Volume 2 4 / N u m b e r 12/December 1992

Marine Pollution Bulletin, Volume24, No. 12, pp. 607-614, 1992. 0025-326X/92 $5.00+0.00
Printed in Great Britain. © 1992PergamonPressLtd

The West Falmouth Oil Spill After

20 Years: Fate of Fuel Oil Compounds
and Effects on Animals
J. M. TEAL*, J. W. FARRINGTON*, K. A. BURNSt, J. J. STEGEMAN*, B. W. TRIPP*, B. WOODIN* and
C. PHINNEY~§
* Woods Hole Oceanographic Institution, Woods Hole, ]VIA 02543, USA
t Bermuda Biological Station for Research, 17 Biological Station Lane, Bermuda GEO1
Environmental Sciences Program, University of Massachusetts-Boston, Habor Campus, Boston, MA 02125, USA
§Current Address: US Geological Survey, Reston, VA 22092, USA.

The barge F/or/da spilled No. 2 fuel oil into Buzzards heavily oiled and another set of sites that had been
Bay, Massachusetts on 29 September 1969. Sediments moderately oiled. At the heavily oiled intertidal site, a
from five of the original stations were sampled in August rainbow sheen that appeared on the surface of the
1989 and analysed for fuel oil hydrocarbons. Two water, and a No. 2 fuel oil odour suggested the
subtidai and one intertidal marsh station showed no continued presence of oil in the marsh. We also took
evidence of fuel oil. One subtidal mud core had traces samples from the original reference sites: 1. a salt marsh
of biodegraded fuel oil at 10-15 cm. One marsh core unoiled in the 1969 spill about 3 km to the south of
contained 10 -6 g g-t dry wt of weathered and bio- Wild Harbor (Little Sippewissett Salt Marsh (SM)), and
degraded fuel oil aromatic hydrocarbons and cycio- 2. an unoiled subtidal station 2.5 km to the west in
alkanes at 5-10 cm with lesser concentrations at 0-5 Buzzards Bay (Fig. 1). None of the oiled or reference
and 10-15 cm. Although present in trace concentra- sites has been directly impacted by spilled oil in the
tions, these hydrocarbons appear to be slightly intervening years, although there have been spills of fuel
inducing cytochrome P4501A in marsh fish (Fundulus oil elsewhere in Buzzards Bay since 1969. The uplands
heteroclitus ). adjacent to these sites are low density residential areas.
In addition to hydrocarbons we analysed the content of
cytochrome P4501A and its activity. Monooxygenase
activities were previously found to be elevated in fish
from the oiled sites (Burns, 1976; Stegeman, 1978).
The long-term fate of spilled oil compounds in
aquatic ecosystems is discussed following each major
spill (NRC, 1985, 1989; Clark, 1982; Jackson et al., Core B-7

1989; Milke, 1990; NOAA, 1988), partly because there

are few long term studies of several years duration of ~ Wild Harbor
the fate and effects of spilled oil. The 650-700 t of No. ~ . ~ Core
30 L r ~ M-,
2 fuel oil spilled by the barge Florida in the West
Falmouth area of Buzzards Bay, Massachusetts in
September of 1969 (Blumer & Sass, 1972; Sanders et
al., 1980; Krebs & Burns, 1977; Burns & Teal, 1979)
have been the subject of some of the more in depth
studies of oil spills (NRC, 1985). Two and three ring
aromatic compounds persisted in intertidal marsh
sediments at a heavily oiled location 6 yr after the spill
(Teal et al., 1978). Because of renewed interest in long

~
Core Sections
term effects of oil spills generated by the many severe o-sc~ Wild Harbor and
spills occurring in 1989, e.g. with the Exxon Valdez spill 5-10 cm Buzzards Bay
10-15 cm
in Prince William Sound, Alaska (Trustee Council, 15-20cm Sampling Stations
1989), we resampled our previous study sites in West 20-25 cm
Falmouth in August, 1989, 20 yr after the spill.
In the Wild Harbor (WH) region we sampled Fig. 1 Bathymetric chart of the Buzzards Bay shoreline showing
sampling stations and sections of subtidal and marsh cores
sediments and organisms from an intertidal marsh and analysed for hydrocarbons. The numbers by the dots are those
a subtidal site that in 1969 were amongst the most used in earlier studies (e.g. Sanders et al., 1980).

607
 Marine Pollution Bulletin

Methods Samples of Fundulus liver from fish killed immedi-

ately after collection, and from WH fish held for various
Sediment sampling and analytical procedures were periods in a clean water laboratory, were analysed for
similar to those we used previously and are described in cytochrome P4501A (see Stegeman, 1992 for nomen-
more detail elsewhere (Burns & Teal, 1979; Teal et al., clature) concentration and activity. Livers were homo-
1978; Farrington et al., 1977; Farrington & Tripp, genized in ice cold 0.05 M Tris/0.15 M KCI,
1975; Farfington et al., 1982). Briefly, we obtained one microsomal fractions prepared by centrifugation and
core from each site by manually inserting a core liner resuspended in Tris buffer with 1 mM EDTA, 1 mM
into the sediment and removing a 9 cm diameter by dithiothreitol, and 20% glycerol (Stegeman et al., 1979).
30 cm deep sediment core. Cores were wrapped in Western immunoblotting was carried out as before
washed foil, frozen and stored until analysis. Frozen (Kloepper-Sams et al., 1987). Microsomal proteins were
cores were cut into 5 cm sections and portions for resolved on polyacrylamide gels and electrophoretically
chemical analysis were subsampled from various core transferred to nitrocellulose. Monoclonal antibody
sections. The measured depth does not account for the 1-12-3 (Park et al., 1986) was the primary antibody
approximately one-third compression of the sample used, the secondary a goat anti-mouse IgG linked to
during penetration of the corer into the sediment. alkaline phosphatase (Bio-Rad). Staining was done with
Individual sediment sub-samples were Soxhlet ex- NBT and BCIP. Purified scup P450E (P4501A1) was
tracted with 1 : 1 methanol: toluene for 1 day followed used as the standard. The amount of stain was
by two separate extractions with 1 : 3 methanol: toluene quantified by densitometry (Master Scan, Scanalytics/
for 1 day each. The combined sulphur-free extracts CSPI, Billerica, MA).
were partitioned into saturated hydrocarbons (fraction- Hepatic microsomal ethoxyresorufin O-deethylase
1) and aromatic hydrocarbons (fraction-2) using silica (EROD) activity was measured spectrophotometrically
gel column chromatography. Each fraction from column as the appearance of resorufin at 572 nm using an
chromatography was analysed by flame ionization gas extinction coefficient of 73 mM -1 cm -1 (Klotz et al.,
chromatography, using a J&W fused silica capillary 1984). Protein content was measured by the method of
column (0.32 mm i.d., by 30 m) coated with Durabond Smith etal. (1985).
D-5 (0.25 ~tm) and programming from a 70 ° injection
temperature to 250 ° at 3 ° mn-~ followed by more rapid
Results
programming to a final temperature of 320°C. Indi-
vidual aromatic hydrocarbons were quantified using Sediments
GC/MS following procedures similar to those described The presence of No. 2 fuel oil hydrocarbons in
in Teal et al. (1978) and Farrington et al. (1982). sediments from the oiled sites is indicated by three
Animals were collected in the immediate vicinity of types of evidence as compared to samples from control
the contaminated intertidal marsh site or in the tidal sites: 1. elevated concentrations of alkanes, cyclo-
creek about 1 m away. Reference animals were alkanes, and aromatic hydrocarbons in the molecular
collected from the reference marsh. Fiddler crabs, Uca weight range for No. 2 fuel oil; 2. composition differ-
pugnax, and marsh mussels, Geukensia demissa were ences in the alkane and cycloalkane fractions indicating
hand picked from the sediment surface. Marsh min- the presence of the very complex mixture of fuel oil
nows, Fundulus heteroclitus were collected with compounds in comparison to the less complicated
minnow traps and softshell clams, Mya arenaria were mixture of biogenic hydrocarbons and biogenic de-
dug from the stream bottom. Hydrocarbon analyses gradation products; and 3. compositional differences in
were done on one composite sample of each species aromatic hydrocarbons in the fuel oil molecular weight
composed of 6-12 individuals. Hydrocarbon analysis range consistent with No. 2 fuel oil aromatic hydro-
began with Soxhlet extraction with methanol using carbon compositions.
added KOH to saponify lipids. Internal standards were The 0-5, 5-10, and 10-15 cm core sections from
added to evaluate recovery in both sediment and animal WH Marsh station M-1 contained aromatic hydro-
samples. After partitioning into hexane, hydrocarbon carbons from No. 2 fuel oil not found in the M-1 15-20
fractions were separated on silica/alumina columns cm section or in control sediments from the nearby
using hexane to elute saturated compounds, followed Sippewissett marsh (Table 1). The greater relative
by 10% ether in hexane, and by 20% methylene abundances of the substituted phenanthrenes (C~-, C2- ,
chloride in hexane. The latter two fractions were and C3-) compared to the unsubstituted phenanthrene,
combined for the aromatic compounds. The fractions fluoranthene, and pyrene in marsh core M-1 sections
were analysed by gas chromatography. The aromatic (Fig. 2a, coupled with the quantitative data of Table 1)
fraction of the animal extracts was also analysed by indicate the presence of aromatic hydrocarbons from
ultraviolet fluorescence (UVF) using marine diesel to No. 2 fuel oil. In the deeper section of core M-1 (15-20
produce a standard response curve. Selected aromatic cm) the relative abundance of phenanthrene, fluor-
fractions were further cleaned by normal phase high anthene, pyrene, benz(a)anthracene and benz(a)pyrene
performance liquid chromatography (HPLC) and the dominates the substituted C~-, C2-, and C3-phen-
reconcentrated aromatic fraction analysed for indi- anthrenes and is indicative of pyrogenic origin of the
viduai aromatics by GC/MS. Methods for tissue aromatic compounds from combustion of fossil fuels,
analysis are described in detail in IOC/UNESCO/ grass and forest fires, and perhaps creosote (NRC 1983,
UNEP (1991). 1985; Wade & Quinn, 1980). The composition of

608
Volume 24/Number 12/December 1992

CORE M 1A - MARSH
Y. m/z 178, 192, 202, 206, 220
loo- O-Sere F! PY

C2Ph

I
E Ph C3Ph
E ~?
t~

"7~ 100- 5 .lOom

e-,

,,.

.~,

101). 10 - 15¢m

~D
, , . , ,

E
~J 1 0 0 15 - 20era
e~

e~
I

C.--
SC~' 1000 1200 1400 161)0 1800 21)00
Time 16:40 20:00 23:20 26:40 30:00 33:20
Fig. 2a Mass chromatograms (Zm/z 178, 192, 202, 206, 220) of
aromatic hydrocarbons in core sections of marsh sediment
samples, August, 1989 Wild Harbor Marsh•

aromatic hydrocarbons found in the bottom of core

M-1 is typical of the composition found at the
Sippewissett Marsh reference station. The surface 0-5
cm of previously lightly oiled marsh station (M-4)
showed no indication of fuel oil hydrocarbons
-H- ¢,O'~
._~ remaining. The mass chromatogram is essentially the
same as for the deep core section (15-20 cm) of station
M-1.
~ The compositions of the alkanes and cycloalkanes
..~,
-~ ~ for sections of core M-1 (Fig. 2b) are consistent with
="d ~ the aromatic hydrocarbon mass spectral data. The
~ resolved peaks at the high-molecular weight end of the
e-
gas chromatograms are n-alkanes of 25-31 carbon
_~ chain length dominated by odd carbon chain lengths,
~__ typical of plant waxes from marsh vegetation. The
cluster of peaks with retention times of 48-55 min is

609
 Marine Pollution Bulletin

Wild Harbor Marsh Sediments and the mass chromatograms for duplicate subsamples
(SIQ. M-'I) of two core sections analysed in this study (M-l, 10-15
O-Scm cm and B-7, 10-15 cm; not shown), show virtually
identical patterns when superimposed. Such similar
patterns have consistently been observed in our
laboratory and elsewhere (Teal et al., 1978; Gearing
& Gearing, 1982). Thus, the differences noted for
composition among the 0-5, 5-10 and 10-15 cm core
U_l sections in marsh core M-1 are real. Differences in
oil concentrations in the replicate analyses can be
attributed to sample heterogeneity due to roots and
large pieces of marsh detritus. While hydrocarbon
concentrations are different for sample replicates, the
relative abundance of compounds (normalized to
benz(a)pyrene) (not shown but easily calculated from
Table 1) shows close agreement for the aromatic hydro-
carbons measured.
Observed differences in concentrations and com-
position between core sections in M-1 indicate spatial
heterogeneities in the original oiling of the marsh
followed by differential water washing of the oil sorbed
to marsh sediments caused by tidal pumping of water;
water washing and mixing by storm events; and

"15-20cm ~~L microbial degradation. Active redistribution processes

were probably continuing in the marsh at the time of
our sampling. The greater relative abundance of
phenanthrene and Cl-phenanthrene relative to the C z-
and C3-phenanthrenes in the 0-5 cm and 10-15 cm
depth intervals in M-1 compared to the 5-10 cm depth
a I . . . . interval; and the higher concentrations for the 5-10 cm
o ~b 2b 3b 4b 5b 6'o 70 interval for the phenanthrenes and other fuel oil
Fig. 2b High resolution gas chromatograms of alkanes and cyclo-
compounds suggests that the 5-10 cm interval may be a
alkanes in the same core sections shown in Fig. 2a. continuing source for fuel oil compounds diffusing and/
or advecting to the other depth intervals via the inter-
stitial water in the marsh. The hole from which core
biosynthetic or early diagenetic alkanes and alkenes M-1 was retrieved filled to the top with water within
found elsewhere in uncontaminated sediments (NRC, minutes of our sampling at low tide.
1985; Blumer & Sass, 1972; Burns & Teal, 1979; The aromatic hydrocarbon data for the marsh core
Farrington et al., 1977; Farrington & Tripp, 1975). The M-1 are consistent with our earlier data for this marsh
unresolved complex mixture of alkanes/cycloalkanes area taken between January 1971 and July 1976 and
(the 'hump' in the chromatograms) characteristic of compared to the original fuel oil aromatic hydrocarbon
biodegraded and weathered fuel oil (NRC, 1983; composition. The original fuel oil aromatic hydro-
Youngblood & Blumer, 1975), is most pronounced in carbon composition was dominated by naphthalene,
the 5-10 cm section and less pronounced or undetected CI-, C2- and C3-naphthalenes. There was a loss of the
in the other sections. Alkane and cycloalkane composi- naphthalene and much of the methyl naphthalenes and
tions and concentrations in the surface samples at C2-naphthalenes relative to the phenanthrenes by
station M-4 and the control station are the same as for January of 1971. By November of 1973 C2-naph-
the deepest section of core M-1 and show no fuel oil thalenes were only barely detectable relative to the
contamination. phenanthrenes and by July of 1976 C2-, and C 3-
Data for the aromatic hydrocarbons and alkanes/ phenanthrenes predominated over the naphthalenes,
cycloalkanes from this study are consistent with what including the C4-naphthalenes. The surface marsh muds
has been reported for No. 2 fuel oil weathering and in 1989 have somewhat higher concentrations of C3-
biodegradation over shorter time intervals of 1 month naphthalenes relative to the C2-, and C3-naphthalenes
to 6 yr for spilled oil in marsh sediments (NRC, 1985; compared to the July 1976 data (Teal et al., 1978). We
Blumer & Sass, 1972; Burns & Teal, 1979), and also do not know whether this is due to spatial heterogeneity
for chronic introduction of water soluble fractions of for oil distributed in this marsh area or if this is due to
No. 2 fuel oil into mesocosms (Wade & Quinn, 1980; remobilization of fuel oil to the surface from deeper in
Gearing & Gearing, 1982). The C2-, and C3-phen- the sediment.
anthrenes and the cycloalkanes are the compounds that Both alkane/cycloalkane data and aromatic hydro-
survived the longest in sediments in those studies. carbon data for subtidal surface sediment samples for
The gas chromatograms of fractions containing the the Buzzards Bay stations B-1 (control station), B-5,
alkanes/cycloalkanes, and the aromatic hydrocarbons, and B-7 indicate only biogenic and early diagenetic

610
Volume 24/Number 12/December 1992

alkanes and alkenes, and pyrogenic source aromatic

hydrocarbons. Trace amounts of alkanes/cycloalkanes
from a weathered and biodegraded No. 2 fuel oil
.4
indicated by the unresolved complex mixture in the gas
chromatogram are present in the 10-15 cm section of e,q

the core at B-7. We found no such trace in the 5-10 cm

section. ~3 (0
e,q e,q
Organisms
The crabs, which burrow into the sediments and feed
on detritus, mud and algae on the sediment surface, I d ,i, 3
showed the greatest indication of contamination by oil
on the basis of GC analysis (Table 2). Figure 3 shows
P,,.
the small unresolved complex mixture (UCM) in the ¢,q
~3
WH animals compared with the SM animals which
lacked a UCM. There was only a trace of oil in the WH (3
crabs as indicated by the clearly detectable biogenic
plant waxes despite the petroleum residues indicated by
the UCM.
The other samples of marsh organisms showed no
consistent difference in hydrocarbon content between
animals from the oiled an unoiled marshes by GC
URE
analyses. All were contaminated with traces of oil with
no differences in UCM patterns. The UVF patterns, Fig. 3 Gas chromatograms of saturated hydrocarbon fractions of
which are also sensitive to hydrocarbon oxidation extracts of fiddler crabs, Uca pugnax. Top: Sippewissett
reference marsh. Bottom: West Falmouth marsh oiled in 1969.
products, showed a slight shift in the emission maxima Labeled peaks are the n-alkanes and isoprenoids typical of
in extracts of crabs and mussels from the oiled marsh as biogenic input from benthic algae (C 15, C 17) and marsh grasses
well as a slightly higher signal (Fig. 4). Although not (C23-C31). URE (UCM) is the unresolved signal due to
residual petroleum hydrocarbons in the fuel oil elution range.
large, this increase in emission between 335 and 360
nm on the synchronous scan is consistent with a slightly
increased content of 3-5 ringed PAH series and/or fuel
oil oxidation products. The C2- and C3-phenanthrenes
were the most persistent of the fuel oil aromatic hydro-
carbons remaining in the sediments. The GC/MS

1
analyses of the crabs from both WH and SM contained
traces of PAHs (Table 3). The reference animals
contained mostly low molecular weight naphthalenes,
while the WH crabs contained traces of the higher -'~ ~'--A~.- 5 nm
molecular weight compounds including chrysene. Such
a difference in composition could also account for the
slight difference in the UVF spectra.
The mussels from both sites would be classified as
relatively clean by coastal monitoring criteria (Burns &
Smith, 1981, 1982; Farrington et al., 1983). The back-
ground contamination appears to be a combination of i I i I i I i I ' I t I i
o
light fuel from boating traffic and PAHs from deposi-
tion of airborne combustion products. -'~ t"-A ~.- 5 nm
The concentration of saturated hydrocarbons in the
Fundulus was very low, as it was in 1975 (Burns & Teal,
1979). There was a slight indication of contamination

TABLE 2
Petroleum hydrocarbon content of animals measured by the size of the
unresolved envelope (UCM) shown on the GC of the fraction
containing the saturated compounds and the ultra-violet fluorescence
of the fractions containing the aromatic compounds (UVF).
' I ' I ' I t I ' I ' I
UCM pgg-~ dry wt UVF ~tgg-t drywt
Genus Oiled Reference Oiled Reference

Uca 1.4 0.3 0.9 0.5 Fig. 4 Synchronous excitation/emission ultra-violet fluorescence
Fundulus 3.6 2.2 43.8 13.5 spectra of extracts from the crabs (top) and mussels (bottom).
Geukensia 8.0 15.2 27.9 19.6 W. Fal. is the oiled marsh, Sipp. is the reference marsh, blank is
Mya 8.6 7.7 31.0 20.8 the complete analytical blank. Note slight differences in the
wavelength of emission maxima.

611
 Marine Pollution Bulletin

TABLE 3 1.2
Polynuclear aromatic hydrocarbons in selected samples based on SIM
GC/MS. Units are ng g-1 dry wt. 1.0

Uca Geukensia 0.8

Compound m/z Sipp. W. Harb. Sipp. W. Harb.
0.6
Naphthalene* 128 0.27 0.07 78.58 90.57
2-Me naphthalene* 142 0.47 0.07 52.00 65.96 0.4
1-Me naphthalene* 142 0.25 63.09 74.97 c~
C2-Naphthalene 156 0.78 43.07 68.27 0.2
C3-Naphthalene 170 0.93 7.96 23.57 trJ
C4-Naphthalene 184 8.61 52.42 0.0
Fluorene* 166 0.09 2.47 4.89
C rFluorene 180 1.66
ESL (control)Little Sippiwissett Wild Harbor
Dibenzothiophene* 184 0.40 0.29 1.13 3.04
C~-Dibenzothiophene 198 0.20 0.01 0.44 1.18 0.09'
C2-Dibenzothiophene 212 3.00 0.08'
C3-Dibenzothiophene 226 1.39
0.07'
Phenanthrene/anthrac* 178 0.54 0.38 10.10 30.61
C t-Phenanthrene 192 0.28 0.30 11.25 11.35 0.06'
Cz-Phenanthrene 206 9.96 <

ii/I
0.05'
Fluoranthene* 202 0.16 3.19 7.66
Pyrene* 202 2.98 7.61 0.04'
Ct-Pyrene/fluoranth 216 5.25 0.03"
Benzanthracene 228 1.35 5.84
Chrysene* 228 0.87 4.67 5.86
0.02-"
Benzo(a)pyrene* 252 0.07 0.22 0.01 -
Perylene* 252 0.27 0.31 0.00
Sum of PAHs 4.12 2.51 302.89 463.66 ESL(control)LittleSippiwissettWildHarbor
*Means compound in calibration mixture so that response factors
are calculated directly. Response factors for others are interpolated as
per UNESCO, 1991.
m/z is the ion used for quantification. )) ~ F.em~le
m~,~

with aromatic compounds in the UVF data. Since

~ 0.8

microsomal cytochrome P4501A in these fish can be

induced by hydrocarbons (Kloepper-Sams et al., 1987),
we examined the amount of this enzyme in Fundulus. 0"6l
P4501A activity measured as EROD activity was
elevated in the WH fish as compared with the SM
reference fish but the latter showed elevated EROD ~ .~ o.o
0.2 4.25
activity compared with other, untreated F. heteroclitus
time in dean water (mo)
held under conditions that reduce the background of
induction (Kloepper-Sams & Stegeman, 1989) (Fig. 5). Fig. 5 Top-EROD (ethoxy resorufin O-deethylase) activity in
Fundulus heteroclitus from the oiled Wild Harbor marsh, a
The same general picture was shown by the amount of reference marsh in Buzzards Bay, Little Sippewissett, and fish
P4501A detected by immunoblot analysis. The differ- held under clean conditions at ESL. Middle-concentrations of
ences between WH and SM fish are only marginally P4501A in the same fish. Bottom--the time course of decrease
in EROD activity in fish held in clean water.
significant, and not very large. The origin of the
apparently elevated EROD or P4501A in the reference
fish is unknown, but could include effects of natural or sediments by physical and biological processes exposed
anthropogenic compounds. Natural products might also the adsorbed oil to both physical removal and to oxygen
be acting in WH. which stimulated degradation. One of the intertidal
To test that the levels of P4501A in WH Fundulus marsh stations had no detectable concentrations of fuel
represented induction and not, for example, a genetic oil. In contrast, the other marsh station contained
difference in P4501A expression, we held fish in clean substantial concentrations of No. 2 fuel oil to a depth of
water and measured EROD activity. Values starting at 1 15 cm. The available knowledge of the fate of spilled oil
nmol min-1 mg -1 had fallen to 0.3 after 2 weeks and to (e.g. Blumer & Sass, 1972) suggests that heavy oiling of
about 0.1 by 13 weeks, indicating that the initial activity this area early in the spill, the high organic content of
was not a fixed character and represented recent the marsh sediments promoting sorption of the oil
induction. compounds, the low energy nature of the marsh
environment, and generally zero oxygen content of the
marsh sediment below the surface all act to retard the
Discussion removal of the oil by physical means of water washing
Of the three subtidal stations from the original oiled and flushing, and by microbial degradation.
areas, only one station shows a trace concentration of The composition of hydrocarbons we found in
weathered and biodegraded No. 2 fuel oil 20 yr after several marsh species is consistent with some fuel oil
the spill. At all of these sites, stirring and flushing of contamination. However, these hydrocarbons were all

612
Volume 24/Number 12/December 1992

at very low levels in all cases, levels that would not Blumer, M. & Sass, J. (1972). Oil pollution: persistence and degrada-
tion of spilled fuel oil. Science 176, 1120-1122.
ordinarily be considered to indicate significant expo- Burns, K. A. (1976). Microsomal mixed function oxidases in an
sure and cannot be unambiguously attributed to estuarine fish, Fundulus heteroclitus, and their inducation as a result
weathered, and degraded No. 2 fuel oil. The only of environmental contamination. Comp. Biochem. Physiol. 53B,
443-446.
conclusion we can confidently make is that animals Burns, K. A. & Teal, J. M. (1979). The West Falmouth oil spill:
today are coming into contact with small amounts of oil hydrocarbons in the salt marsh ecosystem. Estuar. Coast. Mar. Sci. g,
from the sediments contaminated 20 yr ago. However, 349-360.
Burns, K. A. & Smith, J. L. (1981). Biological monitoring of ambient
even these low levels may be responsible for the slightly water quality: the case for using bivalves as sentinel organisms for
greater content of P4501A in the WH fish. monitoring petroleum pollution in coastal waters. Estuar. Coast.
Where does all this leave us with respect to impact ShelfSci. 13,433-443.
Burns, K. A. & Smith, J. L. (1982). Hydrocarbons in Victorian coastal
assessment? There is still some oil present in marsh ecosystems: Chronic petroleum inputs to Wesernport and Port Philip
sediments to a depth of 15 cm, persisting 20 yr after the Bays. Arch. Environ. Contam. Toxicol. 11, 129-140.
spill at least in an area that was heavily oiled in 1969. It Clark, R. B. (1982). The impact of oil pollution on marine populations,
communities and ecosystems: some questions. Philos. Trans. R. Soc.
is there in relatively high concentrations. On the other London B 297, 185-192.
hand, most of the oil has disappeared during this 20 yr. Farrington, J. W., Frew, N. M., Gschwend, P. M. & Tripp, B. W. (1977).
It is virtually all gone from the subtidal sites and even Hydrocarbons in cores of northwestern Atlantic coastal and
continental marine sediments. Estuar. Coast. Mar. Sci. 5,793-808.
from most of the intertidal marsh muds. Our sampling Farrington, J. W., Tripp, B. W., Teal, J., Mille, G., Tjessem, K., Davis, A.
was very limited but we estimate that less than 1% of C., Livramento, J. B., Hayward, N. A. & Frew, N. M. (1982). Bio-
the marsh oiled 20 yr ago is still significantly con- geochemistry of aromatic hydrocarbons in the benthos of micro-
cosms. Toxicology Environ. Chem. 5,331-346.
taminated. The disappearance of the oil must pre- Farrington, J. W. & Tripp, B. W. (1975). A comparison of analysis
sumably involve movement from the anoxic sediments methods for hydrocarbons in surface sediments. In: Marine
where we now find it, to shallower layers of sediment Chemistry in the Coastal Environment, ACS Symposium Series No.
18 (T. M. Church, ed.), pp. 267-284. ACS, Washington, DC.
and/or into the water, oxic environments where the oil Farrington, J. W., Goldberg, E. D., Risebrough, R. W., Marten, J. H. &
could be degraded. Bowin, V. T. (1983). U.S. "Mussel Watch" 1976-1978: An overview
The marsh is now visually no different from other of the trace metal, DDE, PCB, hydrocarbon and artificial radio-
nuclide data. Environ. Science Tech. 17,490-496.
healthy New England salt marshes as long as the oiled Gearing, P. J. & Gearing, J. N. (1982). Behavior of No. 2 fuel oil in the
area is undisturbed. For the first 5-6 yr after the spill, water column of controlled ecosystems. Mar. Environ. Res. 6, 115-
there was no doubt the oil was adversely affecting the 132.
Hall, C. A. S., Howarth, R., Moore, B. & Vorosarty, C. J. (1978).
marsh ecosystem. Twenty years later, the residual Environmental impacts of industrial energy systems in the coastal
effects are extremely small. However, an animal zone. Ann. Rev. Energy 3,395-475.
burrowing into the still contaminated sediments would IOC/UNESCO/UNEP (1991). Determination of petroleum hydro-
carbons in marine sediments. Manual and Guides No. 11 (Rev. 1) (K.
be exposed to oil concentrations that caused significant A. Burns, ed.).
biological effects in the past (Hall et al., 1978). Whether Jackson, J. B. C., Cubit, J. D., Keller, B. D., Batista, V., Burns, K.,
burrowing animals now avoid the oiled area, or are still Caffey, H. M., Caldwell, R. L., Garrity, S. D., Getter, C. D., Gonzalez,
C., Guzman, H. M., Kaufmann, K. W., Knap, A. H., Levings, S. C.
burrowing there and being killed as occurred in the Marshall, M. J., Steger, R., Thompson, R. C. & Weil, E. (1989).
years following the spill (Krebs & Burns, 1977) is Ecological effects of a major oil spill on Panamanian coastal marine
unknown. Such an effect would be most likely to occur communities. Science 243, 37-44.
Kloepper-Sams, P. J., Park, S. S., Gelboin, H. V. & Stegeman, J. J.
with young, overwintering crabs and would be very (1987). Specificity and cross-reactivity of monoclonal and polyclonal
difficult to detect. Given our observations, we conclude antibodies against cytochrome P450E of the marine fish scup. Arch.
that any severe disruption of the marsh sediments in the Biochem. Biophys. 253,268-278.
Kloepper-Sams, P. J. & Stegeman, J. J. (1989). The temporal relation-
area still contaminated could release enough oil to have ships between P450E protein content, catalytic activity, and mRNA
easily observed local effects, the magnitude of which levels in the teleost Fundulus heteroclitus following treatment with
would depend upon the rapidity with which the 13-Naphthoflavone. Arch. Biochem. Biophys. 268,525-535.
Klotz, A. V., Stegeman, J. J. & Walsh, C. (1984). An alternative
released oil was dispersed. Two questions remain: 1. For 7-ethoxy-resorufin O-deethylase activity assay: a continuous visible
what period, between 5 and 20 yr, did the release of oil spectrophotometric method for measurement of cytochrome 17-450
from the sediments continue to have significant and monoxygenase activity. Anal. Biochem. 140, 138-145.
Krebs, C. T. & Burns, K. A. (1977). Long term effects of an oil spill on
readily observable effects on the marsh ecosystem? 2. populations of the salt marsh crab Uca pugnax. Science 197, 484-
Do the very small releases still occurring have a signific- 487.
ant, but now less readily observable effect(s)? Further Milke, J. E. (1990). Oil in the Ocean: The Short- and Long-Term
Impacts of a Spill. Congressional Research Service Report for
analyses of the biota could shed light on the latter Congress 90-356SPR. US Congress, Washington, DC.
question. NOAA (1988). National Marine Pollution Program, Federal Plan for
Ocean Pollution, Research, Development, and Monitoring Fiscal
years 1988-1992. National Ocean Pollution Program Office, Office
We gratefully acknowledge funding support for this project from the of the Chief Scientist, National Oceanic and Atmospheric Adminis-
Donaldson Charitable Trust through the WHOI Coastal Research tration, US Department of Commerce, Washington, DC.
Center, from the US Dept. of Interior Minerals Management Service NRC (1983). Polycyclic Aromatic Hydrocarbons: Evaluation of
(Contract No. 14-12-0001-30393) and from WHOI Sea Grant. We Sources and Effects. National Academies Press, Washington, DC.
thank Mark McCaffrey of WHOI and L. Yelle of BBS for assistance NRC (1985). Oil, in the Sea; Inputs, Fates and Effects. National
with the gas chromatographic analyses and Paul Sherbolom and Academies Press, Washington, DC.
Mohamed Pauzi Zakaria of U. Mass/Boston for assistance with NRC (1989). Using Oil Dispersants on the Sea. National Academies
sediment extractions. We are grateful to our many colleagues who have Press, Washington, DC.
provided advice and input regarding the oil spill studies over the past Park, S. S., Miller, H., Klotz, A. V., Kloepper-Sams, P. J. Stegeman, J. J.
twenty years and to Mr. George Souza, Falmouth Shellfish Warden for & Gelboin, H. V. (1986). Monoclonal antibodies against cytochrome
his long interest. This is contribution No. 7654 from the Woods Hole P-450E from the marine teleost Stenotomus chrysops (scup). Arch.
Oceanographic Institution and No. 1293 from the Bermuda Biological Biochem. Biophys. 249, 339-350.
Station for Research. Sanders, H. L., Grassle, J. E, Hampson, G. R., Morse, L. S., Garner-

613
 Marine Pollution Bulletin

Price, S. & Jones, C. C. (1980). Anatomy of an oil spill: long-term Teal, J. M., Bums, K. & Farrington, J. (1978). Analyses of aromatic
effects from the grounding of the barge Florida off West Falmouth, hydrocarbons in intertidal sediments resulting from two oil spills of
Massachusetts. J. Marine Res. 38, 265-380. No. 2 fuel oil in Buzzards Bay, Massachusetts. J. Fish. Res. Bd. Can.
Stegeman, J. J. (1978). Influence of environmental contamination on 35,510-520.
cytochrome P-450 mixed-function oxydases in fish: Implications for Trustee Council (1989). Exxon Valdez State/Federal Natural Resource
recovery in the Wild Harbor Marsh. J. Fish. Res. Bd. Canada 35, Damage Assessment Plan August, 1989 Public Review Draft.
668-674. Trustee Council, P.O. Box 20792, Juneau, AK 99802.
Stegeman, J. J. (1992). Nomenclature for hydrocarbon inducible Wade, T. L. & Quinn, J. G. (1980). Incorporation, distribution and fate
cytochrome P450 in fish. Mar. Environ. Res. (in press). of saturated petroleum hydrocarbons in sediments from a controlled
Stegeman, J. J., Binder, R. L. & Orren, A. (1979). Hepatic and extra- marine ecosystem. Mar. Environ. Res. 3, 15-33.
hepatic microsomal electron transport components and mixed- Youngblood, W. W. & Blumer, M. (1975). Polycyclic aromatic hydro-
function oxygenases in the marine fish Stenotomus versicolor. carbons in the environment: homologous series in soils and recent
Biochem. Pharmacol. 28, 3431-3439. marine sediments. Geochim. Cosmochim. Acta. 39, 1303-1314.

Marine Pollution Bulletin, Volume24, No. 12,pp. 614-619, 1992. 0025-326X/92$5.00+0.00

Printedin GreatBritain. © 1992PergamonPressLtd

Modelling the Movement of Pollutants

in the UK Shelf Seas
A. J. ELLIOT-F*, A. C. DALE* and R. PROCTORt
* Unit for Coastal and Estuarine Studies, Marine Science Laboratories, Menai Bridge, Gwynedd LL59 5EY,, UK
t Proudman Oceanographic Laboratory, Bidston, Birkenhead, Merseyside L43 7RA, UK

The EUROSPILL oil and chemical spill model has The residual circulation
been extended to include the transport due to residual
currents in the UK shelf seas. Wind-driven circulation The mean wind stress over the UK shelf seas was
patterns for the winter and summer have been derived, derived from estimates of the surface winds at 6 h
and the annual mean (long-term) flow pattern estim- intervals for the period 1955-1986 provided by the
ated. Current speeds during the winter are typically of Norwegian Meteorological Institute. Wind stress was
the order of 0.05-0.10 m s -1, the summer values show computed by a quadratic law using the drag coefficient
a similar spatial pattern but are considerably weaker. formula of Smith & Banke (1975), so that
Trajectories over a period of 100 days are presented for CD = 0.001 × (0.63 + 0.066 × W),
a selection of release points arranged on a regular
latitude/longitude grid. Estimates of patch size after where W is the wind speed in m s-~.
300 days are given for releases from sites of heavy The wind stress vectors were then averaged during
shipping and offshore exploration. Use of the method each month of August and February to produce
for accident simulation is demonstrated by considering climatic estimates of the summer and winter forcing and
the likely spread of Lindane spilled in the English interpolated from the 150 km meteorological grid onto
Channel during March 1989. the hydrodynamic model grid. The sea model solved
the depth-integrated shallow water equations on a
spherical polar grid using a grid size of ~ degree of
latitude by ~ degree of longitude. The model included
M2 tidal currents to ensure realistic bottom friction,
The EUROSPILL oil and chemical spill model (Elliott, and a radiational boundary condition was applied
1991) simulates the effects of wind-driven and tidal around the shelf edge. The model was run to quasi-
transports on the movement of a patch of pollutant. The steady state before the currents were averaged over a
model was developed to provide predictions over time tidal cycle to derive the residual flow. The mean flow
scales of hours to days at the time of an accidental therefore contains contributions both from the non-
release. Consequently it did not include the influence of linear tidal effects and the wind-forcing. The analysis
the mean (residual) currents which only become was made for August (summer) and February (winter)
important over longer time scales. With the inclusion of winds and a mean annual pattern was computed by
the residual flow the model can now be applied to long averaging the two seasonal distributions. Dispersion
time scale problems such as the spread of pesticides and estimates can be made with the choice of either
other long-lived pollutants. summer, winter or annual mean circulation patterns.

614