Resource Partitioning

Resource partitioning (also termed niche differentiation or niche segregation) is the ecological and evolutionary process by which competing species divide shared limiting resources along spatial, trophic, temporal, or morphological axes, thereby reducing interspecific competition below intraspecific competition and enabling long-term coexistence within a shared ecosystem. Under the competitive exclusion principle formalized by Vito Volterra, Alfred J. Lotka, and Georgii F. Gause, two species that occupy an identical ecological niche in a constant environment cannot stably coexist at equilibrium because the species capable of maintaining positive population growth at a lower resource threshold will drive its competitor to local extirpation. Resource partitioning resolves this paradox of biodiversity by separating the realized niches of sympatric guild members across one or more resource utilization axes.

Ecologists distinguish between two primary mechanisms driving resource partitioning across natural communities. Plastic behavioral or ecological sorting occurs within a single generation when subordinate competitors shift their foraging microhabitat, prey selection, or activity schedule in the immediate presence of a dominant competitor—compressing their physiological fundamental niche into a narrower realized niche. Conversely, evolutionary resource partitioning unfolds over multigenerational timescales through ecological character displacement, in which natural selection favors heritable divergence in trophic morphology (such as beak depth in Darwin's finches or carnassial tooth length in felids and mustelids) among sympatric populations compared to allopatric populations.

The Lotka-Volterra Competition Framework and Limiting Similarity

In quantitative community ecology, resource partitioning is evaluated by mapping each species' utilization function, u_i(R), along a continuous or discrete resource gradient R (such as seed diameter, prey body length, canopy foraging height, or soil rooting depth). In the classical formulation developed by Robert H. MacArthur and Richard Levins (1967) and extended by Robert M. May (1973), suppose two competing species utilize a one-dimensional resource spectrum with Gaussian utilization curves characterized by mean resource positions μ₁ and μ₂, separated by niche distance d = |μ₁ − μ₂|, and common niche breadth (standard deviation) w. The symmetric Lotka-Volterra competition coefficient α₁₂—measuring the per-capita competitive effect of species 2 on species 1 relative to intraspecific competition—is given by the convolution of their utilization curves:

α₁₂ = exp( −d² / 4w² )

When two species share identical resource preferences (d = 0), niche overlap is complete (α₁₂ = 1.0), and coexistence is unstable unless demographic rates and carrying capacities are exactly equal. As resource partitioning increases the separation-to-width ratio (d / w), the competition coefficient α₁₂ declines exponentially. In deterministic models coupled to environmental variance, MacArthur and May demonstrated that stable multispecies packing typically requires d / w ≥ 1.0 to 1.5, a theoretical threshold known as limiting similarity, which corresponds to a maximum tolerable interspecific competition coefficient of approximately α ≈ 0.54 to 0.78.

Schoener's Index and Pianka's Symmetric Niche Overlap

In empirical field studies where resource states are categorized into n discrete classes (such as taxonomic prey items in stomach contents or discrete canopy zones), ecologists quantify the degree of resource partitioning using two standard metrics:

1. Schoener's Proportional Similarity Index (PS or C_jk): Introduced by Thomas W. Schoener in 1968 and 1970, this metric measures the shared area beneath two discrete probability distributions where p_ij and p_ik represent the proportions of total resource use by species j and species k falling within resource state i:

C_jk = 1 − 0.5 ∑ |p_ij − p_ik| Values range from 0.0 (complete resource partitioning with zero shared use) to 1.0 (complete overlap), with values below 0.60 generally interpreted in field ecology as biologically significant niche segregation.

1. Pianka's Symmetric Overlap Index (O_jk): Formulated by Eric R. Pianka in 1973 as the geometric mean of the asymmetric MacArthur-Levins overlap coefficients:

O_jk = ( ∑ p_ij p_ik ) / √( ∑ p_ij² ∑ p_ik² ) By comparing observed Pianka overlap matrices against randomized null models (such as Lawlor's RA3 niche-breadth-retaining permutation algorithm), community ecologists test whether observed resource partitioning is statistically greater than expected by chance assembly.

In a comprehensive review of 81 empirical studies encompassing over 250 terrestrial and aquatic animal guilds, Thomas W. Schoener (1974) established a consistent hierarchy among the three primary dimensions of resource partitioning: habitat (spatial) dimension > food-type (trophic) dimension > temporal (diel or seasonal) dimension. Moreover, multispecies guilds frequently partition resources complementarily across two axes—such that species exhibiting high spatial overlap diverge in prey size, whereas species consuming identical prey segregate by microhabitat.

MacArthur's Boreal Forest Warblers (Setophaga)

The most influential field demonstration of spatial and behavioral resource partitioning was conducted by Robert H. MacArthur (1958) in the coastal spruce-fir boreal forests of Maine and Vermont. Five congeneric species of wood-warblers—the Cape May warbler (Setophaga tigrina), yellow-rumped (myrtle) warbler (S. coronata), black-throated green warbler (S. virens), Blackburnian warbler (S. fusca), and bay-breasted warbler (S. castanea)—possess virtually identical body masses (8 to 10 g) and bill dimensions, and all five breed synchronously within the same mature white spruce (Picea glauca) stands while feeding on spruce budworm (Choristoneura fumiferana) and other arboreal insects.

By dividing spruce crowns into six vertical height zones and three radial branch zones (outer needle-bearing tips, middle dense foliage, and inner bare bark and lichen zones) and recording over 8,000 seconds of foraging observations, MacArthur discovered that each species occupied a distinct microhabitat and kinematic foraging niche:

• Cape May warbler (S. tigrina): Concentrated in the uppermost sunlit terminal spire (top 10–20% of the tree) along outer needle tips, executing vertical aerial hawking flights and nectar/sap feeding; its population density tracked episodic spruce budworm outbreaks.
• Blackburnian warbler (S. fusca): Foraged at high-to-upper-middle elevations along the outermost foliage tips, moving radially from the trunk outward toward terminal twigs.
• Black-throated green warbler (S. virens): Occupied the dense middle canopy foliage, combining tangential hopping along lateral branches with brief hovering maneuvers to glean insects from the undersides of needles.
• Bay-breasted warbler (S. castanea): Restricted its foraging to the shaded interior core of the middle canopy, moving slowly and methodically along bare branches and lichen-encrusted limbs.
• Yellow-rumped warbler (S. coronata): Operated as a generalized lower-canopy and understory forager (0 to 4 m above the forest floor), frequently flycatching from low perches and gleaning across forest floor mosses.

Greater Antillean Anolis Lizard Ecomorphs

Across the four main islands of the Greater Antilles (Cuba, Hispaniola, Jamaica, and Puerto Rico), over 110 species of arboreal iguanian lizards in the genus Anolis partition structural vegetation and thermal microclimates with high precision. As synthesized by Ernest E. Williams (1972) and Jonathan B. Losos (2009), local communities of up to 11 sympatric Anolis species segregate along three orthogonal niche axes:

1. Structural Microhabitat (Perch Height and Perch Diameter): Sympatric species sort into six discrete morphological and behavioral ecomorphs—crown-giant, trunk-crown, trunk, trunk-ground, twig, and grass-bush anoles. Twig anoles (such as Anolis valencienni and A. insolitus) possess short limbs and prehensile tails adapted for slow creeping along narrow distal branches (<1 cm diameter), whereas trunk-ground anoles (A. cristatellus and A. sagrei) possess long hindlimbs adapted for rapid sprinting and jumping from broad lower tree trunks to the ground.
2. Thermal Microclimate (Heliothermy vs. Thermoconformity): Even when two trunk-ground or trunk-crown congeners share similar structural perches, they partition the forest floor and canopy by solar insolation—one species basking in sunlit forest gaps (maintaining body temperatures T_b ≈ 31–33 °C) while its sister species occupies shaded deep-forest interiors (T_b ≈ 25–27 °C).
3. Prey Size Partitioning: Within each structural ecomorph guild, sexual size dimorphism and interspecific body-size ratios follow G. Evelyn Hutchinson's 1.3× trophic appendage rule, partitioning arthropod prey by body length. Phylogenetic reconstructions demonstrate that these six ecomorph classes evolved independently on each island via convergent evolution driven by local interspecific competition.

Root Depth, Mycorrhizal Networks, and Nitrogen Partitioning in Plants

Because terrestrial vascular plants are sessile autotrophs that all require solar photons, atmospheric carbon dioxide (CO₂), water (H₂O), and dissolved mineral ions (NO₃⁻, NH₄⁺, H₂PO₄⁻, K⁺, Ca²⁺), early theorists questioned how hundreds of plant species avoid competitive exclusion—a terrestrial analog of G. Evelyn Hutchinson's 1961 "paradox of the plankton." Modern soil ecology and tracer experiments have revealed three belowground resource partitioning mechanisms:

• Vertical Rooting Stratification: In water-limited savannas and deserts, shallow-rooted fibrous grasses intercept episodic summer precipitation pulses in the upper 0 to 30 cm of soil, whereas deep-rooted taproot shrubs and leguminous trees (Acacia, Prosopis) extract groundwater from depths of 2 to 15 meters—formalized in Heinrich Walter's two-layer hypothesis (1971) of tree-grass coexistence.
• Chemical Partitioning of Soil Nitrogen: Using triple-labeled ¹⁵N tracers (¹⁵NH₄⁺, ¹⁵NO₃⁻, and dual-labeled ¹⁵N/¹³C glycine) in nitrogen-limited Arctic tundra and alpine meadows, Robert B. McKane and colleagues (2002) demonstrated that dominant plant species partition nitrogen by chemical form, uptake season, and rooting depth. Shallow-rooted Carex sedges preferentially assimilate nitrate (NO₃⁻) during late summer, deeper Eriophorum vaginatum tussocks absorb free amino acids (glycine) in early spring thaw, and ericoid mycorrhizal shrubs (Vaccinium) hydrolyze recalcitrant organic nitrogen polymers.
• Light Spectrum Partitioning in Aquatic Autotrophs: In marine and freshwater water columns, coexisting cyanobacteria and eukaryotic microalgae partition the photosynthetically active radiation (PAR, 400–700 nm) spectrum using complementary phycobilisome pigments: phycoerythrin-rich strains absorb green photons (540–570 nm) penetrating deep oligotrophic waters, whereas phycocyanin-rich and chlorophyll a/b strains utilize red and blue wavelengths in shallow surface layers.

Observing that sympatric species occupy distinct niches does not by itself prove that active interspecific competition maintains those boundaries. As Joseph H. Connell emphasized in his 1980 critique, "the ghost of competition past," two species may differ in diet or microhabitat either because evolutionary divergence eliminated their overlap millennia ago (leaving zero ongoing competition) or because they evolved different adaptations in allopatry prior to secondary contact.

To distinguish active ecological competition from historical divergence, experimental ecologists perform reciprocal species-removal experiments (such as Connell's 1961 intertidal barnacle removals of Balanus balanoides and Chthamalus stellatus in Scotland, and David Tilman's nutrient-ratio resource competition experiments in diatoms and prairie grasses). When removing Species B triggers an immediate behavioral or demographic expansion of Species A into the vacated microhabitat (competitive release), active ongoing competition is confirmed. Conversely, if Species A fails to expand its niche after Competitor B is removed—and if sympatric populations exhibit heritable morphological divergence relative to allopatric controls—the observed resource partitioning reflects fixed evolutionary character displacement.

1. MacArthur, R. H. (1958). Population ecology of some warblers of northeastern coniferous forests. Ecology, 39(4), 599–619. https://doi.org/10.2307/1931600
2. Schoener, T. W. (1974). Resource partitioning in ecological communities. Science, 185(4145), 27–39. https://doi.org/10.1126/science.185.4145.27
3. McKane, R. B., Johnson, L. C., Shaver, G. R., Nadelhoffer, K. J., Rastetter, E. B., Fry, B., Giblin, A. E., Kielland, K., Kwiatkowski, B. L., Laundre, J. A., & Murray, G. (2002). Resource-based niches provide a basis for plant species diversity and dominance in arctic tundra. Nature, 415(6867), 68–71. https://doi.org/10.1038/415068a
4. Stuart, Y. E., Campbell, T. S., Hohenlohe, P. A., Reynolds, R. G., Revell, L. J., & Losos, J. B. (2014). Rapid evolution of a native species following invasion by a congener. Science, 346(6208), 463–466. https://doi.org/10.1126/science.1257008
5. Chesson, P. (2000). Mechanisms of maintenance of species diversity. Annual Review of Ecology and Systematics, 31(1), 343–366. https://doi.org/10.1146/annurev.ecolsys.31.1.343