Biology · Population Ecology & Life-History Patterns
Density dependence, resource limits and population regulation
After a harsh environmental event, a population is smaller. That observation establishes a decline if the counts are reliable. It does not yet show whether crowding affected survival, whether the habitat changed, or whether a restoring feedback will limit later fluctuations.
Organisms, Environments & Ecological Niches introduces ecological scales, habitat and the distinction between observation and inference. Begin here with the counts and ordinary arithmetic. The mathematics explanations of functions and exponentials are optional preparation for the growth lesson; its multiplier-versus-slope explanation can be read before derivatives. Return to Animal Reproduction, Development & Life Histories and Evolutionary Processes & Adaptation for reproductive outcome stages and the evidence needed to connect demographic differences to selection.
Back to the Biology mapBy the end, you can
Distinguish density responses and environmental effects from evidence that a population is regulated over time.
Ask which per-member outcome depends on density
Density dependence means that a demographic outcome, such as survival, reproduction or net per-capita growth, changes with population density under a specified comparison. If area stays fixed, changes in N also change density. If area changes, abundance alone can conceal the density comparison. Population totals and per-member outcomes are different quantities.
Negative density dependence lowers a per-capita demographic contribution as density rises. For example, sharing a limited resource among more consumers can reduce the contribution available per member. The total number added can nevertheless rise over part of the range because more members contribute. The logistic derivative example makes that distinction visible.
Positive density dependence raises a per-member contribution over a stated range. Increased availability of partners is one possible biological mechanism, but it requires its own evidence. A rising assigned curve does not identify that mechanism. Nor should a positive low-density relationship be extrapolated to unlimited benefit from crowding.
Density responses can point in different directions
Read the diagram in words
The negative series declines 0.27, 0.24, 0.21, 0.18 as size rises 10, 20, 30, 40. The separate positive series rises−0.2, 0, 0.2, 0.4 at those sizes, crossing zero at twenty.
| Assigned population size in a fixed area | Assigned net per-capita rate per time unit |
|---|---|
| 10 | 0.27 |
| 20 | 0.24 |
| 30 | 0.21 |
| 40 | 0.18 |
| Assigned population size in a fixed area | Assigned net per-capita rate per time unit |
|---|---|
| 10 | -0.2 |
| 20 | 0 |
| 30 | 0.2 |
| 40 | 0.4 |
g(N) = 0.3 − 0.003 × N
g(N) = 0.02 × (N − 20)
Zero rate at the assigned threshold: 20
Relationships are supplied only from ten to forty
Higher density lowers this assigned per-capita rate
Higher density raises this assigned low-density rate
A plotted relationship does not identify its biological mechanism
Original alternative closed models in one fixed area, only over supplied sizes ten to forty. These are assigned relationships, not evidence for any species or an identified Allee mechanism. Time-unit rates are instantaneous, not finite-interval multipliers.
One favourable component may not determine the net outcome
An Allee effect can describe improvement in a component of individual fitness as the number or density of conspecifics rises. Conspecifics are members of the same species. A demographic Allee effect concerns the combined population-level result, rather than just one beneficial component.
Better mating success or group protection could be offset by reduced food access or other costs. Evidence for a positive component therefore need not show positive net population growth or a threshold below which decline occurs. The separate positive toy curve here stipulates a net-rate threshold; it is not a demonstration of an Allee mechanism in real organisms.
Random differences in the fates of a few individuals can also cause population counts to vary. Environmental stochasticity concerns varying external conditions; demographic stochasticity concerns chance variation in individual events such as births and deaths. Increased extinction risk at small N is not, by itself, proof that a per-member fitness component declined because fewer neighbours were present.
Keep the shock separate from the subsequent response
An environmental influence need not depend on density in a particular model. In our assigned shock, exactly half of each starting population survives. That is a stipulation for the comparison, not a rule saying that weather, fire or chemical exposure affects every individual equally at all densities.
After the shock, the model applies a finite-interval multiplier that declines with post-shock size. A starts 40→20 and then reaches 24; B starts 80→40 and then reaches 44. The shock survival fractions match, while the later multipliers differ. Environmental and density-related influences can act in sequence or interact rather than belonging to permanently separate biological categories.
These are assigned mechanisms, so the calculations follow the supplied rules. A sequence of observed counts without those rules would not establish them. Changes in resources, age composition, migration or detectability could produce similar count patterns. A density correlation alone can leave the direction and cause of the relationship unresolved.
A shared shock and later density response are separate steps
Read the diagram in words
A goes 40→20→24: twenty shock losses and later multiplier 1.2. B goes 80→40→44: forty shock losses and later multiplier 1.1. Both retain fraction 0.5 in the shock. The later multiplier is evaluated at the post-shock abundance.
| A shared shock and later density response are separate steps | Before the assigned shock | Fraction surviving the shock | Members lost during the shock | After the assigned shock | Later multiplier evaluated at post-shock size | After the later census interval |
|---|---|---|---|---|---|---|
| Assigned starting population A | 40 | 0.5 | 20 | 20 | 1.2 | 24 |
| Assigned starting population B | 80 | 0.5 | 40 | 40 | 1.1 | 44 |
λ(N) = 1.3 − 0.005 × N
One later assigned time interval
Shock and later interval are distinct transitions
Three counts do not establish long-term regulation
Original fixed-area populations, assigned half-survival shock and subsequent one-unit closed census interval. Equal shock survival is a teaching stipulation, not a rule for weather, fire or other abiotic events. No observed causal experiment or proof of long-term regulation.
Restoring tendencies allow variation
Population regulation concerns processes that tend to constrain abundance over time, often through restoring demographic feedbacks. Negative density dependence can contribute: above a model equilibrium growth is lower, and below it growth is higher. Regulation does not require abundance to remain exactly constant at every census.
Density dependence and regulation are related claims rather than interchangeable observations. A particular density response may be too weak, delayed or opposed by other processes to stabilise the complete dynamics. Environmental conditions and the effective equilibrium can also change. A bounded-looking short record is not sufficient proof of a persistent restoring mechanism.
To assess a regulation claim, connect the count record to survival, recruitment, movement, observation uncertainty and the proposed feedback at the appropriate spatial and temporal scales. Historical research on regulation emphasises the challenge of inferring mechanisms from abundance series. The lesson is to state what a record supports, rather than converting every fluctuation into proof of control.
Worked example
A common shock can leave different later multipliers
The shock keeps 0.5 × 40 = 20 of A and 0.5 × 80 = 40 of B. Its assigned losses are 20 and 40 respectively: equal survival fraction does not mean equal number lost.
The later multiplier law is λ(N) = 1.3 − 0.005N, evaluated at the post-shock size. At 20 it gives 1.2, producing 24. At 40 it gives 1.1, producing 44. These are discrete multipliers for the later interval; they are not instantaneous continuous rates r.
The separate positive-rate example has g(N) = 0.02(N−20) over supplied sizes 10 to 40. It gives−0.2 at 10, zero at 20 and 0.2 at 30. This assigned low-density threshold has a different feedback direction from the negative series. Neither finite toy comparison establishes a real population’s long-term persistence.
Check your understanding
If total births increase with N, has positive density dependence of per-capita birth rate been shown?
Show reasoning
No. More members can produce more total births while births per member fall or stay constant. The claim needs a per-member contribution, a density denominator and a defined comparison.
Does improved mating success in larger groups prove that net population growth is positive there?
Show reasoning
No. It can be a favourable fitness component while survival, resource costs or other contributions offset it. A component effect and the complete demographic result require different accounts.
Both assigned populations retain half their members in a shock. Why are their later multipliers 1.2 and 1.1 rather than equal?
Show reasoning
The later rule depends on post-shock N, which is 20 for A and 40 for B. Substituting into 1.3 − 0.005N gives 1.2 and 1.1. Equal assigned shock survival does not remove the later density difference.
What to carry forward
Specify the per-member density response, its range and the other influences before drawing a conclusion about population regulation.
References
Resources supporting this lesson’s explanations and examples.
- Biology 2e (2018; chapter 45, sections 45.1–45.4)Mary Ann Clark, Matthew Douglas and Jung ChoiTextbook chapters
- Population Regulation in Theory and Practice (Ecology 75(2):271–287, 1994; doi: 10.2307/1939533)William W. MurdochResearch abstract
- What Is the Allee Effect? (Oikos 87(1):185–190, 1999; doi: 10.2307/3547011)P. A. Stephens, W. J. Sutherland and R. P. FreckletonOriginal conceptual paper