validation of ucides cordatus as a bioindicator of oil

007) 315–327www.elsevier.com/locate/envint

Environment International 33 (2

Validation of Ucides cordatus as a bioindicator of oil contaminationand bioavailability in mangroves by evaluating sediment

and crab PAH records

Adriana Haddad Nudi a, Angela de Luca Rebello Wagener a,⁎, Eleine Francioni a,1,Arthur de Lemos Scofield a, Carla B. Sette a, Alvaro Veiga b

a Departamento de Química, Pontifícia Universidade Católica do Rio de Janeiro, 22543-900 Rio de Janeiro, Brazilb Departamento de Engenharia Elétrica, Pontifícia Universidade Católica do Rio de Janeiro, 22543-900 Rio de Janeiro, Brazil

Received 6 July 2006; accepted 1 November 2006Available online 6 December 2006

Abstract

This study is aimed at verifying the relevance of Ucides cordatus as a bioindicator of oil contamination and PAH bioavailability in mangrovesediments. For this, crabs and sediment cores were sampled from five mangroves, including an area suspected of contamination derived from anMF380 oil spillage, and analyzed for the 16 PAH in the USEPA priority list as well as for the five series of alkylated homologues. Concentrationsin sediments varied from 35 μg kg−1 in the lower core layer of the control area to 33,000 μg kg−1 in the upper layer of the most contaminated area.Total PAH contents in crabs varied from 206 to 62,000 μg kg−1 and were closely correlated to that in sediments. In general, individual PAHprofiles in both matrices were in good agreement. Phenanthrenes, however, were more predominant in crabs making up to 30–46% of the TotalPAH. Accumulation factors found in the range of 0.7 to 35 were highly variable even after normalizing concentrations for organic carbon and lipidcontent. Survival in highly contaminated environment and reliable record of environmental contamination in the tissue provide evidence thatU. cordatus is an excellent bioindicator for oil in mangroves.© 2006 Elsevier Ltd. All rights reserved.

Keywords: Ucides cordatus; Crabs; PAH; Oil contamination; Sediments; Monitoring

1. Introduction

The majority of Brazilian oil production derived fromoffshore fields, mainly from the Campos Basin, in the state ofRio de Janeiro. This area supplies nearly 80% of the oilexploited in the country. The oil produced is transported,transferred and/or processed in coastal areas, often populatedwith mangrove forests which have been threatened by oil spillsand illegal oil releases. There is a need to develop adequatemeans to monitor oil derived contaminants in these ecosystems,as well as their effects on the mangrove distinctive biota. Thepolycyclic aromatic hydrocarbons (PAH) present in oils are of

⁎ Corresponding author. Tel.: +55 21 3527 1809; fax: +55 21 2239 2150.E-mail address: [email protected] (A. de Luca Rebello Wagener).

1 Present address: Laboratório de Química Ambiental, PETROBRAS/CENPES/PDEDS/AMA, 21941-598 Rio de Janeiro, Brazil.

0160-4120/$ - see front matter © 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.envint.2006.11.001

special concern because many of its constituents are persistent,toxic and show mutagenic activity.

PAH are widely disseminated in aquatic environments as aresult of natural and anthropogenic processes. Petroleum,incomplete combustion of fossil fuels (Neff, 1979) and forestfires are important sources of this group of substances. PAHfrom pyrolytic sources are predominantly non-substitutedparental polyaromatic hydrocarbons, while those present inoils are largely alkylated homologues (Steinhauser and Boehm,1992). Several papers have shown that marine organisms areprone to bioaccumulate these substances, particularly in lipid-rich tissues (Neff, 2002; Carls et al., 2004; Francioni et al.,2005; Dugan et al., 2005).

PAH present in sediments or food are assumed to be lessbioavailable than in water, because bioaccumulation seems toinvolve release in solution before uptake (Pruell et al., 1987).The bioaccumulation of certain PAH by organisms in sediments

mailto:[email protected]

http://dx.doi.org/10.1016/j.envint.2006.11.001

316 A.H. Nudi et al. / Environment International 33 (2007) 315–327

depend, therefore, on their water solubility, water/solid phasepartitioning and Kow (Bierman, 1990; Lamoureux and Browna-well, 1999), as well as on the solid phase properties, as forinstance, organic carbon content and speciation (Rust et al.,2004; Sundelin et al., 2004; Cornelissen et al., 2005). Crus-taceans and bottom dwelling organisms can absorb toxics byingesting contaminated food. Biotic (reproductive stage, age,sex, metabolism, lipid content, etc.) and physical (temperatureand salinity) factors also influence bioaccumulation (Meadoret al., 1995; Neff, 2002) and are responsible for the concen-tration variability observed within a population.

Organisms that accumulate contaminants but are resistant totheir toxicity can be used in monitoring contamination in themarine environment (Cossa, 1989). Among various aquaticorganisms, invertebrates have been preferred for environmentalassessment. They are the major components in all ecosystemsand, because of the usually numerous population, they can besampled for analyses with little damage to population dynamics(Depledge and Fossi, 1994).Mussels are most frequently used asbiomonitors of water contamination because of the abundance,resilience, and high bioaccumulation of several contaminants ofconcern. Bioaccumulation of PAH in contaminated sediments bycrab species has been reported (Pancirov and Brown, 1977;Hellou et al., 1994; Kayal and Connell, 1995; Axys Group,1995; Baumard et al., 1998; Eickhoff et al., 2003; Dugan et al.,2005). Nevertheless, very few address bioaccumulation of al-kylated homologues and none of these reports target thevalidation of crabs as a biomonitor for oil contamination.Also, there are no reports of monitoring oil contamination inmangrove ecosystems by using crabs.

Crabs are more efficient than mussels in biotransformingPAH into polar metabolites, facilitating excretion (Fillmannet al., 2004). The crab hepatopancreas plays a role similar to theliver in vertebrates and is the organ where biotransformationtakes place. Therefore, analyses of contaminants in hepatopan-creas provide information primarily on recent exposure andshould reflect actual levels in the environment.

The crab Ucides cordatus is ubiquitous in inter-tidal habitatalong the Brazilian coast. In mangroves, during low tide,U. cordatus escapes to sediment burrows where it can be direct-ly exposed to contaminants deposited in this environment. Sucha species is a good candidate to be used as biomonitor for PAHcontamination in mangroves which are among the ecosystemsmost vulnerable to oil contamination (Grundlach and Hayes,1987).

Once an oil spill reaches a mangrove, degradation is slow andforest restorationmay take several decades (Tam et al., 2001). Thepersistence of oil in mangrove sediments is also associated to theactivity of crustaceans by which the contaminant may be carrieddown to lower layers of the sediment, where degradation is slowdue to reduced molecular oxygen concentrations. According toIrwin et al. (1997) this effect is more significant for the highmolecular weight PAH. It is recognized that the fate of oils inmangrove and the associated impact will vary for different oilsand depends on the properties of the mangrove area.

The goal of the present study was to investigate thesuitability of U. cordatus as a biomonitor of oil contamination

in mangrove sediments and as an indicator of oil bioavailability.For this, PAH concentrations as well as individual distributionin crab tissue (hepatopancreas) and in sediments were obtainedand compared, as a means to determine: (1) the degree ofproportionality between environmental contamination and bodyburden; and (2) how closely crabs reproduce PAH sourcefingerprinting found in sediments. This work also aimed todetermine PAH levels in a mangrove contaminated by a majoroil spill that took place in January 2000 in Guanabara Bay.

2. Materials and methods

2.1. Study area

Guanabara Bay is an urban estuary surrounded by a densely populated area ofthe city of Rio de Janeiro and of other seven smaller municipalities. This bay wasselected to test the possibility of using U. cordatus as a biomonitor for oil inmangrove sediments owing to: (1) the presence ofmangrove forests, someofwhichwere contaminated in January 2000 by amajor spill ofMF 380 oil; (2) the presenceof several potential oil sources (two oil refineries, an oil terminal, the second largestharbor in the country, several ship yards and marinas, intense ship traffic, urbanrunoff, oil transport and transfer); (3) the role played as a service area to the oilexploitation fields in Rio de Janeiro, providing infrastructure and ancillaryactivities for the petroleum industry.

2.2. Sampling

The strategy was to obtain sediment cores and crab samples from mangroveareas in Guanabara Bay thought to have different PAH contamination levels.Sediments and crabs were collected from five mangrove areas (Fig. 1). Fourareas are in Guanabara Bay (Peteca Channel (PT) sampled in 2004; NovaOrleans (NO); Surui (S); and Piedade (P) sampled in 2003 and 2004) and one,taken as control, is in Guaratiba (G, sampled in 2004). Guaratiba is a ratherpristine area located 60 km west of Guanabara Bay (Fig. 1). The Suruimangrove, situated between two branches of the Surui River, is frontally andlaterally flooded during high tide. In all mangrove areas, except for Piedade,cores and crabs were collected at two stations: station 1 was near the fringe,where higher contamination levels were expected; and station 2 was locatedabout 200–300 m inland from the fringe. In the Peteca, station 1 is in a sitecrisscrossed by underground oil pipelines while station 2 is at the bay rim.

Sediment samples (S) were collected by using a piston corer (Husky DuckEquipments) especially devised for sampling mangrove sediments and equippedwith 1.0×0.07 m sharp edged aluminum tubes. In each station samples were takenfrom 5 different points located 10 m apart. A total of five cores were then availableat each station. Each core was sliced into three layers labeled as: layer 1: 0–28 cm;layer 2: 28–58; and layer 3: 58–90 cm. Thereafter, sediment layers fromcorresponding depths were mixed together to produce three composite samples foreach station. The thicknesses of the first two layers were defined based on the depthintervals from where crabs were collected. Crabs were usually harvested fromburrows up to 40–50 cm deep. In spite of this, a sediment layer below this depthwas sampled to cautiously check for possible influences of the PAH content on thecrabs. Cores from Guaratiba and Peteca had shorter lengths and were sliced intotwo layers of 0–28 cm and 28–58 cm. Samples were stored in clean aluminumboxes at −20 °C. A separate set of cores was used to measure pH, Eh (redoxpotential) and to sub-sample for organic carbon, grain size and water contentdeterminations as described in Section 2.3.

Crabs were collected from the same sites as for sediments with the help ofartisan fishermen. The number of harvested crabs varied from 15 to 30 as a resultof the availability at each site. Animals were removed manually from theburrows and only adult males with carapace size in the range of 60–70 mmweretaken. Females were not sampled to avoid the uncertainties imposed by thereproductive cycle. In the laboratory, mud was removed from each animal beforeweighing and measuring the carapace width with vernier calipers. Hepatopan-creas (H) were removed with the help of surgical tools (decontaminated withdetergent, acetone and dichloromethane) after opening the carapace withscissors. The material collected from the set of 15–30 crabs was lumped together

Fig. 1. Location of the sampling sites in the Guanabara Bay (Piedade (P), Nova Orleans (NO), Surui (S), Peteca (PT)) and in Barra de Guaratiba (G), the control area.

317A.H. Nudi et al. / Environment International 33 (2007) 315–327

before homogenization in an UltraTurrax (Ika Labortechnik-T25 basic), andstorage in clean glass containers at −20 °C. Sub-aliquots for lipid content anddry weight determination were stored separately.

2.3. Determination of ancillary data in sediment

Eh and pH were measured in fresh sediments, immediately after sampling,with a platinum electrode and a combined glass electrode, respectively.Sediment fraction b63 μm was determined gravimetrically after treating thesamples with hydrogen peroxide, drying and sieving through a 63 μm sieve.Organic carbon determination was performed in 10 mg aliquots of the 63 μmfraction, after eliminating carbonates with 2 M HCl (Suprapur Merck) anddrying at 60 °C to constant weight. Measurements were carried out in a TOC-5000A Shimadzu equipped with an SSM-5000A module for solid samples.Water content was determined gravimetrically.

2.4. Determination of PAH in sediments

PAH determinations were based on the USEPA (US Environmental ProtectionAgency) methods 3540C (USEPA, 1996a) and 8270C (USEPA, 1996b). Aliquotsof 10 g of fresh sediment were treated with anhydrous Na2SO4 (Merck) previous-ly decontaminated at 450 °C. To this mixture, 100 ng of p-terphenyl-d14(AccuStandard) were added as surrogate standard before Soxhlet extraction over24 h with a 1:1 mixture of dichloromethane-acetone (UltimAR-Mallinckrodt). Theextract was rotary evaporated and concentrated to about 1 mL under gentle streamof purified N2, prior to solvent exchange to hexane (UltimAR-Mallinckrodt).

The aromatic fraction was cleaned and separated in a glass column (1.3 cm i.d.and 30 cm height) packed in the top with colloidal copper for sulfur removal,followed by 1 g of anhydrous Na2SO4, 1 g of alumina (Merck) 2% waterdeactivated, and 11 g silica gel (Merck, 230–400 mesh) activated at 160 °C.PAH were eluted with 50 mL of a 1:1 hexane-dichloromethane (UltimAR-Mallinckrodt) mixture. The aromatic fraction was then rotary evaporated andconcentrated under nitrogen stream before quantitative transference to a 1 mLclass A volumetric flask. The volume was completed with hexane after adding

100 ng of an internal standard mixture containing naphthalene-d8,acenaphthene-d10, phenanthrene-d10, chrysene-d12 and perylene-d12(AccuStandard).

2.5. Determination of PAH in hepatopancreas

PAH determination in the hepatopancreas homogenates was performed intriplicate according to the procedure described below.

To aliquots of 3 g of tissue homogenate, 30 g anhydrous Na2SO4 previouslydecontaminated at 450 °C were added followed by 100 ng of p-terphenyl-d14,used as surrogate standard. The mixture was Soxhlet extracted with 200 mLdichloromethane over 24 h. The obtained extract was rotary evaporated andconcentrated to approximately 1 mL under N2 stream. The 70% lipid content inthe extracts were reduce in two steps: (1) the extract was first cleaned byadsorption chromatography in a 2.2 diameter×30 cm length glass columnpacked with 20 g of 2% deactivated neutral alumina. After elution with 100 mLof dichloromethane, the extract was rotary evaporated and concentrated underN2 stream to approximately 1 mL before solvent exchange to acetone-cyclohexane (3:7) (HPLC/Spectro, Tedia); (2) the pre-cleaned extract wasfurther purified by gel permeation using a Shimadzu system equipped with aLC-10AD pump, a UV–VIS detector Shimadzu and Shodex CLNpack columns.Acetone-cyclohexane was used for eluting the analytes of interest. The elutionvolumes where calibrated with a GPC EPA mixture (AccuStandard) containingcorn oil, bis(ethylhexyl) phtlalate, methoxychlor, perylene and sulfur. Thepurified extract was rotary evaporated and concentrated under gentle N2 streamto about 1 mL. After volume correction the extract was treated as described initem 2.4 to obtain the aromatic fraction.

2.6. GC-MS analysis, quantification and quality assurance

Determination using a full scan mode proceeded in a Finnigan Trace GC gaschromatograph coupled to a Finnigan Polaris Q Ion Trapmass spectrometer fittedwith a J and W DB-5 MS.ITD capillary column (30 m×0.25 mm×0.25 μm).Helium, the carrier gas, was adjusted at 1.2 mL min−1 and the column


temperature was programmed as follows: 5 min at 50 °C, 50 °C min−1 up to80 °C, 6 °Cmin−1 from 80 °C to 280 °C, and a final hold of 25min at 280 °C. Theinjector temperature was at 250 °C (interface at 300 °C and ion source at 250 °C;electron impact 70 eVand emission current 250 mA), and the injected volume insplitless mode was 2 μL.

Quantification included the following 37 compounds and/or groups ofalkylated compounds: (1) The 16 USEPA PAH: naphthalene (N), acenaphthy-lene (Acen), acenaphthene (Ace), fluorene (F), phenanthrene (Ph), anthracene(Ant), fluoranthene (Fl), chrysene (Ch), pyrene (P), benzo(a)pyrene (BaP), benz(a)anthracene, (BaAnt), benzo(b)fluoranthene (BbFl), benzo(k)fluoranthene(BkFl), indeno(1,2,3-c,d)pyrene (IP), dibenz(a,h)anthracene (DBAnt), benzo(ghi)perylene (BghiPe); and (2) perylene (Pe), dibenzothiophene (DBT), benzo(e)pyrene (BeP), alkylated naphthalenes (C1N, C2N, C3N and C4N), fluorenes(C1F, C2F and C3F), phenanthrenes (C1Ph, C2Ph, C3Ph and C4Ph),dibenzothiophenes (C1DBT, C2 DBT and C3DBT), chrysenes (C1Ch andC2Ch) and pyrenes (C1P and C2P).

Standard solutions in two concentration ranges were used for quantification:one for the low concentration levels (5, 10, 20, 50, 100 ng mL−1) and another forthe higher concentration levels (50, 100, 200, 400 and 1000 ngmL−1). Theywereprepared including 100 ng mL−1 of each deuterated standard (naphthalene-d8,acenaphthene-d10, phenanthrene-d10, chrysene-d12 and perylene-d12), the 16USEPA PAH, 2-methylnaphthalene, 1-methylnaphthalene, dibenzothiophene,and pyrene (AccuStandard). Quantification of the alkylated PAH was based onthe calibration curve of the non-alkylated homolog except in the case of 1 and 2methyl-naphthalene, for which authentic standards were available. Baselines forthe alkylated PAHwere setmanually. Linear correlation coefficientsN0.990wereobtained for the calibration curves in all cases. For each group of 10 analyzedsamples a standard solution of the above compounds was injected in the CG/MSto verify instrumental calibration conditions. For each batch of samples at leastone analytical blank was analyzed.

For sediments, average recoveries of 71–110% (85±11%; n=39) wereobtained for p-terphenyl d14, the surrogate standard added to samples beforethe extraction. As recommended in the EPA method, surrogate recovery wasnot used to adjust analyte concentrations. Accuracy was checked by analyzingthe NIST 383 sediment reference material. Found values fell within theexpected range for all certified compounds except for naphthalene for whichrecovery was low (30%). The quantitative limit for each PAH was 1 μg kg−1 d.w. and the detection limit (USEPA, 2002) for the dibenz (a,h) anthracene, acompound that showed the lowest signal/noise ratio among all determinedPAH, was equivalent to 0.1 μg kg−1 d.w. of sediment. Standard deviationswere below 10%.

For the hepatopancreas, recoveries of the surrogate standard were between 64and 105% (75±21; n=55); the quantitative limit for each PAH was 3.5 μg kg−1

Table 1PAH concentrations in sediment and hepatopancreas of Ucides cordatus from mang

Hepatopancreas a

Date Mangroves Sampling stations Total PAH Σ16 PAH

(September 03) Piedade 1 303±87 85±14Nova Orleans 1 6551±363 445±81

2 925±82 58±4Surui 1 62297±6314 2290±773

2 62398±8291 1744±247(October 04) Piedade 1 330±27 41±4

Nova Orleans 1 2319±172 187±412 2056±177 117±3

Surui 1 46272±3209 1519±392 3316±81 164±13

Guaratiba 1 412±45 84±142 206±33 26±4

Peteca 1 4131±690 386±372 5727±391 516±46

Total PAH and Σ16 PAH are given in μg kg−1 dry weight.a Standard deviation for the hepatopancreas analysis in triplicate.

d.w. and the detection limit was 0.7 μg kg−1 d.w. The relative standard deviationof each individual compound in real samples ranged from 5.1 to 22% (n=5). Thequality assurance procedure included also the determination of PAH in fish tissuereference material IAEA 406. The lowest recovery of 75% in relation to thereference range was obtained for naphthalene. For all the other compoundsrecovered concentrations were in the expected range.

2.7. Statistical evaluation

Factor analysis was performed as an exploratory analysis for the data setobtained. For this the SPSS 8.0 program was used and the Varimax rotation withKeiser normalization was selected to represent the planar projection of the twofirst principal components. Spearman correlation test was used to search forcorrelations (STATISTICA 6.0).

3. Results and discussion

3.1. PAH in sediments

The sediments contained 35 to 60% humidity, and 1.2 (Guaratiba) to8.4% (Peteca and Nova Orleans) organic carbon (Corg). Among themangroves of Guanabara Bay, the Surui showed the lowest values forboth variables as well as the smallest fraction of fine sediments (fractionb63 μm: Surui, 20–60%; Piedade, Nova Orleans and Peteca, 70–98%;Guaratiba, 50–80%). As expected, Corg decreased with depth in thesediment cores. As for pH and Eh, most samples from the first samplingperiod were slightly anoxic (Eh around 0Vup to−0.150 V) or oxic (Eh of0.700 V in Peteca, for example) and pH was 6.86±0.52. In the secondsampling period pH was 6.78±0.18.

Sediments layers from Guanabara Bay mangroves showed a widerange of PAH concentration (Table 1). In the upper sediment layer theTotal PAH ranged from 615 μg kg−1 in Piedade to 33,000 μg kg−1

in Surui, while in Guaratiba (the control area) it was 175 μg kg−1

(Table 1).The nonparametric Spearman test used to verify possible correla-

tions among PAH concentrations, pH, Eh and organic carbon gavenegative results. The lack of correlation between PAH and Corg can beexplained assuming that phase distribution of such compounds iseffectively controlled by specific fractions of Corg, as for instance blackcarbon and humic material (McCroddy et al., 1996; Gustafsson et al.,

roves in Guanabara Bay and Guaratiba

Sediments

(0–28 cm) (28–58 cm) (58–90 cm)

Total PAH Σ16 PAH Total PAH Σ16 PAH Total PAH Σ16 PAH

615 305 658 366 312 1531135 234 671 292 275 461098 217 385 143 387 5931500 1001 22577 873 11618 50933082 913 9920 306 4556 480792 421 352 132 379 116924 171 238 72 181 301446 268 542 163 294 5923540 972 1358 188 1278 1176601 320 2859 229 667 62243 42 223 22 – –175 22 34 10 – –6132 630 1725 211 – –2550 337 592 92 – –


1997; Accardi-Dey and Geschwend, 2002), which may be present in thestudied sediments as different proportions of total Corg. These fractionswere not quantified in the present work.

Themangroves inGuanabaraBay can be ranked according to the PAHcontamination level as follows: SuruiNPetecaNNova OrleansNPiedade.Satellite images obtained after the oil spill of January 2000 demonstratedthat Peteca, in the proximity of the oil spill, and especially Surui had beenseverely contaminated (Bentz and Miranda, 2001). The high concentra-tions found three to four years after the spill in Surui provide indicationsof the oil persistence in the mangrove as will be shown later.

Fig. 2. PAH distribution in sediment from Surui mangrove sampled in: September 2station 2 (S22). Layers 1 (0–28 cm), 2 (28–58 cm), and 3 (58–90 cm). PAH conaphthalene–C4 naphthalene); ACE (acenaphthylene); ACEN (acenaphthene); F (flu(C1 phenanthrene–C4 phenanthrene); Ant (anthracene); Fl (fluoranthene); P (pyrene)C1Ch–C2Ch (C1 chrysene–C2 chrysene); BbFl (benzo[b]fluoranthene); BkFl (benzIP (indeno[1,2,3-cd]pyrene); DBAnt (dibenz[ah]anthracene); BghiP (benzo[ghi]perdibenzothiophene).

In layers 1 and 2 of both Surui stations, alkylated naphthalenes(ΣC1N=15,700 μg kg−1; ΣC2N=14,500 μg kg−1), alkylated pyrenes(ΣC1P=3000 μg kg−1;ΣC2P=2000 μg kg−1), alkylated phenanthrenes(ΣC1Ph=4700 μg kg−1; ΣC2Ph=2300 μg kg−1), alkylated fluorenes(ΣC1F=3200 μg kg−1; ΣC2F=1500 μg kg−1) and alkylated chrysenes(ΣC1Ch=2000 μg kg−1; ΣC2Ch=1100 μg kg−1) are the majorcomponents, as should be expected for sediments contaminated withoil. A decrease in concentration with depth is generally observed thoughmore accentuated in station 2 (Fig. 2) in 2003. At the fringe, tidal inflowadded to crustacean activity may contribute to downward transport of

003 at station 1 (S11) and station 2 (S12), October 2004 at station 1 (S21) andncentration in μg kg−1 d.w. Abbreviations: N (naphthalene); C1N–C4N (C1orene); C1F–C3F (C1 fluorene–C3 fluorene); Ph (phenanthrene); C1Ph–C4Ph; C1P–C2P (C1 pyrene–C2 pyrene); BaAnt (benz[a]anthracene);Ch (chrysene);o[k]fluoranthene); BaP (benzo[a]pyrene); BeP (benzo[e]pyrene); Pe (perylene);ylene); DBT (dibenzothiophene); C1DBT–C3DBT (C1 dibenzothiophene–C3


contaminants, leading to the high PAH concentrations observed insediments from 58–90 cm depth (11,600 μg kg−1 station 1, firstsampling).

In Surui, Total PAH declined from 2003 to 2004 in all sedimentlayers of both stations 1 and 2. In layer 1 the decrease was of 75%.As shown in Fig. 2, the increasing concentration of naphthalene andfluorene homologues from C0 to Cn is typical of weathered oil.Phenanthrenes and dibenzothiophenes in layer 1, however, showpredominantly bell-shaped concentration profiles, usually an indication

Fig. 3. PAH distribution in sediment from mangrove Peteca mangrove sampled inmangrove (control area) at station 1 (G11) and station 2 (G12). Layer 1 (0–28 cmabbreviations).

of recent oil inputs. The ratios C2DBT/C2Ph and C3DBT/C3Ph werethen used to investigate the source of the oil residues present in thesesediments. Such ratios are considered robust tools (Wang et al., 1999;Readman et al., 2002) to identify the source of oils present in sediments.The C2 pair, as well as the C3 pair, shows similar solubility andweathering rates, which cause the concentration ratio to remain constantin time. The calculated ratios for sediments from the two upper layers ofstations 1 and 2 in Surui (0.53bC2DBT/C2Phb0.65; 0.53bC3DBT/C3Phb0.58) are similar to those of the MF380 oil (C2DBT/

: September 2003 at stations 1 (PT11) and station 2 (PT12), and in Guaratiba) and layer 2 (28–58 cm). PAH concentration in μg kg−1 d.w. (see Fig. 2 for


C2Ph=0.59; C3DBT/C3Ph=0.64), in spite of the chronic oilcontamination in the bay (derived from illegal releases and frequentspillage of diverse magnitudes), which interferes with the process ofidentification of a single event. In the deeper sediment layers suchassociation is less straightforward, possibly due to the lowercontributions of MF380 to the mixture of earlier contaminations.

In light of the above results, the bell-shaped concentration dis-tribution of Ph and DBT homologues may be ascribed to a relativelyhigher persistence of these compounds in the mangrove environment.The sandy acidic soil in Surui (among the studied areas only withless than 35% of fine grains in surface sediments) may contribute tostagnation of pollutant concentration.

Fig. 4. PAH distribution in sediment from Nova Orleans mangrove sampled in: Septem(NO21) and station 2 (NO21). Layer 1 (0–28 cm), layer 2 (28–58 cm), and layer 3

Absorption inside organic particles or occlusion in other sedimentcomponents may render the contaminants inaccessible to bacteriaresponsible for degradation. Under these conditions half lives turn outto be longer than expected for degradable substances. The formation oftar droplets with low surface to volume ratios, that occurs in the case ofmassive contamination, also limits the bacteria access to the PAH.Also, it has been shown that the sequential metabolism of PAH bymarine microorganisms leads to preferential utilization of substrates.For instance, Foght et al. (1989), and Stringfellow and Aitken (1995)observed that naphthalene (N) undergoes preferential biodegradation incomparison to phenanthrene (Ph); the presence of fluorene (F) alsoretards Ph degradation. An attempt to verify possible hindrance of

ber 2003 at station 1 (NO11) and station 2 (NO12); in October 2004 at station 1(58–90 cm). PAH concentration in μg kg−1 d.w. (see Fig. 2 for abbreviations).


phenanthrene degradation by these compounds was made here bysearching for trends between phenanthrene (ΣPh) and naphthalene(ΣN)+fluorene (ΣF) concentrations. The Spearman test gave goodcorrelation coefficients for log ΣPh versus log (ΣN+ΣF) in Surui(r2=0.835; pb0.05; n=12) and also when grouping all examinedmangrove areas (r2=0.882; slope of 0.85; pb0.05; n=26). Thepossibility that these trends represent an in situ evidence of inter-dependent PAH degradation must be better investigated. However they,in principal, cannot be ascribed to a common origin of the compounds,since the degradation rates of each single PAH considered separatelydiffer significantly and, therefore, should cancel out such correlations.

As for the differences in total concentration found in 2003 and2004 (Table 1) degradation may not be the only cause, but they mayrather result from the combination of two factors: the difficulty inpositioning with high accuracy the sampling stations under themangrove forest, and the heterogeneous distribution of the oil insediments. In some cores, droplets of oil were visibly present andmay have contributed to the higher values in the 2003 sampling,despite the use of sediment homogenate prepared using 5 cores fromeach sampled station. This aspect underlines the complexityassociated with environmental impact assessment in mangroveareas. It should be emphasized, however, that individual PAHcomposition found in sediments collected in 2004 concurs withthose obtained for samples from 2003 (Fig. 2).

The Total PAH concentration found in Surui three to four years afterthe oil spill is of the order of magnitude of those reported for sedimentsfrom Caribbean coast mangrove in Bahia de Las Minas four years afterthe Galeta Oil Spill (Burns et al., 1994) and from Prince William Soundafter the Exxon Valdez spill (O'Clair et al., 1996). Comparison withother regions, as for instance those given under Readman et al. (2002),is difficult because many of the cited reports address only parental PAHwhich in the present case are not the prominent components.

Individual PAH distribution in Peteca (Fig. 3) differs from that inSurui. The major compounds are alkylated phenanthrenes, pyrenes andchrysenes, while naphthalenes are at the level of some tens of μg kg−1.The distribution profile of homologues indicates that oil residues are

Fig. 5. PAH distribution in sediment from Piedade mangrove sampled in: Septemb(28–58 cm), and layer 3 (58–90 cm). PAH concentration in μg kg−1 d.w. (see Fig

further degraded than in Surui. Peteca has a known history of oilcontamination due to pipeline leakages and other events. This mayhave resulted in adaptation of the microbial community to degrade oilmore effectively than in other examined mangroves. In spite of thedegradation and of the presence of other oil residues, samples fromstation 2 show indicative ratios (C2DBT/C2Ph=0.62; C3DBT/C3Ph=0.61) similar to those of MF380.

The Nova Orleans mangrove showed PAH concentrations morethan one order of magnitude lower than in Peteca or Surui. As shown inFig. 4, phenanthrenes are dominant in the first layer (from Ph=8.3 tomaximum in C3Ph=150 μg kg−1) and the relative proportion ofdibenzothiophenes is higher than in the other stations, in all sedimentsegments.

In Piedade (Fig. 5), the least contaminated mangrove in GuanabaraBay, the individual PAH profile differs significantly from the others byshowing predominance of PAH heavier than fluoranthene. Guaratiba,the control area, also shows a distinct individual PAH profile (Fig. 3) inwhich naphthalenes, fluorene and pyrenes are dominant.

The ratio between the 16 USEPA PAH and the Total PAH can beused to compare sources of hydrocarbon contamination, as highproportions of parent aromatic hydrocarbons are characteristic ofpyrogenic residues. In Nova Orleans, in the intermediate sedimentlayer, the significant contribution of heavier PAH of pyrolytic originleads to a rise in the Σ16 PAH to 50% of the total (Fig. 6). In Piedade,the presence of quasi even proportions of petrogenic and pyrolytic PAHis evident since the ratio of Σ16 PAH/Total PAH ranges from 49–56%,in 2003, and from 30 (in the lowest segment)–53% in 2004. It shouldbe noted that combustion sources are abundant in the surroundings ofall mangrove areas (vehicles and industrial activities are most relevantin the area of Peteca, while biomass burning is more important in theother mangroves). However, combustion derived PAH was generallylow in sediments and in crab tissue. The contribution of Σ16 PAH toTotal PAH increased with decreasing contamination level; therefore, itis less prominent in Surui and Peteca as shown in Fig. 6.

Besides the fraction of Σ16 PAH, there are at least four diagnosticratios usually applied to distinguish petrogenic sources from pyrolytic

er 2003 (P11) and October 2004 (P21) at station 1. Layer 1 (0–28 cm), layer 2. 2 for abbreviations).

Fig. 6. ∑16 PAH/Total PAH (%) in sediments (layer 1) and hepatopancreas of Ucides cordatus from mangroves in Guanabara Bay and Guaratiba. P11 and P21: firstand second samplings in station 1 of Piedade; NO11, NO12, S11, S12: first sampling, stations 1 and 2 in Nova Orleans and Surui; NO21, NO22, S21 and S22: secondsampling station 1 and 2 in Nova Orleans and Surui; PT11 and PT12: first sampling, stations 1 and 2 in Peteca Channel; G11 and G12 first sampling, stations 1 and 2 inGuaratiba.


sources (Wang et al., 1999; Readman et al., 2002), such as: Ph/Ant, Fl/P,C1Ph/Ph; and the Σother 3–6 rings PAH/Σ 5 alkylated PAH series (Σ5alkylated PAH: C0–C4 naphthalenes and phenanthrenes; C0–C3fluorenes and dibenzothiophenes; and C0–C2 chrysenes. Σother 3–6rings PAH: other parental PAHwith 3 to 6 rings plus perylene and benzo(e)pyrene). Because chrysene and anthracene were not found in manysamples, only two ratios were applied here: Fl/P and “Σother 3–6 ringsPAH/Σ 5 alkylated PAH series”. Table 2 shows the results of these twodiagnostic ratios confirming predominance of petrogenic sources inSurui. However, for Peteca, Nova Orleans, and Guaratiba (layer 1) onlyFl/P indicated petrogenic predominance. The diagnostic ratios indicateinputs from both sources in Piedade.

The sediment quality benchmarks proposed by Buchman (1999) wereused to evaluate the potential risk of the measured PAH concentrations tothe benthic biota. In spite of the low fraction of priority USEPA PAH, inSurui several of the 16 PAHwere above TEL (Threshold Effect Level): instation 2, acenaphthene was above PEL (Probable Effect Level=88.9 μgkg−1) and anthracene was close to the PEL (245 μg kg−1); in stations 1and 2, benz(a)anthraceneNTEL (88.81 μg kg−1), chryseneNTEL(107.77 μg kg−1), fluoreneNTEL (21.17 μg kg−1), 2-methylnaphthale-neNTEL (20.21 μg kg−1). In Peteca benz(a)anthracene was also above

Table 2Diagnostic ratios applied to distinguish petrogenic from pyrolytic sources (Fl/P andΣC2Ph and C3DBT/C3Ph)

Shaded areas highlight values indicative of petrogenic contamination.

TEL. The sum of the lowmolecular weight PAH (N,Acen,Ace, F, Ph andAnt) was above TEL (311.7μg kg−1) in the sediments of layers 1 and 2 ofstation 1, and above ERL (552 μg kg−1) in layer 1 of station 2. For thesum of high molecular weight PAH (Fl, P, BaAnt, Ch, total BFl, BaP, IP,DBAnt and BghiPe) only in layer 1 of station 1 the value was higher thanTEL (655.34 μg kg−1). Benchmarks for alkylated homologues are not yetavailable, although non-planar PAH seem to be more susceptible to reactin a biological system. Furthermore, the most important analytes toconsider in damage assessment for oil spills are the alkylatedhomologues, which are more persistent and often more toxic than theparental compounds (Sauer and Boehm, 1991; Irwin et al., 1997).

3.2. PAH in crabs

Weight and carapace length of sampled crabs ranged from 114 to173 g and from 62 to 73 mm, respectively. Lipid content in thehepatopancreas was of 50 to 74% with the lower values found inanimals from Nova Orleans; humidity ranged from 50 to 74%.

Concentration trends of PAH recorded in the hepatopancreas andsediments provide, in general, (see Table 1) similar information on thedifferent contamination status of the mangrove areas. The prevailing

3–6 rings PAH/Σ5 rings alkylated PAH series) and to trace oil identity (C2DBT/


distribution feature of PAH homologues shown in the histogram of Fig. 7is that of increasing concentration from C0 to Cn and, as in sediments,hepatopancreas show, in most cases, predominance of the C2 or C3homolog in the phenanthrene family.

Apart from Piedade, a rather uncontaminated area in Guanabara Bay,in all other samples there is a strong predominance of phenanthrenes(Kow: 4.57 to 6.51 for C0 to C4) which make up 30 to 46% of the TotalPAH in crabs, while in sediments phenanthrenes appeared in lesserproportion, ranging from 6% (Guaratiba— control area) to 34% (Surui).In animals from Nova Orleans, Surui and Peteca dibenzothiophenes,chrysenes, pyrenes and naphthalenes are the other main components. Thehigh naphthalene concentrations found in Surui sediments are alsopresent in the hapatopancreas. Nevertheless, there is an evident dis-criminative mechanism in the uptake and/or elimination that favors thehigher alkylated homologues. As a result, the concentration distributionprofile of the 5 alkylated PAHseries in crab tissue shows closer analogy tothat of weathered oil than in the sediments. This should be expected sincebioaccumulation is influenced, among other factors, by Kows whichincrease from C0 to Cn in all alkylated series.

The decrease in Total PAH from 2003 to 2004 in Surui sediments isalso recorded in the hepatopancreas (75%). This is a strong indicationof the good U. cordatus response as a biomonitor for PAH.

Fig. 7. PAH distribution in hepatopancreas of Ucides cordatus from Guanabara Bayand 2-control area-Guaratiba); P11 and P21: samplings of September 2003 and OcSeptember 2003 (Nova Orleans and Peteca); S11, S12, S21 and S22 samplings of Sein μg kg−1 d.w. (see Fig. 2 for abbreviations).

The diagnostic ratios (Fl/P and “Σother 3–6 rings PAH/Σ 5alkylated PAH series”) suggest that in crabs the petrogenic imprint isamplified in comparison to the sediments. In general, the C2DBT/C2Phand C3 DBT/C3 Ph ratios are not as similar to the MF380 oil ratios asthose found for sediments. Possibly, differences in bioavailability,uptake and metabolization rates of such compounds are responsible forthe observed discrepancy. This implies that, in the application of suchtools to investigate oil source by using crabs, care has to be taken. Itshould be noted that none of the studied mangroves suspected ofcontamination with the MF380 oil, were originally completely free ofinfluences from other oil sources, since Guanabara Bay is chronicallycontaminated by oil. The use of these ratios to access the presence ofcertain oil in environments subjected to inputs from diverse sources isproblematic.

The Spearman test indicated strong correlation (r=0.934; pb0.05)between PAH concentration in hepatopancreas and PAH in the topsediment layer (0–28 cm). For the intermediate layer (28–58 cm;r=0.789; pb0.05) correlation was weaker, but still significant. Asalready mentioned, the specific crab mobility in the sediment columncontributes to push the oil front downwards. Once the contaminant isburied, aerobic degradation becomes less likely, though the burrowingact may eventually carry molecular oxygen deeper in the sediment.

and Guaratiba mangroves. G11, G12: sampling of September 2003 at stations 1tober 2004, station 1 (Piedade); NO11, NO12, PT11 and PT12: samplings ofptember 2003 and October 2004 at stations 1 and 2 (Surui). PAH concentration

Fig. 8. Factorial analysis of PAH distribution in crabs (U. cordatus) andsediments (all layers) from Guanabara Bay and Guaratiba. Projections have beenperformed on the factorial plane (F1×F2). Factor 1 represents 57% of totalvariance of the original data, and factor 2, 20%. Inset shows a zoom on thecluster (H = hepatopancreas and S = sediment).


This aeration, however, is evidently not sufficient to promote fastelimination of the contaminant from the sediments. The result is thatmangrove crabs remain exposed to oil contamination for long periodsof time after a spill.

The factorial analysis of the entire data set revealed 5 factors ofwhich the first two are responsible for 77% of the overall variance.Factor 1 groups mainly 4–5 rings PAH and some of the higheralkylated homologues; factor 2 groups the majority of the 2–3 ringalkylated and parental homologues, Ant and Ace. The 5–6 ring PAH(ID, BghiPe), Fl are grouped in factor 4 which contributes only 6% tothe overall variance. The plot of the two major factors (Fig. 8) showsthe great majority of sediment and hepatopancreas samples in a cluster,demonstrating that the two matrices show similar concentrations.Outside the cluster, in the upper left part of the plot, appear the Suruisamples from station 1, layer 1, that differ from the cluster by factor 2(concentration of compounds in factor 2), and from hepatopancreassamples by factor 1 (higher concentration of compounds in factor 1). Inthe lower part, outside the cluster, are the hepatopancreas from station 1in Peteca, which also differ in concentration of compounds in factor 1.The analysis underlines substantive differences in Surui samples inrelation to others, which are mainly due to higher concentrations oflight alkylated PAH.

The percentage of parental PAH in crabs is lower than in thesediments, and varies from 11–28%, in the least contaminated areas(Guaratiba and Piedade), to 3–10% in the other mangroves. There is an

Fig. 9. Accumulation Factor (AF) for individual PAH; Surui mangrove, stati

evident dominance of alkylated PAH both in sediments and in thehepatopancreas. Nevertheless, bioaccumulation factors (BAF wascalculated using normalization of PAH concentrations to organiccarbon content in sediments, and to lipid content in animals) forindividual compounds are variable from area to area. In Nova Orleansand Peteca, lower molecular weight PAH show higher BAF, while inSurui, station 1, alkylated phenanthrenes and benzo(k)fluoranthene arefavored (see Fig. 9). Preferred bioaccumulation of alkylated PAHshould be expected from the superior Kow and lipophilicity of suchcompounds in relation to their parent homologues (Irwin et al., 1997),although these same properties may render the compounds lessavailable for uptake from the water phase. Lamoureux and Brownawell(1999) ascribe maximum bioaccumulation factors of hydrophobicorganic contaminants, commonly occurring at intermediate Kows(log Kow: 6.0–6.5), to slow desorption of high Kow compounds, andto association of low Kow compounds to resistant phases. In the presentwork no correlation was found between Kows and BAF when evaluat-ing the entire data set from a single station. However, within the seriesof phenanthrenes (4.57b log Kowb6.51), fluorenes (4.18b logKowb5.5), and diobenzothiophenes (4.49b log Kowb5.73) there aregood linear relations in Surui and Nova Orleans. The increase inbioaccumulation with increasing molecular weight in these series canbe seen in Fig. 9 for Surui. The occurrence of such a relation, par-ticularly in the most contaminated samples, can be ascribed to directexposure to free oil droplets in the sediments.

Melzian and Lake (1987) were possibly the first to report prevalenceof alkylated homologues of naphthalene and phenanthrene over parentalPAH in crabs exposed to No. 2 fuel oil. Hellou et al. (1994) also reportpredominance of phenanthrene and alkylated phenanthrene in Chionoe-cetes opilio andHyas coarctatus in relation to sediments fromConceptionBay in Newfoundland. These authors found, in addition, enrichment ofbenz(a)anthracene and chrysene in hepatopancreas, attributing theconcentration differences between crabs and sediments to the composi-tion of the diet, to bioavailability and to the activity of MFO enzymaticsystem. Similar to the findings of the present paper, Dugan et al. (2005)reported that in sand crab tissues (Eremita analoga) from sandy beachesin California, 70% or more of the Total PAH was alkylated homologues.

In the present work the observed results may be explained by: theaccumulation derived from dietary exposure as a result of direct foodcontact with free oil residues found in some sediment, as for instance,in Surui; and the variable fraction of the active organic substratecontrolling bioavailability of PAH in the different areas.

Di Toro et al. (1991) and Baumard et al. (1999) report that BAFshould preferably be used to predict bioaccumulation, since normal-ization for organic carbon in sediments and for lipids in animals oughtto level out variability among sites and animals. Here the variability,expressed as relative standard deviations of BAF and of accumulationfactors (AF=ratio between concentrations in sediment and hepatopan-creas) calculated either using the Σ16 PAH or Total PAH were very

on 1, samplings (▪) 2003 and (□) 2004 (see Fig. 2 for abbreviations).


similar, and equal to 31% and 28%, respectively. The variation of theAF and BAF ratios for Total PAH of 0.5 to 0.05 in Piedade and 5.8 to0.6 in Nova Orleans, showed a linear correlation characterized by theequation AF=10.1BAF+0.58, with a correlation coefficient (r2) of0.941 (n=28; pb0.001). Therefore, normalization for lipid content(62.4±6.5%) and sediment organic carbon (1.21–8.43%) did notfacilitate the observation of accumulation trends or decreased factorvariability among different sites. There was no relation between AF andorganic carbon content in sediments. These observations emphasize themajor influence of organic carbon speciation in controlling bioavail-ability. The relatively low AF in crabs compared to those observed formussels (Hellou et al., 2005) may be explained because the formerpresent a higher level of metabolism that leads to biotransformation andexcretion of PAH as polar metabolites (Nudi, 2005).

Although chemical analysis of PAH in hepatopancreas is morelaborious than in sediments, the information obtained is of higher valuefor risk assessment. Moreover, harvest of animals can be carried outfaster and requires fewer personnel in the field. This is advantageous inmangroves where effects of oil contamination can be aggravated due tothe transit of persons. The decline in PAH concentration from 2003 to2004 in sediments and crabs from Surui, which is credited toheterogeneities in the sediments, is a strong indication that U. cordatusmobility must be restricted to a rather small area. This limited drift rangeis an additional advantage for the use in biomonitoring, first becauseconcentrations in tissues reliably represent the contamination level in awell specified area, and secondly because the number of harvestedanimals for such a purpose can be optimized, so as to reduce impact onpopulations. The results of this study point out U. cordatus as anexcellent bioindicator to evaluate PAH pollution in mangrove areas.

4. Conclusions

The PAH study carried out in sediments and in hepatopan-creas of Ucides cordatus from Guanabara Bay mangroves showthat bioaccumulation of these substances occurs in crabs fromcontaminated areas such as Surui and Channel of Peteca. Apreferential accumulation of parental and alkylated PAH withthree and four rings was observed.

The determination of alkylated PAH homologues wasessential for identifying the sources of aromatic hydrocarbonsin the crabs. Except in Piedade mangrove, that is further awayfrom oil sources, the main source of PAH contamination in thecrabs is unequivocally petrogenic.

The crabs collected in the Surui show elevated concentrationsof oil derived PAH, which concur with observations in thesediments. This implies an uptake of contaminants by crabsdirectly from the sediments, since, given the high Kow of suchsubstances, concentrations in water should be too low (some tensof ng L−1) to explain the high content in the hepatopancreas.Because crustaceans are capable of metabolizing PAH andexcreting the contaminants, the parity between sedimentconcentration and body burden suggests continuous exposureallied to a moderate efficiency of the MFO enzymatic system.The statistical treatment of the entire PAH data set indicates thatthe main contamination of biota occurs in the upper 28 cm of thesediment column.

Survival in a sediment environment which contains morethan 30,000 μg kg−1 PAH proves that U. cordatus is resilient tooil, a primary prerequisite of a good biomonitor for oil con-

tamination in mangroves. Furthermore, both the Total PAHconcentration and the individual PAH profiles found in thehepatopancreas are reliable records of different contaminationlevels in the environment and of contaminant source.

Finally, analysis of hepatopancreas proved that PAH, presentprincipally in the sediments, are bioavailable to the crabs. Theobtained results can also be used as a basis for evaluating riskimposed to humans due to the common practice of ingestingcrab internal organs.

Acknowledgements

The authors are grateful to the Brazilian Research Council(CNPq), to the Rio de Janeiro Research Foundation (FAPERJ)and to the Brazilian Coordination for Capacity Building(CAPES) for financial support. The collaboration of Dr. RenatoCarreira is acknowledged.

References

Accardi-Dey A, Geschwend PM. Assessing the combined roles of naturalorganic matter and black carbon as sorbents in sediments. Environ SciTechnol 2002;36:21–9.

Axys Group. Determination of polyaromatic hydrocarbons (PAH) includingalkylated PAH in tissues. SOP PH-T-01/Ver 2. Sidney, BC, Canada; 1995.

Baumard P, Budzinski H, Garrigues P. Concentrations of PAHs (PolycyclicAromatic Hydrocarbons) in various marine organisms in relation to those insediments and to trophic level. Mar Pollut Bull 1998;36:951–60.

Baumard P, Budzinski H, Garrigues P, Narbonne JF, Burgeot T, Michel X, et al.Polycyclic aromatic hydrocarbon (PAH) burden of mussels (Mytilus sp.) indifferent marine environments in relation with sediment PAH contamination,and bioavailability. Mar Environ Res 1999;47:415–39.

Bentz CM, Miranda FP. Application of remote sensing data for oil spillmonitoring in the Guanabara Bay, Rio de Janeiro, Brazil. Anais X SBSR,Sessão Técnica oral-workshops. São Paulo, Brazil: INPE; 2001. p. 747–52.

Bierman VJJ. Equilibrium partitioning and biomagnifications of organicchemicals in benthic animals. Environ Sci Technol 1990;24:1407–12.

Buchman MF. NOAA screening quick reference tables. Coastal Protection andRestoration Division, NOAA HAZMAT report 99-1. Seattle, Washington:National Oceanic and Atmospheric Administration; 1999.

Burns KA, Garrity SD, Jorissen D, MacPherson J, Stoelting M, Tierney J, et al.The Galeta oil spill. II. Unexpected persistence of oil trapped in mangrovesediments. Estuar Coast Shelf Sci 1994;38:349–64.

Carls MG, Harris PM, Rice SD. Restoration of oiled mussel beds in PrinceWilliam Sound, Alaska. Mar Environ Res 2004;57:359–76.

Cornelissen G, Bucheli TD, Jonker MO, Koelmans A, Vannoort PM. Extensivesorption of organic compounds to black carbon, coal, and kerogen in sedimentsand soils: mechanisms and consequences for distribution, bioaccumulation,and biodegradation. Environ Sci Technol 2005;39:6881–95.

Cossa D. A review of theMytilus spp as quantitative indicators of cadmium andmercury contamination in coastal waters. Oceanol Acta 1989;12:417–32.

Depledge MH, Fossi MC. The role of biomarkers in environmental assessmentin Invertebrates. Ecotoxicol 1994;3:161–72.

Di Toro DM, Zarba CS, Hansen DJ, Berry WJ, Swartz RC, Cowan CE, et al.Technical basis for establishing sediment quality criteria for nonionicorganic chemicals using equilibrium partitioning. Environ Toxicol Chem1991;10:1541–86.

Dugan JE, Ichikawa G, Stephenson M, Crane DB, McCall J, Regalado K.Monitoring of coastal contaminants using sand crabs. California: CentralCoast Regional Water Quality Control Board; 2005.

Eickhoff CV, He SX, Gobas FA, Law FCP. Determination of polycyclic aromatichydrocarbons inDungeness crabs (Cancermagister) near an aluminum smelterin Kitimat Arm. British Columbia, Canada. Environ Toxicol Chem 2003;22:50–8.


Fillmann G, Watson GM, Howsan M, Francioni E, DepledgeMH, Readman JW.Urinary PAH metabolites as biomarkers of exposure in aquatic environ-ments. Environ Sci Technol 2004;38:2649–56.

Foght JM, Fedorak PM, Westlake DWS. Mineralization of 14C-hexadecane and14C-phenanthrene in crude oil: specific among bacterial isolates. Can JMicrobiol 1989;36:169–75.

Francioni EF, Wagener ALR, Scofield AL, Cavalier B. Biomonitoring ofPolycyclic Aromatic Hydrocarbon in Perna perna (Guanabara Bay, Brazil).Environ Forensics 2005;6:361–70.

Gustafsson Ö, Hanghseta F, Chan C, MacFarlane J, Gschwend PM.Quantification of the dilute sedimentary soot phase: implications for PAHspeciation and bioavailability. Environ Sci Technol 1997;31:203–9.

Grundlach ER, Hayes MO. Vulnerability of coastal environments to oil spillimpacts. Mar Technol Soc 1987;J.12:18–27.

Hellou J, Upshall C, Taylor D, O´Keefe P, O´Malley V, Abrajano T. Unsaturatedhydrocarbons in muscle and hepatopancreas of two crabs species. Chio-noecetes opilio and Hyas coarctatus. Mar Pollut Bull 1994;28:482–8.

Hellou J, Steller S, Leonard J, Longiville MA, Tremblay D. Partition ofpolycyclic aromatic hydrocarbons bioaccumulation in mussels: a harbourcase. Mar Environ Res 2005;59:101–17.

Irwin RJ, VanMouwerik M, Stevens L, Seese MD, Basham W. Environmentalcontaminants encyclopedia. Fort Collins, Colorado: National Park Service,Water Resources Division; 1997.

Kayal S, Connell DW. Polycyclic aromatic hydrocarbons in biota from theBrisbane River Estuary, Australia. Estuar Coast Shelf Sci 1995;40:475–93.

Lamoureux E, Brownawell BJ. Chemical and biological availability of sediment-sorbed hydrophobic organic contaminants. Environ Toxicol Chem 1999;18:1733–41.

McCroddy SE, Farrington JW, Gschwend PM. Comparison of the in situ anddesorption sediment-water partitioning of polycyclic aromatic hydrocarbonsand polychlorinated biphenyls. Environ Sci Technol 1996;30:172–7.

Meador JP, Stein JE, Reichert WL, Varanasi U. A review of bioaccumulation ofpolycyclic aromatic hydrocarbons by marine organisms. Rev EnvironContam 1995;143:79–165.

Melzian BD, Lake J. Accumulation and retention of no. 2 fuel oil compounds inblue crab, Callinectes sapidus rathbun. Oil Chem Pollut 1987;3:367–99.

Neff JM. Polycyclic aromatic hydrocarbons in the aquatic environment —sources, fates and biological effects. London: Applied Science PublishersLTD; 1979. 261 pp.

Neff JM. Bioaccumulation in marine organisms — effect of contaminants fromoil well produced water. The Netherlands: Elsevier; 2002. 452 pp.

Nudi, A.H. Avaliação da contaminação de manguezais da Baía de Guanabarautilizando caranguejos Ucides cordatus como bioindicador de poluentes depetróleo e desenvolvimento de metodologias de análises. PUC-Rio, Rio deJaneiro, Brazil, D. Sci. Thesis; 2005. pp. 233.

O'Clair CE, Short JW, Rice SD. Contamination of intertidal and subtidalsediments by oil from the Exxon Valdez in Prince William Sound. In: RiceSD, Spies RB, Wolfe DA, Wright BA, editors. Proceedings of the ExxonValdez oil spill symposium, vol. 18. American Fisheries Society SymposiumPublication; 1996. p. 61–93. Maryland.

Pancirov RJ, Brown RA. Polynuclear aromatic hydrocarbons in marine tissues.Environ Sci Technol 1977;11:989–91.

Pruell RJ, Quinn JG, Lake JL, Davis WR. Availability of PCBs and PAHs toMytilus edulis from artificially resuspended sediments. In: Capuzzo JM,Kester DR, editors. Oceanic processes in marine pollution, vol. 1. Malabar:Robert Krieger Publisher; 1987. p. 97–108.

Readman JW, Fillman G, Tolosa I, Villeneuve JP. Petroleum and PAHcontamination of the Black Sea. Mar Pollut Bull 2002;44:48–62.

Rust AJ, Burgess RM, McElroy AE, Cantwell MG, Brownawell BJ. Influenceof soot carbon on the bioaccumulation of sediment-bound polycyclicaromatic hydrocarbons by marine benthic invertebrates: an interspeciescomparison. Environ Toxicol Chem 2004;23:2594–603.

Sauer T, Boehm P. The use of defensible analytical chemical measurements for oilspill natural resource damage assessment. Proceedings 1991 oil spillconference. Washington, D.C: American Petroleum Institute; 1991. p. 363–9.

Steinhauser MS, Boehm PD. The composition and distribution of saturated andaromatic hydrocarbons in near shore sediments, river sediments and coastalpeat of Alaska Beaufort Sea. Implications of detecting anthropogenichydrocarbons impacts. Mar Environ Res 1992;33:223–53.

Stringfellow WT, Aitken M. Competitive metabolism of naphthalene,methylnaphthalenes, and fluorine by phenanthrene-degrading pseudomo-nads. Appl Environ Microbiol 1995;61:357–62.

Sundelin B, Wiklund A, Lithner G, Gustafsson O. Evaluation of the role of blackcarbon in attenuating bioaccumulation of polycyclic aromatic hydrocarbonsfrom field-contaminated sediments. Environ Toxicol Chem2004;23:2611–7.

Tam NF, Kel Y, Wang XH, Wong YS. Contamination of polycyclic aromatichydrocarbons in surface sediments of mangroves swamp. Environ Pollut2001;114:255–63.

US Environmental Protection Agency (USEPA). SW-846 Test methods forevaluating solid waste, physical/chemical methods; Method 3540C-SoxhletExtraction; 1996a.

US Environmental Protection Agency (USEPA). SW-846 Test methods for eval-uating solid waste, physical/chemical methods; Method 8270C — Semivo-latile organic compounds by gas chromatography/mass spectrometry; 1996b.

US Environmental Protection Agency (USEPA). Definition and procedure forthe determination of the method detection limit–Appendix B to part 136–Revision 1.11. 40 CFR Ch.I (7-1-02 Edition), 2002.

Wang Z, Fingas M, Page DS. Oil spill identification. J Chromatogr A1999;843:369–411.

validation of ucides cordatus as a bioindicator of oil

Documents