Polycyclic Aromatic Hydrocarbons in Louisiana Rivers and Coastal ...

Environmental Forensics, 9:63–74, 2008 C Taylor & Francis Group, LLC Copyright ISSN: 1527–5922 print / 1527–5930 online DOI: 10.1080/15275920801888301

Polycyclic Aromatic Hydrocarbons in Louisiana Rivers and Coastal Environments: Source Fingerprinting and Forensic Analysis Javed Iqbal,1 Edward B. Overton,2 and David Gisclair3 1

W. A. Callegari Environmental Center, Louisiana State University AgCenter, Baton Rouge, LA, USA Department of Environmental Studies, Louisiana State University, Baton Rouge LA, USA 3 Department Louisiana Oil Spill Coordinator Office (LOSCO), Office of the Governor, Baton Rouge, LA, USA 2

Polycyclic aromatic hydrocarbons (PAHs), many of which are toxic and recalcitrant compounds, are ubiquitous in rivers and coastal environments. Anthropogenic introduction of these chemicals into the environment compromises the assessment of cleanup responsibility and environmental damage liability. Natural background and anthropogenic PAHs in Louisiana coast and major rivers were differentiated based on PAH profiles in samples selected from a pool of 3,540 samples collected over a 3-year period. Several groupings of 2- to 6-ring parent and their C1–C4 alkylated PAH homologs were quantified by gas chromatography/mass spectrometry. Sampling stations were delineated in terms of pyrogenic, petrogenic, and biogenic/diagenetic source PAHs. Fragmentograms indicated that petrogenic inputs generally dominated at more stations than pyrogenic and diagenetic inputs. Most of the results reflected multiple sources of contamination, as would be expected. Preexisting environmental forensic techniques were selected and applied to compare several different source differentiation and allocation methods to evaluate PAH sources in samples from a wide geographical coastal system influenced by myriad sources and complex mixing dynamics. This article covers diagnostic ratios and plots utilizing petroleum biomarker constituents, ratios within homologous PAH categories, pollution indices, and qualitative comparisons to reference profiles suspected as PAH sources. Keywords: environmental forensics, source fingerprinting, PAH analyte profiles, petrogenic, pyrogenic, diagenetic

Introduction Petroleum and petroleum-related activities are widespread throughout the world with consequent pollution, particularly in coastal and harbor environments. Most significant are oil spills in rivers and coastal environments, ranging from small leakages to larger accidents. Hence, environmental damages associated with such incidents warrant a comprehensive chemical characterization to defensibly determine the source(s), distinguish spilled oil from background hydrocarbons, and assess impacts on the ecosystem. Sources of Polycyclic Aromatic Hydrocarbons Polycyclic aromatic hydrocarbons (PAHs) in coastal environments are derived from oil spills, oil pollution, and transportation (petrogenic), combustion of organic matter and fossil fuels (pyrogenic), and chemical/biological transformation of natural Received 18 September 2006; accepted 31 August 2007. Address correspondence to Javed Iqbal, W. A. Callegari Environmental Center, Louisiana State University Agriculture Center, 1300 Dean Lee Drive, Baton Rouge, LA 70820, USA. E-mail: [email protected]

organic matter (diagenetic) or biological processes (biogenic) (Neff, 1979, 2002; Boehm, 2006). A typical crude oil may contain from 0.2% to more than 7% total PAHs (Neff, 1979, 2002). Most of the PAHs in petroleum are low molecular weight (LMW) containing two or three fused aromatic rings. High-molecularweight (HMW) PAHs, when present, usually are at low concentrations (Kerr et al., 1999). It has been reported that petrogenic PAHs are highly dominated by alkylated PAHs regardless of the petroleum source or degree of weathering (Boehm et al., 2001; Boehm, 2006; Hawthorne et al., 2006). PAH distribution patterns in pyrogenic PAHs, as mentioned elsewhere (Bjorseth, 1985; Neff, 1979), are characterized by dominance of unsubstituted compounds over their corresponding alkylated homologues and dominance of HMW 4- to 6-ring PAHs over LMW 2- to 3-ring PAHs. The m/z 252 group, containing 5-ring PAHs, is generally considered an indicator of combustion sources. However, within the m/z 252 group benzo[e]pyrene is associated with petroleum (Sporstol et al., 1983; Stout et al., 2002), and the diagenetic PAH system is frequently characterized by high concentrations of perylene, an unsubstituted 5-ringed PAH produced in subtidal sediments by a process known as early diagenesis (Wang et al., 1999). Perylene is also found in some crude oils; e.g., South Louisiana crude oil. 63

64

J. Iqbal et al.

Study Area The Louisiana coastal ecosystem supports diverse biological resources that are both commercially and ecologically important to Louisiana and the nation (Shirley et al., 2005). The ecosystem is also vulnerable to petroleum-related activities because the Gulf of Mexico houses the major infrastructure utilized in the production, refining, and transportation of oil and natural gas. Coastal Louisiana supports almost one third of the crude oil production in the United States, ranking first in the crude oil supply among all states (Cusimano and French, 2003). This article discusses environmental forensics and source fingerprinting aspects of PAHs in samples collected along the Louisiana coast and from major Louisiana rivers. The sample set is comprised of 3,540 samples collected at 1,180 stations in each of 3 years under the Louisiana Baseline Sampling and Analysis program. Each sample presented a unique fingerprint. This article discusses

some general trends in the data and specific parameters that can be used to differentiate between PAH sources, to characterize the source, and to fingerprint a few representative samples.

Material and Methods Sampling and Analysis The study area comprised the entire Louisiana coast along the shoreline of the northern Gulf of Mexico and major rivers in the state including the Mississippi and Red Rivers. Sampling stations also were located at Lake Pontchartrain, the Mississippi Delta region, Grand Isle, Chandeleur and Breton Islands, West Bay, Bay Adams, Barataria Bay, and Round Lake (Figure 1). Stations were chosen based on high oil spill probability rankings, defined as proximity to Gulf/inland shipping lanes, crude oil pipeline hazard points, and areas surrounding oil fields,

Figure 1. Sampling stations in coastal Louisiana and major Louisiana rivers. Fingerprint profiles of several representative samples are shown.

PAHs in Louisiana Rivers and Coastal Environments

refineries, major crude oil tank farms, and other fixed sources. The points were spaced approximately every 3 K, with closer spacing in areas with high probability rankings and wider spacing in areas with low risk of environmental contamination. Composite samples from the top 5 cm of 10-cm-diameter cores were collected three times over 3 years at 1,180 georeferenced points. In water locations sediment samples were taken from areas less than 3 m deep at the visually estimated mean tidal level. A vigorous sample extraction procedure based on previous studies (Short et al., 1996) involved a sequential 38-h tumbling with dichloromethane (DCM) as solvent. Briefly, samples were prepared for extraction by weighing approximately 15 g of wet sample, drying with anhydrous sodium sulfate, and adding 75 mL of DCM. The sample was placed on roller for 16 h followed by a second aliquot of 75-mL DCM, rolled for 6 h, and then a third aliquot of 75 mL was added and the solution rolled for another 16 h. The extract was filtered though anhydrous sodium sulfate and concentrated to 2 mL by rotary evaporation (water bath temperature set at 35◦ C). The extract was transferred to a labeled vial and concentrated to 200 µL using purified, compressed nitrogen. Hexane (4 mL) was added and the solution concentrated again to a final volume of 1 mL. The gas chromatography/mass spectrometry (GC/MS) was programmed at an initial temperature of 50◦ C (hold 3 min), ramped at 6◦ C/min to 120◦ C (hold 0.1 min), followed by 3◦ C/min to 200◦ C (hold 0.3 min), and then 12◦ C/min to 300◦ C (hold 15 min) for a total run time of 65 min. A quantitation method was developed for a selected-ion monitoring procedure using reference materials (South Louisiana Crude(SLC), Fisher Scientific International Inc., Fair Lawn, NJ, USA, Cat. No. SRS953). Target analytes included 34 parent and alkylated PAHs, saturated hydrocarbons (n-C9 to n-C35 carbon units), isoprenoids (pristane and phytane), 17α(H), 21β(H)hopane, and steranes (sum of a series of steranes, methylated sterane, diasterane, and aromatic sterane biomarkers m/z 217; Peters and Moldowan, 1993). This article describes only PAHs and hopane and steranes biomarker compounds. The target analytes provided a sample profile used to both classify contaminant sources and identify crude oil sources if significant contaminant levels were detected. A minimum detection limit (MDL) of 10 ngg−1 (ppb) for each analyte was achieved. Data Quality Assessment Each analytical batch included a procedural blank, certified calibration standards for all the analytes (ULTRAcheck, Ultra Scientific, North Kingstown, RI, USA, Cat. No. FRNH-068), a laboratory-fortified sample spiked with South Louisiana crude (containing the analytes of interest in known concentrations), and duplicate analysis. Reference SLC oil was analyzed every 40 to 50 samples. Since the same SLC oil was utilized during the 3year study period, variability was expected in its chemical composition, especially the LMW PAHs. Results were monitored for 60–140% recovery of internal standards (hexamethylbenzene, d10-anthracene, d12-benzo(a)anthracene), 40–130% recovery of surrogate standards (d8-naphthalene, d10-acenaphthene, d10phenanthrene, d12-chrysene, d12-perylene, o-terphenyl), and

65

40–130% recovery of matrix spike analytes. Laboratory results were subjected to rigorous review and validation (Gisclair and Iqbal, 2005). Forensics tools used included indices, e.g., fossil fuel pollution index (FFPI) (= [ naphthalenes (C0–C4) + diben zothiophenes (C0–C3) + 1/2 phenanthrenes (C0–C1) + phenanthrenes (C2–C4)]/ PAH) (Boehm and Farrington, 1984), diagnostic ratios and plots (Table 1), the presence of biomarkers, and fingerprint profiles. The source and weathering ratios (Table 1) were used for possible source identification because relative concentrations of specific petroleum constituents included in this ratio in different oil types are reported to be source-specific (Douglas et al., 1996). Results of principal components analysis (PCA) applied to statistically explain the suspected sources are given in a separate article (Iqbal et al., 2008).

Results and Discussion Non-alkylated parent PAHs, including most of the United States Environmental Protection Agency priority pollutant PAHs, were detected more frequently at the sites compared with detection of alkylated homologues. However, their mean concentrations were lower than those of their alkylated homologues. Figure 2 shows the mean concentrations and variability of all PAHs in the data set for coastal Louisiana, Louisiana rivers sediments, and SLC. Variability increased with increased alkylation. Figure 2 shows a general pattern of PAHs distribution in the sediment samples and reference SLC. Parent and substituted PAH distributions in the sediment samples were observed as C0 < C1< C2 < C3 < C4, whereas that of the reference SLC oil was observed as C0 < C1 < C2 < C3 > C4. The PAH distributions in the sediment samples predominantly included a mixture of petrogenic PAHs (alkyl>parent; lower 4 to 6 ring) and pyrogenic PAHs (sitespecific parent>alkyl; e.g., pyrenes; high 4 to 6 ring). Perylene, a product of diagenesis and also found in SLC, was the most frequent among HMW-PAHs in the sediment samples during the 3-year sampling interval. As noticed by other investigators (Neff, 1979; Douglas, 1996), the relative degree of individual PAH depletion in weathered sediments decreased with an increasing number of rings, and within a homologous series, decreased with increased alkylation, i.e., C0 < C1 < C2 < C3 < C4 (Figure 2A). Five- and 6ring HMW-PAHs appear accumulated in the sediments of coastal Louisiana and major rivers. Average concentrations of middle molecular weight (MMW)-PAHs were greater than those of 2to 3-ring LMW-PAHs. Anthracene and fluoranthene are reported to be less stable thermodynamically than their isomers, phenanthrene and pyrene, respectively (Baumard et al., 1998; Neff et al., 2005). Anthracene and fluoranthene are produced during rapid, hightemperature pyrosynthesis but are less favored to persist during the slow organic diagenesis leading to the generation of fossil fuels. Phenanthrene, although common to both pyrogenic and petrogenic systems, is an important petroleum-source PAH usually

66

J. Iqbal et al.

Table 1. Environmental forensics and diagnostic tools utilized in this study Environmental forensic tools used and their description Fossil fuel pollution index (FFPI) = [ naphthalenes (C0 –C4 ) + dibenzothiophenes (C0 –C3 )+ 1/2 phenanthrenes phenanthrenes (C2 –C4 )]/ (C 0 –C1 ) + PAH)) = [ naphthalenes (C0 –C4 ) + dibenzothiophenes (C0 –C3 )+ 1/2 phenanthrenes (C 0 –C1 ) + phenanthrenes (C2 –C4 )]/ PAH) Phenanthrene/anthracene (P/A) Fluoranthene/pyrene (Fl/Pyr) Double ratio plot C2dibenzothiophenes/C2-phenanthrenes (C2D/C2P) versus C3dibenzothiophenes/C3-phenanthrenes (C3D/C3P) (Unalk36Ring) plotted against (Alk5Target): anthracene, fluoranthene, pyrene, benzo(a)anthracene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(e)pyrene, benzo(a)pyrene, perylene, indeno(1,2,3-cd)pyrene, dibenzo[a, h]anthracene, benzo(g, h, i)perylene plotted against C1-, C2-, C3-, and C4-naphthalene, fluorene, dibenzothiophene, phenanthrene, and chrysene

Application

Reference

Petrogenic/pyrogenic polycyclic aromatic hydrocarbon (PAHs)

Boehm and Farrington, 1984

Petrogenic/pyrogenic PAHs Petrogenic: P/A < 5 Pyrogenic: P/A > 5 Petrogenic/pyrogenic PAHs Petrogenic: Fl/Pyr < 1.23 based on the reference oil used Pyrogenic: Fl/Pyr > 1.23 Petrogenic source differentiation, and sulfurto–non-sulfur ratio

Neff et al., 2005

Differentiate petrogenic and pyrogenic PAHs

Wang et al., 1998; Wang and Fingas, 1999

found in higher amounts in refined oil products than in crude oil (Baumard et al., 1998; Wang and Fingas, 1999). A diagnostic ratio of phenanthrene to anthracene (P/A) is therefore used to differentiate pyrogenic and petrogenic PAHs assemblages, where a value less than 5 is reported to indicate pyrogenic PAHs and a value greater than 5 shows petrogenic origin (Neff et al., 2005). In another study, the P/A ratio is reported to be 3 to 26 in petrogenic dominant sediment samples (Colombo et al., 1989). We found P/A ratio values between 4.11 and 102.6 in 75% of the reference SLC or sediments spiked with SLC, whereas mean phenanthrene and anthracene concentrations were found 591 ± 57 ngg−1 and 55 ± 11 ngg−1 , respectively. Significant variability in P/A ratio in SLC and sediment spiked with SLC is partly analytical; however, this also reflected matrix effects on recovery of analytes. Based on results of 1,728 sediment samples where both phenanthrene and anthracene were detected above the MDLs, approximately 51% of the samples have a P/A ratio less than 4.11, indicating pyrogenic-dominant assemblages, whereas 49% indicated petrogenic-dominant PAHs. Similarly, the fluoranthene to pyrene ratio (Fl/Pyr) is also reported to differentiate between sediment PAH assemblages containing primarily pyrogenic or petrogenic inputs (Neff et al., 2005). Ratios exceeding a value of one indicate pyrogenic-dominant assemblages, whereas values substantially less than one indicate

Colombo et al., 1989; Neff et al., 2005 Boehm et al. 1997; Burns et al.,1997; Overton, 2004; Page et al., 1996; Stout et al., 2002

petrogenic-dominant PAHs assemblages. This ratio is reported to be 0.6 to 1.4 in Kuwait and Louisiana crude and No. 2 fuel oil (Colombo et al., 1989). All of the reference SLC and approximately 85% of the samples spiked with SLC exhibited Fl/Pyr ratios in the range of 0.50 to 1.23. Based on results of the reference SLC, approximately 50% of the sediment samples had Fl/Pyr < 1.23, suggesting dominant petrogenic inputs and approximately 50% of samples exceeded 1.23 suggesting non-petrogenic input. Among the 50% petrogenic-dominant samples, hopane was detected in 82% and steranes were detected in 58% of the samples at levels above the MDLs. Both hopane and sterane analytes were detected concurrently in approximately 52% of the petrogenicdominant samples. These biomarker analytes correlated significantly (r2 = 0.75, p < 0.05) in the sediment samples. Hopane is a common constituents of crude oil, resistant to biodegradation and considered a quite reliable refractory constituent due to the fact that it is neither produced nor destroyed over significant periods of oil weathering (Prince et al., 1994), although it is known to be biodegradable in some environments. Hopane and sterane biomarkers are reported to be among the most effective internal marker compounds because their linked alicyclic structures provide few pathways for biotransformation (Atlas, 1981; Whittaker et al., 1997). Hopane concentrations in the sediment samples ranged between 0.01 µgg−1 to 3.81 µgg−1 .


67

Figure 2. Polycyclic aromatic hydrocarbon (PAH) fingerprint and variability (environmental and/or analytical); A) Coastal Louisiana sediments: variability increased with increased alkylation; B) fingerprint of reference South Louisiana crude. Arrows indicate the general trend in alkylated PAH homologues.

A double-ratio plot of C2-dibenzothiophene/C2phenanthrene (C2D/C2P) and C3-dibenzothiophene/C3phenanthrene (C3D/C3P) has been used to distinguish sediment PAHs from crude oil released in the Exxon Valdez oil spill at Prince William Sound (Page et al., 1996; Boehm et al., 1997; Burns et al., 1997; Stout et al., 2002). Both of these ratios are reported to be fairly stable over time and sediment depth (Page et al., 1995). The C3D/C3P ratio has also been defined as the petrogenic source ratio and sulfur to non-sulfur ratio (Overton, 2004). Dibenzothiophenes usually are considered primarily petrogenic (Neff et al., 2005). In this study, a double-ratio plot of C2D/C2P versus C3D/C3P for the sediment samples was overlaid on that of the reference SLC oil (Figure 3). The plot indicates higher values for SLC along with some of the

sediment samples, indicating possible crude oil constituents in coastal sediments. C2D and C2P significantly correlated in the sediment samples (r2 = 0.73, p < 0.05) and SLC (r2 = 0.88, p < 0.05). Similarly, C3D and C3P significantly correlated in the sediment samples (r2 = 0.72, p < 0.05) and SLC (r2 = 0.83, p < 0.05). Wang et al. (1998) utilized the ratio of the total priority unsubstituted 3- to 6-ring PAHs (abbreviated here as Unalk36Ring) plotted on the x-axis against the total of 5 target alkylated PAHs (Alk5Target) homologues on the y-axis to discriminate between petrogenic and pyrogenic PAHs. Unalk36Ring thus contains ANT, FLUORANT, PYR, BENZ, BBFL, BKFL, BEP, BAP, PERYL, INDENO, DIBENZANT, BENZOP, while Alk5Target contains C1N–C4N, FLUOR, DBT, PHENAN, and CHRYS

68

J. Iqbal et al.

Figure 3. Double-ratio plot of C2–dibenzothiophenes/C2–phenanthrenes (C2D/C2P) and C3–dibenzothiophenes/C3–phenanthrenes (C3D/C3P) for sediment samples, reference South Louisiana crude oil (SLC), and sediments spiked with SLC. Data points for overlapping samples and reference SLC oil are encircled indicating SLC oil as a possible petrogenic source. Note: Reference sediments cover multiple analysis of the same SLC oil during the 3-year study period.

(Table 2 lists abbreviations for compounds). Petrogenic PAHs are dominated by Alk5Target homologues, whereas pyrogenic PAHs are dominated by Unalk36Ring compounds. A ratio of these two parameters is reported to have good consistency with little interference from natural concentration fluctuation in individual compounds, long-term weathering, and biodegradation (Wang and Fingas, 1999). A histogram of this ratio showed that 51.87% of the samples were dominated by Alk5Target homologues showing petrogenic-dominant PAHs and 48.13% of the sediment samples were dominated by Unalk36Ring. The FFPI is also used to discern pyrogenic and petrogenic PAHs. FFPI is close to zero for mostly pyrogenic PAH assemblages, whereas for sediments containing significant amounts of fossil-fuel PAH constituents, the FFPI is close to one (Boehm, 1984). FFPI values in our study for reference SLC oil ranged between 0.60 and 0.99 and between 0.05 and 0.95 for sediment samples. Approximately 66% of the sediment samples have FFPI values less than 0.50, while 77.46% of the sediment samples have FFPI value less than 0.60. This therefore suggests that at least 34% of the samples contain significant amounts of fossil fuel constituents. Approximately 23% of the samples coincide with the range exhibited by reference SLC.

Mixing-Model Calculations Boehm et al. (1998, 2006) calculated pyrogenic PAHs as the sum of the concentrations of 3-, 4-, 5-, and 6-ring parent PAHs analytes except perylene (equivalent to the sum of ANT, PHENAN, FLUORANT, PYR, BENZ, CHRYS, BBFL, BKFL, BEP, BAP, INDENO, DIBENZANT, BENZOP in Table 2). Total petrogenic PAHs were calculated as the total PAHs minus pyrogenic PAHs

Table 2. List of polycyclic aromatic hydrocarbons and biomarker analytes and their abbreviations Naphthalene (NAPH) C1-Naphthalene (C1N) C2-Naphthalene (C2N) C3-Naphthalene (C3N) C4-Naphthalene (C4N) Fluorene (FLUOR) C1-Fluorene (C1F) C2-Fluorene (C2F) C3-Fluorene (C3F) Dibenzothiophene (DBT) C1-Dibenzothiophene (C1D) C2-Dibenzothiophene (C2D) C3-Dibenzothiophene (C3D) Phenanthrene (PHEN) Anthracene (ANT) C1-Phenanthrene (C1P) C2-Phenanthrene (C2P) C3-Phenanthrene (C3P) Fluoranthene (FLUORANT) Pyrene (PYR) C1-Pyrene (C1PY) C2-Pyrene (C2PY) Benzo[a]Anthracene (BENZ) Chrysene (CHRYS) C1-Chrysene (C1C) C2-Chrysene (C2C) Benzo[b]Fluoranthene (BBFL) Benzo[k]Fluoranthene (BKFL) Benzo[e]pyrene (BEP) Benzo[a]pyrene (BAP) Perylene (PERYL) Indeno(1,2,3-cd)pyrene (INDENO) Dibenzo[a,h]anthracene (DIBENZANT) Benzo[g,h,i]perylene (BENZOP) 17α(H),21β(H)-Hopane (HOPANE) Steranes (STERANE)


69

Figure 4. Pyrogenic, petrogenic, and diagenetic polycyclic aromatic hydrocarbon (PAH) contribution estimated by mixing model calculations (Boehm et al., 1998) in major parishes/subdivisions of coastal Louisiana.

and minus perylene. Perylene is a biogenic PAH reported as a product of early diagenesis in sediments, so its concentration is used as the diagenetic component. These mixing model calculations were applied to estimate pyrogenic, petrogenic, and diagenetic source PAH loads in coastal Louisiana and major rivers, reference SLC oils, and other reference profiles from previous studies. Figure 4 summarizes results of these calculations for all the sediment samples. Results indicated that the petrogenic PAH contribution in coastal Louisiana and major Louisiana rivers averaged 51%, pyrogenic PAHs 36%, and diagenetic/biogenic 14%. Based on this model, petrogenic PAHs

in reference SLC oils averaged 98%. Approximately 50% of the samples showing pyrogenic source PAHs using diagnostic ratios and indices may also incorporate PAHs from diagenetic or biogenic sources. Sampling stations dominated by pyrogenic input (>50%) were characterized by industrial activities and urban runoff; e.g., New Orleans. High petrogenic profiles (>50%) were observed in areas with high oil and gas production, transportation and storage activities; e.g., the lower Mississippi Delta region (52%), which contributes approximately 18% of oil production in the state (Louisiana Mid-Continent Oil and Gas Association [LMOGA], 1999). Diagenetic PAHs (i.e., perylene)

70

J. Iqbal et al.

dominated stations with riparian sediments; e.g., 32% in Red River.

Delta and the reference SLC oil based on the similarity of their FFPI values and relative concentrations of hopane and sterane biomarkers (Figure 5D).

Fingerprint Analysis Samples selected for fingerprint analysis and source identification represented a major segment of an area (parish) or a region. Fingerprint profiles of the suspected source and the samples were examined graphically by overlaying their PAH concentration profiles to display their relative composition. Diagnostic ratios and PCA was then used in identifying a probable match. PCA is not discussed in this article but in a companion manuscript (Iqbal et al., 2008). Fingerprint concentrations were also examined for petrogenic, pyrogenic, and diagenetic source contributions using the mixing model calculations mentioned previously. Figure 1 shows some of the PAH profiles. Petrogenic-Dominant Stations High concentrations of alkylated naphthalene and phenanthrene in an East Baton Rouge sediment sample (sample ID#300913912) were attributed to petrogenic sources (66% on mixing model calculations; Figure 1). East Baton Rouge is reported to contribute approximately 3% of the oil production activities in the state (LMOGA, 1999). Pyrogenic and diagenetic PAHs at this station were calculated to be 29 and 5%, respectively. As another example of petrogenic origin, the fingerprint of a sediment sample collected at an island in the Gulf of Mexico (sample ID#290914621) was compared with that of a reference background offshore sediment (Burns et al., 1997). Both are characterized by bell-shaped profiles for alkylated PAHs, high concentrations of naphthalenes and phenanthrenes, lower concentrations of HMW PAHs, as well as high FFPI values (Figure 5A). High concentrations of hopane and the presence of sterane biomarkers in the sample corroborate petrogenic dominance at the station. Hopane and sterane concentrations in the reference sediment were not available. Another sample collected at a wildlife refuge (sample ID#290921941) resembles that of the SLC reference oil, characterized by bell-shaped patterns of alkylated PAH, FFPI values of 0.96, >98% petrogenic PAH input on mixed model calculations, and the presence of hopane and sterane biomarkers. Alkylated naphthalenes indicated fresh oil contamination while the absence of HMW PAHs and perylene indicated insignificant pyrogenic or biogenic input (Figure 5B). Figure 5C shows weathering process that occurred during the 3 years of sampling at this originally petrogenic station (Plaquemines, ID#290895642). The Fl/Pyr ratio is not significantly affected, whereas the FFPI value changed with a decrease in LMW PAHs and either a build-up of HMW PAHs or a subtle shift in the proportion of LMW and of HMW PAHs. The PAH profile of a primarily petrogenic sample (Station ID 290895511), collected at approximately 60 m from an oil tank and well, was comparable to that of the SLC oil analyzed in the second year of the study. An interesting correlation exists between this petrogenic-dominant station in the lower Mississippi

Pyrogenic-Dominant Stations The fingerprint of a sample (Station ID 290921311) collected at soft marsh drainage and a swamp embankment in the vicinity of a telephone pole resembled that of a telephone pole creosote (Rabalais, et al., 1998). This observation was also based on the FFPI and Fl/Pyr ratios. Naphthalenes were mostly weathered, but the presence of hopane indicated petrogenic input at the station in addition to the dominant pyrogenic input (Figure 6A). A sample collected on the southern bank of Lake Catherine (Station ID 300895911) matches fairly well with reference pyrogenic habitation soot (Neff et al., 2005). Both the sample and the reference PAH profile showed similar Fl/Pyr ratios (1.05 and 1.04, respectively) and low FFPI values (0.22 and 0.14, respectively). The sample is characterized by an absence of dibenzothiophenes and fluorenes, which are primarily petrogenic compounds (Neff et al., 2005), and the presence of perylene indicates some biogenic PAH input at the station (Figure 6B). The fingerprint of another pyrogenic sample (Station ID 290900125), collected on the side of an access road to a pipeline company, was superimposed on that of a most probable reference creosote contaminated sediment (Neff et al., 2005). Both were characterized by similar pyrogenic and petrogenic compositions, Fl/Pyr ratios, FFPI values, and supported by a cluster in a PCA plot (not shown). The station also indicates recent co-contamination with LMW PAHs of petrogenic origin (Figure 6C). Biogenic/Diagenetic Dominant Stations A representative sample collected in Lafourche Parish (Station ID 290902031) in a marsh interior was characterized by 84% diagenetic PAHs, that is, perylene, a product of biogenic and diagenetic processes in sediments of rivers, lakes, anoxic marine environments, and swamps (Figure 1; Boehm, 1997; Venkatesan, 1988). Stations with Mixed Petrogenic and Pyrogenic Input Most of the stations in the lower Mississippi Delta region indicated multiple sources of contamination varying from extreme petrogenic to extreme pyrogenic inputs (Figure 6D). This area is an operational center for the offshore oil and gas industry (5% of the wells in Louisiana are located in this area [LMOGA, 1999]) as well as serving as a sink for a significant influx of contaminants from the Mississippi River. A sample collected at a station (Station ID 290890111, St. Barnard) in the vicinity of a sewage treatment plant and a docking facility exhibited 49 and 48% petrogenic and pyrogenic PAHs, respectively (Figure 1). Oil activities and urban runoff are the suspected sources of contaminants at this station, also characterized by a value of 0.84 for the Unalk36Ring/Alk5Target ratio. The PAH profile of a


71

Figure 5. Fingerprint profiles of selected samples matched with those of the reference profiles of known origin. A) A probable match of a sample with petrogenic origin to a reference background offshore sediments (Burns et al., 1997) with both characterized by bell-shaped profiles and high concentrations of naphthalenes. B) A petrogenic-dominant sample match with a sediment sample spiked with South Louisiana crude oil (SLC), used as a matrix spike. C) Weathering process in an originally petrogenic station during 3 years; fluoranthene-to-pyrene ratio (Fl/Pyr ratio) is not significantly affected; fossil fuel pollution index (FFPI) values changed with a decrease in low-molecular-weight polycyclic aromatic hydrocarbon (LMW PAH) and build-up of high-molecular-weight (HMW) PAH. D) A petrogenic-dominant sample collected approximately 60 m from oil tank and well, matched with a reference SLC oil (analyzed in year 2). Total polycyclic aromatic hydrocarbon (TPAH). Note: Table 2 lists abbreviations for PAH and biomarker analytes.

72

J. Iqbal et al.

Figure 6. Fingerprint profiles of selected samples matched with those of the reference profiles of known origin. A) Fingerprint profile of a pyrogenicdominant sample and suspected pyrogenic source telephone pole creosote (Rabalais et al., 1998). B) Probable match of a pyrogenic dominant sample with reference pyrogenic habitation soot (Neff et al., 2005). C) Fingerprint of another sample superimposed on a most probable reference sediment heavily contaminated with creosote (Neff et al., 2005). D) Example of mixed input at a station in two sampling years showing petrogenic input suspected a fresh diesel reference (Roberts, 1999) in year 1, and pyrogenic input matched an atmospheric dust reference (Burns et al., 1997) in year 2. Fl/Pyr, fluoranthene-to-pyrene ratio; FFPI, fossil fuel pollution index; high molecular weight, HMW; low molecular weight, LMW; middle molecular weight (MMW); PAH, polycyclic aromatic hydrocarbon; total polycyclic aromatic hydrocarbon (TPAH). Note: Table 2 lists abbreviations for PAH and biomarker analytes.


representative sample collected at an industrial spillway (Station ID 300936223, Calcasieu) was suspected to originate primarily from pyrogenic sources (Figure 1). The station location is in an area with nearby industrial and oil activities. Its FFPI value (0.16) indicated predominantly pyrogenic input. However, profiles of LMW and MMW PAHs indicated that the station was contaminated with a mixture of both pyrogenic and petrogenic PAHs. The station was enriched with HMW 4- to 6-ring PAHs. High levels of fluoranthene, a high ratio value of 6.25 for the Unalk36Ring/Alk5Target ratio, and an alkylated chrysene distribution indicate a typical pyrogenic pattern. Brenner et al. (2002) noted that weathering causes pyrogenic products like creosote to be increasingly dominated by 4- to 6-ring PAHs, producing a pattern very similar to urban runoff. Mobility of PAHs also decreases with an increase in the number of rings due to hydrophobicity and preference for the environmental matrix (Murphy and Brown, 2005). Therefore, to further speciate pyrogenic PAHs sources at stations affected by industrial, urban, or other mixed inputs, analytical tools such as PCA coupled with more specific reference PAHs profiles have been very helpful. This point is beyond the scope of this article but does appear in Iqbal et al. (2008).

Conclusion Results indicated that most of the PAHs in coastal Louisiana and major rivers originated from anthropogenic sources averaging approximately 50% from petroleum and petroleum products. Pyrogenic contribution was found to average approximately 36%, whereas natural diagenetic/biogenic processes contributed only 14%. Environmental forensics and source fingerprinting techniques including diagnostic ratios, mixing model calculations, and concentration fingerprint profiling indicated that the generalized sources of contamination could successfully be identified. Detailed information like fragmentograms and timely sampling can add to give defensible source allocations. This current study indicated that a few representative samples could be successfully linked to possible petrogenic and/or pyrogenic sources including South Louisiana crude oil and creosote. PCA, discussed in a separate article, is also a useful tool to distinguish PAHs of different origin.

Acknowledgement The Louisiana Oil Spill Coordinator’s Office (LOSCO), Office of the Governor, provided financial support for this study. The authors are very grateful to the reviewers of this article for their valuable contribution.

References Atlas, R. M. 1981. Microbial degradation of petroleum hydrocarbons: An environmental perspective. Microbiology Reviews 45: 180–209. Baumard, P., Budzinski, H., and Garrigues, P. 1998. Polycyclic aromatic hydrocarbons in sediments and mussels of the western Mediterranean Sea. Environmental Toxicology and Chemistry 17: 765–776.

73

Bjorseth, A. 1985. Sources of emissions of PAH. In Handbook of Polycyclic Aromatic Hydrocarbons. New York: Marcel Dekker, 1–20. Boehm, P. D. 2006. Polycyclic aromatic hydrocarbons. In Environmental Forensics, eds., Morrison, R. D., and Murphy, B. L. MA: Elsevier Academic Press. Boehm, P. D., Douglas, G. S., Burns, W. A., et al. 1997. Application of petroleum hydrocarbon chemical fingerprinting and allocation techniques after the Exxon Valdez oil spill. Marine Pollution Bulletin 34: 599– 613. Boehm, P. D., and Farrington, J. W. 1984. Aspects of the polycyclic aromatic hydrocarbon geochemistry of recent sediments in the Georges Bank Region. Environmental Toxicology and Chemistry 18: 840– 845. Boehm, P. D., Page, D. S., Burns, W. A., Bence, A. E., Mankiewiez, P. J., and Brwon, J. S. 2001. Resolving the origin of the petrogenic hydrocarbon background in Prince William Sound, Alaska. Environmental Toxicology and Chemistry 35: 471–479. Boehm, P. D., Page, D. S., Gilfillan, E. S., Bence, A. E., Burns, W. A., and Mankiewicz, P. J. 1998. Study of the fates and effects of the Exxon Valdez oil spill on benthic sediments in two bays in Prince William Sound, Alaska — I. Study design, chemistry, and source fingerprinting. Environmental Science and Technology 32: 567–576. Brenner, R. C., Magar, V. S., Ickes, J. A., et al. 2002. Characterization and fate of PAH-contaminated sediments at the Wyckoff/Eagle Harbor Superfund site. Environmental Science and Technology 36(12): 2605– 2613. Burns, W. A., Mankiewicz, P. J., Page, D. S., and Parker, K. R. 1997. A principal-component and least-squares method for allocating polycyclic aromatic hydrocarbons in sediment to multiple sources. Environmental Toxicology and Chemistry 16(6): 1119–1131. Colombo, J. C., Pelletie, E., Brochu, C., and Khall, M. 1989. Determination of hydrocarbon sources using n-alkane and polyaromatic hydrocarbon distribution indexes. Case study: Rio de La Plata Estuary, Argentina. Environmental Science and Technology 7(23): 888–894. Cusimano, C. V., and French, T. M. 2003. Developing a Louisiana energy policy. In Proceedings of the 50th Mineral Law Institute, Baton Rouge, LA: Louisiana State University, Law Center. (www.dnr.louisiana. gov/sec/execdiv/TECHASMT/presentations/MineralLawInstituteFinal 040303.pdf) Douglas, S., Gregori, A., Bence, A. E., Prince, R. C., McMillen, S. J., and Butler, E. R. 1996. Environmental stability of selected petroleum hydrocarbon source and weathering ratios. Environmental Science and Technology 30: 2332–2339. Gisclair, D., and Iqbal, J. 2005. A three-year baseline sampling and analysis study of petroleum pollutants in Coastal Louisiana. In Proceedings: 2005 International Oil Spill Conference Institute. American Petroleum Institute, Publications, Englewood, CO. Public No. 4710, paper #273. Hawthorne, B., Miller, D. J., and Kreittinger, J. P. 2006. Measurement of total polycyclic aromatic hydrocarbon concentrations in sediments and toxic units used for estimating risk to benthic invertebrates at manufactured gas plant sites.Environmental Toxicology and Chemistry 25(1): 287– 296. Hunt, J. M. 1995. Petroleum Geochemistry and Geology. New York: WH Freeman and Company. Iqbal, J., Overton, E. B., and Gisclair, D. 2008. PAH Sources in Louisiana Rivers and Coastal Environments: Principal Component Analysis. Environmental Forensics 9: 3 (Accepted). Kerr, J. M., Melton, H. R., McMillen, S. J., Magaw, R. I., and Naughton, G. 1999. Polyaromatic hydrocarbon content in crude oils around the world. SPE-52724. In Proceedings: 1999 SPE/EPA Exploration and Production Environmental Conference (March 1–3, 1999, Austin, TX). Richardson, TX: Society of Petroleum Engineers. LMOGA (Louisiana Mid-Continent Oil and Gas Association). 1999. Louisiana Mid-Continent Oil and Gas Association: Louisiana Oil and Gas Parish Activity Profile—1999. Available at: http://www.lmoga.com (accessed January 2006). Murphy, B. L., and Brown, J. 2005. Environmental forensics aspects of PAHs from wood treatment with creosote compounds. Environmental Forensics 6: 151–159.

74

J. Iqbal et al.

Neff, J. M. 1979. Polycyclic Aromatic Hydrocarbons in the Aquatic Environment. Sources, Fates and Biological Effects. Barking, Essex, UK: Applied Science. Neff, J. M. 2002. Bioaccumulation in Marine Organisms. Effects of Contaminants from Oil Well Produced Water. Amsterdam, The Netherlands: Elsevier. Neff, J. M., Stout, S. A., and Gunster, D. G. 2005. Ecological risk assessment of polycyclic aromatic hydrocarbons in sediments: Identifying sources and ecological hazard. SETAC Integrated Environmental Assessment and Management 1(1): 22–33. Overton, E. B., Ashton, B. M., and Miles, M. S., 2004. Historical polycyclic aromatic and petrogenic hydrocarbon loading in Northern Central Gulf of Mexico shelf sediments. Marine Pollution Bulletin 49: 557–563. Page, D. S., Boehm, P. D., Douglas, G. S., and Bence, A. E. 1995. Exxon Valdez oil spill: Fate and effects in Alaskan waters. In ASTM Special Technical Publication 1219, eds., Wells, P. G. Butler, J. N., and Hughes, J. S. Philadelphia: American Society for Testing and Materials, 41–83. Page, D. S., Boehm, P. D., Douglas, G. S., Bence, A. E., Burns, W. A., and Manciewicz, P. J. 1996. The natural petroleum hydrocarbon background in subtidal sediments of Prince William Sound, Alaska, USA. Environmental Toxicology and Chemistry 15: 1266–1281. Peters, K. E., and Moldowan, J. M. 1993. The Biomarkers Guide. Engelwood Cliffs, NJ: Prentice-Hall. Prince, R. C., Elmendorf, D. L., Lute, J. R., et al. 1994. 17α(H),21β(H)Hopane as a conserved internal marker for estimating biodegradation of crude oil. Environmental Science and Technology 28: 142–145. Rabalais, N. N., Smith, L. E., Henry, C. B., Roberts, P. O., and Overton, E. B. 1998. Long-term effects of contaminants from OCS-produced water discharges at Pelican Island facility, Louisiana. In OCS Study MMS 98–0039. New Orleans, LA: U.S. Department of Interior, Minerals Management Service, Gulf of Mexico OCS Regional Office. Roberts, P. O. 1999. Perylene as an Indicator of Ecological/Environmental Changes. Master’s Thesis, Louisiana State University, Baton Rouge, LA.

Shirley, L., Wooddell, G., Hangleman, R., Gramling, R., and Farris, M. T. 2005. At risk: The human, community and infrastructure resources of coastal Louisiana. Journal of Coastal Research 44(90): 111. Short, J. W., Jackson, T. L., Larsen, M. L., and Wade, T. L. 1996. Analytical methods used for the analysis of hydrocarbons in crude oil, tissues, sediments, and seawater collected for the natural resources damage assessment of the Exxon Valdez oil spill. In Proceedings: Exxon Valdez Oil Spill Symposium, eds., Rice, S. D., Spies, R. B. Wolfe, D. A., and Wright, B. A. Bethesda, MD: American Fisheries Society, 140–148. Sporstol, S., Gjos, N., Lichtenthaler, R. G., et al. 1983. Source-identification of aromatic hydrocarbons in sediments using GC/MS. Environmental Science and Technology 17: 282–286. Stout, S. A., Uhler, A. D., and Boehm, P. D. 2002. Recognition and allocation among multiple sources of PAH in urban sediments. Environmental Claims Journal 13: 141–158. Venkatesan, M. I. 1988. Occurrence and possible sources of perylene in marine sediments—A review. Marine Chemistry 25: 1–27. Wang, Z., and Fingas, M. 1999. Identification of the source(s) of unknown spilled oils. In Proceedings: International Oil Spill Conference. American Petroleum Institute, Publications, Englewood, CO, Publ. 4686, paper #162. Wang, Z., Fingas, M., Landriault, M., and Sigouin, L. 1999. Development in fingerprinting analysis of petroleum hydrocarbons. In Proceedings: International Oil Spill Conference. American Petroleum Institute, Publications, Englewood, CO, Publ. 4686, paper #461. Wang, Z. D., Fingas, M., Sigouin, I., et al. 1998. Quantitative characterization of PAH in burn residue and soot samples and differentiation of pyrogenic PAH from petrogenic PAH—the 1994 Mobile burn study. In Proceedings. 21st Arctic and Marine oil Spill Program (AMOSP). Technical Seminar. Ottawa, ON, Canada: Environment Canada, 673–704. Whittaker, M. D., Simon, J. T., and Pollarad, S. J. T. 1997. Performance assessment of source correlation and weathering indices for petroleum hydrocarbons in the environment. Environmental Toxicology and Chemistry 16(6): 1149–1158.