Plastic debris in the open ocean

Andrés Cózara,1, Fidel Echevarríaa, J. Ignacio González-Gordilloa, Xabier Irigoienb,c, Bárbara Úbedaa,
Santiago Hernández-Leónd, Álvaro T. Palmae, Sandra Navarrof, Juan García-de-Lomasa, Andrea Ruizg,
María L. Fernández-de-Puellesh, and Carlos M. Duartei,j,k,l
a
Área de Ecología, Facultad de Ciencias del Mar y Ambientales, Universidad de Cádiz, Campus de Excelencia Internacional del Mar, E-11510 Puerto Real, Spain;
b
Red Sea Research Center, King Abdullah University of Science and Technology, Thuwal 23955-6900, Kingdom of Saudi Arabia; cAZTI, Arrantza eta
Elikaigintzarako Institutu Teknologikoa, 20110 Pasaia, Spain; dInstituto de Oceanografía y Cambio Global, Universidad de Las Palmas de Gran Canaria,
Campus Universitario de Tafira, 35017 Las Palmas de Gran Canaria, Canary Islands, Spain; eFisioaqua, Las Condes, 6513677 Santiago, Chile; fDepartamento de
Ecología, Universidad de Barcelona, E-08028 Barcelona, Spain; gPakea Bizkaia, 48990 Getxo, Spain; hInstituto Español de Oceanografía, Centro Oceanográfico
de Baleares, 07015 Palma de Mallorca, Spain; iDepartment of Global Change Research, Instituto Mediterráneo de Estudios Avanzados (Universidad de las Islas
Baleares-Consejo Superior de Investigaciones Científicas), 07190 Esporles, Spain; jThe University of Western Australia Oceans Institute and kSchool of Plant
Biology, The University of Western Australia, Crawley, WA 6009, Australia; and lFaculty of Marine Sciences, King Abdulaziz University, Jeddah 21589,
Kingdom of Saudi Arabia

Edited by David M. Karl, University of Hawaii, Honolulu, HI, and approved June 6, 2014 (received for review August 3, 2013)

There is a rising concern regarding the accumulation of floating acquired from seawater through sorption processes [e.g., hy-
plastic debris in the open ocean. However, the magnitude and the drophobic chemicals (14, 15)]. Recent studies provide evidence
fate of this pollution are still open questions. Using data from the that these contaminants can accumulate in the receiving organ-
Malaspina 2010 circumnavigation, regional surveys, and previously isms during digestion (14).
published reports, we show a worldwide distribution of plastic Our awareness of the significance of plastic pollution in the
on the surface of the open ocean, mostly accumulating in the ocean is relatively recent, and basic questions remain unresolved.
convergence zones of each of the five subtropical gyres with
Indeed, the quantity of plastic floating in the ocean and its final
comparable density. However, the global load of plastic on the
destination are still unknown (16). Historical time series of surface
open ocean surface was estimated to be on the order of tens of
thousands of tons, far less than expected. Our observations of
plastic concentration in fixed ocean regions show no significant
the size distribution of floating plastic debris point at important increasing trend since the 1980s, despite an increase in production
size-selective sinks removing millimeter-sized fragments of floating and disposal (3, 16, 17). These studies suggest that surface waters
plastic on a large scale. This sink may involve a combination of fast are not the final destination for buoyant plastic debris in the ocean.
nano-fragmentation of the microplastic into particles of microns or Nano-fragmentation, predation, biofouling, or shore deposition
smaller, their transference to the ocean interior by food webs and have been proposed as possible mechanisms of removal from the
ballasting processes, and processes yet to be discovered. Resolving surface (3, 9, 16).
the fate of the missing plastic debris is of fundamental importance to On the basis of samples collected on a circumnavigation cruise
determine the nature and significance of the impacts of plastic (Malaspina 2010 expedition), on five regional cruises, and avail-
pollution in the ocean. able data from recent studies (3–5, 17–19), we aim to provide a
first-order approximation of the load of plastic debris in surface

T he current period of human history has been referred as the

Plastic Age (1). The light weight and durability of plastic
waters of the open ocean. We also examine the size distribution
of floating plastic debris collected along the circumnavigation to

ENVIRONMENTAL
materials make them suitable for a very wide range of prod- provide insight into the nature of possible losses of floating plastic

SCIENCES
ucts. However, the intense consumption and rapid disposal of from the open ocean surface.
plastic products is leading to a visible accumulation of plastic
debris (2). Plastic pollution reaches the most remote areas of
Significance
the planet, including the surface waters of the open ocean.
Indeed, high concentrations of floating plastic debris have
been reported in central areas of the North Atlantic (3) and High concentrations of floating plastic debris have been re-
Pacific Oceans (4, 5), but oceanic circulation models suggest ported in remote areas of the ocean, increasing concern about
possible accumulation regions in all five subtropical ocean the accumulation of plastic litter on the ocean surface. Since
gyres (6, 7). The models predict that these large-scale vortices the introduction of plastic materials in the 1950s, the global
act as conveyor belts, collecting the floating plastic debris re- production of plastic has increased rapidly and will continue in
leased from the continents and accumulating it into central the coming decades. However, the abundance and the distri-
convergence zones. bution of plastic debris in the open ocean are still unknown,
Plastic pollution found on the ocean surface is dominated by despite evidence of affects on organisms ranging from small
particles smaller than 1 cm in diameter (8), commonly referred invertebrates to whales. In this work, we synthetize data col-
to as microplastics. Exposure of plastic objects on the surface lected across the world to provide a global map and a first-
waters to solar radiation results in their photodegradation, em- order approximation of the magnitude of the plastic pollution
brittlement, and fragmentation by wave action (9). However, in surface waters of the open ocean.
plastic fragments are considered to be quite stable and highly
Author contributions: A.C., F.E., J.I.G.-G., X.I., and C.M.D. designed research; A.C., F.E., J.I.
durable, potentially lasting hundreds to thousands of years (2). G.-G., X.I., B.U., S.H.-L., A.T.P., S.N., J.G.-d.-L., A.R., M.L.F.-d.-P., and C.M.D. performed
Persistent nano-scale particles may be generated during the research; A.C., X.I., B.U., S.N., J.G.-d.-L., and M.L.F.-d.-P. contributed new reagents/analytic
weathering of plastic debris, although their abundance has not tools; A.C., J.I.G.-G., B.U., A.T.P., S.N., and J.G.-d.-L. analyzed data; and A.C., F.E., X.I., and
been quantified in ocean waters (9). C.M.D. wrote the paper.

As the size of the plastic fragments declines, they can be The authors declare no conflict of interest.
ingested by a wider range of organisms. Plastic ingestion has This article is a PNAS Direct Submission.
been documented from small fish to large mammals (10–12). Freely available online through the PNAS open access option.
The most evident effects of plastic ingestion are mechanical [e.g., 1
To whom correspondence should be addressed. E-mail: andres.cozar@uca.es.
gastrointestinal obstruction in seabirds (13)], but plastic frag- This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
ments contain contaminants added during plastic manufacture or 1073/pnas.1314705111/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1314705111 PNAS | July 15, 2014 | vol. 111 | no. 28 | 10239–10244

Results and Discussion the 1970s (106 tons) is 100-fold larger than our estimate of the
The dataset assembled here included 3,070 total samples col- current load of plastic stored in the ocean.
lected around the world (SI Appendix, Table S1). The frequency Examination of the size distribution of plastic debris on the
of occurrence of plastic debris in the surface samples of the open ocean surface shows a peak in abundance of fragments around
ocean was considerably high (88%; Fig. 1). Nevertheless, the con- 2 mm and a pronounced gap below 1 mm (Fig. 3A). Similar pat-
centration of plastic ranged broadly, spanning over four orders of terns are found when the data are analyzed separately by ocean
magnitude across the open ocean. The distribution pattern agreed basin (SI Appendix, Fig. S6). The predominance of fragments in
with those predicted from ocean surface circulation models (6, 7), an intermediate interval (1–5 mm) of the plastic size spectra is
confirming the accumulation of plastic debris in the convergence also a general feature for the oceanic size distributions reported
zone of each of the five large subtropical gyres. Using the high and in the past (5, 8). However, experiments on the fragmentation of
low ranges of spatial concentrations measured within 15 major plastic materials show that the size distribution of fragments
convergence/divergence zones in the global ocean (Fig. 2), we es- generated by a plastic object conforms to a fractal process,
timate the amount of plastic in the open-ocean surface between spreading over several orders of magnitude and below the size
7,000 and 35,000 tons (Table 1). The plastic concentrations per range in our study (25, 26). Cracking patterns of photodegraded
surface area were comparable across each of the five accumulation plastics are observed at multiple scales, from centimeters to few
zones, although the North Pacific Ocean contributed importantly microns (9). Therefore, the progressive fragmentation of the
to the global plastic load (between 33 and 35%), mainly owing to plastic objects into more and smaller pieces should lead to
the size of this gyre. The plastic load in the North Pacific Ocean a gradual increase of fragments toward small sizes. In steady
could be related to the high human population on the eastern coast state, the abundance–size distribution should follow a power law,
of the Asian continent, the most densely populated coast in the with a scaling exponent equal to the spatial dimension of the
world, with one-third of the global coastal population (20). Indeed, plastic objects (i.e., 3, SI Appendix, Fig. S8). Likewise, a stable
the surface plastic concentrations measured in the Kuroshio input and fragmentation of large plastic objects should result in
Current, the western arm of the North Pacific Gyre, can become a steady volume–size distribution. A model based on fragmen-
exceptionally high, including the highest reported for nonaccu- tation, without additional losses, gave an abundance–size distri-
mulation regions (21, 22). bution similar to that sampled, which showed a power exponent
Continental plastic litter enters the ocean largely through of 2.93 ± 0.08, similar to the expected value, but only for size
storm-water runoff, flowing into watercourses or directly dis- classes larger than 5 mm. Below 5 mm, the observed size dis-
charged into coastal waters. Estimating the plastic input to the tribution diverged from that expected from the model (Fig. 3 B
ocean is a complex task. In the 1970s, the US National Academy and C). Because plastic input is progressively transferred toward
of Sciences estimated that the flux of plastic to the world oceans small-size classes by fragmentation, this divergence results from
was 45,000 tons per year (23), equivalent to 0.1% of the global the gradual accumulation of plastic losses. An assessment of
production of plastic (24). Since then, the annual production progressive departures of the observed distribution from a con-
of plastic has quintupled (265 million tons per year in 2010). servative distribution indicates that losses are concentrated
Around 50% of the produced plastic is buoyant (24), and 60–64% around sizes of 2.2 mm (Fig. 3C). Hence, the paucity of frag-
of the terrestrial load of floating plastic to the sea is estimated to ments in the lowest part of the size distribution would be ex-
be exported from coastal to open-ocean waters (7). Despite the plained by the interruption of the downward transfer of plastic at
possible inaccuracies of these numbers, a conservative first-order the millimeter scale, unless there is an abrupt nano-fragmenta-
estimate of the floating plastic released into the open ocean from tion of the millimeter-sized particles directly into pieces of

Fig. 1. Concentrations of plastic debris in surface waters of the global ocean. Colored circles indicate mass concentrations (legend on top right). The map shows
average concentrations in 442 sites (1,127 surface net tows). Gray areas indicate the accumulation zones predicted by a global surface circulation model (6). Dark
and light gray represent inner and outer accumulation zones, respectively; white areas are predicted as nonaccumulation zones. Data sources are described in SI
Appendix, Table S1. Plastic concentrations along the Malaspina circumnavigation and a latitudinal gradient are graphed in SI Appendix, Figs. S4 and S5.

10240 | www.pnas.org/cgi/doi/10.1073/pnas.1314705111 Cózar et al.

 a removal rate much faster than the input into the ocean, with
the reduced global load of surface plastic resulting from a delay
between input and removal. Another requirement is that the sink
must lead to a degradation or permanent sequestration of plas-
tic. Finally, the size distribution of floating plastic debris is evi-
dence for a size-selective loss process or processes.
A selective washing ashore of the millimeter-sized fragments
trapped in central areas of the open ocean is unlikely. Likewise,
there is no reason to assume that the rate of solar-induced
fragmentation increased since the 1980s (3). However, the gap in
the plastic size distribution below 1 mm could indicate a fast
breaking down of the plastic fragments from millimeter scale to
micrometer scale. Recent scanning electron micrographs of the
surface of microplastic particles showed indications that oceanic
bacterial populations may be contributing to their degradation,
potentially intervening in the fragmentation dynamics (27). The
scarce knowledge of the biological and physical processes driving
the plastic fragmentation leaves room for the possibility of a two-
phase fragmentation, with an accelerated breakdown of the
photodegraded fragments with dimension of few millimeters.
A preferential submersion of small-sized plastic, with high
surface:volume ratio, by ballasting owing to epiphytic growth could
also be possible. Once biofouled fragments reach seawater density,
they enter the water column as neutrally drifting or slowly sinking
particles. Biofouled fragments probably are often incorporated
into the sediment in shallow and, particularly, nutrient-rich areas
Fig. 2. Ranges of surface plastic concentrations by ocean. Nonaccumulation
(28), but this may be a less effective mechanism in the deep,
zone (blue boxes), outer accumulation zone (green boxes), and inner accu-
mulation zone (red boxes). The boundaries of the boxes indicate the 25th open ocean (9, 29). Because the seawater density gradually in-
and 75th percentiles, the black lines within the box mark the mean, and the crease with depth, the slowly sinking plastic, marginally exceed-
whiskers above and below the boxes indicate the 90th and 10th percentiles. ing the surface seawater density, should remain suspended at a
Data used in this graph are mapped in Fig. 1. An equivalent analysis for depth where its density is equal to that of the medium. Field ex-
a dataset of plastic concentrations not corrected by wind effects is graphed periments have shown that biofouled plastic debris undergoes a
in SI Appendix, Fig. S3. rapid defouling when submerged, causing the plastic to return to
the surface (29). Defouling in deep water could occur, for ex-
ample, from adverse conditions for the epiphytic organisms (e.g.,
few microns or smaller, allowing passage through the 200-μm decreasing irradiance) or the dissolution of carbonates and opal
mesh net used (SI Appendix, Fig. S9). A sampling bias causing owing to acidic conditions.
the apparent loss in small sizes can be rejected because the size The fourth possible sink is ingestion by marine organisms. The
distribution of nonplastic particles in the same samples followed

ENVIRONMENTAL
size interval accumulating most of plastic losses corresponds to
the characteristic power distribution, with increasing abundances that of zooplankton (mainly copepods and euphausiids). Zoo-

SCIENCES
toward smaller sizes (SI Appendix, Fig. S12). planktivorous predators represent an abundant trophic guild in
Our study reports an important gap in the size distribution of the ocean, and it is known that accidental ingestion of plastic
floating plastic debris as well as a global surface load of plastic occurs during their feeding activity. The reported incidence of
well below that expected from production and input rates. To- plastic in stomachs of epipelagic zooplanktivorous fish ranges from
gether with the lack of observed increasing temporal trends in 1 to 29% (30, 31), and in stomachs of small mesopelagic fish from 9
surface plastic concentration (3, 16, 17), these findings provide to 35% (10, 32). The most frequent plastic size ingested by fish in
strong support to the hypothesis of substantial losses of plastic all these studies was between 0.5 and 5 mm, matching the pre-
from the ocean surface. A central question arising from this dominant size of plastic debris where global losses occur in our
conclusion is how floating plastic is being removed. Four main assessment. Also, these plastic sizes are commonly found in pred-
possible sinks have been proposed: shore deposition, nano- ators of zooplanktivorous fish (30, 31, 33).
fragmentation, biofouling, and ingestion (3, 9). Although a rig- Although diverse zooplanktivorous predators must contribute
orous attribution of losses to each of these mechanisms is not yet to the plastic capture at millimeter scale, the small mesopelagic
possible, our study provides some insights as to their plausibility. fish likely play a relevant role. They constitute the most abundant
To counterbalance the increase in input rates over the past and ubiquitous zooplanktivorous assemblage in the open ocean,
decades, the removal rate of the presumed sink would also have with densities close to one individual per square meter also in the
needed to increase (3). Alternatively, the lack of increasing oligotrophic subtropical gyres (34, 35). Mesopelagic fish live in
trends in surface plastic pollution could also be explained from the middle layer (200–1,000 m deep) of the ocean but migrate to

Table 1. Range of the global load of plastic debris in surface waters of the open ocean
Plastic debris, kilotons North Pacific Ocean North Atlantic Ocean Indian Ocean South Atlantic Ocean South Pacific Ocean Total

Low estimate 2.3 1.0 0.8 1.7 0.8 6.6

Mid estimate 4.8 2.7 2.2 2.6 2.1 14.4
High estimate 12.4 6.7 5.1 5.4 5.6 35.2

Loads by ocean were estimated from the low, mid, and high ranges of plastic concentration measured within major regions in relation to the degree of surface
convergence (nonaccumulation zone, outer accumulation zone, and inner accumulation zone). The ranges of plastic concentration by zones are shown in Fig. 2.

Cózar et al. PNAS | July 15, 2014 | vol. 111 | no. 28 | 10241
 Fig. 3. Size distribution of floating plastic debris
collected during the Malaspina circumnavigation at
calm conditions. (A) Size distribution in abundance
(light blue bars) and abundance normalized by the
width (in millimeters) of the size class (blue circles). (B)
Measured (blue circles) and modeled (red squares) size
distributions of normalized abundance in logarithmic
scale. Modeled distribution was strictly based on frag-
mentation of large plastic items. (C) Measured (blue
circles) and modeled (red squares) size distributions in
normalized volume. Green bars indicate the estimated
losses of plastic volume by size class (Δ i ). After
smoothing the measured distribution with a Weibull
function (black line, R = 0.9979, P < 0.0001), losses by
size were estimated from its progressive departure
from the modeled distribution. Dashed vertical lines
through all three graphs correspond to 1-mm and
5-mm size limits. Because plastic presence declined
for sizes over 10 cm, modeling analysis was applied
up to 10 cm. Note that the largest size class extends
from 10 cm to 1 m, the length of the net mouth.
Measured size distributions are built from the plastic
collected in tows with u* <0.5 cm·s−1 (4,184 plastic
items) to avoid wind-mixing effect. An analysis of the
effect of wind mixing on plastic size distribution is
shown in SI Appendix, Fig. S7, and size distributions
for the whole Malaspina dataset (7,359 plastic items)
are graphed in SI Appendix, Fig. S10.

feed in the surface layer at night. Using the plastic content in single day, depending on whether ingested fragments remain in
stomachs, the reported estimates of standing load of plastic in the fish throughout their complete lifespan or are defecated (32).
mesopelagic fish (32) are on the same order of magnitude as our The plastic fragments ingested by small fish can be transferred to
estimates of free plastic on the surface. The turnover time of the larger predators (31, 33), sink with the bodies of dead fish, or be
plastic contained in mesopelagic fish must vary from 1 y to a defecated. Gut content of mesopelagic fish is evacuated as long

10242 | www.pnas.org/cgi/doi/10.1073/pnas.1314705111 Cózar et al.

viscous feces that assume spheroid shapes while sinking at high Overall, 7,359 plastic items were measured and separated in 28 size classes to
velocities (around 1,000 m·d−1) (36). Hence, microplastic fragments build a size distribution. Size limits of the bins followed a 0.1-log series of l.
could also reach the bottom via defecation, a proposition that re- The width of the uppermost bin extended from 10 cm to the length of the
net mouth (100 cm) to account for all sizes that could be collected by the
quires further quantitative testing.
net. The trapping efficiency of fine particles by the mesh was tested from
Surface losses of large plastic objects by sinking are un- the analysis of the size distribution of nonplastic particles in six tows evenly
accounted for in our fragmentation model (Fig. 3). However, distributed along the circumnavigation (SI Appendix, Fig. S12). Once the
these large objects, included those in the uppermost part of our plastic particles were picked out from the samples, the size distribution of
plastic size spectrum, are commonly observed on the seafloor nonplastic particles was measured by the same methods.
(37) and likely contribute significantly to reduce the global load Wind stress can extend the vertical distribution of floating plastic debris
at the surface. Large plastic objects undergo particular bio- into the surface mixing layer, resulting in underestimation of the plastic
fouling because they can host a wide size range of organisms and concentrations measured by the surface tows (0.25 m deep). Thus, the in-
often show large cavities (e.g., bags, bottles) that facilitate their tegrated plastic abundance from the surface to the base of the wind-mixed
ballasting and subsequent sinking. layer (generally <25 m) was estimated with a model dependent on u* and
the numerical concentrations measured in the surface tows (39). Wind-cor-
In the present study, we confirm the gathering of floating
rected abundances were converted to mass concentrations using a correla-
plastic debris, mainly microplastics, in all subtropical gyres. The tion based on simultaneous measurements of total mass and abundance of
current plastic load in surface waters of the open ocean was plastic in 570 worldwide tows (SI Appendix, Fig. S13).
estimated in the order of tens of thousands of tons (10,000–
40,000). This estimate could be greatly improved through joining Size-Distribution Analysis. A theoretical size distribution of plastic derived
sampling efforts particularly in semiclosed seas (e.g., Mediter- from fragmentation was modeled by assuming steady state (large-objects
ranean) and the southern hemisphere, where existing data are input = small-fragments output, below 0.2 mm). Given that the plastic abun-
scarce. Nevertheless, even our high estimate of plastic load, dance in a given size class depends on the fragmentation of larger plastic
based on the 90th percentile of the regional concentrations, is objects already present, we selected a size class with relatively large plastic
considerably lower than expected, by orders of magnitude. Our (reference bin) and projected the plastic amount measured in this bin toward
smaller and larger size classes (onward and backward in time). Therefore, the
observations also show that large loads of plastic fragments with
normalized abundance (divided by the width of the size-class interval) of the
sizes from microns to some millimeters are unaccounted for in size class i derived from steady fragmentation was modeled as
the surface loads. The pathway and ultimate fate of the missing
plastic are as yet unknown. We cannot rule out either of the Aref · α · lref
3
Aref · lref
3
Afi = = :
proposed sink processes or the operation of sink processes yet to α · li3 li3
be identified. Indeed, the losses inferred from our assessment
likely involve a combination of multiple sinks. Missing micro- We used a standard shape for the plastic fragments having the three principal
plastic may derive from nano-fragmentation processes, rendering axes proportional to l. Thus, α · li3 accounts for the mean volume of the
fragments of i, with α being a shape factor and li the nominal length for the
the very small pieces undetectable to convectional sampling nets,
class i, set at the bin midpoint. Aref is the normalized abundance measured
and/or may be transferred to the ocean interior. The abundance in the reference bin (i = ref). The 20- to 25-mm class was selected as reference,
of nano-scale plastic particles has still not been quantified in the although similar results were obtained by selecting other large-size classes.
ocean (9), and the measurements of microplastic in deep ocean The normalized volume in each size class derived from fragmentation was
are very scarce, although available observations point to a sig- modeled as Vif = Afi · α · li3 = Aref · α · lref
3
, being α = 0.1, a value corresponding
nificant abundance of microplastic particles in deep sediments to flat-shaped volume. Because the steady fragmentation of the large-plastic
(38), which invokes a mechanism for the vertical transport of input results in an even volume–size distribution, deviations of the observed
plastic particles, such as biofouling or ingestion. Because plastic size distribution from a conservative distribution can be related to changes in

ENVIRONMENTAL
inputs into the ocean will probably continue, and even increase, the fragmentation dynamics, inputs of small plastics, or losses (SI Appendix,

SCIENCES
resolving the ultimate pathways and fate of these debris is a matter Fig. S9). Estimating volumes from observed abundances ðVi* = Ai* · α · li3 Þ, and
after smoothing the resulting volume–size distribution to remove small ir-
of urgency.
regularities, the deviations from a conservative distribution (Δi, expressed as
percentage of total) were calculated as
Materials and Methods
   
From December 2010 to July 2011 the Spanish circumnavigation expedition A* * 3
i−1 · li−1 − Ai · li
* − V* 3
Vi−1
Malaspina 2010 sampled surface plastic pollution at 141 sites across the Δi = P  i  = P 

  ,
n  
V * − V * 
n
oceans. Floating plastic was collected with a neuston net (1.0- × 0.5-m mouth, i=1 i−1 i  A · l3 − A · l3 
i=1 i−1 i−1 i i
200-μm mesh) towed at 2–3 knots for periods 10–15 min (total tows 225).
Tow areas were calculated from the readings of a flowmeter in the mouth of where i = 1, 2, . . ., n, with n being the lowest size class (0.2–0.25 mm). The
the net. Wind speed and water surface density were measured during each denominator accounts for the total deviations accumulated across the entire
tow to estimate average friction velocity in water (u*) (39). size range studied. Negative values of Δi are related to net plastic losses and
The material collected by the net was mixed with 0.2-mm-filtered sea- positive values to plastic accumulations. Note that Δi is independent of the
water. Subsequently, floating plastic debris was carefully picked out from the standard plastic shape (α value) used in the computations. Possible variations
water surface with the aid of a dissecting microscope. This examination was of α with size were unable to induce changes in the volume–size distribution
repeated at least twice to ensure the detection of all of the smallest plastic enough to explain the gap found in small sizes, owing to the extreme
particles. To confirm the plastic nature of the material collected in the scarceness of plastic below 1 mm and the geometrical constrain for α, get-
examinations, Raman spectroscopy was applied to a random subset of par- ting the maximum at 0.52 (spherical shape). Observed plastic abundance in
ticles (n = 67). The analysis confirmed the identity of all plastic particles, and the lowest part of the size spectrum was four orders of magnitude lower
polyethylene was found to be the most common polymer type. The vast than expected from fragmentation (Fig. 3).
majority of the plastic items consisted of fragments of larger objects, and The size-distribution analysis is a useful tool to constrain the possible
industrial resin pellets represented only a small fraction (<2%) of all en- dynamics of marine plastic pollution. Nevertheless, the mechanisms leading
countered items. Textile fibers were found only occasionally and were ex- to the observed plastic size distributions still are not entirely understood and
cluded from the analysis because they could be airborne contamination deserve further attention, resolving the size dependence of the sink/sources
from clothing during the sampling or processing (31). processes, as well as testing the framework proposed here (SI Appendix, Fig.
Plastics extracted from the seawater samples were washed with deionized S9) to identify additional processes.
water and dried at room temperature. The total dry weight of the plastics
collected in each tow was recorded. The maximum linear length (l) of the Spatial Analysis. To analyze the global distribution of floating plastic, data
plastic items was measured by high-resolution scanning (SI Appendix, Fig. from the Malaspina circumnavigation were combined with additional re-
S11) and the image processing Zooimage software (www.sciviews.org). Al- gional surveys and recent (from 2006 to date) measurements reported by
ternatively, excessively large plastic objects were measured with a ruler. other researchers after data standardization (SI Appendix, Table S1).

Cózar et al. PNAS | July 15, 2014 | vol. 111 | no. 28 | 10243
Concentrations of plastic per surface-water volume were converted to con- sampling frequency. Overall, 442 grid cells (1,127 net tows) were included in
centrations per surface area from the tow depth, determined according to the wind-corrected dataset (Fig. 1 and SI Appendix, Table S1). Midrange
net type and mouth dimensions (one-half mouth height for neuston nets, regional concentrations were calculated from the averaging of the wind-
three-fourths mouth height for manta nets). Plastic concentrations mea- corrected plastic concentrations within each major zone. High-range re-
sured with mesh sizes larger than 0.2 mm were multiplied by a correction gional concentrations were calculated from the 90th percentile. We used
factor derived from the plastic size distribution measured in the Malaspina a wide confidence interval for the plastic load estimate to address vari-
circumnavigation. For 0.3-, 0.5-, and 1.0-mm mesh sizes, numerical un-
ability and possible inaccuracies in the spatial concentrations of plastic.
derestimation was estimated at 0.4, 2.7, and 21.3%, and mass un-
Low-range concentrations were calculated from the averaging of the di-
derestimation at 0.0, 0.4, and 5.0%, respectively. Data reported in numerical
rect measurements of surface concentrations, without wind correction or
concentrations were converted to mass concentrations by using the global
relationship found between total mass and abundance (SI Appendix, Fig. discards by high wind mixing (noncorrected dataset: 851 grid cells, 3,070
S13). For data reported without wind correction (3–5, 18), we use satellite net tows; SI Appendix, Figs. S2 and S3). Global plastic loads in the open-
winds from the CCMP database (http://podaac.jpl.nasa.gov) to discard samples ocean surface were estimated from high, mid, and low regional concen-
collected with winds speeds larger than 5 m·s−1 (u* ∼0.6 cm·s−1), the threshold trations and surface areas.
above which the effects of wind stress can be significant (39).
The range of the global plastic load in the surface ocean was estimated ACKNOWLEDGMENTS. We thank Pakea Bizkaia and the Chilean Navy, which
from the concentration ranges measured over 15 major zones in relation contributed to the sample collection, and K. L. Law, M. C. Goldstein, M. J. Doyle,
to the degree of surface convergence and by using two different sets of M. Eriksen, J. Reisser, and their collaborators for their available data. We also
measurements, a wind-corrected dataset and a noncorrected dataset. Using thank S. Loiselle and J. Ruiz for his useful suggestions in writing the paper. This
research was funded by the Spanish Ministry of Economy and Competitiveness
a global circulation model (6), nonaccumulation, outer accumulation, and
through the Malaspina 2010 expedition project (Consolider-Ingenio 2010,
inner accumulation zones were delimited in each ocean basin to reduce the CSD2008-00077) and the Migrants and Active Flux in the Atlantic Ocean project
inaccuracies derived from an uneven distribution of measurements. In ad- (CTM2012-39587-C04-01). Original data reported in this paper are freely avail-
dition, plastic measurements were spatially averaged over grid cells of 2° able at http://metamalaspina.imedea.uib-csic.es/geonetwork. This is Campus
in both latitude and longitude to avoid overweight of areas with high de Excelencia Internacional del Mar (CEIMAR) Publication 58.

1. Yarsley VE, Couzens EG (1945) Plastics (Penguin, London). 21. Day RH, Shaw DG, Ignell SE (1990) The quantitative distribution and characteristics of
2. Barnes DKA, Galgani F, Thompson RC, Barlaz M (2009) Accumulation and fragmen- neuston plastic in the North Pacific Ocean, 1985–88. Proceedings of the 2nd In-
tation of plastic debris in global environments. Philos Trans R Soc Lond B Biol Sci ternational Conference on Marine Debris, eds Shomura RS, Godfrey ML (National
364(1526):1985–1998. Oceanic and Atmospheric Administration, Honolulu), pp 247–266.
3. Law KL, et al. (2010) Plastic accumulation in the North Atlantic subtropical gyre. Sci- 22. Yamashita R, Tanimura A (2007) Floating plastic in the Kuroshio Current area,
ence 329(5996):1185–1188. western North Pacific Ocean. Mar Pollut Bull 54(4):485–488.
4. Goldstein MC, Rosenberg M, Cheng L (2012) Increased oceanic microplastic debris 23. National Academy of Sciences (1975) Assessing Potential Ocean Pollutants: A Report
enhances oviposition in an endemic pelagic insect. Biol Lett 8(5):817–820. of the Study Panel on Assessing Potential Ocean Pollutants to the Ocean Affairs Board
5. Eriksen M, et al. (2013) Plastic pollution in the South Pacific subtropical gyre. Mar (National Research Council, Washington, DC).
Pollut Bull 68(1–2):71–76. 24. Association of Plastic Manufacturers (2011) Plastics – the Facts 2011: An analysis of
6. Maximenko N, Hafner J, Niiler P (2012) Pathways of marine debris derived from tra- European Plastic Production, Demand and Recovery for 2010 (Plastic Europe, Brussels).
jectories of Lagrangian drifters. Mar Pollut Bull 65(1–3):51–62. 25. Timár G, Blömer J, Kun F, Herrmann HJ (2010) New universality class for the frag-
7. Lebreton LCM, Greer SD, Borrero JC (2012) Numerical modelling of floating debris in mentation of plastic materials. Phys Rev Lett 104(9):095502.
the world’s oceans. Mar Pollut Bull 64(3):653–661. 26. Kishimura H, Noguchi D, Preechasupanya W, Matsumoto H (2013) Impact fragmen-
8. Hidalgo-Ruz V, Gutow L, Thompson RC, Thiel M (2012) Microplastics in the marine tation of polyurethane and polypropylene cylinder. Physica A 392(22):5574–5580.
environment: A review of the methods used for identification and quantification. 27. Zettler ER, Mincer TJ, Amaral-Zettler LA (2013) Life in the “plastisphere”: Microbial
Environ Sci Technol 46(6):3060–3075. communities on plastic marine debris. Environ Sci Technol 47(13):7137–7146.
9. Andrady AL (2011) Microplastics in the marine environment. Mar Pollut Bull 62(8): 28. Vianello A, et al. (2013) Microplastic particles in sediments of Lagoon of Venice, Italy:
1596–1605. First observations on occurrence, spatial patterns and identification. Estuar Coast
10. Boerger CM, Lattin GL, Moore SL, Moore CJ (2010) Plastic ingestion by planktivorous Shelf Sci 130:54–61.
fishes in the North Pacific Central Gyre. Mar Pollut Bull 60(12):2275–2278. 29. Andrady AL, Song Y (1991) Fouling of floating plastic debris under Biscayne Bay ex-
11. Choy CA, Drazen JC (2013) Plastic for dinner? Observations of frequent debris in- posure conditions. Mar Pollut Bull 22(12):117–122.
gestion by pelagic predatory fishes from the central North Pacific. Mar Ecol Prog Ser 30. Lusher AL, McHugh M, Thompson RC (2013) Occurrence of microplastics in the gas-
485:155–163. trointestinal tract of pelagic and demersal fish from the English Channel. Mar Pollut
12. de Stephanis R, Giménez J, Carpinelli E, Gutierrez-Exposito C, Cañadas A (2013) As Bull 67(1–2):94–99.
main meal for sperm whales: Plastics debris. Mar Pollut Bull 69(1–2):206–214. 31. Foekema EM, et al. (2013) Plastic in north sea fish. Environ Sci Technol 47(15):
13. Azzarello MY, Van-Vleet ES (1987) Marine birds and plastic pollution. Mar Ecol Prog 8818–8824.
Ser 37:295–303. 32. Davison P, Asch RG (2011) Plastic ingestion by mesopelagic fishes in the North Pacific
14. Teuten EL, et al. (2009) Transport and release of chemicals from plastics to the en- Subtropical Gyre. Mar Ecol Prog Ser 432:173–180.
vironment and to wildlife. Philos Trans R Soc Lond B Biol Sci 364(1526):2027–2045. 33. Eriksson C, Burton H (2003) Origins and biological accumulation of small plastic
15. Hirai H, et al. (2011) Organic micropollutants in marine plastics debris from the open particles in fur seals from Macquarie Island. Ambio 32(6):380–384.
ocean and remote and urban beaches. Mar Pollut Bull 62(8):1683–1692. 34. Lam V, Pauly D (2005) Mapping the global biomass of mesopelagic fishes. Sea Around
16. Thompson RC, et al. (2004) Lost at sea: Where is all the plastic? Science 304(5672):838. Us Proj Newsl 30:4.
17. Law KL, et al. (2014) Distribution of surface plastic debris in the eastern pacific ocean 35. Irigoien X, et al. (2014) Large mesopelagic fishes biomass and trophic efficiency in the
from an 11-year data set. Environ Sci Technol 48(9):4732–4738. open ocean. Nat Commun 5:3271.
18. Doyle MJ, Watson W, Bowlin NM, Sheavly SB (2011) Plastic particles in coastal pelagic 36. Robison BH, Bailey TG (1981) Sinking rates and dissolution of midwater fish fecal
ecosystems of the Northeast Pacific ocean. Mar Environ Res 71(1):41–52. matter. Mar Biol 65:135–142.
19. Reisser J, et al. (2013) Marine plastic pollution in waters around Australia: Charac- 37. Pham CK, et al. (2014) Marine litter distribution and density in European seas, from
teristics, concentrations, and pathways. PLoS ONE 8(11):e80466. the shelves to deep basins. PLoS ONE 9(4):e95839.
20. Center for International Earth Science Information Network (2012) National Ag- 38. Van Cauwenberghe L, Vanreusel A, Mees J, Janssen CR (2013) Microplastic pollution
gregates of Geospatial Data: Population, Landscape and Climate Estimates Version 3 in deep-sea sediments. Environ Pollut 182:495–499.
(National Aeronautics and Space Administration Socioeconomic Data and Applica- 39. Kukulka T, Proskurowski G, Morét-Ferguson S, Meyer DW, Law KL (2012) The effect of
tions Center, Palisades, NY). Available at http://sedac.ciesin.columbia.edu/data/set/ wind mixing on the vertical distribution of buoyant plastic debris. Geophys Res Lett
nagdc-population-landscape-climate-estimates-v3. Accessed October 16, 2012. 39(7):L07601.

10244 | www.pnas.org/cgi/doi/10.1073/pnas.1314705111 Cózar et al.