The PAH composition of surface sediments from Stagnone coastal lagoon, Marsala (Italy)

Marine Chemistry 99 (2006) 117 – 127 www.elsevier.com/locate/marchem

The PAH composition of surface sediments from Stagnone coastal lagoon, Marsala (Italy) Loredana Culotta a, Concetta De Stefano b, Antonio Gianguzza a, Maria Rosaria Mannino a, Santino Orecchio a,* a

Dipartimento di Chimica Inorganica, Universita` di Palermo, Viale delle Scienze, Parco D’Orleans II, I-90128, Palermo, Italy b Dipartimento di Chimica Inorganica, Chimica Analitica e Chimica Fisica, Universita` di Messina, Salita Sperone 31, I-98166 Messina Vill. S. Agata, Italy Received 6 December 2004; received in revised form 22 March 2005; accepted 5 May 2005 Available online 18 January 2006

Abstract This paper examines the presence, distribution, nature and sources of 22 Aromatic Hydrocarbons (PAHs), which are important environmentally and toxicologically, in sediments from the Stagnone coastal lagoon at Marsala (Italy). Analysis was performed by gas chromatography/mass spectrometry (GC/MS) with selected ion monitoring (SIM). The total concentration of polycyclic aromatic hydrocarbons ranged from 72 to 18381 Ag/kg of dry matrix. The relative standard deviation (RSD) of the replicates on the concentrations of individual compounds ranged from 5% to 20%. The accuracy of method was estimated by analyzing bblankQ samples added of known quantities of analytes and the recover percentage was 88 F 9%. The detection limit (LOD) of analytical procedure was less than 0.2 Ag/kg d.w. for all analytes. The quantification limit (LOQ) of analytical procedure was less than 0.7 Ag/kg d.w. The resulting distributions and weight ratios of specific compounds are discussed in terms of sampling location and origin of organic matter. A comparison with other studies of total PAHs suggests that the levels are within the concentration ranges already reported by other authors. From an eco-toxicological point of view, total PAH concentrations at seven out of the eight sites studied represent a relatively clean environment when compared to other sites. Organic matter content and PAH concentrations were found to be correlated and the compounds present in Stagnone sediments were shown to be mainly of pyrolitic origin, while a negligible quantity of PAHs may derive from biogenic sources since all the sediments contain perylene traces. There is no evidence of coal-tar contamination. Cluster analysis was carried out in order to discriminate between different PAH origins. D 2005 Elsevier B.V. All rights reserved. Keywords: PAHs; Sediments; Environmental chemistry; Coastal lagoon; Stagnone; Marsala

1. Introduction Polycyclic aromatic hydrocarbons (PAHs) include aromatic molecules containing fused aromatic rings * Corresponding author. E-mail address: [email protected] (S. Orecchio). 0304-4203/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.marchem.2005.05.010

and are of special concern because of their widespread distribution throughout the environment and their often toxic and carcinogenic properties (Hoffman et al., 1984; Pruell and Quinn, 1985). Some PAHs are persistent and toxic to aquatic organisms and bioaccumulate. Fish and higher organisms in the food chain tend to metabolise and excrete PAHs relatively rapidly. PAHs

118

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

can influence the development of liver tumours in several fish species and may adversely affect the reproductive process in fish and other aquatic species. As a consequence of their hydrophobic nature, PAHs in aquatic environments rapidly tend to become associated with particulate matter. Sediments therefore represent the most important reservoir of PAHs in the marine environment. In general the two main sources of PAHs in the environment are fossil fuels, mainly crude oil, and the incomplete combustion of organic materials such as wood, coal and oil. PAHs enter the sea by both atmospheric and aquatic routes, though little information exists on riverine input. In addition to the many domestic and industrial combustion processes in use today, coal-tar containing coating systems are also a major source of PAHs. To these may be added offshore activities, oil spills, offshore installations and shipping exhausts (Howsam and Jones, 1998). Under anaerobic conditions some PAHs can also be derived from biogenic precursors such as pigments and steroids (Wakeham et al., 1979, 1980). PAHs formed by natural processes include perylene, retene, and phenanthrene homologues. They are also formed naturally in forest fires and volcanic eruptions. Background levels of PAHs in the marine environment are a result of biosynthesis and natural oil seeps. Anthropogenic activities are generally recognised to be the most important source of PAH release into the environment. Background values for PAHs in sediments appear to be within the range 0.01 to approximately 1 mg/kg d.w. The highest levels of oil in bottom sediments typically occur in river mouths, estuaries and bays, as well as in areas of regular shipping, oil production and transportation. The purpose of this study was to determine the distribution of PAHs in the superficial sediments of the Stagnone coastal lagoon at Marsala (Italy), to provide data for comparison with other marine systems and to establish whether the compounds are of biogenic or anthropogenic derivation. Because of the favourable physical and environmental conditions it enjoys, Stagnone lagoon is used for salt production and fish farming. The lagoon is characterized by abundant organic detritus, deriving from aquagenic and anthropogenic inputs and anoxic/reduced sediments make it a preferential site for the uptake and preservation of PAHs. Moreover, the coal-tar and/ or creosote used in the fish farming process may be a source of polycyclic aromatic compounds. In the present study, investigations were carried out into the sixteen PAHs identified by the US-EPA as

requiring priority monitoring action within the framework of environmental quality control (Simoneit, 1998). Other non-US-EPA listed PAHs, namely perylene and some methyl derivatives, were also investigated in order to obtain further information on their origins. Perylene is usually considered a marker of the terrestrial origin of organic matter in the sediments (Sportsol et al., 1983; Colombo et al., 1989; Budzinski et al., 1997). In this work, together with PAHs, we analyzed water and organic contents because it has been demonstrated that the concentrations of PAHs in sediments were affected by chemical composition of the samples such organic matter and water content (Kim et al., 1999). Generally, sediments with high water content (40% = muddy) were characterised by high values of PAHs, while sandy (about 20% water) sediments with low PAH content (Belahcen et al., 1997), sediments with high organic content were characterised by high values of PAHs (Witt, 1995). Relationships between PAH concentration and distribution, which are useful for evaluating possible sources of contamination, were defined by comparing the profiles of 22 PAH compounds recovered from sediments. 2. Experimental procedure 2.1. Study site The Stagnone lagoon at Marsala is a shallow basin in western Sicily (378 52V north; 128 28V east) (Fig. 1) with a surface area of about 2000 ha and an average depth of 1.5 m. It is separated from the open sea by a calcarenite platform (Isola Longa). The northern (Bocca S. Teodoro) and southern (Bocca Grande) mouths allow water exchange. Water temperatures (min. 11.2 8C, max. 29.1 8C) and salinities (min. 32.8, max. 47.1) have broader annual amplitudes than the surrounding sea (temperature: min. 14.1 8C and max. 26.4 8C; salinity close to 37). The basin is oligotrophic. No freshwater input is present (Sara` et al., 1999). 2.2. Sampling Eight sediment samples were collected in the spring of 2004. Fig. 1 shows the location of the sampling sites. A total of 200 g of surface sediment were manually collected from each site using a glass container and placed in plastic bags. The samples were immediately refrigerated (4 8C) on site, stored avoiding exposure to light, and then rapidly transported to the laboratory where they were frozen prior to analysis.

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

119

1-4 Tyrrhenian sea

5

8

6 7 Marsala Coastal lagoon

Marsala

Stagnone of Marsala geographic coordinate: 37°52' N 12°28’ E

Sample n°1 collected along the bank of the channel

Fig. 1. Location of sampling sites.

2.3. Chemicals All chemicals used were of analytical grade with high purity. In particular, n-pentane and dichloromethane, from Fluka, were 99.8% pure. Acetone (Envisolv for analysis of dioxins, furans and PCB) from Fluka was z 99.8% pure. HCl suprapur for trace analysis from Fluka, Standard PAH mixtures (EPA 610 PAH mix, lot LA-96245) and perdeuterated internal standards (fortification solution B Lot N8 LA-92479) were from Supelco. 2.4. Sample treatment After 10 min centrifugation (4000 rpm), overlying water was pipetted off and approximately 5 g of sediment were treated with precleaned (Soxhlet extracted with dichloromethane for 24 h) anhydrous Na2SO4

(Fluka). Activated copper (200 mg) was added to the extraction vessel to remove elemental sulphur. The copper (Aldrich 40 mesh, 99.5% purity) was activated with HCl 1M and then washed with water, acetone and CH2Cl2. A solution of a perdeuterated internal standard (benzo(a)anthracene-d12) was added. 2.4.1. Extraction of PAH Various techniques and solvents were tested in order to identify the most efficient extraction procedure. Together with the PAHs analysis, performed on the natural sediment samples, different recovery tests were carried out by using sediments non-containing polycyclic aromatic compounds. These bblankQ sediments were obtained by performing several extraction steps in 48 h on three natural sediments. After the complete PAH extraction (controlled by GC–MS analysis), a known amount of PAH (EPA) standard mixture was

120

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

added to each purified bblankQ sediment. The test-samples obtained in this way were extracted by using three different methods. The extraction tests, together with the average recoveries, calculated for the compounds investigated, and the relative mean deviations are reported in Table 1. The results proved to be the Soxhlet better method because it allowed us to recover the highest percentage of PAHs and are in good agreement with literature data (Song et al., 2002; Oros and Ross, 2004; Giacalone et al., 2004). All samples were extracted in a Soxhlet extractor for 24 h using a dichloromethane-pentane 1 : 1 solvent mixture (Carlo Erba, Milano, Italy). The extracts were filtered through a pre-cleaned Pasteur pipette filled with solvent-rinsed glass wool and pre-cleaned anhydrous Na2SO4, previously rinsed with dichloromethane and concentrated in a rotary evaporator with thermostatic bath at T = 35 (F 0.5) 8C. The final volume was around 2 ml. The last stage in the procedure involved drying the solution containing PAHs under a weak nitrogen flow at room temperature. The dry residue was dissolved in 1 ml solution containing the following perdeuterated internal standards in cyclohexane (0.2 mg/L each): acenaphthene d10; phenanthrene d10, chrysene d12 and perylene d12.

(flow rate 2.6 ml/min) and the interface temperature 325 8C. Analysis was performed in Selected Ion Monitoring (SIM) mode. Identification of the components of the standard mixture was carried out by comparing retention times for each component in the mixture with those of pure components analysed under the same experimental conditions. Identification was confirmed by comparing the spectra of the single components with those stored in the acquisition system library. The identification of PAHs in the solutions extracted from sediment was carried out on the basis of previously determined retention times and confirmed using mass spectra. The PAH content in the sample was quantified relative to the perdeuterated PAHs added to the dry residue. The response factors for different compounds were measured by injecting a mixture containing standard compounds and having the same concentration of perdeuterated PAHs as that used for spiking the sample. The most abundant ion was used for quantification and two other ions were additionally used for confirmation. The list of groups of PAHs formed, the deuterated standards employed, the quantification ion and the confirmation ion for each PAH are shown in Table 2.

2.4.2. PAH analysis Qualitative and quantitative determinations were carried out using a gas chromatograph (Shimadzu mod. GC-17A) coupled with a mass spectrometer (Shimadzu, quadrupole detector mod. GCMS-QP5000) equipped with an acquisition data system (Shimadzu, CLASS 5000). GC separations were achieved on an Equity-5 (30 m  0.25 id, 0.5 Am) fused-silica capillary column from Supelco (Milano, Italy). The injector mode was splitless (0.61 min) and a total flow of 20.6 ml/min was used. The injection of both extracts from samples and standard solutions (1 Al) was performed by hand. The injector temperature was maintained at 280 8C. The GC temperature program was: from 40 8C (2 min) to 100 8C at 40 8C/min, to 200 8C at 10 8C/min, to 325 8C (8 min) at 30 8C/min. The carrier gas was helium

2.5. Water content analysis

Table 1 Results of recovery tests carried out on a mixture of compounds Extraction method

Recovery % (mean of three analysis)

Soxhlet Ultrasound (n-pentane: CH2Cl2 = 1 : 1) Ultrasound (CH2Cl2)

88 F 9 50 F 15 59 F 22

About 2 g of previously centrifuged at 4000 rpm for 10 min (Centrifuge ALC mod. 4218) sediment was dried at 105 8C for one night. The water content was determined by weight loss and was utilized to correlate all the results with dry weight. 2.6. Organic matter Since the following procedure is known to overestimate organic matter content owing to the simultaneous elimination of carbonates, we applied it only after the carbonates had been removed. Sediment samples were treated, homogenized at room temperature and then dried at 60 8C overnight. An aliquot (2– 3 g) was weighed and placed in a platinum crucible. To remove inorganic carbon 6 M hydrochloric acid was added to the sample, which was then heated gradually to about 120 8C on a hot plate for 15 min. The crucible was removed, cooled and the sample treated again with HCl and dried at 120 8C for 30 min. Total organic matter (including water soluble organic carbon fraction) in the sediment was measured by determining the loss of weight after combustion at 450 8C for 4 h.

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

121

Table 2 List of groups of PAHs formed for analysis; the deuterated standards employed (underlined), the quantification ion and confirmation ion for SIM GC–MS mode Group

Chemical

Quantification ion

Confirmation ions

1

Naphthalene 2 methyl naphthalene 1 methyl naphthalene Acenaphthylene Acenaphthene Fluorene Acenaphthene d 10 Phenanthrene Anthracene 2 methyl anthracene 9 methyl anthracene Fluoranthene Pyrene 1 methyl pyrene Benzo(a)anthracene Phenanthrene d 10 Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Chrysene d 12 Perylene Indeno(1,2,3-cd)pyrene Dibenzo(a,h)anthracene Benzo(ghi)perylene Perylene d 12

128 142 142 152 154 166 164 178 178 192 192 202 202 216 228 188 228 252 252 252 240 252 276 278 276 264

64, 115, 115, 76, 152, 164,

127 70 70 151 76 165

188, 188, 96, 96, 101, 101, 108, 114,

89 89 82 82 200 200 94 226

114, 126, 126, 126,

226 250 250 250

126, 277, 279, 277,

250 138 139 138

2

3

4

As concerns the determination of total organic matter, as reported in literature (Borovec, 1996), a good correlation between data obtained by the method employed here and the results obtained by elemental analysis and by other methods has been demonstrated, and this supports the reliability of our results. 2.7. Redox potential This parameter was measured in 10% suspension of sediment in water by using a potentiometer (Crison GLP-22) equipped with an Au electrode combined with an Ag/AgCl reference electrode (B.C. Electronic, Milano). 3. Results and discussion Table 3 shows individual and total concentrations of PAHs detected in Stagnone sediments. The total concentration of the 22 compounds investigated, expressed as the sum of concentrations, P PAHs, varies from 72 to 18381 Ag/kg of dry matrix. Extraction yields, utilizing perdeuterated internal standard (benzo(a)anthracene-d12) was never less than 80% and in most cases almost 100%.

To evaluate the precision of the analysis, three replicates of field samples were analyzed. The relative standard deviation (RSD) of the replicates on the concentrations of individual compounds ranged from 5% to 20%. The detection limit (LOD), estimated as 3 r (three times the background noise) (IUPAC criterion), was similar for all analyzed compounds (less than 0.2 Ag/ kg d.w. for all analytes). The blank values of analytical procedure remained always below the quantification limit (LOQ): 0.7 Ag/kg d.w., estimated as 10 times r. The wide range of PAH concentrations found in the sediments indicates heterogeneous levels of contamination. This can be explained by considering the different compositions (sandy or muddy) of the sediments, which determine different absorption capacities, and the different organic matter content. The concentrations of PAHs found in seven out of eight sites were below those that have often been identified (Long et al., 1995) as biologically unsafe and even low- and medium-risk concentration ranges for PAHs (ERL= 552 Ag/kg and ERM = 3160 Ag/kg, respectively) are above the average concentrations for Stagnone sediments. In only one station (station

122

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

Table 3 Results of analysis (average of three determinations) Sampling sites Compounds

1

2

3

4

5

6

7

8

Naphthalene 2 methyl naphthalene 1 methyl naphthalene Acenaphthylene Acenaphthene Fluorene Phenanthrene Anthracene 2 methyl anthracene 9 methyl anthracene Fluoranthene Pyrene 1 methyl pyrene Benzo(a)anthracene Chrysene Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(a)pyrene Perylene Indeno(1,2,3-cd)pyrene Dibenzo(a,h)anthracene Benzo(g,h,i)perylene Total PAHs a (Ag/kg d.w.) Water b (%) Organic Matter c (%) referred to d.w.

33.5 20.0 19.0 7.9 146.7 164.2 2154.2 537.1 162.7 6.4 2716.7 2243.5 106.4 1401.9 1744.5 1303.9 1320.0 1640.4 364.9 875.0 428.5 983.5 18381 58.0 5.2

60.1 25.8 14.7 2.0 9.1 11.1 167.5 37.0 12.6 0.6 222.0 210.2 11.4 163.2 153.8 140.6 144.4 203.8 53.9 169.1 95.1 204.2 2112 36.8 2.40

31.6 20.8 11.4 2.4 3.4 9.8 99.8 20.0 6.0 b0.2 144.9 106.2 4.2 90.6 95.5 86.8 87.2 111.0 30.8 86.0 54.5 101.0 1204 24 0.80

48.9 16.8 10.2 6.4 10.6 16.7 260.9 71.5 33.0 1.3 403.3 322.5 18.5 244.6 213.6 165.5 212.8 226.7 60.3 171.0 120.8 48.2 2684 43.4 4.20

2.1 2.3 1.9 1.3 0.5 2.0 73.1 10.7 5.1 b0.2 79.2 66.4 4.2 53.9 44.9 34.5 33.7 40.5 12.0 31.3 18.6 36.4 555 25.3 1.00

10.4 11.5 6.0 0.8 1.8 2.4 18.1 3.9 1.4 b0.2 25.3 21.3 1.0 22.7 21.6 20.0 22.8 30.1 8.6 27.2 15.4 29.9 302 22.5 1.00

13.2 2.4 1.9 5.9 3.6 1.9 6.4 3.1 1.6 0.7 10.1 7.3 2.6 7.0 6.8 12.5 5.2 7.4 3.5 6.6 12.7 11.9 133 24.0 0.9

6.8 1.6 1.0 0.9 1.3 0.6 4.3 1.7 0.5 b0.2 5.3 6.1 1.7 3.1 6.1 4.4 3.3 4.1 2.4 4.0 4.8 8.6 72 18.5 1.2

Individual and total PAHs concentration (Ag/kg dry weight)a , water contentc , organic matter contentb carried out in sediments of Stagnone of Marsala. a Relative standard deviations range from 5% to 20%. b Relative standard deviations less than 2%. c Relative standard deviations less than 5%.

1 in Fig. 1) was the ERM exceeded. This station is located in a sheltered area of the lagoon where weaker currents result in a depositional environment. In most of the sampling sites a distribution of around 22 polycyclic aromatic compounds (expressed as a weight percentage) was observed. The water content of these samples ranged from 18% to 58% and the organic matter content from 0.8% to 5.2% (Table 3). Redox potentials for the samples investigated ranged from 348 to 48 mV. A first inspection of Table 3 allows us to make the following observations. According to the literature (Budzinski et al., 1999), two types of sediments are characterized by water content: sandy (about 20% water) and muddy (about 40% water). The latter are known to accumulate hydrophobic compounds to a much greater extent than sandy sediments and overall PAH concentration is lower in sands than in muds. The lower absorption of PAHs by sandy sediments is already well documented and the present results are in good agreement

with literature data (Budzinski et al., 1997). Analysis of the organic matter in sediments showed that concentrations of polycyclic aromatic compounds generally increase with an increase of organic matter content. This trend has already been observed by a number of authors (Laflamme and Hites, 1979; Neff, 1979; Budzinski et al., 1999). With the aim of identifying a relationship between the water or organic matter content and PAH concentrations determined in the sediments investigated, we carried out a linear regression analysis. A linear correlation between PAH total concentration and water or organic matter content was found with r = 0.95 and p = 0.001 for the first case and with r = 0.87 and p = 0.01 for the second one. Station n8 1, which exhibits a strong deviation from the curve in both correlations, was not taken into account in the fits. These results suggest that most of the PAHs identified in the samples originate from the same organic matter and that PAHs accumulate mostly on muddy sediments.

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

3.1. Sources of contaminants Four processes can generate polycyclic aromatic compounds: i) combustion at high temperature; ii) release of petroliferous products; iii) diagenetic processes (degradation of organic matter); iv) leaching from aquatic timberworks. What makes it difficult to accurately identify PAH origins is the fact that there exist a number of possible sources and processes that these analytes can undergo prior to deposition in sediments. The molecular patterns generated by each source, however, are like fingerprints, which makes it possible to hypothesize which processes generate PAHs by studying their distribution in sediments. Pyrolytic sources are characterized by the presence of PAHs over a wide range of molecular weights, while petroleum sources are dominated by the lowest molecular weight PAHs. If polycyclic aromatic compounds are grouped into different classes depending on the number of aromatic rings present in their structure, it can be observed that the PAHs with 4, 5 and 6-rings found in sediments at the sites under investigation constitute 75– 80% of the total.

123

The evidence suggests that PAH contamination in the lagoon might originate mainly from the atmospheric pollution caused by the burning of fossil fuels; pyrogenic polycyclic aromatic hydrocarbons are generally characterized by the dominance of high molecular mass (4–6 rings) PAHs over those with low (2–3 ring) molecular mass (Witt, 1998) (Fig. 2). Sources of PAH pollution in the aquatic ecosystems under investigation were estimated by comparing the distribution indexes of some polycyclic aromatic compounds with their concentration ratios. Phenanthrene / anthracene and fluoranthene / pyrene ratios have commonly been used as a means of determining the main origins of PAHs (Gschwend and Hites, 1981; Budzinski et al., 1997; Ohkouchi et al., 1999). Phenanthrene and anthracene are both structural isomers. In particular, phenanthrene is more thermodynamically stable than anthracene; therefore, in petrogenic PAH pollution the Ph / An ratio is very high, while high temperatures during the combustion process help the formation of anthracene and a lowering of the Ph / An ratio. Because of the differences in reactivity and solubility of these two pairs of isomers, their respective ratios are not expected to remain constant and cannot, therefore, provide a picture of the progress of PAHs from their origins, through environmental transport, to deposition in marine sediments. In particular, the high fluoranthene / pyrene (Fl / Py) (N 0.8–1)

8

7

6

Sites

5

4

3

2

1

0%

20%

40%

2 rings

3 rings

60%

4 rings

5 rings

80%

6 rings

Fig. 2. Contribution of 2, 3, 4, 5 and 6 rings to total PAHs.

100%

124

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

and low phenanthrene / anthracene (Ph / An) ratios (b30) are characteristic of PAHs of pyrolytic origin, while low Fl / Py and high Ph / An ratios indicate PAHs to possibly derive from petroleum (Sicre et al., 1987; Budzinski et al., 1997). A further contribution to the knowledge of PAH origin and the extent of their environmental circulation, is given by analyzing the Fl / Py and Ph / An distribution indexes on sediments from Stagnone coastal lagoons. A phenanthrene / anthracene ratio of about four is typical of urban areas, while larger ratios were more common for remote locations (Ohkouchi et al., 1999). The average ratio for all the sites investigated in this work was 4.1, which is consistent with the presence of a nearby urban site. In conclusion, the analytical data obtained from our investigations of the Stagnone lagoon indicate that the polycyclic aromatic compounds found in all sediments are of pyrolytic origin and that a common cause of contamination, such as atmospheric deposition, may be the main source of the PAHs in the water bodies. In the atmosphere PAHs can be associated with particle phases. The aerosol particles are transported by the wind to distant locations and removed from the atmosphere by rain and dry fallout in seawater. As they sink into the water column, PAHs are attacked by micro-organisms, and a fraction of them should decompose. The PAHs in the Stagnone sediments, however, are more resistant to microbial degradation and therefore accumulate (Witt, 1998).

To assess possible pollution sources, perylene concentrations also have to be taken into account (Tam et al., 2001). In the present study, perylene was found at low concentrations in all the sediments investigated (from 2.0% to 3.3% of total PAHs). The presence of perylene is generally attributed to terrestrial, marine or aqueous biogenic precursors (Yang, 2000; Soclo et al., 2000) and high concentrations of this compound have usually been measured in anoxic sediments with high biological productivity (Budzinski et al., 1997). We therefore believe that the low concentrations of perylene found in all the stations may originate from a negligible contribute of diagenic processes. This assumption is confirmed by the low redox potential values measured in all sediments. 3.2. Cluster analysis To obtain more information on the origins of PAHs and the relationships between them, we conducted a cluster analysis (Forina et al., 2003) on the variables (PAH concentrations and organic matter content). The cluster analysis revealed the presence of two different groups. The dendrogram in Fig. 3 represents the two major groups. The first consists of naphthalene, phenanthrene, pyrene, benzo(a)anthracene, chrysene, benzo(b)fluoranthene, benzo(k)fluoranthene, benzo(a)pyrene, indeno(1,2,3-cd)pyrene, benzo(g,h,i)perylene, dibenzo(a,h)anthracene, and fluoranthene and the second of 2-methylnaphthalene, 1-methylnaphthalene, ace-

Fig. 3. Dendrogram for 22 PAHs and organic matter content: (1) naphthalene, (2) 2-methylnaphthalene, (3) 1-methylnaphthalene, (4) acenaphthylene, (5) acenaphthene, (6) fluorene, (7) phenanthrene, (8) anthracene, (9) 2-methylanthracene, (10) 9-methylanthracene, (11) fluoranthene, (12) pyrene, (13) 1-methylpyrene, (14) benzo(a)anthracene, (15) chrysene, (16) benzo(b)fluoranthene, (17) benzo(k)fluoranthene, (18) benzo(a)pyrene, (19) perylene, (20) indeno(1,2,3-cd)pyrene, (21) dibenzo(a,h)anthracene, (22) benzo(g,h,i)perylene, (23) organic matter content.

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

naphthylene, acenaphthene, fluorene, anthracene, 2methylanthracene, 9-methylanthracene, 1-methylpyrene, perylene and organic matter content. The first cluster is divided into various subgroups. The PAHs present in this cluster are generally of anthropogenic origin (Atanassova and Bru¨mmer, 2004 and references therein). However, fluoranthene, chrysene and phenanthrene are of both anthropogenic and biogenic origin. Benzo(g,h,i) perylene is also known to be a constituent of exhaust fumes (Baek et al., 1991; Wilcke, 2000). The second cluster, can be divided into two subgroups; the first consists of anthracene perylene and 2methylnaphthalene, and the second is composed of 1-methylnaphthalene, fluorene, 2-methylanthracene, acenaphthene, 1-methylpyrene, acenaphthylene, 9methylanthracene and organic matter content. The compounds present in the first subgroup can be considered to be of both anthropogenic and biogenic origin. The low concentrations of perylene that we identified indicate it to be of mainly biogenic origin. According to literature findings (Atanassova and Bru¨mmer, 2004) the PAHs present in the second subgroup can be considered to be of mainly biogenic origin and this is confirmed by their quite low concentrations in sediments (the higher PAH concentrations found at station n.1, as mentioned previously, can be attributed to the fact that this station is located in a sheltered area of the lagoon) and by the fact that are closely related to the organic matter content (see dendogram in Fig. 3). Cluster analysis can be a suitable chemometric tool for differentiating between PAHs of anthropogenic and biogenic origin.

125

As was done for PAH concentrations, a cluster analysis was carried out on the sampling sites. The results show the presence of three clusters (see Fig. 4). The first consists of only one sampling site (site 1). This confirms our experimental results as at this site we found a greater total PAH concentration than at the other sites because of its location in a sheltered area of the lagoon. The second cluster consists of sites 2, 3 and 4 and the third of sites 5, 6, 7, 8. This differentiation is confirmed not only by the geographical position of the sampling sites (the sites present in the second cluster are situated along the inner bank of the lagoon, whereas sites 5–8 are more external) but also by a PAH concentration that is lower in sites 5–8 than in 2–3. The cluster analysis carried out on the sampling sites would therefore suggest that we might reduce the number of sampling sites and consequently the number of analyses performed in future long term monitoring of the lagoon. 4. Conclusion Our main remarks are as follows: a) The results reported here represent the first quantitative investigations of PAHs in superficial sediments from the waters of the Stagnone lagoon. b) The present study made it possible to optimise extraction and analytical conditions for the determination of PAH in lagoon sediments. Under these conditions, recoveries are very good: never less than 80% and in most cases almost 100%. Reproducibility is also satisfactory (relative standard deviation ranged from 5% to 20%).

Fig. 4. Dendrogram for the sampling sites.

126

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127

c) The use of an Equity-5 column allowed complete separation in a shorter space of time (about 25 min) than in previous studies on PAHs (Culotta et al., 2002; Giacalone et al., 2004). d) The greater presence of PAHs with high molecular weights (3–6 rings) in all Stagnone sediments and the Ph / An and Fl / Py ratio values used as PAH distribution indexes demonstrate that most samples owe their PAHs to a predominant single origin, i.e. anthropogenic combustion or pyrolytic processes, while a negligible quantity of PAHs may derive from biogenic sources since all the sediments contain perylene traces (Tam et al., 2001). e) A comparison with other studies of total PAHs suggests that the levels are within the concentration ranges already reported by us (Giacalone et al., 2004) and by other authors (Heit et al., 1988). f) From an eco-toxicological point of view, total PAH concentrations at seven out of the eight sites studied represent a relatively clean environment when compared to other sites (Witt, 1995; Soclo et al., 2000; Giacalone et al., 2004). g) There is no evidence of coal-tar contamination; for coal-tar the Fl / Py ratio is z 1.5 and this ratio is lower for the sediments that were studied in this work. h) Total PAH concentrations were positively correlated with water and organic matter content. i) Cluster analysis would seem to be a valid method for establishing PAH origins and reducing the number of sampling sites and analysis procedures. Acknowledgement We thank ARPA SICILIA (Agenzia Regionale Protezione Ambiente) for financial support (project D.D.G n. 229 del 27.12.02 -Area Tematica n.6 bModellistica e nuove tecnologie applicate alla valutazione dello stato dell’ambiente ed alla protezione ambientaleQ). References Atanassova, I., Bru¨mmer, G.W., 2004. Polycyclic aromatic hydrocarbons of anthropogenic and biopedogenic origin in a colluviated hydromorfic soil of Western Europe. Geoderma 120, 27 – 34 (and references therein). Baek, S.O., Field, R.A., Goldstone, M.E., Kirk, P.W., Lester, J.N., Perry, R., 1991. A review of atmospheric polycyclic aromatic hydrocarbons: sources, fate and behaviour. Water Air Soil Pollut. 60, 279 – 300. Belahcen, K.T., Chaoui, A., Budzinski, H., Bellocq, J., Garrigues, P., 1997. Distribution and sources of polycyclic aromatic hydrocarbons in some Mediterranean coastal sediments. Marine Pollution Bulletin 34, 298 – 316.

Borovec, Z., 1996. The relationship between the content of organic carbon and total organic matter in the stream sediments of Czech rivers. European Water Pollution Control 6, 15 – 20. Budzinski, H., Jones, I., Bellocq, J., Pie´rard, C., Garrigues, P., 1997. Evaluation of sediment contamination by polycyclic aromatic hydrocarbon in the Gironde estuary. Marine Chemistry 58, 85 – 97. Budzinski, H., Letellier, M., Garrigues, P., Le Menach, K., 1999. Optimisation of the microwave-assisted extraction in open cell of polycyclic aromatic hydrocarbons from soils and sediments. Study of moisture effect. Journal of Chromatography 837, 187 – 200. Colombo, J.C., Pelletin, E., Brochn, C., Khall, M., Catoggio, A., 1989. Determination of hydrocarbon distribution indexes—case study: Rio de la Plata estuary, Argentina. Environmental Science and Technology 23, 888 – 894. Culotta, L., Melati, M.R., Orecchio, S., 2002. The use of leaves of Rosmarinus officinalis L. as samplers for polycyclic aromatic hydrocarbons. Assessment of air quality in the area of Palermo. Ann. Chim. (Rome) 92, 837 – 845. Forina, M., Lanteri, S., Armanino, C. (2003) Q-Parvus, Release 3.0: An extendable package of programs for explorative data analysis, classification and regression analysis. Department of Chimica and Tecnologie Farmaceutiche, University of Genova. httpp:// parvus. unige.it. Giacalone, A., Gianguzza, A., Mannino, M.R., Orecchio, S., Piazzese, D., 2004. Polycyclic aromatic hydrocarbons in sediments of marine coastal lagoons in Messina, Italy: extraction and GC/MS analysis, distribution and sources. Polycyclic Aromatic Compounds 24 (2), 135 – 149. Gschwend, P.M., Hites, R.A., 1981. Fluxes of the polycyclic aromatic hydrocarbons to marine and lacustrine sediments in the northeastern United States. Geochimica et Cosmochimica Acta 45, 2359 – 2367. Heit, M., Tan, Y.L., Miller, K.M., 1988. The origin and deposition history of polycyclic aromatic hydrocarbons in the finger lakes region of New York. Water, Air, and Soil Pollution 37, 85 – 110. Hoffman, E.J., Mills, G.L., Latimer, J.S., Quinn, J.G., 1984. Urban runoff as a source of polycyclic aromatic hydrocarbons to coastal waters. Environmental Science and Technology 18, 580 – 587. Howsam, M., Jones, K.C., 1998. Sources of PAHs in the environment. The Handbook of Environmental Chemistry—PAHs and Related Compounds, vol.3-I. Springer, pp. 137 – 174. Kim, G.B., Maruja, K.A., Lee, R.F., Lee, J.H., Koh, C.H., Tanabe, S.S., 1999. Distribution and sources of polycyclic aromatic hydrocarbons in sediments from Kyeonggi Bay, Korea. Marine Pollution Bulletin 28, 166 – 169. Laflamme, R.E., Hites, R.A., 1979. Tetra and pentacyclic, naturallyoccurring aromatic hydrocarbons in recent sediments. Geochim. and Cosmochim. Acta 43, 1687 – 1691. Long, E.R., Mac Donald, D.D., Smith, S.L., Calder, F.D., 1995. Incidence of adverse biological effects within ranges of chemical concentrations in marine and estuarine sediments. Environmental Management 19, 81 – 97. Neff, J.M., 1979. Polycyclic Aromatic Hydrocarbons in the Aquatic Environment. Sources, Fates and Biological Effects. Applied Science Publishers, London, pp. 1 – 262. Ohkouchi, N., Kawamura, K., Kawahata, H., 1999. Distributions of three- and seven rings polynuclear aromatic hydrocarbons on the deep sea floor in Central Pacific. Environ. Sci. Technol. 33, 3086 – 3090. Oros, D.R., Ross, J.R.M., 2004. Polycyclic aromatic hydrocarbons in San Francisco estuary sediment. Marine Chemistry 86, 169 – 184.

L. Culotta et al. / Marine Chemistry 99 (2006) 117–127 Pruell, R.J., Quinn, J.G., 1985. Geochemistry of organic contaminants in Narragansett Bay sediments. Estuarine Coastal and Shelf Science 21, 295 – 312. Sara`, G., Leonardi, M., Mazzola, A., 1999. Spatial and temporal changes of suspended matter in relation to wind and vegetation cover in a Mediterranean shallow coastal environment. Chem. Ecol. 16, 151 – 173. Sicre, M.A., Marty, J.C., Saliot, A., Aparicio, X., Grimalt, J., Albaiges, J., 1987. Aliphatic and aromatic hydrocarbons in different sized aerosols over the Mediterranean Sea: occurrence and origin. Atmospheric Environment 21, 2247 – 2259. Simoneit, B.R.T., 1998. Biomarker PAHs in the environment. The Handbook of Environmental Chemistry—PAH and Related M Compounds, vol. 3-I. Springer, pp. 175 – 222. Soclo, H.H., Garrigues, Ewald, M., 2000. Origin of polycyclic aromatic hydrocarbons (PAHs) in coastal Marine sediments: case studies in Cotonou (Benin) and Aquitaine (France) areas. Marine Pollution Bulletin 40, 387 – 396. Song, Y.F., Jing, X., Fleischmann, S., Wilke, B.-M., 2002. Comparative study of extraction methods for the determination of PAHs from contaminated soils and sediments. Chemosphere 48, 993 – 1001. Sportsol, S., Gros, N., Lichtenthaler, R.G., Gustavsen, K.O, Urdel, K., Ordel, F., Skel, J., 1983. Source identification of aromatic hydro-

127

carbons in sediments using GC/MS. Environmental Science and Technology 17, 282 – 286. Tam, N.F.Y., Ke, L., Wang, X.H, Wong, Y.S., 2001. Contamination of polycyclic aromatic hydrocarbons in surface sediments of mangrove swamps. Environmental Pollution 114, 255 – 263. Wakeham, S.G., Schaffner, C., Giger, W., 1979. Perylene in sediments from the Namibian Shelf. Geochimica et Cosmochimica Acta 43, 1141 – 1144. Wakeham, S.G., Schaffner, C., Giger, W., 1980. Polycyclic aromatic hydrocarbons in recent lake sediment-II. Compounds derived from biogenic precursors during early diagenesis. Geochimica et Cosmochimica Acta 44, 415 – 429. Wilcke, W., 2000. Polycyclic aromatic hydrocarbons (PAHs) in soil— a rewiew. J. Plant Nutr. Soil Sci. 163, 229 – 248. Witt, G., 1995. Polycyclic aromatic hydrocarbons in water and sediments of the Baltic Sea. Marine Pollution Bulletin 31, 237 – 248. Witt, G., 1998. Polycyclic aromatic hydrocarbons in water and sediments of the Baltic Sea. Marine Pollution Bulletin 36, 577 – 586. Yang, G.P., 2000. Polycyclic aromatic hydrocarbons in the sediments of the South China Sea. Environmental Pollution 108, 163 – 171.