Woodland dieback caused by climate change, Adrian Newton.

Key messages:

  • The term resilience has multiple definitions and encompasses several different properties of ecological systems. 
  • Academics, practitioners and policy makers tend to refer to resilience in different ways, with the potential for unintended biodiversity loss [1]
  • UK-wide biodiversity loss will lead to lower resilience of ecosystem functions to future perturbations [2]
  • Trait diversity increases resistance to drought on chalk grasslands. 
  • Measuring and monitoring resilience is challenging and there is a need for indicators linked to the mechanisms underpinning resilience [3]
  • Forest resilience can be measured through the analysis of recovery trajectories [4].
  • Connectivity increases the resilience of temperate forest and shallow lake ecosystems. 

Resilience in environmental policy:

Parties at the 2010 Convention on Biological Diversity agreed a mission that includes to ‘to ensure that by 2020 ecosystems are resilient and continue to provide essential services…’.

Ensuring ecological resilience were part of the commitments in the 2011 Government White Paper

Section 6 of The Environment Act (Wales) places a duty on public authorities to seek ‘to promote the resilience of ecosystems’.

Human wellbeing depends on maintaining ecosystem functions and services in the face of increasing environmental change.  Ecological resilience is increasingly a policy goal, with a need to apply ecological concepts relating to resilience to environmental management.  Yet there is discussion about how to define and measure resilience, with evidence about resilience strongly lagging behind theory.  BESS research is contributing in three areas:

  • Defining resilience
  • Evidence about declining resilience of ecosystem functions
  • Maintaining resilience
What exactly is resilience?
Given the multiple pressures on ecosystems and the need to adapt to environmental change, aiming for ‘resilient’ landscapes is an appealing vision.  The term resilience has become a ‘boundary object’: a concept with slightly different uses and interpretations by different communities [5].  Boundary objects have benefits of bringing people together from different disciplines, as long as there is awareness of the potential for misunderstanding.  Resilience may be used to refer to particular properties of ecosystems, in a more general sense to communicate a goal or in relation to human capital (such as conservation partnerships and local communities) involved in the flow of ecosystem services.  A BESS workshop with researchers Volker Grimm and Hanna Weise highlighted the multiple definitions used, but that resilience essentially encompasses three related elements [5]:
  • Resistance – the ability of a system to remain essentially unaltered after disturbance.
  • Recovery – the ability of a system to return to the previous state after disturbance.
  • Persistence – the ability of a system to maintain a state over time, which depends on both resistance and recovery.
Others consider resistance to be a separate quality to resilience [4].  Theory on ecological resilience suggests that gradual environmental changes and pressures might be having little apparent impact on an ecosystem, until a sudden disturbance such as disease tips the ecosystem into a different set of reinforcing processes.  This can lead to a ‘regime shift’ with a change in the assemblage of species providing ecosystem functions.  This ‘alternative stable state’ might be less favourable in contributing to the ecosystem services needed for societal wellbeing.  Management interventions to return the ecosystem to its previous state might then be more costly than prevention or impossible [6].
Declining resilience of ecosystem functions

Relatively low levels of biodiversity may be sufficient to provide a range of ecosystem services under current environmental conditions [2].  However, the rate and intensity of environmental change is increasing, with future conditions likely to be very different [2].  If species important for underpinning particular services are lost or decline because of environmental change, this could result in sudden declines in ecosystem services [2].  However, if there are multiple species which perform a similar function, but with different responses to environmental changes, then the dominant species performing a particular function could switch.  This is often described as the ‘insurance effect’ provided by biodiversity. 

Credit Joanna Savage

Ellen Fry and colleagues found that having plant species with varying traits increases the resistance of chalk grassland to drought and aids recovery.  These functional traits of plants are characteristics that influence their surroundings and impact on the ability of the system to deliver ecosystem services.

Dark Green Fritillary, Tom Oliver.

Tom Oliver and colleagues looked at trends over four decades in the abundance of animal species that provide ecosystem functions in Great Britain [2].  They found that the resilience of some ecosystem functions are being eroded faster than others [2].  Declines in groups of species that are important for carbon sequestration and decomposition are small and have been offset by arrivals of new species [2].  However, there have been significant net declines in species that provide cultural values, pest control and pollination, suggesting an erosion of the resilience of these ecosystem functions [2]

Dieback in the New Forest with evidence for threshold responses. 

New Forest Beech, Adrian Newton.

Forest dieback is occurring globally and is ongoing in the New Forest.  BESS researchers found that some areas of the New Forest dominated by beech in 1964 had become relatively open grassland by 2014 [7].  The high death rates of large beech and oak trees were associated with changes in other tree species, changes in the plant species at ground level and an increase in grass cover.  Importantly, these changes associated with tree death occurred relatively abruptly.  Only in areas where more than around 40% of the basal area of trees were lost did the composition of other species change and grass cover increase.  These threshold responses occur when there is positive feedback, but they can be difficult to detect without long term monitoring.  This transition in the New Forest does not meet the definition of an alternative stable state required by theory, because some of the disturbance involved is ongoing grazing pressure rather than shorter-term ‘pulse’ disturbance [7].  Likely causes of beech dieback include droughts combined with attack from pathogenic fungi.  Regeneration from seedlings is then prevented by pony and deer grazing.  Beech makes up a large proportion of canopy cover in the native woodlands of southern UK and has a low tolerance for drought, so these UK forests may have low resilience to climate change.

Maintaining resilience

Management options and issues to consider include: 

CBESS research

Awareness of the multiple definitions and uses of the term resilience.  For example, ‘flood resilience’ encompasses many aspects of a local community, while ecological resilience considers the ecosystems providing flood regulation services.  CBESS and BESS small grant research highlights how saltmarsh and mudflats have a direct role in natural flood management and an indirect role through the regulation of greenhouse gases.

Understanding trade-offs between different services and for different groups of people [3].  Scenario setting, such as the DURESS scenarios for upland Wales [8], can highlight the range of drivers involved and societal choices.  There might be trade-offs between managing for short term ecosystem functions and for long term resilience [3]

Credit Debbie Coldwell

Making choices about the level and type of intervention desired by people and appropriate for the landscape and resources available.  For example, one response to beech forest dieback could be to protect seedlings from grazing [7].  An alternative option would be to view the New Forest as adapting to changing conditions, although this would involve accepting reductions or changes in some ecosystem services [7]

Actions intended to increase resilience can meet other needs and goals as part of the Ecosystem Approach [4]Evidence is emerging from BESS research that increasing connectivity in forest landscapes and freshwater shallow lake systems increases resilience.  Interventions that modify functional connectivity for one species can affect different species and processes in different ways [9].  Planting wide buffers of deciduous woodland alongside upland streams can potentially increase resilience and provide other benefits [10]. 

Identifying aims and awareness of the potential for unintended consequences for biodiversity [1].  The particular concept of resilience will influence the management strategy [4].  Management interventions designed to strengthen resilience need to be carefully selected so that they do not result in negative impacts on biodiversity [1]

Identifying the minimum thresholds for ecosystem functions and services that are acceptable to society [3].  This involves interdisciplinary working and involvement of a wide range of stakeholders, such as research by Wessex BESS on grassland restoration. 

Volunteers seeding F3UES annual meadows in Luton, Helen Hoyle.

Actions targeting human and manufactured capital, such as urban design, influence biodiversity and ecosystem functions [11].  For example, restoring park trails maintains biodiversity, erosion control and cultural ecosystem services under climate change [12]

Justifying actions to promote resilience can be challenging when they can appear redundant under current environmental conditions and within the short timescales typically considered during decision making [3]

Monitoring to keep ecosystems within safe operating spaces is currently severely limited and reporting about ecosystem services focuses on the short term [3, 4].  There is a need to develop robust cost-effective indicators linked to the mechanisms underpinning resilience [3]

Mechanisms underpinning the resistance and/or recovery of ecosystem functions to environmental disturbances [3]:

Species level mechanisms

  • Sensitivity to environmental change
  • Intrinsic rate of population increase
  • Adaptive phenotypic plasticity – the ability to change to a changing environment
  • Genetic variability
  • Allee effects - make populations less likely to recover from small numbers

Ecological community level mechanisms

  • A positive correlation between species’ importance for an ecosystem function and their response to a given environmental change will lower resistance
  • Network interaction structure
  • Ecological processes supporting recovery, such as dispersal & establishment
  • Functional redundancy - the insurance effect of biodiversity

Landscape level mechanisms

  • Local environmental heterogeneity
  • Landscape level functional connectivity
  • Potential for alternative stable states
  • Area of natural landscape cover


  1. Newton, A.C., Biodiversity risks of adopting resilience as a policy goal. Conservation Letters, 2016: p. doi: 10.1111/conl.12227.
  2. Oliver, T.H., et al., Declining resilience of ecosystem functions under biodiversity loss. Nature Communications, 2015. 6:10122.
  3. Oliver, T.H., et al., Biodiversity and resilience of ecosystem functions. Trends in Ecology & Evolution, 2015. 30(11): p. 673-684.
  4. Newton, A.C. and E. Cantarello, Restoration of forest resilience: An achievable goal? New Forests, 2015. 46(5): p. 645-668.
  5. Martin, P. Just what is resilience, anyway? Ecology for a crowded planet blog post. 
  6. Walker, B. and D. Salt, Resilience thinking: sustaining ecosystems and people in a changing world. 2006: Island Press.
  7. Martin, P.A., et al., Stand dieback and collapse in a temperate forest and its impact on forest structure and biodiversity. Forest Ecology and Management, 2015. 358: p. 130-138.
  8. Prosser, H., T. Pagella, and I. Durance, Upland Scenarios: what will the future look like? DURESS project report card 2015.
  9. Harrison, L.J., P.C.L. White, and S. Odell, Connectivity and Ecological Networks. Landscape Institute Technical Note 01/2016.
  10. Thomas, S.M., S.W. Griffiths, and S.J. Ormerod, Beyond cool: adapting upland streams for climate change using riparian woodlands. Global Change Biology, 2015. 22(1): p. 310-324.
  11. Cox, D.T.C., et al., Movement of feeder-using songbirds: the influence of urban features. Scientific Reports, 2016. 6: p. 37669.
  12. Tomczyk, A.M., P.C.L. White, and M.W. Ewertowski, Effects of extreme natural events on the provision of ecosystem services in a mountain environment: The importance of trail design in delivering system resilience and ecosystem service co-benefits. Journal of Environmental Management, 2016. 166: p. 156–167.