Bookmark and Share

Regime Shifts in Social-Ecological Systems

Vânia Proença and Juan Fernández-Manjarrés


Human-driven ecosystem changes can lead to “regime shifts”—relatively rapid, sometimes radical disruptions—that pose high risks to human well-being.

Regime shifts:

  • occur due to the loss of ecosystem resilience and/or due to intense shocks,

  • cause the decline of biodiversity and of ecosystem provisioning and regulating functions,

  • are difficult to anticipate with precision, and slow and expensive to reverse, and

  • may be prevented through the enhancement of ecosystem resilience and the mitigation of pressures.

July 2015

What are regime shifts?

Ecosystems consist of components that interact and self-organize through internal feedback mechanisms that maintain the system’s structure and function. Regime shifts occur when systems experience changes in their internal dynamics that push the system into a new state. 1,2

For instance, natural periodic fires help to keep grasslands structure; if fire is suppressed grasslands may shift to encroached shrubland. 3 In humid forests tree evapotranspiration is an important source of humidity for forest persistence. The loss of large tracts of forest may lead to declines in evapotranspiration, which will affect regional rainfall and cause further tree loss, pushing the system into a drier condition. 4

How do regime shifts occur?

Regime shifts occur by means of two main mechanisms that can work together: 2,4

  • cumulative changes in ecosystem properties (e.g., vegetation cover, concentration of pollutants) that lead to a gradual loss of resilience and

  • an intense shock (e.g., hurricane, oil spill) that drastically changes the system state.

Resilience is the capacity of a system to absorb changes and disturbances and self-organize to keep the same structure and function.3, 5 As systems lose resilience they also become more vulnerable to changes (figure 1), hence, less disturbance is needed to push the system into a new regime.

Figure 1

Figure 1. a) Mechanisms driving regime shifts can be depicted using “stability landscapes” 4, 11: valleys represent stability domains, or basins of attraction, in which a system, represented by the ball, is kept by internal feedback mechanisms; b) cumulative changes in system variables can lead to a gradual loss of resilience, here represented by changes in the depth of the valley that becomes shallower up to a point where even a small disturbance can push the system into a new basin of attraction, under a different regime; c) an intense shock can also push the system into a new regime.

When a system shifts into a new regime, it reaches and is kept in a new state by internal feedback dynamics characteristic of that regime (figure 1). This makes the recovery to the previous regime very difficult, especially when lag effects in the system’s response hinder its recovery (i.e., hysteresis). 6,7

In some cases regime shifts are mediated by tipping points 4,6, that is, a non-linear evolution of the system, where an additional small change causes the passing of a threshold and leads to an abrupt change (relatively to the baseline system dynamics). These tipping-point dynamics result from reinforcing feedbacks that amplify the impacts of the drivers of change.

Can we predict regime shifts?

Regime shifts can be anticipated but it is very difficult to define the moment and the spatial scale in which they will occur because the thresholds between regimes are very difficult to define. 8 Therefore, regime shifts often occur unexpectedly. The identification of early warning indicators could help to anticipate regime shifts more precisely, but further research is needed. 9 Early-warning indicators can be detected through statistical analysis of time-series of ecosystem properties. For instance, an increasing trend in the variance of within-year population size of a key species or functional group may suggest a decrease in ecosystem stability and an impending regime shift. 9

The identification of regime shifts and their mechanisms is subject to uncertainty. 4 Not all changes in a system will end in a regime shift. Some changes occur due to the natural fluctuation of the state of the system within a certain regime. 2 For instance, the damage caused by a windthrow in a forest may be perceived as permanent if viewed in a narrow time-frame, but the system may eventually recover to its mature state through secondary forest succession.

How much do we know about regime shifts?

There is ample evidence on past and on-going occurrence of regime shifts. 2,8 In some cases the underlying mechanisms are well described, while in others there is general agreement on the existence of a regime shift even if the mechanisms remain unclear. Furthermore, there are situations in which regime shifts maybe unfolding but evidence is still scarce to support their existence. Examples of regime shifts:

  • Soil salinization. Soils in dryland areas shift from a productive system to a desertified system due to the removal and replacement of natural vegetation and to artificial irrigation that cause the accumulation of salts in the soil 10; the Fertile Crescent in former Mesopotamia is an historical example where agricultural irrigation drove soil salinization. Today, soil salinization is a serious problem in many dryland regions.

  • Freshwater eutrophication. Aquatic systems shift from a clear water (i.e., oligotrophic) state to an eutrophic and low oxygen state due to excessive nutrient inputs (e.g., phosphates in sewage), leading to the death of freshwater organisms sensitive to low oxygen levels and the rise of tolerant ones, such as toxic cyanobacteria.8,11,12 The degree of reversibility varies among cases, from rapid recovery to long-term eutrophication (despite depollution efforts). The mechanisms of recovery are still not fully understood.

  • Amazon forest dieback. The Amazon forest may shift to a savanna-type regime due to interactions between deforestation, the use of fire and climate change.4,8,13 The loss of forest tracts may lead to more frequent and severe droughts, resulting in a feedback loop of forest loss and increased dryness. This feedback mechanism could be further aggravated by climate warming and reduced precipitation caused by global climate change. Model projections suggest that these feedbacks could scale to the regional level if thresholds of deforestation and climate change are met. However, there is uncertainty regarding the value of thresholds and on the ability of forests to tolerate fire and drought.

  • Tropical coral reef bleaching and decalcification. An increase in sea temperature of 2-3ºC above corals optimal temperature, caused by climate change, may lead to coral bleaching (i.e., the expulsion of algae that live inside, “color”, and deliver food - organic compounds - to corals).4,8,14,15 Although corals may survive a bleaching event, they become more vulnerable to additional stressors, such as fishing or pollution, which may cause their death. Increased vulnerability may also be driven by rising concentrations of atmospheric CO2 that lead to ocean acidification and to shortage of carbonate ions dissolved in water, which are essential to build coral skeletons. Local episodes of regime shifts from coral to algae dominated habitats have been observed in the last decades.

How fast and large are regime shifts?

The temporal and spatial dynamics of regime shifts depend on the main processes causing feedbacks that affect variables of the system:

  • Fast variables are generally those of concern to ecosystem users, like crop production, water purity and abundance of harvested species.

  • Slow variables, typically control (but not always) the rates of the processes related to fast variables, and include for example organic content in soils that affect crop productivity and rainfall variation during the growing season.16 The complexity of local to regional transitions is further increased by teleconnections (i.e., long-distance connections between meteorological phenomena), such as El Niño onset and milder and wetter winters in parts of North America.17 Besides, regime shifts can amplify the effects of different drivers by:

  • aggregating in space

  • spreading as contiguous domino effects

  • synergistic dynamics that amplify local and neighboring processes

One example of a potential regional scale regime shift from moist evergreen tropical forest to sparse dry seasonal vegetation, mediated by several interacting local regime shifts, was given recently for South America.4 In this region, deforestation and increased human fires on both sides of the core of the Amazon forest (Eastern Amazonian forests and Andean forests) can permanently change the climate of the Eastern Andean slopes, impeding many species to find a climatic refugia for ongoing climate change.

What are the impacts of regime shifts?

Because people and ecosystems are deeply linked through coupled dynamics, changes in ecosystem state will affect human well-being; examples of impacts include:

  • decline or loss of species due to the degradation of environmental conditions

  • disruption of ecosystem regulating functions, such as climate regulation, fire regulation, flood regulation, soil protection from erosion, coastal protection from storms, regulation of disease spread

  • decline in crop production due to losses in soil fertility, spread of invasive species or climatic changes

  • decline in fish production due to overexploitation, pollution and degradation of shelter, nesting and feeding environments for fishes

  • loss of water quality for drinking and recreation due to pollution and groundwater contamination

  • failure to discover new medicines (many species with medicinal value are still to be found)

  • loss of recreation areas due to ecosystem degradation

  • loss of areas or natural elements with cultural and spiritual value for local communities.

What characteristics make social-ecological systems more or less susceptible to regime shifts?

As stated above, social and ecological systems can co-evolve in the long term. Societies adapt to the services provided by the ecosystems and in turn affect the ecosystem structure and diversity through management decisions.18 Characteristics that enhance resilience of social-ecological systems include:

  • diversity of biological, ecological and social structures and processes;

  • maintenance of redundancy for key provisioning services, both from a social and ecological perspective, like spreading similar economic activities along ecological gradients to avoid ‘putting all the eggs in the same basket’; and

  • promoting network connectivity among actors to provide rich cultural and technological exchange of information regarding human-nature relationships.

However, the structure of the coupling between social and ecological systems can break down rapidly placing these coupled systems in very fragile states. Amongst others, the coupling between social and ecological systems can break down because of:

  • poor governance of resources,

  • political instability,

  • overexploitation of mineral resources (e.g. failed oil states),

  • excessive reliance on a narrow range of ecosystem goods, and

  • illegal harvest of protected species

One striking example is the poaching of species as a response to global markets demand. The scarcer the resource becomes, the higher the price paid, inciting local people in a spiral of illegal actions, supporting corruption and the unsustainable exploitation of resources, as is the case with illegal ivory trade and African elephant poaching.19 Corruption causes social degradation and impairs conservation efforts, thus perpetuating the causes of degradation and keeping locals in poverty traps.19 Conversely, global markets may also have a positive effect on social-ecological systems, for instance, mechanisms of fair trade may aid in providing local communities with stable income, increasing people resilience to environmental variability and enforcing the sustainable use of natural resources.

Is it possible to reverse regime shifts?

Sometimes it is possible to restore a degraded ecosystem into its former state, but it is often difficult, slow and expensive.4,7 Hence, regime shifts are of major policy relevance and justify a precautionary approach to reduce the causes of resilience loss but also a proactive approach to enhance systems resilience.

Today, the main drivers of environmental change are land-use change, climate change, pollution, overexploitation and invasive species. The threat posed by these drivers is aggravated by their cumulative effects but also by the possibility of interactions between the effects.4

How can we prevent regime shifts?

Preventing a pending regime shift requires addressing the various pressures impacting an ecosystem at the scales in which they act, global to local.

  • Concerted actions at the international level. Global scale drivers, such as climate change, require concerted actions at the international level, such as agreements to mitigate reduce greenhouse gases emissions, global mechanisms to control deforestation, or concerted policies to regulate trade chains that promote biodiversity loss.20

  • Control local pressures to increase the resilience to global scale drivers. Global scale dynamics are difficult to predict and manage and reaching international consensus is often challenging14,20, therefore, controlling local pressures, which are more amenable to management, is essential. For instance, coral reefs are under serious threat of climate change, their resilience to sea water warming and acidification can be greatly improved if additional stressors are controlled, namely coastal pollution, destructive fishing methods and overfishing.4,14

  • Safeguard ecosystems from local pressures. This could be achieved through the adoption of sustainable practices in agriculture, forestry and fisheries, the establishment of protected areas, and by raising people’s awareness about risks and opportunities for action. 8

  • Enhance ecosystems resilience through restoration. Ecological restoration, such as reforestation, contributes to reverse degradation (reducing the risk of a regime shift) but also to enhance ecological complexity (i.e., the network of interactions between species and their environment) and therefore, the ability to deal with uncertainty and buffer unexpected changes.3

  • Improve the adaptive capacity of social systems. Informed, flexible and inclusive decision-making is needed to react timely to social and ecological feedbacks. 21 This requires knowledge sharing, learning from past experiences, and the ability to move from business-as-usual unsustainable pathways to innovative forms of human-nature interaction.

  • Improve knowledge on regime shifts. This should be an overarching goal, including scientific research continued development of regime shift databases, technology development, and educational outreach.

Vânia Proença is a postdoctoral researcher at the University of Lisbon. Her research interests revolve around biodiversity patterns and processes, and the use of biodiversity modelling and indicators to assess biodiversity change and its effects on ecosystem functions and services. She was a co-editor of the Portugal Millennium Ecosystem Assessment report, and has participated in international initiatives on biodiversity and ecosystem services, such as IPBES and GEO BON.

Juan Fernández-Manjarrés is a researcher at the University of Paris Ecology, Systematis and Evolution CNRS UMR 8079 laboratory. His research interests are centered in developing concepts and tools to understand the adaptation of forest social–ecological systems to climate change. In particular, his current research involves the evaluation of assisted migration of forests as a public policy and ecological restoration tool.

Regime Shifts in Social-Ecological Systems

Regime Shifts Database:

Regime shifts in lake ecosystems:

Resilience Alliance:

Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES):

Intergovernmental Panel on Climate Change (IPCC):

Convention on Biological Diversity (CBD):

Global restoration network (volunteer):

Fairtrade International:

Regime Shifts Database (add data):

  1. Scheffer, M. & Carpenter, S. R. 2003 Catastrophic regime shifts in ecosystems: linking theory to observation. Trends Ecol. Evol. 18, 648–656.
  2. Rocha, J., Biggs, R. & Peterson, G. 2014 Regime shifts: what are they and why do they matter? Regime Shifts Database,
  3. Holling, C. S. 1973 Resilience and stability of ecological systems. Annu. Rev. Ecol. Syst. 4, 1–23.
  4. Leadley, P. et al. 2014 Interacting regional-scale regime shifts for biodiversity and ecosystem services. BioScience 64, 665–679.
  5. Carpenter, S. R. & Brock, W. A. 2008 Adaptive capacity and traps. Ecol. Soc. 13, 40.
  6. Lenton, T. M., Held, H., Kriegler, E., Hall, J. W., Lucht, W., Rahmstorf, S. & Schellnhuber, H. J. 2008 Tipping elements in the Earth’s climate system. Proc. Natl. Acad. Sci. 105, 1786–1793.
  7. Hughes, T. P., Carpenter, S., Rockström, J., Scheffer, M. & Walker, B. 2013 Multiscale regime shifts and planetary boundaries. Trends Ecol. Evol. 28, 389–395.
  8. Leadley, P., Pereira, H. M., Alkemade, R., Fernandez-Manjarres, J., Proenca, V., Scharlemann, J. P. W. & Walpole, M. 2010 Biodiversity Scenarios: Projections of 21st century change in biodiversity and associated ecosystem services. Montreal: Secretariat of the Convention on Biological Diversity.
  9. Biggs, R., Carpenter, S. R., & Brock, W. A. 2009 Turning back from the brink: detecting an impending regime shift in time to avert it. Proc. Natl. Acad. Sci. 106, 826-831.
  10. Schofield, R., Thomas, D. S. G. & Kirkby, M. J. 2001 Causal processes of soil salinization in Tunisia, Spain and Hungary. Land Degrad. Dev. 12, 163–181.
  11. Carpenter, S. R., Ludwig, D. & Brock, W. A. 1999 Management of eutrophication for lakes subject to potentially irreversible change. Ecol. Appl. 9, 751–771.
  12. Smith, V. H. & Schindler, D. W. 2009 Eutrophication science: where do we go from here? Trends Ecol. Evol. 24, 201–207.
  13. Davidson, E. A. et al. 2012 The Amazon basin in transition. Nature 481, 321–328.
  14. Hoegh-Guldberg, O. et al. 2007 Coral reefs under rapid climate change and ocean acidification. Science 318, 1737–1742.
  15. Pandolfi, J. M., Connolly, S. R., Marshall, D. J. & Cohen, A. L. 2011 Projecting coral reef futures under global warming and ocean acidification. Science 333, 418–422.
  16. Walker, B. H., Carpenter, S. R., Rockstrom, J., Crépin, A.-S. & Peterson, G. D. 2012 Drivers, “slow” variables, “fast” variables, shocks, and resilience. Ecol. Soc. 17, 30.
  17. Stevenson, S., Fox-Kemper, B., Jochum, M., Neale, R., Deser, C. & Meehl, G. 2012 Will there be a significant change to El Niño in the twenty-first century? J. Clim. 25, 2129–2145.
  18. Liu, J. et al. 2007 Complexity of coupled human and natural systems. Science 317, 1513–1516.
  19. Bennett, E. L. 2015. Legal ivory trade in a corrupt world and its impact on African elephant populations. Conservation Biology 29, 54-60.
  20. Lenzen, M., Moran, D., Kanemoto, K., Foran, B., Lobefaro, L. & Geschke, A. 2012 International trade drives biodiversity threats in developing nations. Nature 486, 109–112.
  21. Westley, F. et al. 2011 Tipping toward sustainability: Emerging pathways of transformation. AMBIO 40, 762–780.


Understanding Science