EPA-600/3-77-092
November 1977
Ecological Research Series
HYDROCARBONS IN SEDIMENTS AND BENTHIC
ORGANISMS FROM A DREDGE SPOIL
DISPOSAL SITE IN RHODE ISLAND SOUND
Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Narragansett, Rhode Island 02882
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/3-77-092
November 1977
HYDROCARBONS IN SEDIMENTS AND BENTHIC
ORGANISMS FROM A DREDGE SPOIL
DISPOSAL SITE IN RHODE ISLAND SOUND
Paul D. Boehm and James G. Quinn
Graduate School of Oceanography
University of Rhode Island
Kingston, Rhode Island 02881
Grant No. R803415
Project Officer
Peter Rogerson
Environmental Research Laboratory
Narragansett, Rhode Island 02882
ENVIRONMENTAL RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
NARRAGANSETT, RHODE ISLAND 02882
-------
DISCLAIMER
This report has been reviewed by the Environmental Research Laboratory,
Narragansett, U. S. Environmental Protection Agency, and approved for
publication. Approval does not signify that the contents necessarily reflect
the views and policies of the U. S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
n
-------
FOREWORD
The Environmental Research Laboratory of the U.S. Environmental
Protection Agency is located on the shore of Narragansett Bay, Rhode Island.
In order to assure the protection of marine resources, the laboratory is
charged with providing a scientifically sound basis for Agency decisions on
the environmental safety of various uses of marine systems. To a great
extent, this requires research on the tolerance of marine organisms and their
life stages as well as of ecosystems to many forms of pollution stress.
In addition, a knowledge of pollutant transport and fate is needed.
This report describes a three-year study to investigate the spatial
distribution of hydrocarbons both in surface sediments from upper Rhode
Island Sound and in a commercially important shellfish from the area, the
ocean quahog (Acartia islandica). An attempt is made to distinguish the
regular hydrocarbon geochemistry of Rhode Island Sound, defined by background
hydrocarbon distributions and inputs from Narragansett Bay and adjacent
coastal areas, from the input due to mobilization of hydrocarbons from a
deposited dredge spoil during the five years since the disposal activity has
ceased.
in
-------
ABSTRACT
The hydrocarbon contents of surface sediments, sediment cores, and
ocean quahogs (Arctica islandica) from Rhode Island Sound have been determined.
Hydrocarbon concentrations in surface sediments normally range from 1.0 to
56.1 yg/g, largely dependent on sediment type and sedimentation rates.
However, concentrations up to 301 yg/g are observed in surface samples from
a dredge spoil deposit located in the study area. Based on 1) qualitative
and quantitative hydrocarbon distributions in the sediments, 2) the hydro-
carbon to organic carbon ratio, and 3) the ratio of the concentration of a
prominent cycloalkene compound to organic carbon, the normal hydrocarbon
geochemistry of the region is defined. Using these criteria, the effect of
the dredge spoil deposit (containing 5 to 20 x 103 metric tons of hydro-
carbons) is seen to be insignificant beyond 2 km from the disposal site.
Hydrocarbon contents of the ocean quahog do not reflect the sediment
distributions qualitatively or quantitatively. Throughout the study area
the clams' hydrocarbon contents vary by a factor of 2.5 (2.6 to 6.4 yg/g wet)
while the sediment concentrations vary by two orders of magnitude. The
hydrocarbon assemblage in the clams exhibits a lower boiling point distribu-
tion than that in the sediments.
Key components of the surface sediments are two cycloalkene compounds
of molecular weight 344 and 348. Their concentration covaries very signi-
ficantly with the organic carbon content of the sediment. A major component
of Arctica is another related cycloalkene of molecular weight 342. This
compound is not present in the sediment.
A sediment core from the area shows a decreasing concentration of
hydrocarbons and a decreasing percentage of unresolved components (UCM) with
increasing depth. It is proposed that the rapid increase in the quantity
of the UCM observed at a certain depth within the sediment, can serve as a
chemical marker in the recent sedimentary record. This marker corresponds
to the onset of the industrial revolution and the increased usage of
petroleum products.
This report was submitted in fulfillment of Grant Number R803415 under
the sponsorship of the Environmental Protection Agency. The work cited in
this report was completed as of January 1977 and is included in the Ph.D.
thesis of Paul D. Boehm (University of Rhode Island, 1977). This work has
also been submitted (March, 1977) for publication to the Journal, Estuarine
and Coastal Marine Science in a paper entitled, "Benthic Hydrocarbons of
Rhode Island Sound," by Paul D. Boehm and James G. Quinn.
-------
CONTENTS
Foreword iii
Abstract iv
List of Figures vi
List of Tables vii
Acknowledgements viii
I. Introduction !
II. Methods and Materials 5
III. Results 9
IV. Discussion 28
V. References 35
-------
FIGURES
No. Page
1. Location of study area with sampling station numbers
and sediment types 3
2. Representative gas chromatograms of R. I. Sound
Arctica islandica specimens (a); R. I. Sound surface
sediment (b); Providence River surface sediment
adjacent to sewage outfall (c) 12
3. Mass spectra of unknown compounds 14
4. Relation of organic carbon content of sediment to
concentration of compound Y (HC 344) in sediment 16
5. Relation of organic carbon content of sediment
to its total hydrocarbon content 18
6. Gas chromatograms of hydrocarbons in sediment core
from station 17 20
7. Gas chromatograms of hydrocarbons in sediment core
from station 4 21
-------
TABLES
No. Page
1. Hydrocarbon content of Rhode Island Sound
surface sediments 10
2. Relevant data on cycloalkene compounds 15
3. Hydrocarbon concentrations relative to organic
carbon contents of R. I. Sound and Narragansett
Bay sediments 19
4. Hydrocarbon content of Rhode Island Sound sediment
cores 22
5. Hydrocarbon concentrations in Arctica islandica of
Rhode Island Sound 25
6. Arctica hydrocarbon intercalibration study 27
vn
-------
ACKNOWLEDGMENTS
We would like to express our appreciation to the following individuals
who contributed to the various analytical aspects of this study:
Dr. Chris Brown (URI), Infrared spectra; Dr. Michael Missakian (URI),
NMR spectra; Dr. Nelson Frew (WHOI), Curt Norwood (EPA), GC/MS spectra;
Andy Sweatt (URI), Organic carbon analyses. Dr. Gerry Pesch and Bruce
Reynolds (EPA, Narragansett) aided with the sampling and all shipboard
activities, as did Don Wilcox, the skipper of the Hazel II. We also thank
Sheldon Pratt (URI) for his expert advice on the benthic biology of
Rhode Island Sound. We also appreciate the comments of Dr. John Farrington
(WHOI) on the content of this manuscript. Finally, we wish to thank
Anthony Paulson (URI) with whom we had useful discussions on the potential
significance of the PCB-hydrocarbon relationship in sediment cores.
-------
SECTION I
INTRODUCTION
Rhode Island Sound is that region of coastal water separating
Narragansett Bay from the open waters of the Atlantic Ocean, It is bounded
on the west by Block Island Sound and on the east by Buzzards Bay and
Vineyard Sound (Fig. 1). Between December 1967 and September 1970, R. I.
Sound received 6.3 x 1$ cubic meters of dredge spoil originating in the
Providence River, which flows into the Narragansett Bay estuary south of
the city of Providence (Saila et al., 1972). Coming from an area of
considerable shipping activity and municipal sewage input, the dredge spoil
contained large amounts of associated anthropogenic hydrocarbons (Schultz,
1974; Farrington and Quinn, 1973a, b). This spoil was dumped in a 1.9
kilometer square disposal site located approximately 6.5 kilometers south
of Newport, R. I. (Fig. 1).
It is the purpose of this study to investigate the spatial distribu-
tion of hydrocarbons both in upper R. I. Sound surface sediments and in
the commercially important shellfish from the area, the ocean quahog
(Arctica is!andica). In doing so, we attempt to distinguish the regular
hydrocarbon geochemistry of R. I. Sound, defined by background hydrocarbon
distributions and inputs from Narragansett Bay and adjacent coastal areas,
from the input due to mobilization of hydrocarbons from the deposited
dredge spoil during the five years since the disposal activity has ceased.
GENERAL DESCRIPTION OF STUDY AREA
Physical Circulation
An investigation of the current systems in R. I. Sound (Shonting,
1969) indicated that while the upper layer motion is characterized by a
net westward migrating drift and bottom motions by anticyclonic swirls
with small net displacement, the currents in this region are dominated by
the semidiurnal tide. The net transport of water due to the semidiurnal
tide is small and neither the net tidal transport velocity nor the
magnitude of the instantaneous bottom current are effective in resuspend-
ing sediment (0.1 and 0.3 knots respectively) (Saila et al., 1972).
Wave induced orbital motion of bottom waters at a depth of 30 meters
can exceed that velocity K0.3 knots = 15 cm/sec) theoretically necessary
to resuspend normally deposited unconsolidated sediments in the 0.1 to
1.0 mm size range, especially during winter months. However, Saila et al,
(1972) concluded that while resuspension activity by waves should be
-------
:ure 1. Location of study area with sarrnlinc station numbers and
sediment types.
types:
I sanay silt
silty sand
sand
-------
7I°30'
25'
20'
A ATI-ANTIC
„., BLOCK I OCEAN
33
Kilometers
I I I J._ I |._. .1 I I 1 1 I | L.__
-------
important at the depth of the dump site (2/-30.5 meters), the mode of
deposition of the spoil, plus its mechanical characteristics, have created
an erosion resistant deposition consisting of sediment that can not be
considered unconsolidated.
Sediment Characteristics of R. I. Sound
An extensive survey of R. I. Sound surface sediments by McMaster
(1960) revealed that the study area as shown in Figure 1 consisted of well
sorted fine sand. A large area of silty sand occurs adjacent to the R. I.
mainland and a tongue of this finer sediment lies adjacent to the disposal
site on the southwest. A core of sandy silt is found south of the West
Passage of Narragansett Bay and adjacent to the mainland. McMaster's
study indicates that clay sized particles do not accumulate in this area,
indicating that the finer sediment, presumably of higher organic content,
is transported farther out to sea. In general the quantity of sediment
available for deposition in R. I. Sound is small and storm surges which
resuspend unconsolidated sediment in the Narragansett Bay/R. I. Sound system
probably play significant roles in defining sediment character and accumula-
tion rates (Collins, 1976). The silt deposited in the settling basin
adjacent to the mainland is probably transported out of the Bay south and
westward by near surface currents with coarser silt settling east of the
mainland and finer materials being transported around Pt. Judith into
Block Island Sound (Collins, 1976).
Dredge Spoil Disposal
During the 34 month period when the dredging of the Providence River
navigation channel took place, disposal of mud of high organic content
(^4% organic carbon; Farrington and Quinn, 1973a) was followed by disposal
of material of lower organic content. Thus silty sediment containing
roughly 1-6 mg hydrocarbon/g (Farrington and Quinn, 1973a, Van Vleet and
Quinn, 1977) was buried by silt and sand of lower organic content (il%
organic carbon) and lower hydrocarbon content. A large volume of the spoil
consisted of compact material deposited prior to the time when man first
significantly affected the estuary (Saila et al., 1972) and most likely
consisted of 0.01-0.05 mg hydrocarbons/gram. The resulting 5.5 meter high
mound of spoil deposited in the Sound at a water depth of 29.5-30.5 meters
has a diameter of approximately 1.6 kilometers.
3
Based on the above we estimate that within this mound lies 5-20 x 10
metric tons of hydrocarbons. The post depositional fate of these hydro-
carbons is the focus of this study.
-------
SECTION II
METHODS AND MATERIALS
SAMPLING
Throughout the sampling program special precautions were taken to
avoid any contamination from sampling and storage containers. Surface
sediment samples at 21 stations in the study area were obtained on two
cruises dating 6/27/75 and 7/7/75. A Smith-MacIntyre grab sampler
obtained a relatively undisturbed 0.25 m2, 10 cm deep sediment sample which
when brought on board was subsampled randomly throughout the sample with
a stainless steel trowel. In this manner approximately 500 g (wet weight)
of sediment were obtained and stored in solvent prerinsed glass jars with
aluminum foil-lined caps and were frozen at -20°C approximately 6 hours
later. Two replicate grab samples were taken at each station^ and sub-
sampling duplicates of each grab were obtained, making a total of 4 samples
for analysis from each station.
Sediment cores were obtained at stations 4 and 17 as they represented
stations inside and outside of the disposal area respectively. Cores of
approximately 40 cm length were obtained using a gravity corer with 60 kg
of weight. The plastic core liners containing the sediment were frozen
at -20°C until subsampling was performed. When cut into sections the
sediment which had come in contact with the core liner was shaved off and
discarded.
Ocean quahog specimens were collected on three different sampling
dates, 3/6/75, 6/27/75, 7/7/75 from many of the stations shown in Figure 1.
Benthic quahog sampling stations were coincident with surface sediment
stations except where animals were absent or in low abundance due to
natural population variations in the area. Sampling was accomplished
using a rocking chair dredge. The animals were placed in plastic bags and
chilled over ice on board and then frozen back in the laboratory at -20°C
until analyzed.
Navigation was accomplished using Loran C which is accurate to
within 200 meters in this region.
-------
ANALYTICAL METHODS
Extractions
Sediment and sediment cores--
Samples were thawed overnight and then wet sieved by slurrying with
distilled water, sieving through a 1 mm stainless steel screen and vacuum
filtering through a Whatman 90 filter. Analysis of the filtrate showed that
negligible quantities of hydrocarbons are lost by wet sieving in distilled
water; i.e. these compounds remain associated with the sediments. Wet
sieving removes both shell fragments and any macrofauna and also insures
a relatively homogeneous sample from which one or two subsamples of
approximately 50 grams wet weight each were obtained for analysis. A 10
gram subsample was transferred to a tared beaker and dried at 100°C for 4
hours to obtain moisture content.
Methods of extracting hydrocarbons from sediments are discussed by
Farrington and Tripp (1975), and Rohrback and Reed (1975).
The extraction method chosen was a simultaneous saponification-
extraction technique using 0.5 N KOH in anhydrous methanol, benzene and
distilled water in volume ratios of 2:1:0.2 and using 500 ml of total
solvent to extract a 50 gram wet weight sediment sample. This procedure
both extracts and saponifies the sample in one step. Samples with 50 yg
n~^20 (eicosane) added as an internal standard were extracted under reflux
for 2 hours in a 1 liter round bottom flask.
Doubling the extraction time had no effect on the extraction
efficiency. Re-refluxing the sediment residue from a 2-hour extraction for
an additional 2 hours with fresh solvent yielded less than 1% additional
hydrocarbon material.
After extraction, the mixture was filtered through a preignited
(4 hrs at 450°C) Whatman GF/C filter and the residue extensively rinsed
with petroleum ether. The extraction mixture was combined with the rinses
and 100 ml distilled water added. The benzene-petroleum ether phase (non-
saponifiable fraction) was isolated and the methanol-water phase extracted
twice with 100 ml portions of petroleum ether. The combined benzene
and petroleum ether extracts were combined and taken to dryness on a rotary
flash evaporator at 30°C or less.
Arctica islandica--
Approximately 200 grams (2-3 animals) of tissues and fluids were
removed from thawed specimens and homogenized in a Waring Blendor. To
the homogenate (^90% water) was added 50 yg n-C™ internal standard, 250 ml
of benzene and 100 ml 0.5 N KOH/methanol. The saponification-extraction
proceeded under reflux for 2 hours whereupon the benzene layer was de-
canted. We examined the effect of changing the solvent ratios, saponifying
for 2 hours with 0.5 N KOH/MeOH followed by benzene extraction, doubling
the refluxing time, and doubling the normality of KOH and found no
significant changes in the quantities of hydrocarbons extracted. We did
-------
find, however, that a methanol-benzene extraction under reflux was only, on
the average, 50% as efficient as was the saponification-extraction which
digests the tissues. This result is most likely due to insufficient homo-
genation using the blender. Farrington and Medeiros (1975) employed a high
speed Virtis homogenizer and found the extraction method applied to their
homogenate to be less critical than our large discrepancy would imply.
The extracted non-saponifiable compounds in the benzene phase were
taken to dryness on a rotary flash evaporator.
Chromatographic Procedures
Due to the large amount of pigment material extracted from the sedi-
ments in the nonsaponifiable fraction, an initial cleanup procedure was
employed prior to thin layer chromatographic (TLC) separation. This
involved charging the residue from the above sediment extract in a small
volume of petroleum ether:benzene (95:5) to a 6 mm i.d. column containing
1 gram of alumina deactivated with 5% water. Using 5 bed volumes (^5 ml)
of this 95:5 mixture was sufficient to elute both aliphatic, olefinic and
aromatic hydrocarbons while leaving most pigment material adsorbed.
Our TLC separation of hydrocarbon from non-hydrocarbon compounds has
been described previously (Quinn and Wade, 1974). Briefly, preparative
TLC on Silica Gel G was used for sediment and Arctica samples. The plates
were developed in a system of petroleum etheriNfyOH (100:1) and visualized
by bromothymol blue indicator or UV light. The total hydrocarbons were
isolated by scraping the region defined by separately co-chromatographed
phenanthrene (Rf - 0.5) and n-C?6 (Rf - 1.0). This method of separation
also allows for the isolation of any class of hydrocarbons within the total
hydrocarbon region by visually isolating that region of the plate correspond-
ing to a given spotting standard. A common problem encountered in column
chromatographic separation is the collection of methyl esters along with
the aromatic hydrocarbon fraction. In theory, if a sufficient amount of
water is present in the extraction mixtures, saponification should proceed
with no methyl esters produced via transesterification. However, if
methyl esters are present they are easily separated from the hydrocarbons
by TLC, and an occasional poor separation due to deactivation of the silica
gel is quickly noted.
The isolated hydrocarbons, obtained by extraction of the silica gel,
were analyzed on a Hewlett Packard model 5711 gas chromatograph (GC)
equipped with dual flame ionization detectors. Two meter stainless steel
(2.2 mm i.d.) packed columns of 12 or 15% FFAP on chromosorb W(HP) were
used and hydrocarbon concentrations determined by comparing the area above
baseline with that of the internal standard. Two non-polar columns, 2 m
stainless steel (2.2 mm i.d.) 4% Apiezon L on chromosorb W(HP) and 2 m
stainless steel (2.2 mm i.d.) 2% OV-1 on Anakrom Q, were also used in this
study. Hydrocarbons eluting between n-C-|4 and n-C34 were routinely
quantified. Further details of the GC methodology are found in Quinn and
Wade (1974).
-------
Organic Carbon Analyses
Approximately 10 g of wet sieved sediment were treated overnight with
3 N HC1 to dissolve any carbonate material. The acid was removed by
filtration and the sediment was rinsed with distilled water and dried for
4 hours at 100°C. The dried sediment was ground to a fine powder and
duplicate analyses performed on a Carlo Erba Model 1100 elemental analyzer.
Gas Chromatography/Mass Spectrometry (GC/MS)
GC/MS analyses of selected hydrocarbon extracts were performed on a
Finnigan 1015 quadrupole mass spectrometer coupled to a Varian 1400 gas
chromatograph equipped with a 10 meter SE-30 glass capillary column.
Infrared and Nuclear Magnetic Resonance Spectroscop.y
Infrared (IR) spectra of selected hydrocarbon samples were obtained
on a Perkin Elmer model 521 spectrophotometer. NMR (proton) spectra were
obtained using varian CFD-20 with a micro proton probe.
PCB Analysis (Polychlorinated biphenyls)
Alumina and silicic acid column chromatography was used to isolate
PCB material from the total non-saponifiable extract. Electron capture
GC (Tracor Microtek 220, 63Ni detectors) on OV-17/QF-1 and SE 30/QF-l
columns, was used to analyze the PCB fraction. The chromatograms were
quantified relative to an Aroclor 1254 standard.
Blanks
Throughout this study reagent and procedural blanks were determined
and the reported values have been corrected for these blanks. Arctica
sample concentrations were routinely 10 to 30 times the blank value and
sediment concentrations ranged from 2 to 430 times the blank value averag-
ing 83 times higher. Samples analyzed without the internal standard showed
that natural levels of n-Co were not detectable.
-------
SECTION III
RESULTS
SURFACE SEDIMENTS
The quantities of hydrocarbons present in the surface sediments of
R. I. Sound are presented in Table 1. The values range from a low of
1.0 yg/g found in the coarse sands of station 21 having an organic carbon
content of 0.72 mg/g, to a high of 301 Mg/g found on the southern flank
of the dredge spoil site (Station 4) with an organic carbon concentration
of 8.65 mg/g.
The precision of the values given in Table 1 should be considered as
follows. The precision of the analytical method and chromatographic
analysis is ±5%. Analyzing duplicate samples from the same subsample
(i.e. same wet sieved subsample) yields an average precision of ±9.4%.
Comparing the results from two separate subsamples of the same grab sample
indicates that the subsampling precision is ±18.5% on the average. Re-
plicate grab samples taken at the same station vary by ±26.3%. Farrington
(1971) reported an average variability of ±29% for hydrocarbon analyses of
replicate sediment dredge haul samples from Narragansett Bay.
f* DFl'?Urc 2 l1]"5^3*68 representative gas chromatograms of hydrocarbons
from R. I. Sound Arctica islandica. R. I. Sound surface sediment and
Providence River surface sediments adjacent to the Field's Point sewage
outfall (2a, b, c respectively). R. I. Sound surface sediments consist of
/T°KT \\ T of an unresolved complex mixture (UCM) of hydrocarbons
Uable [). This UCM consists of chromatographically coeluted naphthenic,
naphthenoaromatic, and aromatic hydrocarbons and is characteristic of an
anthropogenic assemblage of hydrocarbons (Farrington and Meyers, 1975;
Farrington et al., 1977a). Superimposed above the UCM are two prominent
peaks (X and Y, Fig. 2b). These compounds contain only carbon and hydro-
gen as revealed by infrared spectroscopy. In addition, combined gas
chromatography/mass spectrometry (GC/MS) reveals that they are cycloalkenes
of molecular weight 348 (C25 H48) and 344 (C25 H44) in the order of elution.
These compounds are identical to those reported by Farrington et al.
U977a) in Buzzards Bay and Gulf of Maine sediments and appear to be very
similar, if not identical to those reported by Gearing et al. (1976) in
faulf Coast sediments. The cycloalkenes are also observed in the surface
sediments of Narragansett Bay as confirmed by GC/MS (this study). The
sedimentary compound reported by Farrington and Quinn (1973a) having a
S«n«°V 65 !!! A?^Z°n L of 2023 appears to be the 344 compound. The
mass spectra of the 344 compound (HC 344) is given in Figure 3 and further
-------
"able 1. Hydrocarcon content of Rhode Island Sound
surface sediments.
Station no.
1 (0-4 cm)
1 (4-3 cm)
2
3
4
8
9
15
16
17
18
19
20
21
30
31
40
43
44
45
46
50
total hydrocarbons
(ug/g dry weight)
18.4
184
116
53.5
301
21.0
104
46.1
96.2
43.9
22.9
21.1
4.0
1.0
29.7
7.6
56.1
27.1
20.8
46.8
13.9
51.2
Organic
Carbon
(mg/g dry wt.)
_ _
n.o
5.62
4.51
8.65
1.55
5.24
3.13
3.76
4.90
3.75
3.60
G.20
0.72
3.54
0.88
5.36
4.65
2.12
3.73
2.38
5.79
% UCM
_ _
86.1
95.0
92.6
97.0
86.5
94.0
92.0
95.0
89.0
90.0
89.5
86.0
72.5
93.4
91.2
90.5
85.5
9-T.O
91.1
91.0
92.5
HC344
(ug/g dry)
0.13
0.27
0.53
1.01
0.65
1.51
1.19
1.04
1.38
1.01
1.01
0.10
0.03
1.03
0.19
1.51
1.08
0.88
0.94
0.69
1.55
UCM = unresolved complex mixture of hydrocarbons.
10
-------
Figure 2. Representative gas chromatograms of R. I. Sound
Arctica islandica specimens (a); R. I. Sound surface
sediment (b); Providence River surface sediment adjacent
to sewage outfall (c) (G. C. data: 2 meter stainless steel
column containing 12% FFAP on Chroraosorb W(HP), temperature
programmed from 100-250°C at 8°C per minute)
Note: 20 = internal standard (n-Cn )
25-31 = n-alkanes of that carbon number
X = cycloalkene, (MW = 348)
Y = cycloalkene (MW = 344)
W = unidentified cycloalkene
Z = cycloalkene (MW = 342)
Q 3 squalene
UCM 3 unresolved complex mixture of hydrocarbons
Arrow indicates point of UCM maximum.
11
-------
UJ
CO
z
o
a.
(O
UJ
a:
a:
o
\-
o
UJ
I-
LLl
O
a
RHODE ISLAND SOUND
Arctica islandica
RHODE ISLAND SOUND
SURFACE SEDIMENT
PROVIDENCE RIVER
SURFACE SEDIMENT
INCREASING TIME AND TEMPERATURE*
12
-------
information on HC 344 and HC 348 is tabulated in Table 2. Farrington et al
(1977a) have reported similar spectra. The mass spectra reveal alkene,
cycloalkane and terpene features with the 348 compound containing one ring
and the 344 possibly two rings.
The most prominent compound, HC 344, appears to consist of a bicyclic
core with two branched aliphatic side chains. The geochemical stability
of HC 344 implies that at least one of the two double bonds is exocyclic
and is stabilized by the ring structure. Its infrared spectrum reveals two
olefinic structures; trisubstituted (815 cnr') and vinyl (885, 910 cnrl).
The lack of a ^970 cnr1 absorption indicates the absence of any trans-
disubstitution. Proton NMR spectrometry indicates a ^-C = CH? structure.
One of the two protons in this vinyl group appears to be deshielded
(6 = 6.8 ppm) by an adjacent olefin (perhaps cyclic). This downfield
shift (6.8 ppm) is indicative of a conjugated structure. The chemical shift
of the other proton is 4.7 ppm (6).
Quantities of HC 344 vary from station to station (Table 1) but are
highly correlated with the organic carbon content of the sediment
(r = 0.955) as shown in Fig. 4. Note that the four stations not showing
this strong correlation, numbers 1, 2, 3, 4, are found within the disposal
site (Fig. 1).
HC 344 and HC 348 are found throughout Narragansett Bay and Rhode
Island Sound sediments. Analyses of phytoplankton, zooplankton, particulate
matter and dissolved hydrocarbons, do not reveal either of these compounds,
although the ocean quahog Arctica islandica (to be discussed below) and
Mercenaria mercenaria the large bivalves of the Sound and Bay respectively
contain these and other similar structures.
Figure 2b shows that the resolved n-alkanes n-C25, C27, C29 and C3i
are also present in the sediments. However,the resolved components, mainly
consisting of these alkanes which have their origins in terrestrial plants,
(Farrington and Meyers, 1975), and the two cycloalkenes, constitute at
most 15% and more typically 10% of the surface sediment total hydrocarbons.
Throughout Narragansett Bay the UCM is a major feature of recent
sedimentary hydrocarbons (Farrington and Quinn, 1973a) as well as of
suspended hydrocarbons in the water column (Schultz, 1974). A comparison
of the chromatograms of R. I. Sound sedimentary hydrocarbons Fig. 2b with
those from the Providence River, Figure 2c, reveals that the latter shows
considerably greater quantities of low boiling resolved and unresolved
features, typical of those from the Fields Point sewage treatment plant
effluent (Farrington, 1971; Farrington and Quinn, 1973b; Van Vleet and
Quinn, 1977). However, analyses of surface sediments from areas of
Narragansett Bay proper (Farrington and Quinn, 1973a; Wade unpublished data;
Van Vleet and Quinn, 1977) indicate that these lower boiling components
are rapidly lost through weathering and an assemblage similar in boiling
point distribution to R. I. Sound sediments (Fig. 2b) is found in the Bay,
and elsewhere in the Providence River
13
-------
HC 344
i*JMlM«uJi
;8 1(1) 118 IM 13d 2M 118 728 Z.M IK) 7S« M8 MB II
M 100 110 119 134 IM ISO
M4
304
320 JM 319
HC 342
M4
M 30 103 110 lid 138 118 IU ICd 179 I8J IV) ZO) Zli
w)
|i'lni'n|»i'l'
78 ioa iao
i.ii.|.nm,r.
io »a iw 3w ba
Figure 3. Mass spectra of unknown compounds; Compound Y; HC 344; sediments and Arctica
Compound I/ HC 342; Arctica.
-------
Table 2. Relevant data on cycloalkene compounds.
en
Mass Spectral Data
Retention Indices
Compound Location
X Sediments,
Arctica
X
(hydrogenated)
Y Sediments,
Arctica
Y
(hydrogenated)
Z
Arctica
Z
(hydrogenated)
Molecular
Weight
348
350
344
348
342
350
Low Mass
Fragments
41,55,57,67,69,81,
83,85,95,97,109,
111,123
43,55,57,69,71,83,
85,97,99,111,113,125
41,43,55,57,69,81,
83,93,95,97,107,
109,121,123
NA
41,43,55,57,67,69,
79,81,83,93,95,107,
109.121.123
NA
High Mass
Fragments
207,235,
250,266
210,238,
266,280
231,247,
259,273,
329
NA
205,217,
259,273
NA
0
12% FFAP
2030
2020
2170
2065
2330
2010
(ovats, ±5) 2%
4% Apeixon L OV-1
2025
2125
2020 2080
2070 2145
2170
2110
Compound designation refers to Figure Z.
NA = not analyzed
-------
Y= 0.264 X+0.07
= 0.955
1-LJ—I—I 1 l l I i I i i i i
0
0
Figure 4.
10.0
ORGANIC CARBON (mg/g)
Relation of organic carbon content of sediment to concentration of
compound Y (h'C 344) in sediment.
12.5
-------
The relation between surface sediment total hydrocarbon and organic
carbon contents is illustrated in Fig. 5. Excluding stations 1, 2, 4 9
from the spoil site and station 16 adjacent to the site, hydrocarbon '
concentrations correlate positively (r = 0.855) with organic carbon content
Including these five stations decreases the correlation coefficient to
r = 0.72 but more importantly results in an unrealistic regression intercept
of minus 41 yg hydrocarbon/g. Farrington's (1971) stations located at the
entrance to Narragansett Bay and on the Continental Shelf south of R I
Sound (40°50'N, 71°00'W) fall very close to the regression line shown in
Fig. 5. The ratio of hydrocarbon to organic carbon may be a useful para-
meter in comparing both the nature and health of the sedimentary environment
in different areas as shown in Table 3. Ratios vary from highs of 5.55 in
the Providence River and 10.7 within the dredge spoil deposit (having its
source in the Providence River), to 2.03 in lower Narragansett Bay, decreas-
ing to 0.96 in Rhode Island Sound. The surface sediments in the disposal
area compare well with lower Narragansett Bay values for this parameter
For comparison, other nearshore and continental shelf values from the region
are given in Table 3. Buzzards Bay appears to be related to adjacent
Rhode Island Sound when considering this parameter (0.63 vs. 0.96).
c+ *• Theinhydrocarbon and organic carbon values at our Station 40 and at
Station WR of Farrington and Quinn (1973a) are in excellent agreement
(hydrocarbons = 56.1 and 60 Mg respectively; organic carbon = 5.36 and
b.b3 mg/g respectively). The two stations are in very close proximity to
one another and in spite of different extraction procedures utilized on
samples taken 5 years apart, similar results are obtained.
SEDIMENT CORES
Figures 6 and 7 illustrate the differences between the core from
Station 17 and that from the flanks of the dredge spoil deposit (Station 4).
Below 18 cm in the Station 17 core (Fig. 6) the resolved components,
notably the n-alkanes n-C2s, C2Z, Cog and Coi emerge from the unresolved
background and the percentage of the UCM, as well as the total hydrocarbon
concentration, decreases with depth (Table 4). As the total hydrocarbon
content and the percentage of UCM decrease with depth in the core, the sum
of the odd n-alkanes, n-C25 to n-C31, does not decrease markedly (Table 4;
Figure 6). The quantity of HC 344 does decrease with depth, while the
organic carbon concentration, after sharply falling off below the top 8 cm,
remains constant with depth.
Unlike the Station 17 core, the core from Station 4 (Fig. 7) shows a
rapid increase of both total hydrocarbon concentration and organic carbon
content with depth (Table 4). The order of magnitude increase in hydro-
carbon concentration within the dredge spoil mound can be ascribed to the
fact that the origin of much of the spoil comes from the Providence River
sediments which contain 1-6 mg/g of total hydrocarbon (Farrington and
Quinn, 1973a; Farrington, 1971; Van Vleet and Quinn, 1977). Furthermore,
the sequence of spoil deposition resulted in the more highly contaminated
spoil being blanketed by cleaner material from areas of the Providence
navigation channel, lower in hydrocarbon concentration (Saila et al., 1972).
17
-------
160
J20
o
CD
o:
o so
o
(T
< 40
O
(301) 4 (184)
16
Y=9.09IX +0.265
r=0.855
J L
1
0 5 10
ORGANIC CARBON (mg/g)
Figure 5. Relation of organic carbon content of sediment to
its total hydrocarbon content. Points denoted by open
circle from Farrington (1971).
18
-------
Table 3. Hydrocarbon concentrations relative to organic
carbon contents of R. I. Sound and Narragansett
Bay sediments.
Location Hydrocarbon:Organic carbon
Average Range
2
Providence River 5.55 5.54-5.56
Upper Narragansett Bay3 5.19 3.17-7.21
Lower Narragansett Bay 2.03 1.76-2.31
R. I. Sound (this study)5 0.96 0.14-2.00
Spoil Site (surface grabs)6 2.35 1.67-3.48
Spoil Site (core 0-45 cm)7 10.7 5.14-17.9
Buzzards Bay8 0.63
Gulf of Maine8 0.29
o
Hudson Canyon 0.19
Concentration of (hydrocarbons [mg/g]/organic carbon [mg/g])
2
x 10 or percentage of organic carbon as hydrocarbon.
2
Stations FP, E, (Farrington and Quinn, 1973a).
3
Stations £„, D (Farrington and Quinn, 1973a).
4
Stations A, B, C (Farrington and Quinn, 1973a).
Excluding stations 1, 2, 4, 9, 16.
This study, stations 1, 2, 4, 9, 16.
Values increase with increasing depth in core.
8Farrington et al. (1977a).
19
-------
RHODE ISLAND SOUND
STATION 17
0-8 cm
20
UJ
en
z
o
Q.
C/3
UJ
01
cr
o
i-
o
LiJ
H
LJ
Q
RHODE ISLAND SOUND
STATION 17
18-28 cm
RHODE ISLAND SOUND
STATION 17
28-38 cm
INCREASING TIME AND TEMPERATURES
Figure 6. Gas chromatograms of hydrocarbons in sediment core
from Station 17. X and Y and 20 defined in Figure 2.
G. C. conditions as in Figure 2.
20
-------
20
RHODE ISLAND SOUND
STATION 4
25 0-5 cm
UCM
UJ
co
z
o
OL
CO
UJ
tr
cc.
o
i-
o
UJ
I-
UJ
Q
RHODE ISLAND SOUND
STATION 4
15-25 cm
UCM
RHODE ISLAND SOUND
STATION 4
35-45 cm
INCREASING TIME AND TEMPERATURES
Figure 7. Gas chromatograms of hydrocarbons in sediment core
from Station 4. X and Y and 20 defined as in Figure 2
b. c. conditions as in Figure 2.
21
-------
Table 4 Hydrocarbon Content of Rhode Island Sound Sediment Cores
(Dry weight basis)
Core No.1
17:
17:
17:
17:
4:
4:
4:
4:
4:
0-8
8-18
18-28
28-38
0-5
5-15
15-25
25-35
35-45
1 ULQ 1
Hydrocarbons
(ng/g)
38.5
21.1
4.6
2.4
184
486
1100
2400
2270
,..rM2 n~C25, 27,29, 31
*UCM (Wg/g)
93.8 0.483
93.4 0.445
79.7 0.340
38.8 0.300
96.0
98.0
98.0
98.0
98.6
HC 344
(ug/g)
1.49
0.51
0.24
0.13
0.68
0.44
nd
nd
nd
Organic
Carbon PCB
(mg/g) (ng/g)
3.50 8.5
2.49 3.1
2.51 nd
2.61 nd
3.57
7.33
13.5
13.6
14.4
Core No. = Station Number:Depth Range(cm)
2UCM = Unresolved Complex Mixture of hydrocarbons.
nd = None Detected
-------
The location of Station 4 is on the southern flank of the dredge spoil mound
(Fig. 1) and consists of the earlier deposited material in contrast to the
residual gravelly lag material located atop (0-4 cm) the mound (Station 1)
of markedly lower hydrocarbon concentration (18.4 yg/g, Table 1).
The 35-45 cm section of core No. 4 (Table 4) contains 12% aromatic
hydrocarbons as separated by thin layer chromatography and analyzed by gas
chromatography. This value is slightly lower than that of the Providence
River surface sediments (^18%) probably due to weathering effects. By
contrast, the surface sediments of R. I. Sound contain lesser quantities of
aromatics (2-9% of the total).
The Station 17 core was analyzed for its PCB content (Table 4). PCB
compounds were present through the top two sections and disappear below
18 cm. The surface value (8.5 ppb) is two orders of magnitude lower than
Providence River surface sediment values (Paulson, personal communication).
The gas chromatographic pattern of the hydrocarbons in the Station 4
core, Fig. 7, mainly shows a broad UCM assemblage. The prominent resolved
feature in the 0-5 cm section, HC 344, is either absent deeper within the
core or else is swamped by the UCM input. At the point of origin of the
spoil material, the Providence River, HC 344 is not observed in the sediment
gas chromatograms (Farrington and Quinn, 1973a; Van Vleet and Quinn, 1977).
Although not exhibiting the strikingly low boiling resolved features which
are typical of Providence River sediment samples adjacent to the municipal
sewage outfall (Van Vleet and Quinn, 1977), (Fig. 2c), core 4 does contain
low boiling unresolved features that are not commonly found in samples
outside of the spoil disposal site (Fig. 7, 15-25 cm vs Fig. 2b and Fig. 6).
ARCTICA ISLANDICA
A typical gas chromatogram of the hydrocarbons found in the ocean
quahog of Rhode Island Sound is shown in Fig. 2a. These hydrocarbons
differ from those in the adjacent sediments. First, there are at least five
biogenic hydrocarbons present in the clams. Examination of these compounds
by combined GC/MS reveals that HC^n and ^344 also are found in the clams.
The mass spectra of these compounds and their hydrogenation products are
identical to the cycloalkenes found in the sediments. Furthermore the
predominant peak in Fig. 2a, (Z), appears to be another related cycloalkene
of molecular weight 342. This compound appears to have four double bonds
and one ring as it hydrogenates to a molecular weight of 350. Its mass
spectra is given in Fig. 3 (HC 342). Additional relevant data on retention
indices are presented in Table 2. Other compounds in the €29-^24 range
include as many as five other cycloalkenes of undetermined molecular weight
and two phytadienes which upon hydrogenation yield a substantial phytane
peak.
A more fundamental difference between the Arctica and sediment chromat-
ograms lies in the elution range of the UCM. As the arrows indicate in
Fig. 2, the Arctica UCM gives a maximum detector response at approximately
a retention time corresponding to n-C23, Wni1e tne sediment UCM, at the
23
-------
same temperature and gas flow rate conditions, peaks at n-Coq- Similar
differences have been noted in Narragansett Bay (Boehm and Quinn, un-
published; Farrington and Quinn, 1973a) where the common bay quahog
Mercenaria mercenaria exhibits a lower UCM boiling point distribution than
do the sediment hydrocarbons in which the organism lives. Teal and
Farrington (1977) also noted that the hydrocarbon boiling point distribution
in mussels is lower than that in the adjacent sediment. Other than the
coexistence of HC 344 and HC 348 in the clams and sediments, Arctica does
not reflect the sedimentary hydrocarbon assemblage.
Arctica specimens from R. I. Sound contain on the average 81% UCM in
their hydrocarbon assemblage, and contain 9-17% aromatic hydrocarbon
material as separated by TLC procedures outlined in Boehm and Quinn (1974).
Hydrocarbon concentrations in whole Arctica samples range from 2.60 to
6.37 yg/g wet weight (Table 5). Thus, the range of concentrations varies
by only a factor of 2.5 throughout the study area whereas analyses of
surface sediments show a range of concentrations over two orders of magni-
tude. The lack of correlation between surface sediment hydrocarbon con-
centrations and Arctica values is best illustrated by the low station 4
Arctica and high sediment values. The highest clam values are found nearer
the mouth of Narragansett Bay (Sta. 44 = 6.37 yg/g) and at Station 8, 0.5 km
southeast of the disposal site (6.12 yg/g). For comparison M. Mercenaria
specimens contain approximately 14-16 yg/g at a Providence River station
(Farrington and Quinn, 1973a) and 4.1-10.0 ug/g at Station C midway down
the Bay and 2.9 to 3.5 yg/g at Station A in Fig. 1 (Table 5).
There is no systematic variation in Arctica's hydrocarbon concentra-
tion with distance from the dredge spoil sTEe^Die nature of the gas
chromatograms remains constant throughout the study area. In general higher
hydrocarbon concentrations are seen at stations ranging from the mouth of
Narragansett Bay, Station 46, (5.52 yg/g'wet), midway between the mouth and
the disposal site, Station 44 (6.37 yg/g wet) to the disposal site itself,
Stations 1, 3, 8 and 9 (4-6 yg/g wet). Concentrations then appear to drop
off on the NW to SE transect (Station 8; 6.12 yg/g wet to Station 22;
2.88 yg/g wet). Concentrations drop off less markedly in the southwest
direction from the spoil site, although due to low faunal abundances at
Stations 30 and 31 sampling gaps appear.
An interlaboratory intercalibration was undertaken between this
laboratory and that of Dr. J. Farrington at the Woods Hole Oceanographic
Institute. Approximately 700 g of Arctica homogenate (blended tissues and
fluids of ten animals) from Sta. 16 were prepared and subsamples analyzed
by both laboratories. Both laboratories employed a saponification-extraction
of a homogenized sample under reflux. Quantification by gas chromatography
was accomplished using the internal standard method. The main differences
in the two methodologies is that a column chromatographic separation of
hydrocarbons from other lipid material is employed by the WHOI group
(Farrington and Medeiros, 1975). Two fractions, fj representing the pentane
eluted hydrocarbons, and fg, that fraction of hydrocarbons eluted by pentane:
benzene (4:1) are isolated. Our separation is by TLC in a developing system'
24
-------
Table 5. Hydrocarbon concentrations in Arctica jj_s]andica
of Rhode Island Sound.
Total hydrocarbons
Station (yg/g wet weight)
1
3
4
8
9
15
16
17
18
19
20
21
22
25
33
44
46
C
A
5.75
5.45
2.95
6.12
4.04
2.84
5.32
5.53
4.57
3.06
3.62
2.62
2.88
2.60
4.05
6.37
5.52
6.542
4.1-10. I2'3
3.504
2.903'4
Precision for above values averages ±20%.
2
Represents concentrations in Mercenaria mercenaria
specimens from Narragansett Bay. For station location
see Farrington and Quinn (1973a).
Values from Farrington and Quinn (1973a).
4
Represents concentrations in f4. mercenaria specimens
-from Narragansett Bay. For station location see Fig. 1
25
-------
of petroleum ether and only the total hydrocarbon fraction is generally
isolated and analyzed. Both GC analyses were carried out on packed columns.
As shown in Table 6, the qualitative nature of the chromatograms is quite
similar showing the same percentages of resolved and unresolved components.
Total hydrocarbon values are within 40% of each other which we considered
to be in reasonable agreement. This is the first such published intercali-
bration that we are aware of using whole organisms.
OTHER HYDROCARBON ANALYSES
Qualitative hydrocarbon determinations were performed on several size
fractions of plankton from lower Narragansett Bay, on two amphipod species
Unciolo irrorata and Leptocheirus pinguis habitating R. I. Sound sediments,
on worms (Nereis sp.) in R. I. Sound sediments and on suspended particulate
matter in lower Narragansett Bay. We were looking for traces of the cyclo-
alkenes from these sources and found none. The particulate hydrocarbons
and the phytoplankton (25 y net) contain 21:6 (heneicosahexaene) as a
prominent component (Lee et al., 1970; Schultz, 1974). Its identity was
confirmed in this study by combined GC/MS. One species of amphipod
(IL. pinguis) contains a prominent polyunsaturated component having the same
retention index as 21:6 (2375 on FFAP). Other polyunsaturated hydrocarbons
are found in the atnphipods and pristane and squalene are major components
of the Nereis worms.
26
-------
Table 6. Arctica Hydrocarbon Intercalibration Study.
G. C. Column
Total Resolved Components
(yg/g wet weight)
Total Unresolved Components
(yg/g wet weight)
Total Hydrocarbons
(yg/g wet weight)
Percent Resolved
Percent Unresolved
U.R.I.
12% FFAP
2 m stainless steel
1.06
4.26
5.32 ± 0.33
(3 analyses)
20
80
W.H.O.I.
OV-17
2 m glass
1.54
5.90*
7.44 ± 0.09
(2 analyses)
21
79
sum of contributions of fraction 1 (f ) and fraction 2 (fj.
27
-------
SECTION IV
DISCUSSION
HYDROCARBON BIOGEOCHEMISTRY
In considering the hydrocarbon geochemistry of Rhode Island Sound one
is faced with a dilemma. The effect of the dredge spoil disposal sequence,
from 1967 to 1970, introducing sediment of high hydrocarbon and organic
matter content, must be examined. However, no R. I. Sound sediment samples
taken prior to the initiation of disposal are available for hydrocarbon
analyses. Thus one must reconstruct the hydrocarbon geochemistry from
measurable, diagnostic parameters which one can use to differentiate a
"normal" from an "abnormal" Narragansett Bay/R. I. Sound system. To this
end we will consider the quantitative and qualitative hydrocarbon chemistry,
trends for a single compound, HC 344, and the organic carbon content of the
sediments.
The unresolved low boiling hydrocarbons found in the Providence
municipal sewage effluent (Farrington, 1971), in Providence River sediments
adjacent to the outfall (Fig. 2c), and within the dredge spoil deposit
(Fig. 7) are not seen in surface sediments outside the disposal area
(Fig. 2b). These components are soon weathered if exposed to the sediment
water interface or resuspended in the water column as evidenced by their
disappearance with increasing distance from the Fields Point sewage out-
fall (Van Vleet and Quinn, 1977). However, if large scale movement of
dredge spoil material had taken place in R. I. Sound we would see evidence
of these components at stations near the disposal site (e.g. station 17).
This is not the case (Fig. 2b and 6).
The absolute quantities of hydrocarbons in the R. I. Sound surface
sediments (Table 1) can not be used directly to evaluate dredge spoil
hydrocarbon movement because of the variations in sediment type and sedi-
mentation rates throughout the study area (McMaster, 1960; Collins,
personal communication).
Considering the hydrocarbon:organic carbon relation (HC/OC) (Table 3),
while the hydrocarbons comprise 5.5% of the organic carbon in the polluted
Providence River sediments, this value falls off rapidly through the
estuary decreasing from 5.5% to about 2% at Station A at the mouth of the
Bay. At Station 40 and Station 46 in R. I. Sound the value decreases to
1.5 and 0.6% respectively. It appears that most of the hydrocarbons are
associated with particles that settle in the Bay. The silty material,
only some of which settles at the mouth of the West Passage of Narragansett
28
-------
Bay (Collins, 1976) is depleted in hydrocarbons relative to total organic
carbon, due to inputs of organic carbon from natural sources which contain
a smaller amount of hydrocarbons relative to Providence River inputs. The
fine particulate material reaching R. I. Sound waters has also been subjected
to greater microbial alteration of its HC/OC ratio probably due to both its
longer residence time in the water column and its greater surface area per
unit weight. The distribution of particulate hydrocarbons in Narragansett
Bay, which shows a decreasing concentration with increasing distance from
the Providence River, correlates well with that predicted from a circulation
model of the Bay (Schultz, 1974) using the upper Providence River area as
the only source of hydrocarbons to the estuary. However, dilution by
natural sources of organic carbon within the Bay would alter the HC/OC ratios
of the particulate natter. Zafiriou (1973) determined the 14C content of
the hydrocarbons from Station E-l of the Providence River, and reported an
average age of 24,400 years indicating that the hydrocarbon chemistry at
this station was defined by 80-97% anthropogenic material and 3-20% recently
biosynthesized compounds.
The covariance of surface sediment hydrocarbon and organic carbon
concentrations (Fig. 5) and the relative uniformity of the HC/OC ratio
throughout the R. I. Sound study area are in sharp contrast to the anomalous
behavior of stations 1, 2, 4, 9 and 16 which are located at or adjacent to
the disposal site. The quantity of organic carbon in the sediment is a
function of the sedimentation rate as well as other factors. Since total
hydrocarbon concentrations covary with organic carbon it appears that
hydrocarbons reach the sediments via normal sedimentation in the Sound.
The sedimentation rate throughout the study area probably varies considerably
due to physical factors, and controls the absolute hydrocarbon concentrations.
However, a consideration of the HC/OC ratio normalizes this variable and
permits a comparison of different stations. Based on this argument, the
only sampling stations to be affected by an input of dredge spoil material,
which has ten times the HC/OC ratio (Table 3), are stations 1, 2, 4, 9 at
the disposal site and station 16, one kilometer to the southwest.
Figure 4 illustrates that the biogenic compound, HC 344, covaries very
significantly with organic carbon at all stations except 1, 2, 3, and 4.
This compound (to be discussed below) serves as an excellent marker of
normal biogenic and/or diagenetic activity in the sediments of this region.
When used in conjunction with organic carbon to normalize the heterogeneity
of the sediments, (Fig. 4), it is apparent that the effect of the high
hydrocarbon dredge spoil material on the benthic environment of R. I. Sound
is confined to an area within 1-2 km of the disposal site.
The apparent lack of contribution of dredge spoil material to R. I.
Sound's hydrocarbon geochemistry, as evidenced in the preceding paragraphs
is a surprising result. Enough spoil was deposited to theoretically cover
the entire study area (^150 km2) with a 4.2 cm layer of sediment with a
high hydrocarbon content. However, it appears that much of the material
settled rapidly and cohesively upon dumping and that little has been re-
suspended since. Alternatively it can be postulated that resuspension of
dredge spoil has taken place, but that this material is carried out of the
29
-------
study area by bottom currents. Gordon (1974) found that 99% of a dredge
spoil deposited in Long Island Sound settled cohesively and initially re-
mained within a 120 meter radius of the impact point. Visual observations
of the R. I. Sound dredge spoil disposal site (Saila et al., 1972) indicated
that an erosion resistant deposit resulted and large scale resuspension had
not occurred. This observation is verified chemically by this study. The
presence of lag deposits on the top of the deposit indicates that some
erosion has most likely occurred. However, if erosion has proceeded, there
is no evidence that the eroded material settled within the study area.
Analyses of Arctica islandica indicate that their hydrocarbon contents
are probably due to a combination of ingestion of particles from the water
column and biosynthesis. The most important influence on their hydrocarbon
content appears to be distance from Narragansett Bay and hydrocarbon sources
therein. Quantitative and qualitative distributions of hydrocarbons in these
clams suggests a small (if any) input of dredge spoil hydrocarbons. It is
possible that the initial disposal sequence introduced hydrocarbons to the
clams, but in the five year interim before sampling for this study commenced,
the Bay's influence has been reasserted. Certainly, the hydrocarbon content
of Arctica islandica throughout the study area is comparable to that in
Mercenaria^ mercenaria commonly harvested in lower Narragansett Bay (Table 5).
All animals contain some indication of an anthropogenic assemblage of
hydrocarbons, having their probable source in chronic inputs from the
Providence River and urban atmospheric fallout within the Narragansett Bay
drainage basin.
The lower boiling point distribution of hydrocarbons exhibited in the
clams relative to the sediments, Fig. 2a and 2b, may be explained by
1) differing sources; 2) biochemical degradation or modification;
3) selective uptake by the clams; 4) selective solubilization of certain
hydrocarbons out of the sediments followed by uptake by the clams, and
5) a variation of the qualitative hydrocarbon assemblage with sediment grain
size and selective ingestion of certain resuspended particle sizes by the
clams. While none of these explanations can be completely ruled out, the
gas chromatographic similarity of particulate hydrocarbons in the water
column (Schultz, 1974) with those in the sediments casts doubt on the first
explanation. However, Thompson and Eglinton (1976) found that some
estuarine benthic diatoms exhibited a lower boiling range of aliphatic
hydrocarbons than was seen in the sediments. Their ingestion by the clams
could result in a lower boiling mixture of unresolved hydrocarbons in the
clams' fluids and tissues.
Biochemical degradation of anthropogenic hydrocarbons is not known for
bivalves although the possibility of their metabolic alteration of hydro-
carbon patterns has been discussed (Lee et al., 1972; Stegeman and Teal,
1973). Solubilization phenomena in sea water (Boehm and Quinn, 1973) can
result in selective mobilization (Boehm and Quinn, 1974) and selective up-
take (Boehm and Quinn, 1976) of classes of hydrocarbons by marine filter
feeders, but details of the partitioning of hydrocarbon species between a
soluble state, a solubilized (colloidal) state and the adsorbed state are
not known. Lee (personal communication) found most hydrocarbon material in
30
-------
a controlled ecosystem experiment (CEPEX) using No. 2 fuel oil to be adsorbed
to particles. However, a possibility that should be studied, is that given
the relatively high dissolved organic carbon (DOC) of interstitial waters,
sediment resuspension or normal leaching by bottom currents could effect a
solubilization of adsorbed hydrocarbons by this level of DOC (Boehm and
Quinn, 1973). Furthermore, low level dissolution (thermodynamic solubility)
of certain hydrocarbons corresponding to a clam-like assemblage may occur
and these hydrocarbons taken up via equilibration across outer membranes of
the clam.
A qualitative as well as quantitative fractionation of lipid materials
adsorbed to sediment particles can occur in sediments. Finer particles
generally are richer in adsorbed organic matter, which includes hydrocarbons,
on a comparable weight basis because of a larger surface area. In addition,
hydrocarbon distributions can vary qualitatively with grain size within a
given sediment (Thompson and Eglinton, 1977) and the adsorption of hydro-
carbons can be a function of mineral type (Meyers and Quinn, 1973).
However, the sediments in much of the study area are comprised of well sorted
sands and the contribution of silt and clay to total sediment is small
(McMaster, 1960). Sediment size variation and selective uptake by the clams
probably can not alone account for the observed hydrocarbon pattern
variations between the clams and the sediments. Also, given the strong
affinity of hydrocarbons for solid surfaces, ingestion of whole sediment
particles may not result in any transfer of hydrocarbons from the sediment
particle to the clam tissues. These particles, with their adsorbed hydro-
carbons, may be rejected by the clam without hydrocarbon transfer.
Farrington and Quinn (1973a) suggested that the patterns observed in
bivalves may reflect a preserved state of the original suspended matter while
diagenetic changes in the sediments could result in the observed higher
boiling hydrocarbon assemblage. However, except for certain biogenic com-
pounds there is no evidence that the chromatographic character of the
particulate hydrocarbons differs from the distribution in the sediments
(Schultz, 1974; this study). Farrington and Quinn (1973a) also mentioned
that a selective uptake process by the clams could account for the observed
hydrocarbon patterns. Teal and Farrington (1977) noted that filter feeding
pelecypods from chronically polluted waters exhibit hydrocarbon distribu-
tions appreciably lower in GC retention time than those in the surface
sediments. However, Uca pugnax, the fiddler crab, a deposit feeder, seemed
to directly reflect hydrocarbon distributions in the sediments.
CYCLOALKENE COMPONENTS
Two of the three main compounds identified in this study, HC 348 and
HC 344 (X and Y of Fig. 2) are identical to those examined by Farrington
and Tripp (1975), and Farrington et al. (1977a) in other estuarine and
continental shelf sediments. The major component HC 344 (C25HA4)> probably
having a bicyclic core with two aliphatic side chains, is not found in the
water column. It is probably produced in the sediments via biochemical
synthesis or through diagenetic alteration of a sedimented biogenic compound
or suite of compounds'. Despite its olefinic structure, it is relatively
31
-------
stable, being found at least to a depth of 40 cm within the sediment (see
Fig. 6). Its concentration strongly covaries with organic carbon (Fig. 4).
Diagenetic alteration of biosynthesized material (e.g. e-carotene, 049^ or
squalene, 030^0) via chain cleavage and cyclization could possibly result
in a series of "25 cycloalkene components due to abiogenic processes. De novo
synthesis of these compounds by marine microorganisms or their formation
via secondary metabolism might also account for the covariance of organic
carbon and the cycloalkenes. However, biosynthetic pathways leading to
formation of these compounds are not clear, and chemical precursors which
yield these cycloalkenes as secondary metabolites are unknown. Whatever
their source, these compounds are key components in the sediments and
theoretically could be used as indicators of sedimentary processes once more
information on their structure and kinetics of their degradation are known.
Other cycloalkenes are found in the bivalve Arctica islandica (Fig. 2a)
and the major biogenic component HC 342 (0^42) is not present in the
sediment and is indeed quite labile as indicated by its rapid disappearance
from unrefrigerated samples. It is obviously protected from degradation
within the organism. Whether HC 342 is synthesized by Arctica or by micro-
organisms in its gut, or whether its taken up as an early labile precursor
of HC 344 and HC 348 and thereby preserved is not known.
An explanation of the sediment/Arctica cycloalkene relationship that
can not be excluded is that this bivalve, and perhaps other species in this
region, are controlling the cycloalkene content of the sediment. HC 344,
348 and 342 (among other cycloalkenes) may be produced in the clams via
biosynthetic pathways. The fecal material from these bivalves may contribute
significantly to the low organic carbon content of the surface sediments,
and may introduce cycloalkenes of biosynthetic origin. The more labile
structures are soon degraded in the sediments leaving HC 344 and HC 348.
SEDIMENT CORES
The hydrocarbon profiles in Core 17 (Fig. 6) are strikingly similar
to those of Farrington et al. (1977a) and Wakeham and Carpenter (1976).
The decreasing concentration of unresolved components with increasing depth
is not due to postdepositional alteration. The relative constancy of the
C25 through Cai n-alkanes with depth (Table 4) illustrates that degradation
and mobilization of hydrocarbons within the sediments has not been very
significant, a conclusion in agreement with that of Farrington et al. (1977a).
These components (n-alkanes) would be more readily degraded (Ahearn and
Meyers, 1973) and/or solubilized by interstitial organic matter (Boehm and
Quinn, 1973) than would components comprising most of the UCM. However,
the UCM decreases markedly with depth, not the n-alkanes.
A rapid appearance of an anthropogenic assemblage comprising the UCM
in our undated core 17 (Fig. 6) and in the dated cores of Farrington et al.
(1977a) in Buzzards Bay and Wakeham and Carpenter (1976), in Lake
Washington, probably coincides with the increased usage of petroleum which
occurred at the onset of the industrial revolution in North America at the
32
-------
end of the nineteenth century. This marker is seen within the 18-28 cm
section of our Station 17 core. Likewise, PCBs first appear in the 8-18 cm
section of the same core. Manufacturing of PCBs began 45 years ago (Hubbard,
1964). Their presence in this 8-18 cm section may reflect both normal
deposition plus mixing from the surface section by bioturbation, but absence
of PCBs below this section rules out large scale mixing. Based on the in-
faunal assemblage in R. I. Sound (Saila et al., 1972; Pratt, personal
communication) bioturbation is unlikely to proceed to a depth greater than
10 cm. Therefore the hydrocarbon profiles of Fig. 6 and Table 4 are not a
result of large scale mixing throughout the 38 cm section.
The exact depth in the core at which the hydrocarbon and PCB markers
occur is not known due to the rather broad sampling interval that we chose
within the core. Furthermore, bioturbation would tend to blur these chemical
markers. The depth of both markers (d) is a function of the sedimentation
rate(s) and the depth to which mixing occurs (m):
d = ts + m,
where t is the historical time elapsed since the introduction of the markers.
Two simultaneous equations, one for hydrocarbons and one for PCBs can be
solved, giving a sedimentation rate(s) and a bioturbation depth (m).
By utilizing two suitable markers we essentially eliminate the bioturbation
blurring effect as both markers would have been influenced equally when
within the mixing zone. Then we can compute a sedimentation rate which
actually accounts for deposition between the two markers. Allowing for the
sampling interval uncertainty (18-28 cm for the hydrocarbons and 8-18 cm
for the PCBs) and for variations in t, (75-100 years for hydrocarbon
introduction, and 30-45 years for PCBs), sixteen solutions are obtained for
s and m which indicate the possible ranges for these parameters. Rejecting
negative m solutions and sedimentation rates of zero, the possible deposi-
tion rate (s) ranges from 1.4-3.3 mm/year and the biological mixing depth
from 1.4-13.8 cm at this station in R. I. Sound.
A complete analysis of the recent pollen content of this same core
(C. Bernabo, personal communication) indicated that the entire core contained
greater than 10% herb pollen, indicative of a post agricultural revolution
assemblage. Thus the entire core is younger than ^1700 which yields a
minimum sedimentation rate for the 38 cm core of 1.4 mm/yr, in good agreement
with the range calculated above from chemical markers.
Such a sedimentation rate, obtained by using two well spaced chemical
markers, which behave similarly in the sediments (PCB and hydrocarbons),
agrees favorably to the sedimentation rate of 2.9 mm/yr established by 210Pb
geochronology on a Buzzards Bay core, some 40 km to the east of our sampling
site (Farrington et al., 1977b).
The UCM does not disappear completely with age (depth) as natural
pyrolytic sources (e.g. forest fires) (Youngblood and Blumer, 1976)
probably contributed aromatic and naphthenic hydrocarbon compounds to the
atmosphere and hence the sediments via fallout and runoff, long before the
increased usage of petroleum began to dominate the sedimentary hydrocarbon
33
-------
assemblage. We suggest that the recent sedimentary record in estuarine
and coastal areas, on continental shelves and in freshwater bodies around
and within industrialized regions, contains an accurate marker of increased
UCM concentrations. This marker and those of individual compounds which
may be chromatographically resolved using capillary columns (e.g. see
Dastillung and Albrecht, 1975), could serve as useful chemical horizons in
sediments where natural petroleum seepages have not occurred. Used in
conjunction with another anthropogenic marker having a different depositional
history (e.g. PCBs), these horizons could then be used to obtain a sedimenta-
tion rate of that sediment deposited subsequent to the input of anthropogenic
hydrocarbons.
34
-------
SECTION V
REFERENCES
Ahearn, D. G. and S. P. Meyers. 1973. The microbial degradation of oil
pollutants. Publication No. LSU-SG-73-01 Louisiana State University,
322 pp.
Boehm, P. D. and J. G. Quinn. 1973. Solubilization of hydrocarbons by the
dissolved organic matter in seawater. Geochimica et Cosmochimica
Acta 37:2459-2477.
Boehm, P. D. and J. G. Quinn. 1974. The solubility behaviour of No. 2
fuel oil in seawater. Marine Pollution Bulletin 5:101-104.
Boehm, P. D. and J. G. Quinn. 1976. The effect of dissolved organic matter
in seawater on the uptake of mixed individual hydrocarbons and Number 2
fuel oil by a marine filter-feeding bivalve Mercenaria mercenaria.
Estuarine and Coastal Marine Science 4:93-105.
Collins, B. P. 1976. Suspended material transport: Narragansett Bay Area,
Rhode Island. Estuarine and Coastal Marine Science 4:33-44.
Dastillung, M. and P. Albrecht. 1975. Molecular test for oil pollution
in surface sediments. Marine Pollution Bulletin 7:13-15.
Farrington, J. W. 1971. Benthic lipids of Narragansett Bay - Fatty acids
and hydrocarbons. Ph.D. Thesis, University of Rhode Island, 141 pp.
Farrington, J. W. and G. Medeiros. 1975. Evaluation of some methods of
analysis for petroleum hydrocarbons in marine organisms, Proceedings
of the Joint EPA API USCG Conference on the Prevention and Control
of Oil Spills, 115-121.
Farrington, J. W. and P. A. Meyers. 1975. Hydrocarbons in the marine
environment. Chapter 5 in Environmental Chemistry Vol. 1, 109-137
G. Eglinton (ed.), Specialists Periodical Report, The Chemical
Society, London.
Farrington, J. W. and B. W. Tripp. 1975. A comparison of analysis methods
for hydrocarbons in surface sediments. ACS Symposium Series No. 18,
Marine Chemistry in the Coastal Environment, T. M. Church (ed.),
p. 267-284.
35
-------
Farrington, J. W. and J. G. Quinn. 1973a. Petroleum hydrocarbons in Narra-
gansett Bay, I. Survey of sediments and clams (Mercenaria mercenaria).
Estuarine and Coastal Marine Science 1:71-79.
Farrington, J. W. and J. G. Quinn. 1973b. Petroleum hydrocarbons and fatty
acids in wastewater effluents. Journal Water Pollution Control
Federation 45:704-712.
Farrington, J. W., N. M. Frew, P. M. Gschwend and B. W. Tripp. 1977a.
Hydrocarbons in cores of Northwestern Atlantic coastal and continental
margin sediments. Estuarine and Coastal Marine Science (in press).
Farrington, J. W., S. M. Heinrichs and R. Anderson. I977b. Fatty acids and
Pb-210 geochronology of a sediment core from Buzzards Bay, Massachusetts.
Geochimica et Cosmochimica Acta 41:289-296.
Gearing, P., J. N. Gearing, T. F. Lytle and J. S. Lytle. 1976. Hydrocarbons
in 60 northeast Gulf of Mexico shelf sediments: a preliminary survey.
Geochimica et Cosmochimica Acta 40:1005-1017.
Gordon, R. B. 1974. Dispersion of dredge spoil dumped in near-shore
waters. Estuarine and Coastal Marine Science 2:349-358.
Hubbard, H. L. 1964. Chlorinated biphenyl and related compounds. In:
R. E. Kirk and D. F. Otherm (eds.), Encyclopedia of Chemical
Technology 5:289, second edition.
Lee, R. F., J. C. Nevenzel, G. A. Paffenhoffer, A. A. Benson, S. Patton and
T. E. Kavanagh. 1970. A unique hexaene hydrocarbon from a diatom
(Skeletonema costatum). Biochimica et Biophysica Acta 202:386-388.
Lee, R. F., R. Sauerheber and A. A. Benson. 1972. Petroleum hydrocarbons:
uptake and discharge by the marine mussel Mytilus edulis. Science
177:344-346.
McMaster, R. L. 1960. Sediments of Narragansett Bay system and Rhode
Island Sound, Rhode Island. Journal of Sedimentary Petrology 30:249-
274.
Meyers, P. A. and J. G. Quinn. 1973. Association of hydrocarbons and
mineral particles in saline solution. Nature 244:23-24.
Quinn, J. G. and T. L. Wade. 1974. Hydrocarbon analyses of IDOE inter-
calibration samples of cod liver oil and tuna meal. Marine Memorandum
Series Number 33, University of Rhode Island, 8 pp.
Rohrback, B. G. and W. E. Reed. 1975. Evaluation of extraction techniques
for hydrocarbons in marine sediments, Technical Report, Publication
No. 1537, Institute of Geophysics & Planetary Physics, University of
California at Los Angeles, 23 pp.
36
-------
Saila, S. B., S. D. Pratt and T. T. Polgar. 1972. Dredge spoil disposal
in Rhode Island Sound. Marine Technical Report Number 2, University
of Rhode Island, 48 pp.
Schultz, D. M. 1974. Source formation and composition of suspended lipoidal
material in Narragansett Bay, Rhode Island, Ph.D. Thesis, University
of Rhode Island, 205 pp.
Shonting, D. H. 1969. Rhode Island Sound square kilometer study, 1967:
flow patterns and kinetic energy distribution. Journal of Geophysical
Research 74:3386-3395.
Stegeman, J. J. and J. M. Teal. 1973. Accumulation, release and retention
of petroleum hydrocarbons by the oyster Crassostrea virginica.
Marine Biology 22:37-44.
Teal, J. M. and J. W. Farrington. 1977. A comparison of hydrocarbons in
animals and their benthic habitats. Bulletin Conseil Pour Texplora-
tion de la mer, (in press).
Thompson, S. and G. Eglinton. 1976. The presence of pollutant hydrocarbons
in estuarine epipelic diatom populations. Estuarine and Coastal
Marine Science 4:417-425.
Thompson, S. and G. Eglinton. 1977. The fractionation of a recent sediment
for organic geochemical analysis. Geochimica et Cosmochimica Acta
(submitted).
Van Vleet, E. S. and J. G. Quinn. 1977. Input and fate of petroleum hydro-
carbons entering the Providence River and Upper Narragansett Bay from
wastewater effluents. Environmental Science and Technology (submitted),
Wakeham, S. G. and R. Carpenter. 1976. Aliphatic hydrocarbons in sediments
of Lake Washington. Limnology and Oceanography 21:711-732.
Youngblood, W. W. and M. Blumer. 1976. Polycyclic aromatic hydrocarbons
in the environment: homologous series in soils and recent marine
sediments. Geochimica et Cosmochimica Acta 39:1303-1314.
Zafiriou, 0. C. 1973. Petroleum hydrocarbons in Narragansett Bay
II. Chemical and Isotopic Analysis. Estuarine and Coastal Marine
Science 1:81-87.
37
-------
TECHNICAL IU:i'O!-.T DATA
(I'lrasc tcail lmJntclilrli>ii:l
._EPAr 600/3- 77-092
------- |