Hydrocarbons in intertidal Sediments and Mussels from Prince William Sound, Alaska, 1977-1980: Characterization and Probable Sources

NOAA Technical Memorandum NMFS-AFSC-9

U.S. DEPARTMENT OF COMMERCE

National Oceanic and Atmospheric Administration

National Marine Fisheries Service

Alaska Fisheries Science Center

January 1993

Hydrocarbons in Intertidal Sediments

and Mussels from Prince William

Sound, Alaska, 1977-1980:

Characterization and Probable Sources

John F. Karinen, Malin M. Babcock, Donald W. Brown,

William D. MacLeod, Jr., L. Scott Ramos, and Jeffrey W. Short

NOAA Technical Memorandum NMFS

The National Marine Fisheries Service's Alaska Fisheries Science Center

uses the NOAA Technical Memorandum series to issue informal scientific and

technical publications when complete formal review and editorial processing

are not appropriate or feasible. Documents within this series reflect sound

professional work and may be referenced in the formal scientific and technical

literature.

The NMFS-AFSC Technical Memorandum series of the Alaska Fisheries

Science Center continues the NMFS-F/NWC series established in 1970 by the

Northwest Fisheries Center. The new NMFS-NWFSC series will be used by

the Northwest Fisheries Science Center.

This document should be cited as follows:

Karinen, J. F., M. M. Babcock, D. W. Brown, W. D. Macleod, Jr.,

L. S. Ramos, and J. W. Short. 1993. (revised December 1994).

Hydrocarbons in intertidal sediments and mussels from Prince William

Sound, Alaska, 1977-1980: Characterization and probable sources.

U.S. Dep. Commer., NOAA Tech. Memo. NMFS-AFSC-9, 70 p.

Reference in this document to trade names does not imply endorsement by

the National Marine Fisheries Service, NOAA.

NOAA Technical Memorandum NMFS-AFSC-9

Hydrocarbons in intertidal Sediments

and Mussels from Prince William Sound,

Alaska, 1977-1980: Characterization

and Probable Sources

John F. Karinen

, Malin M. Babcock

, Donald W. Brown

William

D. MacLeod, Jr.

, L. Scott Ramos

, and Jeffrey W. Short

Alaska Fisheries Science Center

Auke Bay Laboratory

11305 Glacier Highway

Juneau, Alaska 99801-8626

Northwest Fisheries Science Center

Environmental Conservation Division

2725 Montlake Boulevard East

Seattle, WA 98112-2097

U.S.

DEPARTMENT

OF COMMERCE

Barbara Hackman Franklin, Secretary

National Oceanic and Atmospheric Administration

John A. Knauss, Administrator

National Marine Fisheries Service

William W. Fox, Jr., Assistant Administrator for Fisheries

January 1993

This

document is available to the public through:

National Technical Information Service

U.S. Department of Commerce

5285 Port Royal Road

Springfield, VA 22161

Notice to Users of this Document

This document is being made available in .PDF format for the convenience of users; however, the

accuracy and correctness of the document can only be certified as was presented in the original hard

copy format.

iii

We collected and analyzed samples of sediments and mussels

(Mytilus trossulus) for alkane and aromatic hydrocarbons from

eight sampling stations adjacent to the oil tanker vessel

transportation corridor through Prince William Sound, Alaska,

during the period from 1977 to 1980,

to determine baselines prior

to the start of oil tanker movement through the Sound. We‘

evaluated interannual variability of these analytes using a two-

factor analysis of variance of logarithm-transformed hydrocarbon

concentrations determined in duplicate samples collected in June

1977 and in June 1978 at six of the stations.

Intra-annual

variability was evaluated using analyses of duplicate samples

collected in May,

June, and August 1978 at seven of the stations.

In addition, total organic carbon and grain size distribution was

determined in the sediment samples, the lipid content was

determined in the mussel samples, and the surface seawater

temperature and salinity was determined for each sampling

station.

The hydrocarbon analyses indicated chronic, low-level

hydrocarbon contamination that probably originates from small

fuel spills,

ballast water discharges, and fuel-combustion

exhaust emissions of occasional vessel activity adjacent to three

of the sampling stations: Constantine Harbor, Rocky Bay, and

Mineral Flats, in decreasing order of contamination,

respectively.

Sediments at these three stations were

contaminated by aromatic hydrocarbons found at concentrations

that were generally less than 10 ng/g dry sediment weight, but

above detectable limits (< 1.0 ng/g).

In contrast, the remaining

five stations showed no indication of petroleum hydrocarbon

contamination, primarily because detected aromatic hydrocarbons

were present only sporadically and at concentrations that were

generally near detection limits.

Both perylene,

which was found

at concentrations well above detection limits at all stations

outside Port Valdez, and phenanthrene, which was also found

sporadically at all sampling stations may have natural sources.

Concentrations of aromatic hydrocarbons were frequently too low

at most of the sampling stations to allow evaluation of intra-

and interannual variability.

Concentrations of individual n-alkanes varied substantially

in sediments and in mussels.

The most abundant n-alkanes in

sediments included normal alkanes with an odd number of carbon

atoms and a molecular weight greater than tetradecane (C-14).

Concentrations of these n-alkanes were generally in the range of

10 to 100 ng/g dry sediment weight and exceeded 1,000 ng/g at

Constantine Harbor.

The most abundant n-alkanes in mussels

included decane (C-10) through heptadecane (C-17), and pristane,

at concentrations generally ranging from 10 to over 1,000 ng/g

dry tissue weight.

Sources of alkanes in sediments included terrigenous plant

waxes,

marine plankton,

and possibly marine macrophytic algae at

all the stations; petroleum-derived alkanes were also found at

Constantine Harbor.

Terrigenous plant waxes in sediments were

indicated by high abundances of odd-numbered carbon n-alkanes of

molecular weight greater than nonadecane (C-19) compared with

even-numbered carbon n-alkanes in these sediments, and by slight

but significant intra-annual variability of these odd-numbered

carbon alkanes in sediments, which probably arose from seasonal

deposition of senescent leaves.

Marine planktonic and algal

sources of pristane and normal alkanes were indicated by the

presence of these alkanes in sediments and in mussels, and by the

relatively high abundances of pristane; pentadecane (C-15), and

heptadecane (C-17) in sediments and in mussels.

The concentrations of pristane, pentadecane (C-15), and

heptadecane (C-17) varied significantly in sediments, in mussels,

or in both, intra-annually or interannually.

Pristane

variability in sediments and in mussels was significantly

correlated and was probably due to variability of populations of

calanoid copepods in Prince William Sound.

Neither pentadecane

variability nor heptadecane variability were correlated in

sediments and mussels,

suggesting multiple biological sources of

these alkanes.

These results indicate that, except in areas affected by

localized vessel traffic,

intertidal sediments and mussels in

Prince William Sound were remarkably free of petroleum-

contaminant hydrocarbons during the period of this study.

The

hydrocarbons found in sediments and mussels unaffected by vessel

traffic can be adequately explained by known, natural sources.

As a result,

sediments and mussels contaminated by crude oil from

the Exxon Valdez oil spill should be particularly apparent due to

the general absence of other confounding sources of petroleum

hydrocarbons.

CONTENTS

Introduction

.......................

Materials and Methods

..................

Sampling Stations

.....

.............

Sample Collection

...................

Sediment -

Physical Measurements.

...........

Mussel tissue -

Dry-Weight and Lipid Determination.

Chemical Analysis

...................

Extraction Procedure.

.................

Fractionation Into Hydrocarbon Classes.

........

Gas Chromatography.

..................

Confirmation of Aromatic Hydrocarbon Analyte Identities

Gas Chromatography/Mass Spectrometry

...

. . . .

Data Analysis

.................

. . . .

Results

.....................

Intra-annual variation - Sediments.

......

Interannual variation - Sediments

.......

Intra-annual variation - Mussels.

.......

Interannual variation - Mussels

........

Correlation of Alkanes in

Sediments and Mussels

Discussion . . .

. .

. . . .

. . . . . . . . .

. .

Acknowlegments .

. .

. . . .

. . . . . . . . .

. .

Citations . . .

. .

. . . . . .

. . . . . . .

Appendix . . . .

. .

. . . . . .

. . . . . . .

. . . .

INTRODUCTION

The oil spill that resulted from the March 1989 grounding of

the oil tanker vessel Exxon Valdez provides a unique opportunity

for the study of marine oil pollution effects because the spilled

crude oil polluted a large geographic area that was previously

considered pristine.

Large-marine oil spills more commonly occur

along well established tanker routes that are already measurably

polluted by oil,

where the effects of a particular oil spill are

confounded with effects of prior spills or of chronic oil

pollution sources.

In contrast, the Exxon Valdez oil spill is

the only large spill along a recently established major oil

tanker route,

so the fate and effects of the spilled oil should

be more clearly discernable.

The only sources of confounding hydrocarbons in the areas of

Prince William Sound, Alaska,

impacted by the spill are naturally

occurring hydrocarbons and anthropogenic hydrocarbons from

occasional boating activity in the Sound or due to long-range

atmospheric transport. Naturally occurring hydrocarbons may

include:

alkane hydrocarbons produced from terrigenous and marine

biological sources such as plant waxes (Kolattukudy 1976;

Eglinton et al.

1962) and phytoplankton (Blumer et al. 1971;

Clark and Blumer 19,67);

polynuclear aromatic hydrocarbons (PAHs)

generated by forest fires (Hites 1981; Farrington et al. 1977;

Youngblood and Blumer 1975) and subsequently precipitated into

the drainage basin of the Sound; perylene produced from

unidentified but probably natural sources (Venkatesan 1988, and

citations therein);

and aliphatic and aromatic hydrocarbons

originating from natural oil seeps.

Anthropogenic hydrocarbons

in the Sound can result from chronic small spills of fossil fuels

associated with boating and shipping directly into seawater, and

from fossil fuel combustion products subsequently precipitated

that originate from boating and shipping, or from distant

industrial centers (Lake et al.

1979; Lunde and Bjorseth 1977).

We sampled intertidal sediments and tissues of mussels

Mytilus trossulus in Prince William Sound during a 4-year, period

beginning in 1977 to establish the levels and variability of

hydrocarbons in these matrixes prior to any large-scale pollution

'events.

The likelihood of such large-scale pollution events

within the Sound increased substantially in, July 1977 with the

large volume of crude oil transported after the opening of the

trans-Alaska oil pipeline,

which connects the Prudhoe Bay oil

field with a tanker terminal at Valdez, Alaska.

Our specific objectives were to determine the levels,

intra-

annual variability,

and interannual variability of selected

alkane hydrocarbons and PAHs in intertidal sediments and in

M. trossulus tissues at a network of sampling stations over the

I-year sampling period,

and if possible to identify the likely

sources of hydrocarbons found. We chose sediments and mussels to

facilitate comparison with the large body of data on hydrocarbons

in these two matrixes in the literature.

We present our results

now to facilitate evaluation of hydrocarbon analyses of these

matrixes in Prince William Sound conducted in the aftermath of

the Exxon Valdez oil spill.

MATERIALS AND METHODS

Sampling Stations

Prince William Sound is one of North America's largest tidal

estuary systems occupying about 6,500 km on the southern coast

of Alaska.

There is a mean precipitation of 4.6 m at some

locations (Arctic Environmental Information and Data Center 1977)

which is sufficient enough to depress salinities within the Sound

and influence adjacent oceanic currents.

The prevailing ocean

current enters the Sound through Hinchinbrook Entrance in the

south-central part of the Sound and exits mainly through Montague

Strait in the southwest (Galt et al. 1991).

Intertidal sediment and mussel samples were collected from

four to eight stations in Prince William Sound, Alaska, during

the period May 1977 to August 1980.

The latitude, longitude, and

collection dates for each station are listed in Table 1, where

the station numbers correspond with those in Figure 1, and where

entries for sediment temperature and seawater salinity in Table 1

indicate samples were collected.

The eight sampling stations

bracket the oil tanker route inside Prince William Sound to the

Alaska pipeline terminal in Port Valdez (Fig. 1).

The eight

stations were not all sampled prior to the June 1978 sampling.

The sampling stations are subject to dissimilar

environmental influences.

Two stations,

Dayville Flats and

Mineral Flats,

are located inside Port Valdez near the oil tanker

terminal and the city of Valdez, Alaska, respectively (Fig. 1).

The surface waters of Port Valdez receive sufficient fresh water

from direct precipitation and glacial melt-water to depress

salinities substantially, particularly in late summer (Table 1,).

In addition,

the glacial melt-water bears a high sediment load

that is deposited in the Port.

Although remote from human

activity, the station at Siwash Bay is similarly affected by

fresh and glacier melt-water.

The other stations are not

directly exposed to glacial melt-water.

The stations at Constantine Harbor and at Rocky Bay are

exposed to the effects of occasional vessel traffic.

Constantine

Harbor was the site of an early Russian settlement, and more

recently has been used as a port of refuge for vessels seeking

Table l.--Sample station locations, sampling dates, salinities and temperatures.

Salinities (%) and temperatures (°C) are given for each sediment and

mussel sample collection date at each sampling station.

Missing values for salinity and temperature indicate no samples collected. M=

measurement not taken.

The sample station numbers refer to those

in Figure 1.

1977

1978

1979

1980

*Mussels only collected

Figure l.--

Maps

of the study area and of the sampling stations

within the study area.

The sample station numbers on

these maps are those listed in Table

shelter from the frequent storms in the Gulf of Alaska.

Although

an excellent harbor,

it is very poorly flushed due to its shape.

Rocky Bay is fished commercially for several species, resulting

in occasionally dense vessel traffic.

The stations at Bligh Island, Naked Island, and Olsen Bay

are rarely exposed to human activities and are exposed to similar

oceanographic conditions that are typical of most of the Sound.

Sample Collection

Sediment collection transect lines (30 m) were located

parallel to the water line from the -0.75 m to +0.75 m tide

levels.

Sediment cores were collected using a rinsed,

hydrocarbon-free stainless steel cookie cutter. Sediment samples

were collected in triplicate at each site by compositing 10 cores

(diameter 3.2 cm x depth 1.25 cm) taken at random along the 30-m

transect for each sample.

Composite sediments were placed in

dichloromethane-rinsed glass jars and were frozen within 2-3

hours of collection.

Mussel collection transects were located in mussel bands

parallel to the water line,

usually just above the sediment

transects (-+l m tide level).

Mussel samples were collected in

triplicate by taking approximately 30 2-5 cm mussels (enough to

produce ≥ 10 g tissue) at random along the 30-m transect.

The

collected mussel samples were placed into dichloromethane-rinsed

glass jars and were frozen within 2-3 hours of collection.

Sediment

- Physical Measurements

Total organic carbon was determined by the method of Jackson

(1958).

Sediment grain size distribution was determined using

standard sieves and pipetting methods (Krumbein and Pettijohn

1938).

Sediment dry weight was determined gravimetrically by

measuring the weight lost after drying for 24 hours at 100°C.

Mussel Tissue -

Dry Weight and Lipid Determination

Mussel tissue dry weight was determined gravimetrically by

measuring the weight lost after drying for 24 hours at 120°C.

Percent lipid of the tissue was determined by the method of

Hanson and Olley (1963) and is reported on a dry weight basis.

Chemical Analysis

Sediment and mussel tissue samples were analyzed for normal

alkanes having 10 to 30 carbon atoms (C-10 to C-30), pristane,

phytane,

and the PAHs listed in Table 2.

Pristane and phytane

refer to 2,6,10,14-tetramethylpentadecane and 2,6,10,14-

tetramethylhexadecane, respectively.

Concentrations are reported

as ng/g dry weight.

Not more than two samples were analyzed from

any triplicate of samples collected.

Sediment and mussel tissue samples were processed for

analysis in batches consisting of 8-10 samples, a reagent blank,

and a spiked reagent blank.

The alkane dodecylcyclohexane (DCH)

was added to the samples and to the spiked blank, but not to the

reagent blank, to verify recovery of alkane analytes.

An aliquot

of the hydrocarbons listed in Table 2 was added to the spiked

blank to estimate losses during analysis. The samples, reagent

blank, and spiked blank were processed identically.

Extraction Procedure

Hydrocarbons in sediment samples were extracted using a

ball-mill tumbler extraction procedure described fully in Brown

et al. (1980).

An aliquot of DCH was added to a 100 g wet

sediment sample,

which was subsequently dewatered by swirling the

sample with two successive aliquots of 50 ml methanol per aliquot

in a 1 L bottle.

The two methanol aliquots were then decanted

and combined in a 600 ml beaker, and 100 ml of 2:l

dichloromethane:

methanol was added to the bottle containing the

sediment sample.

The bottle was sealed with a Teflon-lined

screwcap and rolled on a ball-mill tumbler for 16 hours

(overnight).

The extract was decanted into the 600 ml beaker

containing the methanolic extracts, together with 5 ml

dichloromethane used to rinse the sample and bottle.

The

dichloromethane-methanol sediment extraction step was repeated

twice,

first for 6 hours, then for 16 hours, with 100 ml each

time.

The combined extracts were filtered through a coarse

fritted-glass filter into a 1 L separatory funnel and extracted

with 500 ml distilled water to remove methanol.

The aqueous

phase was separated and back-extracted with 20 ml

dichloromethane,

and the dichloromethane phases were combined and

filtered through a 19 mm id chromatography column containing

20 ml of activated silica gel covered with a 1 cm layer of sand.

The dichloromethane filtrate and rinses were concentrated and

exchanged into 2 ml of hexane under reflux.

Hydrocarbons in whole mussel tissue were extracted using a

procedure described fully by Brown et al.. (1979).

Whole tissues

from samples of mussels were mechanically homogenized for at

least 30 seconds.

A 10 g aliquot of the homogenate was combined

with 6 ml of 4 N sodium hydroxide in a 40 ml centrifuge tube

capped with a Teflon-lined screw cap, shaken for 1 minute, and

digested at 30°C for 18 hours.

A 15 ml aliquot of diethyl ether

Table 2.--Identities

and abbreviations of alkane and aromatic hydrocarbon

analytes determined in sediment and in mussel tissue samples

this study.

was added to the centrifuge tube after it had cooled to room

temperature,

and the re-capped tubes were shaken for 1 minute and

centrifuged for 10 minutes.

The ether phase was transferred to a

30 ml bottle, and the sample was re-extracted with 10 ml of

diethyl ether.

The combined ether extracts were dried with 0.5 g

sodium sulfate, transferred to a concentrator tube, then

concentrated and exchanged into 2 ml of hexane under reflux.

Fractionation Into Hydrocarbon Classes

Alkane and aromatic hydrocarbons were separated by column

liquid chromatography using silica gel.

The silica gel (Davison

grade 923) was 100-200 mesh activated at 150°C for 24 hours.

The

chromatography column was packed with 7 g silica gel covered with

a 1 cm layer of sand in dichloromethane, and then washed with

40 ml of petroleum ether.

The sample in 2 ml hexane was loaded

onto the top of the column and eluted with 15 ml petroleum ether,

then 3 ml 1:4 dichloromethane:petroleum ether, then 25 ml 2:3

dichloromethane:petroleum ether.

The first 18 ml of eluate

collected contained the alkane analytes, and the next 25 ml of

eluate contained the aromatic analytes.

Elemental sulfur was

removed from these eluates with copper metal that was activated

through rinsing with concentrated hydrochloric acid, then

methanol.

The eluates were prepared for analysis by gas

chromatography (GC) by concentration to 0.7 ml under reflux,

addition of 1.0 ml internal standard (4 ng/ul hexamethylbenzene

(HMB) in hexane),

and re-concentration to 0.7 - 1.0 ml.

Gas Chromatography

The alkane and aromatic hydrocarbon extract fractions were

analyzed using GC with a hydrogen flame ionization detector (FID)

to measure each analyte.

The GC instrument conditions were the

same for the alkanes and aromatics.

The GC analysis was

performed on a Hewlett-Packard model 5840A gas chromatograph

equipped with a 30 m long by 0.25 mm id glass capillary column

coated with a dimethylsiloxane polymer (SE-30).

A 2 ul aliquot

of sample was injected into the injection port operated in the

splitless mode at 280°C.

The split valve was opened after 0.3

minutes.

The carrier gas was helium, the initial column

temperature was 40°C for 5 minutes, but then was raised at a rate

of 4°/minute to 270°C.

The detector was operated at 300°C and

used nitrogen make-up gas.

Analyte peak area response was

determined using an electronic integrator.

Hydrocarbon analyte, concentrations in the extracts injected

into the GC were determined by comparison of the ratio of analyte

FID response and internal standard (HMB) FID response for a

sample with the same ratio for a hydrocarbon standard.

The

concentration of hydrocarbon analyte in the original sample was

calculated as the ratio of the amount of analyte present in the

original sample and the equivalent dry weight

the sample.

The

equivalent dry weight of a sample is the product of the sample

wet weight and the ratio of wet and dry weights of that sample,

which were determined using a 10-20 g subsample for sediments and

a 3 g subsample for mussel tissue homogenates.

Detection limits were estimated on the basis of the minimum

measurable instrument response,

response factors for standards,

and sample dry weight.

These estimates of detection limit

concentrations are indicated by the preceding symbol "<" for each

analyte not detected in each sample in Tables A-l through A-4 in

the Appendix.

Confirmation of Aromatic Hydrocarbon Analyte Identities

Gas

Chromatography/Mass Spectrometry

The identity of aromatic hydrocarbons detected and measured

by GC/FID were confirmed by GC/MS analysis in each sample matrix

at each sample station.

Data Analysis

Intra-annual trends of selected alkane analytes were

evaluated using a two-factor analysis of variance (ANOVA), where

factors included seven sample stations and the three sample

collection periods listed in Table 1 for 1978.

Data from 1978

were evaluated because more of the sample stations were sampled

and analyzed in duplicate than in other years, resulting in a

balanced ANOVA with maximum power.

The sample station not

included for the ANOVA was the Mineral Flats station in Port

Valdez,

which was not sampled in May of 1978.

The analytes

selected for evaluation included: pristane, phytane, and the

normal alkanes from C-12 through C-31 in sediments; pristane and

the normal alkanes from C-12 through C-17 in mussels were also

included.

The remaining alkane analytes and all the aromatic

analytes in both sediments and mussels were only sporadically

above detection limits,

precluding evaluation by ANOVA.

Interannual trends of selected alkane analytes were

evaluated using a two-factor ANOVA, where factors included six

sample stations and two sample collection periods: June 1977 and

June 1978.

Data from June of these years were evaluated because

more of the sample stations were sampled and analyzed in

duplicate than the other months,

resulting in a balanced ANOVA

with maximum power.

The sample stations not included for the

ANOVA were the Olsen Bay station and the Mineral Flats station in

Port Valdez,

which were not sampled in June of 1977.

The

analytes selected for evaluation included: pristane, phytane, C-

15, and the normal alkanes from C-17 through C-31 in sediments;

pristane and the normal alkanes from C-12 through C-17 in mussels

were also included.

The remaining alkane analytes and all the

aromatic analytes in both sediments and mussels were only

sporadically above detection limits,

precluding evaluation by

ANOVA.

The association of an alkane analyte in sediments with

mussel tissues was evaluated using the Pearson product-moment

correlation coefficient.

Data from all the sampling stations and

from all the sampling dates were included for the correlation

coefficient calculations.

Correlation coefficients were

determined for C-15,

C-17, and pristane because these were the

only three hydrocarbon analytes consistently present in both

sediments and mussels.

The original alkane concentrations, c, were transformed as

ln(c + 1),

and the correlation coefficients and the ANOVAs were

calculated using the transformed values.

RESULTS

The concentrations of alkane and of aromatic hydrocarbon

analytes found in sediments and mussels are listed in Tables A-l

through A-4 of the Appendix.

The percent total organic carbon

(TOC) is also listed in Table A-l for each sediment sample where

determined,

and the percent lipid is also listed in Table A-2 for

each mussel sample where determined.

The sediment grain size

distribution is listed in Table A-5 for each sediment sample

where determined.

Percent TOC, percent lipid, and sediment grain

size are presented in these tables to facilitate comparison with

future work.

Following is a summary and statistical

evaluation of the hydrocarbon results listed in these tables.

Alkanes

Sediments

Mean concentrations of individual alkane analytes were

generally on the order of 1 to 10 ng/g dry sediment, ranging up

to about 100 ng/g,

at all sites except Constantine Harbor

(Fig. 2).

Mean alkane concentrations were consistently elevated

by a

factor of about 5 or more at Constantine Harbor compared

with the other sites (Fig. 2).

These mean concentrations are

averages of sample duplicates and of all sampling periods.

Sediments at Constantine Harbor contained numerous

unidentified alkane hydrocarbons compared with sediments at the

other sampling stations.

Representative chromatograms of alkanes

in sediment samples at Constantine Harbor, Dayville Flats, and

Olsen Bay in June 1978 are presented in Figure 3.

Note the

presence of numerous unidentified peaks in the Constantine Harbor

chromatogram that are generally absent in the Dayville Flats and

Olsen Bay chromatograms of Figure 3. Constantine Harbor is the

only sampling station where these unidentified peaks are

typically present.

Concentrations of normal alkanes having an odd number of

carbon atoms generally predominate over adjacent even-numbered

alkanes at all eight sites,

with exceptions in the C-10 to C-14

range of the normal alkanes (Fig. 2). The sum of the mean

concentrations of odd C-15 thru C-31 account for 71 - 89%,

depending on the site,

of the sum of the mean concentrations of

all the alkanes,

with maxima at C-15 to C-19 and at C-27 to C-31.

Pristane and phytane are relatively minor constituents of

alkanes in sediments.

Mean pristane concentrations were highest

at Naked Island,

Constantine Harbor, and Rocky Bay, with mean

concentrations of 43.6 ng/g, 42.6 ng/g, and 24.4 ng/g,

respectively.

However,

even at these three stations, the mean

pristane concentration was less than 8% of the sum of the mean

concentrations of all the alkanes at the respective stations.

Phytane concentrations were consistently highest at Constantine

Harbor at a mean concentration of 9.07 ng/g; mean phytane

concentrations at the other stations range from 1.04 to

3.43 ng/g.

Intra-annual Variation - Sediments

In 1978,

the sampling period was at least a highly

significant factor (P < 0.01) for six alkane analytes, based on

the intra-annual two-factor ANOVAs (Table 3).

The ANOVA

interaction term was not significant for two of these alkanes,

pristane and C-20,

indicating uniform changes of these two

alkanes across the sampled stations.

These two alkanes were

consistently lower in August 1978 than in May 1978 at each

sampled station (Table 4).

The ANOVA interaction term was

significant for the other four alkanes, C-13, C-14, C-16, and

C-18, (Table 3),

indicating station-specific significant changes

of these alkanes.

These four alkanes increased from May 1978 to

August 1978 at Bligh Island and at Siwash Bay, but decreased

during this same period at the other sampled stations (Table 4).

Sampling period is a significant factor (P < 0.05) for eight

other alkane analytes at stations sampled in 1978, and the ANOVA

interaction term is either not significant or is marginally

significant (P

-0.05) for these analytes (Table 3). These

alkanes include phytane, C-19, C-21, C-22, C-24, C-27, C-29, and

Table 3 .--Summary analysis of variance table for intra-annual variation of

logarithmically transformed alkane concentrations in sediments

during 1978 at seven sampling stations.

The excluded station is

Mineral Flats because it was not sampled in May 1978.

The two-

factor ANOVA includes 1978 sampling period and sampling stations as

the two factore,

and is performed for each alkane analyte listed

below independently.

Also listed for each alkane analyte below are

the ANOVA F-ratios of the sampling period mean square (df = 2) and

the error mean square (df = 21), the sampling station mean square

(df =

6) and the error mean square,

and the interaction mean square

of sampling periods and sampling stations (df = 12) and the error

mean square.

The error mean square is itself listed for each alkane

analyte to allow the reader to reconstruct the full ANOVA table for

each alkane.

These ANOVA's are fully balanced, with two

observations for each combination of sampling period and sampling

station.

The symbols *, **,

and *** associated with the F-ratios

indicate significance (P < 0.05),

high significance (P < 0.01), and

very high significance (P <0.00l), respectively. Alkane

concentrations of Table A-l were transformed as ln (c + 1), where c

is the concentration listed in Table A-l, prior to the ANOVA

calculation.

Alkane

Month Station

2,21

6&21

Interaction

12,21

MSe

Table 4.

--Proportional change from May 1978 to August 1978 of

alkane analytes in sediments at seven sampling stations.

The excluded station is Mineral Flats because it was

not sampled in May 1978.

Proportional change is

calculated as (A - B)/B for each alkane analyte and

for each included sampling station where

and B are

mean alkane analyte concentrations of August 1978 and

May 1978,

respectively, using the data in Table A-l.

Bligh

Const.

Dayville Naked

Olsen

Rocky

Siwash

Alkane

Island Harbor

Flats

Island

Bay

Figure 3.--

Representative chromatograms of alkane hydrocarbons

in sediments at Constantine Harbor, Dayville

Flats,

and Olsen Bay, June 1978.

C-31.

These alkanes are consistently lower in August, 1978 than

in May,

1978 at all the stations sampled except Bligh Island and

Constantine Harbor,

and they are frequently lower at Bligh

Island.

Note that sampling station is consistently a very highly

significant factor (P < 0.001) for each alkane listed in Tables 3

and 4.

Interannual Variation - Sediments

Very highly significant station-specific changes of C-17 and

of pristane were observed when June 1977 and June 1978

concentrations were compared,

based on the interannual two-factor

ANOVAs (Table 5).

Concentrations of C-17 were consistently

higher at each station in June 1978 than in June 1977, by

multiples ranging up to about 18 (Table 6).

Concentrations of

pristane were higher at most stations by factors ranging up to

about 11; pristane concentrations were unchanged at Rocky Bay and

decreased by 34.3

% at Constantine Harbor (Table 6).

Changes of

the other alkane analytes listed in Tables 5 and 6 were either

insignificant or were marginally significant (P -0.05).

Note that sampling station is consistently a very highly

significant factor (P < 0.001) for each alkane listed in Tables 5

and 6.

Mussels

Mean concentrations of individual alkane analytes are

generally on the order of 10 to 100 ng/g dry mussel tissue,

although some alkanes are not detected at some stations, and

others are detected at concentrations substantially higher than

100 ng/g (Fig. 4).

In general,

alkane analytes ranging from C-10

through pristane are the most abundant in mussels at all the

sampling stations.

Pristane,

C-15, and C-17 together account for

62 -

87% of total alkanes, depending on the station.

Phytane is

detected sporadically at concentrations less than 100 ng/g except

at Olsen Bay,

where it is detected once only at 520 ng/g (May

1979,

see Table A-2).

Intra-annual Variation - Mussels

In 1978 the sampling period is at least a highly significant

factor (P < 0.01) for C-12, C-14, C-15,

and pristane based on the

intra-annual two-factor ANOVAs (Table 7).

By far the most

significant change was for pristane,

which consistently declined

at all stations from concentrations ranging up to several

thousand ng/g in May 1978 to concentrations less than 100 ng/g in

Table 5.--

Summary analysis of variance table for interannuel variation of logarithmically transformed alkane

concentrations in sediments from June 1977 to June 1978 at six sampling stations.

The excluded

stations are Mineral Flats and Olsen Bay because these were not sampled in June 1977. The included

stations were all sampled in June 1977 and in June 1978.

The two-factor ANOVA includes the year of

the June sampling period and sampling stations as the two factors, and is performed for each alkane

analyte listed below independently.

Also Listed for each alkane analyte below are the ANOVA F-ratios

of the sampling period mean square (df =

1) and the error mean square (df = 121, the sampling station

mean square (df = 5) and the error mean square, and the interaction mean square of sampling periods

and sampling stations (df =

5) and the error mean square.

The error mean square is itself listed for

each alkane analyte to allow the reader to reconstruct the full ANOVA table for each alkane.

These

ANOVAs are fully balanced, with two observations of alkane concentration for each combination of

sampling period and sampling station.

The symbols *,**, and *** associated with the F-ratios

indicate significance (P < 0.051, high significance (P < 0.011, and very high significance

(P < 0.001), respectively.

Alkane concentrations of Table A-l were transformed as ln (c + 1), where

c is the concentration listed in Table A-l, prior to ANOVA calculation. The alkane analytes C-10

through C-14, and C-16, are not included in this table because the concentrations of these analytes

are frequently below detection limits, and thus transform

zero, which compromises the homoscedastic

assumptions of the ANOVA.

Alkane

Year

1,12

Station

5,12

Interaction

5,12

MSe

Only four sampling stations were included in this ANOVA; the additionally excluded station is Constantine

Harbor, because the analytical results were not reported for this alkane analyte at this station in

June

1977

results (see Table A-l, Constantine Harbor).

The degrees of freedom for the sampling station mean square, the

interaction mean square, and the error mean square are 4, 4, and 10, respectively, for this alkane in this table.

Table 6.

--Proportional change from June 1977 to June 1978 of alkanes in

sediments at six sampling stations.

The excluded stations are

Mineral Flats and Olsen Bay because they were not sampled in

June 1977.

Proportional change is calculated as (A - B)/B for

each alkane analyte and for each included sampling station

where A and B are mean alkane analyte concentrations of June

1977 and June 1978, respectively,

using the data in Table A-l.

The alkane analytes included in this table are those listed in

Table 5.

Alkane

Bligh

Island

Const.

Dayville

Harbor

Flats

Naked

Rocky

Island

Bay

Siwash

Bay

= not available;

June 1977 values below detection limits.

August 1978 (Tables 8 and A-2).

This seasonal decline of pristane

concentrations were repeated in 1979 at all stations (Table A-2).

The

ANOVA interaction term was not significant for C-12 and for C-14 (Table

7)l

indicating uniform declines of these two alkanes across the sampling

stations (despite the slight rise of the mean concentration of C-13 at

Siwash Bay; see Table 8).

The ANOVA interaction term was significant for

C-15 (Table 7) due to the decline of the C-15 mean concentration at Olsen

Bay and the increase of the C-15 mean concentration at the other

stations.

Note that sampling station is a significant factor only for pristane

and C-15 in Table 7.

Interannual Variation - Mussels

Very highly significant changes of C-15 and pristane were observed

when June 1977 and June 1978 concentrations were compared, based on the

interannual two-factor ANOVAs (Table 9). Pristane consistently increased

at each sampling station in June 1978 compared with June 1977 by

multiples ranging up to ten-fold depending on sampling station (Table

10).

C-15 consistently decreased at each sampling station by factors

ranging up to 4 depending on sampling station (Table 10).

Note that sampling station is a significant factor only for pristane

and C-17 in Table 9.

Correlation of Alkanes in Sediments and Mussels

Only pristane was significantly correlated in sediments and mussels,

with a coefficient of correlation r = 0.344 (n = 90, P < 0.001).

Correlation coefficients for C-15 and for C-17 were 0.028 (n = 97) and -

0.053 (n = 95), respectively,

and were clearly not significant (

> 0.5).

II. Aromatics

Sediments

At three of the sampling stations,

a few aromatics were frequently

present at concentrations well above detection limits.

These stations

included Constantine Harbor, Mineral Flats, and Rocky Bay (see Table A-3

for these stations).

Perylene and phenanthrene were the most abundant

aromatic analytes at these three stations,

together accounting for 31 -

54%

of total aromatic analytes (Fig. 5).

Sediments at Constantine Harbor contain numerous unidentified

aromatic hydrocarbons compared with sediments at the other sampling

stations.

Representative chromatograms of aromatics in sediment samples

at Constantine Harbor, Dayville Flats,

and Olsen Bay in June 1978 are

Table 7.

--Summary analysis of variance table for intra-annual

variation of logarithmically transformed alkane

concentrations in mussels during 1978 at seven stations.

The excluded station is Mineral Flats because it was not

sampled in May 1978.

The two-factor ANOVA includes 1978

sampling period and sampling stations as the two

factors,

and is performed for each alkane analyte listed

below independently.

Also listed for each alkane

analyte below are the ANOVA F-ratios of the sampling

period mean square (df = 2) and the error mean square

(df = 21),

the sampling station mean square (df = 6) and

the error mean square,

and the interaction mean square

of sampling periods and sampling stations (df = 12) and

the error mean square.

The error mean square is itself

listed for each alkane analyte to allow the reader to

reconstruct the full ANOVA table for each alkane. These

ANOVAs are fully balanced, with two observations for

each combination of sampling period and sampling'

station.

The symbols *, **,

and *** associated with the

F-ratios indicate significance (P < 0.05), high

significance (P < 0.01),

and very high significance

(P < 0.001),

respectively.

Alkane concentrations of

Table A-l were transformed as ln(c + 1), where c is the

concentration listed in Table A-l, prior to the ANOVA

calculation.

The alkane analytes C-18 through C-31 and

phytane are not included in this table because the

concentrations of these analytes are frequently below

detection limits,

and thus transform to zero, which

compromises the homoscedastic assumptions of the ANOVA.

Alkane

Month

Station

Interaction

1,21

6,21

12,21

MSe

Table 8.

--Proportional change from May 1978 to August 1978 of

alkanes in mussels at seven sampling stations.

The

excluded station is Mineral Flats, because it was not

sampled in May 1978.

Proportional change is calculated

as (A -

B)/B for each included alkane analyte, and for

each included sampling station, where A and B are mean

alkane analyte concentrations of August 1978 and May

1978,

respectively,

using the data in Table A-l. The

alkane analytes included in this table are those listed

in Table 7.

Alkane

Bligh

Island

Const.

Harbor

Dayville

Flats

Naked

Island

Olsen

Bay

Rocky

Bay

Siwash

Bay

(+):

mean May 1978 =

0; mean August 1978 >0

Table 9.

--Summary analysis of variance table for interannual

variation of logarithmically transformed alkane

concentrations in mussels from June 1977 to June 1978 at

six sampling stations.

The excluded stations are Mineral

Flats and Olsen Bay because these were not sampled in May

1977.

The included stations were all sampled in June 1977

and in June 1978. The two-factor ANOVA includes the year'

of the June sampling period and sampling stations as the

two factors,

and is performed for each alkane analyte

listed below independently.

Also listed for each alkane

analyte below are the ANOVA F-ratios of the sampling period

mean square (df =

1) and the error mean square (df = 12),

the sampling station mean square (df = 5) and the error

mean square,

and the interaction mean square of sampling

periods and sampling stations (df = 5) and the error mean

square.

The error mean square is itself listed for each

alkane analyte to allow the reader to reconstruct the full

ANOVA table for each alkane.

These ANOVAs are fully

balanced, with two observations of alkane concentration for

each combination of sampling period and sampling station.

The symbols *, **,

and *** associated with the F-ratios

indicate significance (P < 0.05), high significance

(P < 0.01),

and very high significance (P < 0.001),

respectively.

Alkane concentrations of Table A-l were

transformed as ln(c + 1),

where c is the concentration

listed in Table A-1, prior to ANOVA calculation. The

alkane analytes C-18 through C-31, and phytane, are not

included in this table because the concentrations of these

analytes are frequently below detection limits, and thus

transform to zero,

which compromises the homoscedastic

assumptions of the ANOVA.

Alkane

Year

1,12

Station

Interaction

5,12

MSe

Table 10.

--Proportional change from June 1977 to June 1978 of

alkanes in mussels at six sampling stations.

The

excluded stations are Mineral Flats and Olsen Bay

because they were not sampled in June 1977.

Proportional change is calculated as (A - B)/B for each

alkane analyte and for each included sampling station

where A and B are mean alkane analyte concentrations of

August 1978 and May 1978, respectively, using the data

in Table A-l.

The alkane analytes included in this

table are those listed in Table 9.

Alkane

Bligh

Island

Const.

Harbor

Dayville

Naked

Flats

Island

Rocky

Bay

Siwash

Bay

(+) :

mean 1977 =

0; mean 1978 >0

presented in Figure 6.

Note the presence of numerous unidentified

peaks in the Constantine Harbor chromatogram that are generally

absent in the Dayville Flats and Olsen Bay chromatograms of Figure 6.

These unidentified peaks are most numerous and largest at Constantine

Harbor,

followed by Rocky Bay and Mineral Flats in decreasing order.

Note also the prominence of identified aromatic analytes which lack

alkyl substituents in the Constantine Harbor chromatogram of Figure-

The variety of identified aromatic analytes usually detected

decreases as follows:

Constantine Harbor > Rocky Bay > Mineral

Flats. Most aromatic analytes are usually detected at elevated

concentrations at Constantine Harbor and Rocky Bay, the exceptions

being the mononuclear aromatics, benzothiophene, anthracene, and

benzo[a]pyrene (and 2,3,5-trimethylnaphthalene at Rocky Bay), which

are sporadically detected.

Aromatic analytes usually detected at

Mineral Flats include the naphthalenes (except 2,3,5-

trimethylnaphthalene), fluorene, dibenzothiophene, phenanthrene,

fluoranthene, and pyrene (Fig. 5).

At five of the sampling stations,

all aromatic analytes except

perylene were near or below detection limits (Table A-3).

These

stations included Bligh Island, Dayville Flats, Naked Island, Olsen

Bay,

and Siwash Bay.

Aromatic analytes most frequently detected at

these stations included phenanthrene, 2-methylnaphthalene, perylene,

naphthalene,

and 1-methylnaphthalene, in order of decreasing

frequency of detection among these stations (Table A-3).

Although

not detected at Dayville Flats, perylene accounted for 43 - 79% of

the aromatic analytes at the remaining four of these stations (Fig.

5) l

In contrast, benzothiophene, benzanthracene, and benzo[e]pyrene

were never detected at these five stations; 2,3,5-

trimethylnaphthalene, dibenzothiophene, and chrysene were each

detected once only at concentrations that were less than 0.7 ng/g.

The low and sporadic concentrations of aromatics preclude

evaluation of intra-

or interannual trends.

Mussels

Aromatic analytes were rarely detected in mussels.

Aromatic

analytes were detected only 18 times,

which included only five

analytes (Table A-4).

Naphthalene was detected nine times-and 1-

methylnaphthalene was detected six times.

Iso-propylbenzene, n-

propylbenzene,

and benzothiophene were each detected once only.

Naphthalene or 1-methylnaphthalene were detected most often at

Dayville Flats (four times), Rocky Bay (four times), Bligh Island

(three times),

and Constantine Harbor (three times).

Aromatic analytes in mussels were detected most often in 1977,

when detection limits are substantially lower than for succeeding

years.

Of the 18 instances of aromatic analyte detection in mussels,

14 occur in 1977 (Table A-4).

Detection limits of aromatic analytes

for the 1977 samples were usually less than 3 ng/g, whereas these

detection limits were usually greater than 4 rig/g for samples from

succeeding years (Table A-4).

DISCUSSION

Trace level hydrocarbon contamination was evident at the

Constantine Harbor, Mineral Flats, and Rocky Bay sampling stations.

Contamination was indicated by the presence and diversity of aromatic

hydrocarbons repeatedly found in sediments at these stations, which

were generally absent at the other sampling stations.

Sediment contamination levels were highest at Constantine

Harbor,

and the contamination was probably the result of a

combination of sources.

The highest concentrations and greatest

diversity of aromatic analytes and of other unidentified aromatic

hydrocarbons was found in the sediments of this station.

The

presence of elevated concentrations of PAHs containing more than

three rings and lacking alkyl substituents in these sediments (but

not including perylene, see below),

indicates a pyrolytic source for

these PAHs.

Lower concentrations of alkyl-substituted 3- and 4-ring

PAHs compared with corresponding unsubstituted homologues may be

inferred from the chromatograms of these samples (Fig. 6), which show

the unsubstituted homologues as the most prominent peaks.

Prominence

of these higher molecular weight PAHs is indicative of a pyrolytic

source and atmospheric transport to these sediments (Hites 1981,

Lunde and Bjorseth 1977,

Youngblood and Blumer 1975), and this source

and transport mechanism was further supported by the general absence

of PAHs in mussel tissues at the Constantine Harbor station.

Waterborne transport of petroleum-derived hydrocarbons to

Constantine Harbor sediments was probably a second source of

contaminant hydrocarbons.

This source was indicated by the

consistently elevated levels of lower molecular weight normal-

alkanes, phytane,

unidentified branched alkanes, and aromatics found

in Constantine Harbor sediments.

Phytane is associated with

petroleum and with ancient sediments (Blumer and Snyder 1965,

Oro et al.

1965) and is usually absent from modern unpolluted

sediments,

although it may be produced by biochemical processes or

found in modern sediments in special situations (Ikan et al. 1975,

Nissenbaum et al. 1972).

Concentrations of lower molecular weight

normal-

and branched-alkanes and aromatics are enriched in refined

petroleum products compared with crude oil.

The general absence of

these aromatic hydrocarbons in Constantine Harbor mussels suggests

that the source is not petroleum hydrocarbon seepage from a

terrestrial or sub-marine source because a nearly continuous seepage

should result in a nearly constant influx of petroleum hydrocarbon

contamination that would have been detected in these mussels.

The most likely source of all the contaminant hydrocarbons found

in Constantine Harbor sediments is from sporadic fuel spills and fuel

combustion, exhaust emissions of marine vessel traffic.

Constantine

Harbor is an excellent natural anchorage that is sometimes used as a

harbor of refuge by vessels of all sizes, including commercial

vessels,

seeking protection from violent storms that frequently occur

in the Gulf of Alaska.

The larger vessels usually keep their engines

running at idle while at anchor,

and they may also discharge ballast

water.

The contaminant hydrocarbons found in Constantine Harbor

probably arose from these emissions and discharges over the past

several decades.

Contaminant hydrocarbons found at Rocky Bay and at Mineral Flats

probably arose from sources similar to those at Constantine Harbor.

Rocky Bay is the site of occasional commercial fishing, and Mineral

Flats is adjacent to both the boat harbor at the city of Valdez and

the oil tanker loading terminal in Port Valdez.

There was no evidence of detectable contaminant hydrocarbons at

the other five sampling stations of this study.

The perylene and

phenanthrene detected at these stations probably arose from natural

sources;

both have been detected in unpolluted sediments worldwide

and in sediment core samples that pre-date industrial activity

(Venkatesan 1988, Hites et al. 1980, Farrington et al. 1977, Hites et

al. 1977).

Perylene has been reported as the predominant PAH in some

Alaskan sediments,

and has been significantly correlated with the sum

of C-27 and C-29 in these sediments,

which was taken as evidence of a

terrigenous source (Venkatesan and Kaplan 1982).

Our results are

consistent with these observations;

the highest concentrations of

perylene,

C-27, and C-29 occurred at Constantine Harbor.

The other

aromatic hydrocarbons found in sediments and in mussels at these

stations were present only sporadically, and at concentrations near

detection limits.

Similarly, phytane and many of the normal alkanes

having an even number of carbon atoms were found at these stations at

concentrations near detection limits,

further corroborating the

general absence of petroleum hydrocarbon contamination.

Of the normal alkanes of molecular weight higher than C-19 in

sediments, the high abundance of normal alkanes having an odd number

of carbon atoms compared with those having an even number of carbon

atoms at all the sampling stations of this study indicates

terrestrial plants as primary sources of these alkanes (Kolattukudy

1976, Eglinton and Hamilton 1967, Eglinton et al. 1962).

These

alkanes may be transported by senescent leaves of beach grasses,

upland shrubs,

and trees such as alder (Alnus rubra), to intertidal

sediments where they may be pulverized and incorporated into the

organic carbon compartment of the surface sediments.

These plants

are common in Prince William Sound and may form dense stands adjacent

to or in the upper intertidal zone.

Primarily terrestrial sources of

these hydrocarbons was further supported by the low and sporadic

concentrations of these alkanes found in the mussels of these

sampling stations.

The presence and abundance of normal alkanes of molecular weight

lower than C-20 and pristane in sediments and in mussel tissues

suggests sources of these hydrocarbons that are primarily marine.

These normal alkanes are common constituents of marine bacteria

(Oro

et al. 1967),

blue-green algae (Winters et al. 1969), planktonic and

macrophytic algae (Clark and Blumer 1967, Blumer et al. 1971), and

pristane is biochemically synthesized from phytol by several calanoid

copepod species (Avigan and Blumer 1968, Blumer et al. 1964).

Incorporation of these alkanes into intertidal sediments could arise

from deposition of carcass fragments of these organisms after death,

while incorporation into mussels could arise from ingestion of

phytoplankton as food,

and possibly from ingestion of detrital

material derived from zooplankton.

The high concentrations, variability,

and the different patterns

of variability of pristane,

C-15, and C-17 in sediments and in

mussels indicates distinct biological sources of these alkanes.

The

very high May concentrations of pristane in mussels, followed by

consistent and dramatically lower concentrations in June and still

lower in August,

suggests accumulation of pristane by ingestion of

detrital material derived from carcasses of calanoid copepods.

Blumer et al. (1964) found concentrations of pristane in

Calanus finmarchicus, C. glacialis,

and in C. hyperboreus approaching

1% dry weight in the northwestern Atlantic ocean, and in analogous

Pacific ocean species such as C. marshalae, Neocalanus plumchrus, and

N. cristatus.

Calanus marshalae and N. plumchrus are abundant in

Prince William Sound (Cooney 1987).

The very high May concentrations of pristane in mussels were

highest at Rocky Bay and Naked Island, which are the two stations

most exposed to the prevailing ocean current that enters the Sound at

Hinchinbrook Entrance and exits through Montague Strait.

This

current carries inorganic nutrients from deep-water upwelling off the

continental shelf into the Sound, which supports dense planktonic

blooms in the spring,

followed by blooms of zooplankton.

It is

remarkable, however,

that pristane concentrations in mussels were

quite low in 1977;

this probably reflects the population dynamics of

the calanoid copepods.

The significant correlation of pristane in mussels and in

sediments among the sampling stations and periods further suggests

calanoid copepods as the source of pristane in the sediments.

The high intra-

and interannual variability of C-17 in

sediments, but not in mussels,

suggests multiple sources of this

alkane in these matrixes.

C-17 is the predominant normal alkane of

many inter-

and sub-tidal red algae (Rhodophyceae) (Clark and Blumer

1967),

which on senescence may be incorporated into the organic

compartment of intertidal sediments but would not be available to

mussels.

The relatively high concentrations of C-17 in mussels,

together with less intra-annual variability may be due to the

ubiquity of this alkane in phytoplankton ingested by mussels (Blumer

et al. 1971).

The high intra-

and inter-annual variability of C-15 in mussels,

but not in sediments,

suggests multiple sources of this alkane in

these matrixes that are different than those of C-17.

Although C-15

varied greatly from year to year,

and generally increased from May to

August in mussels,

it was not clear why corresponding variation in

sediments is not evident.

In conclusion,

except in areas affected by localized vessel

traffic,

intertidal sediments and mussels in Prince William Sound

were remarkably free of petroleum-contaminant hydrocarbons during the

period of this study.

The hydrocarbons found in sediments and

mussels unaffected by vessel traffic can be adequately explained by

known, natural sources.

a result,

sediments and mussels

contaminated by crude oil from the Exxon Valdez oil spill should be

particularly apparent,

due to the general absence of other

confounding sources of petroleum hydrocarbons.

ACKNOWLEDGMENTS

The authors

thank D.L. Fisher, A.J. Friedman, D.D. Gennero, K.L.

Grams, K.E.

Kreps,

P.G. Prohaska, D.G. Burrows, K.A. Culler, J.F.

Morado,

P.P. Murphy, R.G. Jenkins, T.I. Scherman, and O. Maynes for

collecting and analyzing samples,

and L.A. Quintrell for assistance

in manuscript preparation.

CITATIONS

Arctic Environmental Information and Data Center.

1977.

Climatic atlas of the outer continental shelf waters and

coastal regions of Alaska. Volume 1: Gulf of Alaska.‘

439 p.

Avigan, J.,

and M. Blumer.

1968.

On the origin of pristane in

marine organisms. J. Lipid Res. 9:350-352.

Blumer, M.,

M. M. Mullin, and D. W. Thomas.

1964.

Pristane in

the marine environment.

Helgol. Wiss. Meeres. 10:187-201.

Blumer, M., R. R. L. Guillard, and T. Chase.

1971.

Hydrocarbons

of marine phytoplankton. Mar. Biol. 8:183-189.

Blumer, M., and W. D. Snyder.

1965.

Isoprenoid hydrocarbons in

recent sediments: presence of pristane and probable absence

of phytane. Science 150:1588-1589.

Brown, D. C., L. S. Ramos, M. Y. Uyeda, A. J. Friedman, and W. D.

MacLeod.

1980.

Ambient temperature extraction of

hydrocarbons from marine sediment - comparison with boiling-

solvent extractions. In L. Petrakis and F. T. Weiss

(editors),

Petroleum in the marine environment, p. 313-326.

ACS Symp. Ser. 185, American Chemical Society, Washington,

D. C.

Brown, D. C.,

A. J. Friedman, D. G. Burrows, G. R. Snyder, B. G.

Patten,

W. E. Ames, L. S. Ramos, P. G. Prohaska, D. D.

Gennero,

D. D. Dungan, M. Y. Uyeda, and W. D. MacLeod.

1979.

Investigation of petroleum in the marine environs of

the Strait

Juan de Fuca and Northern Puget Sound, 107 p.

NOAA Tech. Memo. ERL-MESA, EPA-600-7-79-164.

Clark, R. C., and M. Blumer. 1967.

Distribution of N-Paraffins

in marine organisms and sediment.

Limnol. & Oceanogr.

12:79-87.

Cooney, R. T.

1987.

Zooplankton.

In D. W. Hood and S. T.

Zimmerman (editors), The Gulf of Alaska - Physical

environment and biological resources.

Minerals Management

Service Pub. No. OCS Study, MMS 86-0095.

U. S. Government

Printing Office, Washington, D. C.

Eglinton, G.,

and R. J. Hamilton.

1967.

Leaf epicuticular

waxes.

Science 156:1322-1335.

Eglinton, G.,

A. G. Gonzalez, R. J. Hamilton, and R. A. Raphael.

1962.

Hydrocarbon constituents of the wax coating of plant

leaves:

a taxonomic survey.

Phytochemistry 1:89-102.

Farrington, J. W., N. M. Frew, P. M. Gschwend, and B. W. Tripp.

1977.

Hydrocarbons in cores of Northwestern Atlantic

coastal and continental margin sediments.

Estuarine Coast.

Mar. Sci.

5:793-808.

Galt, J.A.,

W.J. Lehr, and D.L. Payton.

1991.

Fate and

transport of the Exxon Valdez oil spill.

Environ. Sci.

Technol.

25:202-209.

Hanson,

S. W. F.,

and J. Olley.

1963.

Application of the Bligh

and Dyer method of lipid extraction to tissue homogenates.

Biochem. J. 89:101-102.

Hites, R. A.

1981.

Sources and fates of atmospheric polycyclic

aromatic hydrocarbons. In E. S. Macias and P. K. Hopke

(editors),

Atmospheric aerosol, source/air quality

relationships, p. 187-196.

ACS Symp. Ser. 167, American

Chemical Society, Washington, D. C.

Hites, R. A.,

R. E. LaFlamme,

and J. W. Farrington.

1977.

Sedimentary polycyclic aromatic hydrocarbons: the historical

record.

Science 198:829-831.

Hites, R. A.,

R. E. LaFlamme, and J. G. Windsor.

1980.

Polycyclic aromatic hydrocarbons in marine/aquatic

sediments:

their ubiquity.

In L. Petrakis and F. T. Weiss

(editors), p. 289-311.

Petroleum in the Marine

Environment.

ACS Symp. Ser.

185, American Chemical Society,

Washington, D. C.

Ikan, R.,

M. J. Baedecker, and I. R. Kaplan.

1975.

Thermal

alteration experiments on organic matter in recent marine

sediment - II. Isoprenoids.

Geochem.

Cosmochim. Acta

39:187-194.

Jackson, M. L.

1958.

Soil chemical analysis.

Prentice-Hall,

Inc.

Englewood Cliffs, New Jersey.

498 p.

Kolattukudy, P. E.

1976.

Chemistry and biochemistry of natural

waxes.

Elsevier, New York.

459 p.

Krumbein, W. C., and F. J. Pettijohn.

1938.

Manual of

sedimentary petrography.

Appleton Century Crofts, Inc.

New

York, New York.

549 p.

Lake, J. L., C. Norwood, C. Dimock, and R. Bowen.

1979.

Origins

of polycyclic aromatic hydrocarbons in estuarine sediments.

Geochim.

Cosmochim. Acta 43:1847-1854.

Lunde, G., and A. Bjorseth.

1977.

Polycyclic aromatic

hydrocarbons in long-range transported aerosols.

Nature

268:518-519.

Nissenbaum, A.,

M. J. Baedecker, and I. R. Kaplan.

1972.

Organic geochemistry of Dead Sea sediments.

Geochim.

Cosmochim. Acta 36:709-727.

Oro, J.,

T. G. Tornabene, D. W. Nooner, and E. Gelpi.

1967.

Aliphatic hydrocarbons and fatty acids of some marine and

freshwater microorganisms.

J. Bacterial. 93:1811-1818.

Oro, J.,

D. W. Noonor, A. Zlatkis, S. A. Wikstrom, and E. S.

Barghoorn.

1965.

Hydrocarbons of biological origin in

sediments about two billion years old.

Science 148:77-79.

Venkatesan, M. I.

1988.

Occurrence and possible sources of

perylene in marine sediments - a review.

Mar.

Chem. 25:1-

27.

Venkatesan, M. I., and I. R. Kaplan.

1982.

Distribution and

transport of hydrocarbons in surface sediments of the

Alaskan outer continental shelf.

Geochim.

Cosmochim. Acta

46:2135-2149.

Winters, K., P. L. Parker,

and C. Van Baalen.

1969.

Hydrocarbons of blue green algae: geochemical significance.

Science 163:467-468.

Youngblood, W. W., and M. Blumer.

1975.

Polycyclic aromatic

hydrocarbons in the environment: homologous series in soils

and recent marine sediments.

Geochim.

Cosmochim. Acta

39:1303-1314.

APPENDIX

Table A-l

.--Concentrations of alkane analytes found in sediment samples at each sampling period and sampling station of this study. Concentration

units are ng alkane/g dry sediment weight.

Concentrations less than detection Limits are indicated by the symbol <, followed by the

detection limit estimate of that alkane analyte in that sample.

Identities of abbreviated alkanes are given in Table 2 of the text.

Also listed is total organic carbon (TOC) for each sample where determined, presented as g carbon/g dry sediment weight times 100%.

MAY

JUNE JUNE OCT OCT MAY MAY

JUNE

AUG AUG MAY

AUG AUG

1977 1977 1977 1977 1977 1977 1978

1978

1978 1978

1978 1979 1979

1980

MEAN

Bligh Island

Table A-l

.--Continued.

MAY MAY

JUNE

MAY MAY

JUNE JUNE

AUG

AUG MAY

AUG AUG

1977 1977 1977 1977 1978 1978

1978 1978

1978 1978 1979

1979 1980

MEAN

Constantine Harbor

Table A-l.--Continued.

MAY MAY JUNE

JUNE

OCT

MAY MAY

JUNE

AUG AUG MAY AUG AUG

1977 1977 1977 1977

1977 1977 1978 1978 1978 1978

1978 1978 1979 1979 1980 MEAN

Dayville Flats

Table A-l

.--Continued.

MAY MAY

AUG AUG MAY

AUG AUG

1978

1978 1978 1978 1979

1979 1980

MEAN

Mineral Flats

Table A-l .--Continued.

MAY

JUNE

JUNE OCT

OCT

MAY

MAY JUNE

JUNE

AUG AUG

MAY

AUG

1977 1977

1977

1977 1977 1977

1978

1978 1978

1978

1978 1978

1979 1979

1980

MEAN

Naked Island

Table A-l

--Continued.

MAY

JUNE

AUG

AUG MAY AUG

AUG

1978 1978 1978 1978

1978

1979 1979 1980

MEAN

Olsen Bay

Table A-l

.--Continued.

JUNE

MAY MAY JUNE

JUNE

AUG AUG MAY AUG

1977 1977 1978 1978 1978

1978

1978 1978

1979

1980 MEAN

Rocky Bay

Table A-l

.--Continued.

JUNE

OCT

OCT MAY

MAY

JUNE JUNE

AUG AUG MAY AUG AUG

1977

1977 1977

1977 1978

1978 1978

1978

1979

1979 1980

WEAN

Siwash Bay

Table A-2:-Concentrations of alkane analytes found in mussel tissue samples at each sampling period and sampling station of this study.

Concentration units are ng alkane/g dry tissue weight.

Concentrations Less than detection limits are indicated by the symbol <,

followed by the detection Limit estimate of that alkane analyte in that sample.

Identities of alkanes abbreviated below are given in

Table 2 of the text.

Listed is percent lipid content for each sample where determined, presented as g lipid/g dry tissue weight

times 100%.

MAY MAY

JUNE

OCT

MAY MAY

JUNE

AUG

MAY AUG AUG

1977 1977 1977 1977 1977 1977 1978 1978 1978 1978 1978

1978

1979 1979 1980 MEAN

Bligh Island

Table A-2:-Continued.

MAY MAY

JUNE

MAY MAY

JUNE

JUNE AUG

AUG

MAY AUG

AUG

1977 1977 1977

1977 1978 1978

1978 1978 1978

1978

1979

1980

MEAN

Constantine Harbor

Table A-2.

--Continued.

MAY

JUNE

OCT

MAY HAY

JUNE JUNE

AUG AUG MAY AUG

AUG

1977 1977

1978 1978

1978

1978 1978

1979 1979

1980 MEAN

Dayville Flats

Table A-2. --Continued.

JUNE

JUNE AUG AUG

MAY

AUG AUG

1978 1978

1979 1979

1980

MEAN

Mineral Flats

Table A-2. --Continued.

MAY MAY JUNE

JUNE

AUG AUG

MAY AUG

AUG

1978 1978 1978 1978 1978

1978

1979 1979 1980 MEAN

Olsen Bay

Table A-2

.--Continued.

JUNE JUNE

MAY

JUNE

AUG

MAY

AUG AUG

1977

1978

1978 1978

1979

1980

MEAN

Rocky Bay

Table A-2.

--Continued.

JUNE JUNE

OCT

OCT MAY

MAY

JUNE

AUG AUG

MAY

AUG

1977 1977 1977 1977 1978

1978 1978

1978

1978 1979 1979 1980

MEAN

Siwash Bay

Table A-3.--

Concentrations of aromatic enalytes found in sediment samples at each sampling period and sampling station of this study. Concentration

units are ng alkane/g dry sediment weight.

Concentrations less than detection limits are indicated by the symbol <, followed by the

detection limit estimate of that aromatic analyte in that sample.

Identities of aromatics abbreviated below are given in Table 2 of the

text.

MAY MAY

JUNE

OCT OCT MAY MAY

JUNE

JUNE AUG AUG MAY

AUG AUG

1977 1977

1977

1977 1978

1978 1978

1978

1979

1979 1980

MEAN

Bligh Island

Table A-3.

--Continued.

MAY MAY JUNE JUNE MAY MAY JUNE

JUNE

AUG

MAY

AUG

1977

1977 1977

1977 1978

1978 1978

1978 1979

1979 1980

MEAN

Constantine Harbor

Table A-3.--Continued.

MAY MAY

JUNE

OCT

OCT MAY

MAY

JUNE

AUG

MAY

AUG

1977 1977

1977

1977 1977

1977

1978 1978

1978

1979 1979

1980

MEAN

Dayville Flats

Table A-3. --Continued.

JUNE

AUG

AUG MAY

AUG

1978 1978

1978 1978 1979

1979

1980

MEAN

Mineral Flats

Table A-3.--Continued.

MAY MAY

JUNE

AUG AUG

MAY

MAY JUNE

JUNE

AUG AUG HAY AUG

AUG

1977

1977 1977 1977 1977

1978

1978 1978

1978 1979 1979

1980

MEAN

Naked Island

Table A-).--Continued.

MAY

JUNE

AUG

MAY

AUG

1978

1978 1978 1978

1978

1978 1979

1979

1980 MEAN

Olsen Bay

Table A-3.

--Continued.

JUNE

AUG AUG

JUNE

AUG AUG

MAY

AUG

1977

1977 1977

1977 1978

1978

1979

1980 MEAN

Rocky Bay

Table A-3. --Continued.

JUNE

JUNE OCT OCT MAY MAY

JUNE

AUG

MAY AUG

AUG

1977 1977 1977

1977 1978 1978

1978

1978 1979 1979

1980

MEAN

Siwash Bay

Table A-4.--

Concentrations of aromatic analytes found in mussel samples at each sampling period and sampling station of this study. Concentration

units are ng alkane/g dry mussel tissue weight.

Concentrations less than detection limits are indicated by the symbol <, followed by the

detection limit estimate of that alkane aromatic in that sample.

Identities of aromatics abbreviated below are given in Table 2 of the

text.

MAY

JUNE

JUNE OCT

OCT

MAY MAY

JUNE

AUG

MAY

AUG AUG

1977

1977 1977 1977

1978 1978

1978

1979

1979 1980 MEAN

Bligh Island

Table A-4 .--Continued.

MAY MAY JUNE JUNE

MAY MAY

JUNE

AUG

AUG MAY

AUG

1977 1977 1977

1977 1978 1978

1978 1978 1978

1978 1979 1979

1980 MEAN

Constantine Harbor

Table A-4. --Continued.

MAY

JUNE

OCT OCT MAY

MAY

JUNE

JUNE AUG AUG MAY

AUG AUG

1977 1977 1977 1977 1977 1977 1978 1978 1978 1978

1978

1978 1979 1979

1980

MEAN

Dayvi11e Flats

Table A-4.--Continued.

JUNE

AUG

MAY AUG AUG

1978

1979 1979

1980

MEAN

Mineral Flats

Table A-4.

--Continued.

MAY

JUNE

JUNE OCT

OCT MAY MAY

JUNE

AUG AUG

MAY

AUG

1977

1977 1977

1977

1977 1977

1978

1978 1978

1978

1979 1979

1980

MEAN

Naked Island

Table A-4. --Continued.

MAY

JUNE

JUNE AUG

AUG MAY

AUG AUG

1978

1978 1978

1978 1979

1979

1980

MEAN

Olsen Bay

Table A-4

.--Continued.

JUNE JUNE MAY

MAY JUNE

JUNE

AUG

MAY

AUG

1977

1977 1978

1978 1978

1978

1979 1979

1980 MEAN

Rocky Bay

Table A-4.

--Continued.

JUNE JUNE

OCT

MAY

MAY JUNE JUNE

AUG

AUG MAY AUG

AUG

1977

1977 1977

1977 1978

1978 1978

1978 1979

1979 1980

MEAN

Siwash Bay

Table A-5.

Table of sediment grain-size distribution for sediment samples.

(PHI) Sediment grain size as fraction percent

Sampling

period

<-2

-2 to +0

+0 to +2 +2 to +4

+4 to +8

>+8

Mean Sand/Mud

Bligh Island

Constantine Harbor

Dayville Flats

Table A-5

.--Continued.

(PHI) Sediment grain size as fraction percent

Sampling

period

<-2

-2 to +0 +0 to +2 +2 to +4 +4 to +8

>+8 Mean Sand/Mud

Table A-5

.--Continued.

(PHI) Sediment grain size as fraction percent

Sampling

period <-2

-2 to +0

+0 to +2

+2 to +4 +4 to +8

>+8

Mean

Sand/Mud

RECENT TECHNICAL MEMORANDUMS

Copies of this and other NOAA Technical Memorandums are available from the

National Technical Information Service, 5285 Port Royal Road, Springfield, VA 22167

(web site: www.ntis.gov). Paper and microfiche copies vary in price.

AFSC-

8 WING, B. L. 1993. Winter oceanographic conditions in the eastern Gulf of Alaska, January-February

1986, 53 p. NTIS No. PB93-158335.

7 ARMISTEAD, C. E., and D. G. NICHOL. 1993. 1990 bottom trawl survey of the eastern Bering

Sea and continental shelf, 190 p. NTIS No. PB93-156677.

6 WOLOTIRA, R. J., JR., T. M. SAMPLE, S. F. NOEL, and C. R. ITEN. 1993. Geographic and bathymetric

distributions for many commercially important fishes and shellfishes off the West Coast of North America,

based on research survey and commercial catch data, 1912-84, 184 p. NTIS No. PB93-167682.

5 GUTTORMSEN, M., R. NARITA, J. GHARRETT, G. TROMBLE, and J. BERGER. 1992. Summary

of observer sampling of domestic groundfish fisheries in the northeast Pacific Ocean and Eastern Bering

Sea, 1990, 281 p. NTIS No. PB93-159085.

4 GUTTORMSEN, M., R. NARITA, and J. BERGER. 1992. Summary of U. S. observer sampling

of joint venture fisheries in the northeast Pacific Ocean and Eastern Bering Sea, 1990, 78 p. NTIS No.

PB93-127546.

3 JOHNSON, P. A., S. D. RICE, and M. M. BABCOCK (compilers). 1992. Impacts of oil pollution

and Prince William Sound studies: Bibliography of 1960-91 publications and reports, Auke Bay Laboratory,

98 p. NTIS No. PB93-114064.

2 KAJIMURA, H., and E. SINCLAIR. 1992. Fur seal investigations, 1990, 192 p. NTIS No. PB93-

109080.

1 MERRICK, R. L., D. G. CALKINS, and D. C. MCALLISTER. 1992. Aerial and ship-based surveys

of Steller sea lions (Eumetopias jubatus) in southeast Alaska, the Gulf of Alaska, and Aleutian Islands

during June and July 1991, 41 p. NTIS No. PB92-235928.

F/NWC-

216 ZENGER, H. H., JR., and M. F. SIGLER. 1992. Relative abundance of Gulf of Alaska sablefish

and other groundfish based on National Marine Fisheries Service longline surveys, 1988-90,

103 p. NTIS No. PB92-222843.

215 KINOSHITA, R. K., B. M. K. BROOKE, L. E. QUEIROLO, and J. M. TERRY. 1992. Economic status of

the groundfish fisheries off Alaska, 1990, 99 p. NTIS No. PB92-187699.

214 ANTONELIS, G. A. 1992. Northern fur seal research techniques manual, 47 p. NTIS No. PB92-191824.

213 BAKKALA, R. G., W. A. KARP, G. F. WALTERS, T. SASAKI, M. T. WILSON, T. M. SAMPLE,

A.M. SHIMADA, D. ADAMS, and C. E. ARMISTEAD. 1992. Distribution, abundance, and biological

characteristics of groundfish in the eastern Bering Sea based on results of U.S.-Japan bottom trawl and

hydroacoustic surveys during June-September 1988, 362 p. NTIS number pending.