1. A RESILIENCE APPROACH TO RESOURCE MANAGEMENT AND GOVERNANCE Brian Walker

2. Why is it, that despite the best of intentions.…

3. Western Australian wheatbelt

4. rangelands in Australia

5. alternate states in N. Hemisphere lakes

6. Proposition 1 – there are two paradigms for managing natural resource systems: - MSY (maximum sustainable yield, in one form or another) - Resilience management and governance

7. extreme events can be ignored Problem assumptions with MSY problems from different sectors don’t interact (but partial solutions don’t work) change will be incremental and linear (in fact, it’s mostly lurching and non-linear) keeping the system in some optimal state will deliver maximum benefits. (There is no sustainable “optimal” state of an ecosystem, a social system, or the world. It is an unattainable goal.)

8. Key assumptions underlying Resilience Management and Governance: - SESs have non-linear dynamics and can exist in alternate stability domains (regimes) - They go through different phases of organisation and behaviour

9. Proposition 2 - Sustainable use and development rests on three capacities of social-ecological systems: - resilience - adaptability - transformability

10. Resilience “The capacity of a system to absorb disturbance and re-organise so as to retain essentially the same function, structure and feedbacks – to have the same identity” (remain in the same system regime)

11. Adaptability The capacity of actors in the system (people, in SESs) to manage resilience : (i) change the positions of thresholds between alternate regimes (ii) control the trajectory of the system – avoid crossing a threshold (or engineer such a crossing)

12. Transformability The capacity to become a fundamentally different system when ecological, social and/or economic conditions make the existing system untenable.

13. resilience places an emphasis on thresholds between alternate regimes of a system

14. reversible, with hysteresis

17. A THRESHOLD OCCURS WHERE THERE IS A CHANGE IN FEEDBACKS

18. ∆Pw = 0 (courtesy Steve Carpenter)

19. Alternate regimes in lake ecosystems

20. The net effect of increasing level of grazing on rangeland composition Feedback change: effect of grass biomass (grazing effects) on the intensity of fire

21. Feedback changes involved in regime shifts controlling variableassociated feedback changes Rainfall – evapotranspiration; leaching; water table level Temperature – soil moisture (E/T); germination (microclimate); symbiosis (coral bleaching) Nutrients – O 2 in water (decomposition); competition (plant species) Acidification - calcification (diatoms) Vegetation - water interception (cloud forests); infiltration rates; water tables; nutrients (legumes); soil temp (insulation) Landscape cover - immigration/emigration rates; reproduction; survival Herbivory - regeneration; competition; fire (fuel) Harvesting - recruitment (depensation); E/T (forests) Frequency - seed bank viability; regeneration time (fires, fallows) Predation - recovery (depensation); herbivore behaviour (spiders, wolves) The economy - income:cost ratios; debt:income ratios (markets) Social preference - subsidies / taxes (through change in government)

22. Feedback changes involved in regime shifts controlling variableassociated feedback changes Rainfall – evapotranspiration; leaching; water table level Temperature – soil moisture (E/T); germination (microclimate); symbiosis (coral bleaching) Nutrients – O 2 in water (decomposition); competition (plant species) Acidification - calcification (diatoms) Vegetation - water interception (cloud forests); infiltration rates; water tables; nutrients (legumes); soil temp (insulation) Landscape cover - immigration/emigration rates; reproduction; survival Herbivory - regeneration; competition; fire (fuel) Harvesting - recruitment (depensation); E/T (forests) Frequency - seed bank viability; regeneration time (fires, fallows) Predation - recovery (depensation); herbivore behaviour (spiders, wolves) The economy - income:cost ratios; debt:income ratios (markets) Social preference - subsidies / taxes (through change in government)

23. www.resalliance.org - thresholds database

24. What determines resilience? - Diversity - Modularity (connectedness) - Tightness of feedbacks - Openness – immigration, inflows, outflows - Reserves and other reservoirs (seedbanks, nutrient pools, memory) - Overlapping institutions - Polycentric governance

25. Rangeland in eastern Australia - lightly grazed photo by Jill Landsberg

26. abundance of grass species in lightly grazed rangeland Relative abundance (%)

27. Plant attributes determining ecosystem production (available data) height mature plant biomass specific leaf area longevity leaf litter quality height mature plant biomass specific leaf area longevity leaf litter quality

28. Functional similarities between dominant and minor species

29. Heavily grazed site, close to water - 4 of the dominant species gone - 3 replaced by functional analogues Without the response diversity (the existence of the functional analogues), a regime shift would occur – different controls and processes

30. - Ecosystem performance is promoted by high functional diversity (complementarity) - Resilience is promoted by high response diversity (in rainforests, coral reefs, lakes, rangelands) ( cf Elmqvist etal; Response diversity, ecosystem change and resilience. Frontiers in Ecology and Environment)

31. ( ) Supply of ecosystem services as a function of ecosystem state B - lake services (fish, recreation) as a function of phosphate in mud A - rangeland services (wool production from grazing) as a function of shrubs Vc - critical threshold demarcating a shift from one attractor to another. SYSTEM “REGIMES” VS OPTIMAL STATES

32. N Goulburn Broken Catchment – Victoria, Australia

33. % veg cleared in upper catchment % veg cleared in lower catchment equilibrium water table stable & below surface equilibrium water table stable at surface equilibrium water table stable & at surface Simulated clearing history

34. ( ) Supply of ecosystem services as a function of ecosystem state B - lake services (fish, recreation) as a function of phosphate in mud A - rangeland services (wool production from grazing) as a function of shrubs Vc - critical threshold demarcating a shift from one attractor to another. SYSTEM “REGIMES” VS OPTIMAL STATES

35. N Goulburn Broken Catchment - Victoria

37. Can revegetation happen fast enough to avoid equilibrium? Is it worth trying? Should they transform to some other kind of SES, to a different way of making a living?

38. Multiple regime shifts

39. Sacred Forests – alafaly, Androy region, Madagascar

40. Tombs in sacred forests

41. Sacred forest approx. 150 ha surrounded by agricultural land (mainly beans and some maize)

42. Thresholds in the Androy region of Madagascar

43. (Causse Mejan in France, a catchment in SE Australia, southern Madagascar, Western Australia wheatbelt) Multiple, interacting thresholds Kinzig et al (2006, in press) Special Issue of Ecology & Society

44. general vs. specified resilience resilience of what, to what? the “HOT” model (J. Doyle, SFI)

45. There is a cost to resilience short term cost of foregone extra ‘yield’ vs. cost of being in an alternate regime (cost x prob. of a regime shift) cost of maintaining resilience (relatively easy to estimate) vs. cost of not maintaining it (difficult to estimate)

46. Managing resilience - what determines Adaptability? – social capacity (capital) leadership trust - human capital (skills, education, health) - financial resources - natural capital -ongoing learning -?

47. What determines Transformability? still learning – but seem to be three classes: - preparedness to change (understanding, ‘mental models’, vested interests in not changing) - capacity to change (leadership, knowledge, capitals, higher level support) - options for change (diversity, knowledge of thresholds, experiments, markets,)

48. Managing resilience, and/or achieving a transformation, requires knowing where, how and when to intervene Where - knowledge of potential and actual regime shifts in the system How - knowledge of the resilience and transformability attributes When - dictated by i) external events (crises and windows of opportunity) and ii) the phases of the system’s dynamics at various scales

49.  Ω r K r: growth / exploitation resources readily available K: conservation things change slowly; resources ‘locked up’  release things change very rapidly; ‘locked up’ resources suddenly released  re-organization/renewal system boundaries tenuous; innovations are possible

50.  Ω r K

51. Panarchy – hierarchies of linked adaptive cycles

52. Dangers of K-phase behaviour - increases in “efficiency” (remove apparent redundancies, OSFA solutions) - subsidies not to change (rather than to change) - sunk costs effects - increased command-and-control (less and less flexibility) - pre-occupation with process (more and more rules) - novelty suppressed, little support for experimentation - rising transaction costs (“metabolism” vs. growth) - increasing consequences of partial solutions resilience declines

53. Success in intervening in a social-ecological system at a particular focal scale depends on where it is in the adaptive cycle, and the states of the system at higher and lower scales policy (rules, laws) financial (subsidies, infrastructure, technology, education..) management actions / techniques

54. The main messages: sustainability ≈ resilience, adaptability, transformability policy and management need to focus on the attributes that determine these three properties alternate regimes are the norm; pay attention to feedbacks that change suddenly at some level of a controlling (slow) variable “optimal” states reduce resilience -- top-down, command- and-control management doesn’t work for very long you cannot understand or manage a social-ecological system (SES) by focussing at only one scale, or using solutions that are only ecological, social, or economic. adaptive cycles and panarchy effects determine the success of interventions

55. Final proposition: Indicators of system performance (including biodiversity status) interact. Unless their changes are explicitly related to each other (especially slow and fast variables) they cannot be integrated, and they are of limited (dangerous?) use in policy development.