Ecosystem management as a wicked problem

Abstract

Ecosystems are self-regulating systems that provide societies with food, water, timber, and other resources. As demands for resources increase, management decisions are replacing self-regulating properties. Counter to previous technical approaches that applied simple formulas to estimate sustainable yields of single species, current research recognizes the inherent complexity of ecosystems and the inability to foresee all consequences of interventions across different spatial, temporal, and administrative scales. Ecosystem management is thus more realistically seen as a “wicked problem” that has no clear-cut solution. Approaches for addressing such problems include multisector decision-making, institutions that enable management to span across administrative boundaries, adaptive management, markets that incorporate natural capital, and collaborative processes to engage diverse stakeholders and address inequalities. Ecosystem management must avoid two traps: falsely assuming a tame solution and inaction from overwhelming complexity. An incremental approach can help to avoid these traps.

Full text

REVIEW
Ecosystem management
as a wicked problem
Ruth DeFries
1
*and Harini Nagendra
2
Ecosystems are self-regulating systems that provide societies with food, water, timber, and
other resources. As demands for resources increase, management decisions are replacing
self-regulating properties. Counter to previous technical approaches that applied simple
formulas to estimate sustainable yields of single species, current research recognizes the
inherent complexity of ecosystems and the inability to foresee all consequences of
interventions across different spatial, temporal, and administrative scales. Ecosystem
management is thus more realistically seen as a “wicked problem”that has no clear-cut
solution. Approaches for addressing such problems include multisector decision-making,
institutions that enable management to span across administrative boundaries, adaptive
management, markets that incorporate natural capital, and collaborative processes to
engage diverse stakeholders and address inequalities. Ecosystem management must avoid
two traps: falsely assuming a tame solution and inaction from overwhelming complexity. An
incremental approach can help to avoid these traps.
People modify and manage ecosystems to
provide food, energy, building materials,
and other resources, as well as to filter wa-
ter, control infectious diseases, decompose
wastes, and connect with nature. Ecosys-
tem managers who oversee the provision of these
resources and services to society range from gov-
ernment administrators, policy-makers, and indus-
try officials to farmers, fishers, and foragers. They
collectively manage many types of ecosystems,
including forests, grasslands, lakes, rivers, coastal
areas, farms, protected areas, and cities. In this
Review, we assess how views toward ecosystem
management have changed over time and what
approaches can guide ecosystem management in
the changing ecological and socioeconomic reali-
ties of the 21st century.
U.S. Department of Agriculture Forest Service
Chief F. Dale Robertson coined the term “ecosystem
management”in 1992, describing an ecological ap-
proach to “blend the needs of people and envi-
ronmental values in such a way that the National
Forests and Grasslands represent diverse, healthy,
productive and sustainable ecosystems”(1). Long
before this formal definition, preindustrial tradition-
al societies embraced many principles of ecosys-
tem management. Indigenous and local knowledge
evolved to suit local conditions, with social insti-
tutions based on ecological understanding of the
local dynamics of the resource base (2,3). Two
of many examples are the pre-Columbian Incan
prohibition against harvesting guano, a valuable
fertilizer, during bird nesting season (4)andthe
widespread existence of sacred groves that pro-
vide refuge for biodiversity (5).
For most of the industrialized age, extraction
of maximum yields of fish, timber, and other re-
sourceswastheprevalent paradigm. Science to
manage forests emerged in Germany in the latter
18th century, with a single focus on timber pro-
duction. This focus spread throughout the indus-
trialized world—for example, to the United States,
which managed national forests for efficient pro-
duction of timber to support the growing housing
industry after World War II (6). Similarly, mana-
gers used fisheries science to maximize catch on
the basis of historical statistics of single species
(7,8). The purpose of management was to achieve
a theoretical sustained yield and efficiently manage
only those species with commercial value. The ap-
proach was based on the concept that ecosystems
exist in equilibrium and that once an ecosystem
DeFries et al., Science 356,265–270 (2017) 21 April 2017 1of6
1
Department of Ecology, Evolution and Environmental Biology,
Columbia University, New York, NY 10027, USA.
2
School of
Development, Azim Premji University, Bengaluru, India.
*Corresponding author. Email: rd2402@columbia.edu
Corridors assist with complex ecosystem
management. Canada’s Bow Valley Provincial
Park is part of a wildlife corridor that attempts to
balance local resource use and ecosystem
health across borders.
PHOTO: STOCKTREK IMAGES/NATIONAL GEOGRAPHIC CREATIVE
on April 20, 2017http://science.sciencemag.org/Downloaded from

reaches a climax state, yields can be controlled
and sustained indefinitely (9).
Ecological research in the 1970s and 1980s
revealed, however, that ecosystems are not in a
continual state of stable equilibrium. Rather, dis-
turbances such as fire, wind, floods, and drought
are integral to maintaining healthy ecosystems
(10,11). This shift in thinking implied that managing
ecosystems was more complex than extracting
predetermined sustainable yields. Disturbances are
necessary rather than harmful, viability of com-
mercially relevant species depends on the whole
ecosystem, and populations vary spatially and tem-
porally. In short, ecosystems function as complex,
dynamic systems with nonlinear responses to inter-
nal and external forces, feedbacks across space and
time, thresholds, and inherent unpredictability (12).
Another fundamental shift from the maximum-
yield paradigm came from the realization that
human societies depend on ecosystems for well-
being and services other than commodities. In
2005, the Millennium Ecosystem Assessment
called attention to the multiple services provided
by ecosystems, including provisioning (e.g., food),
regulating (e.g., water filtration and decomposi-
tion of wastes), supporting (e.g., soil formation),
and cultural services (e.g., recreation) (13). Eco-
systems perform several functions simultaneously.
A forest can sequester carbon from the atmosphere,
provide habitat for biodiversity, constitute a sacred
and recreational space, and produce timber.
Awareness of human behavior and social in-
stitutions as integral parts of ecosystems has also
been growing. For example, the spread of infec-
tious disease depends on human behaviors, such
as contact with others, as well as population dy-
namics of disease vectors. These interactions give
rise to complex dynamics (14). Moreover, in a global-
ized world, trade and exchanges between distant
regionsalternormsandrulesofresourceextrac-
tion in increasingly unpredictable ways (15).
The realization that ecosystems behave as
complex systems, with humans as a component,
has upended the notion that managers can pre-
dictably obtain resources from ecosystems by
following simple formulas and exerting top-down
control. Problems in ecosystem management, such
as reducing mortality from infectious diseases (14)
and improving water quality affected by non–point
source pollutants (16), have proven to be much
less tractable than once thought.
Complexity gives rise
to wicked problems
Becauseecosystemsareinherentlydynamicand
largely unpredictable complex systems, ecosystem
management is a “wicked problem”(17–19). The
concept of wicked problems arose more than
30 years ago in response to the dominance of
top-down, expert-driven technical and engineer-
ing solutions to thorny issues in public policy,
such as poverty alleviation and unemployment
in urban communities (20). Wicked problems are
inherently resistant to clear definitions and easily
identifiable, predefined solutions. In contrast, tame
problems, such as building an engineered struc-
ture, are by definition solvable with technical solu-
tions that apply equally in different places. Wicked
problems have been described in many disci-
plines, including public health, political science,
business management, urban and regional plan-
ning, and natural resource management.
Rittel and Webber (19) have defined 10 pri-
mary characteristics of wicked problems, in-
cluding the elusiveness of a final resolution, no
definitive test for a solution, and no generalizable
solution that applies in all cases. Wicked prob-
lems are seemingly intractable and subject to
multiple interpretations.
Heifetz (21) classified problems in terms of
their wickedness. Type I problems are technical
in nature and have clearly defined questions and
mechanical, straightforward solutions (i.e., they
are tame). Type II problems are clearly definable
but have no clear-cut solution. Solutions to type
IIproblemsareonlyproposalsthatmustbetested
and refined on the basis of outcomes. Type III
problems have neither clear-cut definitions nor
technical solutions. Type III problems are the most
wicked and require continual learning to formulate
the problem and adaptively work toward solutions.
In ecosystem management, researchers have
identified many types of wicked, non–type I prob-
lems (Table 1). Wicked problems arise from one
or a combination of multiple dimensions (20):
complexity and interdependency of components,
which create feedbacks and nonlinear responses
to management interventions; uncertainty of risks
and unintended consequences; divergence in values
and decision-making power of multiple stakehold-
ers; and mismatches in spatial and temporal
scales of ecological and administrative processes.
Tame, type I problems such as controlling point
source pollution are amenable to technical solu-
tions. Wicked, non–type I problems that involve
inherently unpredictable complex ecological sys-
tems, compounded by human behavior and socio-
economic complexities, require incremental and
adaptive approaches to continually reframe their
definition and develop incremental solutions.
Increasing wickedness
in the 21st century
Several realities of the 21st century make ecosystem
management an increasingly wicked problem.
For most of human history, ecosystems essentially
functioned as self-regulating adaptive systems with
self-organizing properties that evolved through
long-term interactions between populations and
their environments (12). Human societies bene-
fited from these properties to obtain resources.
In the 21st century, humans have increasingly
used approaches that replace or supplement eco-
system functions—such as pesticides that replace
the ecological function of natural predators of
pest species and fertilizers that augment nutri-
ent cycling, a fundamental role of ecosystems.
These approaches often do not mimic the self-
regulating properties of ecosystems. The mismatch
gives rise to unintended consequences, such as
the loss of natural predators of pest species and
the accumulation of excess nutrients in water-
ways that receive runoff from fertilized fields. Such
human management has helped unprecedented
numbers of people to escape extreme poverty and
DeFries et al., Science 356,265–270 (2017) 21 April 2017 2of6
Table 1. Examples of wicked problems in ecosystem management.
Management problem Example of management
question Reasons for wickedness
Control of infectious
disease
Where and when will
outbreaks and spread of
infectious diseases occur? Will
vaccinations and other
control strategies be effective?
Nonlinear population dynamics of
disease vectors and hosts; feed-
backs between disease incidence
and human behavior (14)
.................................... .................................. ..................................... .................................. ..................................... .................................. .
Non–point source
pollution
Which sources (e.g., agricultural
or urban runoff) and pollutants
should managers target to
improve water quality?
Complex causality from multiple
pollution sources; long lag times in
system response (16)
.................................... .................................. ..................................... .................................. ..................................... .................................. .
Fire management
at urban-wildland
interfaces
Should managers
suppress fires?
Unintended consequences of
increased fuel loads and fire
severity (66)
.................................... .................................. ..................................... .................................. ..................................... .................................. .
Conservation of
biodiversity
How much and where should
places be protected from
other sectors (e.g., mining,
roads, fishing)?
How to protect species with
ranges outside protected
areas?
How to rectify inequities for
local communities that lose
access to resources?
Differences in values (35);
mismatch in ecological and
administrative boundaries (42);
divergent objectives of
stakeholders (56)
.................................... .................................. ..................................... .................................. ..................................... .................................. .
ECOSYSTEM EARTH
on April 20, 2017http://science.sciencemag.org/Downloaded from

undernourishment (22), but it often lacks the built-
in feedback loops that lead to self-regulation.
Another reason for the increasing wickedness
of ecosystem management is the separation in
space between the production and consumption
of resources. As a result of these “teleconnections,”
locations that experience the consequences of
resource use are disassociated from those where
demand originates. For example, tropical forests
are cut down to enable export of feed and vege-
tableoilsacrosscontinents(23). Decoupling deci-
sions from local impacts reduces the likelihood of
self-regulating feedbacks that would change man-
agement practices when negative impacts arise,
such as political pressure to control local pollution.
Ostrom (24) famously identified eight design
principles for successful management of common
pool resources such as fisheries and forests. These
principles are applicable when small groups have
local control over decisions, information, and
institutional arrangements and thus can change
management practices. With teleconnections, pos-
sibilities for applying many of the Ostrom princi-
ples break down. For example, communities often
no longer have opportunities to match rules gov-
erning the use of common resources to local needs
and conditions, ensure that outside authorities re-
spect the rule-making rights of community mem-
bers, or access means for dispute resolution.
Third, concern about inequalities in access to
ecosystem resources is becoming more common,
particularly where subsistence communities that
depend on local ecosystems for fishing, forest pro-
duct collection, or grazing are negatively affected
by conservation or land acquisitions. To address
these concerns, ecosystem management approaches
are becoming more inclusive—for example, by en-
suring that protected area management involves
collaborating with local communities (25). These
shifts are welcome from human rights and equity
perspectives, but they increase wickedness for eco-
system management by adding stakeholders with
divergentviews,norms,goals,power,andinfluence.
Systemic approaches
to wicked problems
Many researchers have called for systemic ap-
proaches to ecosystem management to replace
previous, equilibrium-based methods for extract-
ing maximum yields (26–28). These approaches
are detailed in the following sections and Table 2.
Multisector decision-making
Stated goals for management of landscapes and
seascapes often aim to provide a single ecosys-
tem service. For example, protection of forests
in the Catskills aims to filter water for use by
New York City (29), and scenic value was the jus-
tification for establishing Yellowstone, the first
U.S. national park (30). In reality, landscapes
and seascapes often provide multiple services
simultaneously, including marketed services such
as timber and fisheries and nonmarketed services
such as flood protection, erosion control, and cli-
mate regulation through carbon storage. Manage-
ment decisions can lead to trade-offs or synergies
among ecosystem services (31). Advanced models
allow evaluation of the trade-offs and synergies
from different management scenarios to inform
decision-making (32,33).
However, sector-wise administrative structures
limit abilities to weigh outcomes of management
decisions that affect sectors outside a given sector ’s
mandate. For example, the decision to establish a
protected area might lie within the jurisdiction
of an environment ministry, but the repercus-
sions for local people who would be excluded from
using resources from a newly established protected
area might lie outside the purview of that ministry.
With “stovepiped”decision-making, opportunities
to account for multiple ecosystem services affected
by management decisions are limited. At a na-
tional or subnational scale, spatial planning to
compare outcomes of land use scenarios for multi-
ple ecosystem services can overcome sector-wise
decision-making; an example is national-scale
planning to determine where economically im-
portant oil palm plantations could be located
with minimal carbon emissions in the highly
forested country of Gabon (34). Multifunctional
ecosystem management also requires multilevel
governance systems that recognize the importance
of state, community, and private ownership (35).
Decision-making across
administrative boundaries
Ecological processes encompass spatial scales that
often transcend administrative boundaries. Nation-
al and provincial boundaries can cut across water-
sheds, airsheds, and home ranges of mammals
and birds. Consequently, ecosystem management
in one country or province can affect ecosystem
services in other jurisdictions. For example, winds
transport polluted air from fires in Indonesia to
downwind countries, with major public health
consequences (36). Also, protected areas common-
ly do not encompass the full geographic extent of
migration patterns, as is the case for grizzly bears
in western North America (37) and endangered
tigers in central India (38). Further, water impound-
ments can trap sediment upstream, leading to
downstream loss of wetlands, as has occurred in
the Mississippi delta (39).
In the absence of governance arrangements
that span administrative boundaries, managers
lack incentives, authority, and mechanisms for
considering the consequences of their decisions
beyond their own jurisdictions. Institutional mech-
anisms can address this mismatch in ecological
and administrative boundaries; examples include
the Mekong River Commission (40) and the As-
sociation of South East Asian Nations (ASEAN)
Agreement on Transboundary Haze Pollution (41).
Institutional mechanisms for landscape-scale con-
nectivity for wildlife movement include the Yellow-
stone to Yukon corridor (42)andtheMesoamerican
Biological Corridor (43).
Adaptive management
The essence of adaptive management is learning
by doing and recognition of uncertain outcomes.
Adaptive management requires an explicit con-
sideration that the future may be unknowable and
DeFries et al., Science 356,265–270 (2017) 21 April 2017 3of6
Table 2. Approaches to address ecosystem management as a wicked problem.
Approach Problem to address Examples of
implementation Obstacles
Multisector
decision-
making
Services from multifunc-
tional landscapes and
seascapes are not
factored into decisions
about single sectors
National-level spatial
planning (34); mul-
tilevel governance (35)
“Stovepiped”adminis-
trative structures
................................... ................................... .................................... ................................... .................................... ................................... .
Decision-
making across
administrative
boundaries
Ecological processes
transcend administrative
boundaries
River basin com-
missions (40);
large-scale corridor
planning (42,43)
Managers lack incen-
tives and authority
to consider other
jurisdictions
................................... ................................... .................................... ................................... .................................... ................................... .
Adaptive
management
Learn-by-doing when
outcomes of decisions are
uncertain because of com-
plex system dynamics
Ecosystem restora-
tion; fisheries
management (48)
Inflexible bureaucra-
cies; lack of monitoring
................................... ................................... .................................... ................................... .................................... ................................... .
Incorporating
natural capital
and ecosystem
services in
markets
Externalities are not
included in economic
accounting systems
Payments for
ecosystem services;
certification;
inclusive wealth
accounting (50)
Difficulty in deter-
mining value of
nonmarketed ecosys-
tem services
................................... ................................... .................................... ................................... .................................... ................................... .
Balancing
ideologies and
political reali-
ties of diverse
stakeholders
Politics and different
expectations of ecosystem
management lead to log-
jams in decision-making
Collaborative plan-
ning (67)
Differences in ideology
and values; political
realities
................................... ................................... .................................... ................................... .................................... ................................... .
on April 20, 2017http://science.sciencemag.org/Downloaded from

predictions have limited reliability. Key features
of adaptive management are monitoring, re-
assessing initial plans, redefining goals on the
basis of new evidence, social learning, and col-
laborations (44–46). Planning needs to be geared
toward flexible decision-making, with nimble
management structures that are capable of swift
changes (47). Examples include engagement of
multiple stakeholders to determine water alloca-
tions in the congressionally mandated Florida
Everglades ecosystem restoration program and
management of salmon in the Columbia River
basin in the northwestern United States (48).
Adaptive management to address wicked prob-
lems is intuitively appealing, but scientific and
institutional barriers hamper implementation.
Entrenched bureaucracies and social and legal
limits of authorities constrain opportunities for
flexible decision-making and the ability to change
course once a policy is put into place. Ideally, ac-
tive adaptive management would be conducted
through controlled experiments to identify causes
and effects among policies and outcomes, but such
experiments are costly and time-consuming. Mo-
nitoring systems, which are essential for adaptive
management, have suffered from lack of funding
and leadership (48,49).
Incorporating natural capital and
ecosystem services in markets
Economic systems reward short-term production
and consumption of natural resources. Changes
in nonmarketed ecosystem services (such as water-
shed protection) and natural capital (such as stocks
of minerals, energy sources, and forests) are ex-
ternalities that are not factored into traditional
economic accounting systems. Consequently, mar-
kets do not provide incentives to value ecosystem
services and natural capital (50).
Approaches to correct these market failures
target different decision-makers. At a national
level, tax policies and environmental regulations
provide incentives or penalties to corporations
and other natural resource users. Some countries
have adopted programs to incentivize households
and landowners to value natural capital. For ex-
ample, China’s reforestation program promotes
conversion from croplands and barren lands to
forests and grasslands. Costa Rica’s national pay-
ment scheme for ecosystem services has reversed
deforestation trends (50,51). The inclusive wealth
index, although yet to be used in standard prac-
tice, incorporates the value of natural capital into
national accounting systems to complement stan-
dard accounting systems that disregard natural
capital (52).
At a regional level, payment for ecosystem ser-
vices by the beneficiaries incentivizes the providers’
use of natural resources. Such payments are most
commonlyappliedinthecaseofupstreamwater-
shed protection to provide clean water to down-
stream users (e.g., New York City) (29). At an
individual consumer level, product certifications
and labels allow consumers to identify products that
are produced in ways that conform to guidelines
aimed at protecting naturalcapital and ecosystem
services (53). Many corporations have incorporated
the value of nature into their supply chains—for
example, with no-deforestation pledges (54).
Market mechanisms have been criticized for
exacerbation of inequities, particularly in places
with unequal power relations and coercive redis-
tribution of property rights (55). In addition, pro-
cesses to assess effectiveness of these mechanisms
and ensure that negative externalities are not dis-
placed to other locations need to be consistently
applied. Although a potentially powerful approach,
the appropriateness of market mechanisms de-
pends on the local socioeconomic and govern-
ance context.
Balancing ideological differences
among stakeholders
The challenge of understanding the perspectives
of diverse stakeholders contributes to the wick-
edness of ecosystem management. Ecosystem man-
agement decisions that may seem to be a simple
matter of setting scientific limits on resource use
frequently fail because of the political process of
decision-making, differing values and norms, and
power imbalances.
The history of protected areas to conserve
nature, for example, is fraught with differences
in ideologies and values among diverse stake-
holders, including conservationists, extractive
industries, and local communities (56). Conserva-
tionists historically have aimed to set aside lands
and waters from human use, which caused con-
flicts with industries aiming to extract resources
and with communities dependent on local re-
sources for livelihoods. Managers have attempted
to reconcile these goals through approaches such
as mixed-use management, integrated conserva-
tion and development projects, community-based
management, and eco-development. Such efforts
have yet to resolve this wicked problem, although
dialogs that recognize divergent aims are increas-
ing. For example, continual engagement among
policy-makers, communities, and researchers in
East Africa aims to balance pastoral livelihoods
with wildlife conservation (57).
A way forward
No two wicked problems are alike, and the above
approaches apply differently depending on the
problem and the ecological and socioeconomic
contexts. Efforts to control infectious disease and
non–point source pollution call for adaptive man-
agement because the population dynamics and
complexities of human behavior are inherently
DeFries et al., Science 356,265–270 (2017) 21 April 2017 4of6
Fig. 1. Decision flowchart for wicked problems in ecosystem management. Such an approach can help to avoid trap A (falsely applying a technical,
tame solution to a wicked problem) or trap B (inaction from overwhelming complexity).
GRAPHIC: ADAPTED BY K. SUTLIFF/SCIENCE
“Explicit recognition of the
underlying reasons that a
wicked problem occurs...could
help identify incremental
interventions that avoid
either oversimplifying a
problem or inaction from
overwhelming complexity.”
ECOSYSTEM EARTH
on April 20, 2017http://science.sciencemag.org/Downloaded from

nonlinear. A multijurisdiction planning approach
is needed for conservation of migrating species.
Each problem requires careful analysis to identify
the main sources of wickedness and the approaches
that might be appropriate for incremental solutions.
From a governance perspective, policy-makers
can approach these problems in ecosystem man-
agement through mechanisms that promote insti-
tutional “fit”or “interplay.”Institutional fit aligns
institutions to spatial, temporal, and functional
scales of different parts of an integrated system,
such as river basin commissions that encompass
multiple jurisdictions to cover a watershed (58).
Institutional fit is challenged by the reality that
an appropriate spatial fit for one problem, such
as water, may lead to spatial misfits with other
associated problems, such as wildlife movements
(59). Conversely, policies can promote institutional
interplay to include multiple institutions, each
with their own legacies and cultures of opera-
tion, and accept the associated messiness (60).
Institutional interplay, such as national-level land-
use planning with multiple sectors, requires policies
that shift institutional interactions biased toward
dominance, separation, and merger toward nego-
tiated agreements and systemic change (61). Both
institutional fit and interplay are likely to be messy.
Decisionsaboutwhichapproachismoreeffective
and tractable depend on the context.
There is no single or best solution to wicked
problems in ecosystem management. Two types
of traps can curtail incremental, partial improve-
ments. First, there is a risk of oversimplifying a
problem and assuming that a technical solution
will fix a wicked problem (trap A). For example,
approaches to provide food security run this risk.
In the 20th century, fertilizer, irrigation, and plant
breeding vastly increased the amount of food pro-
duction, reduced the cost of food, and alleviated
famine. However, the explosive increase in produc-
tionalsohadunintendedconsequences,including
underpinning the current rise in obesity, reducing
the nutritional content of cereals (62), creating in-
equities in access to food, and causing environmen-
tal problems such as soil degradation, fertilizer
runoff, and greenhouse gas emissions. A continued
solefocusontechnicalsolutionstoproducemore
food would overlook the myriad socia l, economic,
and environmental dimensions of the problem.
Conversely, there is a risk of making a problem
overly complex (trap B). Managers trained in
technical problem-solving can be ill-equipped to
confront complex social processes. The result can
be inaction from the inability to identify an in-
cremental, partial solution. The long-standing
problem of balancing needs of local people with
conservation in and around protected areas risks
falling into a trap of inaction (56).
A middle ground between the two traps can be
foundbyusingtheprinciplesofadaptiveman-
agement (Fig. 1). To complement an adaptive
approach, analysis of a problem in ecosystem
management, whether it is tame or wicked, and
the primary reasons for the wickedness (if ap-
plicable) can help identify an initial, tractable in-
stitutional or technical intervention, whose outcome
can be readily tracked. Analysis that considers di-
vergent stakeholders and possible unintended con-
sequences from the outset can help to avoid trap
A. An initial experimental intervention, including
monitoring and reassessment, helps to avoid trap
Bandfindsolutionsappropriateforthecontext
(63). Successive interventions are based on a more
complete understanding of the problem, reasons
for wickedness, and realistic institutional possi-
bilities for interventions.
Efforts to protect habitat for species whose
ranges transcend boundaries of protected areas,
such as the decades-long Yellowstone to Yukon
(Y2Y) Conservation Initiative in North America,
are one illustration of how this process could work
(Fig. 2). Y2Y was founded in the early 1990s to
promote conservation of large carnivores over an
area of 120 million hectares, encompassing three
U.S. states, four Canadian provinces, and many
jurisdictions of Canadian and U.S. land manage-
ment agencies, Canadian First Nations territories,
and U.S. Native American tribal lands. The con-
temporary history of the area includes agricul-
ture and resource extraction. Initially, Y2Y was
a loose network of conservation organizations and
individuals that advocated land-use change con-
ducive to large carnivore conservation through a
science-driven plan. The initial effort encountered
difficulties in implementing this vision because
of conflicts over property rights and divergent
values from those of local communities and indus-
try. Y2Y then reshaped its vision to harmonize
“the needs of people with those of nature”and
adopted governance structures to enhance parti-
cipation by local partners (64). The consortium has
enabled wildlife crossings and protected status
for land throughout the corridor (65). These govern-
ance strategies could evolve further, with the
role of local partners moving from participation
toward coproduction of governance strategies.
Explicit recognition of the underlying reasons
that a wicked problem occurs in a particular geo-
graphic, institutional, and cultural setting could
help identify incremental interventions that avoid
either oversimplifying a problem or inaction from
overwhelming complexity. Analyses of empirical
case studies, identification of possible incremen-
tal solutions, and context-appropriate, tractable
metrics to assess progress are all needed to ad-
dress understudied wicked problems in ecosystem
management.
REFERENCES AND NOTES
1. M. R. Kaufmann et al., “An ecological basis for ecosystem
management”(Rocky Mountain Forest and Range Experiment
Station Research Paper, U.S. Department of Agriculture, 1994).
2. F. Berkes, J. Colding, C. Folke, Ecol. Appl. 10,1251–1262 (2000).
3. M. Tengö, E. S. Brondizio, T. Elmqvist, P. Malmer,
M. Spierenburg, Ambio 43, 579–591 (2014).
4. G. J. Leigh, The World's Greatest Fix: A History of Nitrogen and
Agriculture (Oxford Univ. Press, 2004).
5. A. A. Ormsby, S. A. Bhagwat, Environ. Conserv. 37,320–326 (2010).
6. W. B. Kessler, H. Salwasser, C. W. Cartwright Jr., J. A. Caplan,
Ecol. Appl. 2, 221–225 (1992).
7. D.Ludwig,R.Hilborn,C.Walters,Ecol. Appl. 3,548–549 (1993).
8. A. Sáenz‐Arroyo, C. M. Roberts, Fish Fish. 9, 316–327 (2008).
9. C. S. Holling, G. K. Meffe, Conserv. Biol. 10, 328–337 (1996).
10. K. Robbins, in The Laws of Nature: Reflections on the Evolution
of Ecosystem Management Law and Policy, K. Robbins, Ed.
(Univ. of Akron Press, 2012), pp. 1–24.
11. P. S. White, S. Pickett, in The Ecology of Natural Disturbance and
Patch Dynamics, S. Pickett, P. S. White, Eds. (Academic Press,
1985), pp. 3–13.
12. S. Levin et al., Environ. Dev. Econ. 18, 111–132 (2013).
13. Millennium Ecosystem Assessment, Ecosystems and Human
Well-Being: Synthesis (Island Press, 2005).
DeFries et al., Science 356,265–270 (2017) 21 April 2017 5of6
Reevaluate
goals and
strategies
Monitor
outcome
at systemic
level
Approximation
of reasons for
wickedness
Approximation of
tractable, incremental
approach toward
solution
Wicked
How to provide
habitat for large
carnivores with
ranges outside
PAs: Is problem
tame or wicked?
Mismatch in ecological and
administration boundaries
covering mutiple jurisdictions;
multiple sectors (mines, timber,
local communities) use corridors;
divergent values of stakeholders
(conserve vs. extract resources)
Lack of policy
change; lack of
engagement from
local communities
and extractive sector
Collaboratively
reframe vision
and governance
to incorporate
local priorities
Science-based
conservation plan
Fig. 2. Flowchart for managing an example wicked problem. This example illustrates the wicked problem of providing habitat for large carnivores
outside protected areas (PAs) through efforts such as the Yellowstone to Yukon (Y2Y) Conservation Initiative (42).
GRAPHIC: ADAPTED BY K. SUTLIFF/SCIENCE
on April 20, 2017http://science.sciencemag.org/Downloaded from

14. H. Heesterbeek et al., Science 347, aaa4339 (2015).
15. M. L. Pace, J. A. Gephart, Ecosystems 20,44–53 (2016).
16. A. R. Rissman, S. R. Carpenter, Daedalus 144,35–47 (2015).
17. F. Berkes, Fish Fish. 13, 465–476 (2012).
18. D. Ludwig, Ecosystems 4, 758–764 (2001).
19. H. Rittel, M. M. Webber, Polity 4, 155–169 (1973).
20. B. W. Head, Public Policy 3, 101 (2008).
21. R. A. Heifetz, Leadership Without Easy Answers, vol. 465
(Harvard Univ. Press, 1994).
22. M. Roser, E. Ortiz-Espina, “Global extreme poverty”(Our World
in Data, 2017); https://ourworldindata.org/extreme-poverty/.
23. J. Liu et al., Ecol. Soc. 18, 26 (2013).
24. E. Ostrom,Governing the C ommons (Cambridge Univ. Press, 2015).
25. L. Porter-Bolland et al., For. Ecol. Manage. 268,6–17 (2012).
26. K. K. Davies, K. T. Fisher, M. E. Dickson, S. F. Thrush,
R. Le Heron, Ecol. Soc. 20, 37 (2015).
27. R. E. Grumbine, Conserv. Biol. 8,27–38 (1994).
28. J. Sayer et al., Proc.Natl.Acad.Sci.U.S.A.110,8349–8356 (2013).
29. S. L. Postel, B. H. Thompson, in Natural Resources Forum,
vol. 29 (Wiley Online Library, 2005), pp. 98–108.
30. R. Nash, Am. Q. 22, 726–735 (1970).
31. R. DeFries, J. Foley, G. P. Asner, Front. Ecol. Environ 2,
249–257 (2004).
32. B. Reyers, P. J. O’Farrell, J. L. Nel, K. Wilson, Landsc. Ecol. 27,
1121–1134 (2012).
33. G. Rodríguez-Loinaz, J. G. Alday, M. Onaindia, J. Environ.
Manage. 147, 152–163 (2015).
34. M. E. Burton et al., Conserv. Lett. 10.1111/conl.12265 (2016).
35. H. Nagendra, E. Ostrom, Int. J. Commons 6, 104–133 (2012).
36. S. Koplitz et al., Environ. Res. Lett. 11, 094023 (2016).
37. F. Craighead, Track of the Grizzly (Sierra Club Books, 1979).
38. T. Dutta, S. Sharma, B. McRae, P. S. Roy, R. DeFries,
Reg. Environ. Change 16,53–67 (2015).
39. M. D. Blum, H. H. Roberts, Nat. Geosci. 2, 488–491 (2009).
40. J. Dore, L. Lebel, Environ. Manage. 46,60–80 (2010).
41. D. Heilmann, J. Curr. Southeast Asian Aff. 34,95–121 (2015).
42. C. C. Chester, Environ. Sci. Policy 49,75–84 (2015).
43. M. B. Holland, in Climate and Conservation, J. A. Hilty,
C. C. Chester, M. S. Cross, Eds. (Island Press, 2012), pp. 56–66.
44. D. R. Armitage et al., Front. Ecol. Environ 7,95–102 (2009).
45. C. S. Holling, Adaptive Environmental Assessment and
Management (John Wiley and Sons, 1978).
46. National Research Council, Adaptive Management for Water
Resources Project Planning (National Research Council, 2004).
47. D. E. Schindler, R. Hilborn, Science 347, 953–954 (2015).
48. P. J. Balint, R. E. Stewart, A. Desai, Wicked Environmental
Problems: Managing Uncertaintyand Conflict (Island Press, 2011).
49. C. J. Walters, Ambio 36, 304–307 (2007).
50. A. D. Guerry et al., Proc. Natl. Acad. Sci. U.S.A. 112, 7348–7355
(2015).
51. Z. Ouyang et al., Science 352, 1455–1459 (2016).
52. University of the United Nations–International Human Dimensions
Programme on Global Environmental Change, UN Environment
Programme, Inclusive Wealth Report 2014: Measuring Progress
toward Sustainability (Cambridge Univ. Press, 2014).
53. J. C. Milder et al., Conserv. Biol. 29, 309–320 (2015).
54. P . M. Kareiva, B.W. McNally, S. McCormick, T. Miller, M. Ruck elshaus,
Proc. Natl. Acad. Sci. U.S.A. 112,7375–7382 (2015).
55. E. Gómez-Baggethun, R. Muradian, Ecol. Econ. 117,217–224 (2015).
56. R. DeFries, in Science, Conservation, and National Parks,
S. Bessinger, D. Ackerly, H. Doremus, G. Machlis, Eds. (Univ. of
Chicago Press, 2017), pp. 227–246.
57. R. S. Reid et al., Proc. Natl. Acad. Sci. U.S.A. 113, 4579–4584
(2016).
58. G. Epstein et al., Curr. Opin. Environ. Sustain. 14,34–40 (2015).
59. T. Moss, in How Institutions Change (Springer, 2003), pp. 85–121.
60. E. S. Brondizio, E. Ostrom, O. R. Young, Annu. Rev. Environ.
Resour. 34, 253–278 (2009).
61. O. Young, Ecol. Soc. 11, 27 (2006).
62. R. DeFries et al., Science 349, 238–240 (2015).
63. E. Ostrom, M. A. Janssen, J. M. Anderies, Proc. Natl. Acad. Sci.
U.S.A. 104, 15176–15178 (2007).
64. C. Wyborn, Glob. Environ. Change 30,56–67 (2015).
65. Yellowstone to Yukon Conservation Initiative, The Yellowstone
to Yukon Vision: Progress & Possibility (Yellowstone to Yukon
Conservation Initiative, 2014); https://y2y.net/publications/
y2y_vision_20_years_of_progress.pdf.
66. D. E. Calkin, J. D. Cohen, M. A. Finney, M. P. Thompson, Proc.
Natl. Acad. Sci. U.S.A. 111, 746–751 (2014).
67. A. H. Toomey, A. T. Knight, J. Barlow, Conserv. Lett. 10.1111/
conl.12315 (2016).
10.1126/science.aal1950
DeFries et al., Science 356,265–270 (2017) 21 April 2017 6of6
ECOSYSTEM EARTH
on April 20, 2017http://science.sciencemag.org/Downloaded from

(6335), 265-270. [doi: 10.1126/science.aal1950]356Science
Ruth DeFries and Harini Nagendra (April 20, 2017)
Ecosystem management as a wicked problem
Editor's Summary
This copy is for your personal, non-commercial use only.
Article Tools
http://science.sciencemag.org/content/356/6335/265
article tools:
Visit the online version of this article to access the personalization and
Permissions http://www.sciencemag.org/about/permissions.dtl
Obtain information about reproducing this article:
is a registered trademark of AAAS. ScienceAdvancement of Science; all rights reserved. The title
Avenue NW, Washington, DC 20005. Copyright 2016 by the American Association for the
in December, by the American Association for the Advancement of Science, 1200 New York
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last weekScience
on April 20, 2017http://science.sciencemag.org/Downloaded from