Contribution
A conceptual distinction that ecology lacked. Persistence is a different property from return to equilibrium, it is located at the boundary of a domain of attraction rather than at its centre, and it can be undermined by the very management that maximises constancy.
Key insight
How fast a system returns to its equilibrium tells you nothing about how close it is to the edge beyond which it will not return. Optimising the first can wear away the second.
Assumptions
- Ecological systems are typically far from equilibrium and driven by random events, so equilibrium-centred analysis misses what matters for persistence.
- Realistic population models have lags, nonlinear functional responses, contagious attack and reproduction thresholds, which create bounded domains of attraction.
Key results
- Definitions. Resilience “determines the persistence of relationships within a system” and measures their ability “to absorb changes of state variables, driving variables, and parameters, and still persist”. Stability is “the ability of a system to return to an equilibrium state after a temporary disturbance”.
- Cases. A small Italian crater lake that kept one state for millennia and flipped after the Via Cassia was built (after Hutchinson), Great Lakes fisheries, spruce budworm cycles, salmon runs jumping between levels, fire-driven forest sequences.
- Trade-off. Variable environments give low stability with high resilience, and homogeneous ones the reverse.
- Measures. Resilience by the area of the domain of attraction and the height of its lowest boundary point; stability by slopes and speeds near the equilibrium.
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Stability and resilience are distinct properties | strong | conceptual, with model behaviours that separate them |
| C2 | Distinct domains of attraction are not uncommon in closed ecosystems | moderate | case histories; Holling notes the sample is biased |
| C3 | High variability can go with high resilience | moderate | budworm, forest insects (Watt’s survey) |
| C4 | Managing for a stable maximum yield can erode resilience | weak | argued from the framework and suggestive cases |
Concepts
- domain of attraction — the region of state space from which trajectories return to a given equilibrium or cycle.
- resilience — the persistence of a system’s relationships under change; a property of the domain, not of the equilibrium.
- stability — speed and smoothness of return to equilibrium.
Connections
Builds on May’s stability analyses and on Holling’s own predation models, and is the source of the resilience literature that Scheffer & Carpenter (LIT-726) work in.
Bearing on the record
Source of THEORY-132. Holling’s resilience is about staying within a domain of attraction. Aubin’s viability (THEORY-134) is about staying within a constraint set given apart from the dynamics. The two are not the same: a regime can be resilient (hard to dislodge) while its constraint set is being violated, which is the owner’s protective lock-in. For the essay’s §6.1, Holling is the earlier and more widely cited source of “stability is not persistence”, and Aubin is the formal machinery.
Limitations
Case histories, not controlled tests. The OCR text is poor, and the quantitative measures are proposed rather than applied.
Open questions
How to measure resilience in the field before it is lost, a problem Scheffer and colleagues later took up as early warning signals.