COMMENTREPRODUCIBILITY A call

to shun predatory journals p.326

FILM Biomechanics adviser to Finding Dory in conversation p.325

PHYSICS A fond history of the Cavendish, a lab with few rivals p.323

POLICY Rubric for prioritizing action on the Sustainability Development Goals p.320

How will the 10  billion people expected to be living on Earth by 2050 obtain sufficient and nutri-

tious food? This is one of the greatest chal-lenges humanity faces. Global food systems must supply enough calories and protein for a growing human population and pro-vide important micronutrients such as iron, zinc, omega-3 fatty acids and vitamins.

Deficiencies of micronutrients — so called because the body needs them only in tiny amounts — can increase the risks of perinatal and maternal mortality, growth retardation, child mortality, cognitive def-icits and reduced immune function1. The associated burdens of disease are large. Forty-five per cent of mortality in children

under five is attributable to undernutrition; nutritional deficiencies are responsible for 50% of years lived with disability in chil-dren aged four and under1.

Fish are crucial sources of micronutri-ents, often in highly bioavailable forms. And fish populations are declining. Most previous analyses have considered only how people will be affected by the loss of protein derived from fish. We calculate that this is the tip of the iceberg. Combining data on dietary nutrition, and fish catch, we predict that more than 10% of the global popula-tion could face micronutrient and fatty-acid deficiencies driven by fish declines over the coming decades, especially in the develop-ing nations at the Equator (see ‘Troubled

Waters’. This new view underlines the need for nutrition-sensitive fisheries policies.

NUTRITIONAL RISKPresently, 17% of the global population is zinc deficient, with some subpopulations being particularly at risk1. Nearly one-fifth of preg-nant women worldwide have iron-deficiency anaemia and one-third are vitamin-A defi-cient1. We estimate that 845 million people (11% of the current global population) are poised to become deficient in one of these three micronutrients if current trajectories in fish-catch declines continue.

Considering nutrients found only in foods derived from animals, such as vita-min B12, and DHA omega-3 fatty acids

Fall in fish catch threatens human health

Christopher Golden and colleagues calculate that declining numbers of marine fish will spell more malnutrition in many developing nations.

Women from a traditional sea-harvesting community fishing in Mozambique.

MIC

HA

EL D

. KO

CK

1 6 J U N E 2 0 1 6 | V O L 5 3 4 | N A T U R E | 3 1 7©

2016

Macmillan

Publishers

Limited.

All

rights

reserved.

(almost exclusively derived from meat consumption, see Supplementary Informa-tion; go.nature.com/25oll0p), we calculate that 1.39 billion people worldwide (19% of the global population) are vulnerable to deficiencies because fish make up more than 20% of their intake of these foods by weight.

IMPACT ASSESSMENTTo make this sobering new assessment, we coupled two databases from 2010, the most recent year for which both had data. The new Global Expanded Nutrient Supply (GENuS) database combines food balance sheets (total quantity of food production and imports minus livestock feed, post-harvest losses, and exports) and production or trade data from the Food and Agriculture Organization of the United Nations (FAO) with estimates of food group intake by age and sex2. It esti-mates per capita edible supplies for 225 foods, paired with regional food composition tables to infer nutrient supplies by country. GENuS is supported by the Bill and Melinda Gates Foundation and the Winslow Foundation.

We categorized populations as nutrition-ally vulnerable if their nutrient supply was less than double the estimated average require-ment (EAR), and if they derived from fish more than 10% of their vitamin A or zinc, or more than 5% of their iron. We chose to double the EAR for two reasons. First, even in countries where the national average intake is greater than the EAR, large variability of intake may still mean that a significant part of the population is eating less well. If we had used the EAR as our threshold, more than 50% of a nation’s population would need to be deemed at risk of nutritional deficiency, which we feel is an irresponsibly high pro-portion required for raising an alarm. Second, our GENuS-derived estimates measure food supply rather than food intake, and are gener-ally regarded as overestimates of true intake2.

The Sea Around Us database, released in 2016, provides a portrait of marine fisheries catch between 1950 and 2010 for every coastal nation3. Over 15 years, a team of researchers in every coastal country collated informa-tion from government documents, academic research and maritime records to reconstruct the numbers of fish caught. This database measures the contribution of subsistence, artisanal and industrial marine fisheries to food supply at the country level more accu-rately than previous estimates3. This database was funded by the Pew Charitable Trust and the Paul G. Allen Family Foundation and is maintained by staff at the University of British Columbia in Vancouver, Canada.

These global marine catch data are alarming. Conservative estimates by the FAO characterize global fisheries as stable, but acknowledge that global catch has declined by 0.38 million tonnes per year since 1996 (ref. 4). The Sea Around Us estimates summarized

earlier this year3 paint a much bleaker picture, in which fish catch peaked in 1996 and has been falling by 1.22 million tonnes (roughly 1%) per year since — three times faster than the decline reported by FAO. The degrada-tion of marine habitat by destructive fishing practices, industrial pollution, climate change and coastal development for urbanization and aquaculture is likely to further degrade ocean ecosystems and reduce fisheries yields. Such patterns call into question the ability of wild fisheries to support future demand for fish.

PERFECT STORMThe health impacts of fishery declines will hit some places harder than others. A perfect storm is brewing in the low-latitude develop-ing nations. This is where human nutrition is most dependent on wild fish, and where fish-eries are most at risk from illegal fishing, weak governance, poor knowledge of stock status, population pressures and climate change.

Sharp declines in the health of fisheries were first described in the twentieth century in high-latitude regions where industrial fishing began, such as the northwest Atlantic Ocean. Developed countries have compensated with intensive agricultural production, by import-ing goods (including fish), vitamins, supple-ments and fortified foods. Since the 1990s, the major declines in fish stocks have been in lower latitudes and developing nations. This rapid degradation probably results from the increasing industrialization of fisheries, poor governance and the accelerated expansion of foreign fishing in these regions.

These sensitivities in the tropics will be only exacerbated by climate change. Ocean warm-ing and shifts in net primary production are likely to drive remaining fish and shellfish species from low to high latitudes, potentially reducing catch globally by more than 6% and by as much as 30% in some regions (such as the tropics) by 2050 relative to recent dec-ades5. Fish will also probably get smaller in the tropics: ocean warming and associated declines in oxygen content are projected to reduce the average biomass of fish communities by around 20% during this period6. Coral reefs, essential ecosystems for many tropical coastal subsist-ence and artisanal fisheries, will be heavily degraded by warming and ocean acidifica-tion. Mangroves — nurseries for many fish that are crucial in developing nations — con-tinue to decline. Also under threat are global inland freshwater fisheries, another crucial source of nutrition and livelihood for hun-dreds of millions of people around the world7.

In these same regions, fish play an impor-tant part in the avoidance of diseases associ-ated with malnutrition. Nearly all countries that depend heavily on fish for nutrition are situated in the developing world (46 of 49); we

defined these as nations in which more than 20% of the population’s animal-based food by weight is seafood. Furthermore, countries with the highest levels of undernourishment and the weakest governance are often net exporters of seafood to well-nourished coun-tries with strong governance8.

Poor people have fewer alternatives to make up for these impending shortfalls in access to micro nutrients. Meat, eggs, vitamin supplements and imported fish can be pro-hibitively expensive. Communities are often forced to rely on what they can harvest locally or on less-healthy processed foods.

FARMED FISHCould global increases in fish farming meet the nutritional shortfalls we predict for poor equatorial populations? With today’s produc-tion and distribution patterns, we think not. Aquaculture has expanded globally by more than an order of magnitude over the past three decades9. Farmed fish exceeded wild catch destined for human consumption for the first time in 2014 (ref. 10). However, aqua-culture has not yet developed significantly in many low-income countries where food and technology is in short supply (particularly sub-Saharan Africa and the Pacific Islands11). Such regions still largely depend on domes-tic, subsistence and artisanal fishing. These small-scale fisheries include the pirogue fleets of West Africa, the spearfishers of the world’s coral reefs, and shore-based gathering of shellfish in mangrove creeks.

Where aquaculture is growing, much of it has been aimed at wealthier consumers in domestic cities or in international mar-kets, rather than local rural areas12. Globally, developing countries export higher value fish (caught and farmed) and import lower-value food from industrial fisheries in developed countries13. For example, shrimp, tilapia and Mekong catfish grown in developing and transitional countries such as Bangladesh, China, Indonesia, Ecuador, Thailand and Vietnam are mostly exported to the wealthy countries of Europe and North America, or consumed by the growing middle-classes in the megacities of these economies. The bene-fits of these export-oriented industries to live-lihoods and nutrition of the coastal poor are unclear and cannot be inferred from national seafood balance data or national economic growth statistics13. Moreover, commercial aqua culture can displace coastal and inland fisheries, and small-scale aquaculture produc-ers may not always profit from engagement in global value chains.

Farmed fish may also be of lower nutri-tional value12. Aquaculture species that are most affordable, such as carp, are often not as rich in omega-3 fatty acids as are the wild species currently accessible to impoverished communities, such as sardines and macker-els. Oil-rich wild fish are also the basis for

3 1 8 | N A T U R E | V O L 5 3 4 | 1 6 J U N E 2 0 1 6

COMMENT

“Poor people have fewer alternatives.”

© 2016

Macmillan

Publishers

Limited.

All

rights

reserved. ©

2016

Macmillan

Publishers

Limited.

All

rights

reserved.

Most reliant on �sh and most vulnerableto micronutrient malnutrition

Reliant on �sh and vulnerable tomicronutrient malnutrition

Less reliant and less vulnerable

No data

<–20% –20% to 0% >20%

Projected percentage change in maximum marine catch potential by 2050 relative to 2000 levels

0% to 20%

In the low-latitude developing nations, human nutrition is most dependent on wild �sh, and �sheries are most at risk from illegal �shing, weak governance, poor knowledge of stock status, population pressures and climate change. These countries urgently need e�ective strategies for marine conservation and �sheries management to rebuild stocks for nutritional security.TROUBLED WATERS

Guyana

Suriname

Peru

Angola

Mozambique

Malawi

Sri Lanka

Bangladesh

Indonesia

Gambia

Benin

GhanaIvory

Coast

UAE

PhilippinesLaos

Thailand

Malaysia

Congo

NigeriaCameroon

GuineaHaiti

Bahamas

Belize

Fiji

In Bangladesh, much farmed high-value �sh is exported to wealthier nations. Smallholder systems, including �sh farmed in �ooded rice �elds, have improved local food security.

Sub-Saharan African regions still largely depend on domestic, subsistence and artisanal �shing.

In developing small island states of the Paci�c, wild �sheries will move poleward because of a rise in sea temperature, and aquaculture in deltas and �oodplains will be a�ected by rising sea levels.

FrenchPolynesia

Maldives

aquaculture feeds. Because supplies of these wild-fish have little potential for expan-sion under current management policies, plant-based feeds are increasingly used, further altering the nutritional composition of farmed fish. And aquaculture generally focuses on fewer species than those caught from the wild. A global fish supply dominated by aquaculture, as it is currently practiced, would lead to a drop in the diversity, and thus nutritional quality, of many diets12.

Yet, when explicitly planned to improve local well-being, aquaculture can be a crucial contribution to local diets and economies. For example, in Bangladesh, smallholder integrated systems in which fish are raised in rice paddies have improved local food security14. Small indigenous fish, rich in nutrients, can be grown for household con-sumption in ponds together with carp, tila-pia or catfish as cash crops. Less-intensive and more-diverse forms of aquaculture may have the most potential to meet the nutrition and food-security needs of the poor.

This promise may, however, be constrained by lack of suitable sites. Both inland and coastal production are under increasing pres-sure from urbanization and industrialization. In the same small island states likely to be hit hardest by shifts in the distribution of marine species5, prospects for increasing aquaculture are at best mixed. Much aquaculture pro-duction is concentrated in river deltas and

floodplains, brackish-water lagoons and other low-lying tropical coastal areas. These areas will be heavily affected by sea-level rise, ocean acidification and increased storm intensity. Non-intensive aquaculture technologies face many challenges and are currently minor con-tributors to global production.

NEXT STEPSNext-generation models that integrate data on human health and fisheries, such as those explored here, with climate and population models need to play an important part in estimating the human health burdens of envi-ronmental change and the enormous public health dividends of natural-resource manage-ment. These models can also be used to iden-tify hotspot countries that urgently need more effective strategies for marine conservation and fisheries management to rebuild stocks for nutritional security3.

Aquaculture also needs reform so that undernourished people can access nutritious products. The following can help: boosting investment in less intensive and domestically oriented aquaculture of cheap and nutritious species; farming species lower in the food chain to reduce dependence on wild-caught fish meal; and allocating coastal land and water resource rights to small-scale aquacul-ture (as has already been done for fisheries).

The analytical methods currently in use to inform fisheries and aquaculture policies

require refinement. Data on food-price fluc-tuations are needed for local economic mod-els of fish supply and dietary substitution, and empirical research is required to under-stand and model how populations around the world will adapt to changes in fish sup-ply and market prices. We need better data on freshwater fisheries and aquaculture, as well as on the nutrient composition of foods and nutritional status of more human pop-ulations around the world. Improvements should include separating data on farmed and wild fish to better characterize vulner-ability to micronutrient deficiencies.

Addressing these emerging problems will require new interdisciplinary partner-ships among fisheries scientists, aquacul-ture technologists, ecosystem managers, nutrition and public-health specialists, development economists, granting agen-cies and policymakers. As a first step, new funding streams are required to support the emerging discipline of planetary health, dedicated to characterizing and quantifying the human health impacts of accelerating global environmental change. For example, our work has been supported by the Well-come Trust, the US National Socio-Environ-mental Synthesis Center and the Rockefeller Foundation. A second step would be more interaction between health agencies (such as the World Health Organization, the United Nations children’s fund UNICEF, and health

SO

UR

CE:

V. L

AM

, G. R

EYG

ON

DEA

U, M

. SM

ITH

& W

. CH

EUN

G

1 6 J U N E 2 0 1 6 | V O L 5 3 4 | N A T U R E | 3 1 9

COMMENT

© 2016

Macmillan

Publishers

Limited.

All

rights

reserved. ©

2016

Macmillan

Publishers

Limited.

All

rights

reserved.

ministries) and ocean-management agencies (such as the FAO, the UN Envi-ronmental Programme, regional fisheries management organizations, and minis-tries of fisheries and the environment).

Mitigating losses of biodiversity and income have been at the heart of fisheries-management policies. In our view, there should be a much stronger emphasis on human health. This would mirror recent shifts in agricultural policy that respond to rising burdens of diet-related diseases.

These policy changes are possible. We believe that improvements in fisheries management and marine conservation can serve as nutritional delivery mechanisms. A meta-analysis of nearly 5,000 fisheries worldwide found that applying sound management reforms to global fisheries could increase catch by more than 10%15. Without these changes, the health of the poor is at risk. ■

Christopher D. Golden is a research scientist at the Harvard T.H. Chan School of Public Health, and associate director of the Planetary Health Alliance at the Harvard University Center for the Environment, Harvard University, Cambridge, Massachusetts, USA. Edward H. Allison, William W. L. Cheung, Madan M. Dey, Benjamin S. Halpern, Douglas J. McCauley, Matthew Smith, Bapu Vaitla, Dirk Zeller and Samuel S. Myers.e-mail: golden@hsph.harvard.edu

1. Black, R. E. et al. Lancet 382, 427–451 (2013).2. Smith, M. R., Micha, R., Golden, C. D.,

Mozaffarian, D. & Myers, S. S. PLoS ONE 11, e0146976 (2016).

3. Pauly, D. & Zeller, D. Nature Comm. 7, 10244 (2016).

4. Food and Agriculture Organization of the United Nations. The State of World Fisheries and Aquaculture (SOFIA) 2010 (FAO, 2011).

5. Cheung, W. W. L. et al. Ecol. Model. 325, 57–66 (2016).

6. Cheung, W. W. L. et al. Nature Clim. Change 3, 254–258 (2013).

7. Youn, S. J. et al. Glob. Food Sec. 3, 142–148 (2014).

8. Smith, M. D. et al. Science 327, 784–786 (2010).

9. World Bank. Fish to 2030: Prospects for Fisheries and Aquaculture (World Bank, 2014).

10. OECD/Food and Agriculture Organization of the United Nations. OECD–FAO Agricultural Outlook 2015–2024 (2015); available at http://dx.doi.org/10.1787/agr_outlook‑2015‑en

11. Hall, S. J. et al. Proc. Nat. Acad. Sci. 110, 8393–8398 (2013).

12. Thilsted, S. et al. Food Policy 61, 126–131 (2016).

13. Asche, F., Bellemare, M. F., Roheim, C., Smith, M. D. & Tveteras, S. World Dev. 67, 151–160 (2015).

14. Dey, M. M. et al. Food Policy 43, 108–117 (2013).

15. Costello, C. et al. Proc. Nat. Acad. Sci. USA 113, 5125–5129 (2016).

Full author details and Supplementary Information accompany this article online at go.nature.com/25oll0p.

AM

IT D

AVE/

REU

TER

S

3 2 0 | N A T U R E | V O L 5 3 4 | 1 6 J U N E 2 0 1 6

COMMENT

Solar lights are used by vendors in rural western India, where lack of electricity has stymied development.

Next month in New York, the United Nations’ 2030 Agenda on Sustain-able Development will have its

first global progress review. Adopted by the UN General Assembly in 2015, the agenda represents a new coherent way of think-ing about how issues as diverse as poverty, education and climate change fit together; it entwines economic, social and environ-mental targets in 17 Sustainable Develop-ment Goals (SDGs) as an ‘indivisible whole’.

Implicit in the SDG logic is that the goals depend on each other — but no one has spec-ified exactly how. International negotiations gloss over tricky trade-offs. Still, balancing interests and priorities is what policymak-ers do — and the need will surface when the goals are being implemented. If countries ignore the overlaps and simply start try-ing to tick off targets one by one, they risk perverse outcomes. For example, using coal to improve energy access (goal 7) in Asian

Map the interactions between Sustainable Development Goals

Måns Nilsson, Dave Griggs and Martin Visbeck present a simple way of rating relationships between the targets

to highlight priorities for integrated policy.

© 2016

Macmillan

Publishers

Limited.

All

rights

reserved. ©

2016

Macmillan

Publishers

Limited.

All

rights

reserved.

COMMENTSUPPLEMENTARY INFORMATION

S U P P L E M E N T A R Y I N F O R M A T I O N | 1 6 J U N E 2 0 1 6 | N A T U R E | 1

Supplementary information to: Fall in fish catch threatens human health (Comment in Nature 534, 317–320; 2016)

Christopher D. Golden, Edward H. Allison, William W. L. Cheung, Madan M. Dey, Benjamin S. Halpern, Douglas J. McCauley, Matthew Smith, Bapu Vaitla, Dirk Zeller & Samuel S. Myers

In order to estimate the number of people worldwide nutritionally vulnerable to declines in fish catch, the GENuS model was used to estimate nutrient intake for 34 age-sex groups across all countries for which there were data1. Using the newly published catch re-constructions from the Sea Around Us database2, we more accurately reconstructed fisheries catch in each country in the world, including within the small-scale subsistence and artisanal sectors which are greatly underestimated in previous data. In addition to information on marine food resources from the Sea Around Us database, the model used intra-household allocation data from recent food consumption datasets to predict the distribution of food across age-sex groups. We then used GENuS nutrient composition data for the fish species, as well as for all other foods, to predict nutrient intake for each age-sex group.

These intakes were then used in conjunction with estimated average requirements (EARs) of nutrients to estimate the fraction of the population newly vulnerable to nutritional deficiencies following a shock to wild-capture fisheries. Because there is no standard for a threshold value to define a food as being important to the nutrient supply, following internal discussion we chose 10% for zinc and vitamin A as a reasonable value to demonstrate a high reliance on fish. On average, countries derive <1% of their caloric supply and ~5% of their total protein from fish; 10% was chosen to reflect the greater importance of fish to micronu-trient supplies, while being congruent with their contributions to macronutrient supplies as well. Regarding iron, because the type of iron supplied by fish and other animal source foods is a heme-type iron and much more easily absorbed, we used a lower value (5%) to reflect the disproportionately large importance of fish-based iron to achieving dietary sufficiency. DHA omega-3 fatty acids were deemed to be nearly exclusively obtained from animal source foods because de novo conversion rates of ALA (obtained from many vegetarian sources) to DHA omega-3 fatty acids (obtained directly only from animal products) has been estimated to occur at a rate of roughly 0.1%3.

To model vulnerability, we used Monte Carlo simulations— for each of a 1,000 itera-tions, we estimated how many people are vulnerable and deficient for each country, as well as summed for the whole world. Therefore, we have 1,000 individual country estimates for those vulnerable and deficient, as well as 1,000 estimates for the global total, and for all of these, we then calculate median and 95% uncertainty intervals. This means our “Global” es-timate represents the median of the sums, not the sum of the medians, which are not equal (See Supplementary Table).

COMMENT

2 | N A T U R E | 1 6 J U N E 2 0 1 6 | S U P P L E M E N T A R Y I N F O R M A T I O N

SUPPLEMENTARY INFORMATION

We are currently improving our global estimate of nutritionally vulnerable populations by integrating the above procedure with more detailed ecological and economic modeling. The Sea Around Us catch reconstructions will be inputted into a dynamic bioclimatic enve-lope model4 (DBEM) to predict future distributions and catch of fish species to the year 2050 that are driven by changes in ocean conditions under climate change scenarios, as well as by changes in catch effort.

We will then use these projections as supply inputs into a partial equilibrium model to evaluate the interaction of producers, consumers, and foreign traders in generating annual predictions of market clearing quantities of a wide variety of fish species5. Non-fish commodi-ties are determined outside of the system but assumptions about trends in such exogenous factors—including increases in income per capita, policy changes, technological improve-ments, and so on—can be incorporated into the model. Given available data to parameterize the supply and demand functions, especially elasticities characterizing producer and con-sumer responses to price changes, the model can be used at any geographic scale. Our work aims at global coverage of all fisheries-dependent populations at high risk of undernutrition and other health problems, largely through the aggregation of high-resolution models at the country and sub-country levels. To characterize fisheries dependence, we will need to have a better understanding of how populations substitute other foods for seafood in the context of fish declines, and to what degree populations can compensate for lost fish through dietary alternatives, fortification, and supplementation. Without accounting for these factors, we may be overestimating fisheries dependence in the current estimation. Yet, GENuS only cov-ers 152 countries, and there are several without data (Cambodia, Gabon, Vietnam, Myanmar, Kiribati, etc.) which we would expect to be vulnerable to fish declines. This, then, represents an underestimation of global vulnerability.

The economic outputs—projected quantities of various fish species available for consumption at the population level for every year to 2050—are the inputs for our refined dietary modeling approach. We assume that all artisanal and industrial catches are poten-tially available to the global trade market6 (realizing such global access may vary by country and over time) and thus captured within the import-export model. Subsistence catch enters directly into local dietary input. In the future, we hope to model dietary intake at subnational levels, applying additional strata (socioeconomic class, rural/urban location, etc.) as needed. This could include the analysis of vulnerable populations such as indigenous people who may have greater subsistence reliance on fisheries. Without these subnational analyses, many vulnerable populations around the world were unaccounted for in the current analysis. In the next iteration of the analysis, we will examine the sensitivity of these results to changes in aquaculture productivity, population growth, and GDP growth. We will assess the nutritional

COMMENT

S U P P L E M E N T A R Y I N F O R M A T I O N | 1 6 J U N E 2 0 1 6 | N A T U R E | 3

SUPPLEMENTARY INFORMATION

vulnerability of the future global population under expected dietary intakes of seafood from our projections of what fish catch will be in the future.

1. Smith, M. R., Micha, R., Golden, C. D., Mozaffarian, D., & Myers, S. S. PLoS ONE 11, e0146976 (2016).

2. Pauly, D., & Zeller, D. Nature Comm. 7, 10244 (2016).

3. Hussein, N. et al. J. Lipid Res. 46, 269-280 (2005).

4. Cheung, W. W. L., et al. Ecol. Model. 325, 57-66 (2016).

5. Dey, M. M., Briones, R. M, & Ahmed, M. Aqua. Econ. Manag. 9, 113–139 (2005).

6. Tveterås, S., et al. PLoS ONE 7, e36731 (2012).

COMMENT

4 | N A T U R E | 1 6 J U N E 2 0 1 6 | S U P P L E M E N T A R Y I N F O R M A T I O N

SUPPLEMENTARY INFORMATION

Supplementary References

Fish and Nutrition1. Black, R. E. et al. Lancet 382, 427–451 (2013).2. Thilsted, S. et al. Food Policy 61, 126–131 (2016).3. Smith, M. R., Micha, R., Golden, C. D., Mozaffarian, D., & Myers, S. S. PLoS ONE 11, e0146976 (2016).4. Youn, S. J. et al. Glob. Food Sec. 3, 142–148 (2014).5. The World Bank. Hidden harvest: the global contribution of capture fisheries. The World Bank Report 66469-GLB (World Bank, 2012).6. Srinivasan, U. T., Cheung, W. W. L., Watson, R., Sumaila, U. R. J. Bioecon. 12, 183–200 (2010).7. Bell, J. D. et al. Nature Clim. Change 3, 591–599 (2013).

Current Fisheries Status1. Pauly, D., & Zeller, D. Nature Comm. 7, 10244 (2016).2. FAO. The State of World Fisheries and Aquaculture (SOFIA) 2010 (2011).3. Costello, C. et al. Science 338, 517–552 (2012).4. McCauley, D. J., et al. Science 347, 1255641 (2015).5. Merino, G., et al. Global Environ. Chang. 22, 795–806 (2012).6. Jackson, J. B. C., et al. Science 293, 629–637 (2001).7. FAO. The world’s mangroves 1980-2005. FAO Forestry Paper 153 (FAO, 2007).8. Hamilton, S.E. & Casey, D. Glob. Ecol. Biogeogr. http://dx.doi.org/10.1111/geb.12449 (2016).

Fisheries Economics1. World Bank. Fish to 2030: Prospects for Fisheries and Aquaculture. The World Bank Report Number 83177-GLB. (World Bank, 2014).2. Smith, M. D. et al. Science 327, 784–786 (2010).3. Asche, F., Bellemare, M. F., Roheim, C., Smith, M. D., & Tveteras, S. World Dev. 67, 151–160 (2015).4. OECD/Food and Agriculture Organization of the United Nations. OECD-FAOAgricultural Outlook 2015 (OECD Publishing, 2015); available at http://dx.doi.org/10.1787/agr_outlook-2015-en

Climate Change and Fisheries1. Cheung, W. W. L. et al. Ecol. Model. 325, 57–66 (2016).2. Cheung, W. W. L. et al. Nature Clim. Change 3, 254–258 (2013).3. García Molinos, J. et al. Nature Clim. Change http://dx.doi.org/10.1038/nclimate2769(2015).4. Hoegh-Guldberg, O. et al. Science 318, 1737–1742 (2007).

Aquaculture1. Cao, et al. Science 347, 133–135 (2015).

COMMENT

S U P P L E M E N T A R Y I N F O R M A T I O N | 1 6 J U N E 2 0 1 6 | N A T U R E | 5

SUPPLEMENTARY INFORMATION

2. Marshke, M. & Betcherman, G. Environ. Dev. Sust. 1–22 (2015).3. Bjørndal, T. et al. Aquacult. Econ. Mgmt 19, 148–173 (2015).4. Salem Jr, N. & Eggersdorfer, M. Curr. Op. Clin. Nutr. Metabol. Care 18, 147–154 (2015).5. Dey, M. M. et al. Food Policy 43, 108–117 (2013).6. Fiedler, J. L., Lividini, K., Drummond, E., & Thilsted, S. H. Aquaculture 452, 291–303 (2016).7. Pomeroy, R., Dey, M. M., & Plesha, N. Aquacult. Econ. Mgmt 18, 303–324 (2014).8. Jansen, H. M. et al. Aquacult. Int. 24, 753–756 (2016).9. Alexander, K. A. et al. Aquaculture 443, 16–23 (2015).10. Diana, J. S. et al. Bioscience 63, 255–262 (2013).

Fisheries Management1. Hall, S. J. et al. Proc. Nat. Acad. Sci. USA 110, 8393–8398 (2013).2. Costello, C. et al. Proc. Nat. Acad. Sci. USA 113, 5125–5129 (2016). 3. Sale, P. F. et al. Mar. Poll. Bull. 85, 8–23 (2014).4. Jentoft, S. Maritime Stud. 13, 1–15 (2014).

COMMENT SUPPLEMENTARY INFORMATION

6 | N A T U R E | 1 6 J U N E 2 0 1 6 | S U P P L E M E N T A R Y I N F O R M A T I O N

Supplementary Figures

% c

hang

e in

MC

P

<-20

-20-0

0-20

>20

A.

COMMENTSUPPLEMENTARY INFORMATION

S U P P L E M E N T A R Y I N F O R M A T I O N | 1 6 J U N E 2 0 1 6 | N A T U R E | 7

Vitamin A

Zinc

Nutrient Supply as Proportion of EAR

Iron

3.59

2.92

2.50

2.32

2.11

1.80

1.49

1.21

0.92

0.44

5.09

3.44

2.88

2.35

1.88

1.54

1.29

1.04

0.74

0.28

7.25

3.02

2.53

2.29

2.14

1.99

1.78

1.56

1.31

0.97

0.59

3.22

4.47B.

COMMENT

8 | N A T U R E | 1 6 J U N E 2 0 1 6 | S U P P L E M E N T A R Y I N F O R M A T I O N

0

0.02

0.03

0.05

0.07

0.12

0.18

0.26

0.40

0.75Proportion of ASF as Fish (by weight)

A) The figure shows projected change in maximum catch potential by 2050 (average be-tween 2041–2060) relative to 2000 (average between 1991–2010) by Exclusive Economic Zones of the world’s coastal countries under the RCP 8.5 scenario. The projections are ob-tained from the Dynamic Bioclimate Envelope Model driven by the mean ensemble out-puts from NOAA’s Geophysical Fluid Dynamics Laboratory Earth System Model (GFDL ESM 2G), the Institut Pierre-Simon Laplace Climate Model (IPSL-CM5A-MR) and the Max Planck Institute Earth System Model (MPI-ESM-MR) [1].

B) National average nutrient supplies as proportion of each country’s population-weighted estimated average requirements (EAR). Nutrient supplies are from 2010 and are esti-mated from the GENuS nutrient supply model [2]. EARs were estimated based on FAO/WHO, IZiNCG, and US IOM recommendations by age and sex group [3-6], and popula-tion weighted using the UN World Population Prospects [7] to estimate national-average values. For age-sex categories that included pregnant and lactating women, the numbers for each were estimated using the crude birth rate from [7] and a 40-week gestational period; the average duration of breastfeeding in each country was based on WHO surveys [8], and estimates for countries without data were interpolated using regional aver-ages. For zinc, we chose from between the two possible EARs for each age-sex group based on each country’s phytate:zinc ratio (<=18 or >18), measured by [9]. For iron, there are four potential EARs based on the bioavailability of diets, which we sorted each country based on dietary criteria [10,11]: “5% bioavailability” — low meat intake (<50 g/day); “10% bioavailability” — moderate meat (50-150 g/day), low fruit & vegetable intake (<300 g/day); “12% bioavailability” — moderate meat (50-150 g/day), moderate-high fruit & vegetable intake (>300 g/day); “15% bioavailability” — high meat (>150 g/day) and fruit/vegetable intake (>300 g/day).

C.

SUPPLEMENTARY INFORMATION

COMMENT

S U P P L E M E N T A R Y I N F O R M A T I O N | 1 6 J U N E 2 0 1 6 | N A T U R E | 9

C) Fish and seafood consumption as a proportion of total animal-source food (ASF) intake. Fish/seafood and total ASF intake are measured as grams per day in 2010, estimated from [2]. Animal source foods include meat, offals, fish, seafood, dairy, eggs, and animal fats.

1. Cheung, W. W. L. et al. Ecol. Model. 325, 57–66 (2016).2. Smith, M. R., Micha, R., Golden, C. D., Mozaffarian, D., & Myers, S. S. PLoS ONE 11,e0146976 (2016).3. Vitamin and mineral requirements in human nutrition: report of a joint FAO/WHO ex-pert consultation, Geneva (2004). 4. Allen, L., de Benoist, B., Dary, O., Hurrell, R. Guidelines on food fortificationwith micronutrients (WHO/FAO, 2006). 5. Hotz, C. Food Nutr. Bull. 28 (3 suppl.), S430–S453 (2007). 6. Iron in Dietary Reference Intakes for Vitamin A, Vitamin K, Arsenic, Boron, Chromium,Copper, Iodine, Iron, Manganese, Molybdenum, Nickel, Silicon, Vanadium, and Zinc.(National Academy of Sciences, Institute of Medicine, 2001). 7. World Population Prospects: The 2012 Revision, Volume I: Comprehensive Tables. (United Nations. Department of Economic and Social Affairs, Population Division, 2013). 8. WHO Global Data Bank on Infant and Young Child Feeding (2009).http://www.who.int/nutrition/databases/infantfeeding/en/ 9. Wessells, K.R., Brown, K.H. PLoS ONE 7, e50568 (2012). 10. Hurrell, R., Egli, I. Am. J. Clin. Nutr. 91 (suppl.), 1461S–7S (2010). 11. Hallberg, L., Rossander-Hulthen, L. Am. J. Clin. Nutr. 54, 1047–58 (1991).

SUPPLEMENTARY INFORMATION