Resilience to climate change: translating principles into practice


Correspondence author. E-mail:


‘Resilience’ has been a subject of ecological theory and investigation over many years. It has also become a common objective of climate change adaptation across the whole range of human activities. Climate change adaptation within a conservation framework draws on both of these histories, and it is not surprising that increasing resilience is frequently an overarching objective in adaptation strategies and principles; many of which have been published by conservationists and ecologists in recent years (Heller & Zavaleta 2009; Mawdsley, O’Malley & Ojima 2009). It is a word that resonates with policy makers and scientists alike. Nevertheless, as ecologists working on the boundary between science and practice, we have found it a slippery concept and translating a strategic commitment to ‘increasing resilience’ into effective, on-the-ground, action presents many challenges. This is partly because resilience has a range of meanings and is not used consistently and partly because there are substantial uncertainties around the best way of enhancing resilience (however defined) in any particular situation. It is not enough to know that several different approaches may increase resilience to climate change; we need to know which are the most efficient and effective in particular circumstances if we are to prioritize scarce resources.

In this perspective, we ask whether ‘resilience’ is a useful concept with which to frame climate change adaptation and set out an approach to bridging the gap between conceptual thinking and practical action. This is a topic that matters in its own right – climate change is a critical challenge for conservation – but it is also an interesting example of what is necessary to really apply ecological principles to practical problems.

Meanings of resilience

Two broad understandings of resilience can be identified in the ecological literature (Gunderson 2000):

  • 1 the amount of disturbance that an ecosystem can withstand without changing self-organized processes and structures;
  • 2 the return time to a stable state following a perturbation.

The understanding of resilience as the return time to a stable state – recovery from disturbance – parallels the meaning of the term in material science. It has also entered wider usage as a term to describe the abilities of systems, organizations and people to recover from an adverse impact. Within ecology, this usage is often contrasted with ‘resistance’, the ability of a system to resist a disturbance and remain unchanged. Resilience in the sense of ‘the amount of disturbance that an ecosystem can withstand’ can be traced back to Holling (1973). Once the resilience of a system is exceeded, a tipping point is reached and a new stable state is entered into, from which it may not be possible to return to the former state. This is perhaps closest to the sense in which ‘resilience’ is used in wider thinking on climate change adaptation. An influential example from the Intergovernmental Panel on Climate Change (IPCC 2007) is: ‘the ability of a social or ecological system to absorb disturbances while retaining the same basic structure and ways of functioning, the capacity for self-organization, and the capacity to adapt to stress and change’. Other understandings of resilience tend to be variants on these themes; for example, a recent review of ecological networks in England (Lawton et al. 2010) combined both, defining a resilient ecological network as ‘capable of absorbing, resisting or recovering from disturbances...’

As climate change is an ongoing process, withstanding ongoing disturbance, rather than recovering from individual events, is generally a more helpful concept. Nevertheless, recovery from extreme climatic events, which are likely to become more frequent with climate change, is a component of this. For practical purposes, we advocate a pragmatic approach: accept that resilience is a broad term encompassing a series of related concepts and ensure that the intended meaning is clearly explained when applied to specific situations.

It is helpful to contrast building resilience with adaptation by transformation (Poiani et al. 2011) in which conservation action actively moves a system to a changed state that it is better suited to the changing circumstances. An intermediate approach of accepting or accommodating change can also be identified in which inevitable changes are accepted rather than promoted. We can therefore set out a spectrum of adaptation responses:

  • Resilience – Accommodation – Transformation

A good example of transformation is the managed realignment of coasts. There are a number of schemes in the UK in which sea defences have been removed or established further inland to allow natural processes to reshape the coastline and create new intertidal habitats, which also provide protection against storm surges (Dixon et al. 2009). With rising sea levels, this is often more cost-effective than building new hard sea defences (e.g. sea-walls), as well as creating new habitats. Managed realignment has proved controversial where it threatens private property and, in general, transformational approaches raise more concerns than resilience approaches. Nevertheless, other contentious transformational approaches, such as species translocation, are being actively considered by conservationists in a variety of organizations (Minteer & Collins 2010). In future, more radical transformation measures might also include ‘land zonation’ or ‘sparing’, in which food production is prioritized in those areas that are most productive in a changed climate and conservation is prioritized in areas of low agricultural productivity.

Some authors (Heller & Zavaleta 2009; Poiani et al. 2011) distinguish between resistance and resilience strategies for climate change adaptation, to separate strategies that seek to preserve the status quo from those that accept a degree of change to make the system as a whole more able to absorb change. Resistance strategies in this sense are not about building-resistant ecosystems but about resisting change by active intervention. This is a meaningful distinction, but we have avoided this use, because of the potential confusion with the ecological concept of a resistant community or ecosystem, which could also be encompassed within the concept of resilience as the capacity to absorb disturbance.

The evidence base for decision-making

Table 1 summarizes the main approaches to increasing resilience to climate change. They range from current best practice (‘reduce other pressures’) to substantial enhancements of protected site networks. They raise all sorts of questions for the conservation manager. How strong is the evidence that they will be effective? Which are most effective in which circumstances? Are there potential negative impacts? Are they practically feasible? A good evidence base is essential and will need to be assessed on a case-by-case basis. This is something that may take a considerable time but some general issues are worth noting.

Table 1.   Generic measures that have been proposed to increase resilience to climate change
Reduce other pressures on biodiversity
Increase the number of protected sites
Increase the size of individual protected sites
Provide buffer areas around protected sites
Improve the functional connectivity between sites
Protect/create cool microclimates and potential refugia for species
Maintain or increase the habitat heterogeneity at site and landscape scales
Maintain species diversity within communities
Protect natural processes
Promote the potential for natural genetic exchange between populations
Control invasive species

Climate change adaptation is about trying to pre-empt future change: we cannot easily fall back on tried and tested methods and uncertainty is inevitable. Many climate change adaptation principles are primarily based on ecological theory, rather than practical experience, for example considerations of network design draw on island biogeography and metapopulation theory. There is, however, empirical support for many of the measures in Table 1, even if it falls short of the textbook replicated experiment. For example, the value of protecting refugia – cool or moist locations where species may survive – is supported by palaeoecological data (Willis et al. 2010). This is a component of resilience which is being promoted in a variety of contexts, including dry Australian ecosystems (Lindenmayer et al. 2010; Prober et al. 2011). There are also case studies in which reducing other pressures reduces the impacts of climate variations. Pearce Higgins et al. (2010) showed that blocking drainage channels and predator control additively reduced the climate sensitivity of golden plovers Pluvialis apricaria. Wilmers & Getz (2005) presented evidence that restoring the food web by the reintroduction of wolves (Canis lupus) to Yellowstone buffered scavenger species against changing climate impacts.

There remain major uncertainties. The most frequently advocated approach to climate change adaptation is increasing connectivity of ecological networks to facilitate species in moving to new locations, either locally or further afield (Heller & Zaveleta 2009). This would also promote functioning metapopulations with exchange of individuals between habitat patches, increasing the chances of recovery after extreme events. There is, however, much debate about the overall effectiveness of enhancing connectivity as an adaptation measure, and the value of physical corridors in particular has been questioned (e.g. Hodgson et al. 2009; Doerr, Barrett & Doerr 2011). A key issue is the contrasting dispersal capacity of different species. Rare species often have very low dispersal abilities (or very specialized habitat requirements) and are unlikely to benefit from enhanced connectivity, whereas common invasive species may disperse more easily and present a threat to native species. Fausch et al. (2009) discuss the strategy of isolating native salmonids from non-native species in headwater streams in Montana, USA and the difficult trade-off with population size that this requires. Similarly, linking populations with different locally adapted genotypes may increase the chances of genetic adaptation, but this must be balanced against the risk of genetic homogenization.

An incomplete evidence base is not a reason for inaction: we do not have time our side: waiting for greater certainty will often make adaptation harder. Climate change adaptation is essentially about risk management. In many cases, promoting transformation will be risky, but enhancing resilience will be a relatively safe way forward. Uncertainty does, however, need to be deliberately factored into planning. Basic steps include looking for solutions with relatively few risks of adverse impacts and potential additional benefits (often termed ‘no regrets’ measures). An important but often overlooked action is to ensure that good monitoring is in place to judge the effectiveness of interventions. This is an essential component of an adaptive management approach in which management is developed and adjusted on the basis of experience. It leads naturally to a model of research in which practical conservation actions double up as field experiments and close working between scientists and conservationists is the norm.

An approach to developing resilience to climate change

So, how does the conservation practitioner approach the task of increasing resilience to climate change? We summarize our approach in Table 2.

Table 2.   Key questions to guide the development of ecological resilience to climate change
1) What are your conservation objectives? What are you ultimately trying to protect or enhance?
2) What are the key pressures that threaten these conservation objectives? Both the direct impacts of climate change and their interactions with other factors may be important.
3) Is ongoing persistence, or recovery from extreme events, the aim of increasing resilience?
4) At what spatial scale are you aiming for resilience – and is this the most appropriate scale?
5) How much change would be acceptable, or even desirable?
6) What is known about natural variability: is there a baseline against which success can be judged?

The starting point is a clear understanding of what the conservation objectives are in any particular circumstance. Are there particular species or communities that are a priority for protection or recovery? Are we concerned about particular areas? What are the legal or policy contexts? In the UK, as in many other countries, we are increasing recognizing the importance of healthy ecosystems and the benefits they provide to people. Ecosystem services, such as water supply, recreation or historic landscapes, may be conservation objectives, as well as the protection of species and biological communities.

Establishing our high-level objectives will shape the adaptation measures that are needed to increase resilience. If protecting ecosystem services is the main objective, they may not depend on a narrowly defined assemblage of species. Similarly, many community-level attributes may be maintained without the presence of all the original species. For example, a species-rich calcareous grassland in northern England would continue to fulfil national conservation criteria if a proportion of typical northerly species were replaced by others with a more southerly distribution. Where maintaining specific species in a particular place is not the overriding objective, manipulating or accepting change in species composition may in fact be one of the most effective ways to make communities and ecosystems more resilient. Perhaps the best example is in forests where the planting or encouragement of species better adapted to a changing climate is widely advocated (Thompson et al. 2009). This increases the chances that forest cover will remain and ecosystem services, such as timber production and carbon sequestration, will be maintained; it also protects the conditions that support the ground flora and animal communities. From a conservation perspective, this is preferable to the loss of the canopy if climate-sensitive canopy trees die.

Where species are the objects of conservation, there is an analogous issue with genotypes. The survival of a species in its current location will be made more likely by evolutionary change in response to climate change, although this may be at the expense of distinctive local varieties. In considering the different levels of ecological organization (genotypes – species – communities – ecosystems), there is a general tendency for change to occur first at the lower levels of organization (Fig. 1). Whilst resilience may legitimately be considered at any level, it is likely to be a more effective strategy at the community and ecosystem level, because it is enhanced by the capacity for change at genotype and species levels.

Figure 1.

 Schematic illustration of the relationship between the extent of ecological change and the magnitude of climate change at different levels of organization. Ecosystem processes and services and communities have a higher intrinsic level of resilience as they are not conditional on particular species. This opens up opportunities to manipulate their resilience by managing species composition.

Spatial scale is an important dimension to consider in scoping practical adaptation measures. Climate change may bring changes in the species found on a particular nature reserve (which in England might cover a few hundred hectares). However, species lost from one site may thrive at a different one. For many purposes, resilience at a landscape scale, which takes account of the network of sites, will be more important than resilience at a site scale. In a similar way, managed realignment represents a transformation in the habitats and appearance of a particular area of coastline but, from the perspective of maintaining coastal habitats along the coast as a whole, it promotes resilience. This is not to say that local considerations are never important, unique conditions, such as soil type or management history, may be of overriding importance as may local opinion.

A number of temporal considerations are also important in planning practical adaptation. The background variability in populations, communities or ecosystem attributes is essential to understanding the limits of tolerance and judging the success of adaptation measures. A historical perspective is also crucial to understanding present situations. For example, knowledge of historical frequency and recovery time is essential to understanding resilience to extreme events such as fire (Noss 2001).

Developing resilience to climate change requires conservationists to be well versed in both ecology and climate change, whilst retaining their expertise in the particular ecosystems they are responsible for. In Natural England, we are developing training courses and other ways of upskilling our staff to meet this challenge.

Is resilience a useful concept for climate change adaptation?

Used in a broad sense, resilience is a useful concept, if conservation managers understand their system (which can still be a significant limitation in many ecosystems around the world) and have explicit adaptation goals. They should also consider the spectrum of changes from ‘resilience’ to ‘transformation’ that might occur, or be needed, over time at different levels of ecosystem organization and at different spatial scales.

Perhaps the most useful aspect of resilience as a concept is that it conveys the aim of enabling the persistence of broadly similar communities or ecosystem services, as opposed to accepting or promoting change in response to climate change. This is more about the mindset of the conservationist than the properties of the natural environment, but this is not to belittle it: conservation is a human activity. As a risk management strategy, building resilience will often be a better approach than either doing nothing or promoting substantial change. There is, however, a caveat to this: as the impacts of climate change become more pronounced and organisms exhibit ecological, physiological, behavioural and genetic responses, the more often we will come up against the limits of resilience. International agreements have consistently aimed to limit global temperature rise to 2 °C, but emissions are currently on track for a rise of the order of 4 °C this century (New et al. 2010). A point may be reached where we will need to be much more open to transformational approaches if we are to stand a chance of maintaining overall biodiversity and safeguarding ecosystem services. Preparing for this possibility is a profound challenge to scientists and conservationists alike.


Our thinking on resilience has benefitted from the insights of many colleagues with skills spanning the full range from practical conservation to theoretical research.


The authors work for Natural England, the government conservation agency for England. Mike Morecroft leads on Climate Change at Natural England. He is a Senior Visiting Research Associate at Oxford University and previously led a research group at the Centre for Ecology and Hydrology, Wallingford. Humphrey Crick has worked at the British Trust for Ornithology and then at Natural England on a wide range of climate change related issues – with over 25 publications in this area – and is currently specializing in farmland ecology and the wider countryside. Simon Duffield is a research entomologist by background, who has more recently worked on environmental futures and as an agri-environment scheme advisor. His research interests include invertebrate spatial dynamics and knowledge transfer. Nicholas Macgregor works on the development of conservation strategies, with a focus on climate change adaptation and large-scale conservation; he is chair of the cross-European ENCA Climate Change Group.