Deep-Sea Research I 79 (2013) 96–105

Contents lists available at SciVerse ScienceDirect

Deep-Sea Research I

0967-06http://d

☆ThisCommomits nothe orig

n CorrE-m

journal homepage: www.elsevier.com/locate/dsri

Debris in the deep: Using a 22-year video annotation database tosurvey marine litter in Monterey Canyon, central California, USA$

Kyra Schlining n, Susan von Thun, Linda Kuhnz, Brian Schlining, Lonny Lundsten,Nancy Jacobsen Stout, Lori Chaney, Judith ConnorMonterey Bay Aquarium Research Institute, 7700 Sandholdt Road, Moss Landing, California 95039-9644, USA

a r t i c l e i n f o

Article history:Received 15 January 2013Received in revised form3 May 2013Accepted 16 May 2013Available online 28 May 2013

Keywords:LitterDeep seaSubmarine canyonsGISUSACaliforniaMonterey CanyonUnmanned vehicles

37/$ - see front matter & 2013 The Authors. Px.doi.org/10.1016/j.dsr.2013.05.006

is an open-access article distributed undens Attribution-NonCommercial-No Derivativen-commercial use, distribution, and reproductinal author and source are credited.esponding author. Tel.: +1 8317751784; fax: +ail address: schlin@mbari.org (K. Schlining).

a b s t r a c t

Anthropogenic marine debris is an increasing concern because of its potential negative impacts onmarine ecosystems. This is a global problem that will have lasting effects for many reasons, including: (1)the input of debris into marine environments is likely to continue (commensurate with populationincrease and globalization), (2) accumulation, and possibly retention, of debris will occur in specific areasdue to hydrography and geomorphology, and (3) the most common types of debris observed to date willlikely persist for centuries. Due to the technical challenges and prohibitive costs of conducting researchin the deep sea, little is known about the abundance, types, sources, and impacts of human refuse on thisvast habitat, and the extreme depths to which this debris is penetrating has only recently been exposed.We reviewed 1149 video records of marine debris from 22 years of remotely operated vehicledeployments in Monterey Bay, covering depths from 25 m to 3971 m. We characterize debris by type,examine patterns of distribution, and discuss potential sources and dispersal mechanisms. Debris wasmost abundant within Monterey Canyon where aggregation and downslope transport of debris from thecontinental shelf are enhanced by natural canyon dynamics. The majority of debris was plastic (33%) andmetal (23%). The highest relative frequencies of plastic and metal observations occurred below 2000 m,indicating that previous studies may greatly underestimate the extent of anthropogenic marine debris onthe seafloor due to limitations in observing deeper regions. Our findings provide evidence thatsubmarine canyons function to collect debris and act as conduits for debris transport from coastal todeep-sea habitats.

& 2013 The Authors. Published by Elsevier Ltd. All rights reserved.

1. Introduction

The increasing rate of anthropogenic marine debris input intothe ocean is of global concern due to economic and ecologicalimpacts to coastal, pelagic, and benthic habitats (Ryan andMoloney, 1993; Laist, 1997; Derraik, 2002; Thompson et al.,2004; Barnes and Milner, 2005; Sheavly and Register, 2007;Moore, 2008; Gregory, 2009; Ryan et al., 2009; Bergmann andKlages, 2012). Sources, fates, and impacts of marine debris are notwell understood due to the wide variety and abundance of debrisencountered and the difficulty of conducting research in, andenforcing management of, such vast and variable habitats (Barneset al., 2009; NAS, 2009; Ryan et al., 2009; Ramirez-Llodra et al.,2011). Marine debris was technically defined by UNEP (2009) as

ublished by Elsevier Ltd. All rights

r the terms of the CreativeWorks License, which per-

ion in any medium, provided

1 8317751620.

“any persistent manufactured or processed solid material dis-carded, disposed of or abandoned in the marine environment”,and described perhaps more acutely by Moore and Allen (2000) as“a visible expression of human impact on the marine environ-ment”. These impacts, so clearly manifested on shorelines andsandy beaches are not yet as obvious on the seafloor, but are likelyto be as detrimental to ecosystem function (Hess et al., 1999;Moore and Allen, 2000; Lee et al., 2006; Katsanevakis et al., 2007).In addition to direct organism impacts such as entanglement oringestion (Laist, 1997; Pham et al., in press), debris may serve astransport (Barnes and Milner, 2005; Moore, 2008) and habitat foralien invasive species (Carter and Gregory, 2005; Gregory, 2009;NAS, 2009) thereby altering natural community composition andthreatening biodiversity. Toxic chemicals from debris may leachinto the surrounding environment and accumulation of debris canaffect current flow and circulation for infaunal organisms(Goldberg, 1997).

Due to the extreme technical challenges and prohibitive costsof conducting research in the deep sea, few studies have examinedthe types, quantities, or impacts debris may have on the habitats

reserved.

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105 97

beyond the continental shelf (Ramirez-Llodra et al., in prepara-tion). Much of the existing data on benthic debris comes from theexamination of incidental litter by-catch in trawl surveys con-ducted for commercial fisheries management purposes (June,1990; Hess et al., 1999; Galgani et al., 2000; Keller et al., 2010;Watters et al., 2010) and thus have been limited to relativelyshallow water habitats as it is expensive, risky, and time consum-ing to trawl in deep, rugged terrain such as found withinsubmarine canyons (Spengler and Costa, 2008; Pham et al., inpress). Little is known about what happens to deep-sea debristhrough time (Galgani et al., 2000; Lee, et al., 2006; Watters et al.,2010; Bergmann and Klages, 2012) and less than a dozen studieshave looked at debris below 500 m (June, 1990; Galil et al., 1995;Galgani et al., 1996, 2000; Keller et al., 2010; Miyake et al., 2011;Mordecai et al., 2011; Bergmann and Klages, 2012; Wei et al., 2012;Pham et al., in press; Ramirez-Llodra et al., in preparation). Galganiet al. (2000) observed trends in deep-sea pollution over time usinglarge-scale benthic trawls combined with video surveys in multi-ple locations off the European coast. They found the distribution ofmarine debris to be extremely variable, and were the first to reporttrash aggregating in submarine canyons, which they attributed tomore stable conditions and decreased water movement at deeperdepths. Miyake et al. (2011) reviewed video surveys and recordeddebris down to 7216 m depth in Ryukyu Trench. They noted thatlitter had accumulated in deep-sea trenches and depressions andthe majority of items were composed of plastic. From deepremotely operated vehicle (ROV) video surveys in submarinecanyons off of Portugal, Mordecai et al. (2011) concluded thatsubmarine canyons may act as a conduit for the transport ofmarine debris into the deep sea.

Submarine canyons are globally numerous, complex featureswith elaborate patterns of hydrography, flow, and sedimenttransport and accumulation; however, they remain poorly studied(De Leo et al., 2010; Ramirez-Llodra et al., 2010). Monterey Canyon,in central California, is the largest and best-studied submarinecanyon on the west coast of North America (Breaker andBroenkow, 1994; Petruncio et al., 1998; Greene et al., 2002; Paullet al., 2005, 2010). Over a 22 year period (1989–2010), theMonterey Bay Aquarium Research Institute (MBARI), in MossLanding, California, recorded more than 18,000 h of high-resolution video from ROV dives, down to depths of 4000 m. Ourpresent study takes advantage of MBARI's long-term, deep-seavideo database to characterize marine debris observations inMonterey Canyon and surrounding areas. We also utilize decadesof physical oceanographic measurements and research on canyondynamics to elucidate the transport mechanisms influencingmarine debris distribution within Monterey Bay.

2. Materials and methods

All dive footage was recorded by the ROVs Tiburon and DocRicketts on the R/V Western Flyer and by the ROV Ventana on theR/V Point Lobos. The majority of MBARI's dive expeditions took placewithin the greater Monterey Bay area. MBARI's Video Annotation andReference System (VARS; Schlining and Stout, 2006) was queried forall “trash” annotations recorded in the database from January 1, 1989to January 31, 2011; the resulting observations were carefullyreviewed to more specifically identify the types of debris encoun-tered and to note debris/organism associations (Fig. 1). The VARSdatabase contains over 3.9 million biological, chemical, and geologi-cal observations combined with ancillary metadata (CTD, oxygen,and position) from ROV dive footage.

Marine debris observations were normalized by dive effort(total ROV observation time on the seafloor within 50 m2 gridcells) for our study area, which included the greater Monterey Bay

area, south to Davidson Seamount (Fig. 2, upper inset). Normalizeddata were mapped in ArcGIS (ESRI 2011, ArcGIS Desktop: Release10. Redlands, CA: Environmental Systems Research Institute) tovisualize the distribution and relative abundance of debris overhigh-resolution multibeam bathymetry (Caress et al., 2008) of thestudy area (Fig. 2). Rugosity, the ratio of surface area to planar area,was derived from bathymetry with 50 m2 grid cells (Jenness,2012). To simplify analysis, rugosity was converted to slope anglesthat would give the equivalent ratio of surface area to planar area.Slope angles were classified in 5-degree bins from 0 to 40 degrees.ROV effort was binned with respect to pixels occupied by the ROVas well as by the time that the ROV occupied each slope class. Weanalyzed the overall distribution of trash relative to the slopeclasses. χ2 tests were conducted to determine if debris wasrandomly distributed with respect to slope. The distributions usedfor comparison to the observed distributions were created byrandomly selecting effort from the pixels that the ROV hadobserved. The number of selected pixels was equal to the numberof trash sightings in the particular category. Each test was repeated1000 times to account for variability in the random selection.

Debris observations were characterized using 16 broad cate-gories including: abandoned research equipment, battery, cloth-ing, other fabric, glass, concrete, manufactured wood, militarydebris (e.g., unexploded ordinance), paper, rope, rubber, shipwreckage, plastic, metal, fishing debris (traps, nets, and otherequipment obviously related to fishing practices), and unidentifieddebris items (items that were too difficult to identify or did not fitinto any of the other categories, e.g., mop head). We quantified thepercentage frequency of marine debris items by category for ourstudy grid. Although likely associated with fishing activities, ropewas placed in a separate category because it was not possible todetermine specific use from video. Items that remained visible onthe seafloor over time were counted only when initially observed.

Normalized observations were used to plot the six mostcommon categories of debris (plastic, metal, rope, unidentifieddebris, glass, and fishing debris) over time. Data were partitionedby ROV platform because ROV Ventana was depth-limited to1850 m while ROVs Tiburon and Doc Ricketts were capable to4000 m. Frequency of occurrence for these major debris categorieswere then plotted by depth and slope class. χ2 tests wereconducted to determine if debris categories were randomly dis-tributed with respect to depth and slope class using ROV effort asexpected values and debris occurrences as observed values.

3. Results

3.1. Distribution of marine debris in Monterey Canyon

A total of 1537 trash observations were recorded in the VARSdatabase of which 1149 were located within in our study grid.Correlating ROV position with high-resolution multibeam bathy-metry within GIS, we found that 0.24% (48,581 of the 20,672,56150 m2 cells within the grid) of the seafloor in our survey area hadbeen visited by MBARI's ROVs (Fig. 2). Marine debris items wereobserved within 726 of those grid cells or 1.49% of the surveyedarea. Litter was observed at depths from 25 m (PVC pipe) to3971 m (plastic bag). Debris was not randomly distributedthroughout Monterey Canyon. The null hypothesis for the χ2 teston rugosity (that the expected and the observed distributions ofdebris would be the same across all slope classes) was rejected forboth ROV effort binned by area (df¼7, x2(0.10)¼665.2392) and effortbinned by time (df¼7, x2(0.10)¼1470.7525). When normalized byROV effort, litter was evenly distributed across 0–301 slopes;however, litter was encountered more frequently on the steepestslopes of 30–401 (Fig. 3). In many instances, litter was observed

Fig. 1. Examples of marine debris items observed on MBARI ROV dives: (a) plastic chip bag, 3506 m in Monterey Canyon; (b) aluminum can, 1529 m at Axial Seamount;(c) rope crab pot “ghost fishing”, 1091 m in Astoria Canyon; (d) plastic bag wrapped around a deep-sea gorgonian, 2115 m in Astoria Canyon; (e) lost fishing rope, 999 m inMonterey Canyon; (f) foreign glass soda bottle, 1727 m at Davidson Seamount; (g) shoe with rockfish, 472 m in San Gabriel Canyon; (h) tire with rockfish, anemone, and seacucumber, 868 m in Monterey Canyon; (i) cardboard (paper) with an undescribed species of the sponge Hyalonema, 3950 m offshore of Santa Barbara.

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–10598

clustered just below the edge of canyon walls or on the outside ofcanyon meanders (Fig. 2). This pattern of distribution closelymatches the pattern of distribution of natural debris such as wood,sea grasses, and kelp in the VARS database (Fig. 3).

Within the canyon we identified four major locations of debrisaccumulation based on our normalized data. The area with thehighest relative abundance of trash (600 m, Fig. 2, circle 1)occurred on an upper section of the southern canyon wall. Weobserved plastic items (buckets), a tire, rope, and monofilamentline in this location (41 items observed in six visits). Circle 2(1800 m, Fig. 2), located on the outside of the sharpest meander inthe canyon, is the site of an emplaced whale-fall experiment. Here,the majority of trash was plastic bags, with further down-canyonmovement obstructed by the whale carcass and by shallowdepressions (58 items observed in nine visits). Circle 3 (300 m,Fig. 2) identifies a site just off the continental shelf break on theupper section of a canyon wall to the north. Debris there consistedof old fishing rope, beverage containers, plastic, metal, and a tire(22 items observed in three visits). A fourth area (beyond thedetailed extent in Fig. 2) was in the deep canyon axis at aninterface with the northern wall. This is the location of a naturalwhale fall at 2893 m, with plastic bags, plastic buckets, tools, and a55-gallon drum located nearby (29 items observed in eight visits).

3.2. Characterization of debris

Over half of all debris items observed were categorized asplastic (33%) or metal (23%) (Table 1). Within plastics, 54% of theobservations were plastic bags (N¼203). Of the metal observa-tions, 67% were some type of can (i.e. aluminum, steel, or tin;N¼180). Other commonly observed debris categories includedrope, glass, fishing debris, and unidentified debris. We did notobserve increasing or decreasing trends in trash abundance overall

or within any of these six major categories (Fig. 4) between 1989and 2010, but other trends were noted. Once the deeper-divingROVs Tiburon and Doc Ricketts were put into service, more debriswas recorded and these observations were strongly dominated byplastic and metal (Fig. 4). A χ2 analysis of distribution by depthshows that depth plays a significant role in the distribution ofplastic, metal, rope, and unidentified debris (χ2 analysis rejectedthe null hypothesis that each category was randomly distributedwith respect to depth (α¼0.1)). Plastic, metal, and unidentifieddebris were found more frequently in deeper depths (2000–4000 m), while the opposite was true for rope, which was foundmost frequently above 2500 m (Fig. 5). There also appeared to be agreater abundance of fishing debris at shallower depths, howeverthis trend was not significant (α¼0.1, 65 of 1000 tests acceptedH0). Depth was not a significant factor determining the distribu-tion for glass (results were inconclusive with the χ2 test rejected753 of 1000 times). Debris was found significantly more frequentlyat the steepest slope classes (35–451) across all major categories(χ2 analysis rejected the null hypothesis that each category wasrandomly distributed with respect to slope class (α¼0.1)), with theexception of fishing debris, which was found most frequently at35–401, but missing altogether from the 40–451 category (Fig. 6).

3.3. Debris/organism associations

Video frame grabs existed for 1088 of overall debris observa-tions and these images were analyzed for debris/organism asso-ciations (Fig. 1). Debris was observed in close contact withmegafauna (invertebrates or demersal fishes) in 37% of the images(Table 2). Common organisms noted on or near debris itemsincluded hydroids, anemones, asteroids, serpulid worms, crinoids,holothurians, and rockfish. Most organisms appeared to be usingthe debris as habitat; however, several occurrences of obvious

Fig. 2. Distribution of marine debris observed in Monterey Bay (N¼1149). MBARI ROV surveys over the 22-year study period are shown in red. The relative abundance oftrash was normalized by the amount of time spent searching the seafloor; the largest circles depict areas of trash accumulation which tend to occur on the outside walls ofcanyon meanders where high-energy water flow and erosion occur. The main study grid (upper inset) extended to the abyssal plain and included Davidson Seamount, about130 km to the southwest.

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105 99

detrimental impacts were noted including entanglement in ghostfishing gear or plastic bags (Fig. 1). No assessment could be madefor 5% of the observations due to distance from camera and othervisual obstructions.

4. Discussion

4.1. Distribution of marine debris in Monterey Canyon

The majority of trash items observed during 22 years of ROVsurveys in the greater Monterey Bay area were focused within theMonterey Canyon system. Hydrography, geomorphology, andanthropogenic activity determine the amount, type, and locationof debris reaching the seafloor. Previous studies have attributedthe majority of seafloor debris to marine vessel sources (June,1990; Moore and Allen, 2000; Keller et al., 2010). Certainly bottlesand cans are dumped into Monterey Bay from recreational orfishing vessels and discarded plastic bags and other light trashitems may get blown across the sea surface or float in themidwater. The surface flow in Monterey Bay is typically cyclonic,while the top 25–150 m is anticyclonic (Breaker and Broenkow,1994). Internal waves in the canyon affect the sediment transport(Petruncio et al., 1998) and could transport lighter litter items as

well. Often the flow is downcanyon during rising tides andupcanyon during the falling tide, however, the predominant watermovement is upward from the canyon onto the continental shelf(as measured in the top several hundred meters), with periodicreversals at all depths (Breaker and Broenkow, 1994). Trash fromoffshore sources could be transported shoreward over the con-tinental shelf, and eventually wash up on shore or sink to theseabed. Currents in Monterey Bay would affect marine debristransport in complex and unpredictable ways depending onseasonal circulation, tide, wind, weather, river discharge, canyontopography, plus other factors (Breaker and Broenkow, 1994;Petruncio et al., 1998).

Advancing technologies, such as ROVs and sophisticated navi-gation and mapping tools, have enabled detailed investigations ofthe transport mechanisms and geological features within Mon-terey Canyon. This research on canyon dynamics also providesinsight into processes influencing debris deposition and retentionin Monterey Bay. Unpublished data fromMBARI AUV surveys alongthe Monterey Canyon axis revealed that the highest concentra-tions of sediments at and above the depth of the shelf break occurnearshore in winter and are found further offshore in spring (JohnRyan, MBARI, pers. comm.). Debris flowing south from largemetropolitan areas (e.g., the San Francisco bay area, Santa Cruz)may likewise be swept off the shelf and into the canyon,

Fig. 3. The relationship between rugosity and debris distribution in Monterey Canyon. MBARI ROV surveys over the 22-year study period are shown in blue. The distributionof natural debris (drift kelp, wood) in Monterey Bay is indicated by the blue circles; the distribution of anthropogenic marine debris is indicated by the orange triangles. Thered box outlines a “clean” section of the canyon.

Table 1Percentage frequency of total marine debris items by type observedduring MBARI ROV surveys in the greater Monterey Bay area (MB)from 1989 to 2011.

Debris type %MB (N¼1149)

Plastic 33Metal 23Rope 14Unidentified debris 7Glass 6Fishing debris 5Paper 4Other fabric 3Rubber 2Clothing 1Manufactured wood o1Abandoned research equipment o1Ship wreckage o1Military o1Concrete o1Battery o1

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105100

accounting for the notable lack of marine debris observations,either natural or anthropogenic, seen on the continental shelf.Nearshore sediments are transported along the shoreline to thecanyon head via littoral drift where they accumulate until failure isinduced by triggers including storms and mass wasting events(Paull et al., 2005). Down-canyon sediment transport processes areextremely active on the floor of Monterey Canyon. Frequentepisodic mass transport events, such as slumping, turbidity flows,and gravity flows have been documented in the upper canyon axis,with current speeds as fast as 200 cm/s carrying large volumes of

sediment, and research equipment, deeper into the canyon system(Paull et al., 2003, 2005, 2010). In addition, significantly higherlevels of pollutants (e.g., DDT) have been documented in sedi-ments and fish from deep in Monterey Canyon versus nearbycontinental-shelf sediments and fish, underscoring the transportand subsequent accumulation of terrestrial materials in the deepsea (Looser et al., 2000; Hartwell, 2004, 2008). Presumably, thesesame mechanisms transport anthropogenic debris down-canyonwhere specific types of debris, notably plastic and metal items,were observed more frequently with the deeper diving ROVsTiburon and Doc Ricketts. In fact, the highest occurrences of plasticand metal debris in this study were observed from 2000 to4000 m, well below the depths surveyed by all but seven previousmarine debris studies (Galil et al., 1995; Galgani et al., 2000;Miyake et al., 2011; Mordecai et al., 2011; Bergmann and Klages,2012; Wei et al., 2012; Ramirez-Llodra et al., in preparation).

A recent preliminary study by the Bay Area StormwaterManagement Agencies Association (2012) reported that although44% of trash that enters storm drains in the San Francisco Bay areais successfully removed, there are an estimated 43,040,000 l/yearof trash entering Bay Area creeks via the storm-drainage system.In that study plastic, plastic bags, and recyclable beverage contain-ers accounted for 58–75% of the total trash identified, whichagrees with our results. From Cape Mendocino south to San Diego,waves from the northwest create a southward net littoral drift ofsand. Some trash from this large metropolitan area may well betransported all the way to Monterey Bay via known sand transportpaths within a littoral cell (Habel and Armstrong, 1978; Patsch andGriggs, 2006). As Bergmann and Klages (2012) noted, it isimpossible to determine whether the debris is primarily ocean-based or land-based without collecting samples for ground-truth-ing, however we propose that the bulk of the debris in MontereyCanyon comes from terrestrial or coastal sources and may include

Fig. 4. Relative frequency of marine debris observations of the six most common categories shown by year for MBARI's three ROVs (normalized by ROV effort indicated bythe dotted line overlay). ROV Ventana capable to 1850 m; ROVs Tiburon (1998–2007) and Doc Ricketts (2009, 2010) capable to 4000 m.

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105 101

inputs from Monterey, Santa Cruz, and even as far north as SanFrancisco. The distribution patterns we observed for the anthro-pogenic debris appeared to follow the general patterns for naturaldebris (kelp and driftwood) quite closely (Fig. 3). This, coupledwith the high abundance of debris within the canyon relative tothe continental shelf areas, indicates shore- and continental-shelf-based inputs that were subsequently transported into the canyonsystem. Our findings are supported by other studies that havelikewise proposed submarine canyons may act as conduits foranthropogenic debris, in particular transporting light litter items(e.g., plastic bags, metal beverage cans) away from the coast, off ofthe continental shelf, and into deeper waters (Galgani et al., 1996,2000; Barnes et al., 2009; Mordecai et al., 2011; Ramirez-Llodraet al., 2011).

The distribution of marine debris in the canyon was patchy.Much like the concentration of pelagic debris in the North Pacificcentral gyre (Moore et al., 2001) or benthic debris accumulations

in inlets (Hess et al., 1999) and shallow bays (Williams et al., 1993,Katsanevakis and Katsarou, 2004), there are areas on the deepseafloor where debris accumulates such as physical barriers (high-relief outcrops) or in depressions (Galgani et al., 1996; Miyakeet al., 2011). Galgani et al. (2000), Watters et al. (2010), and Weiet al. (2012) reported the highest density of debris in rocky andcanyon habitats. Our video records also showed debris on physicalbarriers, such as whale falls, clam beds, and boulders on theseafloor even in relatively flat terrain. However, the majority of ourobservations of marine debris were concentrated upon the ruggedslopes of Monterey Canyon (Fig. 3), potentially impacting corals,sponges, rockfish, and other sensitive marine organisms thatspecifically use this high relief and hard rock substrate as habitat.Areas with relatively fewer debris observations may be due to (a)reduced sampling effort; (b) physical oceanographic conditionstransporting the trash out of those areas; or (c) litter beingburied by sediment flows. Convergent transport along the bottom

Fig. 5. Relative frequency of occurrence of marine debris observations of the six most common categories shown by depth. Categories significantly influenced by depth aremarked by *(α¼0.1).

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105102

boundary of Monterey Canyon has been related to detrainment ofwaters from the bottom boundary layer and associated transportof suspended sediments out of the convergence zone (Kunze et al.,2012). Persistence of such transport patterns could also exporttrash that is nearly neutrally buoyant, thus creating zones ofrelatively low trash accumulation. Approximate spatial coinci-dence between a convergence zone observed by Kunze et al.(2012) and low trash accumulation in our study points towardthis potential mechanism (Fig. 3, see red box). Our surveys likelyunderestimate the amount of marine debris in the canyon,particularly in the thalweg (with respect to the upper canyonwalls), as frequent mass wasting events undoubtedly bury andrender it unobservable. Unfortunately, the nature of video datacollection during our ROV surveys does not allow for quantitativecalculations other than relative abundance.

4.2. Characterization of debris

It is not surprising that plastic, due to its ubiquitous use inmodern society and remarkable durability, was the most commoncategory of marine debris identified in our seafloor study (1/3 oftotal debris observed). Golik (1997) and Derraik (2002) report thatplastic typically dominates beach (60–80%) and floating-debrissurveys (60–70%). Plastic dominates the benthic debris found inthe Mediterranean Sea off the coast of France (Galgani et al., 1996)and other European coastal regions (Galgani et al., 2000). Even onremote Alaskan (Hess et al., 1999) and deep Arctic seafloors(Bergmann and Klages, 2012), plastic items accounted for themajority of litter recorded. Regardless of how it enters and travels

through the ocean, plastic litter will eventually sink to the seafloorwhere low oxygen and lack of sunlight inhibit degradation rates,resulting in debris that will persist for centuries (Moore, 2008;Barnes et al., 2009; Gregory, 2009; O’Brine and Thompson, 2010).As the plastic fragments into tiny particles (microplastics) it ismore likely to be ingested (Thompson et al., 2004; Browne et al.,2008) thus delivering a source of chemical pollutants to themarine food web with a largely unknown effect (O’Brine andThompson, 2010; Cole et al., 2011; Davidson and Asch, 2011). Thisalarming accumulation of microplastics in the ocean (Thompsonet al., 2004; Cole et al., 2011; Ramirez-Llodra et al., 2011) is notvisible in the ROV video and therefore unaccounted for in thescope of this study.

While Monterey Bay supports thriving commercial and recrea-tional fishing industries, debris from fishing activities was not asdominant in our observations (fishing debris 3%; rope 11%) as inmany previous studies that found fishing-related debris (i.e. netsand monofilament line) to have the highest spatial coverage(Moore and Allen, 2000; Lee et al., 2006; Watters et al., 2010;Keller et al., 2010), even within a marine sanctuary (Chiapponeet al., 2005). This may be due to our video surveys including deep,rugged sites within the canyon where fishermen and researchersalike are unable to trawl. We noted the complete absence offishing debris observations in the steepest slope class (40–451;Fig. 6). The greater abundance of fishing debris we saw in shallowhabitats compared to deep may be because more fishing activityoccurs in shallower habitats and lost gear can easily get hung up inrocky outcrops and may not travel down-canyon as freely asplastic, glass, and metal debris. We acknowledge that our fishing

Fig. 6. Relative frequency of occurrence of marine debris observations of the six most common categories shown by slope class. Categories significantly influenced by slopeclass are marked by *(α¼0.1).

Table 2Percentage of debris/organism associations visiblein frame grabs from all MBARI ROV surveys (1989–2011).

Debris/organismassociations

% of totalN¼1088

No organism contact 58Organism contact 37Unable to determine 5

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105 103

debris observations may be biased because MBARI ROV pilotsactively avoid hazards like discarded fishing gear due to the risk ofentanglement.

4.3. Debris/organism associations

Approximately one third of our debris observations had visibleassociated epibenthic megafauna (37%). By comparison, Bergmannand Klages (2012) noted 67% of the debris items they observedwere entangled or colonized by organisms. Our estimate is likelyconservative as animals may be missed in the video due to smallsize, location underneath or inside debris items, or distance fromthe ROV. We observed a few instances of direct negative impacts oftrash on deep-sea organisms, such as a coral colony wrapped in aplastic bag and several ghost fishing traps and nets with entangleddead organisms (Fig. 1). However, in most cases it appeared thatthe organisms were utilizing the debris items as habitat to hide in,or adhere to, as was also reported by other studies (Backhurst andCole, 2000; Katsanevakis et al., 2007; Watters et al., 2010; Miyake

et al., 2011; Mordecai et al., 2011; Bergmann and Klages, 2012).Multiple studies and anecdotal records have investigated thedetrimental effects of debris on coastal marine organisms, includ-ing entanglement and ingestion (Laist, 1997; Browne et al., 2008).These reports are abundant compared to verifiable data on theeffects of marine debris upon life in the deep sea. A recent researchcollaboration between MBARI and the Monterey Bay NationalMarine Sanctuary presents evidence for shifting structure in adeep (1281 m) seafloor community following the accidental intro-duction of a large shipping container (Josi Taylor, MBARI, pers.comm.), but by and large, the impacts of debris on deep-seaecosystems are unconfirmed and require further study.

4.4. Other areas surveyed

The R/V Western Flyer has also conducted expeditions furtherafield, including 3-month expeditions to Hawai’i and the Gulf ofCalifornia, as well as several expeditions to Southern Californiaand various parts of the Pacific Northwest. It is noteworthy that inall regions MBARI has explored, marine litter was observed in thedeep sea (see Supplementary materials). Although direct compar-isons with our quantitative Monterey Bay results cannot be madedue to the sparse availability of data as well as differences inseafloor features and hydrographic conditions in these otherregions, we did discover that, similar to our Monterey Bayfindings, plastic and metal made up 55–67% of total observationsin these regions, with the exception of the Gulf of California, whichwas dominated by plastic items (79%; see Supplementarymaterials).

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105104

5. Conclusions

Worldwide concern about marine debris has increased over thelast decade as scientists and media outlets have brought the issueinto the public eye. The majority of marine debris reports havebeen based upon coastal and pelagic debris surveys, with recentattention focused on the North Pacific central gyre (Moore et al.,2001) or “The Great Garbage Patch”. Our study demonstrates thatdebris does indeed reach deep ocean habitats, even in a nationalmarine sanctuary, and highlights the paucity of data available tocomprehend the full extent of this issue. The abundance ofinformation on the transport mechanisms and physical featureswithin Monterey Canyon provides insight into processes that canaffect debris deposition and retention in Monterey Bay. Ourresearch suggests that once it reaches the seabed, debris is sweptoff the continental shelf into Monterey Canyon and is confinedthere by topographical barriers; in addition, sediment transportmechanisms act as a conduit, carrying debris (particularly plasticand metal) into deep habitats. There are as many as 660 submarinecanyons worldwide with an estimated 15%, or 99 of them, withnearshore heads that receive substantial coastal sediment inputs(De Leo et al., 2010) potentially facilitating the transport of coastalanthropogenic debris to deep-sea habitats.

Monterey Canyon is one of the best-studied areas of the world'sdeep ocean. Yet, remarkably, in 22 years of exploration, MBARIvisited just 0.24% of the seafloor in our study area. With vastregions of the deep sea remaining unexplored, we can presumethat the amount of marine debris greatly exceeds that reported todate. It is encouraging to see the growing number of researchstudies focusing on deep seafloor litter and bringing this hiddenproblem to light. Multiple studies have concluded that, costs aside,it would do greater damage to the benthic fauna and habitats toremove colonized debris items than to leave them in place(Backhurst and Cole, 2000; Watters et al., 2010; Mordecai et al.,2011). Ultimately, preventing the introduction of litter into themarine environment through increased public awareness remainsthe most efficient and cost-effective solution to this dilemma(Ryan et al., 2009; Watters et al., 2010).

Acknowledgments

We would like to acknowledge the dedicated crews of the R/VsPoint Lobos and Western Flyer and the skilled pilots of ROVsVentana, Tiburon, and Doc Ricketts. This paper benefited fromthoughtful discussions with John Ryan, Ken Smith, and CharliePaull at MBARI and by constructive comments from Eva Ramirez-Llodra and two anonymous reviewers. Nancy Barr at MBARIprovided editing assistance. Our research was funded via theDavid and Lucile Packard Foundation's support for MBARI.

Appendix A. Supplementary materials

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.dsr.2013.05.006.

References

Backhurst, M.K., Cole, R.G., 2000. Subtidal benthic marine litter at Kawau Island,north-eastern New Zealand. J. Environ. Manage. 60, 227–237, http://dx.doi.org/10.1006/jema.2000.0381.

Barnes, D.K.A., Milner, P., 2005. Drifting plastic and its consequences for sessileorganism dispersal in the Atlantic Ocean. Mar. Biol. 146, 815–825.

Barnes, D.K.A., Galgani, F., Thompson, R.C., Barlaz, M., 2009. Accumulation andfragmentation of plastic debris in global environments. Philos. Trans. R. Soc., B,Biol. Sci. 364, 1985–1998.

Bay Area Stormwater Management Agencies Association. 2012. Preliminary base-line trash generation rates for San Francisco Bay Area MS4s, TechnicalMemorandum.

Bergmann, M., Klages, M., 2012. Increase of litter at the Arctic deep-sea observatoryHAUSGARTEN. Mar. Pollut. Bull. 64, 2734–2741, http://dx.doi.org/10.1016/j.marpolbul.2012.09.018.

Breaker, L.C., Broenkow, W.W., 1994. The circulation of Monterey Bay and relatedprocesses. Oceanogr. Mar. Biol.: An Annu. Rev. 32, 1–64.

Browne, M.A., Dissanayake, A., Galloway, T.S., 2008. Ingested microscopic plastictranslocates to the circulatory system of the mussel, Mytilus edulis (L.). Environ.Sci. Technol. 42 (13), 5026–5031.

Carter, R., Gregory, M.R., 2005. Bryozoan encrusted plastic from the continentalslope: eastern South Island, New Zealand. N. Z. Nat. Sci. 30, 49–55.

Caress, D.W., Thomas, H., Kirkwood, W.J., McEwen, R., Henthorn, R., Clague, D.A.,Paull, C.K., Paduan, J., Maier, K.L., 2008. High-resolution multibeam, sidescan,and subbottom surveys using the MBARI AUV D. Allan B. In: Reynolds, J.R.,Greene, H.G. (Eds.), Marine Habitat Mapping Technology for Alaska. Alaska SeaGrant College Program. University of Alaska Fairbanks, pp. 47–70.

Chiappone, M., Dienes, H., Swanson, D.W., Miller, S.L., 2005. Impacts of lost fishinggear on coral reef sessile invertebrates in the Florida Keys National MarineSanctuary. Biol. Conserv. 121, 221–230.

Cole, M., Lindeque, P., Halsband, C., Galloway, T.S., 2011. Microplastics as contami-nants in the marine environment: A review. Mar. Pollut. Bull. 62, 2588–2597.

Davidson, P., Asch, R.G., 2011. Plastic ingestion by mesopelagic fishes in the NorthPacific Subtropical Gyre. Mar. Ecol. Prog. Ser. 432, 173–180.

De Leo, F.C., Smith, C.R., Rowden, A.A., Bowden, D.A., Clark, M.R., 2010. Submarinecanyons: hotspots of benthic biomass and productivity in the deep sea. Proc. R.Soc., B, Biol. Sci. 277, 2783–2792, http://dx.doi.org/10.1098/rspb.2010.0462.

Derraik, J.G.B., 2002. The pollution of the marine environment by plastic debris: Areview. Mar. Pollut. Bull. 44, 842–852.

Galgani, F., Souplet, A., Cadiou, Y., 1996. Accumulation of debris on the deep seafloor off the French Mediterranean coast. Mar. Ecol. Prog. Ser. 142, 225–234.

Galgani, F., Leaute, J.P., Moguedet, P., Souplet, A., Verin, Y., Carpentier, A., Goraguer, H.,Latrouite, D., Andral, B., Cadiou, Y., Mahe, J.C., Poulard, J.C., Nerisson, P., 2000. Litteron the sea floor along European coasts. Mar. Pollut. Bull. 40, 516–527.

Galil, B.S., Golik, A., Türkay, M., 1995. Litter at the bottom of the sea: A sea bedsurvey in the eastern Mediterranean. Mar. Pollut. Bull. 30, 22–24.

Goldberg, E.D., 1997. Plasticizing the seafloor: An overview. Environ. Technol. 18,195–201.

Golik, A., 1997. Debris in the Mediterranean Sea: Types, quantities, and behavior.In: Coe, J.M., Rogers, D.B. (Eds.), Marine debris: Sources, impacts, and solutions.Springer-Verlag, New York, pp. 7–14.

Greene, H.G., Maher, N., Paull, C.K., 2002. Physiography of the Monterey BayNational Marine Sanctuary and implications about continental margin devel-opment. Mar. Geol. 181, 55–82, http://dx.doi.org/10.1016/S0025-3227(01)00261-4.

Gregory, M.R., 2009. Environmental implications of plastic debris in marine settings—entanglement, ingestion, smothering, hangers-on, hitch-hiking and alieninvasions. Philos. Trans. R. Soc., B, Biol. Sci. 364, 2013–2025.

Habel, J.S., Armstrong, G.A., 1978. Assessment and atlas of shoreline erosion alongthe California coast. Calif. Dep. Navig. Ocean Dev., 277.

Hartwell, S.I., 2004. Distribution of DDT in sediments off the central Californiacoast. Mar. Pollut. Bull. 49, 299–305.

Hartwell, S.I., 2008. Distribution of DDT and other persistent organic contaminantsin Canyons and on the continental shelf off the central California coast. Mar.Environ. Res. 65, 199–217.

Hess, N.A., Ribic, C.A., Vining, I., 1999. Benthic marine debris, with an emphasis onfishery-related items, surrounding Kodiak Island, Alaska, 1994–1996. Mar.Pollut. Bull. 38, 885–890.

Jenness, J., 2012. DEM Surface Tools. Jenness Enterprises. Available from: ⟨http://www.jennessent.com/arcgis/surface_area.htm⟩.

June, J.A., 1990. Type, source, and abundance of trawl-caught marine debris offOregon, in the eastern Bering Sea, and in Norton Sound in 1988. In: Shomura, R.S.,Godfrey, M.L. (Eds.), Proceedings of the Second International Conference onMarine Debris, NOAA Technical Memo NMFS-SWF-SC-154, pp. 279–301.

Katsanevakis, S., Katsarou, A., 2004. Influences on the distribution of marine debrison the seafloor of shallow coastal areas in Greece (Eastern Mediterranean).Water, Air Soil Pollut. 159, 325–337.

Katsanevakis, S., Verriopoulos, G., Nicolaidou, A., Thessalou-Legaki, M., 2007. Effectof marine litter on the benthic megafauna of coastal soft bottoms: A manip-ulative field experiment. Mar. Pollut. Bull. 54, 771–778.

Keller, A.A., Fruh, E.L., Johnson, M.M., Simon, V., McGourty, C., 2010. Distributionand abundance of anthropogenic marine debris along the shelf and slope ofthe US West Coast. Mar. Pollut. Bull. 60, 692–700.

Kunze, E., MacKay, C., McPhee-Shaw, E.E., Morrice, K., Girton, J.B., Terker, S.R., 2012.Turbulent mixing and exchange with interior waters on sloping boundaries.J. Phys. Oceanogr. 42, 910–927, http://dx.doi.org/10.1175/jpo-d-11-075.1.

Laist, D.W., 1997. Impacts of marine debris: Entanglement of marine life in marinedebris including a comprehensive list of species with entanglement andingestion records. In: Coe, J.M., Rogers, D.B. (Eds.), Marine debris: Sources,impacts, and solutions. Springer-Verlag, New York, pp. 99–139.

Lee, D.I., Cho, H.S., Jeong, S.B., 2006. Distribution characteristics of marine litter onthe sea bed of the East China Sea and the South Sea of Korea. Estuarine, CoastalShelf Sci. 70, 187–194.

Looser, R., Froescheis, O., Cailliet, G.M., Jarman, W.M., Ballschmiter, K., 2000. Thedeep-sea as a final global sink of semivolatile persistent organic pollutants?

K. Schlining et al. / Deep-Sea Research I 79 (2013) 96–105 105

Part II: organochlorine pesticides in surface and deep-sea dwelling fish of theNorth and South Atlantic and the Monterey Bay Canyon (California). Chemo-sphere 40, 661–670.

Mordecai, G., Tyler, P.A., Masson, D.G., Huvenne, V.A.I., 2011. Litter in submarinecanyons off the west coast of Portugal. Deep-Sea Res. II 58, 2489–2496, http://dx.doi.org/10.1016/j.dsr2.2011.08.009.

Miyake, H., Shibata, H., Furushima, Y., 2011. Deep-sea litter study using deep-seaobservation tools. In: Omori, K., Guo, X., Yoshie, N., Fujii, N., Handoh, I.C., Isobe, A.,Tanabe, S. (Eds.), Interdisciplinary Studies on Environmental Chemistry-MarineEnvironmental Modeling and Analysis Terrapub, pp. 261–269.

Moore, C.J., 2008. Synthetic polymers in the marine environment: A rapidlyincreasing, long-term threat. Environ. Res. 108, 131–139.

Moore, C.J., Moore, S.L., Leecaster, M.K., Weisberg, S.B., 2001. A comparison ofplastic and plankton in the North Pacific central gyre. Mar. Pollut. Bull. 42,1297–1300.

Moore, S.L., Allen, M.J., 2000. Distribution of anthropogenic and natural debris onthe mainland shelf of the Southern California Bight. Mar. Pollut. Bull. 40, 83–88.

National Academy of Sciences (NAS). 2009. Tackling Marine Debris in the 21stCentury. Committee on the Effectiveness of International and National Mea-sures to Prevent and Reduce Marine Debris and its Impacts, Ocean StudiesBoard, Division on Earth and Life Studies, Natural Research Council, NationalAcademy of Sciences, Washington, DC, USA, 219 pp.

O’Brine, T., Thompson, R.C., 2010. Degradation of plastic carrier bags in the marineenvironment. Mar. Pollut. Bull. 60, 2279–2283, http://dx.doi.org/10.1016/j.marpolbul.2010.08.005.

Patsch, K., Griggs, G., 2006. Littoral cells, sand budgets, and beaches: UnderstandingCalifornia's shoreline. Institute of Marine Sciences, University of California,Santa Cruz, 40 pp.

Paull, C.K., Ussler III, W., Greene, H.G., Keaten, R., Mitts, P., Barry, J., 2003. Caught inthe act: The 20 December 2001 gravity flow event in Monterey Canyon. Geo-Marine Lett. 22, 227–232.

Paull, C.K., Mitts, P., Ussler III, W., Keaten, R., Greene, H.G., 2005. Trail of sand inupper Monterey Canyon: Offshore California. Geol. Soc. Am. Bull. 117,1134–1145.

Paull, C.K., Ussler III, W., Caress, D.W., Lundsten, E., Covault, J.A., Maier, K.L., Xu, J.,Augenstein, S., 2010. Origins of large crescent-shaped bedforms within the axialchannel of Monterey Canyon, offshore California. Geosphere 6 (6), 755–774.

Petruncio, E.T., Rosenfeld, L.K., Paduan, J.D., 1998. Observations of the internal tidein Monterey Canyon. J. Phys. Oceanogr. 28, 1873–1903.

Pham, C.K., Gomes-Pereira, J.N., Isidro, E.J., Santos, R.S., Morato, T., Abundance oflitteron Condor seamount (Azores, Portugal, Northeast Atlantic). Deep-Sea Res.II, http://dx.doi.org/10.1016/j.dsr2.2013.01.011., in press.

Ramirez-Llodra, E., Brandt, A., Danovaro, R., De Mol, B., Escobar, E., German, C.R.,Levin, L.A., Martinez-Arbizu, P., Menot, L., Buhl-Mortensen, P., Narayanaswamy, B.E.,Smith, C.R., Tittensor, D.P., Tyler, P.A., Vanreusel, A., Vecchione, M., 2010. Deep,diverse and definitely different: Unique attributes of the world’s largest ecosystem.Biogeosciences 7, 2851–2899.

Ramirez-Llodra, E., Tyler, P.A., Baker, M.C., Bergstad, O.A., Clark, M.R., Escobar, E.,Levin, L.A., Menot, L., Rowden, A.A., Smith, C.R., Van Dover, C.L., 2011. Man andthe last great wilderness: Human impact on the deep sea. PLoS ONE 6 (8),e22588, http://dx.doi.org/10.1371/journal.pone.0022588.

Ramirez-Llodra, E., De Mol, B., Company, J.B., Coll, M., Sardà, F., 2013. Effects ofnatural and anthropogenic processes in the distribution of marine litter in thedeep Mediterranean Sea. Prog. Oceanogr., in preparation.

Ryan, P.G., Moloney, C.L., 1993. Marine litter keeps increasing. Nature 361, 23.Ryan, P.G., Moore, C.J., van Franeker, J.A., Moloney, C.L., 2009. Monitoring the

abundance of plastic debris in the marine environment. Philos. Trans. R. Soc., B,Biol. Sci. 364, 1999–2012.

Schlining, B.M., Jacobsen Stout, N. 2006. MBARI's video annotation and referencesystem. In: Proceedings of the Marine Technology Society/Institute of Electricaland Electronics Engineers Oceans Conference, Boston, MA, pp. 1–5.

Sheavly, S.B., Register, K.M., 2007. Marine debris and plastics: Environmentalconcerns, sources, impacts and solutions. J. Polym. Environ. 15, 301–305.

Spengler, A., Costa, M.F., 2008. Methods applied in studies of benthic marine debris.Mar. Pollut. Bull. 56, 226–230.

Thompson, R.C., Olsen, Y., Mitchell, R.P., Davis, A., Rowland, S.J., John, A.W.G.,McGonigle, D., Russell, A.E., 2004. Lost at sea: where is all the plastic? Science304, 838.

UNEP, 2009. Marine Litter: A Global Challenge. UNEP, Nairobi p. 232.Watters, D.L., Yoklavich, M.M., Love, M.S., Schroeder, D.M., 2010. Assessing marine

debris in deep seafloor habitats off California. Mar. Pollut. Bull. 60, 131–138,http://dx.doi.org/10.1016/j.marpolbul.2009.08.019.

Wei, C.L., Rowe, G.T., Nunnally, C.C., Wicksten, M.K., 2012. Anthropogenic “litter”and macrophyte detritus in the deep Northern Gulf of Mexico. Mar. Pollut. Bull.64, 966–973.

Williams, A.T., Simmons, S.L., Fricker, A., 1993. Off-shore sinks of marine litter: anew problem. Mar. Pollut. Bull. 26, 404–405.