Forum

10-fold background. Despite widespread To address concerns that extinction rates

Planetary Boundaries

criticisms, the tipping-point claim per- are an inappropriate metric, the biodiver-

for Biodiversity:

sists, with recent reproduction of the orig- sity boundary is renamed as ‘biosphere

ii

inal claim [1] and statements that the integrity’ [3]. Two static measures of bio-

Implausible Science,

threshold is ‘not arbitrary’, emerges from diversity replace rates: phylogenetic vari-

Pernicious Policies

‘massive amounts of data’ from many ability and functional diversity. Problems

1, fields, and that ‘no one is saying that of definition apart, reliable estimates for

José M. Montoya, *

’ ‘ 2 the idea is wrong , despite massive anything resembling these are impossible

Ian Donohue, and

’

3 breakthroughs in counting extinctions . to obtain at regional to global scales.

Stuart L. Pimm

As we explain in Box 1, none of these

statements are justified. Confronted with the inappropriateness of

The notion of a ‘safe operating

their measures, we are urged to keep

space for biodiversity’ is vague

Drawing attention to global environmen- using ‘in the interim’ extinction rates –

and encourages harmful policies. tal issues is certainly essential, therefore already shown to be flawed – and a ‘bio-

fi

Attempts to x it strip it of all mean- what harm is there in another approach, diversity intactness index’ [3]. The latter is

fi

ingful content. Ecology is rapidly super cially attractive, even if it has the average abundance of a broad range

gaining insights into the connec- limitations? We show that notions of of species relative to their abundance in

planetary boundaries add no insight into an undisturbed habitat. The boundary is

tions between biodiversity and eco-

our understanding of the threats to set at >90%, assessed geographically

system stability. We have no option

biodiversity and ecosystem functioning, across biomes or other large areas. This

but to understand ecological com-

have no evidence to support them, are proliferation of indices adds no useful

plexity and act accordingly.

too vague for use by those who manage insight. Even if we were able to estimate

biodiversity, and promote pernicious the necessary numbers, their limits are

How Should We Manage Human policies. Attempts to fix these problems arbitrary.

Actions That Harm Biodiversity? strip the original idea of all meaningful

Human actions obviously harm the natu- content, but still plead for the notion Finally, the purported threshold occurs for

ral world and, as we reduce the popula- of a safe operating space. Why is this the response variable of ‘biosphere func-

tions of species and drive some to deeply flawed idea so seductive, tioning’. Neither theory nor empirical data

extinction, we change ecosystems. and what problems arise from its support any threshold of biodiversity

How best should environmental science embrace? below which ecosystem function is

articulate its concerns, set research agen-

das, and advise policies? One solution

Box 1. Why Tipping Points for Biodiversity Are Fatally Flawed

embraces the notion of planetary bound-

The critical global extinction rate is operationally undefined: when the heart of the last individual of a species

aries [1] arguing that global environmental

stops beating, global extinction rate spikes momentarily. Why should this lead to planetary collapse?

‘

processes very generally have tipping Suppose we define the rate ourselves – for example in terms of extinctions per million species [2] averaged

per year or decade. Following the discovery of the Hawaiian Islands by the Polynesians 1500 years ago, they

points’. These are catastrophes involving

eliminated so many species that even the decadal global extinction rate would have been exceptional.

thresholds beyond which there will be

However, why would these extinctions of island endemics cause a collapse that putatively is both global and

rapid transitions to new states that are

only now visible? There would certainly be local consequences of species loss, but why a precipitous local

very much less favorable to human exis- collapse in ecosystems and why would it be global in extent? Furthermore, how might the rate of loss (versus

its size) be responsible?

tence than current states. The associated

notion is that humanity’s ‘business as

Certainly, there are regional physical processes for which empirical data suggests thresholds. Globally their

usual’ can only continue so long as it

existence is far from certain; they do not exist within the terrestrial biosphere in isolation [12]. Models of single

remains within some ‘safe operating

populations and local communities can show thresholds, but these neither deal with extinction rates nor i,ii space’ . global processes.

ii

Indeed, in publications [3], though not in presentations , planetary boundary arguments have moved away

The rate of human-caused extinctions –

from catastrophes, first to rapid transitions, where small changes lead to large effects, then to more gradual

now 100–1000-fold the natural back-

ones. The concession is ‘not all Earth system processes included in the planetary boundary have singular

–

ground rate [2] is one of two of the nine thresholds at the global/continental/ocean basin level’ [3]. Exactly so. This statement admits their arbitrary

global processes deemed to have nature. If anything can happen, then there is no insight gained: gradual change is embraced by entirely

arbitrary and indefinable values where the ‘safe operating space’ is transgressed.

exceeded a purported tipping point of

Trends in Ecology & Evolution, February 2018, Vol. 33, No. 2 71

compromised [4]. Defining a safe operat- species and the many species that hard, we will not need to understand its

ing space for ecosystem function makes depend on them. details. We need not define measures,

even less sense as the spatial scale and terms, processes, responses in opera-

the number of functions analyzed Irrespective of spatial scale, the bound- tional ways. In short, ecological ignorance

increases [5]. aries framework is ill-founded, inoperable, is bliss, if human actions remain within

and can have unexpected detrimental limits.

If Not Global Processes, Then effects on ecosystems.

Local Ones? Reality is different. Nothing changed glob-

‘Nevertheless’, continue the arguments, The Dangers of a Flawed ally in 1989, and this local experience has

‘it is important that boundaries be estab- Worldview many precedents elsewhere, before and

i

lished for these processes’. Why? Per- In an informative example, Rockström after. This cod collapse was unfortunate,

haps, although the planetary boundary reinforces his initial claims arguing that but overfishing is global, as appreciated

framework might add no insights into the collapse of the Newfoundland cod since the 18th century, and the term was

what we know about global human fishery in 1989 represents ‘a very precise first used (for cod) in 1855. Humans overf-

impacts, then its practical utility to envi- tipping point’ of human actions trans- ished, overharvested, overgrazed, defor-

ronmental managers might justify it. gressing global planetary boundaries. ested, polluted, and caused many other

Fatally, the boundaries framework lacks Human actions were apparently within environmental ills long before 1989 and in

clear definitions, or it has too many con- bounds before 1989. The year 1989 many other places. They have extermi-

flicting definitions, does not specify units, was apparently ‘the boundary between nated substantial numbers of species

and fails to define terms operationally, the Holocene and Anthropocene’ – a globally, and especially top predators,

thus prohibiting application by those notion we find particularly specious. The across vast swaths of land and sea,

who set policy or manage natural resour- facts are entirely prosaic: cod landings and have done so for tens of thousands

ces. Moreover, recent reviews indicate averaged about 300 000 tons from the of years.

that tipping points occur only rarely in late 1880s until the late 1950s, spiked at

natural systems [6], while policies related over threefold higher in the 1960s, and Ways Forward

to boundaries are unlikely to be evidence- the stock declined precipitously thereafter How then can environmental science sen-

based. A need for operational definitions [9]. sibly inform those who manage and set

to aid managers is self-evident [7]. policies for the complexity that is nature?

First, there is an acute moral hazard. Elsewhere, we review 42 large organiza-

At regional and local scales, managers Because there is no operational definition tions devoted to global environment man-

and conservation bodies are starting to of ‘safe operating space’, this not only agement and their various aspirational

abandon the boundaries framework. encourages arguments that ‘growth targets [7]. We applaud the Convention

i

Many claim that the adoption of bound- within limits’ is acceptable but also the on Biological Diversity (CBD) and others

aries and associated tipping points as a belief that human actions were once when they define rigorous and operational

policy goal risks biodiversity conservation. environmentally either benign or allowed targets. Good examples are 17% of land

In the case of European forests, it pro- recovery. Worse still, if the planet is not area and 10% of the ocean protected

moted interventions that harmed biodi- obviously collapsing around us, then (CBD Aichi target 11), with the areas

versity [8]. Planting of ‘resilient tree surely we can continue to deplete it. being ‘ecologically representative and

species’ – to climate change, pests, well-connected’, ‘avoiding overfishing’

and disease – and silviculture practices Second, if we suggest that a catastrophe (target 6), and preventing ‘the extinction

to promote such resilience – primarily has taken place and the consequences of known threatened species’ (target 12).

thinning to encourage growth and to are not evident, then how will managers Environmental scientists must seek ways

increase carbon storage – was recom- and policy makers trust the science we to engage policymakers to frame all their

mended to avoid reaching a tipping point do? When bad science informs policies, other aspirations similarly, because some

in forest service provisioning, primarily its future credibility is compromised. are not so clearly defined.

timber production. These recommenda-

tions run counter to biodiversity conser- Third, the planetary boundary framework At the heart of the problem are terms such

vation guidelines. They endanger old- suggests that we can view nature and its as ‘planetary boundaries’, but also ‘sus-

growth forests, veteran trees, and rela- complex ecological processes as a type tainability’, ‘health’, ‘harmony’, and

tively low-productivity native woody of black box – if we do not poke it too others, that are emotionally appealing

72 Trends in Ecology & Evolution, February 2018, Vol. 33, No. 2

but rarely, if ever, defined. They all speak We provide some examples where eco- but to understand that complexity, make

to the urgent need to understand how system change is gradual but is inextri- it operational, and act accordingly.

human impacts change ecosystems, cably tied to biodiversity loss. First, what

when at best we aspire to protect only pollinators can we not afford to lose? Acknowledgments

half of it. We must set policies and estab- Regional declines in native wild pollina- J.M.M. was supported by the French Laboratory of

Excellence Project ‘TULIP’ (ANR-10-LABX-41; ANR-

lish management for the vast tracts of tors compromise the quality and quan-

11-IDEX-002-02).

land and sea that we do not protect. tity of food crops that depend on

Fatally, those who do so often use lan- pollination. Second, how well do spe-

Resources

guage that does not borrow from the cies abundances resist harvesting or

i

Rockström, J. (2015) Abundance with Planetary

existing knowledge about ecosystem removal of top-predators – as we have

Boundaries. Presentation to the International Institute

processes, nor readily translates its aspi- done over much of the land and the

for Applied Systems Analysis (IIASA), March 12th.

fi

rations to those who study them [7]. oceans? How can we ensure that sh- www.youtube.com/watch?v=1WFtCAdCm84

eries and other exploited resources pro- ii

Rockström, J. (2017) Beyond the Anthropocene.

Fortunately, mounting evidence demon- vide reliable yields against a natural Presentation at the World Economic Forum, January

strates the patterns and mechanisms by background of year-to-year variability, 18. www.youtube.com/watch?v=V9ETiSaxyfk

which biodiversity loss alters the provision given economic drivers that require a

1

Theoretical and Experimental Ecological Station, Centre

of functions and the stability of ecosys- minimum annual return and discount

National de la Recherche Scientifique (CNRS) University

tems. We can now assess and monitor the future value of the stock? Third, Paul Sabatier (UPS), Moulis, France

2

School of Natural Sciences, Trinity College Dublin,

how losses in biodiversity affect different how can the functioning of ecosystems

Dublin, Ireland

ecosystems. This in turn allows the effec- and their associated services to humans 3

Nicholas School of the Environment, Duke University,

tiveness of a given environmental policy to persist in the face of climate change, Durham, USA

be determined. The focus must be on particularly when local extinctions

*Correspondence:

appropriate scales and variables that reduce the resistance of ecosystem

[email protected] (J.M. Montoya).

we can measure operationally. It must productivity to climate extremes [11]? https://doi.org/10.1016/j.tree.2017.10.004

recognize and define the multiplicity of

References

human actions and their consequences. Good policy means we have no option

1. World Wildlife Fund (2016) Living Planet Report 2016. Risk

We must create mutual translations of the but to understand the necessary com-

and Resilience in a New Era, WWW International

terms used by empirical ecologists, the- plexity of nature in the environments we 2. Pimm, S.L. et al. (1995) The future of biodiversity. Science

269, 347

oreticians, policymakers, and managers are starting to unravel. However,

3. Steffen, W. et al. (2015) Planetary boundaries: guiding

to describe them [7]. This way forward is acknowledging such complexities is not

human development on a changing planet. Science

shared by researchers within different dis- enough. We need the particulars – the 347, 1259855

4. Cardinale, B.J. et al. (2012) Biodiversity loss and its impact

ciplines: from those interested in the aspects of ecosystem change that we

on humanity. Nature 486, 59–67

dynamics of socioecological systems aim to minimize. Which species are vital

5. Isbell, F. et al. (2011) High plant diversity is needed to

maintain ecosystem services. Nature 477, 199–202

[10] to those centered on biodiversity to which processes, and how these con-

6. Capon, S.J. et al. (2015) Regime shifts, thresholds and

conservation [8]. nect to human social and economic sys-

multiple stable states in freshwater ecosystems; a critical

tems. We must understand how appraisal of the evidence. Sci. Total Environ. 15, 122–130

We know many useful things about these economic losses depend on the species 7. Donohue, I. et al. (2016) Navigating the complexity of

ecological stability. Ecol. Lett. 19, 1172–1185

issues, and theory and empirical studies involved and the ecological communities

8. Newton, A.C. (2016) Biodiversity risks of adopting resil-

mutually reinforce each other. We sug- in which they are embedded. We must ience as a policy goal. Conserv. Lett. 9, 369–376

9. Hutchings, J.A. and Myers, R.A. (1994) What can be

gest a way forward: to address how bio- also understand that the loss of any spe-

learned from the collapse of a renewable resource? Atlan-

diversity loss affects the different facets of cies is a loss of cultural values and poses

tic cod, Gadus morhua, of Newfoundland and Labrador.

Can. J. Fish. Aquat. Sci. 51, 2126–2146

ecosystem change [7] – resilience (how significant moral issues.

10. Dawson, T.P. et al. (2010) Dynamic properties of complex

fast systems recover), resistance (how

adaptive ecosystems: implications for the sustainability of

much they change), variability (how much There are limits to growth. When we harm service provision. Biodivers. Conserv. 19, 2843–2853

they fluctuate over time), and persistence nature, environmental changes some- 11. Isbell, F. et al. (2015) Biodiversity increases the resistance

of ecosystem productivity to climate extremes. Nature

(how long they persist). These measures times kick in immediately and in inevitably

526, 574–577

fi

of change are well-de ned, have units, complex ways that deny the simple and 12. Lenton, T.M. and Williams, H.T. (2013) On the origin of

planetary-scale tipping points. Trends Ecol. Evol. 28, 380–

can be monitored over time, and can seductive notion that, within some limited

382

inform management. They tie to pressing space, whatever the stresses we inflict on

practical problems. nature it will be OK. We have no option

Trends in Ecology & Evolution, February 2018, Vol. 33, No. 2 73