Content uploaded by Robert K Nelson
Author content
All content in this area was uploaded by Robert K Nelson
Content may be subject to copyright.
Environmental Forensics, 6:273–281, 2005
Copyright C
Taylor & Francis Inc.
ISSN: 1527–5922 print / 1527–5930 online
DOI: 10.1080/15275920500194480
The West Falmouth Oil Spill: ∼100 Kg of Oil Found to Persist
Decades Later
Emily E. Peacock,1Robert K. Nelson,1Andrew R. Solow,1Joseph D. Warren,2Jessica L. Baker,3
and Christopher M. Reddy1
1Woods Hole Oceanographic Institution, Woods Hole, MA, USA
2Southampton College, Long Island University, Southampton, NY, USA
3GeoNet Systems, Bournedale, MA, USA
In order to investigate the long-term fate of petroleum hydrocarbons in salt marsh sediments in Wild Harbor (West Falmouth, MA)
impacted by the Florida spill of 1969, 26 sediment cores were collected and analyzed for total petroleum hydrocarbons (TPH). The
results from this effort indicate that the distribution of petroleum hydrocarbons is spatially heterogeneous, oil compounds are generally
located at sediment depths of 4 to 20 cm in areas closest to the banks of the marsh, and ∼100 kg of petroleum residues can be found
to persist in intertidal sediments that were originally the most impacted.
Keywords: Florida spill, West Falmouth, No. 2 fuel oil, petroleum hydrocarbons, sediments
Introduction
For the past three decades, the West Falmouth oil spill has in-
fluenced and shaped our understanding of post-spill environ-
mental impacts and developed the science of oil spill research
in the marine environment (Blumer et al., 1970, 1971; Blumer
and Sass, 1972a, 1972b; Burns and Teal, 1979; Frysinger et al.,
2003; Hampson and Sanders, 1969; Krebs and Burns, 1977;
Reddy, 2004; Reddy et al., 2002; Sanders et al., 1980; Slater et
al., 2005, Teal et al., 1978, 1992; White, Reddy, et al., 2005).
This spill occurred when the barge Florida went aground near
West Falmouth, Massachusetts, and released ∼700,000 l of No.
2 fuel oil into Buzzards Bay on September 16, 1969 (Figure 1).
Immediately after the spill, scientists at Woods Hole Oceano-
graphic Institution began to analyze samples from impacted
sediments and biota using state-of-the-art analytical technol-
ogy, mainly packed-column gas chromatography (Blumer et al.,
1970). These analyses revealed that petroleum hydrocarbons
could be acted upon by a series of weathering processes in-
cluding evaporation, water-washing, and microbial degradation
but also could persist for at least several years in salt marsh
sediments (Blumer et al., 1972b).
The environmental fate of the spilled oil was intensively stud-
ied for the first seven years after the spill (Blumer et al., 1970,
1971; Blumer and Sass, 1972a, 1972b; Burns, 1975; Burns and
Teal, 1979; Michael et al., 1975; Teal et al., 1978) and then
Received 3 January 2005; accepted 23 May 2005.
Address correspondence to Christopher Reddy, Department of
Marine Chemistry and Geochemistry, Woods Hole Oceanographic In-
stitution, Woods Hole, MA 02543, USA. E-mail: creddy@whoi.edu
some sites were revisited 20 (Teal et al., 1992) and 30 years
(Frysinger et al., 2003; Reddy et al., 2002) later. All of these
studies demonstrated that petroleum hydrocarbons did persist for
many years in the intertidal salt marsh sediments of Wild Harbor
(Figure 1), where no human intervention or cleanup was done
after the spill except for water-based emulsifiers applied imme-
diately after the spill. While investigating intertidal marsh sedi-
ments collected in 2000 at the M-1 site in Wild Harbor (Figure
1), Reddy et al.(2002) were also able to provide a highly re-
fined inventory of existing petroleum hydrocarbons using com-
prehensive two-dimensional gas chromatography (GC ×GC).
This novel technology showed that after three decades of expo-
sure to in situ degradative processes, only the n-alkanes were
completely degraded. However, despite some degradation, all
other compound classes typical of No. 2 fuel oil could still be
identified in the sediments at this site (Reddy et al., 2002). Be-
cause of the potential fears associated with the long-term haz-
ards of petroleum hydrocarbons in the marine environment, the
latter results attracted attention by the mainstream media and
comment by the scientific community. Owens (2003) criticized
this work (Reddy et al., 2002) by stating that only one loca-
tion (M-1) from the whole impacted area was investigated and
that likely much of the original cargo had dissipated. Reddy
et al. (2003) responded to Owens’s comments by stating that
their study was originally focused on only sediments at the
M-1 site because it was one of the most studied and contam-
inated sites. Nevertheless, Owens was correct that additional
sediment cores from a wider area would need to be collected
and analyzed before drawing conclusions about the long-term
fate of the total amount of petroleum released from this
spill.
273
274 E. E. Peacock et al.
Figure 1. Map of general area near the grounding of the Florida.
To address Owens’s comments as well as to continue to
study this oil spill, we surveyed Wild Harbor marsh sediments
for petroleum hydrocarbons. Twenty-six cores from intertidal
marsh sediments on the west and east sides of the Wild Harbor
River were collected and analyzed for petroleum hydrocarbons
(Figure 2). These analytical results were evaluated horizontally
and vertically in order to characterize the spatial distribution
of remaining petroleum in the impacted marsh sediments. The
Figure 2. Map of the study area. Marker size represents the maximum
TPH content (mg g−1) found at each sample location, which was generally
at 12–16 cm. All of the core locations are within the intertidal zone.
data were then analyzed using two independent models aimed at
quantifying the volume of petroleum remaining in these sed-
iments. Finally, these results were compared to the volume
of petroleum released during the spill in order to estimate what
percentage of petroleum originally released persists in these im-
pacted sediments.
Methods
Sediment Collection
From November 2002 to March 2004, 26 sediment cores were
collected from intertidal Wild Harbor marsh sediments. Methods
for collecting and splitting sediment cores from this marsh have
been discussed previously (Reddy et al., 2002). Briefly, a 9.5-cm
diameter PVC tube was pushed into the sediments. The area
around the tube was cleared with a shovel and then the tube
was removed and capped. The core was returned immediately to
the laboratory and frozen at −40◦C. After a brief thaw, the core
wasextruded at 4-cm intervals from 0–4 to originally 44–48 cm.
(However after analyzing several of the first cores collected and
finding that oil was usually limited to depths of 0 to 24 cm, we
then only analyzed sections in the 0–4 to 20–24 cm horizons.)
The outer one centimeter of each core was discarded in order
to avoid cross-contamination. Each sample was placed in pre-
combusted glass jars with Teflon-lined caps and frozen until
analysis.
Total Petroleum Hydrocarbons (TPH) Analysis
Three to nine grams of air-dried sediment were spiked with
40 µgofhexatriacontane (n-C36) and extracted with a 9:1
dichloromethane (DCM):methanol mixture by pressurized fluid
extraction (100◦C, 1000 psi). Extracts were reduced in volume,
solvent exchanged into hexane, and charged onto a glass col-
umn (9 cm ×0.4 cm) packed with fully activated silica gel
(100–200 mesh). The column was eluted with 12 ml of a 3:1
mixture of hexane: DCM. The extract was then treated with ac-
tivated copper to remove elemental sulfur and dried with sodium
sulfate. Finally, the extract was spiked with stearyl palmitate (40
µg) and analyzed on a Hewlett-Packard 6890 Series gas chro-
matograph with a cooled injection system (CIS) and interfaced
to both a Hewlett Packard 5973 mass spectrometer and flame
ionization detector (FID). A 1-µL sample was injected splitless
into the CIS, which was temperature programmed from 40 (0.1-
min hold) to 350◦Cat720◦min−1(8 min hold). Compounds
were separated on a glass capillary column (J&W DB-5MS,
60 m, 0.32-mm i.d., 0.25-µm film thickness) with He as the
carrier gas at a constant flow of 1.5 mL min−1. The GC oven
temperature was programmed from 40 (1-min hold) to 80◦Cat
20◦min−1and then from 120 to 320◦Cat5
◦min−1(10-min
hold). The TPH were quantified by integrating the total FID
area of the unresolved complex mixture (UCM) and using re-
sponse factors determined from No. 2 fuel oil standards. The
carbon range of the UCM was variable but always existed be-
tween decane (n-C10) and hexacosane (n-C26), which excludes
The West Falmouth Oil Spill Decades Later 275
most of the larger, biogenic hydrocarbons. Because a subsample
of the original Florida cargo oil is no longer available, we used
the Marine Ecosystem Research Laboratory (MERL) No. 2 fuel
oil as a calibration standard (Reddy et al., 2002). We believe
that the uncertainty of using a fresh oil as a calibration stan-
dard for a weathered oil is minimal, as an FID is based on the
mass of carbon combusted, which should be relatively insensi-
tive to the chemical nature of hydrocarbons (Reddy and Quinn,
1999). In addition, we assume that most, if not all, of the hydro-
carbons in our operationally defined term for TPH is from the
Florida spill. Laboratory blanks of combusted sand were free of
petroleum compounds. Recoveries of No. 2 fuel oil spiked into
combusted sand and analyzed were 70 to 115% with an average
of 95%. Precision, based on the preparation and analysis of four
duplicate pairs of samples, ranged from +0.1 to 10%. Our esti-
mated method detection limit is 0.015 mg of TPH per gram of
dry sediment (Reddy et al., 2002).
Results and Discussion
Overall of Results
To determine the extent and distribution of TPH in Wild Harbor,
we chose to study the boxed area in Figure 1, which is magni-
fied in Figure 2. All of the cores were collected in the intertidal
regions of the marsh. We chose to investigate this area in Wild
Harbor based on historical results (Teal et al., 1992) and un-
published work from our laboratory, which showed that only
sediments from intertidal, and not subtidal, areas has detectable
petroleum hydrocarbons. As the study continued over more than
ayear, additional core locations were selected for analysis based
on our available data. It was found early in this survey that the
TPH content in marsh sediments farther from the Wild Harbor
River channel were minimal so more emphasis was placed on
areas nearest to the shoreline. Because of the sporadic distribu-
tion of the contamination, it is possible that some contaminated
pockets of the marsh were missed. However, based on 30 years
of historical data and all of the new data from numerous sta-
tions that had not been previously sampled, we believe all of the
hot spots have been included in this study. In total, ∼150 sed-
iment samples were analyzed from 26 sediment cores. (Please
note that only data for 22 cores are shown in Figure 2. The four
other cores were outside the study area shown in Figure 2 and
contained very little or no TPH). In terms of TPH content, the
values vary widely between locations and had a UCM eluting
in the range of n-C10 and n-C26,which is typical of partially
weathered No. 2 fuel oil. Core F (Figure 2) had no detectable
TPH (i.e., no detectable oil residues). Thirteen cores contained
sediments with TPH concentrations greater than 1 mg per gram
of dry sediment (mg g−1). Of those, six sediment cores had max-
imum TPH values ranging from 1 to 4 mg g−1and the remaining
seven sediment cores had TPH values that ranged from 4.6 to
14.1 mg g−1. Core H, which was collected in the same vicinity
of the historical sampling site M-1, had a very similar TPH con-
tent as a core collected at M-1 in August 2000 (Reddy et al.,
2002).
Spatial Trends
Spatially throughout the marsh on the west side of the river,
sediments from cores in closer proximity to the water contained
higher TPH concentrations than the sediments obtained from
cores farther from shore (Figure 2). For example, sediments
from core G contained TPH concentrations about an order of
magnitude less than sediments from core H, while sediments
from core F had no detectable TPH (Figure 2). This difference
in TPH content at three sites occurs over a short distance of
only 10 meters indicating that oil contamination has not spread
laterally since the spill. The best supporting evidence for this
conclusion are photographs of the marsh taken immediately after
the spill. The photos reveal that a 2-wide band of oil had sorbed
to the intertidal marsh sediments closest to the banks of the river.
Hence, areas that were the most contaminated shortly after the
spill continue to be the most impacted three decades later. On
the east side of the river, a similar trend was observed. Cores N,
P, Q, T, and S outline a localized hot spot with higher TPH values
in sediments collected closest to the channel. In this area, which
had not been previously sampled, there is a visible depression
in the marsh surface. Assuming this depression existed at the
time of the spill, finding elevated TPH concentrations in these
sediments is consistent with the hypothesis that the horizontal
distribution of sedimentary TPH concentrations is related to the
original deposition of oil in this area.
Vertical Trends
Ve r tically, the maximum TPH value for each sediment core was
consistently between 4 and 20 cm below the surface (see Figure 3
to view four down-core profiles) and most frequently observed
in the 8–12 or 12–16 cm horizons. These depths are reasonably
consistent with the timing of the spill (1969) and an estimated
sedimentation rate of 0.35 cm yr−1(White, Reddy et al., 2005).
The concentrations of TPH that were found in sediments above
(i.e., in surface 0 to 4 cm) and below (20–24 cm) the mid-core
maximums were relatively minimal. Several of the initial cores
were analyzed to a depth of 48 cm, and no oil was detected be-
low 20 cm. The analysis of subsequent cores was then limited
to a depth of 24 cm, despite the presence of oil in the 20–24 cm
horizon in some samples. The maximum depth of detectable
TPH varied, which could be attributed to either a difference in
marsh conditions or compaction of sediments during sampling.
The shape of the UCM and the presence of resolved peaks on
the UCM hump (Figure 4) varied substantially with depth be-
tween cores, suggesting a depth-dependent variance in the nature
of the petroleum hydrocarbons in these sediments. Preliminary
GC-MS and GC ×GC analysis of some samples indicate that the
composition of the UCM differs as depicted by the shape of the
UCM hump, thus supporting our hypothesis. Mechanistically,
we believe the latter results from post-depositional movement
of oil compounds based on their different abilities to partition
between the pore water and sediment organic matter, a hypoth-
esis that is evaluated and discussed elsewhere (White, Xu et al.,
2005).
276 E. E. Peacock et al.
Figure 3. Typical down-core concentration profiles of TPH at Wild Harbor
from locations denoted in Figure 2 as: (a) Core H, (b) Core S, (c) Core E,
and (d) Core D.
Estimating the Total Oil Remaining in the Marsh
The main objective in this study was to estimate the total amount
of oil remaining in the east and west sides of the marsh. To ac-
complish this task, the total amount of oil (identified as TPH)
present in each side of the channel was volumetrically integrated.
In order to account for the low statistical power caused by a less
than ideal number of cores collected from the study area, two
independent approaches were developed and used to estimate
Figure 4. GC-FID chromatograms of Wild Harbor marsh sediment extracts
from Core E: (a) 0–4 cm, (b) 4–8 cm, (c) 8–12 cm, (d) 12–16 cm, (e) 16–
20 cm, and (f) 20–24 cm. The peaks annotated with * and ** are standards.
Besides the standards, compounds eluting after n-C26 are biogenic hydro-
carbons (see Reddy et al., 2002).
this value (Davis, 2002). Our reasoning is that by using inde-
pendent lines of evidence to arrive at the same objective, if these
two values are similar, we obtain a higher degree of confidence
in the calculated volume of petroleum remaining in these marsh
sediments.
The first approach involved the use of a grid and three-
dimensional interpolation to calculate the volumes. First, all of
the data from the cores were entered into a three-dimensional
grid, which considered an area created by a multisided polygon
of the outer perimeter of the cores. Pixel sizes in the grid were
approximately 1m ×1m in the horizontal dimension and 0.02
minthe vertical. A three-dimensional, isotropic, linear interpo-
lation algorithm (MATLAB, The Mathworks, Inc.) calculated
the oil content at each point in the grid that did not have core
data. These data were then integrated volumetrically to produce
an estimate of 22 kg of TPH still present in the intertidal sedi-
ments on the east side and 15 kg of TPH remaining in the inter-
tidal sediments on the west side for the areas sampled. Unfortu-
nately, this method does not allow a means to confidently predict
uncertainty.
The second approach involves a statistical method that in-
corporates our data set and allows for estimating the total oil
remaining and the uncertainty involved. First, the TPH con-
tent throughout each sediment core was integrated over a 24-cm
depth (Figure 5). We then assumed that the TPH (or oil) content
in each core was homogeneous within one square meter, which
we call Y(x)where xis the two-dimensional location of each
core. The total amount of material contained within our study
region R,the boxed regions in Figure 5, is:
YR=
R
Y(x)dx (1)
The West Falmouth Oil Spill Decades Later 277
Figure 5. Map of the study area with the TPH content (g m−2) listed for
each sample location. This was accomplished by integrating the total oil
found in each core from 0 to 24 cm depth of each sample. Hence, Core
E (see Figure 2) was estimated to have 350 g of TPH per m2in the top
24 cm.
Interest centers on estimating YRfrom concentrations
Y1,Y2,...,Ynmeasured at sample locations x1,x2,...,xn.
To estimate YRefficiently and particularly to give some mea-
sure of uncertainty about it, we will assume that Y(x) has an ln
normal distribution with parameters µ(x) and σ2. These param-
eters represent the mean and variance of log Y(x). The corre-
sponding mean and variance of Y(x) are:
β(x)=exp µ(x)+σ2
2(2)
and:
τ2(x)=(exp σ2−1) exp[2µ(x)+σ2] (3)
It follows that an estimate of YRis given by:
ˆ
YR=
R
exp ˆµ(x)+ˆσ2
2dx (4)
where ˆµ(x) and ˆσ2are estimates of µ(x) and σ2, respectively.
To estimate µ(x) and σ2,wewill assume that:
µ(x)=µo+µ1|x−xo|(5)
where µoand µ1are unknown parameters and |x−xo|is the
distance between location xand an unknown location xo. Under
this model, which was chosen on the basis of some preliminary
analysis of the data, mean ln TPH content declines linearly—and
mean TPH content declines exponentially—with distance from
the location xo(Figures 5 and 6). The model can be written as:
log Y(x)=µo+µ1|x−xo|+ε(x) (6)
where ε(x)isanormal error with mean 0 and unknown variance
σ2.Forfixed xo, this model can be fit by simple linear regression.
Let ˆµo(xo) and ˆµ1(xo)bethe corresponding regression estimates
of µoand µ1, respectively, and let RSS(xo)bethe corresponding
residual sum of squares. The final estimate ˆ
xois the value of
xothat minimizes RSS(xo) and the final estimates of µo,µ1,
and σ2are µo(ˆ
xo), µ1(ˆ
xo),and RSS(ˆ
xo)/(n−2), respectively.
It is possible to go beyond point estimation of YRto provide a
measure of uncertainty. Specifically, provided the scale of any
Figure 6. Plot of the ln of TPH mass (mg) in each sediment core versus
distance from ˆ
xo: (a) west side and (b) east side of Wild Harbor marsh. The
TPH content in each core was calculated by multiplying the area of each
core barrel (0.0708 m2)bythe values shown in Figure 5 and converting to
mg.
278 E. E. Peacock et al.
spatial correlation in TPH content is small compared to the scale
of R:
Va r Y R∼
=
R
τ2(x)dx (7)
where τ2(x)isgiven in (3). An estimate Vˆar YRof the variance
in (7) can be found by replacing µo,µ1, and σ2in the expression
for τ2(x)bytheir corresponding estimates.
The statistical method described above estimated that there is
52 and 45 kg of oil (as TPH) that can be found in the sediments
on the east and west sides, respectively, for the study region R.
The uncertainties for the east and west sides were very good and
were +2 and 6 kg, respectively. The location of ˆ
xois depicted
in Figure 5. Shown in Figure 6 is the fit of the model for the two
data sets. These two independent approaches for estimating the
mass of oil (as TPH) remaining in these marsh sediments give
values that are within the same order of magnitude. However,
the statistical model yielded estimates that were approximately
three times larger than the linear interpolation approach. One
factor for this discrepancy was the study region for the statis-
tical model was ∼200% larger than the linear interpolation ap-
proach because the former required a simple rectangular study
area whereas the latter used a multi-sided polygon that enclosed
and only considered the area within the core locations. Hence,
the linear interpolation approach assumed a zero TPH content
beyond the outer perimeter of the polygon. Nevertheless, both
results reveal that a small fraction of oil can be found compared
to the original 595,000 kg of original product that spilled in Buz-
zards Bay. (This mass was calculated by using a specific gravity
of 0.85 and volume of 700,000 l).
Fate of Oil in Wild Harbor Since the Spill
For further discussion, we have chosen to use the larger estimate
from the statistical model because it provided an uncertainty and
because some oil in the sediment may have been missed in our
field sampling. Hence, less than 0.02% of the total original cargo
that was spilled (∼100 kg for both sides) was found in the inter-
tidal sediments of Wild Harbor. It is important to note that the
amount of oil that initially entered Wild Harbor is not known and
a significant amount of the original product likely evaporated or
was transported offshore within the first few weeks of the spill
(Reddy and Quinn, 2001). For example, ∼20, ∼30, and ∼50% of
product evaporated days to months following spills of the Exxon
Valdez (Wolfe et al., 1994), Amoco Cadiz (Gunlach et al., 1983),
and Jessica (Kingston, 2002), respectively. In May 1970, eight
months after the spill, an estimated four tons of oil still remained
in Wild Harbor (Blumer et al., 1971). Although it is unclear how
this value was calculated, only about 3% of that is remaining; it is
difficult to accurately account for all of these losses. The major-
ity of oil was probably lost in the subtidal sediments, which were
originally heavily oiled but are now relatively free of residues
from this spill. However, some of the petroleum hydrocarbons
were also removed in the intertidal sediments and evidence of
this can be observed in traditional one-dimensional packed-
column GC-FID chromatograms of sediment extracts from the
early to mid 1970s (Burns, 1975; Burns and Teal, 1979). These
chromatograms document the loss of lower molecular weight
compounds, presumably from water washing/evaporation, and
resolved compounds from biodegradation. By assuming that
each compound in these chromatograms has a similar FID re-
sponse and comparing the vintage GC chromatograms to recent
GC chromatograms, we estimate that ∼20 and ∼10% of the mass
of hydrocarbons in the intertidal sediments were removed due to
water-washing/evaporation and biodegradation, respectively,
since 1970. Despite this large estimated decrease since 1970,
concentrations of petroleum hydrocarbons in leftover hot spots
in intertidal sediments, such as M-1, are similar to values de-
tected in the mid 1970s (although analytical methods may cause
slight differences) and suggest that the most oiled locations have
persisted the most. This is consistent with our work at M-1,
where we believe that the removal of phytane and potentially
other saturated, branched compounds has not significantly ad-
vanced in the last two decades (Reddy et al., 2002). Obviously,
these results also need to revolve around the subjects of scale and
recovery. Only a small amount of oil exists in a small area but
at high concentrations. Most of Wild Harbor and Buzzards Bay
does not contain any oil residues from the Florida spill. Blumer
et al. (1971) showed that ∼8km
2of Buzzards Bay was initially
oiled after the Florida grounding. Our recent study indicates that
∼0.001 km2of salt marsh (Figure 2) have detectable residues
three decades later, which corresponds to 0.01% of the original
area impacted. Visually, the marsh appears quite healthy, but
studies are now underway to determine whether any plants or
animals continue to be affected.
Why Do Petroleum Residues Continue to Persist
in Wild Harbor?
While other types of shorelines can retain petroleum residues
for several years after a particular oil spill (Hayes and Michel,
1999), numerous studies and reviews have revealed that marshes
are the most capable of long-term oil preservation (Kingston,
2002; Table 1). For example, Baker (1999) compared numer-
ous spills and different shorelines; she illustrated that sheltered
salt marshes, including Wild Harbor, have the longest recovery
times after an oil spill. Consider Winsor Cove marsh sediments,
which are just 4 km away from Wild Harbor and were contam-
inated in 1974 when the Bouchard 65 spilled an undetermined
amount of No. 2 fuel oil. Recent analysis of Winsor Cove shows
persistent contamination, with most of the oil residing in the
top 3 cm (Reddy, 2004). Crude oil spilled by the Amoco Cadiz
on March 16, 1978, was found in the Ile Grande salt marsh in
France after 13 years. The levels there were generally low, with
localized areas of higher amounts of weathered oil, also an at-
tribute of the remaining oil in Wild Harbor (Mille et al., 1998).
Long-term studies of the sheltered lagoon, Black Duck cove in
Chedabucto Bay, Nova Scotia, where the tanker Arrow spilled
The West Falmouth Oil Spill Decades Later 279
Ta bl e 1. Examples of oil spills with petroleum residues persisting for more than 5 years
Name of vessel Date of spill Location Type of coastline Type of oil Size of spill What is left
Florida September
1969
Buzzards Bay,
Massachusetts
Salt marsh No. 2 fuel oil 700,000 l After 30 years, TPH values are
similar to values observed in
mid 1970s (Reddy et al.,
2002). Only n-alkanes
completely degraded
(Frysinger et al., 2003).
Arrow February
1970
Chedabucto Bay,
Nova Scotia,
Canada
Beach with sand
and cobbles
Bunker C 2 million l After 22 years, n-alkanes were
mostly lost, and target PAHs
were degraded. Triterpanes
and steranes remain (Wang
et al., 1994).
Metula August 1974 Straight of
Magellan,
Chile
Marsh and beach Light Arabian
Crude
50,000,000 l light
Arabian Crude
and 2,000,000
l Bunker C
After 24 years, lightly weathered
oil in one marsh and heavily
weathered in one marsh and
one beach, with most
compounds including
biomarkers showing some
degradation (Wang et al.,
2001).
Bouchard 65 October 1974 Buzzards Bay,
Massachusetts
Salt marsh No. 2 fuel oil 42,000–140,000 l After 30 years, partially
weathered diesel fuel remains
with moderate biodegradation
(Reddy, 2004)
Amoco Cadiz March 1978 Ile Grande Salt
Marsh, North
Coast of
Brittany,
France
Salt marsh,
Rocky and
sandy beaches
Crude, Iranian,
and Arabian
light
223 million l After 13 years, hydrocarbon
levels were generally low, with
localized areas of highly
weathered oil in the marsh.
Biomarkers remain (terpanes,
steranes, and diasteranes)
(Mille et al., 1998).
Ruptured storage tank April 1986 East Coast of
Panama, near
Galeta
Mangroves,
seagrasses, and
coral reefs
Medium-weight
crude oil
12–16 million l Exposed oil weathered rapidly,
while oil protected under
sediments retained most
compounds after 5 years
(Burns et al., 1994).
Exxon Valdez March 1989 Prince William
Sound, Alaska
Wide variety of
beaches
Alaskan North
Slope Crude
oil
42 million l Boulders and cobbles continue to
protect oil in the intertidal
zone, inhibiting disturbance
and weathering (Hayes and
Michel, 1999).
Bunker C oil in 1976, found petroleum hydrocarbons at various
stages of degradation after 20 and 30 years (Lee et al., 2003;
Vandermeulen and Singh, 1994). A study done 22 years after
the Arrow spill found that biomarker compounds at this location
persist while other classes of compounds are at least partially
degraded (Wang et al., 1994). Weathered forms of light Ara-
bian Crude oil spilled into the Straight of Magellan from the
tanker Metula remain after 24 years in some protected marshes
and on one beach, while much of the coastline has been cleaned
by natural processes of the harsh climate (Wang et al., 2001).
Mangroves have also been found to retain oil residues in a way
similar to marshes with pockets of relatively high amounts of
petroleum residues (Burns et al., 1994). In a recent review article
assessing the Exxon Valdez oil spill after 14 years (Peterson et
al., 2003), the authors noted that sediments, which are physically
protected from wave action, disturbance, oxygenation, and pho-
tolysis, remain partially weathered oil for years. In addition to
exhibiting all of these characteristics, salt marsh sediments are
ideal for preserving petroleum hydrocarbons because they are
in temperate regions, have a high organic carbon content, and
become anoxic within the top one cm of the surface (Teal and
Teal, 1969). For example, n-alkanes were degraded four times
faster in tropical mangroves than a temperate salt marsh that
experienced below freezing temperatures in the winter (Burns
et al., 1994). The high organic carbon in salt marshes may aid
in preserving hydrocarbon residues for two reasons. First, many
other natural compounds are available and may be more labile
for microbes to respire. For example, 14C analysis of bacterial
phospholipids in a sediment core collected at the M-1 location
in Wild Harbor in 2001 have indicated that the bacteria in the
contaminated sections of the core were respiring recently pho-
tosynthesized material, which has contemporary 14Cvalues and
not petroleum, which has no detectable 14C (Slater et al., 2005).
Such a finding is consistent with the fact that the TPH in the
most contaminated sites in Wild Harbor only account for ∼10%
of the total organic carbon content (White, Reddy et al., 2005).
280 E. E. Peacock et al.
Second, the elevated organic carbon content allows for increased
partitioning into the sediment and out of the porewater, which in
turn reduces the bioavailibility. While evidence for anaerobic hy-
drocarbon degradation does now exist for select hydrocarbons in
No.2 fuel oil (Widdel and Rabus, 1999), there is no “real world”
data indicating that partially weathered hydrocarbon residues,
which persist in Wild Harbor, can be degraded.
Summary
Wild Harbor marsh continues to store petroleum residues
30 years after the Florida spill. Approximately 100 kg of the
original product can be found. The locations that were most heav-
ily contaminated continue to have the most persistent residues.
It is unfortunate that a more complete mass balance for the
spilled oil cannot be estimated, but it is likely that evaporation
and offshore transport of the fuel residing in the water column
days to weeks after the spill were major removal processes. The
results from this study should pave the way for future efforts
that focus on the geochemical and biological factors for contin-
ued persistence of petroleum hydrocarbons in a relatively small
area. Wild Harbor marsh is ideal for studies directed at gauging
whether any biological effects exist and whether the ecosystem
has recovered, which in turn may aid policy makers in decid-
ing when human intervention is necessary at other impacted salt
marshes.
Acknowledgements
We wish to thank Mr. George Hampson, Ms. Helen White,
Mr. Bruce Tripp, Mr. Sean Sylva, Dr. Li Xu, Dr. John Teal,
Dr. James Quinn, and Dr. John Farrington for their advice and
comments related to this effort. Dr. Robert I. Haddad (Applied
Geochemical Strategies, Inc.) provided a very thorough review
of an earlier draft of this manuscript. Ms. Kristan Davison,
Ms. Danielle Rioux, Ms. Emily Hussey, Ms. Jennifer Webber,
Dr. Emma Teuten, and Mr. Erik Gura assisted with fieldwork.
This work was supported by grants from the National Science
Foundation (CHE-0089172), Woods Hole Coastal Ocean Insti-
tute, Environmental Protection Agency (R-830393), Petroleum
Research Fund Type G starter grant, and the Office of Naval Re-
search Young Investigator Award (N00014-04-01-0029). This is
WHOI contribution # 11405.
References
Baker, J. M. 1999. Ecological effectiveness of oil spill countermeasures:
How clean is clean? Pure Appl. Chem. 71:135–151.
Blumer, M., Sanders, H. L., Grassle, J. F., and Hampson, G. R. 1971. A
small oil spill. Environment 13:2–12.
Blumer, M., and Sass, J. 1972a. Indigenous and petroleum-derived hydro-
carbons in a polluted sediment. Mar. Poll. Bull. 3:92–94.
Blumer, M., and Sass, J. 1972b. Oil pollution: Persistence and degradation
of spilled fuel oil. Science 176:1120–1122.
Blumer, M., Souza, G., and Sass, J. 1970. Hydrocarbon pollution of edible
shellfish by an oil spill. Mar. Biol. 5:195–202.
Burns, K. A. 1975. Distribution of hydrocarbons in a salt marsh ecosystem
after an oil spill and physiological changes in marsh animals from the
polluted environment. Massachusetts Institute of Oceanography/Woods
Hole Oceanographic Institution joint program in oceanography, Ph.D.
dissertation, Woods Hole, MA.
Burns, K. A., Garrity, S. D., Jorissen, D., MacPherson, J., Stoelting, M.,
Tierney, J., and Yelle-Simmons, L. 1994. The Galeta oil spill. II. Un-
expected persistence of oil trapped in mangrove sediments. Estuarine,
Coastal and Shelf Science 38:349–364.
Burns, K. A., and Teal, J. M. 1979. The West Falmouth oil spill: Hydrocar-
bons in the salt marsh ecosystem. Estuarine and Coastal Mar. Science.
8:349–360.
Davis, J. C. 2002. Statistics and data analysis in geology. (3rd ed.). Wiley,
Canada.
Frysinger, G. S., Gaines, R. B., Xu, L., and Reddy, C. M. 2003. Resolving
the unresolved complex mixture in petroleum-contaminated sediments.
Environ. Sci. Technol. 37:1653–1662.
Gunlach, E. R., Boehm, P. D., Marchand, M., Atlas, R. M., Ward, D. M., and
Wolfe, D.A. 1983. The f ate of the Amoco Cadiz oil. Science 221:122–129.
Hampson, G. R., and Sanders, H. L. 1969. Local oil spill. Oceanus 15(2):8–
11.
Hayes, M. O., and Michel, J. 1999. Factors determining the long-term per-
sistence of Exxon Valdez oil in gravel beaches. Mar. Poll. Bull. 38:92–101.
Kingston, P. F. 2002. Long-term environmental impact of oil spills. Spill
Sci. and Technol. Bull. 7:53–61.
Krebs, C. T., and Burns, K. A. 1977. Long-term effects of an oil spill on
populations of the salt-marsh crab Uca pugnax.Science 197:484–487.
Lee, K., Prince, R. C., Greer, C. W., Doe, K. G., Wilson, J. E. H., Cobanli,
S. E., Wohlgeschaffen, G. D., Alroumi, D., King, T., and Tremblay, G. H.
2003. Composition and toxicity of residual bunker C fuel oil in intertidal
sediments after 30 years. Spill Sci. and Technol. Bull. 8:187–199.
Michael, A. D., Van Raalte, C. R., and Brown, L. S. 1975. Long-term effects
of an oil spill at West Falmouth, Massachusetts In Proceedings of Joint
Conference on Prevention and Control of Oil Spills, American Petroleum
Institute, San Francisco, CA, pp. 573–582.
Mille, G., Munoz, D., Jacquot, F., Rivet, L., and Bertranc, J.-C. 1998. The
Amoco Cadiz oil spill: Evolution of petroleum hydrocarbons in the Ile
Grande salt marshes (Brittany) after a 13-year period. Estuarine, Coastal
and Shelf Science 47:547–559.
Owens, E. H. 2003. Comment on “The West Falmouth oil spill after thirty
years: The persistence of petroleum hydrocarbons in marsh sediments.”
Environ. Sci. Technol. 37:2020.
Peterson, C. H., Rice, S. D., Short, J. W., Esler, D., Bodkin, J. L., Ballachey,
B. E., and Irons, D. B. 2003. Long-term ecosystem response to the Exxon
Valdez oil spill. Science 302:2082–2086.
Reddy, C. M. 2004. Oil in our coastal back yard. Oceanus 43:30–35.
Reddy, C. M., Eglinton, T. I., Hounshell, A., White, H. K., Xu, L., Gaines,
R. B., and Frysinger, G. S. 2002. The West Falmouth oil spill after 30
years: The persistence of petroleum hydrocarbons in salt marsh sediments.
Environ. Sci. Technol. 36:4754–4760.
Reddy,C. M., Eglinton, T. I., Hounshell, A., White, H. K., Xu, L., Gaines, R.
B., and Frysinger, G. 2003. Response to comment on “The West Falmouth
oil spill after thirty years: The persistence of petroleum hydrocarbons in
marsh sediments.” Environ. Sci. Technol. 37:2021.
Reddy, C. M., and Quinn, J. G. 1999. GC-MS analysis of total petroleum
hydrocarbons and polycyclic aromatic hydrocarbons after the North Cape
oil spill. Mar. Poll. Bull. 38:126–135.
Reddy, C. M., and Quinn, J. G. 2001. The North Cape oil spill: The fate
of hydrocarbons in Point Judith Pond and Rhode Island coastal waters.
Marine Env. Res. 52:445–461.
Sanders, H. L., Grassle, J. F., Hampson, G. R., Morse, L. S., Garner-Price,
S., and Jones, C. 1980. Anatomy of an oil spill: Long-term effects from
the grounding of the barge Florida off West Falmouth, Massachusetts. J.
Mar. Res. 38:265–380.
Slater, G., White, H. K., Eglinton, T. I., and Reddy, C. M. 2005. Determi-
nation of microbial carbon sources in petroleum contaminated sediments
using molecular 14C analysis. Environ. Sci.Technol. 39:2552–2558.
Teal, J. M., Burns, K. A., and Farrington, J. W. 1978. Analyses of aromatic
hydrocarbons in intertidal sediments resulting from two oil spills of No. 2
The West Falmouth Oil Spill Decades Later 281
fuel oil in Buzzards Bay, Massachusetts. J. Fish. Res. Board Can. 35:510–
520.
Teal, J. M., Farrington, J. W., Burns, K. A., Stegeman, J. J., Tripp, B. W.,
Woodin, B., Phinney, C. 1992. The West Falmouth oil spill after 20 years:
Fate of fuel oil compounds and effectson animals. Mar. Poll. Bull. 24:607–
614.
Teal, J., and Teal, M. 1969. Life and death of the salt marsh. New York:
Ballantine Books.
Vandermeulen, J. H., and Singh, J. G. 1994. ARROW oil spill, 1970–90:
Persistence of 20-yr weathered bunker Cfuel oil. Can. J. Fish. Aquat.
Sci. 51:845–855.
Wang, Z., Fingas, M., Owens, E. H., Sigouin, L., and Brown, C. E. 2001.
Long-term fate and persistence of the spilled Metula oil in a marine
salt marsh environment degradation of petroleum biomarkers. Journal
of Chrom. A 926:275–290.
Wang, Z., Fingas, M., and Sergy, G. 1994. Study of 22-year old Arrow
samples using biomarker compounds by GC/MS. Environ. Sci. Technol.
28:1733–1746.
White, H. K., Reddy, C. M., and Eglinton, T. I. 2005. Isotopic constraints
on the fate of petroleum residues sequestered in salt marsh sediments.
Environ. Sci.Technol. 39:2545–2551.
White, H. K, Xu, L., Lima, A. L. C., Eglinton, T. I., and Reddy, C. M.
2005. Abundance, composition, and vertical transport of PAHs in marsh
sediments. Submitted to Environ. Sci. Technol.
Widdel, F., and Rabus, R. 1999. Anaerobic biodegradation of saturated and
aromatic hydrocarbons. Curr. Opin. Biotechnol. 12:259–276.
Wolfe, D. A., Hameedi, M. J., Galt, J. A., Watabayashi, G., Short, J., O’Clair,
C., Rice, S., Michel, J., Payne, J. R., Braddock, J., Hanna, S., and Sale,
D. 1994. The fate of the oil spilled from the Exxon Valdez. Environ. Sci.
Technol. 28:561A–568A.
