Environmental Pollution 225 (2017) 223e231

Contents lists available at ScienceDirect

Environmental Pollution
journal homepage: www.elsevier.com/locate/envpol

Microplastic contamination in Lake Winnipeg, Canada*
Philip J. Anderson a, 1, Sarah Warrack b, Victoria Langen c, Jonathan K. Challis d,
Mark L. Hanson e, Michael D. Rennie a, b, c, *
a

IISD-Experimental Lakes Area, 111 Lombard Ave. Suite 325, Winnipeg, MB R3B 0T4, Canada
Department of Biological Sciences, University of Manitoba, 50 Sifton Road, Winnipeg, MB R3T 2N2, Canada
c
Department of Biology, Lakehead University, 955 Oliver Road, Thunder Bay, ON P7B 5E1, Canada
d
Department of Chemistry, University of Manitoba, 360 Parker Building, Winnipeg, MB R3T 2N2, Canada
e
Department of Environment and Geography, University of Manitoba, 127 Dysart Road, Winnipeg, MB R3T 2N2, Canada
b

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 12 September 2016
Received in revised form
27 January 2017
Accepted 6 February 2017

Microplastics are an emerging contaminant of concern in aquatic ecosystems. To better understand
microplastic contamination in North American surface waters, we report for the ﬁrst time densities of
microplastics in Lake Winnipeg, the 11th largest freshwater body in the world. Samples taken 2014 to
2016 revealed similar or signiﬁcantly greater microplastic densities in Lake Winnipeg compared with
those reported in the Laurentian Great Lakes. Plastics in the lake were largely of secondary origin,
overwhelmingly identiﬁed as ﬁbres. We detected signiﬁcantly greater densities of microplastics in the
north basin compared to the south basin of the lake in 2014, but not in 2015 or 2016. Mean lake-wide
densities across all years were comparable and not statistically different. Scanning electron microscopy with energy dispersive X-ray spectroscopy indicated that 23% of isolated particles on average were
not plastic. While the ecological impact of microplastics on aquatic ecosystems is still largely unknown,
our study contributes to the growing evidence that microplastic contamination is widespread even
around sparsely-populated freshwater ecosystems, and provides a baseline for future study and risk
assessments.
© 2017 Elsevier Ltd. All rights reserved.

Keywords:
Exposure assessment
Plastic ﬁbres
Great Lakes
Contaminant trends

1. Introduction
Microplastic particles have received signiﬁcant attention
recently as an emerging contaminant of concern in Canada and
globally (Eriksen et al., 2013; Ivar do Sul and Costa, 2014; Anderson
et al., 2016). Microplastics are deﬁned as fragments of plastic that
are smaller than 5 mm in any dimension (MSFD GES Technical
Subgroup on Marine Litte, 2013) and have been identiﬁed as a
possible threat to aquatic ecosystems, with the potential for
detrimental impacts on human health (Van Cauwenberghe and
Janssen, 2014). They are considered an emerging contaminant, as
research into their potential risk in aquatic ecosystems and human
health are still not clear (Drewes and Shore, 2001; Younos, 2005).

*

This paper has been recommended for acceptance by Maria Cristina Fossi.
* Corresponding author. Department of Biology, Lakehead University, 955 Oliver
Road, Thunder Bay, ON P7B 5E1, Canada.
E-mail address: mrennie@lakeheadu.ca (M.D. Rennie).
1
Current address: Department of Biology, University of Saskatchewan, 112 Science Road, Saskatoon, SK, S7N 5E2, Canada.
http://dx.doi.org/10.1016/j.envpol.2017.02.072
0269-7491/© 2017 Elsevier Ltd. All rights reserved.

Recent concern around the potential ecological impacts of microplastics has led to proposed legislation to eliminate their use in the
manufacturing of personal care products in the United States by
July 1, 2017 (United States Congress, 2015). In Canada, they were
listed recently as a toxic substance under Schedule 1 of the Canadian Environmental Protection Act (Canada Gazette, 2016).
Australia and the UK have enacted “voluntary” elimination by industry as early as 2017, and many manufacturers have begun
removing them from their products in advance of legislative action.
Microplastics have been documented in water bodies worldwide. In marine ecosystems, microplastic contamination has been
reported around the world in sediment, open water, and in organisms (Ivar do Sul and Costa, 2014). Observations in freshwater
are more rare and have been identiﬁed as a signiﬁcant data gap,
especially in Canada (Anderson et al., 2016). The Laurentian Great
Lakes have recently been shown to have signiﬁcant densities of
microplastics in surface waters (Eriksen et al., 2013), as have other
more remote lakes in Asia (Free et al., 2014) and Europe (Faure et al.,
2015). In addition, microplastics have been reported in sediments
~ eda et al., 2014;
of Lake Ontario and the St. Lawrence river (Castan

 224

P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Corcoran et al., 2015; Ballent et al., 2016). These studies on larger
systems beg the question as to the extent of microplastic contamination in other inland lakes of similar size (Wagner et al., 2014).
Within marine and freshwater food webs, microplastics have been
detected in the guts of organisms at nearly every trophic level, from
copepods and crustacean zooplankton to ﬁlter-feeding invertebrates, and in pelagic and demersal ﬁshes (Thompson et al.,
2004; Eerkes-Medrano et al., 2015; Lusher et al., 2013). It is hypothesized that microplastics can affect the physiological functioning of animals, either through leaching of accumulated organic
pollutants into the stomach lining of ﬁshes that consume them,
physical blockage of the digestive system, or simply by taking up
space that could otherwise be occupied by food (Bakir et al., 2014;
Wright et al., 2013). If microplastics are capable of facilitating bioaccumulation and/or biomagniﬁcation of harmful contaminants
(Koelmans et al., 2013), then there is potential for adverse effects on
humans, especially for the one ﬁfth of the world's population that
relies on ﬁsh and seafood as their primary animal protein source
(Sumaila et al., 2007).
There are two sources of the microplastics commonly found in
aquatic systems: primary and secondary microplastics (MSFD GES
Technical Subgroup on Marine Litter, 2013). Primary microplastics
are those that were produced for use in a wide variety of consumer
and industrial applications (e.g., abrasives in cosmetic scrubs).
Secondary microplastics originate from larger plastic and synthetic
materials when they are broken down by weathering processes
such as UV degradation or machine washing (MSFD GES Technical
Subgroup on Marine Litter, 2013). In marine systems, secondary
microplastics make up the bulk of the microplastic particles,
especially in areas with high population densities (Hidalgo-Ruz
et al., 2012; Browne et al., 2011). Recent studies have demonstrated that wastewater treatment plants are signiﬁcant release
points of microplastic particles, primarily ﬁbres (Mason et al., 2016;
Anderson et al., 2016; Murphy et al., 2016). The relative importance
of primary and secondary microplastics in freshwater systems is
not well documented; primary plastics (“pellets” or microbeads)
dominated the smallest size category (<1 mm) in the Great Lakes,
whereas secondary plastics (fragments) dominated larger size
classes (>1 mm; Eriksen et al., 2013). Many common forms of
microplastics in consumer products (e.g., polypropylene, high/low
density polyethylene) have densities less than water (EerkesMedrano et al., 2015), and polypropylene and polyethylenes have
been shown to be highly susceptible to microparticle generation
through breakdown (Weinstein et al., 2016; Zbyszewski and
Corcoran, 2011). As such, these types of microplastics have the
potential to be generated by larger plastic litter, as well as be carried
long distances through watersheds owing to their densities, making their study in Lake Winnipeg especially relevant.
Lake Winnipeg is the ﬁfth largest Canadian lake, and has the
second largest watershed in Canada at over 982,000 km2, in
which nearly 7 million people live and work (roughly 20% of
the Canadian population). The watershed is approximately 40x
larger than the surface of the lake, and transports water from
across four Canadian provinces and four US states, making the
condition of the lake very much a function of the events that
occur within the watershed (Schindler, 2009). Our objective
was to report initial estimates of microplastic contamination in
Lake Winnipeg, and formally compare these results to reported
concentrations in other large Canadian Lakes (Eriksen et al.,
2013). This work serves an important ﬁrst step towards ﬁlling
the literature gap around sources and distribution of microplastics in Canadian freshwater ecosystems, and understanding
the comparative extent of microplastic pollution across large
freshwater bodies in Canada.

2. Methods
Lake Winnipeg is a shallow (mean depth 12 m), well-mixed (3to
5-year residence time) lake with low visibility (Secchi depths
typically < 1 m). It is subject to wind mixing and frequent algal
blooms due to cultural eutrophication (Waasenaar and Rao, 2012).
Surface waters during relatively calm conditions (winds between 4
and 18 knots) were sampled at twelve locations (Fig. 1) in Lake
Winnipeg between July 25th and October 2nd, 2014; from June 2nd
to June 17th, 2015; and from June 3rd to June 19th, 2016 using
standard collection methods (Eriksen et al., 2013). Speciﬁcally, a
manta trawl (61 cm wide by 18 cm high) was employed with a 3 m
long, 333 mm mesh bag and 333 mm removable cod-end. The net
was trawled along side the research vessel MV Namao using a ﬁxed
crane arm. Tow times ranged between eight and thirty minutes.
Because sampling primarily occurred during the summer months
when algal blooms are common in Lake Winnipeg, the net was
closely monitored during towing to ensure the net did not clog. If
the net appeared to be nearing a state where water ﬂow through
the net was becoming impeded, the tow haul was stopped and the
net was retrieved. The distance sampled by the net was estimated
as the distance travelled over the GPS start and end points of the
ship, as recorded from the on-board ship computer. Tow speeds
were between 2.2 and 3.75 knots. Once the net was retrieved, all
collected material was concentrated into the cod-end of the net and
transferred to a glass sample jar. The net (with the cod-end
removed) and cod-end separately were rinsed thoroughly with
water from the ship's hose between deployments to avoid crosscontamination. The collected material was then preserved in 70%
ethanol until processing in the laboratory. Trawl sites were chosen
opportunistically to coincide with existing sampling sites visited by
the MV Namao, but also selected to represent a range of offshore
and nearshore sites in both the north and south basins of the lake.
Trawls varied from 530 m to 3780 m in length. Surface densities of
microplastics were estimated from the trawl distance multiplied by
the width of the manta trawl net.
In the laboratory, samples were ﬁltered through a 250 mm mesh
brass sieve. Coarse debris (e.g., hand-sized twigs or leaves) were
thoroughly rinsed with deionized water (DI) and removed; typical
rinsing time was approximately ﬁve minutes per sample. The
remaining material on the sieve was then reconstituted to a known
volume (typically 1250 mL). Rinsing of large debris in this manner
may have resulted in some losses of strongly-attached plastic particles, therefore our results might be considered conservative estimates. The reconstituted sample was stirred using a stirring plate
to achieve uniform consistency. A subsample of known volume
(typically 250 mL, or 1/5 of the reconstituted sample) was collected
by submerging the subsample vessel vertically in the centre of the
bulk sample and removing it vertically. The subsample was then
processed using a wet peroxide oxidation (WPO) treatment
(Masura et al., 2015; Mason et al., 2016). Each 250 mL sub-sample
received 20 mL of a 0.05 M Fe (II) solution and 20 mL of 30%
H2O2 while on a 75   C stirring hotplate for 30 min and was covered
loosely with tinfoil to avoid atmospheric deposition during the
reaction (but allowing the sample to de-gas). The Fe (II) solution
was prepared by adding 7.5 g of FeSO4$H2O to 500 mL of water and
3 mL of concentrated sulfuric acid. Additional 20 mL aliquots of 30%
H2O2 were added to the subsample as needed throughout the
stirring and heating process to help degrade organic material in the
samples. Subsamples were left covered to digest for 24 h. Digested
samples were once again ﬁltered through a 250 mm brass sieve,
reconstituted in DI water and were processed. During these procedures, synthetic ﬁbres were not worn by the individuals handling
samples. We visually identiﬁed particles remaining from these
samples that were putatively considered to be plastic based on

 P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

225

Fig. 1. Sites sampled for microplastics on Lake Winnipeg, Canada.

shape and size using a dissecting microscope (6.3e50.0x magniﬁcation); particles were all less than 5 mm and typically larger than
the mesh size of our net (333 mm) in at least one dimension. A count
of the number and type of microplastic particles was recorded.
Microplastic particles were transferred to ethanol in a glass vial
with a rubber stopper for later scanning electron microscopy (SEM)
identiﬁcation. Our search pattern for types of microplastics were
categorized into ﬁve groups: fragments (hard, jagged-edged particles), micro-pellets (hard, rounded particles), ﬁbres (ﬁbrous or
thin uniform plastic strands), ﬁlms (thin, 2-dimensional plastic
ﬁlms), and foam (i.e., Styrofoam-type material).
We evaluated our subsampling protocols by conducting two
subsamples in duplicate (sample W9 and 59, 2014). In each case,
total visual counts of plastics were similar (sample W9: 49 and 59
particles; sample 59: 54 and 70 particles).
Laboratory blanks were conducted to determine possible
contamination from either deposition from air or DI water. DI water
blanks were conducted by running the DI water tap at the University of Manitoba for 60 min on a clean 333 mm brass sieve which
delivered water at a rate of 8 L per minute (480 L total). Four replicates of this procedure yielded counts of 13, 5, 16 and 9 particles
(ﬁbers), suggesting that on average 1 ﬁber is introduced for every
48 L of DI water applied to samples. Given an average rinse of ﬁve
minutes with DI water (at 8L per minute), plus a reconstitution to
1.25 L prior to subsampling, we estimate that 0.85 ﬁbers were
introduced to our samples, on average, from DI water alone. Air

blanks were conducted by leaving a clean beaker of water out on
the lab counter for 24 h. Duplicate air blanks yielded 8 and 7 ﬁbers
over this time period, or 0.3 ﬁbers per hour. Average time for visual
sorting of samples was 4 h. We estimate 1.25 on average were
introduced from lab air. Therefore, 2 ﬁbers per sample were likely
introduced due to laboratory handling procedures, and all sample
counts were reduced by this value.
A subset of 2014 samples (9 of 12 total) identiﬁed visually as
plastic were examined for their elemental composition using SEM
and Energy Dispersive X-ray Spectroscopy (EDS; following Eriksen
et al., 2013). Between 8 and 32 particles from each sample were
chosen at random and placed on double-sided carbon tape, coated
with a thin ﬁlm of evaporated carbon under vacuum and then
imaged using a Hitachi SU-70 ﬁeld emission SEM operating at
20 KV in backscatter mode. Qualitative elemental composition of
particles was conﬁrmed using an Oxford AZtec Energy Dispersive
X-ray Spectroscopy system (EDS). We used both point and areal EDS
scans of samples, as appropriate (based largely on sample size), to
determine elemental composition. To help characterize particles as
plastic or not, we also analyzed a number of known plastic samples
(shopping bag, sealable plastic bag, Styrofoam, plastic fork, plastic
ﬁlm), weathered plastic samples collected from Lake Tamblyn,
Ontario (shopping bag, water bottle, sealable plastic bag) and
rubber stoppers used to hold sorted plastic particles (see
Supplemental Information). We also characterized samples of
cultured Daphnia magna obtained from the Lakehead University

 226

P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Aquatic Toxicology Research Centre, to ensure that organic particles
could be differentiated from plastics (see Supplemental
Information). Daphnia was chosen because cladoceran carapaces
proved to be the most resistant biological material to WPO treatments in our samples; as such, we wished to be able to differentiate
carapaces (or fragments thereof) from putative plastic particles. We
corrected microplastic counts from all samples based on the proportion of non-plastic particles identiﬁed using SEM-EDS.
Statistical analyses were performed using R (ver. 3.2.3, R core
team 2015). Visualization of microplastic density in the lake was
mapped using the packages ‘ggmap’ and ‘ggplot2’ (Kahle and
Wickham, 2013; Wickham, 2009). The density of microplastic
contamination between the north and south basins over years was
examined using two-way ANOVA. We used two-way ANOVA to
compare differences between nearshore and offshore sites over
years (Table 1), and between sites near northern (site 28) and
southern inﬂows (sites 2, 3B and 7) over years of sampling.
Microplastic densities were log-transformed to satisfy the
assumption of normality. Finally, we compared densities of
microplastics in Lake Winnipeg to those reported in the Great Lakes
(Eriksen et al., 2013, their Table 1) using a Kruskal-Wallis test on
ranks followed by Dunn's test for multiple comparisons with a
Bonferroni adjustment.
3. Results
Microplastic particles were found in all samples collected, in all
years (Fig. 2). On average, 23% of particles visually identiﬁed as
plastic were determined by SEM-XDS to be either silicates, iron
oxide (rust) or paint ﬂakes off the vessel (Supplemental
Information). All sample counts across years were adjusted using
this value (i.e., total densities x 0.77). Adjusted densities of blankcorrected microplastics ranged from 748,000 particles/km2 at station 22 near the outﬂow of Lake Winnipeg in 2014 to 53,000 particles/km2 at station 7 in the Winnipeg River Inﬂow in 2014
(Table 1, Fig. 2). The most common type of particles across all sites
were overwhelmingly identiﬁed as ﬁbres, and ﬁlms and foam were
the least common over all three years of data (Fig. 3). No micropellets or beads were identiﬁed over the three years of sampling
conducted over all 12 stations on the lake.
There was a signiﬁcant interaction between year of collection
and basin (north and south) for microplastic densities in Lake
Winnipeg (two-way ANOVA: F2,29 ¼ 3.74, p ¼ 0.036). Densities of
microplastics were signiﬁcantly greater in the north basin than in
the south basin in 2014 (Tukey HSD test, p < 0.05), but not in 2015
or 2016 (p > 0.05, Fig. 4). We found no evidence of differences
between nearshore and offshore sites across all years (p > 0.05).

Similarly, there were no statistically signiﬁcant differences between
microplastic densities among southern and northern sites nearest
inﬂows across years (two-way ANOVA, p > 0.05). There was no
effect of wind velocity (linear regression, p > 0.05) or direction
(linear regression, p > 0.05) at the time of sample collection on our
microplastic densities.
Signiﬁcant differences in microplastic densities were found
between Lake Winnipeg and those previously reported in the Great
Lakes (Kruskal-Wallis test, Х2 ¼ 31.3, p < 0.0001; Fig. 5). Lake
Winnipeg microplastic densities were comparable to those reported in Lake Erie (p > 0.05), but greater than those reported for
Lake Huron (p < 0.0001), and in western Lake Superior (p ¼ 0.003).
4. Discussion
Microplastics are present in Lake Winnipeg at densities similar
to or greater than those currently reported in the Great Lakes basin.
Values were signiﬁcantly greater than those reported on Lakes
Superior and Huron, and similar to those reported for Lake Erie
(Eriksen et al., 2013). Lake Winnipeg also had the highest total
estimated quantity of microplastic (areal densities multiplied by
the lake surface area) compared with any of the Great Lakes surveyed (Table 2). However, comparisons of our data to those reported from the Great Lakes should be made with some caution, as
those published densities for the Great Lakes are based on fewer
samples, are from only a single year of sampling (compared to three
years of data reported here), and are from a limited range of sites
(e.g., Lake Superior samples are for Whiteﬁsh Bay only, and it is
unclear how representative those observations are for the entire
lake, especially in open water). Regardless, the similarity in
microplastic densities between Lake Winnipeg and Lake Erie is
surprising and should be of concern for those charged as stewards
of Lake Winnipeg. Lake Winnipeg has a surface area, mean depth
and residence time that is most similar to Lake Erie when compared
with either Lake Superior or Lake Huron (Table 2). However, the
Lake Erie catchment supports a population base of nearly double
that supported by Lake Winnipeg, in a watershed less than 1/10th
the size. If we assume that municipalities represent the major
source of microplastics in freshwater (Mason et al., 2016), then the
fewer municipalities feeding wastewater to Lake Winnipeg
compared to Lake Erie suggest that (a) municipalities in the Lake
Winnipeg watershed could be generating more microplastics than
those around Lake Erie, and that (b) long-range transport of
microplastics is facilitated on the surfaces of freshwater rivers.
Alternatively, other secondary sources of microplastics may be
present in Lake Winnipeg that are not found in Lake Erie, but this
seems unlikely given the intensity of industrial and commercial

Table 1
The location, microplastic density (blank-corrected), and description of Lake Winnipeg Research Consortium research stations used for this study. Latitude and longitude are
decimal degrees. n/s ¼ no sample taken.
Station

22
28
65
W1
W4
W6
W8
2
7
59
3B
W9

Densities (# per km2)
2014

2015

2016

748,027
231,239
249,208
193,153
225,852
n/s
283,132
108,034
52,508
188,186
161,893
174,438

140,814
191,598
262,436
184,941
266,001
169,475
69,167
139,467
187,805
157,787
215,514
95,628

121,597
186,276
221,902
169,562
293,449
66,788
98,084
279,161
171,741
217,921
148,505
98,416

Latitude

Longitude

Basin

Type

Description

53.61,286
53.21,349
52.17,458
53.38,208
52.89,907
51.62,056
51.77,651
50.43,604
50.68,053
50.69,880
50.48,764
51.02218

97.96,458
99.23,345
97.85,156
98.44,235
98.24,265
97.69,335
96.86,408
96.83,423
96.41,248
96.78,612
96.72,343
96.58,377

North
North
North
North
North
North
North
South
South
South
South
South

Nearshore
Nearshore
Offshore
Offshore
Offshore
Offshore
Offshore
Nearshore
Nearshore
Offshore
Nearshore
Offshore

Norway House (near outﬂow)
Grand Rapids
Near Dauphin River
Between Grand Rapids and outﬂow
North of Reindeer Island
East of reindeer island
Central Narrows
Red River (East)
Winnipeg River
Between Gimli and Victoria Beach
Red River (West)
Northern tip of south basin

 P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Fig. 2. Spatial distribution of microplastic density (number of particles/km2) across
sampling stations between 2014 and 2016 on Lake Winnipeg, Manitoba. Samples were
collected from eleven stations in 2014 and twelve stations in 2015 and 2016.

activity in Lake Erie relative to Lake Winnipeg. Our results certainly
warrant a closer investigation of municipal microplastic inputs and
studies into the long-range transport of these contaminants in large
river systems connected to Lake Winnipeg. Densities of

227

microplastics were also elevated relative to many Swiss lakes, but
similar to those reported for Lake Maggiore and Lake Geneva
(Grand Lac), both of which support much smaller populations in
their watersheds than Lake Winnipeg (Faure et al., 2015).
Densities of microplastics were signiﬁcantly elevated in the
north basin of Lake Winnipeg in 2014, but not in subsequent years.
It is possible that greater densities in the north basin observed in
2014 may be due to greater delivery rates from the inﬂow at Grand
Rapids during this year. The inﬂow at Grand Rapids delivers water
from the Saskatchewan River, which collects wastewater from all
major communities in Alberta and Saskatchewan. Considering only
major cities, (Edmonton, Calgary, Red Deer, Lethbridge and Medicine Hat in Alberta; Saskatoon in Saskatchewan), the Saskatchewan
River delivers water from municipalities totalling 2.3 million people, compared to just over 1 million from major centres in the Red
River (Regina, Brandon, Winnipeg, Grand Forks and Fargo) and
Winnipeg River drainages. The Winnipeg River ﬂows from its
headwaters in Northwestern Ontario to Lake Winnipeg, but does
not pass any major population centers, a common source of
microplastics (Leite et al., 2014). Typical discharges from the Saskatchewan River (mean monthly rate of 556 m3/s; Manitoba Water
Stewardship, 2011) amounts to approximately 25% of the total
inﬂow into the lake. The Winnipeg River (mean monthly rate of
1064 m3/s; Manitoba Water Stewardship, 2011) contributes almost
50% of the total inﬂowing water to the lake and the Red River is the
smallest of the three main tributaries delivering only 16% of the
total lake inﬂow (346 m m3/s; Manitoba Water Stewardship, 2011).
It is also possible that internal currents in 2014 may have also
acted to help concentrate microplastics in the north basin of Lake
Winnipeg. Very little is known about prevailing currents in Lake
Winnipeg, other than the intense action that wind can have on
mixing in the lake (Manitoba Water Stewardship, 2011). However,
satellite imagery is certainly suggestive of potential gyres and circulation patterns within the north basin (Abrahams et al., 2007),
which could act to concentrate ﬂoating debris like microplastics in
the north basin. Timing of collection (fall in 2014 versus summer in
2015e16) may also have had an inﬂuence on microplastic densities;
perhaps conditions in the fall act to generate inter-basin differences
compared with conditions in the lake during the summer. Additional fall samples and/or an examination of samples taken at
multiple time points through the year are required to better understand possible seasonal dynamics of microplastic particles in
Lake Winnipeg. Neither wind velocity nor direction at the time of
sampling explained microplastic particle densities. This is perhaps
because sampling with the manta trawl requires relatively calm
conditions to function properly.
Of the particles we detected in our study, the majority were ﬁbres. The bead-like particles which have been cause for such great
concern in the popular media, and in surface waters in the Great
Lakes (Eriksen et al., 2013) do not appear to be a signiﬁcant source
of microplastic contamination in the surface waters of Lake Winnipeg. Rather, it would seem that the majority of microplastics in
the lake are from either synthetic textiles, the breakdown of larger
particles, or atmospheric fallout. Atmospheric fallout has been
shown to generate between 30 and 300 particles per m2 per day in
Paris, 90% of which were identiﬁed as ﬁbres (Dris et al., 2015). Paris
is a metropolitan area that supports nearly 12 million people, and
Lake Winnipeg supports no major settlements on its shores; while
long-range atmospheric transport of microplastics from major
centres like the city of Winnipeg (60 km south) is possible, it would
likely be more than an order of magnitude less than that reported in
Dris et al. (2015) based on population alone within the city (approx.
700,000), and losses over long-range transport are expected
(though unquantiﬁed). Large particle breakdown has been proposed as a major source of microplastics in freshwater and oceanic

 228

P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Fig. 3. The proportion of the four types of microplastic particles (ﬁbres, fragments, ﬁlm and foam) found at the twelve stations across Lake Winnipeg over all years (2014e2016).
Note difference in y-axis scaling for lower panels. North basin stations are 22, 28, 65, W1, W4, W6, W8; South basin stations are 2, 7, 59, 3B and W9.

environments (Zbyszewski and Corcoran, 2011; Hidalgo-Ruz et al.,
2012). Fibres from synthetic materials would be expected to have
high concentrations in municipal wastewaters compared with
natural environments (Hartline et al., 2016; Mason et al., 2016). A
comparison of microplastic densities and prevalent types near
municipal wastewater inputs compared to the types of large particle plastics found in surveys of shoreline debris and/or large
particle inputs from inﬂows may help provide a better understanding of sources of microplastics in Lake Winnipeg and
elsewhere.
Our results suggest a potential inﬂuence of population density
on microplastic densities in some years; speciﬁcally, greater
microplastic densities in 2014 may be a consequence of greater
inputs from the Grand Rapids inﬂow, which supports the largest
population densities in the watershed. Free et al. (2014) showed
that proper waste management may play a more important role in
microplastic contamination than population size; Lake Hovsgol, a
remote mountain lake within a national park in Mongolia, had
alarmingly-high levels of microplastic contamination given its
isolated location, which the authors ascribe to the breakdown of
on-shore plastic on the lake. An average density of 20,264 particles/
km2 (less than that reported here for Lake Winnipeg, but greater
than reported for lakes Huron and Superior, Table 2) was reported
for Lake Hovsgol, with a low population density in the in surrounding area (Free et al., 2014). While the currently poor state of
waste management practices may have led to the elevated levels of
microplastics observed in Lake Hovsgol, it would be surprising if
waste management practices vary dramatically among the

watersheds of Lake Winnipeg. Agricultural inputs are greatest from
the Red River based on phosphorus loading estimates (Schindler
et al., 2012).
While the sampling in this study is far from a comprehensive
survey of microplastic distribution in Lake Winnipeg, it is more
extensive than published surveys conducted on the Great Lakes
(Eriksen et al., 2013). By providing an initial assessment of microplastics on the lake, we hope to spur future research that will help
to determine sources, transport, effects and seasonal dynamics of
plastic contamination in Lake Winnipeg, which is as great or
greater than that currently reported on the Great Lakes despite
having a much smaller population base.
Anecdotal evidence from Lake Winnipeg suggests that current
sampling methods for ﬂoating microplastics (both here and elsewhere) may be missing a signiﬁcant fraction of actual microplastic
particles in the lake. The standard manta trawl net used for sampling microplastics is 333 mm (Eriksen et al., 2013, 2014). However,
whole-water samples from at least two parts of Lake Winnipeg
have resulted in documented plastic particles, which appear to be
small enough to escape through the mesh size of the currently used
Manta trawl conﬁguration for assessing microplastic contamination (Fig. 6). As such, it should be emphasized the current estimates
of microplastic densities in freshwater lakes in Canada (i.e., our
study and those of Eriksen et al., 2013) are largely of particles
greater than 333 mm, and are very likely underestimates of
microplastic contamination in these systems (Anderson et al.,
2016). While microplastics in this size range may be of signiﬁcance to ﬁsh if ingested, smaller particles may have ecological

 P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

229

Fig. 4. Signiﬁcant interaction between year sampled and mean densities of microplastics in the north and south basins of Lake Winnipeg (two-way ANOVA: F2,29 ¼ 3.74, p ¼ 0.036).
Means plus or minus 1 standard error are presented.

Fig. 5. Lake Winnipeg microplastic densities are similar to Lake Erie, but elevated compared to values reported on Lake Huron and Lake Superior (Kruskal-Wallis test, Х2 ¼ 31.3,
p < 0.0001). Great Lakes data are from Eriksen et al. (2013). Data presented as boxplots, error bars are 95% conﬁdence intervals, while the box represents the interquartile of data
(0.25 and 0.75 quartiles); median values are bold, horizontal lines. Note log-scale on the y-axis.

 230

P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Table 2
Characteristics of Lake Winnipeg relative to other Laurentian Great Lakes with reported densities of microplastic.

Surface area (km2)
Mean depth (m)
Residence time (years)
Population in watershed
Watershed area (km2)
Mean microplastic density
# per km2
(±standard deviation)
Total estimated surface microplastic

Lake Winnipeg

Lake Superior

Lake Huron

Lake Erie

24,514
12
3e5
7,000,000
982,900
193,420
(±115,567)

82,100
147
191
600,000
127,700
5391
(±4552)

59,588
59
22
2,500,000
134,100
2779
(±2440)

25,744
19
2.6
12,000,000
78,000
105,503 (±173,587)

4.74 Â 109

4.43 Â 108

1.66 Â 108

2.72 Â 109

Fig. 6. Images of microplastic material found in Lake Winnipeg from whole-water samples taken (a) near Grand Rapids (inﬂow of the North Saskatchewan River) and (b) a 10-mm
phytoplankton haul from south of Black Island in the south basin (0e3 m depth). Whole-water samples were taken as a 2L surface grab with a wide-mouth bottle at 0e0.5 m. Photo
credit: (a) M. Stainton, (b) H. Kling. Scale bar in panel (a) is 100 mm; Particle in panel (b) measured approximately 300 mm long by 30 mm wide (H. Kling, personal communication).
Both particles would pass through traditional manta trawl nets given the proper orientation.

relevance to ﬁlter-feeding zooplankton (e.g., Cole et al., 2014). We
strongly recommend future work that employs tandem sampling of
nets with different mesh sizes to help better characterize microplastics smaller than those obtained by current methods, to the
extent possible, given limitations of obtaining sufﬁcient water
volumes while avoiding net clogging.
It is possible that the WPO methods we employed, though
widely adopted in the literature as a standard method (Masura
et al., 2015), may have altered or potentially digested some of the
material in our samples. Chemical digestion methods reported
elsewhere using acids and alkaline digestion methods have noted
degradation and potential digestion of ﬁne particles, especially ﬁbres (Claessens et al., 2013; Cole et al., 2014). To our knowledge, no
similar evaluations have been conducted on WPO treatments of
known plastic samples. Certain low-density polymers such as nylon
and low-density polyethylene (LDPE's) are known to be reactive to
30% H2O2 (chemical compatibility database, Cole-Parmer, https://
www.coleparmer.com/Chemical-Resistance, accessed 22 Jan
2017). However, it is unclear if nylon or LDPE's currently pose signiﬁcant environmental concern. By comparison, PET, the compound used in polar ﬂeece, which is of signiﬁcant concern with
regards to potential for environmental contamination through ﬁbre
shedding (Hartline et al., 2016) is highly resistant to 30% H2O2
(Plastics Europe, www.plasticseurope.org, accessed 22 Jan 2017).
Ultimately, our choice of digestion by WPO was guided entirely to
facilitate a comparison of our results with those reported elsewhere
in North America (Eriksen et al., 2013), who used similar methods
and allowed us to demonstrate similar contamination levels in Lake
Winnipeg as those observed previously in the Laurentian Great
Lakes.

In summary, we have demonstrated persistent microplastic
contamination in Lake Winnipeg over a three-year time period, and
identiﬁed these particles as consisting largely of ﬁbres. It is of signiﬁcant relevance that concentrations reported in Lake Winnipeg
are comparable to those observed on the Laurentian Great Lakes.
Our work provides an important baseline for future studies, and is
an important ﬁrst step for future investigation focused on the
identiﬁcation of inputs of microplastics into the lake and rates of
deposition (e.g., transport from rivers draining the lake and the
importance of long-range transport from municipalities, atmospheric deposition, etc), the quantiﬁcation of the presence of this
contamination on aquatic biota and potential impacts on
organisms.

Acknowledgements
Thanks to Mike Stainton and Hedy Kling for providing images of
plastics encountered in Lake Winnipeg whole water and phytoplankton samples and for stimulating discussions that helped to
initiate this work. We sincerely thank Dr. Sonya Havens, Roger
Mollot, Dr. Charles Wong, and Dr. Michael Paterson for assistance
with equipment and laboratory set up, and Dr. Guosheng Wu for
assistance and training on the SEM-EDS at the Lakehead University
Instrumentation Lab. Dr. Karen Scott, Marianne Geisler, Travis
Durhack, and the crew of the Namao provided critical assistance
with sample collection. Three anonymous reviewers provided
helpful comments on a previous draft. Funding for this project was
provided by the Lake Winnipeg Foundation, an NSERC Tier 2 Canada Research Chair to MDR, and NSERC Discovery Grants to MDR
and MLH.

 P.J. Anderson et al. / Environmental Pollution 225 (2017) 223e231

Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.envpol.2017.02.072.
References
Anderson, J.C., Park, B.J., Palace, V.P., 2016. Microplastics in aquatic environments:
implications for Canadian ecosystems. Environ. Pollut. 218, 269e280.
Abrahams, M.V., Mangel, M., Hedges, K., 2007. Predator-prey interactions and
changing environments: who beneﬁts? Phil. Trans. R. Soc. Lond. Ser. B, Biol. Sci.
362, 2095e2104.
Ballent, A., Corcoran, P.L., Madden, O., Helm, P.A., Longstaffe, F.J., 2016. Sources and
sinks of microplastics in Canadian Lake Ontario nearshore, tributary and beach
sediments. Mar. Pollut. Bull. 110, 383e395.
Bakir, A., Rowland, S.J., Thompson, R.C., 2014. Enhanced desorption of persistent
organic pollutants from microplastics under simulated physiological conditions.
Environ. Pollut. 185, 16e23.
Browne, M.A., Crump, P., Niven, S.J., Teuten, E., Tonkin, A., Galloway, T.,
Thompson, R., 2011. Accumulation of microplastic on shorelines worldwide:
sources and sinks. Environ. Sci. Technol. 45, 9175e9179.
Canada Gazette, 2016. Order adding a toxic substance to Schedule 1 to the Canadian
Environmental Protection Act, 1999. 150(13). http://www.gazette.gc.ca/rp-pr/
p2/2016/2016-06-29/html/sor-dors150-eng.php. (Accessed 7 September 2016).
~ eda, R.A., Avlijas, S., Simard, M.A., Ricciardi, A., 2014. Microplastic pollution in
Castan
St. Lawrence river sediments. Can. J. Fish. Aquat. Sci. 71, 1767e1771.
Claessens, M., Van Cauwenberghe, L., Vandegehuchte, M.B., Janssen, C.R., 2013. New
techniques for the detection of microplastics in sediments and ﬁeld collected
organisms. Mar. Pollut. Bull. 70, 227e233.
Cole, M., Webb, H., Lindeque, P.K., Fileman, E.S., Halsband, C., Galloway, T.S., 2014.
Isolation of microplastics in biota-rich seawater samples and marine organisms.
Sci. Rep. 4, 4528.
Corcoran, P.L., Norris, T., Ceccanese, T., Walzak, M.J., Helm, P.A., Marvin, C.H., 2015.
Hidden plastics of Lake Ontario, Canada and their potential preservation in the
sediment record. Environ. Pollut. 204, 17e25.
Drewes, J.E., Shore, L.S., 2001. Concerns about Pharmaceuticals in Water Reuse,
Groundwater Recharge, and Animal Waste. Pharmaceuticals and Care Products
in the Environment. CHAP, pp. 206e228.
Dris, R., Gasperi, J., Rocher, V., Saad, M., Renault, N., Tassin, B., 2015. Microplastic
contamination in an urban area: a case study in Greater Paris. Environ. Chem.
12, 592.
Eerkes-Medrano, D., Thompson, R.C., Aldridge, D.C., 2015. Microplastics in freshwater systems: a review of the emerging threats, identiﬁcation of knowledge
gaps and prioritisation of research needs. Water Res. 75, 63e82.
Eriksen, M., Lebreton, L.C.M., Carson, H.S., Thiel, M., Moore, C.J., Borerro, J.C.,
Galgani, F., Ryan, P.G., Reisser, J., 2014. Plastic pollution in the World's oceans:
more than 5 trillion plastic pieces weighing over 250,000 tons aﬂoat at sea.
PLoS One 9, e111913.
Eriksen, M., Mason, S., Wilson, S., Box, C., Zellers, A., Edwards, W., Farley, H.,
Amato, S., 2013. Microplastic pollution in the surface waters of the Laurentian
Great Lakes. Mar. Pollut. Bull. 77, 177e182.
Faure, F., Demars, C., Wieser, O., Kunz, M., de Alencastro, L.F., 2015. Plastic pollution
in Swiss surface waters: nature and concentrations, interaction with pollutants.
Environ. Chem. 12, 582.
Free, C.M., Jensen, O.P., Mason, S.A., Eriksen, M., Williamson, N.J., Boldgiv, B., 2014.
High-levels of microplastic pollution in a large, remote, mountain lake. Mar.
Pollut. Bull. 85, 156e163.
Hartline, N.L., Bruce, N.J., Karba, S.N., Ruff, E.O., Sonar, S.U., Holden, P.A., 2016. Microﬁber masses recovered from conventional machine washing of new or aged
garments. Environ. Sci. Technol. 50, 11532e11538.
Hidalgo-Ruz, V., Gutow, L., Thompson, R.C., Thiel, M., 2012. Microplastics in the

231

marine environment: a review of the methods used for identiﬁcation and
quantiﬁcation. Environ. Sci. Technol. 46, 3060e3075.
Ivar do Sul, J.A.I., Costa, M.F., 2014. The present and future of microplastic pollution
in the marine environment. Environ. Pollut. 185, 352e364.
Kahle, D., Wickham, H., 2013. ggmap: spatial Visualization with ggplot2. R J. 5,
144e161.
Koelmans, A.A., Besseling, E., Wegner, A., Foekema, E.M., 2013. Plastic as a carrier of
POPs to aquatic organisms: a model analysis. Environ. Sci. Technol. 47,
7812e7820.
Leite, A.S., Santos, L.L., Costa, Y., Hatje, V., 2014. Inﬂuence of proximity to an urban
center in the pattern of contamination by marine debris. Mar. Pollut. Bull. 81,
242e247.
Lusher, A.L., McHugh, M., Thompson, R.C., 2013. Occurrence of microplastics in the
gastrointestinal tract of pelagic and demersal ﬁsh from the English channel.
Mar. Pollut. Bull. 67, 94e99.
Manitoba Water Stewardship, 2011. State of Lake Winnipeg: 1999 to 2007. Environment Canada and Manitoba Water Stewardship.
Mason, S.A., Garneau, D., Sutton, R., Chu, Y., Ehmann, K., Barnes, J., Fink, P.,
Papazissimos, D., Rogers, D.L., 2016. Microplastic pollution is widely detected in
US municipal wastewater treatment plant efﬂuent. Environ. Pollut. 218,
1045e1054.
Masura, J., Baker, J., Foster, G., Arthur, C., 2015. Laboratory Methods for the Analysis
of Microplastics in the Marine Environment: Recommendations for Quantifying
Synthetic Particles in Waters and Sediments. NOAA Technical Memorandum
NOS-OR&R-48.
MSFD Technical Subgroup on Marine Litter, 2013. Guidance on Monitoring of Marine Litter in European Seas.
Murphy, F., Ewins, C., Carbonnier, F., Quinn, B., 2016. Wastewater treatment works
(WwTW) as a source of microplastics in the aquatic environment. Environ. Sci.
Technol. 50, 5800e5808.
Schindler, D.W., 2009. Lakes as sentinels and integrators for the effects of climate
change on watersheds, airsheds, and landscapes. Limnol. Oceanogr. 54,
2349e2358.
Schindler, D.W., Hecky, R.E., Mccullough, G.K., 2012. The rapid eutrophication of
Lake Winnipeg: greening under global change. J. Gt. Lakes. Res. 38, 6e13.
Sumaila, U.R., Khan, A., Watson, R., Munro, G., Zeller, D., Baron, N., Pauly, D., 2007.
The World Trade Organization and global ﬁsheries sustainability. Fish. Res. 88,
1e4.
Thompson, R.C., Olsen, Y., Mitchell, R.P., Davis, A., Rowland, S.J., John, A.W.,
McGonigle, D., Russell, A.E., 2004. Lost at sea: where is all the plastic? Sci. (New
York, N.Y.) 304, 838.
United States Congress, 2015. Microbead-free Waters Act of 2015. https://www.
congress.gov/bill/114th-congress/house-bill/1321 (Accessed 7 September 2016).
Van Cauwenberghe, L., Janssen, C.R., 2014. Microplastics in bivalves cultured for
human consumption. Environ. Pollut. 193, 65e70.
Waasenaar, L.I., Rao, Y.R., 2012. Lake Winnipeg, the forgotten lake. J. Gt. Lakes. Res.
38, 1e5.
Wickham, H., 2009. ggplot2: Elegant Graphics for Data Analysis, ﬁrst ed. Springer
Science & Business Media, New York.
Weinstein, J.E., Crocker, B.K., Gray, A.D., 2016. From macroplastic to microplastic:
degradation of high-density polyethylene, polypropylene, and polystyrene in a
salt marsh habitat. Environ. Toxicol. Chem. 35, 1632e1640.
~ oz, D., Brennholt, N., Bourrain, X., Buchinger, S.,
Wagner, M., Scherer, C., Alvarez-Mun
Fries, E., Grosbois, C., Klasmeier, J., Marti, T., 2014. Microplastics in freshwater
ecosystems: what we know and what we need to know. Environ. Sci. Eur. 26, 12.
Wright, S.L., Thompson, R.C., Galloway, T.S., 2013. The physical impacts of microplastics on marine organisms: a review. Environ. Pollut. 178, 483e492.
Younos, T., 2005. Emerging threats to drinking water quality. Renew. Resour. J. 23, 6.
Zbyszewski, M., Corcoran, P.L., 2011. Distribution and degradation of fresh water
plastic particles along the beaches of Lake Huron, Canada. Water, Air, & Soil
Pollut. 220, 365e372.