1、USANorofpollutants to enter our food supply in quantities much higher than present.March(IOC/UNESCOand causedal. 2012;providing real time updates of the debris field based on numericalmodel results.Since the tsunami, there have been several reports of tsunamidebris encountered in the Pacific Ocean.
2、Detailed reports of debrissightings with positions and descriptions have been made avail-able by the Japanese government: (http:/www.kantei.go.jp/jp/sin-in a bottle, tossed into the sea and washed up on a distantshoreline; modern oceanography has also benefitted from themessages carried by debris dr
3、opped into the ocean. Examplesinclude the work of by Ebbesmeyer and Ingraham (1992, 1994)and Ebbesmeyer et al. (2007) who studied the fate of shoes andchildrens toys lost from transoceanic container ships to refineand calibrate numerical models of oceanic circulation. Debrisentering the ocean as a r
4、esult of a natural disasters has also beenstudied by Doong et al. (2011) who looked at typhoon debris andPrasetya et al. (2011) who investigated the transport of debris inthe vicinity of Banda Aceh following the 2004 Boxing Day tsunami.Corresponding author. Address: 74 Upper Wainui Road, Raglan 3297
5、, NewZealand.Marine Pollution Bulletin 66 (2013) 5358Contents lists available atMarine PollutE-mail address: jborrerousc.edu (J.C. Borrero).Reymond et al. 2012). One aspect of this event not usually associ-ated with a tsunami disaster was the massive influx of debriswashed from the coastline and int
6、o the ocean. The popular mediahas reported extensively on the story of the tsunami debris whileinternational research efforts are underway to study the transportand distribution of the debris field as well as its potential effects onthe environment and maritime transport. Groups such as TheOcean Con
7、servancy and the National Oceanic and AtmosphericAdministration (NOAA) host internet sites devoted the debriswhile researchers at the University of Hawaii International PacificResearch Center (http:/iprc.soest.hawaii.edu) host a websiteoceanrecov.org/tsunami-debris/about.html). In late March 2012, a
8、so called ghost ship from Japan was spotted off of the west coastof British Columbia and subsequently sunk by the US Coast Guard.Throughout 2012, there have been numerous reports of tsunamirelated debris, particularly floats and other objects that sit higherabove the waterline reaching the coast of
9、North America.2. Studies of marine debris transport and accumulationNautical lore has long romanticized the notion of the message1. IntroductionThe great Tohoku tsunami of 11of over 40 m (Mori et al., 2011), devastatedresulted in nearly 20,000 casualtiesBulletin No. 29, 30 September, 2011)out the Pa
10、cific Ocean (Borrero et0025-326X/$ - see front matter C211 2012 Elsevier Ltd. Allhttp:/dx.doi.org/10.1016/j.marpolbul.2012.11.013C211 2012 Elsevier Ltd. All rights reserved.2011, generated runupthe east coast of Japan,Japan Tsunamidamage through-Lynett et al. 2012;gi/kaiyou/hyouryuu/senpaku_eng.html
11、). In September 2011 theRussian training vessel Pallada, encountered tsunami debris including a fishing boat from the Fukushima prefecture approx-imately 300 miles northwest of Midway Atoll. In December 2011a multi-institution team of researchers visited the waters aroundMidway Island to collect deb
12、ris samples and deploy additionaldrifters to track the progress of the debris field (http:/www.Marine debrisThis is important since a significant fraction of the debris will be comprised of plastics, some of which willdegrade into tiny particles and be consumed by marine organisms, thereby allowing
13、adsorbed organicModeling the transport and accumulationby the 11 March 2011 Tohoku tsunamiLaurent C.-M. Lebretona, Jose C. Borrerob,c,aASR Ltd., Marine Consulting and Research, Raglan 3225, New ZealandbUniversity of Southern California, Tsunami Research Center, Los Angeles, CA 90089-2531,ceCoast Ltd
14、., Raglan 3225, New Zealandarticle infoKeywords:Tsunami debrisGarbage patchSubtropical gyreOceanic accumulation zonePlastic pollutionabstractA global ocean circulation modeling debris washed into thedebris is based on coastal populationdata is used to model the transportinvestigate the distributiono
15、cean likely represents thousandsjournal homepage: reserved.is coupled to a particle-tracking model to simulate the transport of float-th Pacific Ocean by the Tohoku tsunami. A release scenario for the tsunamiand measured tsunami runup. Archived 2011/2012 hindcast currentof debris since the tsunami,
16、 while data from 2008 to 2012 is used todebris on timescales up to 4 years. The vast amount of debris pushed intoof years worth of normal litter flux from Japans urbanized coastline.floating debris generatedSciVerse ScienceDirection Bulletincate/marpolbulof these accumulation zones at subtropical la
17、titudes is related tomostly submerged in the water and extra forcing on potentiallycompares wellto observations, there are discrepancies. particularlyin the north, where tsunami debris has been identified outside ofPollutionemerged parts of the debris is neglected. Using this approach leadsto the un
18、derprediction of the transport of debris particles that sithigh above the water line and are subject to additional wind stress.Indeed, at the time of writing of this article, several pieces of largerdebris have reached the north American west coast. However, inthis study, the simulation applies only
19、 to objects that have essen-tially zero windage, which represents an unknown but probablylarge fraction of the total debris field.4. Debris inputFor our simulations, a total of 50,000 tracer particles were re-leased into the model domain. We used two methods to simulateoverlying wind systems and the
20、re is a considerable body of exist-ing literature describing the extent and concentration of debriscontained within them (Carpenter and Smith, 1972; Law et al.,2010). Recently, Lebreton et al. (2012) proposed a methodologyto quantify and track, floating debris from coastal input sourcesto oceanic ac
21、cumulation zones. The debris field caused by the2011 Tohoku tsunami presents an opportunity to combine datadescribing likely debris sources with oceanic circulation models.3. Model and methodOur particle tracking model uses a two-stage process; first ahydrodynamic model solves the equations of motio
22、n to describewater movements throughout the model domain. In the secondstage, virtual particles are introduced into the flow field and al-lowed to move freely through hydrodynamic forcing. For thisstudy, sea surface currents are extracted from the oceanic circula-tion modeling system HYCOM/NCODA (Cu
23、mmings, 2005). The HY-COM model is forced by the US Navys Operational GlobalAtmospheric Prediction System (NOGAPS) and includes windstress, wind speed, heat flux, and precipitation. The model providessystematic archiving of daily ocean circulation on a global scalewith output data archived back to m
24、id-2003. The HYCOM modelis computed on a Mercator grid between 78C176S and 47C176N at 1/12C176resolution. The tsunami debris model grid covers the North Pacificfrom the Equator through 48C176N and from 130C176E to 105C176W and com-prises 1551C2701 grid nodes with an average grid spacing ofC247 km. W
25、hile the full HYCOM model contains 32 vertical layers,we only consider velocities in the surface layer as the principal dri-ver of floating particles.The velocity data extracted from HYCOM are then coupled tothe Lagrangian particle tacking model Pol3DD and used to drivethe dispersion of floating mat
26、erial across the ocean surface. Themodel Pol3DD tracks virtual particles to simulate waterborne dis-persion of material including neutrally buoyant anthropogenicmaterial, larvae, oil spills, outfall discharges and estuarine or beachsediment transport. Pol3DD tracks and stores the origin, age, andtra
27、jectory information of individual particles. The particle trackingmodel uses a second-order accurate advection scheme (Black andGay, 1990) and is described in detail in Lebreton et al. (2012). Sincewind driven currents are already expressed in the HYCOM hydro-dynamic data, no additional wind stress
28、terms are applied to themotion of particles. This model assumes that debris particles areOn a larger scale, the identification and description of oceanicaccumulation zones, garbage patches or gyres, characterized byhigh concentrations of plastic debris (Moore et al., 2001), has at-tracted worldwide
29、media and scientific attention. The formation54 L.C.-M. Lebreton, J.C. Borrero/Marinethe input of debris from the tsunami; thefirst considered a uniformdistribution of material released along the Japanese coast (uniformrelease), while the second approach used a weighting methodthe zone predicted by
30、our model. We attribute this to additionalwindstressactingonpiecesoffloatingdebris,aneffectnotincludedin our assessment. This effect would be more pronounced on largerpieces of debris that sit higher above the water level such as boatsor shipping containers. Indeed, most of these outlying exampleswe
31、re fishing boats or partially submerged shipping containers.This effect would also explain reports of boats and other objectsthat sit above the waterline reaching the west coast of NorthAmerica in early 2012 since these objects are subject to increasedwind stress.The results of the second modeling s
32、tudy are depicted in Figs. 4and 5 which show the integrated number of particle visits per celland the number of particles existing in a cell (expressed as a per-centage of the total number of particles released) at particulartimes. While the overall patterns are similar, there are noticeabledifferen
33、ces in the year-to-year details. Ultimately however, the re-sults suggest that the bulk of the tsunami debris will accumulate inthe eastern North Pacific Ocean between Hawaii and California.This finding is in line withresults from studies by other researchers(for example see: http:/iprc.soest.hawaii
34、.edu/news/marine_and_tsunami_debris/IPRC_tsunami_debris_models.php). The resultsshow that after 6 months, the debris field extends eastward tothe international dateline and that subsequent eastward transportis much slower, requiring another 2.5 years is required for the deb-(weighted release) combin
35、ing measured tsunami runup heights(Mori et al., 2011) with data describing coastal population density(Halpern et al., 2008). The relationship between tsunami heightand population is an important consideration, since plastics, whichexist in higher amounts in populated or urbanized areas, would bethe
36、most buoyant and longest lasting of the debris washed into theocean. Along sparsely populated sections of the coast; the majorityof the debris would consist of plant material, which wouldeventually break down in the ocean environment. Inspection ofFig. 1 shows that the some of the highest measured t
37、sunami runupheights occurred on areas of the coast with relatively lowpopulations.Based on these inputs, we then devised two modeling experi-ments. The first experiment used only 1 year of oceanic currentdata (11 March 201111 March 2012) and compared the distribu-tion of tsunami debris as a function
38、 of the particle release scenario.The second experiment used only the weighted release scenario.Initialising the model on 11 March of each year from 2008 through2011, we ran the simulation through 11 March 2012. This experi-ment was designed as means of assessing the potential inter-annual variabili
39、ty of the debris transport as well as forecastingthe progress of the debris field by using historical data, assumingthat overall oceanic surface currents do not vary significantly fromyear to year.5. Results and analysisFig. 2 illustrates the difference in the predicted distribution oftsunami debris
40、 as a function of the input scheme. In Fig. 3 wecompare the outer limit of the predicted debris field from bothinput scenarios at 1 month intervals to reported sightings oftsunamidebrisfrom coincidenttime intervals.These plots suggeststhat different release scenarios result in differences in thedist
41、ribution of the particle concentrations, but that the outer limitof the debris field predicted by the model for both releases isvirtually identical. While the modeled extent of the debris fieldBulletin 66 (2013) 5358ris to reach 130C176W. Once east of the dateline, the speed of progres-sion of the d
42、ebris field is approximately 5 cm/s (50C176 longitude at40C176N = 4250 km over 2.5 years) which is consistent with the speedPollutionL.C.-M. Lebreton, J.C. Borrero/Marineof the eastward geostrophic flow of the North Pacific Current(Cummins and Freeland, 2007).Finally, in Fig. 6, we show the rate of
43、progress of the debris fieldacross the North Pacific Ocean by plotting the outer limit of thepredicted particle positions at different time intervals. The resultsindicate that progress of the debris field is relatively quick over thefirst year, but then slows considerably, taking an additional 3 yea
44、rsfor the eastern edge of the debris field to approach the NorthAmerican coastline. Indeed, as indicated in Fig. 5, most of thematerial is predicted to stay in circulation in the Pacific. Our resultssuggest that no large scale beaching of tsunami debris will occur,but rather, isolated items of debri
45、s will be shed from the clockwiserotating North Pacific gyre/accumulation zone and pushed towardsFig.1. Northern Tohoku and southern Hokkaido Japan (left), dots along the coastline are scaledfrom that location for the distributed release scenario. Tsunami runup (panel A) is schematized(panel B) is t
46、aken from data provided by Halpern et al. (2008). The black star indicates thecolor in this figure legend, the reader is referred to the web version of this article.)Fig. 2. The absolute difference in the number of particles after 1 year of circulation betweenbinned into 10C210 regions to calculate
47、the difference. The warmer colors represent the presencethe references to color in this figure legend, the reader is referred to the web version ofBulletin 66 (2013) 5358 55land either on the west coast of North America or on the windwardshores of the Hawaiian Islands. While this may not be the mass
48、deposition of huge amounts of material alluded to in press ac-counts of the phenomenon, it is by no means insignificant. Roughorder calculation suggest that even if only 1% of the total tsunamidebris were to be beached, this would equate to about 1 kg ofmaterial per meter of coastline assuming 15,00
49、0 km of coastlinealong western North America and Hawaii.The most interesting part to this story however will be the ulti-mate effect of the Japan tsunami debris on the total mass of mate-rial, particularly plastics, contained in the so-called Great PacificGarbage Patch. While it is virtually impossible to quantify theamount of material input into the Pacific Ocean from the Japanesein size and color according to the logarithm of the number of particles releasedfrom the data published by Mori et al. (2011), coastal population pressu
