FISHERIES

The small world of global marinefisheries: The cross-boundaryconsequences of larval dispersalNandini Ramesh1*, James A. Rising2, Kimberly L. Oremus3

Fish stocks are managed within national boundaries and by regional organizations, butthe interdependence of stocks between these jurisdictions, especially as a result oflarval dispersal, remains poorly explored. We examined the international connectivity of747 commercially fished taxonomic groups by building a global network of fish larvaldispersal. We found that the world’s fisheries are highly interconnected, forming asmall-world network, emphasizing the need for international cooperation. Wequantify each country’s dependence on its neighbors in terms of landed value, foodsecurity, and jobs. We estimate that more than $10 billion in annual catch from 2005to 2014 is attributable to these international flows of larvae. The economic risksassociated with these dependencies is greatest in the tropics.

Marine fisheries supply food and liveli-hoods to millions of people around theworld (1). Though fisheries are typicallymanaged at the scale of national exclu-sive economic zones (EEZs), many fish

populations are connected beyond EEZ bound-aries (2–6). Whereas pelagic species can betracked across international borders as adults(7), nonpelagic populations connect primarilyvia the dispersal of fish eggs and larvae, formsthat cannot yet swim by ocean currents (2, 8).

Larval connectivity patterns have been ana-lyzed at both the regional (6, 9–12) and global(4, 13, 14) levels and have been used to sug-gest changes for spatial management andconservation (12, 15). However, the impacton fisheries of larval connectivity across EEZsis not well understood, even though morethan 90% of the world’s fish are caught with-in EEZs (16).On the scale of a single species or region,

this connectivity can be analyzed empirically

through genetic testing (9, 10). For analyses onlarger scales, dispersal patterns can be estimatedusing biophysical models that combine oceano-graphic data with an understanding of the bi-ology of the stocks (4, 14, 17). Such efforts can bechallenging, because species vary widely in larvaltiming and duration and currents vary with theseasons; therefore, generalizations can be mis-leading. More realistic inputs can be achievedby using life history traits for each species, in-cluding time and place of spawning and larvalduration. Sensitivity analyses can help to en-sure that results are robust to changes in keyassumptions (14), while empirical bounding cansafeguard against predicting unrealistic disper-sion outcomes (6).Network analysis has previously been applied

to marine systems to describe the connectivity ofplankton communities (18), local fishing com-munities (19, 20), and marine reserves (14).Networks of larval flows have been used toidentify “hub” subpopulations for protectionon a regional scale (12).In this study, we combined oceanographic and

life history data for 706 species and 434 generaof commercially harvested fish to estimate theirconnectivity across 249 EEZs and constructeda network representing the larval flows between

RESEARCH

Ramesh et al., Science 364, 1192–1196 (2019) 21 June 2019 1 of 4

1Department of Earth and Planetary Science, Universityof California, Berkeley, Berkeley, CA 94720, USA. 2GranthamResearch Institute, London School of Economics, London,UK. 3School of Marine Science and Policy, Universityof Delaware, Newark, DE 19716, USA.*Corresponding author. Email: nandiniramesh@berkeley.edu

Mexico

Bermuda

Cuba

Nicaragua

Honduras

BahamasTurks and Caicos

Cayman Islands

Venezuela

Colombia - Jamaica

Colombia

Panama

Puerto Rico

British Virgin Islands

Anguilla

Saint Kitts and NevisAntigua and Barbuda

Barbados

Curacao

Saba

Northern Saint-Martin

Aruba

Sint-Eustasius

Martinique

US Virgin Islands

Saint Lucia

Grenada

Saint Vincent and the Grenadines

Trinidad and Tobago

Guyana

United States

Dominican Republic

Jamaica

Haiti

Costa Rica

Bonaire

Saint Pierre and MiquelonCanada

IcelandGreenland

Svalbard

Alaska

Guatemala

SpainPortugal

El Salvador

Suriname

Ecuador

Peru Brazil

French Guiana

Montserrat

Dominica

Guadeloupe

Argentina

Liberia

Equatorial GuineaSao Tome and Principe

Falkland Islands

Uruguay

Chile

Micronesia

Japan

Palau

Indonesia

Russia

South Korea

Philippines

Paracel Islands

Japan - South Korea Conflict Zone

Vietnam

Southern Kuriles

North Korea

Spratly Islands

Conflict ZoneChina

Taiwan

Japan - Korea

Brunei

Italy

Croatia

Turkey

Germany

Denmark

Poland

Sweden

LatviaLithuania

Norway Estonia

Bulgaria

RomaniaGeorgia

Ukraine

Iran

Tunisia

France

Algeria Libya

Malta GreeceAlbania

Cyprus

EgyptIsrael

Montenegro

Morocco

United Kingdom BelgiumNetherlandsIreland

JerseyGuernsey

Madeira

Faeroe IslandsJan Mayen

Canary IslandsWestern Sahara

SyriaLebanon

Saudi Arabia

SudanYemenEritreaUnited Arab Emirates

KuwaitBahrain

Qatar

Somalia

Djibouti

Mozambique

South Africa

Bassas da IndiaIle Europa

Comoro Islands

Mauritius

Kenya

Tanzania

Mayotte

Glorioso Islands

Ile Tromelin

British Indian Ocean Territory

Maldives

Namibia

Seychelles

Juan de Nova IslandMadagascar

Oman

Pakistan

India

Papua New Guinea

Malaysia

CambodiaThailand

Singapore

Myanmar

Kiribati

American Samoa

SamoaNiue

Tonga

Wallis and Futuna

Tuvalu

Fiji

Howland Isl. and Baker Isl.

Solomon Islands

New Caledonia

Nauru

Australia - Papua New Guinea

Australia

Sri Lanka

East Timor

Andaman and Nicobar

Australia - East Timor

Australia/Indonesia Vanuatu

Marshall Islands

Bangladesh

New Zealand

Macquarie Island

Finland

Angola

GabonRepublique du Congo

Sierra LeoneTogoBenin

Ghana

Ivory CoastNigeria

Cameroon

DR Congo

Cape Verde

Western Sahara/MauritaniaMauritania

GambiaSenegal

Guinea BissauGuinea

Reunion

AntarcticaAsia PacificCaribbeanEast AfricaMediterraneanMiddle East

North AmericaNorthern EuropeSouth AmericaSouth AsiaWest AfricaWest Pacific

source clockwise sink

Fig. 1. The network of spawn-attributed catch flows between EEZs. Each EEZ is a node (circle) of the network and its color represents its networkcommunity. The connectors or edges in this network flow clockwise from source to sink, with their thicknesses representing the magnitude of the net flowof caught biomass between the EEZs. Only the edges in the upper tercile of edge weights are shown, for clarity (see SM 3.2 for the full network). The sizeof each node represents its out-degree, i.e., the number of other EEZs for which it acts as a source of fish larvae, including connections not shown in this image.

on July 16, 2020

http://science.sciencemag.org/

Dow

nloaded from

nations. Nations that depend heavily upon theirneighbors for recruitment risk losing part oftheir catch if the fisheries in the source EEZsoutside their jurisdiction are poorly managed.We quantified these risks in economic termsand identified regional “hotspots” of risk forcatch, fishery employment, and food security.We used a particle-tracking system (21) with

time-varying ocean currents (22) and species-specific life histories (23) to simulate the dis-persal of eggs and larvae through a dynamicocean. We placed multiple simulated particlesfor each species based on the timing and loca-tion of that species’ spawning and let them driftfor their larval duration to obtain a probabilisticestimate of species-specific larval trajectories.We used a random-walk parameterization (21)that adds a small velocity at every time step toaccount for turbulent motion at small scales[see section 3.1.2 in the supplementary mate-rials (SM 3.1.2)].We empirically bounded our results by dis-

carding particles that arrived in regions wherethe species is not present in observed catchdata (16). For a given EEZ, catch is attributedbased on the proportion of particles arrivingthere from each spawning country (see SM 1.1).This proportionality forms the core assumptionof our model. We tested our main results with aseries of analyses of sensitivity to this assumption.These included reducing spawn floating dura-tion to account for uncertainties in spawningmortality (2, 24), introducing return adult spawn-ing migration (25) (see SM 3.6), and distinguish-ing different levels of recruitment limitation.We estimated how much of each country’s ob-

served catch comes from its neighbors by con-structing for each species a transitionmatrix thatdescribes the probability of its offspring dispers-ing from one EEZ to another. This transfer ofbiomass between nations’ EEZs is representedas a network in Fig. 1.Each connector of the network represents

net flows of fish from one country to another.Countries that depend on inflows of juvenilefish to maintain their local populations requireinternational cooperation to ensure sustainablefisheries. Our analysis of these flows revealedthat a large proportion ofmarine fisherieswithinEEZs form a single, global network (Fig. 1).We found that the global network of marine

fisheries is a scale-free, small-world network.The scale-free network property, common innatural systems (26), is characterized by an ex-ponential distribution of the number of con-nections from each node (see SM 3.2). Thisexponential degree distribution results in a“hub-and-spoke” structure that is resilient torandom disturbances because of the large num-ber of less-connected countries from which dis-turbances do not easily propagate to other partsof the network. However, a disturbance to anyof the highly connected hubs in a scale-free net-work can affect numerous surrounding nodes.In this context, habitat destruction, overfishing,or environmental change in a hub EEZ couldhave impacts that spread beyond its own bound-

aries. Conversely, targeted efforts to managefisheries within these hub EEZs could benefitmany nations.To demonstrate the relationship between cur-

rents and the network of larval dispersal, wefocused more closely on four regions (Fig. 2).The differences between the regional networksand average current speed arise from the details

of current speeds during spawning, larval du-ration, and empirical observations of speciespresence or catch. The influence of the GuineaCurrent on the connectivity of West Africa’s fish-eries can be seen in the large number of EEZsthat act as sources to their eastward neighbors,especially between Guinea-Bissau and Nigeria.While the strongest connections are typically

Ramesh et al., Science 364, 1192–1196 (2019) 21 June 2019 2 of 4

A EWest Africa

Baltic Sea

Caribbean

B F

C G

DH

Western Pacific

Asia PacificCaribbeanNorth AmericaNorthern EuropeSouth AmericaWest AfricaWest Pacific

source clockwise sink

Communities

Liberia

Canary Islands

Western Sahara

Sierra Leone Togo

BeninGhana

Ivory Coast

Nigeria

Cape Verde

Western Sahara/Mauritania

Mauritania

Gambia

Senegal

Guinea Bissau

Guinea

Russia

Germany

Denmark

Poland

Sweden

Latvia

Lithuania

Estonia

Finland

Norway

MicronesiaPalau

Indonesia

Philippines

Northern Mariana Islands and Guam

Paracel Islands

Spratly Islands

ChinaTaiwan

Brunei

Papua New Guinea

Malaysia

Kiribati

American Samoa

Line Group

Samoa Cook Islands

NiueTonga

Phoenix Group

Wallis and Futuna

Tuvalu

Jarvis Island

Tokelau

Fiji

Solomon Islands

New Caledonia

Nauru

Australia - Papua New Guinea

Australia

Australia - East Timor

Australia/Indonesia

Vanuatu

Norfolk Island

Marshall Islands

Christmas Island

New Zealand

Palmyra Atoll

Wake Island

Pitcairn

French Polynesia

Howland Island and Baker Island

Cuba

Nicaragua

Honduras

Bahamas

Cayman Islands

Venezuela

Colombia - Jamaica

Colombia

Panama

Puerto Rico

British Virgin Islands

Anguilla

Saint Kitts and NevisAntigua and Barbuda

Barbados

Curacao

SabaNorthern Saint-Martin

Aruba

Martinique

U.S. Virgin Islands

Saint Lucia

Grenada

Saint Vincent and the Grenadines

Trinidad and Tobago

Guyana

Dominican Republic

Jamaica

Haiti

Suriname

Brazil

French Guiana

Montserrat

DominicaGuadeloupe

Costa Rica

Fig. 2. Regional currents and community networks. (A to D) The speed (shown in colors, incentimeters per second) and direction (arrows) of ocean surface currents in four regions withinterconnected fisheries (West Africa, Baltic Sea, the Caribbean, and Western Pacific) duringthe month of maximum spawning activity in each (August, May, June, and May, respectively).(E to H) The corresponding subset of the global network encompassed by the four regions.Colors, node sizing, and connector directions are as for Fig. 1. Nodes are arranged toapproximately correspond to geographic locations of the EEZs.

RESEARCH | REPORTon July 16, 2020

http://science.sciencemag.org/

Dow

nloaded from

between adjacent EEZs, many connections alsoextend over longer distances. In contrast, theBaltic Sea has substantially weaker currents.There, the largest outward flows originate fromSweden and Norway, which have the region’slongest coastlines. In the Caribbean, the NorthBrazil Current flows northwestward along theSouth American coast, and consequently manyof the EEZs lying along this current act assources for the Lesser Antilles. Within the LesserAntilles, the density of small EEZs gives rise to ahighly interconnected, complex network struc-

ture. The effect of the northward flow along thisisland chain can be inferred from the larger nodesizes among the EEZs lying in its southern por-tion. In the Western Pacific, strong currentsdominate in the equatorial ocean, with weakercurrents at higher latitudes. The large areas en-compassed by this region’s EEZs mean that, un-like the other regions, most connections arebetween immediate neighbors.The small-world property implies that it is

possible to traverse the global network in a smallnumber of steps, on average. Within this net-

work, there exist smaller clusters or communi-ties that are tightly connected. Most of theseclusters internally exhibit the small-world prop-erty. In theory, this property of the global fish-eries network suggests that disturbances to alarge hub could propagate via cascading effectson the surrounding spokes.A key question is whether disruptions to a

given EEZ actually propagate in this manner. Astock’s response to external shocks depends onboth its population dynamics and mortality fromfishing, which can be affected by management

Ramesh et al., Science 364, 1192–1196 (2019) 21 June 2019 3 of 4

Catch outflows (1000 tons) Value outflows (USD)

0 500 1000 0 300M 600M 900M

IcelandPeru

DenmarkSweden

IndonesiaArgentina

BrazilTaiwan

AustraliaSouth Korea

NorwayPapua New Guinea

RussiaUnited States

United KingdomIndia

ChinaAlaskaJapan

Outflows to other EEZsCatch inflows (1000 tons) Value inflows (USD)

0 500 1000 0 300M 600M 900M

North KoreaSwedenUruguay

DenmarkChile

TaiwanKiribatiTurkey

MyanmarPapua New Guinea

United KingdomPakistanVietnam

JapanNorwayMexico

IndonesiaChina

South KoreaRussia

Inflows from other EEZs

Resilience:

High (> 99%)

Medium (> 95%)

Low (< 95%)

Resilience:

High (> 99%)

Medium (> 95%)

Low (< 95%)

Fig. 3. Countries with highest outflowing and inflowing catches.(Left) Top 20 countries sorted by total outflowing catch (in thousands oftons) and value [in U.S. dollars (USD)] at risk. (Right) Top 20 countriessorted by total inflow of catch (thousands of tons) and value (USD) at risk.

For both catch and landed values, data from 2005 to 2014 were usedand attributed to larvae, by species. Resilience levels represent theestimated decline a population can endure without being consideredvulnerable to local extinction.

Poland

under 129129 848848 53925392 1879718797 66848over 66848

Sweden

Belgium

Netherlands

Poland

Latvia

Estonia

Lithuania

Romania

Finland

North Korea

Bahamas

Sao Tomeand Principe

St. Lucia

Uruguay

Dominica

GuyanaSt. Vincent andthe Grenadines

Barbados

Suriname

GambiaMauritania

Cameroon

Guinea-Bissau

Comoros

Maldives

Pakistan

Bangladesh

Indonesia

Philippines MarshallIslands

Kiribati

NauruPapua New Guinea

TuvaluSolomonIslands

Palau

Catch Value

Food Security

GDP

Labor

< 129129-848848-5,3925,392-18,79718,797-66,848> 66,848N/A

Catch (1000 tons)

Fed. Statesof Micronesia

Fig. 4. Hotspot map showing fishing dependency on spawninggrounds in neighboring waters, by country. Countries are shadedby catch (in thousands of metric tons) at risk, with darker shadesrepresenting higher catches. Icons depict EEZs that are the mostdependent on their neighbors. The catch icon indicates that more than

30% of a country’s catch value is dependent on neighboring spawninggrounds, the GDP icon represents a risk to more than 0.8% of itsGDP, the labor icon represents that more than 1.5% of its jobs arevulnerable, and the food security icon represents a food securitydependence index >1.1%.

RESEARCH | REPORTon July 16, 2020

http://science.sciencemag.org/

Dow

nloaded from

(27). Some fish stocks are biologically capableof replenishing themselves when their numbersdwindle, provided that fishing pressure is re-lieved, reducing the likelihood that disturbanceswill propagate. However, “recruitment-limited”stocks are vulnerable to a decline in spawningpopulation, making it more likely that distur-bances will spread across the network even ifthe receiving fisheries are managed. We adoptedFishBase’s classification of stock resilienceas a proxy for this type of density dependence.For high-resilience stocks, which are generallynot recruitment limited, our measure of stockdependence overestimates the extent to whichstocks will be reduced if recruitment inflows fail.For those stocks classified as having medium orlow resilience, however, we found a strong cor-relation between our simulation’s predictionsand observed variances in stock levels (see SM3.5). Even for countries whose fisheries mostlycomprise non–density-dependent stocks, theselarval inflows serve as a buffer against fisherycollapse within their waters.To contextualize our results, we estimated the

economic significance of the network’s interna-tional connections. First, we considered theamount and value of catch that flows in and outof each EEZ (Fig. 3). Japan, China, and Alaskaare responsible for the greatest outflows, reflect-ing their productive waters. However, havingfewer neighbors makes them smaller hubs(Fig. 1). Indonesia has the most landed value at-tributable to other countries, due to its high-value catch andmany neighbors. The countrieswith the greatest catch inflows are generallythose with the largest fisheries.Next, we identified nations that are potentially

most vulnerable, in socioeconomic terms, tothe management of neighboring waters (seetable S5). In Fig. 4, we highlight countries thatdepend the most on the spawning grounds ofneighbors in terms of their total catch, grossdomestic product (GDP), number of jobs in thefishery industry, and a fishery food security de-pendence index (28). Themost vulnerable nationsare concentrated in the hotspot regions of theCaribbean, West Africa, Northern Europe, andOceania. The risks to national GDP and laborforce are generally highest in the tropics. How-ever, our measure of food security risk alsoidentified a few European nations.

Our analysis showed that about $10 billionworth of annual marine catch may rely ontransnational exchanges of fish offspring. Thesedependencies form a single global network, in-dicating that marine fisheries, even within na-tional boundaries, constitute an interconnected,globally shared resource.This network’s scale-free and small-world prop-

erties imply that fish stocks from a small num-ber of EEZs provide benefits to a large number of“downstream” countries. The most vulnerablenations are clustered in a few hotspot regions(Fig. 4). This pattern lends further support to theuse of international frameworks, such as largemarine ecosystems and marine protected areanetworks (29, 30).Further research is needed to understand how

small-scale coastal processes, larval behavior, andfisheries management affect this connectivity.Beyond the spawning connections studiedhere, national fisheries are interdependent throughthe movement of adult fish, population shiftsunder climate change, and international fishingtreaties. In particular, the role of adult fish mi-gration in driving international connectivity re-mains an important question. While a moredetailed analysis is required to accurately de-scribe dispersal pathways of individual species,this study highlights the role of larval connec-tivity across international boundaries and theneed for multilateral cooperation for sustain-able management of these shared resources.

REFERENCES AND NOTES

1. FAO, “The state of world fisheries and aquaculture:Contributing to food security and nutrition for all” (Food andAgriculture Organization of the United Nations, 2016).

2. R K. Cowen, S. Sponaugle, Annu. Rev. Mar. Sci. 1, 443–466 (2009).3. A. Di Franco et al., Biol. Conserv. 192, 361–368 (2015).4. E. Popova et al., Mar. Policy 104, 90–102 (2019).5. B. P. Kinlan, S. D. Gaines, Ecology 84, 2007–2020 (2003).6. A. S. Kough, C. B. Paris, M. J. Butler 4th, PLOS ONE 8, e64970

(2013).7. B. A. Block et al., Nature 434, 1121–1127 (2005).8. D. A. Siegel et al., Proc. Natl. Acad. Sci. U.S.A. 105, 8974–8979

(2008).9. S. Planes, G. P. Jones, S. R. Thorrold, Proc. Natl. Acad. Sci. U.S.A.

106, 5693–5697 (2009).10. N. K. Truelove et al., Fish. Res. 172, 44–49 (2015).11. M. Dubois et al., Glob. Ecol. Biogeogr. 25, 503–515 (2016).12. J. R. Watson et al., Proc. Natl. Acad. Sci. U.S.A. 108,

E907–E913 (2011).13. S. Wood, C. B. Paris, A. Ridgwell, E. J. Hendy, Glob. Ecol.

Biogeogr. 23, 1–11 (2014).14. M. Andrello et al., Nat. Commun. 8, 16039 (2017).

15. S. D. Gaines, C. White, M. H. Carr, S. R. Palumbi, Proc. Natl.Acad. Sci. U.S.A. 107, 18286–18293 (2010).

16. D. Pauly, D. Zeller, Eds., Sea Around Us: Concepts, Design andData (Univ. of British Columbia, 2015).

17. M. J. Fogarty, L. W. Botsford, Oceanography 20, 112–123(2007).

18. B. F. Jönsson, J. R. Watson, Nat. Commun. 7, 11239 (2016).19. E. C. Fuller, J. F. Samhouri, J. S. Stoll, S. A. Levin, J. R. Watson,

ICES J. Mar. Sci. 74, 2087–2096 (2017).20. E. T. Addicott et al., Can. J. Fish. Aquat. Sci. 76, 56–68

(2019).21. C. B. Paris, J. Helgers, E. van Sebille, A. Srinivasan, Environ.

Model. Softw. 42, 47–54 (2013).22. J. A. Carton, B. S. Giese, Mon. Weather Rev. 136, 2999–3017

(2008).23. R. Froese, D. Pauly, FishBase, Version 11/2014 (2014); www.

fishbase.se/search.php.

24. C. C. D’Aloia et al., Proc. Natl. Acad. Sci. U.S.A. 112,13940–13945 (2015).

25. A. Hastings, L. W. Botsford, Proc. Natl. Acad. Sci. U.S.A. 103,6067–6072 (2006).

26. D. J. Watts, S. H. Strogatz, Nature 393, 440–442 (1998).

27. D. Pauly et al., Nature 418, 689–695 (2002).

28. M. Barange et al., Nat. Clim. Chang. 4, 211–216 (2014).

29. B. S. Halpern, S. E. Lester, K. L. McLeod, Proc. Natl. Acad. Sci. U.S.A.107, 18312–18317 (2010).

30. J. Lubchenco, Stress, Sustainability, and Development of LargeMarine Ecosystems during Climate Change: Policy andImplementation (UNDP and GEF, 2013).

31. N. Ramesh, J. Rising, K. Oremus, The Small World of GlobalMarine Fisheries: The Cross-Boundary Consequences of LarvalDispersal, Version 1.0, Zenodo (2019); http://doi.org/10.5281/zenodo.2636745.

ACKNOWLEDGMENTS

The authors thank D. Dookie, M. Burgess, M. A. Cane,A. Chaintreau, A. Carlisle, J. Cohen, C. Costello, R. Defries,S. Gaines, S. Hsiang, C. Moffat, and C. Szuwalski for comments,suggestions, and references. This work was initiated and partlyconducted at Columbia University in the City of New York, whichprovided computing resources. Funding: N.R. was partiallysupported by the National Aeronautics and Space Administration(NASA) Headquarters under the NASA Earth and Space ScienceFellowship Program, grant NNX-14AK96H. Author contributions:N.R. performed the network analysis and Lagrangian modeling.J.A.R. performed the country-level risk analysis. N.R., J.A.R., andK.L.O. designed the study, collected data, and wrote the paper.Competing interests: The authors declare no competing interests.Data and materials availability: All newly organized data usedin this study and the intermediate and final results data arepublicly available at Zenodo (31). Analysis reproduction code isavailable at https://github.com/openmodels/small-world-fisheries.

SUPPLEMENTARY MATERIALS

science.sciencemag.org/content/364/6446/1192/suppl/DC1Materials and MethodsFigs. S1 to S13Tables S1 to S8References (32–44)

6 September 2018; accepted 23 May 201910.1126/science.aav3409

Ramesh et al., Science 364, 1192–1196 (2019) 21 June 2019 4 of 4

RESEARCH | REPORTon July 16, 2020

http://science.sciencemag.org/

Dow

nloaded from

dispersalThe small world of global marine fisheries: The cross-boundary consequences of larval

Nandini Ramesh, James A. Rising and Kimberly L. Oremus

DOI: 10.1126/science.aav3409 (6446), 1192-1196.364Science 

, this issue p. 1192Sciencemanagement, and food supplies globally.particular hubs of productivity are widely important. Such connectivity has wide-ranging implications for conservation,that global fish populations represent a small-world network where connections across populations are tight and

seemsanalysis to assess the degree to which populations found in one part of the world may have come from another. It model how these currents distribute the fish larvae of more than 700 species. They used networket al.currents. Ramesh

marine fish, perhaps more than any other vertebrate group, are connected across large distances through ocean Countries manage their fisheries as if they were a local resource. To some degree, this may reflect reality, but

A small, interconnected world

ARTICLE TOOLS http://science.sciencemag.org/content/364/6446/1192

MATERIALSSUPPLEMENTARY http://science.sciencemag.org/content/suppl/2019/06/19/364.6446.1192.DC1

REFERENCES

http://science.sciencemag.org/content/364/6446/1192#BIBLThis article cites 33 articles, 7 of which you can access for free

PERMISSIONS http://www.sciencemag.org/help/reprints-and-permissions

Terms of ServiceUse of this article is subject to the

is a registered trademark of AAAS.ScienceScience, 1200 New York Avenue NW, Washington, DC 20005. The title (print ISSN 0036-8075; online ISSN 1095-9203) is published by the American Association for the Advancement ofScience

Science. No claim to original U.S. Government WorksCopyright © 2019 The Authors, some rights reserved; exclusive licensee American Association for the Advancement of

on July 16, 2020

http://science.sciencemag.org/

Dow

nloaded from