbon and polycyclic aromatic hydrocarbons near the EMEP air monitoring station in Aspvreten, Sweden. Environ. Sci. Technol. 2007, 41 (20), 6926−6932. (6) Kannan, K.; Johnson-Restrepo, B.; Yohn, S. S.; Giesy, J. P.; Long, D. T. Spatial and temporal distribution of polycyclic aromatic hydrocarbons in sediments from Michigan inland lakes. Environ. Sci. Technol. 2005, 39 (13), 4700−4706. (7) Lima, A. L. C.; Eglinton, T. I.; Reddy, C. M. A high resolution record of pyrogenic polycyclic aromatic hydrocarbon deposition during the 20th century. Environ. Sci. Technol. 2003, 37, 53−61. (8) Simcik, M. F.; Eisenreich, S. J.; Golden, K. A.; Liu, S. P.; Lipiatou, E.; Swackhamer, D. L.; Long, D. T. Atmospheric loading of polycyclic aromatic hydrocarbons to Lake Michigan as recorded in the sediments. Environ. Sci. Technol. 1996, 30 (10), 3039−3046. (9) Van Metre, P. C.; Mahler, B. J. Trends in hydrophobic organic contaminants in urban and reference lake sediments across the United States, 1970−2001. Environ. Sci. Technol. 2005, 39 (15), 5567−5574. (10) Van Metre, P. C.; Mahler, B. J.; Furlong, E. T. Urban sprawl leaves its PAH signature. Environ. Sci. Technol. 2000, 34, 4064−4070. (11) Jeremiason, J. D.; Eisenreich, S. J.; Paterson, M. J. Accumulation and recycling of PCBs and PAhs in artificially eutrophied Lake 227. Can. J. Fish Aquat. Sci. 1999, 56 (6), 650−660. (12) McVeety, B. D.; Hites, R. A. Atmospheric Deposition of Polycyclic Aromatic-Hydrocarbons to Water Surfaces - a Mass Balance Approach. Atmos. Environ. 1988, 22 (3), 511−536. (13) Schneider, A. R.; Stapleton, H. M.; Cornwell, J.; Baker, J. E. Recent declines in PAH, PCB, and toxaphene levels in the northern Great Lakes as determined from high resolution sediment cores. Environ. Sci. Technol. 2001, 35 (19), 3809−3815. G

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

hydrocarbons at three European background sites. Environ. Sci. Technol. 2005, 39 (9), 2976−2982. (33) Mandalakis, M.; Gustafsson, O.; Reddy, C. M.; Li, X. Radiocarbon apportionment of fossil versus biofuel combustion sources of polycyclic aromatic hydrocarbons in the Stockholm metropolitan area. Environ. Sci. Technol. 2004, 38 (20), 5344−5349. (34) Sheesley, R. J.; Krusa, M.; Krecl, P.; Johansson, C.; Gustafsson, O. Source apportionment of elevated wintertime PAHs by compoundspecific radiocarbon analysis. Atmos. Chem. Phys. 2009, 9 (10), 3347− 3356. (35) Xu, L.; Zheng, M.; Ding, X.; Edgerton, E. S.; Reddy, C. M. Modern and Fossil Contributions to Polycyclic Aromatic Hydrocarbons in PM2.5 from North Birmingham, Alabama in the Southeastern U.S. Environ. Sci. Technol. 2012, 46 (3), 1422−1429. (36) Zencak, Z.; Klanova, J.; Holoubek, I.; Gustafsson, O. Source apportionment of atmospheric PAHs in the western balkans by natural abundance radiocarbon analysis. Environ. Sci. Technol. 2007, 41 (11), 3850−3855. (37) Reddy, C. M.; Pearson, A.; Xu, L.; McNichol, A. P.; Benner, B.; Wise, S. A.; Klouda, G.; Currie, L. A.; Eglinton, T. I. Radiocarbon as a tool to apportion the sources of polycyclic aromatic hydrocarbons and black carbon in environmental samples. Environ. Sci. Technol. 2002, 36, 1774−1782. (38) Mandalakis, M.; Zebuhr, Y.; Gustafsson, O. Efficient isolation of polyaromatic fraction from aliphatic compounds in complex extracts using dimethylformamide-pentane partitionings. J. Chromatogr. A 2004, 1041, 111−117. (39) Pearson, A.; McNichol, A. P.; Schneider, R. J.; von Reden, K. F. Microscale AMS 14C measurement at NOSAMs. Radiocarbon 1998, 40, 61−75. (40) Oldfield, F.; Appleby, P. G. Alternative Approach to PB-210 Based Sediment Dating. Geophys. J. R. Astron. Soc. 1978, 53 (1), 177− 177. (41) Baker, J. I.; Hites, R. A. Siskiwit Lake revisited: Time trends of polychlorinated dibenzo-p-dioxin and dibenzofuran deposition at Isle Royale, Michigan. Environ. Sci. Technol. 2000, 34 (14), 2887−2891. (42) Shen, H. Z.; Tao, S.; Wang, R.; Wang, B.; Shen, G. F.; Li, W.; Su, S. S.; Huang, Y.; Wang, X. L.; Liu, W. X.; Li, B. G.; Sun, K. Global time trends in PAH emissions from motor vehicles. Atmos. Environ. 2011, 45 (12), 2067−2073.

H

dx.doi.org/10.1021/es400272z | Environ. Sci. Technol. XXXX, XXX, XXX−XXX