Environmental Toxicology and Chemistry, Vol. 24, No. 10, pp.

25002508, 2005
q 2005 SETAC
Printed in the USA
0730-7268/05 $12.00 1 .00

EFFECTS OF LONG-CHAIN HYDROCARBONPOLLUTED SEDIMENT ON

FRESHWATER MACROINVERTEBRATES

VINCENT PETTIGROVE* and ARY HOFFMANN

Research and Technology, Melbourne Water Corporation, GPO Box 4342, Melbourne, Victoria 3001, Australia
Centre for Environmental Stress and Adaptation Research, La Trobe University, Bundoora, Victoria 3086, Australia
Centre for Environmental Stress and Adaptation Research, Department of Zoology, University of Melbourne, Parkville,
Victoria 3010, Australia

( Received 10 January 2005; Accepted 24 March 2005)

AbstractHighmolecular weight (.C16) hydrocarbons (HMWHs) are common pollutants in sediments of freshwater systems,
particularly urban water bodies. No sediment quality guidelines exist for total hydrocarbons; more emphasis is placed on polyaromatic
hydrocarbons, the most toxic component of hydrocarbons. A field-based microcosm experiment was conducted to determine whether
unpolluted sediments spiked with synthetic motor oil impair freshwater macroinvertebrate assemblages. Total petroleum hydrocarbon
(TPH) concentrations of 860 mg/kg dry weight significantly increased the abundance of Polypedilum vespertinus and Cricotopus
albitarsis and decreased the abundance of Paratanytarsus grimmii adults (all Chironomidae), whereas TPH concentrations ranging
from 1,858 to 14,266 mg/kg produced a significant reduction in the total numbers of taxa and abundance, with significant declines
in the abundance of nine chironomid taxa. About 28% of water bodies surveyed in urban Melbourne, Australia, had TPH concen-
trations in sediments likely to cause ecological impairment, and about 14% of the water bodies surveyed are likely to have reduced
species richness and abundance. Therefore, HMWHs can be a significant pollutant in urban water bodies. Freshwater sediment
quality guidelines should be developed for this ubiquitous urban pollutant.

KeywordsBioindicators Sediment contamination Chironomidae Microcosms Urban pollution

INTRODUCTION organisms, they can alter the physical properties of sediments,

Hydrocarbons are common anthropogenic pollutants in the
waters and sediments of urban water bodies that are mainly
derived from crude oil (gasoline, kerosene, fuel oil, mineral
oil, and asphalt) [1]. It is estimated that 1.7 to 8.8 million
metric tons of petroleum oils are released into the environment
annually [2]. Major sources of petroleum hydrocarbon pol-
lution include atmospheric deposition from vehicle and in-
dustrial emissions, asphalt degradation and leakage of oils
from vehicles [1,3] and gray water [4].
Petroleum or crude oil is a complex mixture of many hy-
drocarbons that mainly consist of saturated predominantly
straight-chain alkanes, small amounts of slightly branched al-
kanes, cycloalkanes, and aromatics [5]. The aromatic com-
ponents of oils, particularly polyaromatic hydrocarbons
(PAHs), are generally assumed to be the most toxic component
of petroleum hydrocarbons and are generally the focus of field
surveys (e.g., [68]). Heterocyclics may be a major determi-
nant of toxicity in water polluted with weathered distillate oils
[9]. Synthetic motor oils have also become more widely used
in vehicles and industry in past decades (e.g., http://
www.mobil1.com).
Lowmolecular weight hydrocarbons are more soluble, and
therefore more toxic, to aquatic organisms than those of higher
molecular weight [10]. Highmolecular weight hydrocarbons
(HMWHs) have very limited solubility in water and are more
likely to sorb onto soil [11]. Recalcitrant aliphatics and other
HMWHs are the primary hydrocarbon fractions found in sed-
iment [12] and are considered to have a low toxicity potential
[13]. Although HMWHs might not be directly toxic to aquatic
* To whom correspondence may be addressed
(vin.pettigrove@melbournewater.com.au).
2500
coat and smother organisms [14], and contribute to organic
enrichment of sediments [15].
Sediment quality guidelines have been proposed for PAHs
[1618], but not for other hydrocarbon compounds that com-
monly occur in urban sediments. This raises the question of
whether sediment quality guidelines for just PAHs provide an
adequate indication of whether hydrocarbon-polluted sedi-
ments are affecting the health of aquatic ecosystems. There-
fore, the need to investigate the ecological effects of sediments
polluted with HMWHs, but containing low concentrations of
PAHs, is warranted.
A field-based microcosm method (following [19]) was used
to determine whether sediments contaminated by HMWHs,
but not PAHs, affect the macroinvertebrates that can inhabit
these sediments. Macroinvertebrates would appear to be good
indicators of sediment pollution because many taxa are as-
sociated with sediments. Clean sediments and sediments
spiked with four different concentrations of weathered syn-
thetic motor oil were used to describe how the composition
and abundance of macroinvertebrate assemblages can change
with increased levels of hydrocarbon pollution in sediments.
The sediments spiked with synthetic motor oil contained low
levels of PAHs but elevated concentrations of total petroleum
hydrocarbons (TPHs) within the C15 to C36 range (hence
referred to as HMWHs), which is similar to what is found
when HMWHs are exposed to weathering [1922].
The purpose of this investigation was to determine whether
sediments polluted with HMWHs, but with low levels of PAHs,
can severely impair aquatic ecosystems and whether the de-
velopment of sediment quality guidelines for HMWHs is war-
ranted. Finally, to determine the ecological significance of
Effects of hydrocarbons on macroinvertebrates
HMWH pollution on urban aquatic ecosystems, the TPH con-
centrations present in sediments in this experiment are com-
pared to those present in urban streams, drains, and wetlands
in Melbourne, Australia. Melbourne is a good location to as-
sess the contribution of urban runoff to petroleum hydrocarbon
pollution in freshwater sediments because it has separate storm
water and sewerage systems, and spills of sewage into the
storm water system rarely occur.
MATERIALS and METHODS
Design of microcosm experiments
Nonpolluted sediments were collected from Glynns Wet-
land, located approximately 30 km east of Melbourne, Aus-
tralia. These sediments are used as reference sediments in this
experiment because previous experiments have demonstrated
that Glynns Wetland sediments can support a variety of mac-
roinvertebrate taxa [19]. Glynns Wetland sediments were
spiked with four different concentrations of synthetic motor
oil (Mobil 1t 10W-40, Melbourne, Australia). The spiked sed-
iments were thoroughly mixed in a blender for 30 min and
stored for two months (to allow weathering) before being
placed in field-based microcosms. Ten replicate microcosms
were set up for the nonpolluted Glynns Wetland sediment, and
five replicates were set up for the remaining four motor oil
treatments. A larger number of replicates were set up for
Glynns Wetland sediment to increase the statistical power of
comparisons between a nonpolluted sediment and several pol-
luted sediments.
Synthetic motor oils are manufactured from a variety of
synthetic hydrocarbons and nonhydrocarbon organic com-
pounds. The main ingredient in the synthetic motor oil tested
was polyalphaolefins (http://www.mobil1.com), which bind
tightly to sediments, have low bioavailability, and are rela-
tively nontoxic compared with mineral oils, diesel fuel, and
internal olefins [23]. The polyalphaolefins were selected as a
surrogate of HMWH pollution because their low bioavail-
ability and toxicity provide a conservative estimate of the po-
tential effects of hydrocarbon pollution in freshwater sedi-
ments.
The PAH concentrations in sediments from each treatment
were analyzed according to the U.S. Environmental Protection
Agency (U.S. EPA) Method 8270C (http://www.epa.gov/
epaoswer/hazwaste/test/pdfs/8270c.pdf). Volatile petroleum
hydrocarbons (C6C9) were extracted by purge and trap (U.S.
EPA Method 5030B; http://www.epa.gov/epaoswer/hazwaste/
test/pdfs/5030b.pdf) and analyzed according to U.S. EPA
Method 8260B (http://www.epa.gov/epaoswer/haswaste/test/
pdfs/8260b.pdf). Semivolatiles were analyzed according to
U.S. EPA Method 8015B (http://www.epa.gov/epaoswer/
hazwaste/test/pdfs/8015b.pdf). All analyses were conducted by
the Australian Laboratory Services, which is accredited by the
National Association of Testing Authorities, Australia (http://
www.nata.asn), to conduct these tests. Blanks, checks, and
spikes were conducted as part of quality control procedures
for all analyses. Sediments that were analyzed for TPHs were
spiked with TPH/benzene, toluene, ethylbenzene, and xylene
compound surrogates. The recovery of all surrogates was with-
in 70 to 130% of the initial concentration, thus indicating a
satisfactory recovery of these compounds.
The microcosms (20-L polypropylene tanks with 750 g of
sediment and 15 L of filtered water from the wetland) were
placed along the edge of Glynns Wetland. A random block
design was used in the experiment. The microcosms were ar-
Environ. Toxicol. Chem. 24, 2005 2501
ranged into five blocks, with each block containing six mi-
crocosms (two Glynns Wetland sediments and one each of the
four motor oilspiked treatments). Briefly, insects that emerge
from the wetland mate and then randomly lay eggs within the
microcosms. This method provides a unique means of gath-
ering information on the effects of sediment quality on mac-
roinvertebrates in lentic habitats, particularly for indigenous
species that cannot be easily reared in the laboratory. Further
details regarding the method used for the field-based experi-
ment are described by Pettigrove and Hoffmann [19], except
the following modifications were made to collect adults emerg-
ing from the microcosms. Fine 100% nylon elastic mesh (Cool
Catst stockings, Coolaroo, Victoria, Australia) was placed
over each microcosm once emergence from the microcosms
was evident (after 50 d). This mesh was sufficiently fine (,100
mm) to prevent new insects from colonizing the microcosms
and to contain emerged insects within the microcosms, but
sufficiently coarse to allow ventilation for reducing conden-
sation within the microcosms. Adult insects that emerged from
the microcosms were collected with aspirators approximately
once a week. These were stored dry in 1.5-ml Eppendorf tubes
at 48C until they were processed in the laboratory.
The experiment was terminated 88 d after the mesh had
been placed over the microcosms, when most species had com-
pleted their life cycle and few individuals were emerging from
the microcosms. The only species that did continue to emerge
in large numbers at this time was the parthenogenetic [24]
chironomid Paratanytarsus grimmii. Adults of this species
were observed laying fertile eggs within the tanks soon after
emergence. No other species was able to successfully breed
within the confines of the microcosms once these had been
covered with mesh. After the experiment was terminated, re-
maining larvae in the microcosms were collected with a 64-
mm mesh net and stored in 100% ethanol.
Common macroinvertebrate taxa were identified to the low-
est possible taxonomic unit. Dytiscidae larvae and chironomid
adults, pupae, and larvae were identified with the use of keys
by Pettigrove and Hoffmann [19]. Hemiptera [25], Hydrop-
tilidae larvae [26], and Ephemeroptera nymphs [27] were iden-
tified with the use of keys by Anderson and Weir [25], Wells
[26], and Dean and Suter [27]. Molecular markers (following
[28]) were also examined in some species of Chironomidae to
confirm species identification.
Data analyses
All analyses were conducted with SPSS V 11.0 for Win-
dows (Version 11.5, SPSS, Chicago, IL, USA). Because the
biological data exhibited skewed distributions, nonparametric
tests were conducted to compare species abundance between
treatments. Two-tailed MannWhitney U tests were undertak-
en to compare the total number of taxa and individuals between
treatments. A second test was conducted on the total number
of individuals present, excluding P. grimmii, because this spe-
cies dominated the number of individuals present and obscured
other trends in faunal abundance.
Nonparametric KruskalWallis and JonckheereTerpstra
tests were undertaken to determine whether significant differ-
ences occurred in the abundance of each common taxon be-
tween each of the five treatments. It is hypothesized that the
abundance of common taxa gradually declines with increased
HMWH pollution. To test this hypothesis, a JonckheereTerps-
tra test of each common taxon was conducted to determine
whether abundance declined significantly with a gradient of
2502 Environ. Toxicol. Chem. 24, 2005
increased pollution. An alternative hypothesis is that moderate
levels of pollution increase the abundance of some common
taxa and higher levels of pollution decrease abundance. In
these instances, the abundance of a taxon would not vary in
an a priori manner. Therefore, KruskalWallis tests were used
to test whether significant differences in abundance occurred
in each common taxon between treatments.
Two-tailed MannWhitney U tests were conducted in a
sequential manner to explore what concentrations of TPH in
sediments produced significant changes in the total numbers
of individuals, total number of taxa, and abundance of common
taxa along a gradient of HMWH pollution in sediments. In
these tests, treatment 0 was first compared with treatment 1.
If there was no significant difference between these treatments,
they were compared with treatment 2. If there was a significant
difference between treatments 0 and 1, then only treatment 1
was compared with treatment 2. This process continued until
comparisons had been undertaken between all treatments.
An evaluation of sublethal responses to environmental
stress might provide a better early warning system of ecolog-
ical impairment than metrics based on species presence and
abundance [29]. Growth is often chosen as a response variable
to measure sublethal effects because it is easy to quantify and
integrates a suite of biochemical and physiological effects into
a metric that is often linked to individual fitness [30]. The
emergence rates of adult Cricotopus albitarsis were compared
between treatments 0 to 3 to determine whether HMWH pol-
luted sediments produce sublethal stress, as indicated by re-
tarded rates of emergence. This assessment was confined to
C. albitarsis because it was the only species, except for P.
grimmii (which could breed within closed microcosms), in
which .10 adults emerged from at least four treatments. Two-
tailed MannWhitney U tests were conducted in a sequential
manner to explore whether sediments polluted with TPHs de-
layed adult emergence, as indicated by the time taken for adults
to first emerge and for 50% of adults to emerge from each
treatment. The sequential process is the same used to compare
changes in total numbers of individuals, total numbers of taxa,
and abundance of common taxa along a gradient of HMWH
pollution in sediments.
Comparison of experimental TPH concentrations with
field data
Total petroleum hydrocarbon concentrations present in the
experimental treatments were compared with field data col-
lected by the watershed management authority for Melbourne
(Melbourne Water) from urban streams, drains, and wetlands
(including lakes) in the Melbourne region. These data were
used to determine the proportion of urban aquatic ecosystems
likely to show levels of pollution comparable to that found in
the field-based experiments. Total petroleum hydrocarbon sed-
iment concentrations were measured at wadeable (,2 m)
depths from 104 sites in streams, 112 sites on subterranean
drains, and 269 wetlands and lakes. The method for collection
of sediments is detailed in Pettigrove and Hoffmann [19], ex-
cept that unfiltered sediments were used for the analysis of
TPHs. Sediments were also analyzed for PAHs; however, these
data are not presented because .99% of sediments had PAH
concentrations below the threshold effect concentrations
(TEC) [18] and effects range low (ERL) [17] sediment quality
guidelines, indicating that harmful PAH effects are potentially
rare.
V. Pettigrove and A. Hoffmann
Table 1. Total petroleum hydrocarbon (TPH) concentrations (dry wt)
separated into carbon number range in reference sediment (treatment
0) and sediment spiked with motor oil (treatments 14)
TPH treatment (mg/kg)
Carbon no.
Range 0 1 2 3 4
C6C9 ,4 ,4 ,4 ,4 ,4
C10C14 ,25 26 ,25 ,25 38
C15C28 ,50 316 727 1,811 5,580
C29C36 ,50 518 1,131 2,821 8,648
Total TPH ,125 860 1,858 4,632 14,266
RESULTS
Chemical analysis of sediments
The TPH concentrations in sediments ranged from below
detection limits (,125 mg/kg) in nonpolluted Glynns Wetland
sediment (treatment 0) to 860, 1,858, 4,632, and 14,266 mg/
kg in treatments 1, 2, 3, and 4, respectively (Table 1). The C6
to C9 TPH fraction was below the detection limit of 4 mg/kg
in all samples. The TPHs present were typically from the large
carbon fractions: 38 to 39% of the TPHs were in the C15 to
C28 fraction and 60 to 61% in the C29 to C36 fraction.
All 16 PAHs analyzed in these sediments were below the
detection limit of 0.25 mg/kg in all treatments. Therefore, all
treatment sediments were below the available freshwater sed-
iment TEC [18] and ERL [17] sediment quality guidelines,
indicating that harmful effects from PAH pollution are unlikely
to be observed.
Common macroinvertebrate taxa
A total of 36 taxa and 19,499 individuals were collected
from the 30 microcosms used in this experiment. The fauna
was dominated by the chironomid P. grimmii, which repre-
sented 74.9% of all individuals collected. A list of the common
taxa present in this experiment and the number of microcosms
they inhabited are presented in Table 2. The fauna was dom-
inated by Chironomidae (Diptera), and species from the He-
miptera, Coleoptera, Ephemeroptera, and Trichoptera were
also common. Fifteen taxa (P. grimmii, Procladius villosi-
manus, Mesovelia sp., Tasmanocoenis sp., Micronecta annae,
Kiefferulus martini, Necterosoma pencillatum, Hellyethira
sp., Larsia albiceps, Tanytarsus fuscithorax, Polypedilum
vespertinus, C. albitarsis, Cladopelma curtivalva, Polype-
dilum watsoni, and Cladotanytarsus bispinosus) were present
in six or more microcosms.
Comparison of number of taxa between treatments
The median number of taxa in microcosms ranged from 11
per microcosm in treatment 2, to 0 per microcosm in treatment
4 (Fig. 1A). The numbers of taxa decline significantly between
treatments 0, 1 and 2, and treatment 3 (MannWhitney U test,
p , 0.01) and between treatments 3 and 4 (MannWhitney U
test, p , 0.01). Therefore, TPH concentrations between 1,858
and 4,632 mg/kg in sediments reduced species richness, and
concentrations exceeding 4,632 mg/kg led to a severely im-
poverished fauna.
Comparison of number of individuals between treatments
The median number of individuals per treatment ranged
from 1,568 per microcosm in treatment 2 to 0 per microcosm
in treatment 4, as reported in Figure 1B (error bars represent
the 25th and 75th percentiles for each treatment). The number
Effects of hydrocarbons on macroinvertebrates Environ. Toxicol. Chem. 24, 2005 2503

Table 2. List of the most abundant invertebrate taxa, percentage of total abundance, and the number of microcosms the taxon inhabited

Taxon Family/Order No. of individuals % Total abundance No. of microcosms

1 Paratanytarsus grimmii Chironomidae 13,930 71.40 24

2 Procladius paludicola Chironomidae 1,774 9.09 22
3 Mesovelia sp. Hemiptera 1,486 7.62 6
4 Tasmanocoenis sp. Ephemeroptera 528 2.71 13
5 Polypedilum nubifer Chironomidae 299 1.53 3
6 Micronecta annae Hemiptera 154 0.79 6
7 Kiefferulus martini Chironomidae 153 0.78 8
8 Necterosoma pencillatum Coleoptera 144 0.73 10
9 Hellyethira simplex Trichoptera 127 0.65 6
10 Larsia albiceps Chironomidae 112 0.57 19
11 Tanytarsus fuscithorax Chironomidae 112 0.57 11
12 Coelopynia pruinosa Chironomidae 102 0.52 3
13 Polypedilum vespertinus Choronomidae 98 0.50 12
14 Cricotopus albitarsis Chironomidae 88 0.45 16
15 Cladopelma curtivalva Chironomidae 78 0.40 11
16 Polypedilum watsoni Chironomidae 74 0.38 9
17 Oligochaeta 64 0.33 1
18 Cladotanytarsus bispinosus Chironomidae 35 0.18 11
19 Cryptochironomus delfinicus Chironomidae 33 0.17 2

of individuals between treatments 3 and 4 declined signifi-

cantly (MannWhitney U test, p , 0.01), but with no signif-
icant differences between remaining treatments.
Because the number of individuals present is dominated by
P. grimmii, which was observed laying fertile eggs within the
confines of the microcosms, another analysis was conducted
comparing the number of individuals between treatments with
all data, except P. grimmii. With the exclusion of P. grimmi,
the numbers of individuals with increasing TPH sediment con-
centrations from treatment 0 (median 5 178 individuals/mi-
crocosm) to treatment 4 (median 5 0 individuals/microcosm)
reduces progressively (Fig. 1C). Significant declines in the
number of individuals occur between treatments 0 and 1 and
treatment 2 (MannWhitney U test, p , 0.001), and treatments
2 and 3 and treatment 4 (MannWhitney U test, p 5 0.028).
Therefore, TPH sediment concentrations between 860 and
1,858 mg/kg can reduce macroinvertebrate abundance, and
TPH sediment concentrations exceeding 1,858 mg/kg will lead
to further reductions in macroinvertebrate abundance.

Comparison of abundance of common taxa between

Fig. 1. The average number of taxa (A), average number of individuals
(B), and average number of individuals excluding Paratanytarsus
grimmii (C) per microcosm in total petroleum hydrocarbon (TPH)
treatments. The error bars indicate the 25th and 75th percentiles, and
the asterisks indicate data sets are significantly ( p , 0.05) different
from one another.
TPH treatments
Paratanytarsus grimmii was the most abundant taxon col-
lected during this experiment and was present in 24 of the 30
microcosms (Table 2). Large numbers of larvae were present
in many microcosms when the experiment was concluded. This
species is parthenogenetic [25], and adults that emerged from
the microcosms were observed laying fertile eggs within the
confines of the microcosms. The abundance of P. grimmii
adults and larvae was significantly different between treat-
ments (KruskalWallis test; Table 3). Although P. grimmii
adults and larvae were abundant in treatments 0 to 3 (Fig. 2A),
their abundance was significantly lower in treatment 4 than in
other treatments (MannWhitney U test, p 5 0.001). Para-
tanytarsus grimmi occurred in 92% of microcosms in treat-
ments 0 to 3 but was only present in one of the five microcosms
used for treatment 4.
A higher sensitivity to HMWH pollution is apparent in P.
grimmii when only adults are analyzed (JonckheereTerpstra
test; Table 3). The median number of individuals per treatment
ranged from 66.4 individuals/microcosm in treatment 0, to 0
individuals/microcosm in treatment 4 (Fig. 2B). The number
2504 Environ. Toxicol. Chem. 24, 2005 V. Pettigrove and A. Hoffmann

Table 3. Results of KruskalWallis and JonckheereTerpsta tests tini, and P. watsoni declines significantly with increased TPH
determining whether significant differences exist between the pollution (JonckheereTerpstra; Table 3). Abundance and fre-
abundance of common taxa between all total petroleum hydrocarbon
treatmentsa quency of occurrence of C. curtivalva are similar between
treatments 0, 1, and 2, whereas it is absent in treatments 3 and
Jonckheere 4 (Fig. 3). A further significant decline in abundance of C.
Taxon Terpstra KruskalWallis curtivalva occurred between treatments 0, 1, and 2, and treat-
Cricotopus albitarsis NS 0.004 ments 3 and 4 (MannWhitney U test, p 5 0.03). Significantly
Cladotanytarsus bispinosus NS NS fewer P. villosimanus occurred in treatment 2 than in treat-
Cladopelma curtivalva 0.016 NS ments 0 and 1 (MannWhitney U test, p 5 0.004), and sig-
Procladius villosimanus ,0.001 ,0.001 nificantly fewer individuals were present in treatment 3 than
Kiefferulus martini 0.012 NS in treatment 2 (MannWhitney U test, p 5 0.042). Kiefferulus
Larsia albiceps 0.002 0.009
Mesovelia sp. NS NS martini was present in only eight microcosms, but seven of
Micronecta annae NS NS these are from treatments 0 or 1, and one from treatment 2
Necterosoma pencillatum NS NS (Fig. 3). Abundance declined significantly between treatments
Paratanytarsus grimmii (all) NS 0.017 0 and 1 and the remaining treatments (MannWhitney U test,
Paratanytarsus grimmii (adults) ,0.001 ,0.001
Polypedilum vespertinus NS 0.021 p 5 0.011). Polypedilum watsoni was present in 75% of mi-
Polypedilum watsoni 0.009 0.074 crocosms used in treatments 0 and 1 but was present in only
Tasmanocoenis sp. NS NS one microcosm used in treatment 2 and absent in treatments
Tanytarsus fuscithorax NS 0.048 3 and 4 (Fig. 3). Polypedilum watsoni abundance was signif-
a NS 5 not significant.
icantly lower in treatments 2, 3, and 4 compared with treat-
ments 0 and 1 (MannWhitney U test, p 5 0.012).
The abundance of C. albitarsis, P. vespertinus, Larsia al-
of adults present in treatments 1 and 2 was significantly lower biceps, and Tanytarsus fuscithorax differ between treatments
than those present in treatment 0 (MannWhitney U test, p 5 (KruskalWallis tests; Table 3), but these species tended to be
0.046), and adult abundance was significantly lower in treat- most abundant in sediments with low to moderate levels of
ment 3 than in treatments 1 and 2 (MannWhitney U test, p TPH pollution (Fig. 3). This suggests that moderate levels of
5 0.006). Therefore, although the larvae are able to exist in HMWH pollution in sediments might provide an advantage
the more polluted sediments (treatments 13), even low levels for these species. Cricotopus albitarsis abundance was sig-
of TPH pollution (e.g., treatment 1) are sufficient to reduce nificantly higher in treatments 1 and 2 than in treatment 0
adult emergence. (MannWhitney U test, p 5 0.002); significantly higher in
The abundance of C. curtivalva, P. villosimanus, K. mar- treatments 1, 2, and 3 than in treatment 0 (MannWhitney U
test, p 5 0.012); and significantly lower between treatments
1, 2, and 3 and treatment 4 (MannWhitney U test, p 5 0.008).
Polypedilum vespertinus was more abundant in treatments 1
and 2 than in treatment 0 (MannWhitney U test, p 5 0.027)
and more abundant in treatments 1 and 2 than in treatments
3 and 4 (MannWhitney U test, p 5 0.009). Larsia albiceps
and T. fuscithorax were most abundant in treatments 2 and 1,
respectively, but not higher than in treatment 0. Larsia albi-
ceps was less abundant in treatments 3 and 4 than in treatments
0, 1, and 2 (MannWhitney U test, p 5 0.002). Tanytarsus
fuscithorax was less abundant in treatments 2 and 3 than in
treatments 0 and 1 (MannWhitney U test, p 5 0.015).
The abundance of C. bispinosus, Mesovelia sp., M. annae,
N. pencillatum, and Tasmanocoenis sp. between treatments
was not significantly different by either the JonckheereTerps-
tra or KruskalWallis test. Statistical power was limited for
analyzing Mesovelia sp. and M. annae because they were
present in six of the 30 microcosms. Cladotanytarsus bispi-
nosus, N. pencillatum, and Tasmanocoenis sp. were able to
tolerate HMWH-polluted sediments at levels present in treat-
ment 3, but not treatment 4 (Fig. 3), suggesting they are af-
fected by TPH sediment concentrations between 4,632 and
14,266 mg/kg.

Adult emergence rate

Fig. 2. The effects of total petroleum hydrocarbon (TPH) pollution
on abundance and frequency of occurrence of adults and larvae (A)
and adults (B) of Paratanytarsus grimmii. The asterisks indicate data
sets are significantly ( p , 0.05) different from one another.
Adult C. albitarsis emerged from several treatments and
were tested for delayed emergence with increased HMWH
pollution in sediments. The median value for emergence was
between 6 and 9 d after the microcosms were closed to aerial
colonization for treatments 0, 1, and 2, whereas the median
emergence for treatment 3 was 14 d (Fig. 4). Emergence com-
menced in treatments 0, 1, and 2 between 1 and 5 d after the
Effects of hydrocarbons on macroinvertebrates Environ. Toxicol. Chem. 24, 2005 2505

Fig. 3. The effect of total petroleum hydrocarbon (TPH) pollution on the abundance and occurrence of 12 common taxa present in the microcosms.
The bars illustrate the average number of individuals per microcosm and lines represent the proportion of microcosms per treatment in which
that taxon was present. The asterisks indicate data sets are significantly ( p , 0.01) different from one another.
2506 Environ. Toxicol. Chem. 24, 2005
Fig. 4. The cumulative numbers of Cricotopus albitarsis adults that
emerged from treatments 1, 2, and 3 and unpolluted sediment (treat-
ment 0). Day 0 5 day when microcosms were covered with a mesh
to prevent further aerial colonization by insects. Treatment 0 (l);
Treatment 1 (- v -); Treatment 2 (m); Treatment 3 (m).
microcosms were closed, whereas no emergence occurred in
treatment 3 until 14 d afterwards. The difference in the time
required for adults to first emerge from treatments 0, 1, and
2 and treatment 3 was significant (MannWhitney U test, p
5 0.025), but the differences in median emergence times be-
tween treatments were not significant. This delayed first emer-
gence of adults in treatment 3 indicates C. albitarsis in treat-
ment 3 were stressed from HMWH polluted sediments, al-
though the numbers of adults that emerged from treatment 3
was similar to those in treatments 0 to 2.
Critical TPH sediment concentrations
Some taxa (C. albitarsis, P. vespertinus, L. albiceps, and
T. fuscithorax) appeared to have an advantage with low-level
TPH sediment pollution (860 mg/kg in treatment 1). The only
indication that any of the common taxa were stressed by a
TPH sediment concentration of 860 mg/kg was that signifi-
cantly fewer P. grimmi adults emerged from treatments 1 and
2 than from treatment 0. At a TPH concentration of 1,858 mg/
kg (treatment 2), the abundance of some species (P. villosi-
manus and K. martini) declined significantly compared with
nonpolluted and less polluted sediments. Total petroleum hy-
drocarbon sediment concentrations of 4,632 mg/kg (treatment
3) produced significantly lower total numbers of taxa and in-
dividuals, and significant declines in the abundance of C. cur-
tivalva, P. villosimanus, P. watsoni, P. vespertinus, L. albi-
ceps, and T. fuscithorax. Only small numbers of a few taxa
were able to inhabit microcosms with TPH sediment concen-
trations of 14,266 mg/kg (treatment 4), thus indicating that
this level of pollution severely affects aquatic ecosystems.
TPH concentrations in urban sediments
The TPH concentrations present in the treatments used in
this experiment are similar to those found in many urban water
bodies in Melbourne, Australia (Table 4).
More than 25% of sediments from water bodies in the Mel-
bourne region exceed the concentrations present in treatment
1, whereas ,5% exceed levels present in treatment 3. A high
percentage (34.8%) of urban drain sediments exceed TPH con-
centrations present in treatment 1, but few drains had concen-
trations exceeding those present in treatment 3. A larger pro-
portion of wetlands (including lakes) than drains or streams
had TPH sediment concentrations exceeding those found in
treatments 3 and 4.
V. Pettigrove and A. Hoffmann
Table 4. The percentage of urban wetlands, drains, and streams in
Melbourne, Australia, with total petroleum hydrocarbon (TPH)
concentrations present in the sediments that exceed the TPH
concentrations in the treatments used in the field-based experiment
Water bodies affected (%)
Treatment
No. TPH (mg/kg) Streams Drains Wetlands Total
1 .860 23.1 34.8 26.4 27.6
2 .1,858 6.7 19.6 14.9 14.2
3 .4,632 2.9 1.8 6 4.3
4 .14,266 0 0 1.1 0.6
n 104 112 269 485
DISCUSSION
Ecological implications
Urban water bodies are affected by a variety of pollutants,
including heavy metals [31,32], hydrocarbons [1,3], pharma-
ceuticals, hormones, and other organic wastewater chemicals
[33]. Polluted urban aquatic sediments could have a direct
effect on the occurrence and abundance of macroinvertebrates
in water bodies in urban areas [19]. To improve the ecological
health of these systems, it is important to identify the primary
pollutants affecting the biota so that prudent pollution man-
agement strategies can be developed. This is the first study to
use the field-based microcosm method developed by Pettigrove
and Hoffmann [19] that isolates the effect of one type of sed-
iment pollution (HMWHs) on macroinvertebrate occurrence
and abundance.
The synthetic motor oil added to clean sediments produced
a form of hydrocarbon pollution similar to that in urban streams
containing low levels of heavy metals and PAHs, with sedi-
ment quality guidelines indicating that harmful effects from
PAHs are unlikely. Furthermore, these sediments contained
low concentrations of small-chain carbon (C6C9) substances
and a large proportion of HMWHs (C15C36).
Caution must be taken when comparing the results of the
field-based microcosm experiment, in which only synthetic
motor oil was tested, with TPH concentrations measured in
urban sediments. Total petroleum hydrocarbon concentration
data cannot be used to quantitatively estimate environmental
conditions because sediments with the same TPH concentra-
tion can represent very different compositions and very dif-
ferent risks to the environment. Furthermore, some TPH meth-
ods are subject to interference from naturally occurring ma-
terials, such as animal and vegetable oils, which could result
in artificially high reported TPH concentrations [34]. Synthetic
motor oil containing polyalphaolefins was considered a good
indicator of TPHs in sediments because their low bioavail-
ability and toxicity [23] would provide a conservative estimate
of the potential effects of petroleum hydrocarbon pollution in
freshwater sediments. Further field-based experiments are be-
ing conducted to assess the effects of other HMWHs (mineral
oil, diesel, and biodiesel) on benthic macroinvertebrates. It is
anticipated that these studies will help to confirm which TPH
concentrations in the field might suggest ecological impair-
ment in freshwater ecosystems.
Given the results of this study with synthetic motor oil, it
is predicted that TPH sediment concentrations of 860 mg/kg
in urban bodies of freshwater could impair resident ecosys-
tems. Low-level (860 mg/kg) TPH-polluted sediments might
increase abundance of opportunistic species, whereas TPH
concentrations between 860 and 1,870 mg/kg are likely to
Effects of hydrocarbons on macroinvertebrates
reduce the abundance of TPH pollutionsensitive taxa. Total
petroleum hydrocarbon concentrations .1,870 mg/kg are pre-
dicted to lead to more substantial losses in species presence
and abundance.
The loss of macroinvertebrates, as occurred in treatments
2, 3, and 4, would reduce the amount of food available for
fish and other animals that feed on macroinvertebrates. For
example, a nonbreeding adult platypus consumes on the order
of 15 to 28% of its body mass per day to accommodate a daily
energy expenditure of approximately 850 to 1,100 kg/d [35],
with the food intake of a female in late lactation rising to about
three times her daily consumption when not lactating [36]. The
availability of macroinvertebrates could be a major factor lim-
iting their distribution in urban streams [37]. Therefore, it is
hypothesized that HMWH pollution (where TPH sediment
concentrations exceed between 860 and 1,860 mg/kg) would
severely affect predatory organisms that have a direct or in-
direct reliance on macroinvertebrates as a source of food.
How do HMWHs affect macroinvertebrates?
The toxic effects of hydrocarbons is usually evaluated in
terms of the presence of PAHs because these substances are
considered to be the most toxic substances present (e.g., [6
8]). However, the presence of relatively nontoxic HMWHs
could influence organisms in several ways. Direct coating with
oil does not biochemically interfere with cellular activities.
The primary effects are smothering and mechanical interfer-
ence with activities such as movement and feeding [14]. Phys-
ical alterations to sediments from petroleum hydrocarbon pol-
lution might also include increased sediment oxygen demand
from microbial activity, a variety of chemical oxidation pro-
cesses, and a subsequent reduction in the aerobic layer in
sediments [38]. Increased organic enrichment can also occur
as a result of sediments being polluted with hydrocarbons [15],
and this might have led to the increased abundance of oppor-
tunistic species in treatment 1.
The findings of this study of HMWH pollution in freshwater
sediments must be applied with caution to estuarine and marine
systems. A number of factors can contribute to the greater
susceptibility of freshwater benthic ecosystems to the effects
of HMWH pollution than benthic marine ecosystems. Biotur-
bation in a nonpolluted marine environment effectively mixes
the surface sediment layer about 10 cm deep [39], whereas
the freshwater macroinvertebrate infauna is comparatively
small and generates little bioturbation of sediments. High
molecular weight hydrocarbons adsorbed to sediments are
more likely to accumulate in anoxic sediments, in which they
have lower biodegradation rates [40]. Therefore, HMWHs are
more likely to accumulate in freshwater than in marine sedi-
ments. Highmolecular weight hydrocarbons might also be
more widely dispersed in an open marine environment than
in small urban water bodies, resulting in a lower propensity
for TPH concentrations in marine sediments to reach concen-
trations present in urban sediments. For example, between Jan-
uary 19 and 30, 1991, an estimated 10.8 million barrels of oil
were spilled by Iraqi troops into the Arabian Gulf near the
Gulf of Kuwait. The most heavily polluted sediments ranged
from 266 to 1,448 mg/kg [41], which is considerably less than
what is often present in urban bodies of freshwater in Mel-
bourne, Australia.
Implications for sediment quality guidelines
Highmolecular weight hydrocarbons are common pollut-
ants in the sediments of urban bodies of freshwater. If the TPH
Environ. Toxicol. Chem. 24, 2005 2507
concentrations in the field-based experiment have a similar
effect to those in the field, at least 28% of water bodies in
Melbourne, Australia, are likely to be affected by HMWH
pollution in sediments, and this percentage could be even high-
er for two reasons. First, it is possible that TPH concentrations
lower than 860 mg/kg (the lowest concentration tested in this
experiment) could affect other aquatic macroinvertebrates.
Second, changes in the occurrence and abundance of species
appear to be less sensitive measures of stress than the use of
metrics, such as numbers of adults that emerge from micro-
cosms and growth rates (e.g., adult emergence rates/devel-
opment rates). More severe ecological impacts (reduced abun-
dance of HMWH pollutionsensitive species and reduced spe-
cies richness) are predicted to occur in 14.2% of urban water
bodies in Melbourne.
Even although HMWH-polluted sediments are widespread
in urban water bodies, no sediment quality guidelines for the
protection of aquatic ecosystems exist for these substances
because they are nontoxic [10,13]. However, because we
have demonstrated that sediments polluted with nontoxic
HMWHs affect freshwater macroinvertebrate assemblages, the
development of freshwater sediment quality guidelines for
these substances is warranted. The concentrations of TPHs
provide a good indication of the potential effects these con-
taminants have on aquatic ecosystems. A better indicator of
HMWH pollution might become available once the ways that
substances affect macroinvertebrate assemblages have been
identified and once further work is conducted to determine the
effects of mineral oil and other common petroleum hydrocar-
bon pollutants on benthic macroinvertebrates.
CONCLUSION
Highmolecular weight hydrocarbons are widespread pol-
lutants in urban bodies of freshwater. The aquatic macroin-
vertebrates in the field-based experiments were affected by
HMWHs as low as 840 mg/kg. If the TPH concentrations
measured in the field have properties similar to the synthetic
oil used in the experiments, then about 28% of aquatic eco-
systems in urban water bodies in Melbourne, Australia, could
be impaired by HMWH pollution in sediments. A TPH con-
centration of 840 mg/kg in sediments might be a reasonable
interim guideline value to indicate possible ecological im-
pairment. Sediment quality guidelines for just PAHs provide
an inadequate indication of whether petroleum hydrocarbon
polluted sediments are affecting the health of freshwater eco-
systems. Watershed management authorities should recognize
that HMWH-polluted sediments can affect aquatic ecosystem
health and that appropriate actions need to be taken to mitigate
the presence of these pollutants in urban freshwater systems.
AcknowledgementWe are grateful to Jacqueline Ward, Renae Ayres,
and David Sharley, who assisted with field work; Andrew Weeks for
comments on the draft manuscript; and Parks Victoria (Kew, Victoria),
which provided access to Glynns Wetland. Supported by the Austra-
lian Research Council via the Special Research Centre scheme.
REFERENCES
1. Muschack W. 1990. Pollution of street run-off by traffic and local
conditions. Sci Total Environ 93:419431.
2. National Academy of Sciences. 1985. Oil in the SeaInputs,
Fates and Effects. National Academy, Washington, DC, USA.
3. Fam S, Stenstrom MK, Silverman G. 1988. Hydrocarbons in ur-
ban runoff. J Environ Eng ASCE 113:10321046.
4. Yamane A, Nagashima I, Okubo T, Okada M, Murakami A. 1990.
Stormwater runoff of hydrocarbons in the Tama River basin in
2508 Environ. Toxicol. Chem. 24, 2005
Tokyo (Japan) and their fate in the river. Water Sci Technol 22:
119126.
5. Olah GA, Molnar A. 1995. Hydrocarbon Chemistry. John Wiley,
New York, NY, USA.
6. Tay K, Doe KG, Wade SJ, Vaughan DA, Berrigan RE, Moore
MJ. 1992. Sediment bioassessment in Halifax Harbour. Environ
Toxicol Chem 11:15671581.
7. Pelletier MC, Burgess RM, Ho KT, Kuhn A, McKinney RA, Ryba
SA. 1997. Phototoxicity of individual PAHs and petroleum to
marine invertebrate larvae and juveniles. Environ Toxicol Chem
16:21902199.
8. Beasley G, Kneale P. 2002. Reviewing the impact of metals and
PAHs on macroinvertebrates in urban watercourses. Prog Phys
Geogr 26:236270.
9. Barron MG, Podrabsky T, Ogle S, Ricker RW. 1999. Are aromatic
hydrocarbons the primary determinant of petroleum toxicity to
aquatic organisms? Aquat Toxicol 46:253268.
10. Peterson DR. 1994. Calculating the aquatic toxicity of hydro-
carbon mixtures. Chemosphere 29:24932506.
11. Salanitro JP, Dorn PB, Huesemann MH, Moore KO, Rhodes IA,
Rice JLM, Vipond TE, Western MM, Wisniewski HL. 1997.
Crude oil hydrocarbon bioremediation and soil ecotoxicity as-
sessment. Environ Sci Technol 31:17691776.
12. DeLaune RD, Lindau CW, Banker BC, Devai J. 2000. Degra-
dation of petroleum hydrocarbons in sediment receiving produced
water discharge. J Environ Sci Health A 35:114.
13. Payne JF, Fancey LL, Hellou J, King MJ, Fletcher GL. 1995.
Aliphatic hydrocarbons in sediments: A chronic toxicity study
with winter flounder (Pleuronectes americanus) exposed to oil
well drill cuttings. Can J Fish Aquat Sci 52:27242735.
14. Moore SF, Dwyer RL. 1974. Effects of oil on marine organisms:
A critical assessment of published data. Water Res 8:819827.
15. Spies RB, Hardin DD, Toal JP. 1988. Organic enrichment or tox-
icity? A comparison of the effects of kelp and crude oil in sed-
iments on the colonization and growth of benthic fauna. J Exp
Mar Biol Ecol 124:261282.
16. Long ER, MacDonald DD, Smith S, Calder FD. 1995. Incidence
of adverse biological effects within ranges of chemical concen-
trations in marine and estuarine sediments. Environ Manag 19:
8197.
17. Swartz RC. 1999. Consensus sediment quality guidelines for
polycyclic aromatic hydrocarbon mixtures. Environ Toxicol
Chem 18:780787.
18. MacDonald DD, Ingersoll CG, Berger TA. 2000. Development
and evaluation of consensus-based sediment quality guidelines
for freshwater ecosystems. Arch Environ Contam Toxicol 39:20
31.
19. Pettigrove V, Hoffmann A. 2005. A field-based microcosm meth-
od to assess the effects of polluted urban stream sediments on
aquatic macroinvertebrates. Environ Toxicol Chem 24:170180.
20. Jordan RE, Payne JR. 1980. Fate and Weathering of Petroleum
Spills in the Marine Environment. Ann Arbor Science, Ann Ar-
bor, MI, USA.
21. Cram S, Siebe C, Ortz-Salinas R, Herre A. 2004. Mobility and
persistence of petroleum hydrocarbons in peat soils of south-
eastern Mexico. Soil Sediment Contam 13:439458.
22. Suderman K, Thistle D. 2004. The relative impact of spills of
two alternative fuels on the microalgae of a sandy site: A micro-
cosm study. Mar Pollut Bull 49:473478.
23. Candler J, Hebert R, Leuterman AJJ. 1977. Effectiveness of a
10-day ASTM amphipod sediment test to screen drilling mud
V. Pettigrove and A. Hoffmann
base fluids for benthic toxicity. SPE 37890. Proceedings, 1997
SPE/EPA Exploration and Production Environmental Conference,
Dallas, TX, USA, March 35, 1977, pp 6382.
24. Edward DHD. 1963. The biology of a pathogenetic species of
Lundstroemia (Diptera: Chironomidae), with descriptions of the
immature stages. Proc R Entomol Soc Lond A 38:165170.
25. Anderson NM, Weir TA. 2004. Australian Water Bugs. Their
Biology and Identification (Hemiptera-Heteroptera, Gerromor-
pha & Nepomorpha). Apollo Books, Stenstrup, Denmark.
26. Wells A. 1997. A preliminary guide to the identification of larval
Hydroptilidae (Insecta: Trichoptera). Guide 13. Cooperative Re-
search Centre for Freshwater Ecology, Albury, NSW, Australia.
27. Dean JC, Suter PJ. 1996. Mayfly nymphs of Australia. A guide
to genera. Guide 7. Cooperative Research Centre for Freshwater
Ecology, Albury, NSW, Australia.
28. Carew ME, Pettigrove V, Hoffmann AA. 2003. Identifying chi-
ronomids (Diptera: Chironomidae) for biological monitoring with
PCR-RFLP. Bull Entomol Res 93:483490.
29. Cairns J Jr. 2003. Biotic community response to stress. In Simon
TP, ed, Biological Response Signatures: Indicator Patterns Using
Aquatic Communities. CRC, Boca Raton, FL, USA, pp 1321.
30. Newman MC. 1998. Fundamentals of Ecotoxicology. Ann Arbor,
Chelsea, MI, USA.
31. Crunkilton R, Kleist J, Ramcheck J, DeVita W, Villeneuve D.
1997. Assessment of the response of aquatic organisms to long-
term in-situ exposures to urban-runoff. In Roesner LA, ed, Effects
of Watershed Development and Management on Aquatic Eco-
systems. American Society of Civil Engineers, New York, NY,
pp 1833.
32. Pettigrove V, Hoffmann A. 2003. Impact of urbanisation on heavy
metal contamination in urban stream sediments: Influence of
catchment geology. Aust J Ecotoxicol 9:199128.
33. Kolpin DW, Furlong ET, Meyer MT, Thurman EM, Zaugg SD,
Barber LB, Buxton HT. 2002. Pharmaceuticals, hormones, and
other organic wastewater contaminants in U.S. streams, 1999
2000: A national reconnaissance. Environ Sci Technol 36:1202
1211.
34. Weisman W. 1998. Total Petroleum Hydrocarbon Criteria Work-
ing Group Series Volume 1: Analysis of Petroleum Hydrocar-
bons in Environmental Media. Amherst Scientific, Amherst, MA,
USA.
35. Munks SA, Otley HM, Bethge P, Jackson J. 2000. Reproduction,
diet and daily energy expenditure of the platypus in a sub-alpine
Tasmanian lake. Aust Mammal 21:260261.
36. Holland N, Jackson SM. 2002. Reproductive behaviour and food
consumption associated with the captive breeding of platypus
(Ornithorhynchus anatinus). J Zool (London) 256:279288.
37. Serena M, Pettigrove V. 2005. Relationship of sediment toxicants
and water quality to the distribution of platypus populations in
urban streams. J N Am Benthol Soc (in press).
38. Won Sik Shin, Pardue JH, Jackson WA. 2000. Oxygen demand
and sulfate reduction in petroleum hydrocarbon contaminated salt
marsh soils. Water Res 34:13451353.
39. French-McCay DP. 2004. Oil spill impact modelling: Develop-
ment and validation. Environ Toxicol Chem 23:24412456.
40. Leahy JG, Colwell RR. 1990. Microbial degradation of hydro-
carbons in the environment. Microbiol Rev 54:305315.
41. Massoud MS, Al-Abdali F, Al-Ghadban AN, Al-Sarawi M. 1996.
Bottom sediments of the Arabian GulfII. TPH and TOC con-
cepts as indicators of oil pollution and implications for the effect
and fate of the Kuwait oil slick. Environ Pollut 93:271284.