Skip to content
NatureHQ

Ecologyphenomenon

Disturbance

A forest that has not burned in eighty years is not undisturbed. It is disturbed differently.

Anything that clears space or kills organisms — fire, storm, flood, treefall, grazing. Systems are shaped by the regime rather than by single events, and whether it raises or lowers diversity depends on the system.

Disturbance is not the opposite of a functioning system; many systems require it. Fire-prone vegetation carries species whose seeds need heat or smoke to germinate, and excluding fire from them changes the vegetation as surely as burning it too often. What matters is the regime — how frequent, how intense, how large, how patchy — rather than any single event, and a system adapted to one regime can be transformed by a shift to another without any single disturbance being unusual. The idea most people meet here is that diversity peaks at intermediate disturbance: too little, and the best competitor takes over; too much, and only fast colonisers survive. It is elegant, it is among the most cited hypotheses in ecology, and its standing is far shakier than its prominence suggests. The predicted hump-shaped curve is absent in most studies that test for it, and the mechanism has been challenged on logical grounds — reducing competitive exclusion does not by itself maintain diversity unless species also differ in ways that let each recover when rare, which is the requirement coexistence theory imposes and the hypothesis does not supply. NatureHQ carries the hypothesis and the argument for abandoning it, because a reader taught the curve as a finding has been taught something the literature does not support. The safer statement is narrower and still useful. Disturbance affects diversity, sometimes upward and sometimes downward, and which occurs depends on what the species in that system can do — how fast they colonise, how well they tolerate the specific disturbance, and whether the regime has changed from the one they persist under.

Early coverage · 30% complete · reviewed 2026-09-04

What this page covers

Disturbance regimes shape forests, grasslands, reefs, shores and streams. The best-studied are fire in fire-prone vegetation and storm damage on reefs and rocky shores.

Often confused with: Damage, when many systems require disturbance to persist as they are; "Disturbance increases biodiversity", which is a contested general claim

Quick facts

The regime, not the event
Frequency, intensity, size and patchiness — a shift in any changes the system
Not the opposite of health
Fire-adapted plants need fire; excluding it transforms the vegetation
The famous hypothesis is contested
The hump-shaped curve is absent in most tests, and the logic is disputed
What is safe to say
Disturbance affects diversity in both directions, depending on the species present

The regime matters more than the event

And excluding disturbance is itself a change.

A single fire tells you little. What shapes vegetation is the pattern of fires — how often, how hot, how large, and how much unburnt ground is left between. Species persist under a regime, not under an event, and many carry adaptations tuned to one: seeds that need heat, bark that survives a certain intensity, resprouting from underground parts. Change the regime and those adaptations stop matching, whether the change is towards more fire or less.

That is why excluding disturbance is not a neutral act. A century of fire suppression is a change of regime, and it produces vegetation that would not otherwise be there — often with more fuel, so that when a fire does come it is of a kind the system has no adaptation to. The intuitive framing of disturbance as damage and its absence as recovery does not survive contact with these systems.

The hump-shaped curve, and why it is disputed

One of ecology’s most cited ideas, with an uncomfortable evidence base.

The idea that diversity peaks at intermediate disturbance is among ecology’s most cited, and it is disputed — not only because the pattern often is not found, but because the mechanism may not produce it.

Contested

Researchers actively disagree, and the disagreement is substantive.

The intermediate disturbance hypothesis proposes a diversity maximum at intermediate disturbance frequency and intensity. The predicted hump-shaped relationship is absent in the majority of studies testing for it, and the proposed mechanism has been argued not to generate coexistence without additional species differences of the kind coexistence theory requires.

Who this applies to
A general hypothesis about diversity and disturbance, tested across many systems.Do not extend this beyond the taxa listed — the popular version over-reaches.
Studied in
Plantae, Animalia
Why we rate it this way, and what the caveats are
ContestedModerate confidence

That disturbance affects diversity is not disputed. That it produces a general hump-shaped relationship, by the proposed mechanism, is actively argued.

How far it can be extended

Results differ between systems, and the dispute concerns whether the general prediction holds at all.

Caveats

  • Disturbance unquestionably affects diversity; the dispute is about a specific general prediction.
  • The hypothesis remains widely taught, so a reader may have met it presented as an established finding.

Where researchers disagree

  • The hump-shaped pattern is reported in some systems and absent in the majority of studies designed to test for it.
  • The mechanism is argued to require species differences that the hypothesis does not specify, and which do the work when present.

Still unanswered

  • Whether any general relationship between disturbance and diversity exists, or whether it is system-specific throughout.

Last reviewed 2026-09-04

The evidence (2 studies)

The argument for abandoning it is worth understanding because it is about mechanism rather than data. Suppose disturbance does knock back the best competitor. That prevents exclusion for a while, and it does not by itself let a weaker species persist — for that, the weaker species must be able to increase when it is rare, which requires it to differ from the competitor in some way that disturbance interacts with. Where such differences exist they do the explanatory work; where they do not, removing competitive pressure only postpones the outcome.

None of this means disturbance is unimportant. It means the general prediction — diversity highest in the middle, in any system — is not supported, and the useful questions are about particular species traits in particular regimes.

Related

What counts as disturbance

Including things that do not look like damage.

Disturbances, and what each opens up
DisturbanceWhat it removesWhat follows
FireAbove-ground vegetation, litterLight, nutrients, germination cues
Storm or windthrowCanopy trees, in patchesGaps of varying size
FloodVegetation and sedimentBare, wet substrate and new channels
GrazingSelected plants, repeatedlyAdvantage to grazing-tolerant species
TreefallOne canopy treeA small light gap — the commonest forest disturbance

Single treefalls are the ordinary case in most forests and the least dramatic. A gap of one tree changes light, temperature and moisture on a scale of metres, and the composition of a mature forest is substantially a mosaic of gaps of different ages. Disturbance is not usually a catastrophe; it is usually a hole.

The research behind this page

5 studies, newest first. Each one has a page explaining what it found and what it could not show.

2013Trends in Ecology & Evolution

The intermediate disturbance hypothesis should be abandoned

The proposed mechanism does not reliably generate a diversity peak at intermediate disturbance, and the empirical pattern is absent in the majority of studies testing for it.

2001Nature

Catastrophic shifts in ecosystems

Shallow lakes, coral reefs, rangelands and woodlands show evidence of alternative states with abrupt transitions and hysteresis, so that restoring prior conditions is insufficient to restore the prior state.

1978Science

Diversity in tropical rain forests and coral reefs

Diversity was argued to peak at intermediate levels of disturbance, with low disturbance permitting dominance by superior competitors and high disturbance permitting only rapid colonisers.

1977The American Naturalist

Mechanisms of succession in natural communities and their role in community stability and organization

The three mechanisms make different, testable predictions about what happens when early species are removed, and evidence supports different mechanisms in different systems, with inhibition common.

1973Annual Review of Ecology and Systematics

Resilience and stability of ecological systems

The two properties are independent, and a system that returns quickly from small disturbances may reorganise readily under a large one, while a variable system may persist through much greater shocks.

Where to go from here

Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.

How complete this page is, and what it is still missing

NatureHQ publishes its own gaps. This page is at 30% completeness against what we would call a finished subject, and was last reviewed on 2026-09-04. It carries 3 claims and answers 0 mapped search questions.

  • no research from the last few years is attached — check for newer work
  • more experiments could be explained in plain English
  • no popular claim about this subject has been checked yet
  • Fire ecology deserves its own treatment and has none here.
  • Human disturbance regimes, including land-use change, are mentioned only in passing.
  • Disturbance in aquatic and marine systems is under-covered relative to terrestrial.