23 DECEMBER 2016 • VOL 354 ISSUE 6319 1531 SCIENCE sciencemag.org
The reductions in catch may not sound
momentous, but as Cheung et al. show, the
global total hides a lot of regional variation.
Temperate and polar regions may see little
change or even potential increases in fish
abundance and catch, although these gains
could be lost under more extreme climate
change. In stark contrast, under a 3.5°C
temperature increase, the authors forecast
catch reductions of as much as 20 to 80%
for tropical locations. The main reason for
these large losses is the high turnover of ex-
ploited species in tropical locations, where
many people subsist on seafood and depend
on fishing for their livelihood (6). These find-
ings are in line with other model results that
have indicated much larger risks to coastal
habitats, ocean processes, human health, and
marine industries from rising temperatures
and falling ocean pH (7). The model results
thus underline how meeting the Paris target
can help minimize both environmental and
Research at a regional scale has come to
similar conclusions for terrestrial ecosystem
dynamics. In a recent study, Guiot and Cra-
mer (8) combined models and reconstruc-
tions of past climate and ecosystem dynamics
to show that keeping to a 1.5°C temperature
rise is the only means of staying within the
range of conditions seen in the past ~10,000
years (the Holocene) around the Mediterra-
nean. This is important because all human
societies developed under Holocene condi-
tions, which are assumed to mark our plan-
et’s safe operating space for humanity (9).
Modeling, as a tool for defining plausible
options, has a history of more than 4000
years (10). Over the past decade, however,
it has seen a renaissance, particularly in the
diversity and scope of models and how they
are used to inform decision-making. In the
marine domain, trophic ecosystem models
have been joined by qualitative, empirical,
size-based, agent-based, and bespoke models
of varying complexity to study fisheries management and biodiversity conservation in the
context of global change (11, 12). The diversity of approaches helps to address scientific
uncertainty and also provides modeling options even when the available data are poor.
Marine ecosystem modeling thus not only
provides global insights across fisheries (3)
and other marine ecosystem services (7), but
also helps to put fisheries and oceans management on a more sustainable footing (13).
Marine modeling, together with observed rapid changes in the world’s oceans,
has shown that both adverse change and
beneficial human responses can occur
with remarkable speed (7, 13). Constructive change typically has a cost and this
often makes it unattractive, because our
cognitive machinery amplifies the imagined disadvantages of change while un-derappreciating potential benefits (1).
Nevertheless, at both regional and global
scales, models are repeatedly showing us
that overcoming or accepting short-term
costs can lead to long-term gains in improved ecosystem status, food security, and
societal resilience (3, 7, 8, 13, 14).
Current impact models (3, 8) largely focus
on single drivers, such as temperature. Reliable methods of representing the complex
drivers and effects of global change are yet
to be developed. Nonetheless, the futures
that the models present already provide
an impetus to act. Current climate pledges
under the Paris Agreement do not provide
the reductions in greenhouse gas emissions
needed to keep temperatures below 2°C, let
alone 1.5°C (15). But for those people relying on the close to 25 billion seafood meals
lost each year if we let temperatures rise by
3.5°C, the imperative to act seems clear. Humans today are better equipped to make ev-idence-based decisions than at any point in
our collective history, and all that evidence
tells us that deferring the decision to act is
a costly course to take. j
1. D.Kahneman, Thinking,Fastand Slow (Farrar,Strausand
Giroux, New York, 2013).
2. J.Zipes, The Irresistible Fairy Tale: The Cultural and Social
History of a Genre (Princeton Univ. Press, Princeton, NJ,
3. W. W.L.Cheung etal.,Science 354, 1591(2016).
4. IPCC,Summaryforpolicymakers,in Climate Change 2014:
Impacts, Adaptation, and Vulnerability. Part A: Global and
Sectoral Aspects. Contribution of Working Group II to the
Fifth Assessment Report of the Intergovernmental Panel
on Climate Change, C. B. Field et al., Eds. (Cambridge Univ.
Press, Cambridge/New York, 2014), pp. 1–32.
5. IEA, World Energy Outlook,Special Briefingfor COP21
(International Energy Agency, Paris, 2015); www.iea.org.
6. E.Allison et al., Fish Fish. 10,173(2009).
7. J.-P.Gattuso et al., Science 349,aac4722(2015).
8. J. Guiot, W. Cramer, Science 354, 465 (2016).
9. W.Steffen et al., Science 347,736(2015).
10. H.Scichl,in Modeling Languages in Mathematical
Optimization, J. Kallrath, Ed. (Kluwer, London, 2004), chap. 2.
11. J.S.Collie et al., Fish Fish. 17,101(2016).
12. E. A. Fulton, J. S. Link, in The Sea, M. J. Fogarty, J. J. McCarthy,
Eds. (Harvard Univ. Press, Cambridge, 2014), vol. 16, pp.
13. E. A. Fulton et al., PLOS ONE 9, e84242 (2014).
14. J.L.Blanchard et al., J. Appl. Ecol. 51,612(2014).
15. J.Rogelj et al., Nature 534,631(2016).
A local fisherman takes his catch to market on
the tropical Keys Island off the coast of Honduras.
Tropical fish catches are likely to be much reduced
as global temperatures rise.