Skip to content

Community Ecology: Interactions, Food Webs and Succession

Species interactions, Lotka-Volterra competition and predation, food webs, keystone species, diversity indices, island biogeography and succession for CSIR NET.

  • 21 min read
  • Updated
On this page · 15 sections
  1. Prerequisites
  2. The Hook: A Starfish and a Crowbar
  3. Types of Species Interaction
  4. Competition
  5. Predation and Herbivory
  6. Community Structure
  7. Measuring Diversity
  8. Ecological Succession
  9. Memory Tricks
  10. Don’t Confuse
  11. Common Mistakes
  12. Practice
  13. Quick Revision Table
  14. Key Takeaways
  15. What’s Next

Prerequisites

The Hook: A Starfish and a Crowbar

How CSIR NET asks it. Part B: classify an interaction (+/−, +/0), keystone versus foundation species, Gause’s principle, Connell’s barnacles, Clements versus Gleason, facilitation versus inhibition models, hydrosere stages. Part C: Lotka–Volterra outcomes from carrying capacities and α values, predator–prey equilibria, Shannon and Simpson indices, β diversity, species–area calculations, and interpretation of removal experiments.


Types of Species Interaction

Score each partner as + (benefits), − (harmed) or 0 (unaffected):

InteractionSpecies 1Species 2Example
Mutualism++Mycorrhiza; Rhizobium–legume; fig and fig wasp; lichen (fungus + alga)
Commensalism+0Epiphytic orchids on trees; cattle egrets following cattle; barnacles on whales
Competition−−Two Paramecium species for bacteria; plants for light
Predation+−Lion and zebra; insectivorous plants
Herbivory+−Grazers; a special case of exploitation of plants
Parasitism+−Tapeworm; Cuscuta (dodder) on host plants
Amensalism0−Antibiotic-producing Penicillium inhibiting bacteria; allelopathy by black walnut (juglone)
Neutralism00Rarely demonstrable; species with no measurable effect on each other

Two further labels: protocooperation (both benefit but the association is not obligatory) and obligate versus facultative mutualism. Brood parasitism (cuckoos) is covered in Behavioural Ecology.


Competition

Intraspecific competition (within a species) produces density dependence and the logistic curve. Interspecific competition (between species) reduces the fitness of both.

Mechanisms:

  • Exploitation (scramble): indirect, through using up a shared resource.
  • Interference (contest): direct, through aggression, territoriality or chemicals (allelopathy).
  • Apparent competition: two prey species harm each other indirectly by sustaining a shared predator. It looks like competition but involves no shared resource.

The Lotka–Volterra competition model

Each species grows logistically, but the other species also uses up its carrying capacity:

$$\frac{dN_1}{dt} = r_1 N_1\,\frac{K_1 - N_1 - \alpha_{12}N_2}{K_1}\qquad \frac{dN_2}{dt} = r_2 N_2\,\frac{K_2 - N_2 - \alpha_{21}N_1}{K_2}$$

\(\alpha_{12}\) is the competition coefficient: the effect of one individual of species 2 on species 1, in units of species-1 individuals. \(\alpha_{12} = 0.5\) means one N₂ uses as much of species 1’s resource as half an N₁.

Zero-growth isoclines. Species 1 stops growing along the line \(N_1 = K_1 - \alpha_{12}N_2\), which runs from \(K_1\) on the N₁ axis to \(K_1/\alpha_{12}\) on the N₂ axis. Species 2’s isocline runs from \(K_2\) on the N₂ axis to \(K_2/\alpha_{21}\) on the N₁ axis. Comparing the intercepts gives four outcomes:

ConditionOutcome
\(K_1 > K_2/\alpha_{21}\) and \(K_2 < K_1/\alpha_{12}\)Species 1 always wins
\(K_2 > K_1/\alpha_{12}\) and \(K_1 < K_2/\alpha_{21}\)Species 2 always wins
\(K_1 > K_2/\alpha_{21}\) and \(K_2 > K_1/\alpha_{12}\)Unstable equilibrium: either wins depending on starting numbers
\(K_1 < K_2/\alpha_{21}\) and \(K_2 < K_1/\alpha_{12}\)Stable coexistence

Stable coexistence in plain words: each species limits itself more than it limits the other (intraspecific competition exceeds interspecific). When the coexistence point exists, it lies at:

$$N_1^{*} = \frac{K_1 - \alpha_{12}K_2}{1 - \alpha_{12}\alpha_{21}}\qquad N_2^{*} = \frac{K_2 - \alpha_{21}K_1}{1 - \alpha_{12}\alpha_{21}}$$

Competitive exclusion and niche partitioning

Gause’s competitive exclusion principle: two species competing for the same limiting resource cannot coexist indefinitely; one excludes the other. Gause’s experiments (1930s) with Paramecium: P. aurelia excluded P. caudatum when grown together on the same bacterial food, though each grew well alone. P. caudatum and P. bursaria coexisted, because P. bursaria fed in a different part of the tube.

How competitors coexist in nature:

  • Resource partitioning: MacArthur’s warblers in spruce trees forage in different zones of the same tree.
  • Character displacement: where two species overlap (sympatry) they differ more in a resource-related trait than where each occurs alone (allopatry). Beak depth in Galápagos ground finches is the standard example.
  • Temporal partitioning, disturbance that keeps the dominant competitor in check, and spatial heterogeneity.

Connell’s barnacles: fundamental versus realised niche

On the Scottish coast, the barnacle Chthamalus lives high on the shore and Balanus lower down. Connell (published 1961) removed Balanus from rocks: Chthamalus then survived lower down. Removing Chthamalus did not let Balanus move up, because Balanus is limited above by desiccation. So Chthamalus’s fundamental niche extends lower than its realised niche, and the lower limit is set by competition, while the upper limit of Balanus is set by a physical factor. This is the cleanest field demonstration of the two niche concepts and a frequent Part C setting.


Predation and Herbivory

The Lotka–Volterra predator–prey model

With N prey and P predators:

$$\frac{dN}{dt} = rN - aNP \qquad \frac{dP}{dt} = b\,aNP - mP$$

r: prey growth rate in the absence of predators; a: attack rate; b: conversion efficiency of eaten prey into new predators; m: predator death rate.

Setting each equation to zero gives the isoclines and the equilibrium:

$$N^{*} = \frac{m}{a\,b}\qquad P^{*} = \frac{r}{a}$$

Consequences that Part C likes:

  • The equilibrium prey density is set by predator parameters (m, a, b), and the equilibrium predator density by prey growth (r). Enriching the prey’s environment (raising r) increases predators, not prey.
  • Without density dependence the model gives neutral cycles: populations oscillate with amplitude set by starting values, and the predator peak lags the prey peak by about a quarter cycle.
  • Adding prey self-limitation damps the cycles towards a stable point.

Lab and field cases: Gause found that Didinium ate all Paramecium and then starved, unless refuges or immigration were added. Huffaker (1958) obtained sustained predator–prey mite cycles only in a complex arrangement of oranges with barriers and dispersal routes: spatial heterogeneity stabilises. The snowshoe hare–Canada lynx cycle (about ten years, from fur-trading records) involves hare food supply as well as lynx.

Functional and numerical responses

The functional response (Holling, 1959) is the number of prey eaten per predator as a function of prey density:

TypeShapeMechanism
ILinear up to a plateauFilter feeders; no handling-time limit until satiation
IIRises and decelerates to a plateau (hyperbolic)Handling time limits intake; the disc equation \(\dfrac{aN}{1 + a h N}\)
IIISigmoidLow intake at low density (prey switching, search image, refuges), then rapid rise; can stabilise prey

The numerical response is the change in predator numbers with prey density (through reproduction or aggregation).

Prey defences

Crypsis (camouflage), aposematism (warning colours of toxic prey), Batesian mimicry (a harmless mimic copies a harmful or distasteful model: stingless hoverflies with wasp-like yellow and black bands), Müllerian mimicry (two or more harmful species converge on a shared signal), chemical defences of plants (alkaloids, tannins, cardenolides), and induced defences.

Parasitism and coevolution

Parasites are usually smaller than their hosts, often live on or in one host for a long time, and often do not kill it. Microparasites (viruses, bacteria, protozoa) multiply within the host; macroparasites (helminths, ticks) grow but usually do not multiply within it. Hosts and parasites coevolve: the Red Queen hypothesis (Van Valen, 1973) says species must keep evolving just to keep their relative fitness.


Community Structure

Food chains and food webs

A food chain is a linear sequence of feeding (producer → primary consumer → secondary consumer…). A food web is the network of interlinked chains in a community. Two types of chain: grazing (starting from living plants) and detritus (starting from dead organic matter; dominant in forests, where most NPP is not eaten alive).

Food-web statistics:

  • S = number of species; L = number of feeding links.
  • Linkage density = L/S.
  • Connectance = the proportion of possible links realised. With S species, if any species can eat any other (including itself), the number of possible links is S², and C = L/S². (Some authors use S(S − 1)/2 possible links instead; check the definition in the question.)
  • Food chain length is usually short (three to five levels), limited by energy loss at each transfer and by dynamic stability.

Omnivory (feeding at more than one trophic level) and intraguild predation (killing a competitor that also shares prey) blur trophic levels.

Keystone, foundation and other species roles

TermDefinitionExample
Keystone speciesEffect on the community disproportionately large relative to its abundancePisaster (Paine); sea otter in kelp forests
Foundation speciesAbundant, often structural species that create habitat; large effect because of high biomassCorals, kelp, dominant trees
Dominant speciesThe most abundant or highest-biomass speciesSal or teak in Indian deciduous forests
Ecosystem engineerPhysically modifies the environment, creating or changing habitatsBeaver dams; earthworms; elephants opening woodland
Umbrella speciesLarge-area species whose protection protects many othersTiger
Flagship speciesCharismatic species used to win public supportGiant panda, tiger
Indicator speciesPresence or condition signals environmental stateLichens for SO₂ pollution

Trophic cascades

A trophic cascade is an indirect effect of a predator two or more levels down. Sea otters eat sea urchins; urchins graze kelp. Where otters were hunted out, urchins multiplied and kelp forests turned into “urchin barrens”; where otters returned, kelp recovered (work associated with James Estes). Top-down control by predators contrasts with bottom-up control by nutrient supply. Hairston, Smith and Slobodkin’s “green world” hypothesis (1960) argued that the world is green because predators keep herbivores in check.

Disturbance and diversity

The intermediate disturbance hypothesis (Connell, 1978): diversity is highest at intermediate frequency or intensity of disturbance. Too little disturbance lets competitive dominants exclude others; too much lets only fast colonisers survive. Sousa’s experiments with boulders of different sizes on a shore, which overturn at different frequencies, are the standard test.


Measuring Diversity

MeasureFormulaNotes
Species richness (S)Count of speciesIgnores abundance; depends on sample size
Shannon index (H’)\(H' = -\sum p_i \ln p_i\)p_i = proportion of species i; rises with richness and evenness; maximum ln S
Evenness (Pielou’s J)\(J = H'/\ln S\)1 when all species equally abundant
Simpson’s index (D)\(D = \sum p_i^2\)Probability two random individuals are the same species; dominance, falls as diversity rises
Simpson’s diversity\(1 - D\) or \(1/D\)Rise with diversity; 1/D is the “effective number” of equally common species

For finite samples Simpson’s index is sometimes computed as \(\sum n_i(n_i - 1)/[N(N - 1)]\); questions usually state which form to use.

Alpha, beta and gamma diversity (Whittaker)

  • α diversity: within one site or habitat.
  • β diversity: turnover of species between sites, along a gradient.
  • γ diversity: total diversity of a region.
  • Whittaker’s β = γ/ᾱ, where ᾱ is the mean α across sites. It equals 1 when every site has the same species list.
  • Jaccard similarity = c/(a + b − c), where a and b are species counts at the two sites and c the number shared. Sørensen = 2c/(a + b).

Island biogeography

MacArthur and Wilson’s equilibrium theory of island biogeography (1967): the number of species on an island is a dynamic balance between immigration, which falls as the island fills and is lower for distant islands, and extinction, which rises as the island fills and is higher for small islands. Large, near islands hold the most species; small, far islands the fewest. Species composition keeps turning over even at equilibrium.

The species–area relationship:

$$S = cA^{z}\qquad \log S = \log c + z \log A$$

z is often roughly 0.2 to 0.35 for islands and lower for nested areas within a mainland. A rule of thumb that follows from z ≈ 0.3: a tenfold increase in area roughly doubles species number (10^0.3 ≈ 2). Simberloff and Wilson’s defaunation of small mangrove islands in Florida tested the theory experimentally: species numbers returned to near their original values, with different species. The theory underlies reserve design and the SLOSS debate (single large or several small), covered in Biodiversity, Conservation and Policy.

Latitudinal gradient

Species richness generally increases from the poles to the tropics. Proposed reasons include greater area and energy in the tropics, longer evolutionary time without glaciation, higher speciation rates and more niche specialisation. None is accepted as the sole cause.


Ecological Succession

Succession is the directional, more or less predictable change in species composition of a community over time at one place.

TypeStarting pointSpeedExample
PrimaryBare substrate with no soil or previous communitySlow (centuries)Lava flows, glacial moraine, newly exposed rock, sand dunes
SecondarySoil and seed bank remain after disturbanceFaster (decades)Abandoned farmland, forest after fire or clearing
AutogenicChanges driven by the organisms themselves (shading, soil building)Most successions
AllogenicChanges driven by external factors (siltation, climate)Lake filling with sediment
AutotrophicDominated early by producers; biomass risesMost terrestrial successions
HeterotrophicStarts with organic matter, dominated by consumers and decomposers; energy declinesDecomposition of a fallen log or a carcass

Vocabulary: pioneer species (first colonisers), seral stages (the intermediate communities), sere (the whole sequence), climax (the relatively stable end community). Seres are named by the starting habitat: hydrosere (water), xerosere (dry habitat), lithosere (bare rock), psammosere (sand), halosere (saline soil or water).

Hydrosere and lithosere stages

graph TD
    subgraph Hydrosere
    H1["Phytoplankton stage"] --> H2["Rooted submerged stage
Hydrilla, Vallisneria"] H2 --> H3["Rooted floating stage
Nymphaea, Trapa"] H3 --> H4["Reed-swamp stage
Typha, Phragmites"] H4 --> H5["Sedge-meadow stage
Carex, grasses"] H5 --> H6["Woodland stage"] H6 --> H7["Climax forest"] end subgraph Lithosere L1["Crustose lichens"] --> L2["Foliose lichens"] L2 --> L3["Mosses"] L3 --> L4["Herbs and grasses"] L4 --> L5["Shrubs"] L5 --> L6["Climax forest"] end

A hydrosere moves from wetter to drier (mesic) conditions; a xerosere from drier to mesic. Both converge on a mesic climax suited to the regional climate.

Theories of the climax

ViewAuthorIdea
MonoclimaxF. E. Clements (1916)Each climatic region has one climax; the community is like a superorganism developing to maturity
PolyclimaxA. G. TansleySeveral climaxes in one region, set by soil, topography, fire and animals as well as climate
Climax patternR. H. WhittakerVegetation is a continuous pattern of climax types grading with environmental gradients
IndividualisticH. A. Gleason (1926)Species respond individually to the environment; communities are chance assemblages, not superorganisms

Mechanisms: Connell and Slatyer (1977)

ModelEffect of early species on later onesExample situation
FacilitationEarly species make conditions better for later ones (soil building, nitrogen fixation, shade)Primary succession on glacial moraine, where nitrogen-fixing Dryas and alder enrich the soil
ToleranceEarly species have no effect; later species are those that tolerate lower resource levels and simply outlastSome old-field successions
InhibitionEarly species hold space and block later ones until they die or are damagedAlgae on intertidal boulders; later species establish only after disturbance removes residents
AttributeEarly stagesMature stages
BiomassLowHigh
Gross production / respiration (P/R)Greater than 1Approaches 1
Net community productionHighLow
Species diversityLow, risingHigh (may dip slightly at climax)
Food chainsShort, linear, grazingLong, web-like, detritus-based
Nutrient cyclesOpen, leakyClosed, tight
Life historiesr-selected, small, short-livedK-selected, large, long-lived
Niche breadthBroadNarrow
Stability (resistance)LowHigh

Memory Tricks

Don’t Confuse

ThisvsThat
Keystone: big effect, low abundancevsFoundation / dominant: big effect because abundant
Batesian: harmless mimic of harmful modelvsMüllerian: harmful species share a signal
Exploitation competition: through the resourcevsApparent competition: through a shared predator
Primary succession: no soilvsSecondary succession: soil and seed bank present
Clements: superorganism, monoclimaxvsGleason: individualistic, chance assemblages
Type II functional response: handling time, deceleratingvsType III: sigmoid, switching, can stabilise
α diversity: within a sitevsβ diversity: turnover between sites
Amensalism: 0/−vsCommensalism: +/0
Character displacement: divergence in sympatryvsResource partitioning: different use of a shared resource (may not involve trait change)

Common Mistakes


Practice

All questions below are original practice questions written in CSIR NET style. They are not previous-year questions.

Level 1: Foundation

  1. Classify the interaction in which one species is harmed and the other is unaffected.
  2. In the Connell–Slatyer models, which mechanism requires early colonists for later species to establish?
  3. Which index is the probability that two individuals drawn at random belong to the same species?
Answers
  1. Amensalism (0/−).
  2. Facilitation.
  3. Simpson’s index, D = Σpᵢ².

Level 2: Part B Style

Q1 (practice). A community has three species with relative abundances 0.50, 0.25 and 0.25. Its Shannon index (natural log) and Pielou’s evenness are closest to:

  • (a) H’ = 1.04; J = 0.95
  • (b) H’ = 1.10; J = 1.00
  • (c) H’ = 0.63; J = 0.57
  • (d) H’ = 1.50; J = 0.95
Answer and explanation

(a). H’ = −[0.5 ln 0.5 + 2 × 0.25 ln 0.25] = 0.347 + 0.693 = 1.040. Maximum for three species = ln 3 = 1.099, so J = 1.040/1.099 = 0.95. (b) is the value for perfectly even abundances. Simpson’s D for this community is 0.25 + 0.0625 + 0.0625 = 0.375 (1 − D = 0.625; 1/D = 2.67).

Q2 (practice). Two competing species have K₁ = 100, K₂ = 80, α₁₂ = 0.5 and α₂₁ = 0.6. The Lotka–Volterra model predicts:

  • (a) Species 1 excludes species 2
  • (b) Species 2 excludes species 1
  • (c) Stable coexistence at about N₁ = 86, N₂ = 29
  • (d) An unstable equilibrium; the outcome depends on starting numbers
Answer and explanation

(c). K₂/α₂₁ = 80/0.6 = 133.3 > K₁ = 100, and K₁/α₁₂ = 100/0.5 = 200 > K₂ = 80. Each species limits itself more than it limits the other: stable coexistence. N₁* = (100 − 0.5 × 80)/(1 − 0.3) = 60/0.7 = 85.7; N₂* = (80 − 0.6 × 100)/0.7 = 20/0.7 = 28.6.

Q3 (practice). Removal of a predatory snail that makes up less than 2% of the animal biomass on a reef causes one grazer to dominate and the number of macroinvertebrate species to fall by half. The snail is best described as:

  • (a) a foundation species
  • (b) a keystone species
  • (c) a dominant species
  • (d) an ecosystem engineer
Answer and explanation

(b). A small share of biomass with a disproportionately large effect on community composition is the definition of a keystone species. Foundation and dominant species have large effects because they are abundant; an engineer acts by physically modifying the habitat.

Q4 (practice). On an intertidal boulder field, the first algae to settle on cleared boulders persist until they are torn off by waves, and only then do later-successional algae establish. This fits:

  • (a) the facilitation model
  • (b) the tolerance model
  • (c) the inhibition model
  • (d) Clements’ monoclimax model
Answer and explanation

(c). Early occupants hold space and block later species until disturbance or death removes them. In facilitation they would improve conditions; in tolerance they would have no effect.

Level 3: Part C Style

Q1 (practice). A predator–prey system follows the Lotka–Volterra model with prey growth rate r = 0.5 per month, attack rate a = 0.01 per predator per month, conversion efficiency b = 0.1 and predator death rate m = 0.2 per month.

(i) Find the equilibrium prey and predator densities. (ii) A fertiliser treatment doubles r. What happens to N* and P*? (iii) Which population peaks first in each cycle?

Answer and worked solution
  1. N* = m/(ab) = 0.2/(0.01 × 0.1) = 200 prey; P* = r/a = 0.5/0.01 = 50 predators.
  2. With r = 1.0: N* stays 200; P* doubles to 100. Enrichment feeds the predators, not the prey standing stock.
  3. Prey peaks first; predators peak about a quarter cycle later, as they convert the abundant prey into offspring. The prey then crashes under high predation, followed by the predator.

Q2 (practice). Three forest plots along an altitudinal gradient were surveyed. Plot A had species 1–10, plot B species 6–15, and plot C species 11–20 (all lists complete). Which statements are correct?

(A) γ diversity is 20 species and mean α diversity is 10 species. (B) Whittaker’s β diversity is 2. (C) The Jaccard similarity between plots A and B is 0.5. (D) Plots A and C have zero similarity, indicating complete species turnover between them.

  • (a) A, B and D
  • (b) A and C only
  • (c) B, C and D
  • (d) A, B, C and D
Answer and worked solution

(a) A, B and D.

  • (A) The union of lists is species 1–20: γ = 20. Each plot has 10: ᾱ = 10. Correct.
  • (B) β = γ/ᾱ = 20/10 = 2. Correct.
  • (C) A and B share species 6–10 (c = 5). Jaccard = 5/(10 + 10 − 5) = 0.33, not 0.5. (Sørensen would be 2 × 5/20 = 0.5, which is the trap.) Wrong.
  • (D) A (1–10) and C (11–20) share nothing: similarity 0, complete turnover. Correct.

Q3 (practice). For an archipelago, species number follows S = cA^z with z = 0.25. Island P has 40 bird species and an area of 100 km². (i) How many species are expected on an island of 1,600 km² with the same c? (ii) If 90% of the habitat on island P is cleared, what fraction of its species is eventually expected to persist? (iii) Name one assumption that makes the answer to (ii) an approximation.

Answer and worked solution
  1. Area ratio 16; S ratio = 16^0.25 = 2. Expected S = 80 species.
  2. Remaining area = 10% → S ratio = 0.1^0.25 = 0.56. About 56% of species (about 22 of 40) are expected to persist; roughly 44% to be lost eventually (“extinction debt”, since losses take time).
  3. Assumptions: the species–area curve for whole islands applies to habitat loss within one island; species are lost at random with respect to where they live; the remaining habitat is one block rather than fragments; z is the same at small areas. Any one is acceptable.

Quick Revision Table

ItemFact
InteractionsMutualism +/+; commensalism +/0; competition −/−; predation, parasitism +/−; amensalism 0/−
Competition typesExploitation, interference, apparent (via shared predator)
L–V competition coexistenceK₁ < K₂/α₂₁ and K₂ < K₁/α₁₂: self-limitation beats cross-limitation
Coexistence pointN₁* = (K₁ − α₁₂K₂)/(1 − α₁₂α₂₁)
GauseP. aurelia excludes P. caudatum; P. bursaria coexists
ConnellChthamalus excluded from lower shore by Balanus; Balanus upper limit physical
L–V predator–preyN* = m/(ab); P* = r/a; predator lags prey by about 1/4 cycle
Holling responsesI linear; II handling time, disc equation; III sigmoid, switching
MimicryBatesian (harmless copies harmful); Müllerian (harmful share signal)
KeystonePaine, Pisaster; sea otter–urchin–kelp cascade
ConnectanceL/S² (or per stated definition); linkage density L/S
Intermediate disturbanceDiversity peaks at intermediate disturbance (Connell 1978)
Shannon / evennessH’ = −Σpᵢ ln pᵢ; J = H’/ln S
SimpsonD = Σpᵢ² (dominance); 1 − D, 1/D (diversity)
β diversityWhittaker γ/ᾱ; Jaccard c/(a + b − c); Sørensen 2c/(a + b)
Island biogeographyImmigration falls with distance; extinction rises as area shrinks; S = cA^z
SuccessionPrimary (no soil) vs secondary; hydrosere, lithosere stages
Climax viewsClements monoclimax; Tansley polyclimax; Whittaker pattern; Gleason individualistic
Connell–SlatyerFacilitation, tolerance, inhibition
Odum trendsBiomass up; P/R → 1; webs detritus-based; cycles closed

Key Takeaways

What’s Next