Polycyclic aromatic hydrocarbons in sediments of the Adriatic Sea

Polycyclic aromatic hydrocarbons in sediments of the Adriatic Sea

Marine Pollution Bulletin, Vol. 28, No. 3, pp. 159-165, 1994 Copyright © 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0025-...

583KB Sizes 0 Downloads 48 Views

Marine Pollution Bulletin, Vol. 28, No. 3, pp. 159-165, 1994 Copyright © 1994 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0025-326X/94 $6.00+ 0.00


Polycyclic Aromatic Hydrocarbons in Sediments of the Adriatic Sea LICIA G U Z Z E L L A and ADOLFO DE PAOLIS

Water Research Institute, CNR, Brugherio (MI), Italy

The PAH content of the sediments collected during a naval cruise in the Adriatic Sea is reported. The analysis of the results points out the presence of two main polluted areas: the Delta of the Po River and the Gulf of Trieste. The concentrations of the identified compounds in the marine sediments are compared to those measured for other polluted areas, leading to the conclusion that the Adriatic Sea is moderately contaminated by PAH.

Polycyclic aromatic hydrocarbons (PAH) are one of the more significant classes of organic chemicals that in the recent years have given rise to a growing concern regarding harmful effects to man and other living organisms. This concern arises primarily from the fact that some PAH have been shown to be carcinogenic to mammals, while the environmental impact of the majority of the other PAH that are not carcinogenic, is largely unknown. Because of their low solubility and hydrophobic nature, PAH tend to be greatly enriched in the inorganic and organic air particles that, under the action of atmospheric agents, may be transported in all the ecosystems. Thus freshwater and marine sediments often contain concentrations of PAH of higher magnitudes than those in the overlying water. The deposition of suspended particulates transported by rivers may therefore have an increasing effect in the accumulation of PAH compounds. Once deposited in sediments, PAH are less subjected to photochemical or biological oxidation, especially if the sediment is anoxic, Thus, sedimentary PAH tend to be persistent and may accumulate to high concentrations. The analysis of sediment PAH can serve as a useful index of the rates of PAH input to the aquatic environment. Sediment samples have a substantial integrating effect on temporal patterns of PAH input and offer good geographical resolution, especially when current patterns, sediment origin and settling rates are known. In the present investigation the PAH contamination of Adriatic Sea sediments was evaluated collecting 32 grab samples during a naval cruise from Trieste to Bari. The PAH were determined in the fine sediment fraction (i.e. the one with particulate diameter <0.63 ~tm) according to Karickhoff (1979). This fraction, in fact, because of its higher organic carbon content, preferentially accumulates hydrophobic coml~ounds.

Experimental Sampling The Po sediment was collected by means of a Petersen grab from the top 20 cm of the bottom at the closing-basin station of the river near Guarda Veneta (Rovigo) on 14 March 1990 and it was frozen in glass vessels in the dark at -20°C until the analysis. At a later date, marine sediments were sampled by a Shipek grab during a cruise organized by CNR (National Research Council) aboard the oceanographic ship N/O Minerva during April 1990. The sampling area is shown in Fig. 1 while the geographic coordinates, the salinity of the water just above the sediment and the depth of each sampling station are recorded in Table 1. The marine sediments were stored in the same conditions as that of the river. TABLE 1 Date, g e o g r a p h i c coordinates, salinity in the b o t t o m w a t e r a n d d e p t h of the sampling stations. Geographic coordinates Sampling station

1 2 3 4 6 7 8 10 11 13 l5 l7 18 19 20 21 23 24 25 26 27 27A 27B 27c 27D 28 29 30 31 32 33




Salinity (%0)

Depth (m)


41°24"04" 4 1 ° 5 3 ' 1 1 '' 4 2 0 0 9 ' 6 9 '' 4 2 ° 2 9 ' 7 1 '' 42°52'26" 43o40'33 '' 4 3 0 5 8 ' 3 0 '' 4 4 ° 5 7 ' 2 5 '' 4 5 o 1 9 ' 2 4 '' 4 5 ° 3 9 ' 3 5 '' 45042'45 " 45°38"29 '' 4 5 ° 3 9 ' 2 1 '' 4 5 o 3 6 ' 5 0 '' 45033'32 '' 4 5 ° 2 9 ' 3 8 '' 45°18"10 '' 45°09'27" 44059'54 '' 4 4 0 5 6 ' 4 7 '' 44052"69 '' 44053'37 '' 4 4 ° 5 5 ' 1 7 '' 4 4 o 5 4 ' 3 9 '' 44°54"39 `' 44o44'55 '' 44040'52 '' 44°23'29" 44006'66 '' 43053'29 '' 4 3 0 3 8 ' 0 9 ''

16°43"63" 16°32"41 '' 16°03'51 '' 15028"04 '' 14042"82 '' 14010'08 '' 13036'82 '' 12°24'11" 12°59'32 '' 13°40"70 '' 1 3 ° 3 5 ' 1 4 '' 13022'75 '' 13o12"00 '' 13006'66 '' 12053"57 '' 12o43'63 '' 1 2 ° 2 1 ' 4 2 '' 1 2 ° 2 1 ' 4 9 '' 12031"23 '' 12036'05 '' 12o31"79 '' 12o33'98 '' 1 2 ° 3 6 ' 3 9 '' 12034'70 '' 12°34'59" 1 2 ° 2 3 ' 4 9 '' 12°20'41 '' 12o22"93 '' 12o35'47 '' 1 3 ° 0 1 ' 5 3 '' 1 3 ° 3 2 ' 4 9 ''

38.5 38.5 38.6 38.5 38.6 38.5 38.5 38.2 38.1 38.0 38.0 37.9 37.8 37.9 37.9 37.7 38.3 37.7 37.5 37.5 37.3 37.5 n.d. n.d. n.d. 37.2 36.1 36.2 36.4 36.9 37.9

100 105 120 138 254 77 68 33 29 20 11 11 11 10 12 12 12 17 21 25 13 17 23 22 16 11 14 12 11 14 18


8April 8 April


14April 9April 12 April 12April

12 April 13 April 13 A p r i l 13 A p r i l 13 A p r i l 13April 13 April 13 A p r i l 13April 13April 13 A p r i l

13 April 13 A p r i l 13 April 13 A p r i l 13 April 13 April 13 A p r i l 14 A p r i l 14 April 14April 14April 15 April

n.d. = N o t d e t e r m i n e d .


Marine Pollution Bulletin

0 _.~100


V E ! E x Z ~









TARA=,O--' R=°Isl


CALABRIA Fig. I Geographicaldistribution of the sampling stations in the Adriatic Sea.

Extraction The procedure applied for the extraction of polycyclic aromatic hydrocarbons was the following: sediment samples were homogenized, a sub-sample of the sediment (10-20 g) was sieved and only the fraction smaller than 63 ~tm was used for the analyses. This portion was then freeze-dried for 8 h at the following temperatures: -40°C for the base plate and +30°C for the chamber. The organic carbon content of the freeze-dried sediment (1 g of sample was used) was determined by oxidation with K2Cr207 according to Jackson (1958) as modified by Gaudette (1974). A certain amount of the sediment (1 g) was then extracted with pesticide-grade methylene chloride in an ultrasonic bath at 30°C. Three successive extraction steps were performed adding 10, 10 and 4 ml of methylene chloride respect160

ively. The ultrasonic mixing was carried out for 10 min for each step and the solvent was recovered pipetting the liquid fraction from the top of the glass vessel used for the extraction. The three fractions were then joined and the volume of the solvent was reduced by Rotavapor evaporation, keeping the water bath at 45°C. When the volume of methylene chloride was reduced to 2 ml, 1 ml of pesticide-grade acetonytril was added and the extract was finally concentrated to a volume of 1 ml of acetonytril. As blank for the procedure, 1 g of sodium sulphate was extracted in the same way as sediment samples.

Analysis The quantification of the PAH content in the samples was performed with HPLC Varian 5000 equipment and a LiChrospher 100 RPo18 column (length 25 cm,

Volume 28/Number 3/March 1994

internal diameter 4 mm, particle size 5 ~tm). The flow rate was adjusted to 1 ml min -1 and the mobile phase mixture was carried out in the following conditions: time= 0 rain 60: 40% H20: CH3CN; time-- 20 min 100% CH3CN. The maximum elution time was 30 min. A Perkin-Elmer fluorescence spectrometer, LS-4, was used for the simultaneous determination of different PAH compounds. The excitation and emission wavelengths were changed during the analysis according to the programme shown below:


The results of the recovery of polyeydic aromatic hydrocarbons with ultrasonic procedure. Mean of recoveries Amount (~tg) added to 1 g

Recovery %

1 5 1 5 1 5 0.5

58 105 112 95.6 96 133 42 76 76 90 58 65 121 109 111 96.8 56 93 111

Pyrene Fluoranthene Anthracene

Wavelengths Time (min) 0 6.5 19 21

Excitation (nm)

Emission (nm)

290 305 305 305

463 430 500 400

The peak heights and the relative areas were recorded on a Varian 4270 Integrator. The linearity of the fluorescence detector response for each PAH compounds was verified and is reported in Table 2. The detection limits of the analysis were: 25 ng g-i for pyrene and anthracene and 1 ng g-i for the other compounds. TABLE 2

The linearity response of the fluorescence detector for the single polycyclicaromatic hydrocarbons. Linear range [massinjected (ng)] PAH Pyrene Fluoranthene Anthracene Benzo(a)anthracene Benzo(a)pyrene Benzo(e)pyrene Benzo(k)fluroanthene Benzo(b)fluroanthene Indeno(1,2,3-cd)pyrene Picene

Lower limit

Upper limit

1 0.1 1 0.5 0.1 1 0.05 0.1 0.1 1

10 5 10 10 3 10 0.1 5 5 10

Estimation of recovery The estimation of the recovery obtained with the present extraction procedure was evaluated with a spiking experiment. In the test a sewage sludge material, p r e p a r e d by the B C R w o r k i n g g r o u p o n P A H certification, was used. T h e P A H c o n t e n t of the sludge was p r e v i o u s l y d e t e r m i n e d a n d a q u a n t i t y of 1 a n d 5 Ixg of pyrene, f l u o r a n t h e n e , b e n z o ( b , k ) f l u o r a n t h e n e a n d p i c e n e a n d 0.5 a n d 1.5 ~tg of b e n z o ( a ) a n t h r a c e n e , b e n z o ( a , e ) p y r e n e a n d i n d e n o ( 1 , 2 , 3 , c d ) p y r e n e were a d d e d to 1 g d r y wt of sludge. T h e s e samples were extracted a n d a n a l y s e d following the same p r o c e d u r e as the sediments. A l l the spiking steps were r e p e a t e d five times a n d the m e a n recovery for each a d d i t i o n is r e p o r t e d in Table 3.

Results T h e total polycyclic aromatic h y d r o c a r b o n c o n t e n t in the fine s e d i m e n t fraction a n d the organic m a t t e r



Benzo(a)pyrene Benzo(e)pyrene

0.5 1.5 0.5 1.5

Benzo(k)fluoranthene Benzo(b)fluoranthene Indeno(1,2,3-cd)pyrene Picene

1 5 1 5 0.5 1.5 1















l 15








SD = Standard deviation. c o n t e n t ( O M ) of the a n a l y s e d samples are r e p o r t e d in Table 4. N o significant c o r r e l a t i o n exists b e t w e e n the two p a r a m e t e r s . T h e analysis of P A H levels reveals two m a i n p o l l u t e d areas: the D e l t a of the Po R i v e r a n d the G u l f of Trieste. A m a r k e d i n f l u e n c e of the Po River b u r d e n is e v i d e n t in the stations 26, 27, 2 7 A , 27B, 27C, 27D, 28 a n d 29. I n TABLE 4 PAH content (ng g-I dry wt) and % organic matter in sediments (nd= not determined).

Samplingstation Po-Guarda Ven. 1

2 3 4 6 7 8 10 11 13 15 17 18 19 20 21 23 24 25 26 27 27A 27B 27C 27D 28 29 30 31 32 33

Total PAH (ng g-1 dry wt)

% Organic matter

172 97 150 34 56 44 45 30 78 53 527 355 163 164 41 59 65 33 102 102 128 116 240 214 130 346 126 225 37 27 31 92

1.0 0.9 0.6 0.8 0.4 1.0 0.7 0.3 nd 0.6 1.1 1.2 1.3 0.6 0.0 0.6 1.0 0.8 1.0 1.1 1.6 0.9 1.4 1.3 1.5 1.1 1.2 1.4 0.7 0.4 0.4 0.8 161

M a r i n e Pollution Bulletin

The low but measurable levels of PAIl in the open sea sediments might be attributed to the atmospheric transport of air polluted particles more than to the presence of local pollution sources. Our initial hypothesis that the low PAH content of these sediments might be due to a possible dilution of contaminated sediments with pre-industrial deposits has been rejected because it has been demonstrated by Frignani (1989) that the sedimentation rate in the continental platform is not negligible. Frignani in fact has calculated for a station at - 8 5 m of depth off Ancona (near our station 2) a sedimentation rate of 0.56 gcm -2 yr-1. Regarding the distribution of the single identified compounds (Table 5), the predominance of a characteristic PAH, fluoranthene, present in all the analysed samples is evident. This compound is also the main contaminant for 72% of the samples. Other relevant pollutants are: pyrene, benzo(b) and (k)fluoranthene, while anthracene has never been determined in the sediment samples. The percentage distribution of individual PAH on the total content is quite homogeneous for the Delta samples (Fig. 2) with an exception of pyrene that is present only in the Po and 27A sediments. As concerns the PAH distribution in the Gulf of Trieste (Fig. 3), the sediments collected in stations 13 and 15 are very similar, while that corresponding to station 17 contains a high level of benzo(b)fluoranthene and a low level of pyrene with respect to the other ones.

this area the deposition of suspended matter carried by the fiver plays a relevant role in the contamination of coastal sediments. According to Frignani (1991) the mean accumulation rates calculated on the basis of 21°pb and 137Cs activities can vary from 0.53 to 1.8 g cm -2 yr-1 in the Delta area depending on the sampling stations considered. The lowest sedimentation rate was determined from Frignani (1991) in front of the main mouth of the river (0.53 g cm -2 yr-1 in our station 26). In the north and in the south areas, on the contrary, the influence of the fiver flow is less important and the sedimentation rates are higher (1.72 g cm -2 yr -1 in our station 25, 0.85 and 1.8 g cm -2 yr-~ in the station 27B-C and 28 respectively). The deposition of the suspended matter coming from the Po River is drastically reduced southward of station 29 (sedimentation rate 0.26 g cm -2 yr-1) so that the river influence practically disappears in stations 30, 31 and 32: the PAH content of the collected sediments decreases to natural background level (26-37 ng g-t measured in the other coastal samples). In the Gulf of Trieste the most contaminated area is located in the harbour of the city (station 13). The total amount of the identified PAH reaches a concentration of 0.5 ~g g-l, 10 times the background level (30-78 ng g-1 measured in the stations in the open sea). The Gulf of Trieste is a principle naval base with numerous coal and cargo-handling piers and many private marinas. The observed contamination might be explained therefore on the basis of the oil spillings coming from oil tankers and private yachts. Two other sediment samples that show a significantly higher concentration of PAH are those collected in station 33 off Ancona and in station 1 off Bari. These two cities are characterized by intense tanker traffic so that even in this case the oil spillings might be responsible for the observed contamination.

Discussion Comparison with PCB content of the sediment The same sediments utilized in this study were also analysed for PCB content (Galassi et al., 1993). The comparison of the obtained results evidentiates some points in common. Also in the case of PCB a con-

80cu 700 f--




T <

n 40O9

._c 3O _c: 200

10 0

[ f






pyr b(a)an b(e)p b(b)fl b(k)fl b(a)p inp




Fig. 2 C o n c e n t r a t i o n s of the identified P A H (expressed in p e r c e n t a g e on the total content) in the D e l t a sediments.



Volume 2 8 / N u m b e r 3 / M a r c h 1994





o 50-

T <~

a_ 40(I) O3

._ co 30(D

..c 20O









pyr b(a)an b(e)p b(b)fl b(k)fl b(a)p inp

13 ~


15 [~J 17

Fig. 3 Concentrations of the identified PAH (expressed in percenage on the total content) in the Gulf of Trieste sediments.

taminant effect due to the Po River inflow is clear for the Delta area, while the pollution of the Gulf of Trieste is less serious than for PAH. As for PAH, the atmospheric transport of contaminated air particles plays a relevant role in the pollution of sediments collected in open Adriatic Sea. Again the sediment sample off Bari harbour is characterized by higher concentration of PCB and PAH than those sampled in open sea. Despite these evident similarities no statistical significant correlation (simple regression analysis with Statgraphics) exists between PCB and PAH levels (r-- 0.23). Comparison with other contaminated sediments By comparing the results of this study with the ones published for other Italian sediments (for example those collected in some subalpine lakes; Galassi et al., 1992; Provini et al., 1989) it may be concluded that the concentrations of fluoranthene and benzo(a)pyrene in the sediment collected in the harbour of Trieste are very similar to those measured in Comabbio, the most polluted lake among those studied. The PAH levels in the other sediments collected in the Gulf of Trieste and in the Po Delta area are, on the other hand, closer to those determined for the less PAH contaminated lakes, Varese and Monate. As concerns the comparison with other marine sediments, the PAH levels measured for the Po Delta and the Gulf of Trieste can be considered moderately polluted (Neff, 1979). In the review on PAH edited by Neff, the concentration of benzo(a)pyrene was compared for different marine sediments. In this comparison, the contamination of the Po Delta and the Gulf of Trieste is of the same order of magnitude as the one detected by Lalou (1963) for Port-Venders in France. The monography underlined that the highest sedimentary benzo(a)pyrene levels always occur adjacent to regions of high population or intense

industrial activities. However small amounts of benzo(a)pyrene (0.3-6.5 ng g-l) were detected even in sediments collected in relatively remote regions such as Godthaab in Greenland (Mallet et al., 1963) and Clipperton Lagoon in the tropical eastern Pacific Ocean (Niaussat et al., 1975). The comparison with other Adriatic studies is, on the contrary, not so easy because of the scarcity of the published data. In another experiment on the Yugoslavian Adriatic coast (Dujmov et al., 1989), PAH concentrations in the sediments were expressed as chrysene and Kuwait oil equivalents because the quantification was done on the basis of continuous fluorimetric emission spectra. In the case of Yugoslavian sediment samples, the stations located in the open sea, with the only exception of the one off Dubrovnik, contain higher concentrations of PAH in surface sediments than those located close to the island chain. The study confirms the hypothesis that, in absence of local pollutant sources, the distribution of PAH in sediments is related to the sedimentary process in the sea. In the absence of the inflow of rivers or waste water discharges, the sedimentation rates are usually higher in the open sea than near the coast (Frignani et al., 1989). In another study (Marcomini et al., 1986) similar concentrations of PAH were determined in the sediments collected in Northern parts of the Adriatic Sea. More recent deposits and the analysis of the fine fraction instead of the whole sediments could explain the higher levels of PAHs determined in our study in respect with those published by Marcomini et al. (1986). Sources of P A H in the aquatic environment PAIl are nearly ubiquitous microcomponents of freshwater and marine environments, even in regions remote from industrial activities. The phenomenon may


M a r i n e Pollution Bulletin



V ~





!° v??v??????~vv?vvv?????~?Z*~v??~ "S











V ~


Volume 28/Number 3/March 1994

be explained on the basis of two principle hypotheses: atmospheric transport or biosynthesis. It is in fact well known that a wide variety of organic molecules containing fused-ring polyaromatic systems are synthesized by organisms, particularly bacteria, fungi and algae. Conditions suitable for the reductive biosynthesis of PAH from biogenic material may occur frequently in highly anaerobic basin sediments. Orr and Grady (1967) and Aizenshtat (1973) found in marine sediments collected in southern California (USA) concentration of perylene as high as 260 ng g-t dry wt. Perylene concentrations increased with depth in the sediments, precluding a direct depositional origin. It was hypothesized that perylene was formed in anaerobic sediments through reduction of perylenequinones deriving from the degradation of organic material. Benzo(ghi)perylene and coronene are also compounds sometimes found at high concentration in recent anaerobic sediments; they may be formed by processes similar to those previously described (Meinschein, 1959). Sedimentary PAH assemblages formed in this way are typically quite simple in composition. For the other PAH such as benzo(a)pyrene, it was concluded by Neff (1979) that they cannot be synthetized by bacteria but can be bioaccumulated by a factor of about 500 from the culture medium. The investigator suggested that some PAH could originate in the medium from tap water, test chemicals or by autoclaving standard solutions, rather than by bacterial exogenous biosynthesis of PAH. A large number of complex PAH and heteroaromatic compounds are on the contrary formed by pyrolizing coal as in electrical plants or by burning organic matter as in solid waste incinerators. Crude petroleum and most refined petroleum products are extremely complex mixtures of organic aromatic compounds. Petroleum from different sources may vary tremendously in the relative concentrations of the different hydrocarbon types present. However, it might be possible to identify the main PAH sources on the basis of their relative composition, because generally different sources are characterized by different PAIl composition. Buttini (1992) reported the following typical ratios between pyrene and benzo(a)pyrene measured in air particles of different sources: Concentrations of pyrene/benzo(a)pyrene 0.5 3 0.3-0.8 2-12 50-100

Source Refinery and petroleum platform Electrical plant Domestic heating Gasoline exhaust Diesel exhaust

Comparing these values to the ones calculated for our samples (Table 5), it is possible to conclude that all the above mentioned sources may have contributed to some extent to the pollution of the sediments collected in the Adriatic Sea.

Conclusion The results of this investigation lead to the following conclusions: 1. The more polluted areas in the Adriatic Sea are: the Po River Delta and the Gulf of Trieste; 2. For the Delta area the main source of contamination is the Po River; for the Gulf of Trieste, the oil spills from tankers; 3. The measured PAH content of Adriatic Sea sediment samples can be considered moderately low in comparison with other marine sediments. We thank Silvana Galassi, Department of Biology in the University of Milan, for her useful advice in the preparation of this manuscript.

Aizenshtat, Z. (1973). Perylene and its geochemical significance. Geochim. Cosmochim. Acta. 37,559-567. Buttini, P., Latella, A., Pelliccioni, A. & Cabella, R. (1992). Determinazione di metallie VOC in aree urbane per la valutazione del contributo delle sorgenti. In Inquinamento atmosferico, pp. 557572. Ed. Padovafiere, Padova. Dujmov, J. & Sucevic, P. (1989). Contents of polycyclic aromatic hydrocarbons in the Adriatic Sea determined by UV-fluorescence spectroscopy. Mar. Pollut. Bull. 20, 405-409. Frignani, M., Langone, L. & Ravaioli, M. (1989). Interpretation of radionuclide activity-depth profiles in three sediment cores from the Middle Adriatic. Giornale di Geologia. 51, 131-142. Frignani, M. & Langone, L. (1991). Accumulation rates and 137Cs distribution in sediments off the Po River delta and the EmiliaRomagna coast (northwestern Adriatic Sea, Italy). Cont. Shelf Res. 11,525-542. Galassi, S., Provini, A. & Garofalo, E. (1992). Sediment analysis for the assessment of risk from organic pollutants in lakes. Hydrobiologia 235/236, 639-647. Galassi, S., Guzzella, L. & De Paolis, A. (1993). PCB levels in sediments of the Adriatic Sea. Fresenius Envir. Bull. 2, 25-30. Gaudette, H. E., Flight, W. R. & Toner, L. (1974). An inexpensive titration method for the determination of organic carbon in recent sediments. J. Sediment Petrol 44,249-253. Jackson, M. L. (1958). Organic matter determinations for soils. In Soil Chemical Analysis, pp. 205-223. Prentice-Hall, Englewood Cliffs, NJ. Karickhoff, S. W., Brown, D. S. & Scott, T. A. (1979). Sorption of hydrophobic pollutants on natural sediments. Water Research 13, 241-248. Lalou, C. (1963). Contribution to the study of pollution of marine sediments by benzo-3,4-pyrene. Commision Internationale pour l'Explorations Scientifique de la Mer Mediterranee. Rapport et Proces- Verbeux 17, 711-718. Mallet, L., Perdriau, V. & Perdriau, S. (1963). Extent pollution by polycyclic aromatic hydrocarbons of the benzo-3,4-pyrene type in the North Sea and the glacial Arctic Ocean. Bull. Acad. Nat. Med. 147,320-325. Marcomini, A., Pavoni, B., Donazzolo, R. & Orio, A. A. (1986). Combined preparative and analytical use of normal-phase and reversed-phase high-performance liquid chromatography for the determination of aliphatic and polycyclic aromatic hydrocarbons in sediments of the Adriatic Sea. Mar. Chem. 18, 71-84. Meinschein, W. G. (1959). Origin of petroleum. Bull. Amer. Assoc. Petrol GeoL 42,925-943. Neff, J. M. (1979). Polycyclic Aromatic Hydrocarbons in the Aquatic Environment. Applied Science Publishers Ltd, London. Niaussat, P. M., Trichet, J., Heros, M., Luong, N. T. & Ehrhardt, J. P. (1975). Presence of benzo-3,4-pyrene in the water and organic sediments of brackish water of Polynesian atolls. Study of certain associated biotic and abiotic factors. CR Acad. Sci. (Paris) Ser. D. 281, 1031-1034. Orr, W. L. & Grady, J. R. (1967). Perylene in basin sediments off southern California. Geochim. Cosmochim. Acta 31, 1201-1209. Provini, A., Premazzi, G., Galassi, S. & Gaggino, G. F. (1989). Distribution of nutrients, trace elements, PAHs and radionuclides in sediment cores from Lake Varese (Italy). Hydrobiologia 176/177, 213-223.