Microplastics in Four Estuarine Rivers in the Chesapeake Bay, U.S.A. Lance T

Article

pubs.acs.org/est

Microplastics in Four Estuarine Rivers in the Chesapeake Bay, U.S.A. Lance T. Yonkos,*,†,‡ Elizabeth A. Friedel,‡ Ana C. Perez-Reyes,† Sutapa Ghosal,§ and Courtney D. Arthur∥,⊥,∇

† Department of Environmental Science and Technology, University of Maryland, College Park, Maryland 20742, United States ‡ Wye Research and Education Center, University of Maryland, Queenstown, Maryland 21658, United States § Environmental Health Laboratory, California Department of Public Health, Richmond, California 94804-6403, United States ∥ Marine Debris Program, National Oceanic and Atmospheric Administration, Silver Spring, Maryland 20910, United States ⊥ I.M. Systems Group, Rockville, Maryland 20852, United States

*S Supporting Information

ABSTRACT: Once believed to degrade into simple compounds, increasing evidence suggests plastics entering the environment are mechanically, photochemically, and/or biologically degraded to the extent that they become imperceptible to the naked eye yet are not significantly reduced in total mass. Thus, more and smaller plastics particles, termed microplastics, reside in the environment and are now a contaminant category of concern. The current study tested the hypotheses that microplastics concentration would be higher in proximity to urban sources, and vary temporally in response to weather phenomena such as storm events. Triplicate surface water samples were collected approximately monthly between July and December 2011 from four estuarine tributaries within the Chesapeake Bay, U.S.A. using a manta net to capture appropriately sized microplastics (operationally defined as 0.3−5.0 mm). Selected sites have watersheds with broadly divergent land use characteristics (e.g., proportion urban/suburban, agricultural and/or forested) and wide ranging population densities. Microplastics were found in all but one of 60 samples, with concentrations ranging over 3 orders of magnitude (<1.0 to >560 g/ km2). Concentrations demonstrated statistically significant positive correlations with population density and proportion of urban/suburban development within watersheds. The greatest microplastics concentrations also occurred at three of four sites shortly after major rain events.

■ INTRODUCTION Once in aquatic systems, plastics provide substratum for Marine microplastics, a type of marine debris generally sorption of various contaminants including persistent bio-

Downloaded via OLD DOMINION UNIV on February 27, 2019 at 01:37:53 (UTC). accepted as all plastic particles measuring ≤5.0 mm in size, accumulative organics (e.g., PCBs, PBDEs, etc.) and metals 1,6,13−16 have become a growing concern in North American waters (e.g., Cu, Pb). While at sea, plastic fragments can fi See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. since they were rst reported in southern New England coastal accumulate various attachment organisms (e.g., barnacles, waters in the 1970s.1 Across numerous studies, plastic particles polychaete worms, vegetative cysts) potentially functioning to were found to vary in size, shape, chemical composition, and transport invasive17 or pathogenic species.18 As plastics 2 spatial and temporal distribution. Plastics have been detected degrade, their small sizes, varied shapes, and conspicuous in surface waters around the globe and are likely the most 3 colors can lead to ingestion by a wide range of marine abundant form of marine debris. Low density plastics tend to fi fl ’ organisms, including planktonic sh and invertebrates, seabirds, remain buoyant and oat on the water s surface, allowing them 6,19,20 4 and cetaceans. Such ingestion is non-nutritive and raises to travel long distances. It is generally accepted that most 21−23 microplastics in aquatic systems derive from secondary sources obstruction concerns. Moreover, the potential for toxicity (e.g., progressively reductive fragmentation of larger material) arises from leaching of plastic constituents or sorbed contaminants after ingestion and subsequent biomagnification rather than from primary sources (e.g., manufactured particles − and pelletized preproduction materials).5,6 Patterns in variety within aquatic food webs.14,24 27 and concentration suggest proximity to human population 7,8 centers, river mouths, prevailing ocean currents, and weather Received: July 25, 2014 events all influence introduction and dispersion of plastics into Revised: November 7, 2014 marine systems, and determine regions of eventual accumu- Accepted: November 12, 2014 − lation.3,9 12 Published: November 12, 2014

While microplastic marine debris has been extensively sites (Patapsco, Magothy, Rhode, and Corsica Rivers) were − documented in Atlantic and Pacific Ocean gyres,27 29 similar located within navigable tidal regions of estuarine rivers that studies in estuarine and near-shore waters are comparatively varied substantially in watershed size and land use character- scarce.8,10,31,32 The purpose of this study was to collect and istics (Table 1). Watershed margins were delineated using quantify microplastics in surface waters of the Chesapeake Bay to compare with historical concentrations and to establish Table 1. Characteristics of Watersheds Proximate to modern baselines for future monitoring. Collection was Chesapeake Bay Surface Waters Sampled for Microplastics a directed toward presumed sources of microplastics (i.e., between June and December 2011 estuarine regions) rather than eventual regions of accumulation Patapsco Magothy Rhode Corsica (e.g., midocean gyres). The study was also designed to watershed characteristics River River River River investigate spatial and temporal trends within and between population 899 000 32 350 4300 3500 sample locations. Collections occurred across multiple seasons watershed area (km2) 1637 92 67 97 with sample sites chosen according to land use characteristics tidal river/bay area (km2) 123 21.8 12.8 5.6 and urban intensity within watersheds. It was hypothesized that population density 550 351 64 36 microplastics concentrations would: (1) be greater near urban (persons/km2) sources and (2) vary temporally in response to climate/weather total developed (%) 54 59 12 13.5 phenomena. As the largest estuary in the U.S., the Chesapeake urban/industrial (%) 28 5 0 3.1 Bay and associated watershed currently accommodate a human suburban/residential (%) 26 54 12 10.4 population of over 17 million, provide habitat for numerous agricultural/pasture (%) 18 0.5 16 60.4 and diverse fish, waterfowl, and migratory shorebird popula- forested (%)b 17 32 68 24.4 tions, and sustain culturally and economically important aCatchments delineated using digital USGS quarter quad topographic commercial and recreational fin and shellfish fisheries. Under- maps; area calculated using ArcMap 10.0.3 (ESRI, Redlands, standing sources, transport, fate, and potential impacts of California, U.S.A.); land use delineation determined using the 2006 microplastics on the Chesapeake Bay and its resources is an National Land Cover Database;29 population based on 2010 U.S. 30 b important step to further the state of the science about Census. Sum of land use from given categories does not equal microplastics as an emerging pollution issue. 100%; other land use types (e.g., woody wetlands, open water) account for the remainder. ■ MATERIALS AND METHODS digital US Geological Survey quarter quad topographic maps Sample Sites. Surface waters from four estuarine tributaries with catchment areas calculated in ArcMap 10.0.3 (ESRI, within the Chesapeake Bay, U.S.A. were sampled for Redlands, California, U.S.A.). Land cover (percent agricultural, enumeration of microplastics (Figure 1). The four sample urban/suburban, and forest) was calculated for catchments using the 2006 National Land Cover Database.33 Population within watersheds was extrapolated from 2010 U.S. census data.34 The Patapsco River was by far the largest and most urban of estuarine sites, with a population density of 550 persons per sq. km within a 1637 km2 watershed (total population 899 000; Table 1). While the Upper Patapsco watershed is largely rural and forested, the lower Patapsco and Patapsco River Basin contain the entirety of Baltimore, MD, a densely urban city and historically industrial harbor. The three other estuarine systems had significantly smaller watersheds (range 67−97 km2) with varying population densities and proportions of suburban, agricultural and forested land use: the Magothy, predominantly residential (59%); the Corsica, predominantly agricultural (60%); and the Rhode, predominantly forested (78%). Sample Collection, Preparation, Processing, and Quantification. Samples were collected by surface trawl (1.0−2.0 km) using a 70 cm wide manta net with a mesh size of 0.33 mm, designed to capture floating material to a depth of 15 cm (Figure 2A).35 Sites were surveyed by collecting contents from triplicate surface trawls on five discrete occasions between July and December 2011. Therefore, each estuarine site had a total of 15 samples examined for microplastics. Materials captured by trawl were passed through nested 5.0 mm and 0.3 mm stainless steel sieves. Debris retained on the 0.3 mm screen was rinsed using site water into prelabeled jars for future processing and analysis. Trawl distance, water volume sampled, weather (e.g., air temperature, wind speed and direction, recent precipitation), and water quality parameters (e.g., surface water Figure 1. Locations of estuarine sites within the Chesapeake Bay temperature, salinity) were recorded during each sampling sampled for microplastics between July and December 2011. event.

14196 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article

weights calculated to the nearest 0.01 mg. The number of microplastic particles collected was also determined for each trawl by counting total particles collected. Final environmental concentrations of microplastics were calculated both as mass per unit surface area (g/km2) and particles per unit surface area (pieces/km2). Detected microplastics were sorted by category into groups that included: synthetic fibers, thin flexible sheets, hard multicolored fragments of various sizes, preproduction pellets, and extruded polystyrene (e.g., Styrofoam). Validation of visually based micrplastics identification was achieved using Raman microspectroscopy (RMS)analysis,avibrational spectroscopy technique providing compositional information with micrometer-scale spatial resolution, following the methods of Ghosal and Wagner.37 Ten randomly selected small fragments (≤2 mm; the particle category deemed most likely to be misidentified) were analyzed by RMS using a Renishaw inVia Raman microscope (Renishaw Plc., Old Town, Wotton- under-Edge, Gloucestershire, U.K.) equipped with a near- infrared 785 nm diode laser. Statistical Analysis. Spatial and temporal collection of microplastics from Chesapeake Bay sites provided data amendable to comparison by two-way analysis of variance Figure 2. Collection and isolation of microplastics in Chesapeake Bay (ANOVA) using site and sample date as variables. Concen- surface waters via trawl by manta net (A) (photo courtesy Versar, Inc., trations calculated both by microplastic weight (g/km2) and Columbia, MD, U.S.A.); hydrogen peroxide digestion of labile organics 2 ‰ number of particles (pieces/km ) were analyzed by two-way (B); and hyper-saline (300 ) density separation (C). ANOVA followed by an all-pairwise multiple comparison procedure (Holm-Sidak method) to investigate statistical ff Materials collected from all trawls were manipulated to di erences between sites and/or sample dates. Pearson Product isolate and identify microplastics using a combination of Moment Correlation analysis was employed to investigate the chemical, thermal, physical, and mechanical processes (Figure association between microplastics concentration and various 2B,C) modified from the methods of Baker et al.36 Briefly, watershed characteristics (e.g., population density, land use samples were rinsed onto 0.3 mm stainless steel sieves using prevalence). All analyses were performed using SigmaStat fi deionized (DI) water and then transferred to preweighed 600 version 3.5 (Systat Software Inc.) with statistical signi cance mL glass beakers where they were oven-dried at 75 °C for reported at p = 0.05. approximately 24 h. After recording dry weights, contents were treated with 20 mL aliquots each of a 0.05 M Fe(II) solution ■ RESULTS and 30% hydrogen peroxide (H2O2) to facilitate chemical Microplastics were found in all but one of 60 samples (Corsica, digestion of labile organic material (Figure 2B). Mixtures were December 1, 2011) (Figure 3). Where detected, concentrations placed on heated stir-plates maintained at 75 °C and gently stirred. At 30 min intervals, samples were re-examined and additional H2O2 added and heating repeated, as necessary, until all organic material was digested. After rinsing with DI water to remove residual H2O2, sieve contents were rinsed using a hyper-saline solution (300 g/L table salt in DI water) into glass funnels with attached and clamped flexible tubing (Figure 2C). Funnels were filled with the hyper-saline solution, covered, and left undisturbed for approximately 1 h to allow density-separation of remaining solids. After completion of material separation, settled debris was dried in 50 mL beakers, transferred to 110 mm glass Petri Figure 3. Microplastics collected from individual trawls of Chesapeake dishes, and visually examined using a dissecting scope at 7.5× Bay surface waters during the summer and fall 2011 demonstrating the magnification (Stereomaster SKZ-5) to determine presence/ diversity in abundance and variety of constituents including fragments absence of microplastics. Remaining floating materials were of various sizes, preproduction pellets/cylinders, synthetic fibers, thin fl collected separately onto 0.3 mm stainless steel sieves, rinsed exible sheets, and extruded polystyrene (A, Patapsco 8/2; B, Magothy with DI water, covered loosely with aluminum foil and dried at 9/20). ambient temperature for 24 to 48 h. Once dry, sieve contents were examined for microplastics by dissecting scope. Putative of microplastics were generally low and variable between plastics were carefully separated from residual woody/organic replicates and across sample locations and time periods (Table debris, removed using microforceps, placed into prelabeled 2). Measured concentrations ranged over 3 orders of aluminum weigh-boats (preweighed to the nearest 0.01 mg), magnitude (<1.0 g/km2 to 563 g/km2). Plots of microplastics and dried at 75 °C for 12 to 24 h. Total foil weights were concentration (g/km2) (Figure 4) show generally greater determined on a microbalance and resulting microplastics abundances in Patapsco River samples and comparatively lower

14197 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article

Table 2. Microplastics in Chesapeake Bay and Coastal Mid- a Atlantic Surface Waters During Summer and Fall 2011

microplastics concentration sample pieces/km2 (mean ± g/km2 (mean ± site date reps SD) SD) Patapsco 7/7 3 103.3 ± 118.03 8/2 3 130 867 ± 46 258 77.2 ± 12.55 9/19 3 297 927 ± 180 252 238.1 ± 286.45 10/12 3 59 782 ± 27 323 10.6 ± 9.34 11/30 3 132 918 ± 137 518 81.2 ± 105.95 Magothy 7/7 3 28.9 ± 25.34 8/2 3 99 129 ± 19 965 18.2 ± 5.96 9/20 3 259 803 ± 60 150 245.7 ± 271.72 10/12 3 55 416 ± 45 507 27.3 ± 39.95 11/30 3 36 013 ± 20 116 5.2 ± 3.10 Rhode 7/6 3 5.8 ± 3.41 8/3 3 37 415 ± 33 665 3.4 ± 4.67 9/20 3 131 978 ± 118 895 56.1 ± 87.32 10/13 3 81 910 ± 47 344 9.5 ± 7.41 12/1 3 18 574 ± 13 347 3.2 ± 2.21 Corsica 7/6 3 17.0 ± 9.95 8/3 3 92 617 ± 59 645 19.2 ± 16.06 9/20 3 54 539 ± 32 719 11.5 ± 7.03 10/13 3 10 717 ± 6973 2.7 ± 0.86 12/1 3 5534 ± 5134 4.0 ± 5.59 aValues in bold are significantly different within site (ANOVA followed by Holm-Sidak all-pairwise multiple comparison; p = 0.05). abundances in Rhode and Corsica River samples. With the exception of the Corsica River, all sites had peak mean microplastics concentrations during September sampling Figure 4. Concentrations of microplastics in surface water collections (Figure 4). Two-way ANOVA did not indicate a statistically from four Chesapeake Bay tributaries on five occasions between July significant interaction between sample location and sample time and December 2011; mean (log scale; n = 3) and standard deviation on microplastics concentration (p = 0.175) (two-way ANOVA (error bars). results provided in Supporting Information, SI, Table S1). The power of the test, however, was only 0.219 (at α = 0.05), Again, this result was supported whether comparing abundan- suggesting that the negative result be interpreted with caution. ces based on microplastics weight (Figure 5, top right) or Individually, spatial and temporal variables were both number of particles (Figure 5, bottom right). confirmed to have highly statistically significant effects on There was considerable variation in the types of anthro- resulting microplastics concentrations (sample location p = pogenic materials recovered within and between sites (SI Table <0.001; sample time p = <0.001) (SI Table S1). Pairwise S2). Small plastic fragments (0.3−2.0 mm) and flexible sheets comparisons indicate that samples from both Patapsco and were the most frequently encountered materials followed by Magothy Rivers had greater abundances of microplastics than synthetic fibers, extruded polystyrene and larger fragments did samples from Corsica and Rhode Rivers (p < 0.05). The (2.0−5.0 mm). Preproduction pellets occurred with some relationship between microplastics concentration and sample regularity in Patapsco River samples but were generally less location was statistically significant whether comparing abundant or entirely absent at other sites. abundances based on microplastics weight (Figure 5, top left) All randomly selected particles examined by RMS produced or number of particles (Figure 5, bottom left). This result spectra with distinct polyethylene-related peaks, confirming the supports the hypothesis that microplastics concentrations fragments as pieces of polymers (Figure 7). Several particles increase with proximity to more densely urban/suburban displayed additional peaks identified as secondary constituents areas. Pearson Product Moment Correlation analysis of pooled through spectral matching (Figure 7, SI Figure S1). Two data from the four sites further supports this relationship; particles (numbers 4 and 6) contain peaks indicative of cobalt statistically significant positive correlations are found between phthalocyanine, a commercially available dye. The presence of microplastics concentrations and population density (r = 0.33; this dye is consistent with the brightly colored appearances of p = 0.0096) and proportion of urban/suburban development (r the particles. Particle 7, which appears black, is dominated by = 0.32; p = 0.0142) within watersheds (Table 3; Figure 6). two broad peaks indicative of black carbon and is likely a piece Results of pairwise comparisons of sample times were less of burnt plastic. definitive but generally supportive of September sampling yielding greater microplastics concentrations than other sample ■ DISCUSSION times. September peaks were found to be significantly greater Overall, our data are consistent with global reports that than later samples (Oct and Nov/Dec) but not to be microplastics are ubiquitous in marine environments but vary statistically greater than earlier samples (July and Aug). considerably spatially, temporally, and in material composi-

14198 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article

Figure 5. Microplastics concentrations in Chesapeake Bay surface waters sampled between July and December 2011 with values reported as mass/ km2 (top) and pieces/km2 (bottom; July data on pieces/km2 was not available). Concentrations were found to differ significantly when compared both by estuarine system (blue) and by sample date (pink) (two-way ANOVA followed by Holm-Sidak all-pairwise multiple comparison; groups that share the same letter do not differ significantly from each other; p = 0.05). Boxes display median, 25th and 75th percentiles; whiskers display 10th and 90th percentiles.

Table 3. Pearson Product Moment Correlations of (e.g., salinity and temperature),41 as well as climatic and/or Microplastics Concentration with Population and Land Use meteorological conditions (e.g., rain, prevailing and episodic a Characteristics wind),11,42 reviewed by Browne et al.10 Consequences include stratification and differential expression of various microplastics watershed characteristic rp types vertically within the water column11,42 and horizontally population density 0.332 0.0096 from source to sink.43 Such stratification may lead to collection total developed (%) 0.315 0.0142 bias depending on depth of sampling apparatus, proximity to urban/industrial (%) 0.271 0.0360 source(s), and recent wind, rain, or wave energy. suburban/residential (%) 0.237 0.0681 Processing methodologies used to isolate microplastics are agricultural/pasture (%) −0.200 0.125 also variable employing one or more of organic digestion, forested (%) −0.189 0.148 a ffi density separation, sequential sieving, and/or visual sorting Positive correlation coe cients (r) indicate variables tend to increase steps.39 Subsequent identification of microplastics may be together while negative correlation coefficients indicate one variable performed either visually using light or electron micros- tends to decrease while the other increases. Pairs with p-values below 29,41,44 0.050 (bold) are statistically significant. copy, or by analytical methodologies suitable for positive identification and distinction of specific polymers (e.g., Raman spectroscopy (RMS); Fourier-transform infrared spectrosco- tion.9,38 Interstudy comparisons must, however, be made with py).9,37,45,46 In our study, samples were sorted and selected caution as significant inconsistencies in study design result from manually based on visual characteristics with a subset examined differences in sample collection, isolation, enumeration, and 5,39 by RMS for corroboration. While the number of examined reporting protocols. For example, the minimum size of fi captured microplastics in surface and pelagic tow collections is particles was small (n = 10), con rmation that 100% of particles ultimately determined by mesh size, which varies broadly across randomly selected from the most problematic category (small fragments), serves as reasonable validation of visually based studies. In a review of methods employed in collection of fi microplastics in surface waters, the mesh size of our manta net microplastics identi cation and suggests the majority of (0.33 mm) was the most frequently used among 33 studies (n = collected particles are actually plastics. fi 13), but mesh sizes ranged over 2 orders of magnitude from Despite the lack of uniformity in collection and quanti ca- 0.05 mm to 3.0 mm.5,39 Retention of smaller plastics is tion of microplastics in marine and inland waters, certain trends necessarily skewed by filtration sampling. Depth of sample emerge. Studies in U.S. waters and elsewhere generally support collection (based on net size and shape and trawl practice) can a positive correlation between microplastics quantities and also impact capture results. Presumably, buoyant materials will proximity to densely populated or industrial areas, which is not reside at the surface while others will submerge and eventually surprising given the anthropogenic origin of plastic materi- sink. This, however, is a dynamic process influenced variously als.43,44 Eriksen et al.44 reported greater concentrations of by characteristics of the material (e.g., polymer composition, plastics from surface trawls in Lake Erie near the population size, shape, and density)40 and of the aqueous environment centers of Buffalo, NY, Erie, PA, and Cleveland, OH, compared

14199 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article

and Corsica Rivers, with smaller watersheds, little industry, and considerably lower population densities, had significantly lower plastics concentrations. Substantial impervious surfaces, hard- ened and channelized drainage systems, combined sewer overflows, shipping traffic, and industrial waste discharges are all likely contributors to the large quantities of microplastics entering the Patapsco River.47 The Magothy River watershed, intermediate in character with primarily residential land use, moderate population density and little industry, contained the second highest amount of microplastics. The Rhode and Corsica River watersheds are significantly smaller than the Patapsco River watershed and dominated by forest or row crop agriculture, respectively. More land in forested or agricultural land use generally means lower population density, less development, less impervious surfaces, and fewer storm drains. While not found to be statistically significant, evidence from our study hints at such negative correlations between receiving water microplastics concentrations and proportions of watersheds in agricultural (r = −0.20; p = 0.125) and forested (r = −0.19; p = 0.148) land use (Figure 6B; Table 3). Microplastics quantification and distribution data for comparison in estuarine waters of the Mid-Atlantic are very limited. The only Chesapeake Bay locations with microplastics collections reported in the scientific literature are harbor sites in Baltimore, MD and Norfolk, VA samples two decades ago.47 These harbors were specifically targeted because of their numerous combined sewer/stormwater outfalls, high degree of industrialization, and substantial shipping traffic.47 Not Figure 6. Associations between watershed characteristics and micro- surprisingly, both were found to contain abundant plastics plastics concentrations in Chesapeake Bay surface waters: population density (top); land use patterns (bottom); positive correlation (macro and micro) with Baltimore ranking third among 11 U.S. coefficients (r) indicate variables that tend to increase together while harbors for all plastics and for preproduction pellets. The negative correlation coefficients indicate that one variable tends to Houston Shipping Channel had the greatest concentration of decrease while the other increases; only variable pairs with p-values plastics debris among sampled areas, the preponderance (98%) below 0.050 (e.g., population density, urban/suburban land use) are being preproduction plastic pellets.47 At the time, Houston statistically significant. maintained one of the densest plastics production industries in the U.S. with numerous pellet extruding and processing facilities on or proximate to the shipping channel.47 Because the researchers used neuston nets of varying dimensions (e.g., 1 × 2 m; 0.5 × 1 m), collected samples over inconsistent trawl distances, and reported combined micro- and macroplastic concentrations, direct comparison to our study is not possible. To our knowledge, surface water collection from less urban/ industrial Chesapeake Bay waters for the purpose of microplastics detection is completely lacking from the scientific literature. Microplastics concentration and dispersion to surface waters is influenced by wind10,11,48 and rain.42,49 Extreme meteorological events, such as hurricanes and flash flooding are Figure 7. Raman microspectroscopy (RMS) analysis of randomly particularly adept at hastening the transfer of terrestrial debris selected small fragments (<2.0 mm) from Chesapeake Bay surface 9,45 50 water sampling. (A) Stereozoom image of particles mounted on from land to sea. For example, Moore et al. report that adhesive carbon tab on a standard SEM stub and (B) acquired RMS neustonic litter (i.e., surface plastic debris <4.75 mm in spectra of individual particles (numbers correspond to stereozoom diameter) in California coastal surface waters near the mouth of image) along with the reference spectrum for polyethylene (PE). a modified Los Angeles stormwater conveyance system increased 6-fold following a storm. The authors also describe how the precipitation-induced increase in water volume within to significantly less populous regions of Lakes Huron and the river resulted in litter being transported to greater distances Superior. We note a similar relationship in our study where from the river mouth. In a subsequent study measuring correlation analysis indicates positive associations between neustonic plastic at different depths along the California coast, measures of urban intensity (i.e., population density and urban/ Lattin et al.42 found that the concentration of surface plastics development) and microplastics concentration (Figure 6A,B; increased substantially (e.g., from <1.0 pieces/m3 to 18 pieces/ Table 3). The Patapsco River, receiving input from the densely m3) following a storm event, and that the increase was populated and industrial Baltimore Harbor, consistently had the particularly noticeable at a near-shore site. The authors highest overall concentrations of microplastics, while the Rhode conclude that heavy rains increase the amount of plastics

14200 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article debris entering coastal systems, and that strong winds and ■ REFERENCES associated wave action create a vertical mixing within the water (1) Carpenter, E. J.; Anderson, S. J.; Harvey, G. R.; Miklas, H. P.; column that temporarily resuspends plastics. The notion that Peck, B. B. Polystyrene spherules in coastal waters. Science 1972, 17, winds play a significant role in the dispersion of plastics was 749−750. 10 also supported by Browne et al., who found that downwind (2) Wright, S. L.; Thompson, R. C.; Galloway, T. S. The physical estuarine sites generally contained >3 times more microplastics impacts of microplastics on marine organisms: a review. Environ. Pollut. debris than upwind sites. In our study, we detected peaks in 2013, 178, 483−492. microplastics concentrations at three of the four sample site (3) Zettler, E. R.; Mincer, T. J.; Amaral-Zettler, A. Life in the “plastisphere”: microbial communities on plastic marine debris. shortly after two closely occurring major storm events, − Hurricane Irene and Tropical Storm Lee, which brought Environ. Sci. Technol. 2013, 47, 7137 7146. heavy winds and collectively deposited as much as 90 cm (36 (4) Morris, R. J. Plastic debris in the surface waters of the South Atlantic. Mar. Pollut. Bull. 1980, 11 (6), 164−166. in.) of precipitation throughout much of the Chesapeake Bay 51 (5) Arthur, C.; Baker, J.; Bamford, H., Eds. Proceedings of the watershed between August 26 and September 8, 2011. International Research Workshop on the Occurrence, Effects, and Fate of Microplastics peaks observed in September samples were likely Microplastic Marine Debris. Sept 9−11, 2008. NOAA Technical attributable to substantial runoff of terrestrial debris as well as Memorandum NOS-OR&R-30, 2009. resuspension of plastics already in aqueous systems. The only (6) Moore, C. J. Synthetic polymers in the marine environment: A site where microplastics concentration was apparently un- rapidly increasing long-term threat. Environ. Res. 2008, 108, 131−139. affected by these storm events was the Corsica River; possibly (7) Sadri, S. S.; Thompson, R. C. On the quantity and composition of ’ floating plastic debris entering and leaving the Tamar Estuary, because of the watershed s agricultural dominance, low − population density, and limited impervious surfaces. Southwest England. Mar. Pollut. Bull. 2014, 81 (1), 55 60. Results of this study support prioritization of upstream (8) Rech, S.; Macaya-Caquilpan, V.; Pantoja, J. F.; Rivadeneira, M. M.; Jofre Madariaga, D.; Thiel, M. Rivers as sources of marine litter prevention efforts in urban locations, such as the proposed − 52 A study from the SE Pacific. Mar. Pollut. Bull. 2014, 82,66 75. California State Trash Amendments. Studies such as this one (9) Barnes, D. K. A.; Galgani, F.; Thompson, R. C.; Barlaz, M. are critical for establishing baselines for future aquatic pollution Accumulation and fragmentation of plastic debris in global environ- and water quality monitoring, to accurately describe the hazard ments. Philos. Trans. R. Soc., B 2009, 364, 1985−1998. and develop effective prevention and management options. (10) Browne, M. A.; Galloway, T. S.; Thompson, R. C. Spatial Further attention should be given to study designs focused on patterns of plastic debris along estuarine shorelines. Environ. Sci. elucidating patterns in temporal and spatial variability to Technol. 2010, 44, 3404−3409. develop improved models of microplastic sources and fate (11) Kukulka, T.; Proskurowski, G.; Moret-Ferguson, S.; Meyer, D. within estuarine and coastal areas. W.; Law, K. L. The effect of wind mixing on the vertical distribution of buoyant plastic debris. Geophys. Res. Lett. 2012, 39 (7), L07601. (12) Potemra, J. T. Numerical modeling with application to tracking ■ ASSOCIATED CONTENT marine debris. Mar. Pollut. Bull. 2012, 65,42−50. *S Supporting Information (13) Ashton, K.; Holmes, L.; Turner, A. Association of metals with plastic production pellets in the marine environment. Mar. Pollut. Bull. Tables reporting the results of two-way ANOVA tests; a table − reporting a semiquantitative assessment of microplastics variety 2010, 60, 2050 2055. fi fi (14) Koelmans, A. A.; Besseling, E.; Foekema, E. M. Leaching of and abundance; and a gure showing analysis and identi cation plastic additives to marine organisms. Environ. Pollut. 2014, 187,49− of additional RMS peaks from microplastics particles. This 54. material is available free of charge via the Internet at http:// (15) Rochman, C. M.; Hoh, E.; Kurobe, T.; Teh, S. J. Ingested plastic pubs.acs.org/. transfers hazardous chemicals to fish and induces hepatic stress. Sci. Rep. 2013, 3, 3263. ■ AUTHOR INFORMATION (16) Velzeboer, I.; Kwadijk, C. J. A. F.; Koelmans, A. A. Strong Sorption of PCBs to Nanoplastics, Microplastics, Carbon Nanotubes, Corresponding Author and Fullerenes. Environ. Sci. Technol. 2014, 48 (9), 4869−4876. *Phone: (301) 405-7871; fax: (301) 314-9023; e-mail: (17) Gregory, M. R. Environmental implications of plastic debris in [email protected]. marine settings - entanglement, ingestion, smothering, hangers-on, Present Address hitch-hiking and alien invasions. Philos. Trans. R. Soc. B 2009, 364, ∇ 2013−2025. Industrial Economics, Inc. Cambridge, Massachusetts U.S.A. (18) Maso, M.; Garces, E.; Pages, F.; Camp, J. Drifting plastic debris Notes as a potential vector for dispersing Harmful Algal Bloom (HAB) The authors declare no competing financial interest. species. Sci. Mar. 2003, 67 (1), 107−111. (19) Cole, M.; Lindeque, P.; Halsband, C.; Galloway, S. C. ■ ACKNOWLEDGMENTS Microplastics as contaminants in the marine environment: a review. Mar. Pollut. Bull. 2011, 62, 2588−2597. This study was supported by the National Oceanic and (20) Foekema, E. M.; De Gruijter, C.; Mergia, M. T.; van Franeker, J. Atmospheric Administration (NOAA), and funded by a grant A.; Murk, A. J.; Koelmans, A. A. Plastic in North Sea Fish. Environ. Sci. from the National Marine Sanctuary Foundation (NMSF). The Technol. 2013, 47 (15), 8818−8824. scientific results and conclusions, as well as any views or (21) Browne, M. A.; Dissanayake, A.; Galloway, T. S.; Lowe, D. M.; opinions expressed herein, are those of the authors and do not Thompson, R. C. Ingested microscopic plastic translocates to the reflect the views or policies of NOAA or the U.S. Department circulatory system of the mussel, Mytilus edulis (L.). Environ. Sci. Technol. 2008, 42, 5026−5031. of Commerce. The photograph of the manta net was provided (22) Murray, F.; Cowie, P. R. Plastic contamination in the decapod by Versar, Inc., Columbia, MD, U.S.A. Authors thank Sherry crustacean Nephrops norvegicus (Linnaeus, 1758). Mar. Pollut. Bull. Lippiatt and Alan Mearns at NOAA for their constructive 2011, 62, 1207−1217. comments during an initial review of the manuscript and ES&T (23) Lee, K.-W.; Shim, W. J.; Kwon, O. Y.; Kang, J.-H. Size- reviewers for their many helpful suggestions. dependent effects of micro polystyrene particles in the marine

14201 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202 Environmental Science & Technology Article copepod Tigriopus japonicas. Environ. Sci. Technol. 2013, 47 (19), depths near the southern California shore. Mar. Pollut. Bull. 2004, 49, 11278−11283. 291−294. (24) Cole, M.; Lindeque, P.; Fileman, E.; Halsband, C.; Goodhead, (43) Browne, M. A.; Crump, P.; Niven, S. J.; Teuten, E. L.; Tonkin, R.; Moger, J.; Galloway, T. S. Microplastic ingestion by zooplankton. A.; Galloway, T.; Thompson, R. C. Accumulations of microplastic on Environ. Sci. Technol. 2013, 47, 6646−6655. shorelines worldwide: sources and sinks. Environ. Sci. Technol. 2011, (25) Teuten, E. L.; Saquing, J. M.; Knappe, D. R. U.; Barlaz, M. A.; 45, 9175−9179. Jonsson, S.; Bjorn, A.; Rowland, S. J.; Thompson, R. C.; Galloway, T. (44) Eriksen, M.; Mason, S.; Wilson, S.; Box, C.; Zellers, A.; Edwards, S.; Yamashita, R.; Ochi, D.; Watanuki, Y.; Moore, C.; Pham, H. V.; W.; Farley, H.; Amato, S. Microplastic pollution in the surface waters Tana, T. S.; Prudente, M.; Boonyatumanond, R.; Zakaria, M. P.; of the Laurentian Great Lakes. Mar. Pollut. Bull. 2013, 77, 177−182. Akkhavong, K.; Ogata, Y.; Hirai, H.; Iwasa, S.; Mizukawa, K.; Hagino, (45) Thompson, R. C.; Olsen, Y.; Mitchell, R. P.; Davis, A.; Rowland, Y.; Imamura, A.; Saha, M.; Takada, H. Transport and release of S. J.; John, A. W. G.; McGonigle, D.; Russell, A. E. Lost at sea: Where chemicals from plastics to the environment and to wildlife. Philos. is all the plastic? Science 2004, 304 (5672), 838−838. Trans. R. Soc., B 2009, 364 (1526), 2027−2045. (46) Doyle, M. J.; Watson, W.; Bowlin, N. M.; Sheavly, S. B. Plastic (26) Rochman, C. M.; Hoh, E.; Hentschel, B. T.; Kaye, S. Long-term particles in coastal pelagic ecosystems of the Northeast Pacific ocean. field measurement of sorption of organic contaminants to five types of Mar. Environ. Res. 2011, 71,41−52. plastic pellets: Implications for plastic marine debris. Environ. Sci. (47) Redford, D. P.; Trulli, W. R.; Trulli, H. K. Composition of Technol. 2013, 47 (3), 1646−1654. floating debris in harbours of the United States. Chem. Ecol. 1992, 7, (27) Besseling, E.; Wegner, A.; Foekema, E. M.; van den Heuvel- 75−92. Greve, M. J.; Koelmans, A. A. Effects of microplastic on fitness and (48) Collignon, A.; Hecq, J. H.; Glagani, F.; Voisin, P.; Collard, F.; PCB bioaccumulation by the lugworm Arenicola marina (L.). Environ. Goffart, A. Neustonic microplastic and zooplankton in the North Sci. Technol. 2013, 47 (1), 593−600. Western Mediterranean Sea. Mar. Pollut. Bull. 2012, 64, 861−864. (28) Law, K. L.; Moret-Ferguson, S.; Maximenko, N. A.; (49) Moore, C. J.; Lattin, G. L.; Zellers, A. F. Quantity and type of Proskurowski, G.; Peacock, E. E.; Hafner, J.; Reddy, C. M. Plastic plastic debris flowing from two urban rivers to coastal waters and accumulation in the North Atlantic subtropical gyre. Science 2010, 329, beaches of Southern California. J. Int. Coast. Zone Manage. 2011, 11 − 1185−188. (1), 65 73. (29) Law, K. L.; Moret-Ferguson, S.; Goodwin, D. S.; Zettler, E. .; (50) Moore, C. J.; Moore, S. L.; Weisberg, S. B.; Lattin, G. L.; Zellers, DeForce, E.; Kukulka, T.; Proskurowski, G. Distribution of surface A. F. A comparison of neustonic plastic and zooplankton abundance in ’ − plastic debris in the eastern Pacific Ocean from an 11-year data set. southern California s coastal waters. Mar. Pollut. Bull. 2002, 44, 1035 Environ. Sci. Technol. 2014, 48 (9), 4732−4738. 1038. (30) Eriksen, M.; Maximenko, N.; Thiel, M.; Cummins, A.; Lattin, (51) NOAA National Weather Service: 2011 Atlantic Hurricane G.; Wilson, S.; Hafner, J.; Zellers, A.; Rifman, S. Plastic pollution in the Season http://www.nhc.noaa.gov/data/tcr/index.php?season= South Pacific subtropical gyre. Mar. Pollut. Bull. 2013, 68 (1−2), 71−6. 2011&basin=atl. (31) Carpenter, E. J.; Smith, K. L. Plastics on the Sargasso Sea (52) California Environmental Protection Agency, State Water surface. Science 1972, 175, 1240−1241. Resources Control Board Division of Water Quality. Draft Amend- (32) Colton, J. B.; Burns, B. R.; Knapp, F. D. Plastic particles in ments to Statewide Water Quality Control Plans to Control Trash. surface waters of the northwestern Atlantic. Science 1974, 185, 491− 2014. http://www.waterboards.ca.gov/water_issues/programs/trash_ 497. control/docs/trash_sr_061014.pdf. (33) Fry, J.; Xian, G.; Jin, S.; Dewitz, J.; Homer, C.; Yang, C.; Barnes, C.; Herold, N.; Wickham, J. Completion of the 2006 national land cover database for the conterminous United States. Photogram. Eng. Remote Sens. 2011, 77, 858−864. (34) US Census 2010 Interactive Population Map; http://www. census.gov/2010census/popmap/. (35) Lippiatt, S.; Opfer, S.; Arthur,C. Marine debris monitoring and assessment: Recommendation for monitoring debris trends in the marine environment; NOAA Technical Memorandum NOS-OR&R-46, 2013. (36) Baker, J.; Foster, G.; Masura, J.; Arthur, C. Laboratory methods for the analysis of microplastics in the marine environment; NOAA Technical Memorandum NOS-OR&R-48, 2014. (37) Ghosal, S.; Wagner, J. Correlated Raman micro-spectroscopy and scanning electron microscopy analyses of flame retardants in environmental samples: a micro-analytical tool for probing chemical composition, origin and spatial distribution. Analyst 2013, 138, 3836− 3844. (38) Goldstein, M. C.; Titmus, A. J.; Ford, M. Scales of spatial heterogeneity of plastic marine debris in the northeast Pacific Ocean. PLoS One 2013, 8 (11), e80020. (39) Hidalgo-Ruz, V.; Gutow, L.; Thompson, R. C.; Thiel, M. Microplastics in the marine environment: A review of the methods used for identification and quantification. Environ. Sci. Technol. 2012, 46, 3060−3075. (40) Thiel, M.; Hinojosa, I.; Vasquez, N.; Macaya, E. Floating marine debris in coastal waters of the SE-Pacific (Chile). Mar. Pollut. Bull. 2003, 46 (2), 224−231. (41) Andrady, A. L. Microplastics in the marine environment. Mar. Pollut. Bull. 2011, 62, 1596−1605. (42) Lattin, G. L.; Moore, C. J.; Zellers, A. F.; Moore, S. L.; Weisberg, S. B. A comparison of neustonic plastic and zooplankton at different

14202 dx.doi.org/10.1021/es5036317 | Environ. Sci. Technol. 2014, 48, 14195−14202