environmental contamination associated with a marine landfill (‘seafill’) beside a coral reef

Marine Pollution Bulletin 60 (2010) 1993–2006

Contents lists available at ScienceDirect

Marine Pollution Bulletin

journal homepage: www.elsevier .com/locate /marpolbul

Environmental contamination associated with a marine landfill (‘seafill’) besidea coral reef

Ross Jones *

Australian Institute of Marine Science, The UWA Oceans Institute (M096), 35 Stirling Highway, Crawley, WA 6009, AustraliaBermuda Institute of Ocean Sciences (BIOS), 17 Biological Lane, St. George’s, Bermuda

a r t i c l e i n f o

Keywords:CoralReefPOPsMetalsPollutionSediment quality guidelines

0025-326X/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.marpolbul.2010.07.028

* Tel.: +61 8 6369 4015; fax: +61 8 6488 4585.E-mail address: [email protected]

a b s t r a c t

In Bermuda, bulk waste such as scrap metal, cars, etc., and blocks of cement-stabilized incinerator ash(produced from burning garbage) are disposed of in a foreshore reclamation site, i.e., a seafill. Chemicalanalyses show that seawater leaching out of the dump regularly exceeds water quality guidelines for Znand Cu, and that the surrounding sediments are enriched in multiple contaminant classes (metals, poly-cyclic aromatic hydrocarbons, petroleum hydrocarbons, dioxins and furans, polychlorinated biphenylsand an organochlorine pesticide), i.e., there is a halo of contamination. When compared against biologicaleffects-based sediment quality guidelines (SQGs), numerous sediment samples exceeded the low-rangevalues (where biological effects become possible), and for Hg and Zn exceeded the mid-range value(where they become probable). A few metres away from the edge of the 25 acre dump lies a small coralpatch reef, proposed here as most contaminated coral reef in the world.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction metres away from the current edge of the landfill (Fig. 1) but will

In Bermuda, as with many other densely populated small islandnations, waste disposal is a major problem. In the absence of suit-able landfill space, bulk waste (metallic and building waste such ascars and fridges) and municipal solid waste incinerator ash hasbeen dumped in the sea in a ‘foreshore reclamation site’ – effec-tively a marine landfill or ‘seafill’. No attempt is made to controlthe leaching of xenobiotics to the surrounding seawater, for exam-ple by a retaining wall lined with clay or by impermeable mem-branes. Aerial images show the seafill has grown to encompassan area of 25 acres over the last 35 years (Fig. 1). The extent ofthe environmental contamination associated with this disposal op-tion has never been fully evaluated.

The ecology of the harbour that contains the landfill (CastleHarbour) and the history of past anthropogenic disturbances havebeen described in detail by Flood et al. (2005). Briefly, the most sig-nificant biological features are mangroves, seagrasses, patch andfringing coral reefs, some of which are in very close proximity(i.e., metres) to the dump. The reefs in Castle Harbour are de-graded, having been damaged from dredging and land reclamationoperations in the 1940s (Dodge and Vaisnys, 1977; Flood et al.,2005). The dump is located in the north west quadrant of the basin(Fig. 1), in an area of restricted water flow close to comparativelydeep (10–15 m) dredged areas. The nearest reef is only a few

ll rights reserved.

almost certainly be buried in the near future.The marine landfill started operation sometime in the early

1970s, although exact dates are uncertain. Bulk waste, such asscrap metal, cars, buses, mopeds, domestic appliances, construc-tion waste (soils, rubble, and plasterboard), electrical goods, PVCplastics, and used tyres, is bull dozed into the sea at the site. Theco-disposal of municipal solid waste incinerator ash, generatedfrom combustion of household garbage, started at the same areasince the mid-1990s. The ash is composed of bottom ash (consist-ing of primarily coarse, non-combustible materials collected at theoutlet of the combustion chamber) and fly ash (consisting of fineparticulate matter collected by the electrostatic precipitators inthe flue gases, Sabbas et al., 2003). The two ash types are combinedand mixed with cement, and then poured into moulds, producingapproximately 30 � 1 m3 cubed blocks (each weighing 2 tonnes)each day. After curing, the blocks are dumped into the sea and usedto construct containment walls or ‘cells’. The cells form the outerwalls of the dump and are in-filled with bulk waste. In 2002, some60–70,000 tonnes of garbage was processed by incineration and12,180 tonnes of ash was produced. The purpose of the solidifica-tion/cement-stabilization is to reduce the leachability of contami-nants out of the waste mix (Knap et al., 1991a,b; Hjelmar, 1996;Sabbas et al., 2003). Further information on waste managementis available in the Bermuda 2005 State of the Environment report(Anon, 2005).

There have been a few published and unpublished reports ofenvironmental contamination caused by the dump, see for exam-ple Jickells and Knap (1984), Burns et al. (1990), Knap et al.

http://dx.doi.org/10.1016/j.marpolbul.2010.07.028

mailto:[email protected]

http://dx.doi.org/10.1016/j.marpolbul.2010.07.028

http://www.sciencedirect.com/science/journal/0025326X

http://www.elsevier.com/locate/marpolbul

Fig. 1. Location map showing the marine landfill (dump) in Castle Harbour, a semi-enclosed basin on the eastern portion of Bermuda (see inset map). Sediments sampleswere collected 2 m, 20 m, 80 m, 160 m and 680 m from the edge of the dump (encompassing areas of 5398 m2, 15,570 m2, 55,007 m2, 129,298 m2, and 913,250 m2,respectively) on nine transect lines (T1–T9), and at three ‘reference’ sites (R1–R3) 1.6 km from the dump and in the lagoon (L) behind the dump. The dashed lines representthe ‘D’ series and ‘E’ series of samples. Water samples were collected on the NE side of the dump beside a small patch reef and either side of a road-bridge (Causeway)separating Castle Harbour from Grotto Bay. GWA, GWB, GWD represent groundwater sampling sites in the study of Chapman (2008).

1994 R. Jones / Marine Pollution Bulletin 60 (2010) 1993–2006

(1991a,b), Smith and Hellin (1998). These studies describe tracemetal, polychlorinated biphenyl (PCB), and polycyclic aromatichydrocarbon (PAH) enrichment in water and in sediments adjacentto the dump; however, the studies were limited in spatial extent(i.e., generally one or at most a few samples collected immediatelybeside the dump), and although the studies have shown contami-nation (i.e., concentrations above natural background levels), theyhave not provided any information on pollution (i.e., contamina-tion that causes adverse biological effects in the natural environ-ment, Chapman, 2007).

Sediment quality guidelines (SQGs) have been developed andintroduced around the world in regulatory and non-regulatorycontexts to aid in the interpretation of the relationships betweenchemical contamination and measures of adverse biological effects(see for example Burton, 2002; Long et al., 2006). Two commonlyused sets of numerical sediment quality guidelines (SQGs) arethe National Oceanic and Atmospheric Administration (NOAA)guidelines (i.e., effects range low [ERL] and effects range median[ERM] – Long et al., 1995) and the similarly derived Florida Depart-ment of Environmental Protection (FDEP) guidelines (threshold ef-fects level [TEL] and probable effects level [PEL] – MacDonald et al.,1996). These SQGs are based upon statistical analysis of large dat-abases of synoptic toxicity data and sediment chemistry that iden-tify chemical concentrations that are likely to be associated with

biological effects. Both techniques derive three ranges of chemicalconcentrations, including a low range within which adverse bio-logical effects are unlikely to occur (i.e., <TEL or <ERL), a middlerange in which biological effects are possible (i.e., 6ERL and>ERM or 6TEL and >PEL), and a high range within which biologicaleffects are likely to occur (i.e., >ERM or >PEL). SQGs have been usedto rank and/or prioritize contaminated areas or chemicals of con-cern for further investigation (Long et al., 1998), or to evaluate spa-tial patterns of sediment contamination (Crane and MacDonald,2003). Their use in recent years has been encouraged by researchshowing reasonable predictive abilities (Long et al., 1998), but theiruse also reflects a practical need for protective management toolwhere anthropogenic chemicals present a risk to benthic biota(Fairey et al., 2001).

For marine and freshwaters a number of guidelines are available,but the Florida (US) standards for class 3 waters (i.e., water desig-nated for recreation, propagation and maintenance of a healthy,well-balanced population of fish and wildlife) and Australian guide-lines (ANZECC/ARMCANZ, 2000) are most applicable for Bermuda.Both Florida and Australia have extensive coral reef systems, andtheir use in the context of this investigation seems applicable. TheANZECC/ARMCANZ guidelines are mostly derived from single-species toxicity tests on a range of test species. A statistical distribu-tion approach is used to protect a pre-determined percentage of

Table 1Parameters measured in the seawater and sediment samples and methods of analysis.Where ‘A’ represents sites 2 m from the dump, ‘B’ represents samples 20 m from thedump, etc. R = reference sites, and L = lagoon (see Fig. 2).

Analytical methodsreference

Sampling location (seeFig. 1)

Seawater samplesBOD (mg L�1) APHA 5010 (oxygen

metre)1, 2, 3–8, R 1–4 (n = 12)

Sulphate (SO4) APHA 4110, EPA300 (IC) 1, 2, 3–8, R 1–4 (n = 12)Nitrate (as N) APHA 4110, EPA301 (IC) 1, 2, 3–8, R 1–4 (n = 12)Nitrite (as N) APHA 4110, EPA302 (IC) 1, 2, 3–8, R 1–4 (n = 12)Total phosphate as P APHA 4500-P 1, 2, 3–8, R 1–4 (n = 12)Dissolved Cr EPA 3005A, EPA6020A 1, 2, 3–8, R 1–4 (n = 12)

R. Jones / Marine Pollution Bulletin 60 (2010) 1993–2006 1995

species, usually 95% (or 99% in the case of ecosystems with highconservation value such as coral reef environments). Three gradesof guideline trigger values are derived: high, moderate or low reli-ability, and the grade depends on the data available and hence theconfidence or reliability of the final figures.

The hypothesis to be tested in this study is that Bermuda’s ‘mar-ine landfill’ is leaching contaminants (i.e., PCBs, dioxins, metals,hydrocarbons and PAHs) into the surrounding water and that theseare accumulating in sediments to the point at which biological ef-fects are possible. Detected levels are described with respect tobackground enrichment and also to empirically derived sedimentand water quality guidelines.

(seawater) (ICPMS)Dissolved As

(seawater)EPA 3005A, EPA7000BO33 (HVAAS)

1, 2, 3–8, R 1–4 (n = 12)

Dissolved Hg(seawater)

EPA 3005A, EPA 245.7(CVAFS)

1, 2, 3–8, R 1–4 (n = 12)

Al, Cd, Cu, Fe, Pb, Mn,Ni, Zn)

EPA 6020A (ICPMS) 1, 2, 3–8, R 1–4 (n = 12)

TPHC10-C40 CCME (2000) 1, 2, R 1–4 (n = 6)RLMW, RHMW and

RPAHsEPA 3510/3630/8270(Liq–Liq GCMS)

1, 2, 3–8, R 1–4 (n = 12)

Marine sediment samplesDepth (m) SCUBA depth gauge A, B, C, D, E, L, R (n = 49)Grain size Wet/Dry sieving A, B, C, D, E, L, R (n = 49)Moisture (%) ASTM D2794–00 A, B, C, D, E, L, R (n = 49)TOC Nelson and Sommers

(1996)A, B, C, D, E, L, R (n = 49)

Al, Cd, Cr, Cu, Fe, Pb,Mn, Ni, Zn

EPA 3050B/6010B(ICPOES)

A, B, C, D, E, L, R (n = 49)

Hg EPA 3050B/7471A/245.7(CVAFS)

A, B, C, D, E, L, R (n = 49)

As EPA 3050B/7000 series(HVAAS)

A, B, C, D, E, L, R (n = 49)

TPH C10–C40 CCME (2000) (GC-FID) A, R, L (n = 13)PAHs EPA 3540/8270 (GC/MS) T1,T3, T4,T6, T7, T9 A-C ,

R, L (n = 23)PCDDs EPA 1613 (Revision

B)(HRGC/HRMS)A, R, L, 2B-E, 5B-E, 8B-E(n = 25)

PCDFs EPA 1613 (RevisionB)(HRGC/HRMS)

A, R, L, 2B-E, 5B-E, 8B-E(n = 25)

PCBs EPA 1668A (GC/HRMS) A, R, L, 2B-E, 5B-E, 8B-E(n = 25)

Organochlorine (OC)pesticide

EPA 8081(GC/ECD) A, R, L (n = 13)

Organophosphate(OP) pesticide

EPA 8151 (GC/MS) A, R, L (n = 13)

Carbamate pesticide EPA 8318 (LC/MS/MS) A, R, L (n = 13)

2. Materials and methods

2.1. Seawater analysis

Seawater sampling was conducted at the dump and referencesites (Fig. 1) in December 2008 (six samples). Sampling was con-ducted 1 h before low water at 1 m depth a few metres from theedge of the dump. Samples were compared with reference seawa-ter samples taken either side of a road-bridge (‘Causeway’) sepa-rating Castle Harbour from Grotto Bay (Fig. 1). All water sampleswere placed in pre-cleaned amber glass and plastic containers onthe research vessel and stored at 4 �C before analysis (within48 h). Water samples were analysed for heavy metals, PAHs, TotalPetroleum Hydrocarbons TPH (C10–C40) and nutrients based on US-EPA methodologies (Table 1).

All sediment samples were collected by SCUBA divers usingsmall high-density polyethylene (HDPE) scoops. Diver collectionensured the retrieval of undisturbed sediment using non-contam-inating utensils. The surficial (1–2 cm depth) sediment were col-lected over a 10–20 m2 area at distances of 2 m (‘A’ samples),20 m (‘B’ samples), 80 m (‘C’ samples), 160 m (‘D’ samples) and680 m (‘E’ samples) from the edge of the landfill along the nineseparate surveys lines (transects)(Fig. 1). The terms ‘A series’ and‘B series’, etc., are used in the analyses below to describe the sam-ples collected 2 m or 20 m, etc., from the dump edge. Active fillingof bulk waste (cars, fridges, etc.) was occurring at the start of sur-vey lines 8 and 9 at the time of sampling.

For comparative purposes, and to allow interpretation of thesediment chemistry data, three sediment samples were taken1.6 km from the landfill in the central part of Castle Harbour(Fig. 1) and still within the dredged area. These were referred toas reference (‘R’ samples) although they may not represent uncon-taminated sediments (see Section 4). One sediment sample wasalso taken from a semi-enclosed lagoon (L) behind the dump,which connects to Castle Harbour via a shallow channel with a sillat its narrowest point (Fig. 1). At the surface, all samples wereplaced in 2-L HDPE buckets and homogenized with Teflon stirringrods until even in colour and texture, and then distributed intoseparate pre-labelled, acid-washed and solvent-rinsed glass jarswith Teflon�-lined caps for chemical and grain size analysis. Sam-ples were stored on board ship in coolers filled with ice and sam-ples were transferred to a refrigerator (at 4 �C) immediately uponreturn to the laboratory.

Sediments were examined for metals, metalloids, polycyclicaromatic hydrocarbons (PAHs), total petroleum hydrocarbonsTPH (C10–C40), polychlorinated dibenzo-p-dioxins (PCDDs), dib-enzofurans (PCDFs), polychlorinated biphenyls (PCBs), and pesti-cides (organochlorines, organophosphates, and carbamates),based on US-EPA methodologies (see Table 1) and grain size, mois-ture content and total organic carbon (TOC) (Table 1). All chemicalanalyses were conducted by the ALS Laboratory group Analyticaland Chemistry Testing Services, Vancouver Canada. Sediment sam-

ples were analysed in batches, corresponding with sampling eventsand QA/QC procedures to evaluate accuracy and precision of theanalytical data included surrogate recoveries, procedural blanks,blank spike samples, laboratory duplicates, and standard referencematerials, under adherence to ISO/IEC 17025, the main standardused by testing and calibration laboratories, and the ALS internalquality management system.

PAH analyses included US-EPA priority list PAHs (Keith and Tel-liard 1979), including low molecular weight PAHs – acenaphthene(Ae), acenaphthylene (Ay), anthracene (Ant), fluorene (Fluo), naph-thalene (N), phenanthrene (Ph), and high molecular weight PAHs –benzo[a]anthracene (B[a]A), o[a]pyrene (BaP), chrysene (Ch),dibenzo[a,h]anthracene (DBA), fluoranthene (F), pyrene (P),benzo[b,j]fluoranthene (B[b]F), benzo (g,h,i)perylene (Bper),benzo[k]fluoranthene (B(k)F), and indenol(c,d-123)pyrene (IndP).PCDD and PCDF analyses included 17 congeners (2,3,7,8-TCDD,1,2,3,7,8-PeCDD, 1,2,3,4,7,8-HxCDD, 1,2,3,6,7,8-HxCDD, 1,2,3,7,8,9-HxCDD, 1,2,3,4,6,7,8-HpCDD, OCDD, 2,3,7,8-TCDF, 1,2,3,7,8-PeCDF,2,3,4,7,8-PeCDF, 1,2,3,4,7,8-HxCDF, 1, 2,3,6,7,8-HxCDF, 1,2,3,7,8,9-HxCDF, 2,3,4,6,7,8-HxCDF, 1,2,3,4,6,7,8-HpCDF, 1,2,3,4,7,8,9-HpCDF


and OCDF) for which WHO 2005 toxic equivalency factors (TEFs)(Van den Berg et al., 2006) have been determined to allow calcula-tion of toxic equivalent (TEQ) values. For these calculations, concen-trations below the detection limit (DL) were processed at half the DL(Clarke 1998). For polychlorinated biphenyls (PCBs), 209 congenerswere quantified with PCB congeners (IUPAC numbers) 77, 81, 105,114, 118, 107/109, 126, 156, 157, 167, 169, and 189 representing di-oxin-like PCBs (DL-PCBs) in which TEQ values were calculatedaccording to the WHO 2005 TEFs for fish (Van den Berg et al.,2006). PCB results were corrected for surrogate recovery.

Organochlorine (OC), organophosphate (OP) and carbamatepesticides screens were conducted for all sediment samples col-lected immediately beside the dump (i.e., ‘A’ samples) (Table 1),all reference samples and the sample from the Lagoon (Table 1).

Data are presented as mean values of the A, B, C, D and E seriessamples together with the range (minimum and maximum) orstandard deviation (SD, graphs only). For sediment grain size anal-ysis, ternary diagrams of grain size distribution were constructedusing the Wentworth scale (Wentworth, 1922) and the Shepardtriangular coordinate systems for sediments (Shepard, 1954).

To examine spatial patterns of metal contamination, data wasnormalized (also referred to as standardized) by subtracting themean and dividing by the standard deviation to make all values

Table 2Biochemical oxygen demand (BOD), total suspended solids (TSS), nutrients, total petropolycyclic aromatic hydrocarbon (PAH) concentration as mg L�1 (or lg L�1 for ammonia, Tdump (i.e., groundwater see Fig. 1 – data from Chapman, 2008) or seawater collected frocollected in Castle Harbour (see Fig. 1). Data for seawater are expressed as mean ± SD (minisamples (reference seawater).

GWAa GWBa G

BOD 10 17 1TSS 1600 240 3Sulphate 2190 1720 2NH3 (N) 190 290 1Phosphate-P 8 3 3Nitrate-N 20.5 <5 9Nitrite-N <5 <5 <

TPH 5640 770 <

Metals As 100 90 1Cd 5 2 5Cr 90 20 2Cu 130 50 1Fe 16200 9200 1Pb 430 100 1Mn 340 210 1Hg 1.2 0.1 0Ni 50 20 3Zn 1730 1000 1

PAHS Ae <0.4 <0.4 0Ay <0.2 <0.02 <Ant 0.4 <0.02 <Fluo 0.5 0.05 0N <0.2 0.09 0Ph 1.2 0.03 0B[a]A 0.7 0.02 <BaP 0.45 0.017 <Ch 0.6 <0.02 <DBA <0.2 <0.02 <F 1.4 0.04 0P 1.1 0.05 0B[b]F 0.5 0.02 <Bper 0.3 <0.02 <B(k)F 0.4 <0.02 <IndP 0.3 <0.02 <

R PCBs 1.1 2.09 1

PCDF TEQs 23.9 2.09 1PCDD TEQs 25.7 4.03 3

a Data from Chapman (2008).

have a mean of zero and standard deviation of 1 and all contami-nant values varying over roughly the same limits (typically �2 to+2) (Zar 1999). Data were normalized for each metal (Al, As, Cr,Cu, Fe, Mn, Hg, V, and Zn), both down a transect line (i.e., 1A, 1B,1C, 1D, 1E, etc.) (yielding information on how each metal varieswith distance from the dump on a comparable scale), and alsofor all A series, B series, C series sites, etc. (yielding informationon how each contaminants varies between each transect line ona comparable scale).

Ordination techniques (non-metric multidimensional scaling,MDS) based on Euclidean distance were used to examine spatialpatterns in log-transformed metal data using the multivariate sta-tistical analysis package PRIMER version 6 (Clarke and Gorley,2006). The ANOSIM subroutine (one-way ANOSIM) of the PRIMERpackage (Clarke and Gorley, 2006) was used to test whether therewas any significant difference between the sampling sites (A series,B, series, etc.) and reference sites.

2.2. Sediment quality guidelines (SQGs)

Sediment metal (As, Cr, Cu, Hg, Ni, Pb, and Zn), PAHs and totalPCBS concentrations are examined with respect to two sets ofSQGs, the effects range low (ERL)/effects range median (ERM)

leum hydrocarbons (C6–C50 for groundwater or C10–C40 for seawater), metals, andPH, metals, and PAHs) measured in seawater collected from boreholes drilled into them 1 m depth either immediately beside the dump or reference (ambient) seawater

mum–maximum) and represent n = 8 independent samples (beside the dump) or n = 6

WDa Dump Reference

5 <5 <520 8.9 ± 4.5 (3.0–14.3) 13.3 ± 3.7 (7.7–17.1)210 2736 ± 81 (2592–2830) 2895 ± 14 (2880–2910)90 67.8 ± 70.6 (0.02–154) <5

8.7 ± 2.2 (5.2–12.2) 3.3 ± 0.2 (3.1–3.7).7 <0.5 0.13 ± 0.1 (0.1–0.23)5 <0.1 <0.1

250 143.8 ± 123.7 (100–450) <100

00 0.71 ± 0.47 (0.22–1.48) 1.08 ± 0.08 (0.99–1.21)0.03 ± 0.02 (0.02–0.07) 0.02 ± 0.01 (0.02–0.04)

0 <50 <5010 2.8 ± 6.3 (0.12–18.4) 0.39 ± 0.13 (0.27–0.63)8800 <10 <1070 0.3 ± 0.5 (0.05–1.5) 0.48 ± 0.50 (0.13–1.16)90 26.3 ± 24.3 (0.3–56.8) 0.38 ± 0.09 (0.31–0.54).2 <0.01 <0.010 0.5 ± 0.15 (0.3–0.7) 0.26 ± 0.02 (0.23–0.28)250 44.7 ± 72.4 (3.8–204) 0.86 ± 0.28 (0.64–1.4)

.04 0.01 ± 0.002 (0.01–0.02) <0.010.02 <0.01 <0.010.02 <0.01 <0.01.04 0.02 ± 0.006 (0.01–0.03) <0.01.1 0.05 ± 0.05 (0.01–0.15) <0.02.03 0.02 ± 0.007 (0.02–0.04) <0.020.02 <0.01 <0.010.005 <0.01 <0.010.02 <0.01 <0.010.02 <0.01 <0.01.02 0.01 ± 0.002 (0.01–0.02) <0.01.02 0.02 ± 0.007 (0.01–0.03) <0.010.02 <0.01 <0.010.02 <0.01 <0.010.02 <0.01 <0.010.02 <0.01 <0.01

.06 ns ns

.06 ns ns

.68 ns ns


values described by Long et al. (1995) and threshold effects levels(TEL)/probable effects levels (PEL) values described by MacDonaldet al. (1996) SQGs for Polychlorinated dibenzo-p-dioxins (PCDDs)and polychlorinated dibenzo furans (PCDFs) were obtained fromCCME (2002) as toxic equivalents (TEQs), based upon world healthorganization (WHO) 2005 TEF values for fish (van den Berg et al.,2006). Sediments were sorted into three categories (A) no ERL/TEL exceeded, (B) >ERL/TEL but <PEL/ERM (C) >PEL/ERM.

3. Results


Surface (1 m deep) seawater samples collected beside the dumphad elevated levels of ammonia, with a maximum value of154 lg L�1 ammonia-N, �30 � higher than background levels(<5 lg L�1) (Table 2). Total phosphate (as P) levels (maximum12.5 lg L�1) were consistently 2–4� higher than at the referencesites (<3 lg L�1). Nitrate and nitrite were not distinguishable fromambient levels. Elevated concentrations of Zn (maximum204 = lg L�1), Mn (maximum = 57 lg L�1), and Cu (maxi-mum = 18.4 lg L�1) levels were also measured in seawater sam-ples beside the dump, corresponding to a 145�, 105� and 30�increase relative to maximum values recorded in the reference(ambient) seawater samples.

The TPH C10–C40 detection limits of 0.1 mg L�1 was exceeded inonly 1 sample collected from beside the dump (0.45 lg L�1). PAHlevels in most water samples were below the detection limits forthe 16 US-EPA priority PAHs, but in samples 3–8, slightly elevatedlevels of the low molecular weight (LMW) PAHs naphthalene,phenanthrene, pyrene, fluorene, acenapthene, fluoranthene wererecorded. The highest individual PAH level was 0.15 lg L�1 naptha-lene, which was 7.5� higher than in the background (ambient)samples.

Zinc levels in two of the water samples beside the dump (103and 204 lg L�1) exceeded the Florida surface water quality stan-dards for class 3 (marine waters) of 86 lg L�1 and 3 of the samplesexceeded the ANZECC/ARMCANZ (2000) trigger value of 15 lg L�1

(for 95% protection). Copper concentrations (18.4 lg L�1) in one ofthe samples beside the dump exceeded the Florida standards forclass 3 (marine waters) of 3.7 lg L�1and two samples exceededthe ANZECC/ARMCANZ (2000) 9% percentile trigger value of1.3 lg L�1.

Fig. 2. Ternary diagram of sand–silt–clay size distributions (after Shepard 1954)based on the Udden–Wentworth (Wentworth, 1922) US standard classificationscale of sediments.

3.2. Sediment analysis

Most of the sampling sites in Castle Harbour were located with-in an area dredged during the 1940s (Fig. 1) and, with the excep-tion of the shallow (1.8 m deep) sample collected from thelagoon behind the dump, all samples were taken from generally>12 m water depth. Sediments close to the dump were generallygreen/dark brown in pigmentation and easily re-suspended by di-vers. Reference samples were a much paler, light grey colour butwere also very fine and easily re-suspended. The most conspicuousfauna included the upside down jelly fish, Cassiopea xamachana,which was abundant in some cases in large numbers on most ofthe B, C and D series of sites, especially in the more easterly tran-sects (line 5–9).

All sediment samples had relatively high silt/clay (<63 lm) con-tent (i.e., median = 69%, range 25–90%) with the exception of thesample from the lagoon which was predominantly sand. Ternarydiagrams for the textural classification of sediments on the basisof sand/silt/clay ratios show most sediments were clayey-silt andsandy-silt and silt sand (Fig. 2). Moisture content averaged 55%in all samples and TOC ranged from �2% to 3% for most sediments

(Table 2) and levels in samples closest to the dump (average = 2.8%,range 1.6–4.1%) were comparable to TOC levels at the referencesite 1.6 km away (average = 2.1, range 1.6–2.8%). Overall, therewas no evidence for a decrease in TOC levels with increasing dis-tance from the dump.

Chemical data summaries are shown in Table 3. With the excep-tion of Ni, all metals showed a trend of decreasing concentrationwith increasing distance from the dump (Table 3, Figs. 3A–F and8A). In the A samples (collected 2 m from the dump edge) Zn, Cu,Hg were 10–20 � higher (i.e., enriched) than at the reference sites(R samples) decreasing to �1.5 � higher at a distance of 680 mfrom the dump (Figs. 3 and 8A). Fe, Mn, As, Ch and Pb were 3–5 � higher in the A samples compared with the reference (R) sam-ples decreasing to < 1.3 � in samples 680 m from the dump (Figs. 3and 8A).

A non-metric Multidimensional Scaling (MDS) ordinance plot oflog-transformed metal concentrations of the sediments based onEuclidean distance shows a clear contamination gradient fromthe right to the left hand side of the ordination (Fig. 4). A one-way ANOSIM test produces a global R value of 0.364, suggestingthere are distinct differences between the a priori designatedgroups of samples which are based on their distance from thedump (i.e., A or B samples, etc.). Further pair-wise testing indicatessignificant (p < 0.05) differences between the Reference sites (R)and all other groups of sites except the ‘E’ series which were col-lected 680 m from the dump.

Individual PAH compounds ranged from below detection limitsto a maximum 500 lg kg�1 (pyrene) with a range of 125–3085 lg kg�1 DW for the summed concentrations of the 16 priorityUS PAHs (Table 3). HMW PAHs were found in greater concentrationthan LMW PAHs (Table 3, Fig. 5A and B) and PAHs showed a cleartrend towards decreasing concentrations with distance from thedump (Figs. 5A–C and 8B). RUS-EPA 16 Priority PAHs were�10 � higher (i.e., enriched) in the A samples, decreasing to �4�higher at a distance of 170 m from the dump (Table 3). The Phen-anthrene [Ph]/anthracene [ant] ratio (Gschwend and Hites, 1981)plotted against the fluoranthene [F] to pyrene [P] ratio (see Sicreet al., 1987), shows the data points fall mostly in the pyroliticquadrant (i.e., Ph/Ant < 10 and F/P > 1) rather than petrogenicquadrant (i.e., Ph/Ant > 15 and F/P < 1) (Fig. 5E).

Table 3Mean (min–max) metal and metalloid concentration PAH, PCB, PCDDs and PCDFs and WHO-TEQ concentrations in sediment samples in Castle Harbour, where ‘A (2 m)’ refers tothe ‘A’ series of sediment samples which were collected 2 m away from the edge of the dump, etc. (see Fig. 1).

Lagoon A (2 m) B (20 m) C (80 m) D (160 m) E (680 m) R (1.6 km)

Depth (m) 1.8 12.3 (8.9–15.2) 12.2 (8.9–15.2) 13.8 (8.2–15.8) 14.2 (8.2–15.8) 11.9 (7.0–15.8) 15.8 (15.6–15.8)Moisture (%) 62.7 53.2 (46.1–66.7) 55.7 (44.9–60.8) 58.1 (46.7–62.6) 54.9 (40.2–61.7) 50.5 (39.7–62.0) 49.9 (46.9–53.7)TOC (%) 4.1 2.8 (1.6–4.1) 2.6 (1.6–3.4) 1.7 (0.7–2.7) 2.0 (1.2–2.7) 1.5 (1.2–2.0) 2.1 (1.6–2.8)% Silt and clay 66. 62 (31–85) 59 (28–99) 69 (25–95) 81 (56–94) 79 (52–97) 78 (73–84)

Metals and metalloid (mg kg�1, lg g�1)Al (g kg�1) 4.6 10.1 (5.7–15.6) 9.2 (4.2–17.6) 6.4 (3.3–11.2) 3.9 (1.2–5.2) 2.5 (0.84–3.8) 2.0 (1.4–2.6)Arsenic 8.6 11.2 (6.8–28.9) 8.1 (5.4–10.7) 6.9 (5.7–8.2) 5.6 (3.2–6.7) 3.9 (2.3–5.3) 3.4 (2.7–4.3)Cadmium <8 2.1 (0.50–8.7) 1.1 (0.50–2.0) 1.1 (0.50–1.8) 1.2 (0.25–2.5) 1.3 (0.75–2.5) 1.3 (0.75–1.8)Chromium 33.5 52.4 (33.0–75.0) 51.2 (29.5–86.4) 41.4 (25.5–64.8) 29.3 (10.6–38.2) 21.8 (10.8–30.1) 18.7 (15.7–21.3)Copper 46.8 57.7 (15.8–159) 29.7 (8.5–74.4) 15.7 (10.6–19.7) 9.4 (3.0–12.5) 4.8 (1.5–8.5) 3.8 (3.3–4.5)Iron (g kg�1) 9.23 7.0 (4.1–11.1) 6.3 (3.1–10.5) 4.6 (3.5–6.9) 2.9 (1.3–3.6) 1.9 (0.8–2.6) 1.5 (1.1–1.9)Lead <45 106 (30.0–259) 52.6 (30.0–143.0) 50.0 (30.0–75.0) 46.7 (15.0–75.0) 50.0 (30.0–60.0) 35.0 (30.0–45.0)Manganese 77.3 96.9 (45.7–165) 76.8 (31.4–140) 55.8 (36.0–88.6) 36.2 (14.4–44.6) 24.8 (9.8–36.2) 24.2 (20.3–27.2)Mercury (lg kg�1) 271 390 (116–1040) 288 (76.0–712) 171 (99.7–232) 106 (43.2–140) 62.3 (29.2–91.5) 39.7 (31.5–50.2)Nickel <15 8.9 (5.0–21.0) 6.9 (5.0–7.5) 8.3 (5.0–12.5) 7.8 (2.5–12.5) 8.3 (5.0–10.0) 5.8 (5.0–7.5)Zinc 1160 563 (148–1380) 325 (75.4–703) 162 (102–214) 88.2 (31.1–111) 42.1 (16.6–68.4) 26.9 (22.2–32.5)

Polycyclic aromatic hydrocarbons (PAHs) (lg kg�1, ng g�1)RLMW PAHs 55 236 (55.0–435) 113 (50.0–325) 81.7 (40.0–165) ns ns 41.7 (35.0–50.0)RHMW PAHs 385 843 (185–1840) 450 (155–1210) 294 (135–580) ns ns 53.3 (40.0–70)RTotal PAHs 440 1079 (240–2135) 563 (230–1535) 376 (175–745) ns ns 95.0 (80.0–120)R16 US-EPA 685 1469 (330–3085) 777 (305–2035) 530 (250–1025) ns ns 135 (125–150)

Polychlorinated biphenyls (PCBs) (lg kg�1) or ng kg�1 (pg g�1) for TEQsRPCBs 130 100 (15.0–450) 93.7 (25.0–190) 37.7 (25.0–45.0) 7.8 (6.3–10.0) 3.0 (2.0–3.6) 1.1 (0.8–1.4)

WHO-2005 toxic equivalents (TEQs) based on toxic equivalency factors (TEFs) for fishR-TEQPCBs 1.9 1.78 (0.05–7.41) 1.54 (0.07–3.42) 0.63 (0.08–0.94) 0.02 (0.02–0.03) nd nd

Lagoon A (2 m) B (20 m) C (80 m) D (160 m) E (680 m) R (1600 m)

Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) (ng kg�1, pg/g)2378 TeCDD <0.5 0.14 (0.05–0.38) 1.1 (0.15–2.9) 0.20 (0.15–0.25) 0.10 (0.10–0.10) 0.13 (0.10–0.20) 0.12 (0.05–0.15)12378 PeCDD <0.4 0.63 (0.05–2.0) 0.85 (0.10–2.3) 0.17 (0.10–0.20) 0.05 (0.05–0.05) 0.07 (0.05–0.10) 0.08 (0.05–0.10)123478 HxCDD <0.6 0.81 (0.17–2.3) 1.2 (0.35–2.8) 0.49 (0.25–0.98) 0.44 (0.15–0.80) 0.10 (0.10–0.10) 0.26 (0.10–0.58)123678 HxCDD 3.3 5.3 (1.6–15.0) 6.3 (2.8–13.0) 2.8 (1.5–5.0) 1.2 (0.95–1.50) 0.46 (0.15–0.63) 0.48 (0.10–1.20)123789 HxCDD 1.9 2.4 (0.72–6.1) 3.2 (1.3–6.7) 1.4 (0.75–2.3) 0.74 (0.61–0.89) 0.35 (0.10–0.54) 0.40 (0.10–1.0)1234678 HpCDD 66 145 (37.0–380) 154 (71–290) 62.7 (42.0–100) 23.0 (21.0–24.0) 10.8 (7.5–14.0) 7.5 (3.7–15.0)OCDD 520 1350 (300–3700) 1370 (610–2400) 547 (360–850) 170 (150–190) 86.7 (59.0–110) 51.0 (29.0–93.0)Dioxins 720 1640 (384–4490) 1722 (775–3075) 400 (461–1071) 229 (205–250) 116 (81.3–149) 73.7 (41.2–137)2378 TeCDF 5.8 0.66 (0.10–2.2) 0.97 (0.1–2.6) 0.20 (0.15–0.25) 0.10 (0.05–0.20) 0.08 (0.05–0.15) 0.08 (0.05–0.10)12378 PeCDF 0.4 1.0 (0.05–3.2) 1.3 (0.1–3.6) 0.47 (0.20–1.0) 0.10 (0.05–0.20) 0.07 (0.05–0.10) 0.08 (0.05–0.10)23478 PeCDF 2.9 1.2 (0.05–5.4) 2.2 (0.1–6.4) 0.60 (0.15–1.5) 0.07 (0.05–0.10) 0.05 (0.05–0.05) 0.05 (0.05–0.05)123478 HxCDF 1.2 2.6 (0.58–8.7) 3.9 (1.2–9.0) 1.4 (0.79–2.3) 0.59 (0.53–0.69) 0.15 (0.05–0.23) 0.20 (0.05–0.51)123678 HxCDF 1.2 2.6 (0.60–7.6) 3.9 (1.4–8.8) 1.0 (0.53–1.7) 0.44 (0.35–0.59) 0.17 (0.05–0.29) 0.20 (0.05–0.51)123789 HxCDF <0.4 0.84 (0.05–3.7) 1.7 (0.25–4.1) 0.39 (0.15–0.81) 0.18 (0.05–0.28) 0.05 (0.05–0.05) 0.05 (0.05–0.05)234678 HxCDF 1.9 4.0 (0.90–11.0) 5.5 (1.6–13.0) 1.9 (1.1–3.2) 0.80 (0.63–0.97) 0.28 (0.05–0.40) 0.32 (0.05–0.85)1234678 HpCDF 13 38.2 (11.0–91.0) 40.7 (16–82) 14.7 (10–23) 5.7 (5.3–5.9) 2.1 (1.6–2.8) 2.2 (0.71–5.1)1234789 HpCDF 0.78 2.6 (0.45–8.1) 3.8 (1.0–9.2) 1.1 (0.30–1.90) 0.27 (0.15–0.46) 0.12 (0.10–0.15) 0.23 (0.10–0.48)OCDF 18 79.1 (19.0–220.0) 78.0 (27–150) 33 (19–52) 9.3 (8.00–11.0) 3.3 (2.1–4.9) 3.0 (0.25–6.9)RFurans 160 213 (44.3–651) 240 (62–515) 47.9 (50.6–143) 24.8 (21.0–28.6) 8.5 (5.1–13.0) 10.6 (1.2–28.0)RPCDDs/Fs 880 1853 (429–5141) 1961 (837–3590) 448 (512–1214) 253 (230–271) 125 (86–162) 84.4 (42.4–165)

WHO-2005 toxic equivalents (TEQs) based on toxic equivalency factors (TEFs) for fishR-TEQPCDD/F 3.9 5.37 (1.43–15.46) 7.70 (2.42–17.81) 1.49 (1.75–3.89) 0.95 (0.93–0.97) 0.54 (0.50–0.60) 0.53 (0.38–0.83)


Total PCB levels (i.e., R209 congeners) decreased with increas-ing distance from the dump, from an average of 100 ± 138 lg kg�1

DW (mean ± SD, n = 9 samples) in samples closest to the dump (i.e.,A samples) to 1.1 ± 0.3 lg kg�1 DW (mean ± SD, n = 3 samples) atthe reference sites (R, Table 3, Fig. 6A). The highest total PCB con-centration was 450 lg kg�1 DW in one of the A series of samplescollected 2 m from the dump edge (Table 3, Fig. 6A).

The tetra-, penta- and hexa-CBs were the most dominant PCBhomologues in the sediments, comprising �65–75% of the total(Fig. 6B). In the sample from the lagoon, the mono-tri, and tetra -CBs were most dominant, comprising >90% of all homologues(Fig. 6B). The top ten most numerically dominant congeners in sed-iments in Castle Harbour were PCB 110 > PCB 90/101 > PCB118 > PCB 138 > PCB 153/168>, PCB 43/52, PCB 95, PCB 87, PCB147/149, PCB 70 (Fig. 6C), whilst in the sediment sample collectedfrom the lagoon (L) the most common congeners were PCB 43/52,PCB 48/49, PCB28, PCB44, PCB31, PCB15 (Fig. 6C).

Total dioxin (PCDD + PCDF) levels decreased with increasingdistance from the dump, from an average of 1853 (max 5141)ng kg�1 DW in samples closest to the dump to 84.4 (max 165)ng kg�1 DW (mean ± SD, n = 3 samples) at the reference sites (R,Table 3, Fig. 7A and B). Sediments beside the dump and at the ref-erence sites 1.6 km from the dump were dominated by OCDD (70–73%), >HpCDD (14–15%), HpCDF (4–5%) > OCDF (�4%) > HxCDD(2.5%) homologues (Fig. 7C).

TEQs for PCBs and dioxins ranged from a maximum of 7.41 and17.81 ng kg TEQs, respectively, beside the dump to 0.01 and0.38 ng kg�1 TEQs, respectively, at the reference site (Table 3,Fig. 7D and E). Overall, PCDD/PCDFs contributed 2–5� more tothe total TEQs than PCBs. The highest total TEQs (i.e., Rdioxin-likePCBs + PCDD and PCDFs TEQs) were 22.87 and 21.23 in sample 2Aand B (2 m and 20 m from the dump) (Table 3, Fig. 7D and E).

Total petroleum hydrocarbons (TPH C10–C40) were only mea-sured at a limited number of sites (Table 1), and were only greater

Fig. 3. (A–H) Metal concentrations (mg kg�1 DW) in sediments from Castle Harbour (A–D, E, R – see Fig. 2), where each bar represents an individual sediment sample,numbered left to right from 1 to 9 (samples A–E), 1–3 (R samples. Inset figures represent the mean metal concentration ± SD n = 3–9 samples per location. Horizontal linesrepresent (where available) the TEL (threshold effects level), ERL (effects range low), PEL (probable effects level) and ERM (effects range median) (see text).

Fig. 4. Two-dimensional non-metric multidimensional scaling (MDS) ordination oflog-transformed metal concentrations of the sediments surrounding the dumpwhere the numbers (1–9) refer to which transect lines the samples were taken from(see Fig. 2) and the symbols refer to the distance from the dump.


than the detection limits (200 mg kg�1 DW) in 4 of the 6 sitesimmediately beside the dump (max = 500 mg kg�1 DW), and inthe lagoon behind the dump (L, 345 mg kg�1 DW) (Table 3).

Of the 19 organochlorine pesticides examined for in the A seriesof sediment samples (2 m from the dump), (including aldrin, DDD,DDE, DDT dieldrin, endrin, heptachlor, and mirex), 17 organophos-phate pesticides (including chlorpyrifos, diazinon, parathion,mevinphos, and malathion), and 10 carbamate pesticides (includ-ing aldicarb, carbofuran, and carbaryl), only the organochlorinescis-chlordane (7 lg kg�1 DW), trans-chlordane (7 lg kg�1 DW),and nonachlor (6 lg kg�1 DW) (i.e., 20 lg kg�1 DW Total chlor-dane) were detected, and only in sample 6 A. This sample was closeto where active in-filling was occurring. No pesticides were de-tected in the sediment from the lagoon behind the dump.

Metal data were normalized (see Section 2) to permit analysesof spatial patterns of metal contamination. When normalized onthe basis of distance from the dump, metal levels are clearly higher

closer to the dump and decrease systematically with distance fromthe dump (Fig. 8C). When normalized on the basis of transect line,metal concentrations showed an inconsistent pattern tending to behighest along transects 2, 7 and 8 and lower on transect lines 3, 4(Fig. 8D).


Multiple low-range SQG (TEL and ERL) thresholds were ex-ceeded for metals in sediment sampled close to the dump (i.e., Aand B series and the sample from the lagoon) (Fig. 3, Table 4);for As, and especially Hg and Zn low-range thresholds were alsoexceeded in some of the C samples (collected 80 m from the dump,Fig. 3, Table 4), and for a single D sediment sample for Hg (TELonly). No thresholds were exceeded for the E series of samples orsamples collected at the reference locations (R samples, Fig. 3,Table 4).

Multiple mid-range thresholds (PEL and ERMs) were exceededfor Zn and Hg (and in one instance the Cu PEL) in sediment samplesimmediately beside the dump, and for Zn and Hg in some samplescollected 20 m from the dump (Fig. 3, Table 4). The sediment sam-ple from the lagoon exceeded the As, Cu, Hg and Zn low-rangethresholds and the Hg high range threshold (Fig. 3, Table 4).

For PAHs, ERL thresholds were exceeded for acenapthene,anthracene, fluorene, phenanthrene, benzo(a)anthracene in sam-ples closest to the dump, acenapthene and fluorene 20 m fromthe dump, and fluorene in a samples collected 80 m from the dump(Table 4). The low range (ERL) threshold was also exceeded for thesum of the high molecular weight (HMW) PAHs in a single samplecollected from beside the dump (Fig. 5B, Table 4). No ERM guide-lines were exceeded for any individual PAHs or LMW, HMW or to-tal PAHS (RPAHs) (Table 4).

Low-range SQGs were exceeded for RPCBs (209 congeners) insediments up to the C series of sampling sites (80 m from thedump) but not in the D and E series (Fig. 6A, Table 4). ERL/ERMor TEL/PEL SQGs are not available for PCDDs and PCDFs (Longet al., 1995; MacDonald et al., 1996); however, an interim sedimentquality guideline (ISQG) for marine sediments of 0.85 ng kg�1 TEQ

Fig. 5. (A–E) PAHs. (A) Low molecular weight PAH (LMW) and (B) high molecular weight (HMW) PAH (as defined in Long et al., 1998) and (C) total (i.e., LMW and HMW) PAHconcentrations (lg kg�1 DW) in sediments collected 2 m, 20 m and 80 m (A, B, C, respectively) from the Castle Harbour dump site (see Fig. 1), where R = reference sites located1.6 km from the dump. Inset figures represent mean ± SD n = 6 samples at each location or 3 at the reference sites. Horizontal lines represent (where available) the TEL(threshold effects level), ERL (effects range low), PEL (probable effects level) and ERM (effects range median) (see text). (D) HMW and LMW PAH composition of the 16 US-EPApriority PAHs in sediment sample 2 m away from the dump edge (shown as an average of all nine samples ± SD (E) double ratio plots of phenanthrene/anthracene versusfluoranthene/pyrene (see text).

Fig. 6. PCBs. (A) Total PCB concentration (lg kg�1 DW) (inset Fig. represents mean ± SD n = 9 samples at each location or 3 at the reference sites (B) mono-tri (1–3), tetra (4),penta (5), hexa (6), hepta (7), octa (8) and nona-deca (9–10) chlorinated PCB homologue profiles in sediments from various distances from the dump (samples A through E)and at the reference site (R) located 1.6 km from the dump and from the lagoon (L) behind the dump (see Fig. 1). (C) Total PCBs (R209 congeners) as IUPAC congener numbers,in samples collected beside the dump (A), 20 m away from the dump (B) and 80 m away (C) and in the lagoon (L) behind the dump – note different y-axis scales.


and a probable effects level (PEL) of 21.5 ng kg�1 TEQ (based onWHO 1998 TEFs for fish) have been published by the CanadianCouncil of Ministers of the Environment (CCME, 2002). These

guidelines were derived using similar techniques to NationalStatus and Trends Program (NSTMP) of MacDonald et al. (1996)(CCME, 2002). Many of the Castle Harbour sediment samples

Fig. 8. (A) Average enrichment of sediments relative to the reference sites (R, seeFig. 1) by (A) metals and (B) PCBs, PCDDs/PCDFs and PAHs, as a function of distance(m) from the edge of the dump. Mean ± SD metal (Al, As, Cr, Cu, Fe, Mn, Hg, and Zn)concentration after normalizing for each metal down a transect line (i.e., 1A, 1B, 1C,1D, 1E, etc.) which yields information on how each metal varies with distance fromthe dump on a comparable scale, and for all A series, B series, C series sites, etc.,which yields information on how each metals varies on each transect line.

Fig. 7. Dioxins/furans. (A) Total PCDD concentration (ng kg�1 DW) and (B) PCDF concentration (ng kg�1 DW) in sediments from Castle Harbour (Fig. 1), where each barrepresents an individual sediment sample and where A–E represent distances from the dump site – see Fig. 1. Reference samples (R) were collected 1.6 km from the dump. (C)2, 3, 7, 8 chlorine-substituted congener % in a sample 2 m from the dump (A sample) and a sediment samples from the site (R). Inset figures represent the meanconcentration ± SD (n = 9, or n = 3 samples for the reference sites). Horizontal lines represent the CCME (2002) ISQG for TEQs.


exceed the ISQG including all sediment samples 170 m from thedump (Table 4); however, no samples exceeded the mid-rangethreshold.

The ERL and ERM SQGs for total chlordane (0.5 and 6 lg kg DW)were exceeded in the single sediment sample in which the pesti-cide was detected.

4. Discussion

Concurrent measurements of seawater beside the marine land-fill (seafill), measurements of ground water inside the dump, andmeasurements of chemical concentrations in sediments surround-ing the dump present a very clear and coherent picture of the leak-ing/leaching of contaminants and accumulation in surroundingsediments. Many of the contaminants (chlordane, PCBs, and diox-ins/furans) are on the original ‘dirty-dozen’ list of persistent organ-ic pollutants (POPs), identified under the Stockholm convention(Kaiser and Enserink, 2000), and many contaminants have reachedlevels where biological effects are possible and in some cases prob-able. Comprehensive assessments of the occurrence and distribu-tion of contaminant in tropical waters are lacking (Fernandezet al., 2007), and especially areas close to reefs. A few metres awayfrom the leading edge of the seafill there is a small, denuded coralpatch reef (32�21045.3800N, 64�41046.8100W). Given the 25 acre sizeof the seafill, the proximity of this reef to the seafill, and given theevidence of leaching of contaminants described in this report, thereef is proposed here as being the most contaminated in the world.


Seawater nutrient, TPH, metal and PAH concentrations in CastleHarbour are either very low or below detection limits, as expectedgiven the direct connection to oligotrophic open ocean seawatercharacteristic of the Sargasso Sea. However, water collected atlow tide from 1 m depth immediately beside the dump was en-riched in ammonia and phosphate, some PAHs and several metalsincluding Cu, Mn and especially Zn. Whilst ammonia is well recog-nized as one of the most significant long-term pollutants fromlandfills (Christensen et al., 2001), the Castle Harbour dump doesnot receive biodegradable waste and the source of the ammoniais presently unknown. The enrichment of seawater by metals isconsistent with the predominantly metallic waste disposed of at

Table 4Exceedances of metal, PAHs low molecular weight (LMW) PAH, high molecular weight (HMW) PAH, RPAHS, RPCBs, RWHO-TEQs PCDDS/PCDFs TEL/PEL and ERL/ERM sedimentquality guidelines (SQGs) in marine sediment samples from Castle Harbour (samples A through E, and R and L, where R refers to reference samples and L refers to a sample takenfrom the lagoon behind the dump, see Fig. 2). ns = not sampled.

% of samples <ERL, >ERL and <ERM, >ERM

>TEL >ERL >PEL >ERM A B C D E R L

Metals (49 samples)Arsenic 37% 27% 0% 0% 1/8/0 6/3/0 8/1/0 9/0/0 9/0/0 3/0/0 0/1/0Chromium 18% 2% 0% 0% 9/0/0 8/1/0 9/0/0 9/0/0 9/0/0 3/0/0 1/0/0Copper 35% 16% 2% 0% 4/5/0 7/2/0 9/0/0 9/0/0 6/0/0 3/0/0 0/1/0Mercury 57% 43% 4% 4% 2/6/1 2/6/1 3/6/0 9/0/0 9/0/0 3/0/0 0/1/0Zinc 51% 47% 22% 16% 1/4/4 1/5/3 3/6/0 9/0/0 9/0/0 3/0/0 0/0/1

PAHs (22 samples)Ae 27% 14% 0% 0% 4/2/0 5/1/0 6/0/0 ns ns 3/0/0 3/0/0Ay 0% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0Ant 18% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0Fluo 14% 23% 0% 0% 3/3/0 5/1/0 5/1/0 ns ns 3/0/0 3/0/0mN 0% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0N 0% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0Ph 23% 5% 0% 0% 5/1/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0RLMW PAHs 14% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0B[a]A 23% 5% 0% 0% 5/1/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0BaP 23% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0Ch 18% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0DBA 32% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0F 32% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0P 18% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0RHMW PAHs 18% 5% 0% 0% 5/1/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0RTotal PAHs 9% 0% 0% 0% 6/0/0 6/0/0 6/0/0 ns ns 3/0/0 3/0/0

PCBs (25 samples)R-PCBs 52% 52% 8% 8% 3/6/1 0/3/1 0/3/0 3/0/0 3/0/0 3/0/0 0/1/0

PCDDs/FR-WHO-TEQPCDD/F 76% 0% 0/9/0 0/3/0 0/3/0 0/3/0 3/0/0 3/0/0 0/1/0


the dump. The levels of copper and especially zinc in the seawaterbeside the dump were highly elevated, with the highest concentra-tion (204 lg Zn L�1) �140� greater than ambient levels. On multi-ple occasions water from beside the dump exceeded the Cu andespecially Zn Florida class 3 guidelines for marine waters and theAustralian (ANZECC/ARMCANZ) water quality guidelines for sensi-tive habitats. For organic contaminants such as TPHs, and for somehigh molecular weight PAHs, the levels in the seawater surround-ing the dump were only slightly above detection limits, but werenonetheless detectable as compared with background seawaterlevels.

4.2. Sediment analyses

Sediments around the dump had high silt content and weremostly classified as clayey-silt and sandy-silt, consistent with pastdredging activities in Castle Harbour (see Flood et al., 2005). TOClevels were 2–3%, with no evidence of higher organic content insediments closer to the dump. In more comprehensive surveys ofTOC levels in �100 sediment samples collected in inshore andnearshore areas of Bermuda (conducted at the same time of thepresent study) TOC levels were 2.7% ± 2.3 (mean ± SD), but withvalues as high as 18.5% in protected creeks and bays (Jones unpub-lished data). On this basis, and considering that the dump does notreceive organic waste, it is likely it is not contributing to organicenrichment of the sediments in Castle Harbour.

Sediments neighbouring the landfill were screened for the ma-jor contaminant classes (TPHs, metals, pesticides, PAHS, and PCBs,PCDDs/PCDFs). The results show a clear pattern of localized enrich-ment and a systematic decrease in contamination with increasingdistance from the dump, i.e., there is a halo of contaminationaround the dump. Contaminants found to be elevated in thegroundwater studies of the dump (see Chapman, 2008) were alsofound to be elevated in seawater leaching out of the dump, and

subsequently found to be elevated in neighbouring sediments.Thus, for example, groundwater levels of Zn are 1700 lg L�1, con-sistent with high levels in seawater measured just outside of thedump (up to 200 lg L�1 as compared with a background level of1 lg L�1), and consistent with high levels found in the sedimentsbeside the dump where levels up to 1380 mg kg�1 were measured(compared with background levels of �25 mg kg�1). Some discrep-ancies between the concurrent studies of groundwater/seawater/sediment contamination include elevated Cd, Pb and especiallyFe levels in the groundwater which was not measured in the adja-cent water. Nevertheless, these metals were elevated in the sedi-ments surrounding the dump. Since there is no local agriculture,industry and little urbanization on the Castle Harbour shoreline,and since there are no outfalls, few storm drains and no rivers thatdrain into the basin, the contamination can almost exclusively beattributed to the release of contaminants from the dump, i.e., itis a local point source of pollution.

Assessing the areal extent of which the dump has contaminatedCastle Harbour, and estimating the enrichment or concentrationfactor (CF) of contaminants, depends upon accurately knowingthe natural background level (NBL), i.e., the level in the absenceof human activity. The choice of control or reference sites in thisstudy was difficult because the dump is located close to previouslydredged areas, composed of fine, predominantly sandy or clayey-silt. Outside of the dredge scar sediments were mostly sand or siltysand. Since contaminant concentrations generally increase withdecreasing grain size in most estuarine and coastal sediments (Lor-ing, 1991), it was necessary to sample sediments within thedredged scar to obtain at least comparable sediment grain sizecharacteristics to the other samples collected nearer the dump.The location of the reference sites (i.e., NBL) in this study wastherefore a compromise between attaining the maximum distancefrom the dump as possible, and yet remaining within the originaldredge scar, hence permitting sampling comparable sediment


grain size distributions. Any specified concentration factors for allcontaminant classes should therefore be considered minimum val-ues, appreciating there could still be contamination from the dumpat the reference sites.

With the exception of Nickel, all metals showed a clear trend ofdecreasing concentration with increasing distance from the dump.Zn, Cu and Hg concentrations beside the dump were 10–20� high-er than at the reference sites followed by Fe, Mn, As, Pb, and Chwhich were 3–6� higher than at the reference sites. Jickells andKnap (1984) have previously described metal enrichment aroundthe dump, albeit in a single sediment sample taken close to thedump edge. Interestingly, they considered a value of 250 mg kg�1

DW for Zn as erroneous, but given the high zinc levels in the sea-water and sediment surrounding the dump shown here (i.e., upto 1380 mg kg�1), the result seems very plausible. Analyses of sed-iment metals concentrations in samples collected�250 m from theedge of the dump in 1990 (hence also before ash block dumpingstarted) also describe much higher values of Zn (up to233 mg kg�1), Cu (up to 16.2 mg kg�1) and Pb (up to 59.1 mg kg�1)than at control locations in Castle Harbour (Knap et al., 1991a,b).Overall it appears likely that the sediments around the dump sitehave been contaminated by metals from the dump shortly afterits inception.

PAH levels in surficial sediments surrounding the dump showeda clear spatial trend of decreasing concentration with increasingdistance from the dump edge. PAHs are ubiquitous environmentalpollutants, entering the environment from unburned petroleum(petrogenic PAHs) as oil spills or natural seepages, by incompletecombustion of biomass (pyrogenic or pyrolitic PAHs), or biosynthe-sis and diagenesis (Lima et al., 2005). Some PAHs are highly carcin-ogenic or mutagenic (IARC, 1983). Due to their high octanol–water(Kow) partition coefficients PAHs have a tendency to adsorb to sed-iments (Mackay et al., 1992) and because of their persistence(IARC, 1983) they can also accumulate in bottom sediments. Thedistribution pattern and presence in the water samples is consis-tent with the dump as the source of the contamination. PAHs insamples collected 80 m away from the edge of the dump – whichwas the most distant sampling point – were still >2� the level re-corded in the reference samples (1.6 km from the dump).

PCB concentrations in sediments around the dump were �100�higher than in the background (reference) sites, and decreased sys-tematically with distance from the dump. PCBs were produced be-tween 1929 and 1978 as transformer fluids, hydraulic fluids andother industrial products (Bedard, 2003); they are highly persis-tent, cancer causing (Mayes et al., 1998) and affect the immune,reproductive, nervous and endocrine systems (Mayes et al., 1998;Aoki, 2001; Faroon et al., 2001). Due to their hydrophobic proper-ties, they tend to adsorb to natural organic matter in sedimentsand have bioaccumulative properties (Field and Sierra-Alvarez,2008). PCBs are one of the ‘dirty-dozen’ group of POPs identifiedby United Nations Environment Program (Kaiser and Enserink,2000).

Polychlorinated dibenzo-p-dioxins (PCDDs) and polychlori-nated dibenzofurans (PCDFs) (collectively ‘dioxins’) were foundin all sediment samples ranging in concentration from up to a max-imum concentration of �5100 ng kg�1 DW. Dioxins are omnipres-ent in the global environment. They are formed in mostcombustion systems, including waste incineration (such as MSW,sewage sludge, medical waste, and hazardous wastes), burning offuels (coal, wood, and petroleum products) and cement kilns, andcan be formed from primary and secondary metal operationsincluding iron ore sintering, steel production, and scrap metalrecovery (Olie et al., 1977). Dioxins can also be formed as by-prod-ucts from the manufacture of chlorine bleached wood pulp andagrochemicals. Dioxins are highly hydrophobic and are tightly ad-sorbed by soil and sediments (Kao et al., 2001). They are also one of

the ‘dirty-dozen’ group of persistent organic pollutants (POPs)identified by the United Nations Environment Program (Kaiserand Enserink, 2000). As with the other contaminant classes therewas a clear contamination gradient across Castle Harbour withhighest concentrations closest to the dump, decreasing to levelsof �40 ng kg�1 DW approximately 1.6 km from the dump edge. Re-cently, values of <2 ng kg�1 DW (total dioxins) have been mea-sured in fine sediments of the central Bermuda lagoon (Jones inprep) suggesting the sites in Castle Harbour designated as refer-ence sites could also have been contaminated by the dioxins fromthe dump (see above).

The organochlorine (OCs) pesticide chlordane was detected inone sediment sample immediately beside the dump. Technicalchlordane is a mixture of 147 compounds with c-(trans-) chlor-dane, a-(cis-) chlordane, and trans-nonachlor totaling approxi-mately 25 wt.% of the technical mixture (Dearth and Hites, 1991).All three of these components were detected in the contaminatedsample. Chlordane is toxic and carcinogenic and persistent in theenvironment, with a half life of 20–30 years (Dearth and Hites,1991). It was banned by the US-EPA in 1988 and is one of the‘dirty-dozen’ group of persistent organic pollutants (POPs) (Kaiserand Enserink, 2000). Chlordane was often used on lawns, gardensand house foundations (Harner et al., 2004), but has also been de-tected in Bermuda in rain water (Knap et al., 1991a,b) and air sam-ples (Harner et al., 2006), originating from source regions in NorthAmerica and/or Africa (i.e., trans-Atlantic transport). The contami-nated sediment sample was collected close to an area where activein-filling (dumping) was occurring and most likely represents localcontamination from building site debris/rubble rather than illegaldumping of waste pesticide or by atmospheric deposition.

TPH C10–C40 (a mixture of many different petroleum com-pounds that originate from crude oil) were detectable in severalsediment samples collected a few metres from the dump edge,but were below detection limits in all samples collected 20 mand 80 m away from the dump edge. Overall, sediment contamina-tion by mineral oils, although present close to the dump, does notappear to be significant as for other contaminant classes.

4.3. Source apportionment

The PAH profiles of Castle Harbour sediments indicate a pre-dominance of high as opposed to low molecular weight PAHs. Sev-eral molecular indices based on ratios of selected PAHconcentrations have been suggested as tools for discriminating be-tween pyrogenic (=pyrolitic) or petrogenic sources of PAHS (Read-man et al., 2002). For example a phenanthrene/anthracene ratios(Gschwend and Hites (1981) of >15 and a fluoranthene/pyrene ra-tio (Sicre et al., 1987) of >1 suggests a petrogenic source of PAHs,whilst ratios of <10 and <1, respectively, suggest more pyrogenicsources (see also Readman et al., 2002). The double ratio plots ofphenanthrene/anthracene versus fluoranthene/pyrene suggest amostly pyrogenic signature of the PAHs in Castle Harbour. Similarprofiles to those described here in Castle Harbour have been de-scribed in many parts of the world (see McCready et al., 2006and references therein) reflecting high temperature combustionas the chief source of PAHs as opposed to petrogenic sources, i.e.,fossil fuel spillages (LaFlamme and Hites, 1978).

Castle Harbour sediments were dominated by tetra-, penta- andhexa-chlorinated PCB homologues. Congener specific analysis insediments showed PCB110, PCB153/168, PCB138, PCB118, PCB90/101 are the most common, constituting �70–80% of the total.Burns et al. (1990) also reported the same dominance in individualPCB congeners in a study of PCB contamination of a single sedi-ment sample taken from beside the dump in the late 1980s. Mostof these congeners are part of the ‘ICES7’ (i.e., PCBs identified bythe International Council for the Exploration of the Sea as the most


commonly detected in environmental samples, ICES, 1986). Theseare also some of the most globally produced PCB congeners (Brei-vik et al., 2002) and the homologue and congener pattern analysis(and the high levels of PCB 110) suggest contamination by Aroclor1254 (see Johnson et al., 2008).

The PCB homologue profiles of the sediment collected from thelagoon behind the dump was unusual as compared to the othersamples collected, containing a higher proportion of mono-triand tetra-chlorinated PCBs and proportionately less penta- andhepta-chlorinated PCBs. Microorganism mediated anaerobic bio-transformations of PCBs by reductive dechlorination is a wellknown degradation pathway, resulting in an accumulation of low-er chlorinated congeners (Brown et al., 1987, 1988; Natarajan et al.,1996; Bedard, 2003; Field and Sierra-Alvarez, 2008). Manyresearchers have suggested that complete removal of PCBs fromcontaminated sediments will rely on the sequential activities ofanaerobic dechlorination and aerobic breakdown of the biphenylstructure by microorganisms (i.e., Abramowicz, 1996; Evanset al., 1996). This unusual congener profile in the lagoon sedimentscould be due to anoxia as the sediments were black and anoxic, asopposed to the olive green/brown presumably oxic sediments (seealso McCready et al., 2006) typically collected elsewhere in CastleHarbour.

Dioxin homologue profiles and 2,3,7,8 chlorine-substituted con-gener pattern analysis (‘fingerprint’ or ‘signature’ analysis) has beenwidely used to identify sources of dioxin contamination (Cleverlyet al., 1997) and in particular to determine whether soil PCDD/Fin the vicinity of incinerators were due to emissions from the plant(Domingo et al., 2001). Homologue profile analysis of the sedimentsindicated a dominance by OCDD (70%), H7CDD (15–20%) andH6CDD (5–10%), and these three homologue groups collectivelyaccounted for 90–95% of the PCDD/F concentrations. Similar pro-files, involving dominance by more chlorinated homologues hasbeen recorded in incinerator ash collected from the incinerator pitat Tynes Bay, where 30–40% are OCDD followed by H7CDD (19–26%), with the remaining homologues made up to 2–10%, HpCDF7–10% and HxCDD (Roethel, 2006). However, a similar profile hasalso been found in Bermuda soil (Peters, 2005), and this profile issimilar to that observed in various urban and rural background soilsin the US and not necessarily attributed to incinerator ash. It islikely that the high OCDD dominance reflect the higher stabilityof OCDD in soils sediments than the other PCDD/F homologues.

Overall, the ‘fingerprint’ or ‘signature’ analysis was equivocalwith respect to identifying the incinerator ash as the source ofthe dioxins in the Castle Harbour sediments, especially since thedump also accepts soil in the form of landscaping waste/debris.There are also likely to be many sources of dioxins when consider-ing the nature of the material that is dumped. When consideringthe levels of dioxins found in the incinerator ash, 8171–176,000 ng kg�1 DW sediments (Roethel 2006), as compared withBermuda soil (up to 1300 ng kg�1 DW Peters 2005), and also giventhe volume of ash disposed of at the dump in close proximity towhere the sediments were sampled, it seems unreasonable to ex-clude the incinerator ash as a source of some of the sedimentarylevels of dioxins.

Recent studies have shown that the seawater levels rise and fallwithin the dump in synchrony with the external tide (Chapman,2008) and there is no tidal dampening, as is known to occur in Ber-muda’s central Ghyben–Herzberg freshwater lens (Thomson andFoster, 1986). This suggests that the dump is highly perforate toseawater. Contaminants levels in seawater around the dump wereclearly higher than ambient levels but, nevertheless, quite low(with the possible exception of zinc). Collectively, these observa-tions suggest a comparatively high flux of low level contaminantsemanating from the dump. Metals tend to absorb to sediments andmost of the organic contaminant classes have high octanol–water

(Kow) partition coefficients signifying a tendency also to adsorb tosediments. Noticeably, groundwater samples with high TSS levels(see Chapman, 2008) had higher contaminant levels (i.e., PCDD/Fs and most metal ions), also suggesting sorption (i.e., adsorption,absorption, surface complexation, surface precipitation and ion ex-change) of contaminants onto particulate material within thegroundwater. Collectively, these observations suggest contami-nants may be exiting the dump associated with particulate mate-rial and accumulating in neighbouring sediments. Once outsideof the dump, sediment resuspension by wind and swells will serveto re-distribute sediments around Castle Harbour.


Many of the sediment samples from Castle Harbour exceededSQGs for a number of metals, PCBs, PAHs, and PCDDs/Fs suggestingbiological effects are possible. In the present study more SQGswere exceeded for metals than for PAHs, consistent with the pre-dominantly metallic waste associated with the dump. The numberof SQGs exceeded decreased systematically with increasing dis-tance from the dump, also consistent with the contamination gra-dient described previously. Whilst the majority of the guidelinesthat were exceeded were the low-range thresholds, for mercuryand zinc many samples also exceeded the mid-range (ERM andPEL) values suggesting greater probability of toxic effects.

SQGs were not intended to be used on a pass-fail basis, nor werethey intended as regulatory criteria or standards. They cannot alsobe used to state unequivocally whether biological effects will occurabove a threshold, or equally will not occur below a threshold;rather, they are intended as guidelines for use in interpretingchemical data and for concentrations that, if exceeded, shouldprompt further action usually with a decision-tree approach (seeChapman and Anderson, 2005). Further actions could include tox-icity testing of sediments and/or benthic community analyses and/or bioaccumulation tests, which collectively culminate in a furthercourse of action. The overall conclusions from the SQG analyses isthat contaminant levels in sediments surrounding the dump areclearly at levels at which biological effects could occur, althoughfurther testing would need to confirm this.

Leachate from municipal landfills contain complex mixtures ofinorganic macrocomponents (cations and anions), heavy metalsand xenobiotic compounds (XOCs) (Christensen et al., 2001; Slacket al., 2005). Although Bermuda’s seafill only receives householdwaste after it has been incinerated (and cement stabilized), it hasreceived commercial and mixed industrial waste (including wasteelectrical and electronic equipment such as computers and televi-sions) for the past 35 years. The leachate is likely to be a very com-plex mixture of many xenobiotics, and the chemical analysed inthe present study are only likely to be a small portion of thoseemanating from the dump.

Based on field and laboratory-scale data, conventional munici-pal landfills subject to percolating freshwater are known to under-go a complex series of decomposition phases which affects theleachate volume and composition and potential for aquifer con-tamination (reviewed by Kjeldsen et al., 2002). This knowledgehas been used to manage and understand the long-term behaviourof landfills, but it is not clear how transferable any of this informa-tion is to Bermuda’s seafill. Currently the seafill does not have engi-neered liners or containment systems, or leachate collection andprocessing systems. There are also no on-going monitoring orassessment programs, nor attempts to understand the complexbiogeochemistry occurring in the dump. Consequently, the long-term sustainability of the disposal plan is unclear.

In summary, of the chemicals targeted, the results of thisstudy present a very clear and coherent picture of environmentalcontamination by the seafill. Analyses of groundwater indicate


contamination by different classes of chemicals consistent with thenature of the material being disposed of at the dump (i.e., primarilymetal waste). Contaminants found to be elevated in the ground-water were also found to be elevated in seawater leaching out ofthe dump, and subsequently found to be elevated in neighbouringsediments. Four of the contaminants in the sediments (dioxins,furans, PCBs and the organochlorine chlordane) are on the ‘dirty-dozen’ list of persistent organic pollutants (POPs) that have beentargeted by international convention as priority pollutants for thereduction and elimination. Contaminant levels in water leachingout of the dump regularly exceed water quality guidelines in Floridaand Australia where there are equivalent marine habitats (i.e., coralreefs environments) and contaminant levels in sediments exceedSQGs in Florida and Australia where biological effects are possibleand, for zinc and mercury, where biological effects are probable.

Acknowledgements

Funding was provided by Bermuda Government Ministry of theEnvironment, Khaled Bin Sultan Living Oceans Foundation and Ber-muda Institute of Ocean Sciences (BIOS). Dieb Birkholz (AnalyticalChemistry and Testing Services, ALS laboratory Group) is acknowl-edged for providing advice on chemical analyses. The views andopinions expressed in this paper are those of the author.

References

Abramowicz, D.A., 1996. Aerobic and anaerobic biodegradation of PCBs – a review.CRC Crit. Rev. Biotechnol. 10, 241–249.

Anon, 2005. Government of Bermuda, Ministry of the Environment 2005 State ofthe environment report. Produced by the Department of Communication andInformation. Island Press, Bermuda.

ANZECC and ARMCANZ, 2000. Australian and New Zealand Guidelines for Fresh andMarine Water Quality. Australian and New Zealand Environment ConservationCouncil and Agriculture and Resource Management Council of Australia andNew Zealand, Canberra, ACT.

Aoki, Y., 2001. Polychlorinated biphenyls, polychlorinated dibenzo-p-dioxins, andpolychlorinated dibenzofurans as endocrine disrupters – what we have learnedfrom Yusho disease. Environ. Res. 86, 2–11.

Bedard, D.L., 2003. Polychlorinated biphenyls in aquatic sediments: environmentalfate and outlook for biological treatment. In: Haggblom, M.M., Bossert, I.D.(Eds.), Dehalogenation: Microbial Processes and Environmental Applications.Kluwer Academic Publishers, Boston, MA, pp. 443–465.

Breivik, K., Sweetman, A., Pacyna, J.M., Jones, K.C., 2002. Towards a global historicalemission inventory for selected PCB congeners—a mass balance approach 1.Global production and consumption. Sci. Total Environ. 290, 181–198.

Brown, J.F., Bedard, D.L., Brennan, M.J., Carnahan, J.C., Feng, H., Wagner, R.E., 1987.Polychlorinated biphenyl dechlorination in aquatic sediments. Science 236,709–712.

Brown, M.P., Bush, B., Rhee, G.Y., Shane, L., 1988. PCB dechlorination in HudsonRiver sediment. Science 240, 1674–1675.

Burns, K.A., Ehrhardt, M.G., MacPherson, J.M., Tierney, J.A., Kananen, G., Connelly, D.,1990. Organic and trace metal contaminants in sediments, seawater andorganisms from two Bermuda harbours. J. Exp. Mar. Biol. Ecol. 138, 9–34.

Burton, G.A., 2002. Sediment quality criteria in use around the world. Limnology 3,65–76.

CCME, 2002. Canadian sediment quality guidelines for the protection of aquatic life:summary tables. In: Canadian Environmental Quality Guidelines, 1999,Canadian Council of Ministers of the Environment, Winnipeg.

Chapman, D., 2008. A Soil and Groundwater Characterisation Study of the BermudaAirport Landfill. Master Thesis, Royal Holloway University of London, UK,135pp.

Chapman, P.M., 2007. Determining when contamination is pollution – weight ofevidence determinations for sediments and effluents. Environ. Int. 33, 492–501.

Chapman, P.M., Anderson, J., 2005. A decision-making framework for sedimentcontamination. Integr. Environ. Assess. Manage. 1, 163–173.

Christensen, T.H., Kjeldsen, P., Bjerg, P.L., Jensen, D.L., Christensen, J.B., Baun, A.,Albrechtsen, H.-.J., Heron, G., 2001. Biogeochemistry of landfill leachate plumes.Appl. Geochem. 16, 659–718.

Clarke, J.U., 1998. Evaluation of censored data methods to allow statisticalcomparisons among very small samples with below detection limitobservations. Environ. Sci. Technol. 32, 177–183.

Clarke, K.R., Gorley, R.N., 2006. PRIMER v6. User Manual/Tutorial. PRIMER-E,Plymouth, UK. 192pp.

Cleverly, D., Schaum, J., Schweer, G., Becker, J., Winters, D., 1997. The congener profilesof anthropogenic sources of chlorinated dibenzo-p-dioxins and chlorinateddibenzofurans in the United States. Organohalogen Comp. 32, 430–435.

Crane, J.L., MacDonald, D.D., 2003. Applications of numerical sediment qualitytargets for assessing sediment quality conditions in a US Great Lakes area ofconcern. Environ. Manage. 32, 128–140.

Dearth, M.A., Hites, R.A., 1991. Complete analysis of technical chlordane usingnegative ionization mass spectrometry. Environ. Sci. Technol. 25, 245–254.

Dodge, R.E., Vaisnys, J.R., 1977. Coral population and growth patterns: response tosedimentation and turbidity associated with dredging. J. Mar. Res. 35, 715–730.

Domingo, J.L., Schumacher, M., Granero, S., De Kok, H.A.M., 2001. Temporal variationof PCDD/PCDF levels in environmental samples collected near an old municipalwaste incinerator. Environ. Monit. Assess. 69, 175–193.

Evans, B.S., Dudley, C.A., Klasson, K.T., 1996. Sequential anaerobic–aerobicbiodegradation of PCBs in soil slurry microcosms. Appl. Biochem. Biotechnol.57–58, 885–894.

Fairey, R., Long, E.R., Roberts, C.A., Anderson, B.S., Phillips, B.M., Hunt, J.W., Puckett,H.R., Wilson, C.J., 2001. An evaluation of methods for calculating mean sedimentquality guideline quotients as indicators of contamination and acute toxicity toamphipods by chemical mixtures. Environ. Toxicol. Chem. 20, 2276–2286.

Faroon, O., Jones, D., de Rosa, C., 2001. Effects of polychlorinated biphenyls on thenervous system. Toxicol. Ind. Health 16, 305–333.

Fernandez, A., Singh, A., Jaffe, R., 2007. A literature review on trace metals andorganic compounds of anthropogenic origin in the Wider Caribbean Region.Mar. Pollut. Bull. 54, 1681–1691.

Field, J.A., Sierra-Alvarez, R., 2008. Microbial transformation and degradation ofpolychlorinated biphenyls. Environ. Pollut. 155, 1–12.

Flood, V.S., Pitt, J.M., Smith, S.R., 2005. Historical and ecological analysis of coralcommunities in Castle Harbour (Bermuda) after more than a century ofenvironmental perturbation. Mar. Pollut. Bull. 51, 545–557.

Gschwend, P.M., Hites, R.A., 1981. Fluxes of polycyclic aromatic hydrocarbons tomarine and lacustrine sediments in the northeastern United States. Geochim.Cosmochim. Acta 45, 2359–2367.

Harner, T., Pozo, K., Gouin, T., Macdonald, A.M., Hung, H., Cainey, J., Peters, A., 2006.Global pilot study for persistent organic pollutants (POPs) using PUF diskpassive air samplers. Environ. Pollut. 144, 445–452.

Harner, T., Shoeib, M., Diamond, M., Stern, G., Rosenberg, B., 2004. Using passive airsamplers to assess urban-rural trends for persistent organic pollutants. 1.Polychlorinated biphenyls and organochlorine pesticides. Environ. Sci. Technol.38, 4474–4483.

Hjelmar, O., 1996. Disposal strategies for municipal solid waste incinerationresidues. J. Hazard. Mater. 47, 345–368.

IARC, 1983. International agency for research on cancer. Polynuclear aromaticcompounds, Part 1, chemical, environmental and experimental data. IARCMonogr. Eval. Carcinog. Risk. Chem. Hum. 32, 1–453.

ICES, 1986. Report of the ICES Advisory Committee on Marine Pollution. Co-operative Research Report of the International Council for Exploration of theSea, vol. 167, pp. 26–29.

Jickells, T.D., Knap, A.H., 1984. The distribution and geochemistry of some tracemetals in the Bermuda coastal environment. Estuar. Coast. Shelf Sci. 18, 245–262.

Johnson, G.W., Hansen, L.G., Hamilton, M.C., Fowler, B., Hermanson, M.H., 2008. PCB,PCDD and PCDF congener profiles in two types of Aroclor 1254. Environ.Toxicol. Pharmacol. 25, 156–163.

Kaiser, J., Enserink, M., 2000. Environmental toxicology: treaty takes a POP at thedirty dozen. Science 290, 2053.

Kao, C.M., Chen, S.C., Liu, J.K., Wu, M.J., 2001. Evaluation of TCDD biodegradabilityunder different redox conditions. Chemosphere 44, 1447–1454.

Keith, L.H., Telliard, W.A., 1979. Priority pollutants I – a perspective view. Environ.Sci. Technol. 13, 416–423.

Kjeldsen, P.I., Barlaz, M.A., Rooker, A.P., Baun, A., Ledin, A., Christensen, T.H., 2002.Present and Long-Term Composition of MSW Landfill Leachate: a review. Crit.Rev. Env. Sci. Tec. 32, 297–336.

Knap, A., Binkley, K.S., Artz, R.S., 1991a. The occurrence and distribution of traceorganic compounds in Bermuda precipitation. Atmos. Environ. 22, 1411–1423.

Knap, A.H., Cook, C.B., Cook, S.B., Simmons, J.A.K., Jones, R.J., Murray, A.E., 1991.Marine Environmental Studies to Determine the Impact of the Mass BurnIncinerator Proposed for Tynes Bay, Bermuda. Prepared for Ministry of Worksand Engineering, Hamilton, Bermuda.

Laflamme, R.E., Hites, R.A., 1978. The global distribution of polycyclic aromatichydrocarbons in recent sediments. Geochim. Cosmochim. Acta 42, 289–303.

Lima, A.L., Farrington, J.W., Reddy, C.M., 2005. Combustion-derived polycyclicaromatic hydrocarbons in the environment – a review. Environ. Forensics 6,109–131.

Long, E.R., Field, L.J., MacDonald, D.D., 1998. Predicting toxicity in marine sedimentswith numerical sediment quality guidelines. Environ. Toxicol. Chem. 17, 714–727.

Long, E.R., Ingersoll, C.G., MacDonald, D.D., 2006. Calculation and uses of meansediment quality guideline quotients: a critical review. Environ. Sci. Technol. 40,1726–1736.

Long, E.R., MacDonald, D.D., Smith, S.L., Calder, F.D., 1995. Incidence of adversebiological effects within ranges of chemical concentrations in marine andestuarine sediments. Environ. Manage. 19, 81–97.

Loring, D.H., 1991. Normalization of heavy-metal data from estuarine and coastalsediments. ICES J. Mar. Sci. 48, 101–115.

MacDonald, D.D., Carr, R.S., Calder, F.D., Long, E.R., Ingersoll, C.G., 1996.Development and evaluation of sediment quality guidelines for Floridacoastal waters. Ecotoxicology 5, 253–278.


Mackay, D., Shiu, W.Y., Ma, K.C., 1992. Polynuclear aromatic hydrocarbons,polychlorinated dioxins, and dibenzofurans. In: Illustrated Handbook ofPhysico-Chemical Properties and Environmental Fate for Organic Chemicals,vol. II. Lewis Publishers, Boca Raton.

Mayes, B.A., McConnell, E.E., Neal, B.H., Brunner, M.J., Hamilton, S.B., Sullivan, T.M.,Peters, A.C., Ryan, M.J., Toft, J.D., Singer, A.W., Brown, J.F., Menton, R.G., Moore,J.A., 1998. Comparative carcinogenicity in Sprague-Dawley rats of thepolychlorinated biphenyl mixtures aroclors 1016, 1242, 1254, and 1260.Toxicol. Sci. 41, 62–76.

McCready, S., Birch, G.F., Long, E.R., Spyrakis, G., Greely, C.R., 2006. Relationshipsbetween toxicity and concentrations of chemical contaminants in sedimentsfrom Sydney Harbour, Australia, and vicinity. Environ. Monit. Assess. 120, 187–220.

Natarajan, M.R., Wu, W.M., Nye, J., Wang, H., Bhatnagar, L., Jain, M.K., 1996.Dechlorination of polychlorinated biphenyl congeners by an anaerobicmicrobial consortium. Appl. Microbiol. Biotechnol. 46, 673–677.

Nelson, D.W., Sommers, L.E., 1996. Total carbon, organic carbon and organic matter.In: Bartels, J.M., Bigham, J.M., Sparks, D.L., Page, A.L., Helmke, P.A., Loeppert,R.H., Soltanpour, P.N., Tabatabai, M.A., Johnston, C.T., Sumner, M.E. (Eds.),Methods of Soil Analysis. Part 3. Chemical Methods. SSSA; ASA, Madison, pp.961–1010.

Olie, K., Vermeulen, E.L., Hutzinger, O., 1977. Chlorodibenzo-p-dioxins andchlorodibenzofurans are trace components of fly ash and flue gas of somemunicipal incinerations in the Netherlands. Chemosphere 6, 455–459.

Peters, A., 2005. Environment Bermuda: Soil and Water Tank Survey 2005. BermudaBiological Station for Research, Inc. Special Report to the Department ofEnvironmental Protection, Bermuda Government, 135pp.

Readman, J.W., Fillmann, G., Tolosa, I., Bartocci, J., Villeneuve, J.-P., Catinni, C., Mee,L.D., 2002. Petroleum and PAH contamination of the Black Sea. Mar. Pollut. Bull.44, 48–62.

Roethel, F.J., 2006. Tynes Bay Waste Treatment Facility – Environmental and PotentialWorker Exposure to Municipal Solid Waste Combustor Ash. Final Report to theGovernment of Bermuda Ministry of Works and Engineering, 91pp.

Sabbas, T., Polletini, A., Pomi, R., Astrup, T., Hjelmar, O., Mostbauer, P., Cappai, G.,Magel, G., Salhofer, S., Speiser, C., Heuss-Assbichler, S., Klein, R., Lechner, P.,2003. Management of municipal solid waste incineration residues. WasteManage. 23, 61–88.

Shepard, F.P., 1954. Nomenclature based on sand–silt–clay ratios. J. Sediment.Petrol. 24, 151–158.

Sicre, M.A., Marty, J.C., Saliot, A., Aparicio, X., Grimalt, J., Albaiges, J., 1987. Aliphaticand aromatic hydrocarbons in different sized aerosols over the MediterraneanSea: occurrence and origin. Atmos. Environ. 21, 2247–2259.

Slack, R.J., Gronow, J.R., Voulvoulis, N., 2005. Household hazardous waste inmunicipal landfills: contaminants in leachate. Sci. Total Environ. 337, 119–137.

Smith, S.R., Hellin, D.C., 1998. Environment Bermuda: Marine Environmental ImpactStudies on the Thermal Effluent Released from the Tynes Bay Incinerator andAsh Block Disposal in Castle Harbour. Bermuda Biological Station for ResearchInc., Special Publication, No. 35, 210pp.

Thomson, J.A.M., Foster, S.S.D., 1986. Effect of urbanization on the groundwater oflimestone islands: an analysis of the Bermuda case. J. Inst. Water Eng. Scientists40, 527–540.

Van den Berg, M., Birnbaum, L.S., Denison, M., De Vito, M., Farland, W., Feeley, M.,Fiedler, H., Hakansson, H., Hanberg, A., Haws, L., Rose, M., Safe, S., Schrenk, D.,Tohyama, C., Tritscher, A., Tuomisto, J., Tysklind, M., Walker, N., Peterson, R.E.,2006. The 2005 World Health Organization reevaluation of human andmammalian toxic equivalency factors for dioxins and dioxin-like compounds.Toxicol. Sci. 93, 223–241.

Wentworth, C.K., 1922. A scale of grade and class terms for clastic sediments. J. Geol.30, 377–392.

Zar, J.H., 1999. Biostatistical Analysis, fourth ed. Prentice-Hall Inc., New Jersey.

environmental contamination associated with a marine landfill (‘seafill’) beside a coral reef

Documents