Biology · Communities, Species Interactions & Food Webs

Disturbance, succession, assembly and changing communities

After a disturbance, two communities can follow different trajectories even if their early counts look similar. The later composition depends on what remains, what arrives, the environment and the interactions that follow. A changing community is not necessarily climbing toward one predetermined endpoint.

Start with Population Ecology & Life-History Patterns, especially population change and per-capita rates. Organisms, Environments & Ecological Niches provides useful background on environmental conditions, dispersal and incomplete detection. Keep the distinction between a supplied model and evidence from a real community in view throughout.

Back to the Biology map

By the end, you can

Explain conditional community assembly and distinguish explicitly defined resistance, recovery and variability from a universal notion of stability.

Disturbance changes conditions and leaves different legacies

A disturbance can alter living organisms, resources or the physical setting. Its consequences depend on extent, intensity, timing and the organisms' traits. Some organisms or structures may remain, while others are lost. The affected place need not become biologically empty.

Primary succession is conventionally used for assembly on a newly exposed substrate with little developed soil or comparable biological legacy; secondary succession follows a disturbance where substantial legacies, such as soil or surviving organisms, remain. These broad categories help describe starting conditions. They do not determine the exact subsequent species order or a timetable for recovery.

Species pools, arrival and local conditions all matter

A species pool supplies possible arrivals from a wider area. Arrival alone does not establish a persistent local population: conditions, interactions and demographic replacement can limit establishment. Surviving organisms also influence the space and resources encountered by newcomers.

Priority effects occur when the order or timing of arrival changes later outcomes. Early occupants might make conditions more favourable, draw down a resource or hinder establishment. History can therefore matter alongside environmental filtering. A difference between two trajectories does not, by itself, identify priority effects: the environment and arrivals could differ in several ways.

Two possible assembly trajectories

Different outcomes are compatible with several explanations. These records do not isolate arrival order or imply a guaranteed climax.

Read the diagram in words

The common start is 60, 30, 10. The first path passes through 30, 50, 20 to 20, 30, 50. The second passes through 50, 20, 30 to 45, 15, 40. Each record totals one hundred. Arrows mark possible chronological transitions, not inferred causal mechanisms.

Individual count
Two possible assembly trajectories	Record time	Species A	Species B	Species C	Included total
Common initial record	0	60	30	10	100
First path, middle record	1	30	50	20	100
First path, later record	2	20	30	50	100
Second path, middle record	1	50	20	30	100
Second path, later record	2	45	15	40	100

Conditions and history are not resolved by these counts
Later observation, not a guaranteed endpoint
Common initial record → First path, middle record
First path, middle record → First path, later record
Common initial record → Second path, middle record
Second path, middle record → Second path, later record

Individuals

Assigned time units

Original possible trajectories with no mechanistic transition rule. No assigned arrival sequence, causal priority test, real succession data or extrapolated endpoint.

Several mechanisms can produce community change

Facilitation describes earlier occupants improving conditions for some later organisms. Tolerance describes later organisms persisting under conditions that limit earlier ones without requiring those earlier occupants to help them. Inhibition describes occupants hindering establishment until conditions or occupancy change. These mechanisms can apply to different pairs and stages within the same community.

The branching teaching record gives two possible paths from the same initial counts. One later record is 20, 30, 50; the other is 45, 15, 40. Each still contains one hundred individuals. The divergence is assigned, with no model selecting its cause. Neither final snapshot is certified as a stable endpoint, and another disturbance or changed conditions could redirect either path.

State what resists and what recovers

Resistance concerns how little a defined response changes after a disturbance. Recovery concerns its return toward a defined reference through time; one operational use of resilience focuses on that return. Other uses emphasize maintaining a regime rather than speed of return. The word alone is therefore insufficient: name the response, reference, time and criterion.

Our response baseline is one hundred. Community A first falls to ninety, while B falls to sixty. A retains a larger fraction of its baseline. B later reaches one hundred within the recorded period, while A reaches only ninety-seven. Under our stated recovery band of 99–101, B returns earlier. Better resistance and earlier recovery need not belong to the same community.

Smaller initial change, later observed recovery

Not observed is not infinite recovery time. Listed observations do not certify unobserved intervals or future values.

Read the diagram in words

A's listed values are 100, 90, 92, 95 and 97; B's are 100, 60, 80, 100 and 100. Retained fractions at time one are 0.9 and 0.6. B first meets the 99–101 band with later listed values remaining inside at time three, two units after time one. A has no qualifying recovery in the record.

Observation time
Observation time	Community A	Community B
0	100	100
1	90	60
2	92	80
3	95	100
4	97	100

Reference response
Reference response	100
Recovery band, 99–101	99 — 101
Disturbance between the first two records	0.5
Observation time · Fraction retained at first post-disturbance observation	1

Fraction retained at first post-disturbance observation
Community A / Community B	Fraction retained at first post-disturbance observation	Initial relative drop	First qualifying listed time	Time since first post-disturbance observation
Community A	0.9	0.1	Not observed within the record	Not observed within the record
Community B	0.6	0.4	3	2

First listed value within the band with all subsequent listed values also within the band.

Assigned community-response units

Assigned time units

Original complete listed observations; no values between points or after the last point are specified. Recovery uses the stated band and subsequent listed records only.

A stable total can conceal changing composition

In four equally spaced records, species X has counts 8, 12, 8, 12 and Y has 12, 8, 12, 8. Each changes, but their combined count stays twenty. An aggregate measure can therefore be constant even when its components vary. Choosing the total rather than composition changes the stability question.

For these complete four-value sequences, each species has mean ten and population standard deviation two, giving a coefficient of variation of 0.2. The total has mean twenty and standard deviation zero. The calculation uses four as the variance denominator because it describes the whole supplied sequence, not an unbiased estimate from a random sample. It does not show that adding species caused greater stability in real communities.

A constant total with varying species counts

The variability metric refers to this defined quantity and four-value record. It does not establish a causal biodiversity–stability relationship.

Read the diagram in words

X has 8, 12, 8, 12 and Y has 12, 8, 12, 8. Each has mean ten, variance four, standard deviation two and coefficient of variation 0.2. Their sums are twenty at all four times, with mean twenty and zero variance and coefficient of variation.

Observation time
Observation time	Species X	Species Y	Included total
0	8	12	20
1	12	8	20
2	8	12	20
3	12	8	20

Four-value variance denominator
Four-value variance denominator	4

Changing species composition
Series	Mean across four observations	Population variance across four values	Population standard deviation	Coefficient of variation
Species X	10	4	2	0.2
Species Y	10	4	2	0.2
Included total	20	0	0	0

mean = sum(values)/4; variance = sum((value-mean)^2)/4; CV = sqrt(variance)/mean

Individuals

Assigned time units

Original four complete equally spaced observations. Population variance divides by four; not an unbiased estimator, random sample, causal diversity test or prediction.

A record's endpoint is also an evidence boundary

Resistance, recovery, variability and persistence can provide different information. A community can retain total abundance while losing species, or recover a measured function with a different composition. A temporary return to a reference value does not establish every aspect of ecological stability.

The assigned time series support statements about their listed observations. They do not establish what happens between observations or beyond the last one. A before-and-after sequence can describe a disturbance-associated change, but attributing its cause requires an adequate comparison and attention to concurrent influences. Succession explanations should preserve those limits rather than turn a few stages into a guaranteed climax ladder.

Worked example

Keep resistance and recovery on their own scales

At the first post-disturbance observation, A retains 90/100 = 0.9 of the reference response and B retains 60/100 = 0.6. Under this retained-fraction measure, A has greater resistance. Equivalently, their initial relative drops are 0.1 and 0.4.

Recovery is defined as the first listed return within 99–101 that remains within that band at subsequent listed observations. B qualifies at time three and stays at one hundred at time four. Relative to the first post-disturbance observation at time one, its observed recovery interval is two time units.

A does not qualify by time four. That means recovery under this criterion was not observed within the record; it does not mean recovery is impossible or takes infinite time. The criterion also cannot certify unobserved values between observations or predict either community's future.

Check your understanding

Two paths end with different species proportions. Does this prove that arrival order caused the difference?
Show reasoning
No. The constructed paths assign outcomes without a causal rule. Priority effects are one possible mechanism, but environment, remaining organisms and arrival differences also need consideration. The endpoint contrast alone does not isolate a cause.
Which community has greater resistance, and which has earlier observed recovery under the stated rule?
Show reasoning
A has greater resistance because it retains 90% rather than 60% at the first post-disturbance observation. B meets the recovery band at time three and remains within it at the later listed observation. A has no qualifying recovery within the supplied record.
Does a constant total of twenty mean the two-species community is unchanged?
Show reasoning
No. X and Y alternate between eight and twelve, so composition changes while their total remains constant. The zero variability applies only to the defined aggregate count over these four observations.

What to carry forward

Community change depends on context and history; stability claims must name the response, metric and observation period.

References

Resources supporting this lesson’s explanations and examples.