Ecosystem Resilience Calculator

Resilience is the amount of disturbance a system can absorb while still reorganising into the same configuration of function, structure and feedbacks. It is not the same as resistance, which is the ability to avoid changing under stress, nor stability, which concerns the speed of return to equilibrium. Because resilience cannot be measured directly, it is inferred from proxies. This index uses five. Shannon diversity is normalised by dividing by 4.0, a sensible ceiling since field values above 4 are rare and most temperate communities fall between 1.5 and 3.5. Functional redundancy and connectivity are entered as percentages, recovery rate reflects observed regeneration after past disturbance, and disturbance frequency is converted to a score of 100 minus twice the number of events per century, so a system disturbed 50 times per century scores zero. The weights place a quarter each on diversity and recovery rate, a fifth on redundancy, and 15 percent each on disturbance and connectivity. Adaptive capacity is reported separately as the mean of diversity and redundancy, because those two determine whether replacement species exist when conditions change. Two limitations deserve attention. The linear disturbance penalty assumes all disturbance is harmful, which contradicts the intermediate disturbance pattern in which fire-adapted or flood-adapted communities lose diversity when disturbance is suppressed. And composite scores cannot locate thresholds: a system may appear moderately resilient right up to a regime shift, which is why slowing recovery from small perturbations, known as critical slowing down, is watched as an early warning signal.

The term entered ecology through Crawford Stanley Holling, whose 1973 paper Resilience and Stability of Ecological Systems in the Annual Review of Ecology and Systematics separated two ideas that had been conflated. Engineering resilience measured how quickly a system returned to a single equilibrium; ecological resilience measured how much disturbance it could take before flipping to an entirely different stable state. Holling drew on budworm outbreaks in Canadian boreal forest and on fisheries collapse, and the distinction mattered because a system can return quickly to normal after small shocks yet sit close to a threshold beyond which recovery does not occur. Stuart Pimm sharpened the competing definitions of stability in a 1984 Nature review. The diversity side of the argument is older and was contested. Charles Elton argued in 1958 that diverse communities resist invasion and collapse; Robert May showed mathematically in 1973 that complexity does not automatically confer stability. Empirical support arrived through David Tilman grassland plots at Cedar Creek, Minnesota, where more diverse plots proved more stable during the 1987 to 1988 drought, and Shigeo Yachi and Michel Loreau formalised the insurance hypothesis in 1999. Applied frameworks followed: the Resilience Alliance formed in 1999, Marten Scheffer and colleagues published on catastrophic regime shifts in Nature in 2001 and on generic early warning signals in 2009, and Brian Walker with Holling, Stephen Carpenter and Ann Kinzig set out the adaptive cycle and adaptability in 2004.