The Science of the Total Environment 229 Ž1999.

65]71

Quantification of polycyclic aromatic hydrocarbons in

soil at McMurdo Station, Antarctica

David Mazzeraa,U , Terry Hayes b, Douglas Lowenthal a , Barbara Zielinskab

a
Atmospheric Sciences Center (ASC), Desert Research Institute, 2215 Raggio Parkway Reno, NV 89512-1095, USA
b
Energy and En¨ ironmental Engineering Center (EEEC), Desert Research Institute, 2215 Raggio Parkway Reno,
NV 89512-1095, USA

Received 1 September 1998; accepted 1 March 1999

Abstract

Polycyclic aromatic hydrocarbons ŽPAHs. are a group of compounds that have attracted much attention over the
past several years. The United States Environmental Protection Agency ŽUSEPA. has identified numerous PAHs as
known or probable human carcinogens. The quantity of PAHs in the environment has dramatically increased, with
the majority emitted from fossil fuel combustion sources. Surface soil samples were collected at McMurdo Station,
Antarctica Ž778519 S, 1668419 E., during peak summer activity and analyzed for PAHs. PAHs were detected at several
locations, with maximum concentrations for naphthalene, acenaphthene, acenaphthylene, and fluoranthene at 27 000,
17 800, 15 700, and 13 300 mgrkg, respectively. Results suggest anthropogenic activities may be contributing to
increased levels of PAHs present in McMurdo soils. Q 1999 Elsevier Science B.V. All rights reserved.

Keywords: PAHs in soils; McMurdo Station; Antarctica

1. Introduction tion Agency ŽUSEPA, 1985. has been the driving

Polycyclic or polynuclear aromatic hydrocar-
bons ŽPAHs. have been the focus of much atten-
tion over the past decade. The identification of
numerous unsubstituted PAHs as priority pollu-
tants by the United States Environmental Protec-
U mobile sources, power plants. and practices such
Corresponding author. Tel.: q1-775-677-3203; fax: q1-
775-677-3157; e-mail: davem@dri.edu
force for both development of new and improve-
ment of existing analytical methods for detection
of PAHs in air, soil, sediment, and water.
PAH compounds are introduced into the envi-
ronment by natural and anthropogenic Žman-
made. sources. Natural sources include forest
fires. Anthropogenic sources and activities that
generate PAHs include fossil fuel combustion Ži.e.
as agricultural burning. A study by Basu and

0048-9697r99r$ - see front matter Q 1999 Elsevier Science B.V. All rights reserved.
PII: S 0 0 4 8 - 9 6 9 7 Ž 9 9 . 0 0 0 6 5 - 0
66 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71
Saxena Ž1987. has shown a correlation between
increases in PAH concentrations in environmen-
tal media with sites where fossil fuels and refining
waste residuals were used or produced. On a
smaller scale, everyday activities such as wood-
burning stoves and fireplaces, meat grilling, and
the use of gasoline-powered equipment Žlawn
mowers. may contribute to total PAH contribu-
tions in urban areas ŽFreeman and Cattell, 1990..
Concentrations of PAHs in ‘ambient’ soils have
been measured since the early 1970s. Virtually all
surface soils have measurable PAH concentra-
tions due mainly to the transport of these com-
pounds in the atmosphere and subsequent dry
and wet deposition processes as well as scrubbing
of these organic molecules by forest canopies and
deposit in soils via litterfall ŽMatzner, 1984.. Jones
et al. Ž1989. have shown typical concentrations in
forest soils range from 5 to 100 m grkg Žparts per
billion, ppb. and result from both natural and
anthropogenic sources. Results from analysis of
soil samples collected and preserved from 1880 to
the present have shown a general increase in
PAH concentrations, with levels peaking during
the 1950sr1960s ŽMenzie et al., 1992.. As one
may expect, urban and metropolitan areas typi-
cally have higher concentrations of PAHs due to
an increase in point and non-point sources as well
as an increase in fossil fuel combustion and other
incineration processes.
Antarctica is considered to be one of the last
pristine environments on earth. However, even in
such remote environments there exists evidence
of human contamination. While humans have
been present in Antarctica for decades, the
growing numbers in recent years increase the
potential for contamination. During the austral
summer, McMurdo serves as a major base for
many of the outlying field camps and research
bases. McMurdo’s population exceeds 1000 dur-
ing peak activity.
The purpose of this study is to examine the
levels of PAHs found in Antarctic soils located in
the vicinity of McMurdo Station, the United States
base located on Ross Island, Antarctica Ž778519 S,
1668419 E.. Soil samples collected were analyzed
for a suite of PAHs in order to assess the impact
of human activity on the level of these com-
pounds present in such a remote location. These
observations will contribute to the growing
database of contamination impacting the environ-
ment on a global scale.
2. Materials and methods
2.1. Soil sampling
Collection of soil samples Ž n s 20. was per-
formed in January 1997, during peak seasonal
activity at McMurdo. Surface soil grab samples
Ž0]2 inches. were collected at 18 locations around
McMurdo Station ŽFig. 1.. Two additional sam-
ples were collected from the exposed soils at the
Lake Hoare Long Term Ecological Research
ŽLTER. basecamp, located approximately 90 km
northeast and downwind of McMurdo in Taylor
Valley, Antarctica. All soil samples were stored
frozen in certified clean amber glass containers
with minimal light exposure until the time of
extraction and analysis.
2.2. Extraction and analysis
The soil samples were Soxhlet-extracted sepa-
rately with dichloromethane ŽDCM. using a mod-
ified EPA Method 3540B for a minimum of 18 h
ŽUSEPA, 1994.. Approximately 30 g of the soil
samples were mixed with approximately 5 g of
oven-dried sodium sulfate and placed in an ex-
traction thimble. The extracts were concentrated
by rotary evaporation at 208C under gentle vac-
uum to 1 ml and filtered through a 0.20-m m
Anatop filter ŽWhatman International, Ltd.., rins-
ing the sample flask twice with 1 ml DCM. Ap-
proximately 50 m l of acetonitrile was added to the
sample and the DCM was evaporated under a
nitrogen gas stream. The final volume was ad-
justed to 1 ml with acetonitrile. The sample was
then analyzed by electron impact ŽEI. GCrMS,
using a Hewlett-Packard 5890 GC equipped with
a 7673A automatic sampler and interfaced to a
5970B mass selective detector ŽMSD.. Injections
Ž1-m l. were made in splitless mode onto a 60-m=
0.25-mm ID DB-5 fused-silica capillary column
 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71 67

Fig. 1. Ross Island and McMurdo Station, Antarctica, with soil sample locations.
68 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71

ŽJ &W Scientific Inc... Identification and quanti-
fication of the PAH were made by multiple ion
detection ŽMID., monitoring the molecular ion of
each PAH, and deuterated PAH standards.
Prior to extraction, the following deuterated
internal standards were added to each soil sam-
ple: naphthalene-d 8 ; acenaphthylene-d 8 ; phenan-
threne-d 10 ; anthracene-d 10 ; chrysene-d 12 ; fluor-
anthene-d 10 ; pyrene-d 10 ; benzw axanthracene-d 12 ;
b en zo w e x pyren e-d 1 2 ; b en zo w a x pyren e-d 1 2 ;
benzow k xfluoranthene-d 12 ; and benzow ghi xpery-
lene-d 12 . NIST SRM 1647 Žcertified PAH., with
the addition of deuterated internal standards and
authentic standards of compounds not present in
the SRM, were used to make calibration solu-
tions. Calibration curves for the GCrMSrMID
quantification were made for the molecular ion
peaks of the PAH using the corresponding
deuterated species Žor the deuterated species most
closely matched in volatility and retention charac-
teristics as internal standards.. The soil samples
were corrected for moisture content and results
were reported in m grkg on a dry weight basis.
Practical quantitation limits ŽPQLs. were calcu-
lated based on minimum detection limits ŽMDLs..
MDL values are based on seven replicate analy-
ses of the lowest standard for each PAH of inter-
est. The final MDL values are three times the
standard deviation of the replicate sets. The com-
pound-specific PQLs were then calculated by tak-
ing the MDLs and accounting for an average soil
sample mass of 0.0309 kg, on a dry-weight basis,
for each PAH. Compound-specific PQLs are listed
in Table 1 for samples with concentrations de-
tected above the MDLs.
3. Results and discussion
PAH concentrations detected in soils in the
vicinity of McMurdo Station, Antarctica, are listed
in Fig. 2. The highest concentrations detected
were in soils located in unpaved roadbeds near
the sea water intake Žlocation 5. and the gasoline
pumps Žlocation 10.. It is most likely that soil
samples collected in these areas were contami-

Table 1
Number of detections, MDL, PQL, average, and maximum Žin ppb. per PAH a

No. detectedr Abbreviation MDLc PQLd Averagee Maximumf Site

totalb ID

Naphthalene 4r20 NAP 0.027 0.88 6700 " 1300 27 000 " 2600 5
Biphenyl 3r20 BIP 0.012 0.39 640 " 190 1710 " 360 5
Acenaphthylene 3r20 ACY 0.063 2.0 5500 " 3500 15 700 " 6300 5
Acenaphthene 3r20 ACE 0.030 0.97 6000 " 730 17 800 " 1270 5
Phenanthrene 4r20 PHE 0.0060 0.194 160 " 15 540 " 30 5
Fluorene 3r20 FLE 0.0150 0.49 580 " 38 1590 " 69 5
Anthracene 2r20 ANT 0.0060 0.194 2500 " 370 5000 " 650 5
Fluoranthene 5r20 FLA 0.0180 0.58 2700 " 250 13 300 " 430 5
Pyrene 6r20 PYR 0.0120 0.39 28 " 3.4 132 " 6.4 5
Benzw axanthracene 3r20 BAA 0.030 0.97 510 " 0.78 1420 " 1.38 10
Chrysene 3r20 CHR 0.0150 0.49 570 " 90.0 1630 " 160 5
Benzow e xpyrene 1r20 BEP 0.0090 0.29 250 " 45 250 " 45 5
Dibenzow ahq ac xpyrene 1r20 DAP 0.039 1.26 320 " 1.8 320 " 1.8 5
Benzow b xchrysene 1r20 BBC 0.021 0.68 2700 " 2500 2700 " 2500 5
Coronene 1r20 COR 0.069 2.2 670 " 3.2 670 " 3.2 5
a
Uncertainty values are analytical uncertainties for individual samples or standard deviations for more than one sample.
b
Number of detections above minimum detection limit ŽMDL. out of 20 soil samples collected.
c
Minimum detection limit equals three times the standard deviation of the replicate analysis of lowest standards.
d
PQL, practical quantitation limit, in m grkg soil Žppb.. Refer to text, Section 2.2.
e
Samples with single detection, ‘average’ value equals value reported for single detection.
f
This column presents the maximum concentration detected with the corresponding site identified in the next column.
 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71 69

nated by spillage of fuels or oils from vehicles tectable levels of PAHs of interest: downslope
containing PAHs of interest. In addition, the fol- from the power plant Žlocation 1.; the dirt road-
lowing soil sample locations all contained de- way ramp from the ice aircraft runway Žlocation

Fig. 2. Summary of PAH concentrations detected in soil Žppb. per sample location.
70 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71
3.; the lower helicopter pad Žlocation 6.; an un-
paved roadway in the retrograde yard Žlocation
12.; the vicinity of Tank 98, a JP-8 diesel fuel
storage tank Žlocation 14.; and the unpaved con-
vergence of several roadways between buildings
136 and 120 Žlocation 20..
The two soil samples collected from the Taylor
Valley did not have any detectable levels of PAHs.
The presence of a Long Term Ecological Re-
search basecamp at Lake Hoare represents a
community of approximately 4]10 scientists dur-
ing the austral summer. Activities that may pro-
duce PAHs in this area include helicopter loading
and unloading, electrical generation from a
diesel-fueled generator, burning of solid sewage
in small incinerator toilets, and use of small all-
terrain vehicles. Samples were collected on the
upwind and downwind sides of the helicopter pad,
using the same protocol as for the McMurdo
samples. While the sensitivity of analyses may
have played a role in detecting any PAHs present
in these samples Žsee Table 1 for MDLs and
PQLs., one may conclude that activities at Lake
Hoare are not of sufficient magnitude to generate
detectable quantities of PAHs. In addition, these
results suggest any downwind transport of PAHs
from McMurdo is not detectable.
Regarding McMurdo soils, the following are
high-probability reasons for the presence of PAHs
detected at each of the above mentioned loca-
tions: Ž1. soils near the power plant are likely to
contain PAHs deposited from stack emissions
from the diesel generators as the prevailing winds
are generally downslope towards McMurdo
Sound; Ž2. the dirt ramp to the ice runway is a
site of heavy vehicular traffic during the runway’s
operations and PAHs are most likely present
from both light and heavy duty vehicles burning
gasoline and diesel; Ž3. the lower helicopter pad
is the sight of the most helicopter operations, and
PAHs present are likely due to helicopter exhaust
and or re-fueling operations; Ž4. the retrograde
yard functions as a location for the recycling,
destruction and packaging for transport of used
materials where heavy equipment, operating on a
regular basis, may be a likely source for PAHs
detected. While tank 98 is used for storage of
JP-8 fuel, it is difficult to determine the likely
source of PAHs detected in this vicinity. Simi-
larly, it is difficult to determine the likely source
of pyrene, a PAH, detected between buildings 136
and 120, unless vehicular traffic in this location
has acted as a source of pyrene in soils. Building
136 has been used as office space and repair and
construction shops. The types of designated pollu-
tants stored included solvents, oils and lubricants,
paints and primers, and copier tonerŽASA, 1994..
Table 1 summarizes the results of the 20 soil
samples collected. Pyrene had the greatest num-
ber of detections above the MDL Ž6 total., fol-
lowed by fluoranthene Ž5 total. and phenanthrene
and naphthalene Ž4 total.. The highest average
and maximum concentrations detected were for
naphthalene Žave s 6700 " 1300 m grkg, max s
27 000 " 2600 m grkg. followed by acenaphthene
Ž ave s 6000 " 730, m ax s 17 800 " 1270
m grkg . ,acenaphthylene Ž ave s 5500 " 3500
m grkg, max s 15 700 " 6300 m grkg. and fluo-
ranthene Žave s 2700 " 250 m grkg, max s 13 300
" 430 m grkg.. These ranges in concentrations
between different PAH species may be due to
different proportions formed during combustion
processes. Proportions of each PAH species pro-
duced depends on several conditions, for exam-
ple, fuel type and composition, and combustion
temperature.
The data from this study suggest that there is a
measurable anthropogenic input of PAH into the
local environment at McMurdo Station. Kenni-
cutt et al. Ž1995. report sediment PAH levels in
McMurdo Sound in the ranges of 6267]6339 ppm
for the austral Summers of 1990r1991 and
1992r1993. The predominant explanation for such
high concentrations in the sound immediately ad-
jacent to McMurdo is from the runoff from adja-
cent land areas. The concentrations we have re-
ported are within the range of concentrations
typically reported for forest soils Ž5]100 ppb.
ŽMenzie et al., 1992. with the exception of several
of the PAHs reported in the 1000]27 000 m grkg
Žppb. range Žsee maximum concentrations re-
ported above.. One must keep in mind that these
samples represent a general survey of the range
of PAHs that are present in soils in the vicinities
of Lake Hoare and McMurdo Station, Antarctica.
The concentrations detected may over-represent
 D. Mazzera et al. r The Science of the Total En¨ ironment 229 (1999) 65]71
actual PAH concentrations if one takes into ac-
count random sampling of point sources Že.g. oil
and fuel spills. that are small in number but may
contain high levels of PAHs. Conversely, concen-
trations may be underestimated if one takes into
account physical and biological degradation
processes Žlikely to be minimal in this type of
environment. as well as dilution and dispersion
from non-point combustion sources.
4. Conclusions
The results of this study indicate that human
activities have contributed to the overall levels of
PAHs contaminating McMurdo Station, Antarc-
tica. This study provides a potential source for the
high levels of PAHs found in sediments in Mc-
Murdo Sound as reported by Kennicutt et al.
Ž1995.. PAHs from local soils transported in runoff
from snow melt is a plausible explanation for
elevated concentrations of these compounds in
McMurdo Sound sediments. Further studies on
sources of PAHs at McMurdo are needed to help
reduce any potential elevation in concentrations
of these high-profile semi-volatile organic com-
pounds.
Acknowledgements
The authors gratefully acknowledge the use of
71
the inorganic and organic lab facilities at the
Desert Research Institute, run under the direc-
tion of Dr Judith Chow and Dr Barbara Zielin-
ska, respectively.
References
ASA. Master Permit Application ŽFinal.. Antarctic Support
Associates ŽASA. and Consoer Townsend & Associates,
vol. 1, 27 June 1994.
Basu D, Saxena J. Comparison of drinking water mutagenicity
with leaching of polycyclic aromatic hydrocarbons from
water distribution pipes. Chemosphere 1987;16:2595]2612.
Freeman D, Cattell F. Woodburning as a source of atmo-
spheric polycyclic aromatic hydrocarbons. Environ Sci
Technol 1990;24:1581]1585.
Jones K, Stratford J, Waterhouse K et al. Increases in the
polynuclear aromatic hydrocarbon content of an agricul-
tural soil over the last century. Environ Sci Technol
1989;23:95]101.
Kennicutt II M, McDonald S, Sericano J et al. Human con-
tamination of the marine environment } Arthur Harbor
and McMurdo Sound, Antarctica. Environ Sci Technol
1995;29:1279]1287.
Matzner E. Annual rates of deposition of polycyclic aromatic
hydrocarbons in different forest ecosystems. Water Air Soil
Pollut 1984;21:425]434.
Menzie C, Potocki B, Santodonato J. Exposure to carcino-
genic PAHs in the environment. Environ Sci Technol
1992;26:1278]1284.
United States Environmental Protection Agency ŽUSEPA..
Evaluation and estimation of potential carcinogenic risks of
polynuclear aromatic hydrocarbons: carcinogen assessment
group. Washington, DC: Office of Health and Environmen-
tal Assessment, 1985.
USEPA. SW846 reference methodology: method 3540B-Soxh-
let Extraction. Revision 2, September 1994.