12 Oct 2004 11:54 AR AR229-ES35-20.tex AR229-ES35-20.sgm LaTeX2e(2002/01/18) P1: GJB
10.1146/annurev.ecolsys.35.021103.105711
Annu. Rev. Ecol. Evol. Syst. 2004. 35:557–81
doi: 10.1146/annurev.ecolsys.35.021103.105711
Copyright
c
2004 by Annual Reviews. All rights reserved
First published online as a Review in Advance on August 5, 2004
REGIME SHIFTS,RESILIENCE, AND BIODIVERSITY
IN
ECOSYSTEM
MANAGEMENT
Carl Folke,
1,2
Steve Carpenter,
2,3
Brian Walker,
4
Marten Scheffer,
5
Thomas Elmqvist,
1
Lance Gunderson,
6
and C.S. Holling
7
1
Department of Systems Ecology, Stockholm University, SE-106 91 Stockholm,
Sweden; email: calle@system.ecology.su.se; thomase@ecology.su.se
2
Beijer International Institute of Ecological Economics, Royal Swedish Academy
of Sciences, Stockholm, Sweden
3
Center for Limnology, University of Wisconsin, Madison, Wisconsin 53706;
email: srcarpen@wisc.edu
4
Sustainable Ecosystems, CSIRO, Canberra, ACT, 2601, Australia;
email: Brian.Walker@csiro.au
5
Aquatic Ecology and Water Quality Management Group, Wageningen Agricultural
University, Wageningen, The Netherlands; email: Marten.Scheffer@wur.nl
6
Department of Environmental Studies, Emory University,
Atlanta, Georgia 30322; email: lgunder@emory.edu
7
16871 Sturgis Circle, Cedar Key, Florida 32625; email: holling@zoo.ufl.edu
KeyWords alternate states, regime shifts, response diversity, complex adaptive
systems, ecosystem services
■ Abstract We review the evidence of regime shifts in terrestrial and aquatic envi-
ronments in relation to resilience of complex adaptive ecosystems and the functional
roles of biological diversity in this context. The evidence reveals that the likelihood
of regime shifts may increase when humans reduce resilience by such actions as re-
moving response diversity, removing whole functional groups of species, or removing
whole trophic levels; impacting on ecosystems via emissions of waste and pollutants
and climate change; and altering the magnitude, frequency, and duration of distur-
bance regimes. The combined and often synergistic effects of those pressures can
make ecosystems more vulnerable to changes that previously could be absorbed. As
a consequence, ecosystems may suddenly shift from desired to less desired states in
their capacity to generate ecosystem services. Active adaptive management and gover-
nance of resilience will be required to sustain desired ecosystem states and transform
degraded ecosystems into fundamentally new and more desirable configurations.
1543-592X/04/1215-0557$14.00 557
12 Oct 2004 11:54 AR AR229-ES35-20.tex AR229-ES35-20.sgm LaTeX2e(2002/01/18) P1: GJB
558 FOLKE ET AL.
INTRODUCTION
Humanity strongly influences biogeochemical, hydrological, and ecological pro-
cesses, from local to global scales. We currently face more variable environments
with greater uncertainty about how ecosystems will respond to inevitable increases
in levels of human use (Steffen et al. 2004). At the same time, we seem to chal-
lenge the capacity of desired ecosystem states to cope with events and disturbances
(Jackson et al. 2001, Paine et al. 1998). The combination of these two trends calls
for a change from the existing paradigm of command-and-control for stabilized
“optimal” production to one based on managing resilience in uncertain environ-
ments to secure essential ecosystem services (Holling & Meffe 1996, Ludwig et al.
2001). The old way of thinking implicitly assumes a stable and infinitely resilient
environment, a global steady state. The new perspective recognizes that resilience
can be and has been eroded and that the self-repairing capacity of ecosystems
should no longer be taken for granted (Folke 2003, Gunderson 2000). The chal-
lenge in this new situation is to actively strengthen the capacity of ecosystems to
support social and economic development. It implies trying to sustain desirable
pathways and ecosystem states in the face of continuous change (Folke et al. 2002,
Gunderson & Holling 2002).
Holling (1973), in his seminal paper, defined ecosystem resilience as the mag-
nitude of disturbance that a system can experience before it shifts into a different
state (stability domain) with different controls on structure and function and distin-
guished ecosystem resilience from engineering resilience. Engineering resilience
is a measure of the rate at which a system approaches steady state after a pertur-
bation, that is, the speed of return to equilibrium, which is also measured as the
inverse of return time. Holling (1996) pointed out that engineering resilience is
a less appropriate measure in ecosystems that have multiple stable states or are
driven toward multiple stable states by human activities (Nystr¨om et al. 2000,
Scheffer et al. 2001).
Here, we define resilience as the capacity of a system to absorb disturbance and
reorganize while undergoing change so as to retain essentially the same function,
structure, identity, and feedbacks (Walker et al. 2004). The ability for reorganiza-
tion and renewal of a desired ecosystem state after disturbance and change will
strongly depend on the influences from states and dynamics at scales above and
below (Peterson et al. 1998). Such cross-scale aspects of resilience are captured in
the notion of a panarchy, a set of dynamic systems nested across scales (Gunderson
& Holling 2002). Hence, resilience reflects the degree to which a complex adaptive
system is capable of self-organization (versus lack of organization or organization
forced by external factors) and the degree to which the system can build and in-
crease the capacity for learning and adaptation (Carpenter et al. 2001b, Levin 1999).
Several studies have illustrated that ecological systems and the services that
they generate can be transformed by human action into less productive or other-
wise less desired states. The existence of such regime shifts (or phase shifts) is
an area of active research. Regime shifts imply shifts in ecosystem services and
12 Oct 2004 11:54 AR AR229-ES35-20.tex AR229-ES35-20.sgm LaTeX2e(2002/01/18) P1: GJB
REGIME SHIFTS IN ECOSYSTEMS 559
consequent impacts on human societies. The theoretical basis for regime shifts has
been described by Beisner et al. (2003), Carpenter (2003), Ludwig et al. (1997),
Scheffer & Carpenter (2003), and Scheffer et al. (2001).
Here, we review the evidence of regime shifts in terrestrial and aquatic ecosys-
tems in relation to resilience and discuss its implications for the generation of
ecosystem services and societal development. Regime shifts in ecosystems are
increasingly common as a consequence of human activities that erode resilience,
for example, through resource exploitation, pollution, land-use change, possible
climatic impact and altered disturbance regimes. We also review the functional
role of biological diversity in relation to regime shifts and ecosystem resilience. In
particular, we focus on the role of biodiversity in the renewal and reorganization
of ecosystems after disturbance—what has been referred to as the back-loop of the
adaptive cycle of ecosystem development (Holling 1986). In this context, the in-
surance value of biodiversity becomes significant. It helps sustain desired states of
dynamic ecosystem regimes in the face of uncertainty and surprise (Elmqvist et al.
2003). Strategies for transforming degraded ecosystems into new and improved
configurations are also discussed.
REGIME SHIFTS AND DYNAMICS OF RESILIENCE
IN ECOSYSTEMS
Ecosystems are complex, adaptive systems that are characterized by historical de-
pendency, nonlinear dynamics, threshold effects, multiple basins of attraction, and
limited predictability (Levin 1999). Increasing evidence suggests that ecosystems
often do not respond to gradual change in a smooth way (Gunderson & Pritchard
2002). Threshold effects with regime shifts from one basin of attraction to another
have been documented for a range of ecosystems (see Thresholds Database on the
Web site www.resalliance.org). Passing a threshold marks a sudden change in feed-
backs in the ecosystem, such that the trajectory of the system changes direction—
toward a different attractor. In some cases, crossing the threshold brings about a
sudden, sharp, and dramatic change in the responding state variables, for example,
the shift from clear to turbid water in lake systems (Carpenter 2003). In other cases,
although the dynamics of the system have “flipped” from one attractor to another,
the transition in the state variables is more gradual, such as the change from a
grassy to a shrub dominated rangeland (Walker & Meyers 2004). In Table 1, we
provide examples of documented shifts between alternate states and expand on
some of them in the text.
Tem per ate Lakes
Lake phosphorus cycles exhibit multiple regimes, each stabilized by a distinctive
set of feedbacks. Generally, two regimes have attracted the most interest, although
deeper analyses have revealed even greater dynamic complexities (Carpenter 2003,
12 Oct 2004 11:54 AR AR229-ES35-20.tex AR229-ES35-20.sgm LaTeX2e(2002/01/18) P1: GJB
560 FOLKE ET AL.
TABLE 1 Documented shifts between states in different kinds of ecosystems
Ecosystem type Alternate state 1 Alternate state 2 References
Freshwater systems
Temperate lakes Clear water Turbid eutrophied water Carpenter 2003
Game fish abundant Game fish absent Post et al. 2002, Walters &
Kitchell 2001, Carpenter 2003
Tropical lakes Submerged vegetation Floating plants Scheffer et al. 2003
Shallow lakes Benthic vegetation Blue-green algae Blindow et al. 1993, Scheffer
et al. 1993, Scheffer 1997,
Jackson 2003
Wetlands Sawgrass communities Cattail communities Davis 1989, Gunderson 2001
Salt marsh vegetation Saline soils Srivastava & Jefferies 1995
Marine systems
Coral reefs Hard coral dominance Macroalgae dominance Knowlton 1992, Done 1992,
Hughes 1994, McCook 1999
Hard coral dominance Sea urchin barren Glynn 1988, Eakin 1996
Kelp forests Kelp dominance Sea urchin dominance Steneck et al. 2002, Konar & Estes 2003
Sea urchin dominance Crab dominance Steneck et al. 2002
Shallow lagoons Seagrass beds Phytoplankton blooms Gunderson 2001, Newman et al. 1998
Coastal seas Submerged vegetation Filamentous algae Jansson & Jansson 2002, Worm et al. 1999
Benthic foodwebs Rock lobster predation Whelk predation Barkai & McQuaid 1988
Ocean foodwebs Fish stock abundant Fish stock depleted Steele 1998, Walters & Kitchell 2001,
de Roos & Persson 2002
12 Oct 2004 11:54 AR AR229-ES35-20.tex AR229-ES35-20.sgm LaTeX2e(2002/01/18) P1: GJB
REGIME SHIFTS IN ECOSYSTEMS 561
Forest systems
Temperate forests Spruce-fir dominance Aspen-birch dominance Holling 1978
Pine dominance Hardwood dominance Peterson 2002
Hardwood-hemlock Aspen-birch Frelich & Reich 1999
Birch-spruce succession Pine dominance Danell et al. 2003
Tropical forests Rain forest Grassland Trenbath et al. 1989
Woodland Grassland Dublin et al. 1990
Native crab consumers Invasive ants O’Dowd et al. 2003
Savanna and grassland
Grassland Perennial grasses Desert Wang & Eltahir 2000, Foley et al. 2003
vandeKoppel et al. 1997
Savanna Native vegetation Invasive species Vitousek et al. 1987
Tall shrub, perennial grasses Low shrub, bare soil Bisigato & Bertiller 1997
Grass dominated Shrub dominated Anderies et al. 2003, Brown et al. 1997
Arctic, sub-Arctic systems
Steppe/tundra Grass dominated Moss dominated Zimov et al. 1995
Tundra Boreal forest Bonan et al. 1992, Higgins et al. 2002
