characterization and source of oil contamination on the beaches and seabird corpses, sable island,...

www.elsevier.com/locate/marpolbul

Marine Pollution Bulletin 52 (2006) 778–789

Characterization and source of oil contamination on the beachesand seabird corpses, Sable Island, Nova Scotia, 1996–2005

Zoe Lucas a,*, Clive MacGregor b

a P.O. Box 64, Halifax, Nova Scotia, Canada B3J 2L4b 3852 Rockhead Court, Halifax, Nova Scotia, Canada B3K 6B5

Abstract

During April 1996–May 2005, 2343 oiled seabird corpses were recorded in beach surveys conducted on Sable Island, Nova Scotia.One hundred eighty-three samples of oil were collected from the beaches and from the feathers of bird corpses. Gas chromatographic(GC/FID) analysis was used to identify generic oil type and likely marine source. During this period, at least 74 marine oil dischargeevents were probably responsible for beached pelagic tar and contamination of seabird corpses found on Sable Island, of which77.0% were crude oils, 14.9% were fuel oils, and 8.1% were bilge oil mixtures. While fuel and bilge oils may be discharged by all vesseland platform types, crude oil discharges are associated with tanker operations. This study demonstrates that oiling of the sea from tank-ers remains a serious wildlife issue in the Northwest Atlantic.� 2005 Elsevier Ltd. All rights reserved.

Keywords: Oil pollution; Oil identification; Shipping; Tanker operations; Monitoring; Beached bird surveys; Northwest Atlantic; Sable Island

1. Introduction

Oil enters the marine environment from natural seeps,land runoff, vessels, pipelines and offshore petroleumexploration and production platforms. Common sourcesof oil pollution in the offshore North Atlantic region aremarine shipping activities, offshore oil and gas operations,and long-range industrial transport (National ResearchCouncil, 2003). Marine shipping activities discharge oilby several processes: catastrophic losses during accidents,periodic discharges from poor or illegal tanker operations,and chronic bilge discharges from general cargo vessels orother vessels.

Discharges, such those associated with the Erika andPrestige disasters, are rare, but when they do occur, theygenerally result in the release of crude oil. However, inthe offshore waters of Atlantic Canada, catastrophic spillswhich occurred during accidents involving the tanker

0025-326X/$ - see front matter � 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.marpolbul.2005.11.023

* Corresponding author.E-mail address: [email protected] (Z. Lucas).

Arrow (1970), the oil-barge Irving Whale (1970), and thetanker Kurdistan (1979) (Brown, 1991), resulted in dis-charges of fuel oil. During the 1996–2005 study period,only one catastrophic discharge occurred in the region.The Flare, which broke up roughly 200 km east of CapeBreton, Nova Scotia, in 1998, was a bulk grain carrier withan estimated 650 metric tonnes of fuel and lube oil onboard (R. Percy, Environment Canada, personal communi-cation). In terms of tonnes of oil, total spillage from suchcatastrophic incidents is generally less than that releasedas illegal discharges associated with tank washings, dirtyballast and bilge pumping (National Research Council,2003; GESAMP, 2004).

Discharged crude oils originating from tankers can becontaminated with distillate fuel oils such as diesel that isused as a cleaning solvent for removing residues of crudeoil from vessel cargo tanks. Such tank washings are heldin slop tanks, but may ultimately be discharged to theocean to avoid the cost of disposing ashore at an approvedoil reception facility. Vessels of all types may also dischargebilge waters containing mixtures of lube and fuel oils (Bun-ker or distillates). Some container ships operate on heavier

mailto:[email protected]

Z. Lucas, C. MacGregor / Marine Pollution Bulletin 52 (2006) 778–789 779

fuel oils (e.g. Bunker oils) on long runs but switch to lighterfuel oils when operating near port regions. Heavy fuel oilsleave residues in the tanks that must be periodicallyremoved, sometimes by washing with a light fuel oil. Ifdischarges of this residue occur the resulting oil may becharacterized as Bunker fuels containing lighter fuel oilssuch as diesel.

Discharges from offshore oil and gas operations (explo-ration and production platforms) on the Scotian Shelfcould include light crude oils such as condensate, and spe-cialty products that may contain oil (e.g. drilling fluids).

Marine discharges of oil, exposed at sea for many weeksto months, would generally be observed in Scotian Shelfwaters as ‘‘pelagic tar’’. Highly weathered pelagic tars, or‘‘tarballs’’, are floating balls of tar 5–150 mm diameter,having a weathered hard exterior, and less weathered semiflowable interior (Van Vleet et al., 1984; Savage and Ward,1984; Butler et al., 1998). The weathered outer coating oftarballs at sea makes them less likely to adhere to birdfeathers, but they reflect the occurrence of oiling eventson the sea surface.

Sable Island (44N, 60W), the emerged portion of theSable Island Bank on the outer edge of the Scotian Shelf,is situated in a region with a high level of marine trafficheading to and from ports in Atlantic Canada, the St. Law-rence Seaway, the United States, and Europe (Fig. 1). Theisland is surrounded by major shipping routes, withconcentrations of activity southwest through northwest ofthe island (Lock et al., 1994; Wiese and Ryan, 2003). Ves-sels on the shipping lanes are thought to discharge signifi-cant amounts of oil into the Northwest Atlantic(Chardine, 1991; Wiese and Ryan, 2003; NationalResearch Council, 2003). Also, the Sable Island Bank isat the centre of development and production of offshorepetroleum resources (now primarily gas) on the ScotianShelf, with production platforms located south, southwestand east of the island.

Fig. 1. Typical annual ship track pattern of container, cargo and tankervessels travelling to or through Atlantic Canada destined to Canadianports during 1990–2000 (adapted from Wiese and Ryan, 2003).

Sable Island’s unique offshore location and proximity toareas of heavy marine traffic and energy production activ-ities make it a useful platform for monitoring marine issues(e.g. Lucas, 1989; Lucas and Hooker, 2000; Lucas andDaoust, 2002). Since 1993, surveys for beached oiled birdshave been conducted on the island (Lucas, 2003). Thispaper presents the results of gas chromatographic analysisused to generically identify oil types in samples of pelagictar and oiled feathers recovered from the beach and fromoiled seabird corpses during beach surveys conducted dur-ing 1996–2005. Improved scientific knowledge of sources ofoil discharges to the Northwest Atlantic could help guideeffective pollution policies and monitoring, and enable reg-ulators to target surveillance on vessel types most com-monly suspected of discharging oil at sea.

2. Methods

2.1. Field methods

Sable Island is a 45 km-long sand bar with a maximumwidth of roughly 1.5 km. The shoreline comprises longuninterrupted sand beaches on both the north and southsides. Surveys for beached oil and oiled seabirds were usu-ally conducted once every 30–40 days, and observationswere made during routine travel on the beach between sur-veys. Species identification, corpse condition and extent ofoiling were recorded for seabird specimens. When possible,the time since death was estimated based on freshness oftissues and degree of scavenging. The oiling rate is the frac-tion of oiled birds of the total number of dead birds, intactand otherwise, found on the beach.

Samples of beached pelagic tar and oiled feathers, repre-senting contamination on various seabird species, and atvarious beach locations on the island, were collected. Asample set comprised one or more samples collected duringa beached bird survey, or during an interval between sur-veys. An attempt was made to sample from the freshestspecimens (i.e. the most recently oiled) based on conditionof the corpse and the development of crust formed by sandadhering to the oil. The samples were packaged in alumin-ium foil or in glass jars, labelled, kept frozen for periodsranging from one week to three months, and delivered tothe laboratory for gas chromatographic analysis (SeatechLtd, now Maxxam Analytics, Bedford, Nova Scotia). Sam-ples of oil found on the beach and bird corpses on SableIsland were identified as representing separate dischargeevents on the basis of date, bird corpse condition, and oilcharacteristics.

2.2. Laboratory methods

2.2.1. Analysis of oil specimens collected on Sable Island

All oil specimens were solid samples (pelagic tar or oiledseabird feathers) and were dissolved directly in dichloro-methane. This extract, filtered to remove solids, wasinjected on a glass capillary column (SPB-1) in an HP

780 Z. Lucas, C. MacGregor / Marine Pollution Bulletin 52 (2006) 778–789

5890 Gas Chromatograph with Flame Ionization Detector(GC/FID). Outputs from the GC were retrieved onHPChem station, and chromatograms were produced andassessed manually. Concurrently standard oils such asBunker C, Scotian shelf crude and condensate, marine die-sel, and hydrocarbon n-alkane marker standards (C6–C16and C18–C32) were run under the same conditions. Thispermitted identification of the n-alkane peaks in the chro-matograms. N-alkane peaks were not quantified: only rela-tive concentration calculations based on peak height wereundertaken.

2.2.2. Laboratory evaporative weathering of oils and oil

mixturesTo assess the effect of evaporative weathering, 14 oils

and mixtures of oils representative of marine dischargesto the North Atlantic were weathered in a lab (�20 �C)in weighing dishes (�20 ml/dish) and chromatograms pro-duced at 0, 1, 7 and 28 days of evaporative weathering. Arepresentative evaporative weathering chromatogram ser-ies for crude oil is provided in Fig. 2.

2.2.3. Chromatogram interpretation

Standard oil identification uses a tiered analyticalapproach with basic gas chromatographic procedures usedto generically identify the oils before more specific proce-dures are used to compare spills with suspect sources(Wang et al., 1997; Wang and Fingas, 2003). In this study,robust chromatographic indicators were used to identify oilspecimens collected from the beach and oiled seabirds bygeneric oil type:

• Distillate fuels.• Bunker C type or heavy fuels.• Crude oils.• Condensate from production on the Scotian Shelf.• Bilge oil mixtures of fuels (distillate or heavy) and lubri-

cating oils.

Five basic chromatographic indicators were comparedagainst oil evaporative state:

• Hydrocarbon range (HCR) as n-alkanes for all n-alkanechromatographic peaks greater than 10% of the highestn-alkane chromatographic peak.

• Maximum n-alkane hydrocarbon chromatographic peak(MHCP).

• Unresolved residual material (URM) maximum for thechromatographic envelope under the n-alkane peakswas identified by its n-alkane number.

• Ratio of URM height (i.e. height to baseline correctedfor drift) to MHCP height expressed as a percent (tonearest 5%).

• Presence or absence of a bimodal distribution of hydro-carbon peaks or unresolved materials indicating pres-ence of more than one oil type, such as might occur inbilges.

The HCR, MHCP, URM and URM/MHCP ratio asgeneric oil indicators for the nine test oils are provided inTable 1. In summary, evaporatively weathered oils hadunique indicators:

• Weathered Diesel oils upper end HCR never more than22, MHCP 16–17, URM 17–18, ratio 40%.

• Weathered Condensate upper end HCR never morethan 26, MHCP 16, URM 17, ratio 20%.

• Weathered Bunker C upper end HCR never more than25, MHCP 17, URM 19 but ratio 40% (highest exceptlubricating oil).

• Weathered Crude oil upper HCR 26–29, MHCP 17,URM 24–25, ratio 25%.

• Lubricating oil no HCR or real MHCP, URM 24 andratio >300%.

The indicators (Table 1) for Bunker C (a heavy fuel oil)and crude oil demonstrate how the oils could be identifiedby this technique, as follows: weathered Bunker C had anHCR 14–25, an MHCP of 19, a URM of 19 and the ratiowas 40%; weathered Crude had an HCR of 14–27, anMHCP of 17, a URM of 24–25, and the ratio was 15–25%. The chromatograms were also interpreted for thepresence of two or more oil types (i.e. bimodal peakdistribution). This would indicate a mixture of oils andwas important in determining if bilge oil discharges werepresent.

Although Table 1 indicates that it would be possible todistinguish between different types of crude oil, the analy-ses used in this study do not permit more than generic iden-tification. For example, for Hibernia crude and heavyArabian crude the difference in Hydrocarbon Range(HCR) at the upper end is 27 and 29 at 28 days. If theend hydrocarbon in the range, or any parameter, is off byone unit (e.g. an error in determining end of range), 27could be 28, and 29 could be 28, thus the HCR for thetwo crude oils would be the same. Further, the URM toMHCP ratio is accurate within 5%, so 15% and 25% couldboth be 20%, and again no difference for Hibernia or heavyArab crude. More detailed analysis would provide thegreater accuracy required.

In this study different crude oils cannot be reliably dis-tinguished, but it is possible to separate fuel oils andcrudes. In Bunker C fuel oil and heavy Arab crude, theend of the HCR is similar at 25 and 27, but the URM is19 and 25, respectively. This is too large for an error of 1unit to affect the interpretation. The ratio URM to MHCPis 40% and 25% for Bunker and crude, respectively and thisis also too large for an error of 5% to affect interpretation.Thus crude oils and fuel oils can be easily distinguishedbecause two of four interpretative parameters are suffi-ciently different to allow such an interpretation, while forHibernia and heavy Arab crude the four parameters aretoo close together. Differences in key parameters are usedto separate generically different oil types, but cannot sepa-rate oils within types (e.g. heavy crude versus light crude).

Fig. 2. A representative evaporative weathering chromatogram series for crude oil.


The degree of weathering was assessed by comparison tothe evaporated test chromatograms and described as: slight(<25% loss), moderate (25–50% loss), severe (50–75% loss)and very severe (>75% loss).

3. Results

3.1. Oiled seabirds and collection of oil samples

In 93 beach surveys conducted on Sable Island duringApril 1996–May 2005, 6326 bird corpses were recovered:3971 alcids, 979 shearwaters, and 1376 other species com-bined (e.g. gulls, gannets, loons). Of these 2262 (57.0%)alcid corpses, 19 (1.9%) shearwater corpses, and 62

(4.5%) other species, were oiled. Most alcid corpses,83.4%, occurred in December through April, and shear-waters accounted for a lesser peak in summer, with most,89.4%, being found in June through August. Oiled seabirdcorpses were recorded in 90.5% (38 of 42) of surveys inNovember through April, and in 70.6% (36 of 51) of sur-veys conducted in May through October.

During April 1996–May 2005, 46 sets of oil samples,comprising 183 individual samples, were collected. Ofthese, 13 were samples of pelagic tar found on the beach,and 170 were samples of oil recovered from seabirdcorpses. Sets were comprised of 1–13 samples, dependingon amount of beached pelagic tar, and number and condi-tion of oiled bird corpses available. In 35 (76.1%) sets, two

Table 1Oil standards (static evaporation @20 �C over 28 days)

Oil type Daysevaporation

Evaporation% remaining

HCRa MHCPb URMc Ratio URM toMHCP (%)

Comments

Gasoline 0 0 6–12 8 NA NADiesel 0 0 8–19 13 13 25Marine Diesel 0 0 8–20 13 14 25 Naval Distillate G3P11MCondensate 0 0 6–18 7 NA NA Thebaud condensateBunker C 0 0 11–23 16 17 25Arab light crude 0 0 6–23 7 NA NA URM poorly defined 17–25Arab heavy crude 0 0 6–24 7 25 <10 URM poorly defined 17–25Hibernia crude 0 0 6–27 13 25 10 URM poorly defined 17–25Lubricating oil 0 0 NA 24 24 400 No peaks all URM

Gasoline 1 4 11–13 12 NA NA Very severely weatheredDiesel 1 68 10–20 13 14 25 Moderately weatheredMarine Diesel 1 83 10–21 13 15 30 Slightly weatheredCondensate 1 48 10–22 13 13 10 Severely weatheredBunker C 1 98 11–23 16 18 40Arab light crude 1 76 11–25 13 25 15 URM poorly definedArab heavy crude 1 82 10–25 13 25 15 URM poorly definedHibernia crude 1 78 9–28 13 25 10 URM poorly definedLubricating oil 1 99 NA 24 24 440 No peaks all URM

Gasoline 7 0 NA NA NA NA All evaporatedDiesel 7 46 12–20 13 15 25 Severely weatheredMarine Diesel 7 65 12–21 13 15 40 Moderately weatheredCondensate 7 39 12–23 14 15 10 Severely weatheredBunker C 7 95 14–24 17 19 40 Slightly weatheredArab light crude 7 71 13–27 15 25 15 URM poorly definedArab heavy crude 7 77 12–26 15 24 20 URM poorly definedHibernia crude 7 70 13–28 15 25 15 URM poorly definedLubricating oil 7 99 NA 24 24 400 No peaks all URM

Diesel 28 31 13–21 15 16 35 Severely weatheredMarine Diesel 28 47 14–22 17 18 40 Severely weatheredCondensate 28 30 14–26 16 17 20 Severely weatheredBunker C 28 91 14–25 17 19 40 Slightly weatheredArab light crude 28 66 14–26 17 24 20 Moderately weatheredArab heavy crude 28 73 14–27 17 25 25 Moderately weatheredHibernia crude 28 64 14–29 17 25 15 Moderately weatheredLubricating oil 28 99 NA 24 24 330 Slightly weathered

a Hydrocarbon range (HCR) defined as chromatographic peaks >10% by height of maximum hydrocarbon chromatographic peak.b Maximum hydrocarbon peak (MHCP) is highest chromatographic peak.c Unresolved material maximum (URM) is the envelope under the resolved chromatographic where it has highest response above baseline (including

baseline drift).


or more samples were collected. Oil samples were recoveredyear-round, with 65.2% (n = 30) of sets [and 76.5%(n = 140) of samples] collected during December throughApril.

Oil samples were recovered from the corpses of 13 spe-cies: Northern Fulmar Fulmarus glacialis (6), GreaterShearwater Puffinus gravis (14), Northern Gannet Sula

bassanus (1), Herring Gull Larus argentatus (2), IcelandGull Larus glaucoides (1), Great Black-backed Gull Larus

marinus (2), Black-legged Kittiwake Rissa tridactyla (2),Razorbill Alca torda (14), Common Murre Uria aalge

(19), Thick-billed Murre Uria lomvia (52), Dovekie Alle alle

(11), and Atlantic Puffin Fratercula arctica (10). Five oiledCommon Loons Gavia immer were sampled, and another31 oil samples were taken from bird corpses (mostly largeauks) that could not be identified to species.

Of the 170 oiled bird corpses sampled, 35 (20.6%) hadprobably died within a week of being found. Of these, nine

(one Common Loon, one Razorbill, one Common Murre,four Thick-billed Murres, one Black-legged Kittiwake, andone Greater Shearwater) had been observed alive on thebeach and had died within 24 h. The remaining 135 birdssampled likely died at sea or on the beach one to fourweeks prior to collection. Live oiled seabirds were observedin 10% of beach surveys, and usually in small numbers(<10 birds). In 91.9% (68 of 74) of beach surveys in whichoiled birds were recorded, no oil or pelagic tar was foundon the beach.

3.2. Occurrence of pelagic tar on the beach

No visible slicks were recorded during 1996–2005, how-ever, beached oil in the form of pelagic tar particles wasobserved on 11 occasions. Of these, five incidents com-prised no more than one or two isolated lumps of severelyweathered tar. Significant amounts of pelagic tar, ranging


from 10 to >100 particles/m, washed ashore in December1996, March 1997, February 1998, January 1999, Decem-ber 2001, and July 2003, and in all cases the tar particleswere soft and sticky, without a noticeably hard crusty outercoating.

In mid-December 1996 particles of pelagic tar, 0.5–2.0 cm in diameter, washed ashore along most of the northshoreline, with more on the western part of the beach. Atthis time, no tar was found on the south side of the island.In early March 1997 pelagic tar was found on the westernhalf of the south beach. The tar appeared to have strandedsome time earlier, possibly in February, subsequently bur-ied by windblown sand and then exposed again. No tar wasfound on the north beach. Relatively few oiled bird corpseswere found, although many of those that were oiledoccurred in the patches of beached tar and had probablycome ashore with the tar particles. No pelagic tar sampleswere collected during the 1996 and 1997 occurrences.

In February 1998, pelagic tar came ashore on both thenorth and south sides of the island. The tar was first notedon February 7 when particles, 1.0–8.0 cm in diameter,accumulated along the tide line on the north beach. Onthe following day the entire north and south sides of theisland were inspected, and no pelagic tar was found onthe south side, however tar had continued accumulatingon the north side. Although beach inspections found oiledbird corpses on both north and south sides of the island, nopelagic tar was found on the south side until February 12,and by February 20, tar accumulations on the south wereextensive. During this period oil contamination wasobserved on >25 flying gulls, and on most seals hauledout on the island, including grey Halichoerus grypus, har-bour Phoca vitulina, harp Pagophilus groenlandica, andhooded Cystophora cristata seals. The seals likely becameoiled when they came ashore. Also, oil was seen on theunderparts and legs of overwintering Ipswich sparrowsPasserculus sandwichensis princeps feeding in the driftlineon the beach. In the subsequent beached bird survey,75% of seabird corpses were oiled, and contaminated birdsincluded Northern Fulmars, gulls, Black-legged Kittiwakesand five alcid species.

In late January 1999, pelagic tar (0.5–2.0 cm) washedashore along the eastern half of north beach. No tar wasfound on the western half of the north beach or on thesouth beach. During the February 1998 and January1999 occurrences, a total of nine samples were collected,five of pelagic tar from the beach, and four of oiled feathersfrom seabird corpses. All these oils were identified as slightto moderately weathered crude oil.

In early December 2001, pelagic tar washed ashore onthe western quarter of north beach. None was found onthe south beach. Two pelagic tar samples were recoveredfrom the beach, and seven from oiled seabird corpses. Allwere identified as severely weathered fuel oil.

In late July 2003, pelagic tar (0.5–2.0 cm) washed ashoreon the western quarter of the south beach. Small amountsof tar were noted on other areas of south beach, but accu-

mulations were not heavy anywhere. No tar was found onthe north beach. One pelagic tar sample was collected fromthe beach and was identified as moderately weatheredcrude oil.

3.3. Identification of oil

In 179 of 183 samples collected, three generic categories(marine sources) were identified (Table 2):

1. Crude oils discharged from tankers (e.g. ballasted cargotanks, slop tanks etc.).

2. Heavy Fuel oils (e.g. Bunker C type) discharged by acci-dent during fuel transfer operations or during tankwashing.

3. Bilge water or slop tank discharges from any sort of ves-sel (e.g. lubricating or hydraulic oils combined with fueloils, distillate or Bunker etc.).

Fig. 3 shows a sample chromatogram of crude oil froman oiled bird sampled in 2001. Of the 179 resolvable sam-ples, 139 (77.7%) were identified as crude oil, 23 (12.8%)were fuel oil, and 17 (9.5%) were bilge oil. Platform pro-duction oils were not identified in any of the samples col-lected. Interpretation of GC/FID analyses indicated thatthe 179 samples of resolvable oil collected on Sable Islandduring 1996–2005 likely represented a minimum of 74 sep-arate discharge events (Tables 2 and 3), of which 77.0%were crude oil, 14.9% were fuel oil, and 8.1% were bilgemixtures.

In four of the 183 samples collected, no resolvable oilwas detected. Although the four samples, all from birdcorpses, had both the texture and colour of oil, the chro-matogram results were inconclusive suggesting that if thecontamination was oil, it was severely weathered.

4. Discussion

4.1. Occurrence of oiled seabirds

During 1996–2005, numbers and species composition ofoiled seabirds recorded on Sable Island reflected the sea-sonal distribution of species. Numbers peaked in winter,and alcids comprised the majority of corpses found, reflect-ing the predominance of Arctic-breeding birds, such asThick-billed Murres and Dovekies, overwintering in theregion (Brown, 1985; Lock et al., 1994). Although few wereoiled, most shearwater corpses were found in June throughAugust, and this is consistent with the pelagic distributionof large numbers of shearwaters on the Scotian Shelf dur-ing the summer months (Lock et al., 1994).

Although oiled seabird corpses were recorded in 70.6%of surveys conducted in May through October, and in90.5% of surveys in November through April, this doesnot accurately reflect seasonal occurrence of oiled birds.On Sable Island seabird corpses can be rapidly buried bywindblown sand, and exposed days to months later. Thus

Table 2Seventy-four discharge events identified in 179 oil samples collected on Sable Island during 1996–2005

Date Number of samples Matrix Comments ID codea

Apr 96 2 Alcids Crude, some diesel. Slightly weathered. May be tank washing 1Apr 96 1 Alcid Bunker fuel oil. Slightly weathered 2Feb 98 5 Fulmar, beach Crude. Slight and moderate weathering 1Jan 99 4 Alcids, beach Crude. Slight, moderate and severe weathering 1Feb 99 6 Alcids, loons, gull Crude, some diesel. Slight, moderate and severe weathering.

May be tank washing1

Mar 99 2 Alcids Crude, some diesel. Slight weathering 1Apr 99 3 Alcids Crude. Moderate and severe weathering 1Apr 99 1 Alcid Crude. Slight weathering 1Apr 99 1 Fulmar Crude. Slight weathering 1May 99 1 Alcid Crude. Severe weathering 1Jun 99 2 Shearwaters Crude. Moderate and severe weathering 1Jun 99 1 Beach Crude. Slight weathering 1Jun 99 1 Shearwater Lube oil 3Dec 99 5 Alcids Crude. Slight and moderate weathering 1Dec 99 1 Alcid Crude. Slight weathering 1Dec 99 1 Shearwater Bunker. Moderate weathering 2Dec 99 1 Alcid Bunker. Slight weathering 2Feb 00 1 Fulmar Crude. Severe weathering 1Feb 00 1 Alcid Bunker. Slight weathering 2Apr 00 3 Alcid, loon, gull Mixture of solvents, lubes, crude. Slight and moderate weathering 3Apr 00 5 Alcids, loon Crude. Slight, moderate and severe weathering 1May 00 1 Alcid Crude and diesel. May be tank washing 1Aug 00 2 Alcids Crude. Slight weathering 1Dec 00 1 Alcid Crude. Moderate weathering 1Dec 00 2 Alcids Diesel and crude. Slight weathering. May be tank washing 1Jan 01 1 Alcid Crude. Severe weathering. Poor ID 1Jan 01 1 Alcid Crude. Severe weathering 1Jan 01 1 Alcid Fuel oil 2Jan 01 1 Alcid Mixture of solvents and diesel. Moderate weathering 3Mar 01 2 Alcids Crude. Moderate weathering 1Mar 01 4 Alcids Bunker + lube. Slight weathering 3Mar 01 1 Alcid Crude. Moderate weathering 1Mar 01 2 Alcids Crude. Severe weathering. Poor ID 1Mar 01 1 Alcid Crude. Moderate weathering 1Apr 01 1 Bird Crude. Moderate weathering 1Apr 01 1 Alcid Crude. Severe weathering. Poor ID 1Apr 01 6 Alcids Crude. Slight and moderate weathering 1Jun 01 2 Shearwater, gull Crude. Severe weathering 1Jun 01 1 Alcid Crude. Severe weathering 1Dec 01 2 Shearwater, gannet Crude. Severe weathering 1Dec 01 9 Alcids, gull, beach Diesel. Severe weathering 2Feb 02 6 Alcids Bunker + little lube. Slight and moderate weathering.

Some high URM and high/shifted URM3

Feb 02 1 Alcid Crude. Moderate weathering. High URM 1Mar 02 8 Alcids, fulmar, beach Crude + little lube. Slight, moderate and severe weathering 1Apr 02 2 Alcid, gull Crude. Moderate and severe weathering 1Jun 02 1 Shearwater Lube oil 3Jul 02 1 Shearwater Bunker. Moderate weathering. Possibly Bunker C residuals 2Aug 02 1 Beach Crude. Slight weathering 1Sep 02 1 Shearwater Bunker. Slight weathering 2Oct 02 3 Shearwaters Bunker. Slight to moderate weathering 2Dec 02 5 Alcids Crude. Moderate weathering 1Jan 03 1 Alcid Crude. Severe weathering 1Mar 03 1 Alcid Crude. Severe weathering 1Mar 03 1 Alcid Crude. Moderate weathering 1Mar 03 1 Alcid Bunker. Slight weathering 2Apr 03 2 Alcid, fulmar Crude. Moderate weathering 1May 03 5 Alcids Crude and heavy crude. Moderate and severe weathering 1Jul 03 1 Beach Heavy crude. Moderate weathering. Pelagic tar 1Dec 03 4 Alcids, gull Bunker. Slight weathering. Some with very high URM 2Feb 04 1 Alcid Crude. Severe weathering 1Mar 04 5 Alcids, gull Crude. Moderate and severe weathering 1Mar 04 1 Alcid Crude. Moderate weathering 1


Table 2 (continued)

Date Number of samples Matrix Comments ID codea

Mar 04 6 Alcids Crude. Moderate weathering 1May 04 12 Alcids, fulmar Crude. Slight and moderate weathering 1May 04 1 Alcid Crude. Slight weathering 1Jan 05 2 Alcids Crude. Moderate weathering 1Mar 05 3 Alcids Crude. Moderate weathering 1Mar 05 1 Alcid Crude. Severe weathering 1Mar 05 1 Alcid Crude. Severe weathering 1Mar 05 3 Alcids Crude. Moderate weathering 1Apr 05 3 Alcids Crude. Moderate weathering 1Apr 05 2 Alcid, beach Crude. Moderate weathering 1May 05 2 Alcids Crude. Moderate weathering 1May 05 1 Beach Crude. Moderate weathering 1

a Code 1 = crude oil; code 2 = fuel oil; code 3 = bilge oil.

Fig. 3. A sample chromatogram of crude oil recovered from an oiled Thick-billed Murre corpse found on the Sable Island in April 2001.


some oiled birds recorded during the surveys were up toseveral months old, having been missed during previoussurveys. Such specimens account for occurrences of oiledbird corpses in many of the summer and autumn surveys.When these specimens are included in the survey totalsfor the period in which they likely washed ashore, oiledbirds occur in only 35% of surveys conducted in May toOctober.

Table 3Annual occurrence of discharge events identified by generic oil source ofsamples collected on Sable Island, Nova Scotia during 1996–2005

Year Crude oil Fuel oil Bilge Total

1996 1 1 21998 1 11999 11 2 1 142000 6 1 1 82001 12 2 2 162002 5 3 2 102003 6 2 82004 6 62005 9 9

Total 57 11 6 74

No samples were collected during 1997.

Fewer alcids, a seabird group highly vulnerable to oil pol-lution (Chardine, 1995), are in the offshore waters duringsummer, and this would account, in part, for the much lowernumber of oiled birds found on Sable Island in summer.However, large numbers of shearwaters feed in ScotianShelf waters during June through September (Lock et al.,1994), and flocks of as many as 5000 birds have been sightedfeeding or sitting on the water near Sable Island (McLaren,1981). Although shearwaters spend less time on the sea sur-face than alcids (Chardine, 1995), and thus are less vulnera-ble to oiling, the low oiling rate observed for these speciessuggests that they do not encounter significant amounts ofoil in waters around Sable Island during the summer.

Seasonal fluctuations in numbers and species composi-tion of beached oiled seabird corpses also reflected prevail-ing winds. Bird corpses afloat in the open ocean arestrongly influenced by wind speed and direction, andreported average rates of bird corpse drift range from2.2% to 4.6% of the wind speed (Hope Jones et al., 1970;Bibby and Lloyd, 1977; Bibby, 1981). In winter, prevailingwinds over the Scotian Shelf are from the northwestthrough northeast, and in summer winds are primarilyfrom the southwest through southeast (Meteorological Ser-vice of Canada). If wind is the primary force transporting


floating bird corpses, it is possible that many oiled birdcorpses found on the beach in winter are those that haveencountered oil northwest through northeast of SableIsland, and those found in the summer are mostly birdscontaminated southwest through southeast of the island.However, ocean currents may also affect corpse drift. Sev-eral studies have indicated the presence of gyres or gyre-like tendencies over the major outer banks, including theSable Island Bank (Hannah et al., 1996, 2001), and this isfurther complicated by the strong seasonal variationobserved in the gyre over the Sable Island Bank (Hannahet al., 2001). Gyres are thought to entrain floating materialsuch as plastic litter, oil, seaweed, bird and seal corpses,and other debris, some of which is eventually washedashore on the island (Lucas, 1989; Lucas and Hooker,2000). Also, a small proportion of oiled birds reach SableIsland while they are still alive, having swum or flown, insome cases against prevailing winds and currents. Thismay be a more significant factor for more aerial speciessuch as shearwaters.

The small proportion of live oiled seabirds observed onSable Island during 1996–2005 contrasts with findings insome Newfoundland surveys where a much larger propor-tion (i.e. 50 to 60%) of oiled birds are found alive (Char-dine and Pelly, 1994). Chardine and Pelly (1994) notedthat since oil probably kills birds quickly in cold waters,many of the birds found in Newfoundland likely encoun-tered oil relatively close to shore. The low number of liveoiled birds recorded on Sable Island also suggests that sea-birds do not encounter significant amounts of oil near theisland.

4.2. Occurrence of pelagic tar on the beach

In beach surveys elsewhere, marine spills are most oftenindicated by oiled birds rather than by oil washed ashore(e.g. Chardine et al., 1990; Fowler and Flint, 1997). Simi-larly on Sable Island, in over 90% of surveys in which oiledbirds were found, no oil or pelagic tar was found on thebeach. The last oil slick recorded on the island was in Feb-ruary 1970 (Brown et al., 1973) coinciding with the NovaScotia Arrow spill. During the last 30 years no oil slickshave been reported on Sable Island (G. Forbes, Officer-in-charge, Sable Island Station, personal communication).All beached oil has occurred on Sable Island as particlesof pelagic tar lacking a hard weathered outer coating.The lack of noticeable weathering suggests discharge in anorthern climate (MacGregor, 1975).

On the six occasions when significant amounts ofstranded pelagic tar were observed between 1996 and2005, five occurred in winter, during December throughFebruary, and in four of these incidents oil came ashoreeither exclusively, or first, on the north side of the island.In the single summer occurrence, tar was found only onthe south side of the island. The location of tar particlesbeached on the island is generally consistent with seasonalwind patterns. Oil and pelagic tar in the open ocean is influ-

enced by wind speed and direction (Brown et al., 1973;National Research Council, 1985), as these factors deter-mine the conditions of the surface water layer. Thus, gen-erally, in winter wind-driven oil reaching Sable Islandwould be expected to come from the northwest throughnortheast and wash up on the north beach, and in summerwould come from the southwest through southeast andwash up on the south beach. However, movement of pela-gic tar, like that of floating seabird corpses, is likely affectedby the complex circulation in the region.

In terms of amount of pelagic tar and area of shorelineaffected, the February 1998 event, comprised of crude oil,was the largest oiling to occur during 1996–2005. This inci-dent occurred within two weeks of the break-up and sinkingof the Flare east of Cape Breton, and roughly 300 km north-east of Sable Island, however the oil samples recovered onthe island were identified as slight to moderately weatheredcrude oil and were thus unrelated to the Flare sinking.

4.3. Sources of oil contamination

In this study, the marine oil discharge sources identifiedwere crude oil, heavy fuel oils and some bilge discharges.While fuel losses and bilge discharges can occur with allvessel and platform types, crude oils are tanker discharges,from cargo washings or slop tanks. Over 75% of both indi-vidual oil specimens recovered on Sable Island (Table 2),and inferred discharge events (Table 3) were comprised ofcrude oil, and thus originated with tankers. Some of thesedischarges may be legal. MARPOL 73/78 permits dis-charge of oil at 30 litres/nautical mile beyond 50 nauticalmiles of the coast, or outside a specified fishery zone(National Research Council, 2003). Discharges of thisamount can produce a slick on the sea surface since at thisrate of discharge the oil would easily exceed its limitedsolubility in seawater (National Research Council, 2003;M. Balaban, Transport Canada, personal communication).

No birds oiled by light or mid-range distillate fuels wererecovered on Sable Island during the 1996–2005 study.This suggests that fuel oil spills are less frequent and/orsmall, and evaporate and disperse rapidly. Similarly nocondensates from gas and oil production were recoveredfrom samples collected on Sable Island. The reasons forthis are probably similar to those for distillate oils. Thusonly three of five potential oil discharges from marine oper-ations were identified: heavy fuel oils, crude oils and bilgeoils.

Studies at other locations in the Northwest Atlantic(Wiese and Ryan, 2003) report that 90% of oil contamina-tion recovered from seabirds is heavy fuel oil mixed withlubricant oils, a mixture typical of bilge waste, and that‘‘oil found on seabirds in Atlantic Canada originates fromthe illegal pumping of waste oil and oil-water mixtures ofbilges from large trans-Atlantic tankers and cargo/con-tainer vessels’’. Lock and Deneault (2000) contend thatonly about 1% of oil samples from beached birds recoveredin southeastern Newfoundland have been identified as


crude oils, and report that most oil released from ships off-shore is bilge discharge. These and other sources reportthat most oiled seabirds were killed by oil mixtures consid-ered typical of what accumulates in the bilges of largeocean going ships (e.g. Wiese, 2002). The differencebetween the findings in the above studies and thosereported here could reflect circumstances particular to theSable Island Bank area.

Two other possible sources of crude oil discharge in theScotian Shelf region are natural oil seeps, and leakage fromrusting tanker and supply vessels sunk during WWII.Although there are many natural petroleum seeps in theScotian Shelf, all known seeps are releasing gas (J. Mac-Donald, Nova Scotia Petroleum Directorate, personalcommunication). No crude oil seeps have been reported,but if some unknown oil seeps exist, the crude oil releasedwould be Cohasset-Panuke, a light oil which would evapo-rate rapidly and would not form tar balls or the thick blackoiling observed on seabirds. Oil leakage from sunkenWWII vessels has been reported in the south Pacific (Gil-bert and Nawadra, 2002; Hadfield, 2003). Off Canada’seast coast, during WWII, crude oil was shipped from Ven-ezuela and Aruba to refineries in Montreal, and unrefinedcrude was also shipped from North America to Britain inconvoys dispersed over the Scotian Shelf and Grand Banksareas (J.M. Milner, Military and Strategic Studies Pro-gramme, University of New Brunswick, personal commu-nication). Tankers were prime targets for Germansubmarines, and there were particularly heavy lossesduring 1942 (J.M. Milner, personal communication). Seasurface slicks formed by natural oil seeps or leakage fromsunken WWII vessels would be indicated by multipleobservations of slicks at one or more particular locations.We are not aware of such phenomena reported from aerialpatrols and remote sensing programs conducted off Can-ada’s east coast (see Fingas and Brown, 2002).

Oils characteristic of losses from production platformswere not identified in samples collected during the 1996–2005 study. A single sample, collected in early 1999, wasreported as comprised of Cohasset Crude (Lucas, 2003),but when data were reviewed, this sample was identifiedas moderately weathered crude oil of unknown origin.

Analyses of samples collected on Sable Island have beenreviewed to identify oil chromatograms that are similar andthus probably from the same event. There are, however,significant problems in interpreting the data. In some casesoil may be from a single discharge event but because theoiled seabirds took varying amounts of time to reach SableIsland exact matching may not be evident. For example,samples of crude oil recovered from two or more birdcorpses may show different degrees of weathering but thisdoes not confirm that the birds were oiled in differentslicks. The birds may have been oiled in the same slickbut varied in their time and location of death, and thelength of time at sea before reaching Sable Island. As well,with unusual mixtures of oils, or very low concentrations ofone or more minor components, it is not always possible to

clearly identify the source of all samples of oil contamina-tion found on beach and birds. Low levels of lubricatingoils mixed with crude oils may be difficult to identifybecause their hydrocarbon ranges are partly concurrent.However, lubricating oils with fuel oils would be easilyidentified since their hydrocarbon ranges are not concur-rent. The GC/FID chromatograms may also experienceinterferences from waxes or natural oils on the seabirdsthemselves, or other surfactants (Wang et al., 1997). Sincethese compounds are natural in origin and not petrogenicthey would not contain regular n-alkane peaks. Neverthe-less, using chromatogram data in combination with fielddata (including date, bird species, corpse condition), ithas been possible to identify separate discharge events.

While three generic oil types were found on seabirdcorpses during the study period, only crude oil and fueloil were recorded in samples of pelagic tar. Tars foundon the beach in February 1998 and January 1999 were bothcrude oil from single discharge events.

All samples of oiled seabird feathers collected in theDecember 2001 set were contaminated with oil originatingfrom a single discharge event. However, of the 35 sets inwhich multiple samples were collected during 1996–2005,19 (54.3%) included oils from two or more different dis-charge sources. For example in January 2001, four alcidcorpses were recovered from the beach on the same date,and GC/FID analyses indicated that all four had encoun-tered different oil sources: two different crude oils, one fueloil, and one bilge mixture (Table 2). This is consistent withfindings elsewhere (Furness and Camphuysen, 1997) whichsuggest that the source of most contamination is chronic oilpollution comprising various types of oil from poor andillegal day-to-day operations in the marine shipping andoil industries.

The results of GC/FID analyses reported above indicatethat the 170 oil samples collected from bird corpses during1996–2005 represent >70 different discharge events. A totalof 2343 oiled bird corpses were recovered during this per-iod, and thus a simple extrapolation would suggest that900 or more oil discharge events may have contributed tocontamination of birds at sea during the eight year studyperiod.

Both beach location and seasonal occurrence of pelagictar and oiled seabirds washed ashore on Sable Island, sug-gest that oil pollution is more serious in waters northwestthrough northeast of the island, and/or that more seabirdsare exposed to marine oil pollution northwest throughnortheast of the island. However, the island is located inthe middle of high-density shipping routes (Fig. 1). It ispossible that some oil discharged southwest through south-east of Sable Island is transported northward past theisland, and then moved back to the area during winter.

5. Conclusions

GC/FID analyses of the 183 samples collected on SableIsland during 1996–2005 indicate that a large proportion of


seabird mortality observed at Sable Island is caused by dis-charges from crude oil tankers. Although annual numbersof discharge events marked by beached pelagic tar andoiled birds recorded on Sable Island in 2003–2005 werelower than the apparent peak of 16 events in 2001 (Table3) it cannot be concluded that crude oil and heavy fueloil discharges in the region are decreasing. This study hasdemonstrated that oil samples recovered in the same loca-tion and on the same date may represent several dischargeevents, thus it is likely that unless samples are collectedfrom all oiled birds washed ashore, many discharge eventswill be missed.

In the context of the Northwest Atlantic there has beena lack of data concerning generic sources of oil found onshorelines and seabirds. Although beach surveys do notyield information on actual numbers of birds oiled, theydo provide information about trends in oiling rates and dis-charge sources of oil contamination. The low occurrence ofoil found on the beach in surveys in which oiled birds werefound confirms that monitoring of beached birds is usefulfor detecting evidence of many oil discharges not identifiedas slicks on the sea surface. The results presented here pro-vide a significant data base for drawing generalized conclu-sions on the generic sources of oil fouling the seabirdsrecovered in the beached bird survey program conductedon Sable Island. Continued sampling and analysis of oiledbirds from Sable Island should permit extension of thisdata set and help determine whether educational and regu-latory efforts addressing tanker and other shipping opera-tions will result in cleaner seas.

Acknowledgements

The beached bird survey program on Sable Island wasfunded by Sable Offshore Energy Project, ExxonMobilCanada and EnCana Corporation. We thank Gerry Forbesand the staff of the Sable Island Station (MeteorologicalService of Canada, Environment Canada), and the NovaScotia Petroleum Directorate, for logistical support pro-vided throughout the study. We are also grateful for theadvice and assistance provided by Peter Wells, RogerPercy, Art Cook, and Mike Balaban, and thank FrancisWiese and Pierre Ryan for use of their figure.

References

Bibby, C.J., 1981. An experiment on the recovery of dead birds for theNorth Sea. Ornis Scandinavica 12, 261–265.

Bibby, C.J., Lloyd, C.S., 1977. Experiments to determine the fate of deadbirds at sea. Biological Conservation 12, 295–309.

Brown, R.G.B., 1985. The Atlantic alcidae at sea. In: Nettleship, D.N.,Birkhead, T.R. (Eds.), The Atlantic Alcidae. The Evolution, Distri-bution and Biology of the Auks Inhabiting the Atlantic Ocean andAdjacent Water Areas. Academic Press, pp. 384–425.

Brown, R.G.B., 1991. The natural history of oil spills. In: White, J., Frink,L., Williams, T.M., Davis, R.W. (Eds.), The Effects of Oil on Wildlife.International Wildlife Rehabilitation Council Publication, Suisun, CA,pp. 121–124.

Brown, R.G.B., Gillespie, D.I., Lock, A.R., Pearce, P.A., Watson, G.H.,1973. Bird mortality from oil slicks off Eastern Canada, February–April 1970. Canadian Field-Naturalist 87, 225–234.

Butler, J.N., Wells, P.G., Johnson, S., Manock, J.J., 1998. Beach tar onBermuda: recent observations and implications for global monitoring.Marine Pollution Bulletin 36, 458–463.

Chardine, J.W., 1991. Newfoundland: crossroads for seabirds andshipping in the NW Atlantic. In: White, J., Frink, L., Williams,T.M., Davis, R.W. (Eds.), The Effects of Oil on Wildlife. InternationalWildlife Rehabilitation Council Publication, Suisun, CA, pp. 113–120.

Chardine, J.W., 1995. The distribution and abundance of aquatic birds inCanada in relation to the threat of oil pollution. In: Frink, L. (Ed.),Wildlife and Oil Spills: Response, Research, and Contingency Plan-ning. Tri-State Bird Rescue & Research Inc., pp. 23–36.

Chardine, J.W., Pelly, G., 1994. Operation Clean Feather: reducing oilpollution in Newfoundland waters. Technical Report Series No. 198,Canadian Wildlife Service, Atlantic Region, 43pp.

Chardine, J.W., Elliot, R.D., Ryan, P.C., 1990. Beached birds asindicators of marine oil pollution in Newfoundland. In: ProceedingsTwelfth Canadian Waste Management Conference, October 3–5, 1990.St. John’s Newfoundland, pp. 141–150.

Fingas, M.F., Brown, C.E., 2002. Review of oil spill remote sensors. In:The Seventh International Conference on Remote Sensing for Marineand Coastal Environments, Miami, Florida, 20–22 May 2002.

Fowler, A.C., Flint, P.L., 1997. Persistence rates and detection probabil-ities of oiled king eider carcasses on St. Paul Island, Alaska. MarinePollution Bulletin 34, 522–526.

Furness, R.W., Camphuysen, C.J., 1997. Seabirds as monitors of themarine environment. ICES Journal of Marine Science 54, 726–737.

GESAMP, 2004. Estimates of oil entering the marine environment fromsea-based activities. GESAMP Reports and Studies No. 75, May 2004.164pp.

Gilbert, T., Nawadra, S., 2002. Risk of marine spills in the Pacific IslandsRegion. In: The 9th International Spillcon Conference, Manly Sydney,Australia, 16–19 September 2002.

Hadfield, P. 2003. Ecowar looming with sunken wartime wrecks. NewScientist, 2383, February 23, 2003.

Hannah, C.G., Loder, J.W., Wright, D.G., 1996. Seasonal variation of thebaroclinic circulation in the Scotia-Maine region. In: Aubrey, D.G.,Friedrichs, C.T. (Eds.), Buoyancy Effects on Coastal and EstuarineDynamics, Coastal and Estuarine Studies, vol. 53. American Geo-physical Union, Washington, DC, pp. 7–29.

Hannah, C.G., Shore, J.A., Loder, J.W., Naimie, C.E., 2001. Seasonalcirculation on the Western and Central Scotian Shelf. Journal ofPhysical Oceanography 31, 591–615.

Hope Jones, P., Howells, G., Reese, E.I., Wilson, J., 1970. Effect ofHamilton Trader oil on birds in the Irish Sea in May 1969. British Birds63, 97–110.

Lock, A.R., Deneault, L., 2000. Aerial surveillance to detect illegallydumped oil. Backscatter, 14–20.

Lock, A.R., Brown, R.G.B., Gerriets, G.H., 1994. Gazetteer of marinebirds in Atlantic Canada. Canadian Wildlife Service, EnvironmentalConservation Branch, Environment Canada, Atlantic Region, 137pp.

Lucas, Z., 1989. Monitoring persistent litter in the marine environmenton Sable Island, Nova Scotia. Marine Pollution Bulletin 24, 192–199.

Lucas, Z., 2003. Beached bird surveys on Sable Island, 1993–2002. 2003.Offshore Oil and Gas Environmental Effects Monitoring Workshop,May 26–30, 2003. Bedford Institute of Oceanography, Dartmouth,Nova Scotia.

Lucas, Z., Daoust, P.-Y., 2002. Large increase of harp seals (Phoca

groenlandica) and hooded seals (Cystophora cristata) on Sable Island,Nova Scotia, since 1995. Polar Biology 25, 562–568.

Lucas, Z., Hooker, S.K., 2000. Cetacean strandings on Sable Island, NovaScotia, 1970–1998. Canadian Field-Naturalist 114, 45–61.

MacGregor, C. 1975. Weathering of oil spilled on cold, oceanic water withparticular reference to sulphur containing compounds. Unpublished


M.Sc. thesis, Department of Oceanography, Dalhousie University,Halifax, 139pp.

McLaren, I.A., 1981.The Birds of Sable Island, Nova Scotia. Proceedingsof the Nova Scotia Institute of Science 31, 84pp.

National Research Council, 1985. Oil in the Sea. Inputs, Fates and Effects,first ed. National Academy Press, Washington DC, 602pp.

National Research Council, 2003. Oil in the Sea III: Inputs, Fates, andEffects. National Academy Press, Washington DC, 265pp.

Savage, M., Ward, C.H., 1984. Chemical characteristics of laboratoryproduced tar balls. Marine Pollution Bulletin 15, 174–178.

Van Vleet, E.S., Sackett, W.M., Reinhardt, S.B., Mangini, M.E., 1984.Distribution, sources and fates of floating oil residues in the EasternGulf of Mexico. Marine Pollution Bulletin 15, 106–110.

Wang, Z., Fingas, M., 2003. Development of oil hydrocarbon fingerprintingand identification techniques. Marine Pollution Bulletin 47, 423–452.

Wang, Z., Fingas, M., Landriault, M., Sigouin, L., Feng, Y., Mullin, J.,1997. Using systematic and comparative analytical data to identify thesource of an unknown oil on contaminated birds. Journal ofChromatography 775, 251–265.

Wiese, F.K., 2002. Estimation and impacts of seabird mortality fromchronic marine oil pollution off the coast of Newfoundland. Unpub-lished Ph.D. thesis, Department of Biology, Memorial University ofNewfoundland, St. John’s, 321pp.

Wiese, F.K., Ryan, P.C., 2003. The extent of chronic marine pollution insoutheastern Newfoundland waters assessed through beached birdsurveys 1984–1999. Marine Pollution Bulletin 46, 1090–1101.

characterization and source of oil contamination on the beaches and seabird corpses, sable island,...

Documents