Are coastal resources of NW Portugal fingerprinting hydrocarbons released from the Prestige accident?

¿Están los recursos costeros del NO de Portugal mostrando las huellas específicas de los hidrocarburos derramados por el Prestige?

Ana Maria Ferreira^1*, Cristina Micaelo¹ and Carlos Vale¹

¹ IPIMAR / Instituto de Investigação Agrária e das Pescas Av. Brasilia, 14449-006 Lisboa Portugal. * E-mail: amfer@ipimar.pt

Recibida en enero de 2003;
aceptada en febrero de 2003.

Abstract

Seventy-one samples of mussels, sardine, hake, blue whiting and pouting from the NW Portuguese coast, collected between 18 November 2002 and 6 January 2003, were analyzed for total hydrocarbons, individual n-alkanes and major isoprenoids, pristane and phytane. Hydrocarbons (expressed on chrysene equivalents) registered in whole soft tissues of mussels were below 0.5 µg g^-1, with a few exceptions (maximum 1.6 µg g^-1). Relatively low levels and sporadic enhanced values were also found in fish fillet samples. In spite of this, analyses of individual n-alkanes and major isoprenoids showed that 10-61% of the samples (according to the species) exhibited these compounds in the proportions of the Prestige fuel signature. Similar diagnostic ratios in biological samples and Prestige fuel and n-alkane composition corroborate the fingerprint hypothesis in coastal resources from NW Portugal.

Key words: hydrocarbons, mussels, fishes, Prestige, Portugal.

Resumen

Se analizaron los hidrocarburos totales, n-alcanos individuales e isoprenoides mayores, pristano y fitano, en 71 muestras de mejillones, sardina, merluza, bacaladilla y faneca de la costa NO de Portugal, recolectadas entre el 18 de noviembre de 2002 y el 6 de enero de 2003. Las concentraciones de hidrocarburos (expresadas en equivalentes de criseno) registradas en los tejidos blandos de los mejillones fueron menores a 0.5 µg g^-1, salvo algunas excepciones (máx. 1.6 µg g^-1). En las muestras de filetes de pescado también se encontraron niveles bajos e incrementos esporádicos de hidrocarburos. A pesar de ello, los análisis de n-alcanos individuales e isoprenoides mayores mostraron que entre el 10 y el 61% de las muestras (dependiendo de la especie) tuvieron estos compuestos en las mismas proporciones que la huella específica del fuel del Prestige. Las razones diagnósticas similares en muestras biológicas y en el fuel del Prestige y la composición de n-alcanos corroboran la hipótesis de la huella específica dejada por este accidente en los recursos de la costa NO de Portugal.

Introduction

When petroleum products are accidentally discharged into the sea and chemical oil fingerprinting is searched in the ecosystem, selected groups of compounds are usually analyzed: (i) individual saturated hydrocarbons including n-alkanes (C₁₀-C₄₀) and selected isoprenoids pristane and phytane; (ii) polycyclic aromatic hydrocarbons, specifically alkylated homologues, and (iii) biomarker compounds (Faksness et al., 2002). Among several organisms, mussels are commonly used to monitor the levels of hydrocarbons and other contaminants in coastal environments (Granby and Spliid, 1995). In spite of fish mobility, residues in their tissues provide general information of contaminant dispersion through the food chain in the coastal zone (Parga-Lozano et al., 2002). In addition, assessment of fish residues is important in terms of human consumption.

This paper is the first attempt to assess the impact of the Prestige oil spills on coastal resources from the NW Portuguese coast. This preliminary study reports concentrations of hydrocarbons (chrysene equivalents), individual n-alkanes, major isoprenoids and fingerprint diagnostic indices in mussels, sardine, hake, blue whiting and pouting.

Material and methods

From 18 November 2002 to 6 January 2003, a few days after fuel-oil was discharged by the Prestige, the mussel Mytilus galloprovincialis, sardine Sardine pilchards, hake Merlucius merlucius, blue whiting Micromesistius poutassou and pouting Trisopterus luscus were repetitively captured on the NW Portuguese coast. Mussels (n = 29) were collected from intertidal areas of Caminha, Praia de Âncora, Viana and Aveiro, and sardines (11), blue whiting (13), pouting (10) and hake (8) were obtained from the commercial fleet operating on the NW Portuguese coast and from a campaign of the research vessel Noruega. All the organisms were measured, weighed and dissected in the laboratory. Composite samples of whole soft tissues of 25 mussels and of 3-5 muscles of each fish species were prepared for each sampling period and station. Hydrocarbons were analyzed in composite samples using two different analytical methodologies. Approximately 10 g of biological samples were extracted in soxhlets with hexane during 6 h. The extracts were cleaned up with florisil and the fluorescence intensity measured at 360 nm with excitation at 310 nm. Values were expressed in chrysene equivalents, a crude estimation of total hydrocarbon. In order to analyze the individual resolved n-alkanes and major isoprenoids, 10 g of the samples were soxhlet extracted with hexane/dichloromethane (4:1); the extracted clean-up was performed in a column filled with 8 g silica and 8 g alumina (pre-treated). The analyses were done by gas chromatography with flame ionization detector (GC/FID). A calibration curve with chrysene standard was used to calculate the hydrocarbons. For resolved n-alkanes, a standard containing a homologous series from C₁₂ to C₃₆ and pristane was used. All results are expressed in wet weight.

Results and discussion

The time evolution of hydrocarbon concentrations, expressed as chrysene equivalent, in whole soft tissues of mussels, and in muscle of sardine, hake, blue whiting and pouting is presented in figure 1. Most of the mussel composite samples showed levels below 0.5 µg g^-1 and peak concentrations occurred occasionally (maximum 1.6 µg g^-1). These levels are far below the concentration registered in mussels in 1994 (12 µg g^-1) after the near-shore oil spills by the vessel cercal (IPIMAR, 1995). Hydrocarbon concentrations in the fish species analyzed showed a similar pattern: relatively uniform levels and sporadic peak values. Muscle of sardine presented broader concentration ranges than of hake and pouting, and only one composite sample of blue whiting showed a peak maximum. These values ranged in the same concentration intervals reported for the same species caught after the cercal accident. Since the cercal spill occurred close to the intertidal area and accumulation of hydrocarbons in fish was minor, levels registered in the species analyzed after the Prestige accident were interpreted as low to moderate.

n-Alkanes and isoprenoids

The petroleum products discharged into the sea by the Prestige accident are chemically characterized by individual saturated hydrocarbons, including the n-alkanes C₁₂-C₃₅ and the isoprenoids pristane and phytane. The C₂₀-C₂₆ are highly abundant and C₃₂-C₃₅ minor components (CEDRE, 2002). The sum of individual resolved n-alkanes (C₁₄-C₃₆), pristane and phytane in the biological samples are presented in figure 2. In mussels, two periods of higher levels due to the count of the majority of resolved n-alkanes were found: 18-20 November and 5-13 December 2002. The higher values recorded in the first period resulted from larger contributions of the C₃₀-C₃₆ group (maximum 60% of the total) that represents mainly terrestrial inputs (Bouloubassi and Saliot, 1993). These values probably reflect storm-water runoff. The terrestrial contribution in the second period was smaller and n-alkanes existing in the Prestige oil showed larger expression. The compounds C₁₅, C₁₇ and pristane, which are characteristic of plankton material (Zhou et al., 1996), account for more than 80% in most of the sardine samples. The highest amount was on 19 November due to extreme values of pristane (92% of the total). The composition of the compounds analyzed was relatively uniform except on five occasions, showing composition of n-alkanes comparable to the Prestige signature. The majority of the resolved n-alkanes were quantified in hake and blue whiting. Sixty percent of hake and seventy-five percent of blue whiting registered n-alkanes in the proportions of the Prestige signature, while the rest of the samples showed higher proportions of shorter-chain compounds. Higher values occurred in the same periods for the two species: 25-28 November and 4-6 December. Since only 40% of the pouting samples presented all individual n-alkanes, the distribution pattern was less defined.

Diagnostic ratios

The fingerprint of the oil spills in biological tissues was assessed by the quantification of all the individual n-alkanes and three diagnostic ratios (Faksness et al., 2002): pristane/ phytane (Pr/Pt), C₁₇/Pr and C₁₈/Pt. The ratios in the original fuel of the Prestige and the emulsion in seawater determined by CEDRE were 0.9, 2.1 and 1.7, respectively (CEDRE, 2002). The individual compounds present in the samples and the ranges of diagnostic ratios are presented in table 1. Ratios in the samples analyzed varied in broad intervals and the n-alkanes characteristic of Prestige fuel were often present. Snedaker et al. (1995) observed that some of these indices are more helpful in discriminating between petroleum-contaminated and non-contaminated sediments, than for biota. However, the presence of either even- or odd-numbered carbon, which is contrary to the general rule of odd-numbered carbon predominance in marine biotic systems, indicates the influence of petroleum contamination. When n-alkane composition and two of the three diagnostic ratios in the samples were comparable to the characteristics of the Prestige fuel, it was considered that biological material fingerprints the oil spills. Despite the lack of hydrocarbon "mousse" in the intertidal zone of the NW Portuguese coast, more than 50% of the mussels analyzed appear to fingerprint the presence of fuel. As a sessile filter-feeder, the mussel has to filter a large amount of water and thereby registered exposure to soluble and particle-associated compounds. The percentage of fingerprinted fish samples varied from 10% for pouting to 61% for blue whiting. This range may be attributed to various factors, such as food web, habitat (pelagic and demersal species), bioaccumulation capacities, etc.

Acknowledgements

Several colleagues contributed sampling and analytical assistance: F. Pombal, I. Santos, D. Luz, M. Caetano, P. Antunes, R. Granja, M. Martins, N. Fonseca, J. Raimundo, P. Pereira, J. Canário, P. Brito.

References

Bouloubassi, I. and Saliot, A. (1993). Investigation of anthropogenic and natural organic inputs in estuarine sediments using hydrocarbon markers (NAH, LAB, PAH). Oceanologica Acta, 16: 145-161.         [ Links ]

CEDRE (2002). Identification du fuel du Prestige selon les protocols et recommendations du groupe de travail européen CEN BT/TF 120 oil identification. Rapport No. GC.02-15, 10 pp.         [ Links ]

Faksness, Liv-Guri, Weiss, H.M. and Daling, P.S. (2002). Revision of the Nordtest Methodology for oil spill identification. SINTEF Rep. No. STF66 A02028, 110 pp.         [ Links ]

Granby, K. and Spliid, N.H. (1995). Hydrocarbons and organochlorines in common mussels from the Kattegat and the Belts and their relation to condition indices. Mar. Pollut. Bull., 30: 74-82.         [ Links ]

IPIMAR (1995). Impacte do derrame de crude pelo navio "Cercal". Relat. Cient. Téc., Inst. Port. Invest. Marít., 6, 63 pp.         [ Links ]

Parga-Lozano, C.H., Marrugo-González, A.J. and Fernández-Maestre, R. (2002). Hydrocarbon contamination in Cartagena Bay, Colombia. Mar. Pollut. Bull., 44: 71-74.         [ Links ]

Snedaker, S.C., Glynn, P.W., Rumbold, D.G. and Corcoran, E.F. (1995). Distribution of n-alkanes in marine samples from southeast Florida. Mar. Pollut. Bull., 30: 83-89.         [ Links ]

Zhou, S., Ackman, R.G. and Parsons, J. (1996). Very long-chain aliphatic hydrocarbons in lipids of mussels (Mytilus edulis) suspended in the water column near petroleum operations off Sable Island, Nova Scotia, Canada. Mar. Biol., 126: 499-507.         [ Links ]