โ† Wiki Index / Ecology & Biosphere / Central Place Foraging
๐ŸŒฟ Ecology & Biosphere

Central Place Foraging

Central Place Foraging is a mathematical extension of optimal foraging theory that predicts how animals returning gathered resources to a fixed nest, burrow, or colony optimize their prey size selectivity and load size as a function of round-trip travel distance.

Central Place Foraging (CPF) is a theoretical and empirical framework in behavioral ecology and optimal foraging theory that analyzes the foraging decisions of organisms that travel outward from a fixed spatial base (the central placeโ€”such as an avian nest, a mammalian burrow or cache, a beaver lodge, or an ant or honeybee colony) to harvest resources in patchy habitats and transport those resources back to the central place for storage, consumption, or provisioning of offspring. Formulated in 1979 by American ecologists Gordon H. Orians and Niels E. Pearson (and expanded by Thomas W. Schoener in 1979 and Alejandro Kacelnik in 1984), Central Place Foraging modifies classical continuous-foraging models (such as Charnov's 1976 Marginal Value Theorem and the Contingency Prey Model) to account for the asymmetric energetic and time costs of outbound search travel versus inbound loaded transport.

Whereas a free-ranging, non-central-place forager (such as an antelope grazing across a savanna; see Jarmanโ€“Bell Principle) consumes food items at the exact location where it encounters them, a central place forager must invest a round-trip transit time Ttravel = tout + tin (proportional to radial distance D from the nest) for every foraging bout. Orians and Pearson proved that incorporating round-trip transit time yields two fundamental predictions across the biosphere:

  1. In single-prey loaders (animals that carry only one prey item per trip), the minimum acceptable prey size (and diet selectivity) increases monotonically with distance D from the central place.
  2. In multiple-prey loaders (animals that gather a mouthful, crop-load, or corbicular load of multiple items per trip), the optimal load size (L*) and patch residence time (t*) increase monotonically with distance D from the central place.

Single-Prey Loaders vs. Multiple-Prey Loaders (Orians & Pearson, 1979)

Orians and Pearson divided central place foragers into two distinct functional classes based on how captured prey items are transported back to the nest:

1. The Single-Prey Loader Model (Distance-Dependent Prey Size Selectivity)

A single-prey loaderโ€”such as an osprey (Pandion haliaetus) carrying a fish in its talons, a great tit (Parus major) or pied wagtail (Motacilla alba) carrying a single large caterpillar in its beak, a solitary digger wasp (Sphex or Bembix) carrying a paralyzed orthopteran to its brood burrow, or a beaver (Castor canadensis) dragging a felled sapling to its pondโ€”must return to the central place as soon as it captures and decides to keep a single item.

For a single-prey loader foraging in a patch at round-trip travel time Ttravel from the nest, each prey item i of net energy value ei requires pursuit/handling time hi inside the patch plus the entire return and outbound travel time Ttravel. Thus, the effective profitability of item i is no longer simply ei / hi (as in MacArthur and Pianka's classical diet model), but rather:

ProfitabilityCPF(i) = ei / (hi + Ttravel)

As distance D (and hence Ttravel) from the central place becomes large (Ttravel โ‰ซ hi), the denominator (hi + Ttravel) becomes nearly equal for small and large prey alike. Carrying a tiny prey item back across a 2 km flight wastes the entire round-trip transit time Ttravel for a paltry energetic return! Consequently, the Single-Prey Loader Model predicts the Distance-Size Effect: foragers hunting far from the central place must ignore small prey items that they readily accept near the nest, specializing strictly on large, high-energy prey at long distances.

Forager Functional ClassTransport ConstraintDiminishing Returns Mechanism Inside PatchPredicted Effect of Increasing Distance (D / Ttravel)Classic Experimental / Field Taxa
Single-Prey LoaderCaptures and transports exactly 1 indivisible item per tripNone inside patch; each capture terminates the foraging tripMinimum acceptable prey size & selectivity increase with distance DOspreys (Pandion), Merlins (Falco), Beavers (Castor), Sphecid Wasps
Multiple-Prey Loader (Bill / Cheek Pouch)Accumulates N items in bill, cheek pouches, or claws per tripLoading effect: holding N items in the bill impairs capture of item N + 1Optimal load size L* and patch time t* increase with distance DEuropean Starlings (Sturnus), Puffins (Fratercula), Chipmunks (Tamias)
Multiple-Prey Loader (Crop / Nectar / Honeybee)Pumps fluid nectar or seeds into internal crop or corbiculaeResource depletion in flowers + mass-dependent flight metabolic costOptimal volume L* increases (or optimizes Enet / Espent efficiency)Honeybees (Apis mellifera), Bumblebees (Bombus), Hummingbirds
Dive / Surface Oxygen ForagerSurface air (= central place); dives to benthic prey patchNon-linearly rising physiological O2 depletion & lactate costDeeper dives (Ttravel) require longer bottom foraging times t*Weddell seals (Leptonychotes), Cormorants (Phalacrocorax), Sea Otters

2. The Multiple-Prey Loader Model (The Loading Curve and Optimal Load Size L*)

A multiple-prey loaderโ€”such as a European starling (Sturnus vulgaris) probing turf for leatherjacket larvae (Tipula paludosa), an Atlantic puffin (Fratercula arctica) lining up sandeels crosswise in its bill, an eastern chipmunk (Tamias striatus) stuffing acorns into its internal cheek pouches, or a leaf-cutting ant (Atta)โ€”collects multiple items during a single patch visit before flying or running home.

Inside the patch, cumulative load energy L(t) as a function of patch foraging time t exhibits diminishing returns (โˆ‚L/โˆ‚t > 0, โˆ‚2L/โˆ‚t2 < 0) not only because the patch is depleted (compare Giving-Up Density), but also because of the mechanical loading effect: when a starling or puffin already holds 4 wriggling larvae in its beak, its probing efficiency drops and the risk of dropping held larvae increases as it tries to catch a 5th larva. Applying the Marginal Value Theorem tangent construction to the loading curve L(t), the parent bird maximizes its rate of food delivery to the nestlings (Rdelivery = L(t) / [Ttravel + t]) by departing the patch at residence time t* where:

โˆ‚L(t*) / โˆ‚t = L(t*) / (Ttravel + t*)

Because a longer round-trip travel time Ttravel shifts the tangent origin (โˆ’Ttravel, 0) farther to the left on the time axis, the point of tangency slides higher up the concave loading curve L(t). Thus, multiple-prey loaders traveling farther from the central place spend longer in the patch (t*) and bring back significantly larger loads (L*) per trip.

Classic Empirical Tests: Kacelnik's Starlings and Jenkins's Beavers

1. Alejandro Kacelnik's (1984) Operant Starling Experiments

In 1984, Alejandro Kacelnik conducted one of the most precise quantitative tests of Central Place Foraging in behavioral ecology by training wild parent European starlings (Sturnus vulgaris) provisioning nestlings in nestboxes at Wytham Woods, Oxford, to collect mealworms dropped down a vertical pipe onto an experimental wooden feeder placed at distances ranging from 8 meters to 600 meters from their nestboxes. By controlling the exact inter-worm delivery schedule on a microcomputer to mimic a concave natural loading curve (simulating the progressive difficulty of holding 1, 2, 3, ..., 8 mealworms in the beak), Kacelnik tested whether parent starlings maximized:

  • Gross energy delivery rate to the nestlings: L(t) / (Ttravel + t),
  • Net family energy gain (accounting for the parent's flight metabolic cost cflight Ttravel and the chicks' basal metabolic expenditure), or
  • Energetic efficiency (Edelivered / Eexpended).

As feeder distance increased from 8 m to 600 m, the starlings systematically increased their average load size from ~3 mealworms per trip to ~7 mealworms per trip, closely tracking the curve predicted by maximizing net family rate of energy gain!

2. Beaver Tree-Size Selectivity Along Distance Gradients

North American beavers (Castor canadensis) are archetypal terrestrial central place foragers: they swim out from their central pond lodge, walk inland up a forest bank, fell trees (Populus aspen, Salix willow, Betula birch), and drag branches back to the pond for winter bark caching and dam construction. Field studies by Stephen H. Jenkins and Mark A. McGinley & Thomas G. Whitham (1985) demonstrated that beavers alter both species selectivity and trunk-diameter cutting rules strictly according to CPF predictions: near the water's edge (0โ€“10 m), beavers cut saplings and trees across all trunk diameters indiscriminately; at 30 to 100 m inland, where overland hauling costs and wolf/coyote predation risk (see Giving-Up Density) are high, beavers become highly selective in accordance with CPF models.

3. The Honeybee Nectar Paradox and Energetic Efficiency (Schmid-Hempel et al., 1985)

When Paul Schmid-Hempel, Alejandro Kacelnik, and Alasdair I. Houston (1985) tested CPF predictions on nectar-collecting worker honeybees (Apis mellifera) flying between their hive (where they dance to recruit nestmates; see Information Center Hypothesis) and artificial flower patches, they uncovered an instructive anomaly: bees departed flower patches with crops only partially full, matching the smaller load size L* that maximizes energetic efficiency (Enet / Eflight cost) rather than the larger load size that maximizes net rate of energy delivery (Enet / T)! Schmid-Hempel et al. proved why: a worker honeybee has a fixed lifetime wing-muscle metabolic budget (~800 km of total cumulative flight before irreversible flight-muscle senescence and wing wear cause death). Because heavy nectar loads dramatically increase metabolic power expenditure during flight, maximizing energetic efficiency (Egained per unit metabolic work) maximizes the total lifetime honey contribution of the sterile worker to the colony's inclusive fitness.

Halo Depletion (Ashmole's Halo) Around Colonial Central Places

When thousands or millions of central place foragers aggregate at a single breeding colony (such as cliff-nesting seabird colonies of murres Uria, gannets Morus, and puffins, or leaf-cutting ant nests Atta), their collective radial foraging creates a zone of severe resource depletion surrounding the central placeโ€”known in marine ornithology as Ashmole's Halo (proposed by N. Philip Ashmole in 1963 and verified via satellite tracking by Stephen C. Votier and Kyle H. Elliott). As the colony grows in population size (N), near-colony prey are depleted, forcing parents to commute ever greater distances D, which increases Ttravel, lowers chick provisioning rates, and exerts stabilizing density-dependent regulation on maximum colony size.

Frequently Asked Questions

How does Central Place Foraging differ from Charnov's Marginal Value Theorem?

Charnov's classical Marginal Value Theorem assumes a continuous, nomadic forager that moves sequentially from Patch 1 โ†’ Patch 2 โ†’ Patch 3, eating food inside each patch as it goes. Central Place Foraging (Orians & Pearson, 1979) requires the forager to make a round-trip excursion (Central Place โ†’ Patch โ†’ Central Place) on every bout and carry the harvested load on the return leg. Thus, in CPF, travel time is doubled (tout + tin), the return trip incurs a mass-dependent transport cost, and diminishing returns inside the patch can arise purely from mechanical mouth/beak loading constraints even when the patch itself is not being depleted.

Why do diving mammals and seabirds obey Central Place Foraging models underwater?

In 1985 and 1988, regan C. Kramer and Alasdair I. Houston & John M. McNamara recognized that air-breathing aquatic diversโ€”such as Weddell seals (Leptonychotes weddellii), blue whales, cormorants, and diving ducksโ€”are physiological central place foragers for whom the water surface is the central place (where the resource collected is atmospheric O2) and the seafloor or deep scattering layer is the foraging patch! Because diving to greater depths requires longer vertical transit times (Ttravel), CPF theory accurately predicts that deeper dives require longer bottom foraging times and longer surface oxygen-recovery intervals.

What is the "prey-switching vs. self-feeding" rule in parent central place foragers?

Formulated by Alasdair I. Houston (1987) and verified in field studies of aquatic insects, wheatears, and terns, a provisioning parent foraging in a distant patch splits its captures into two streams: small prey items with low CPF transport profitability (ei / [hi + Ttravel]) are swallowed immediately on the spot by the parent to fuel its own metabolic flight costs, whereas large, high-value prey items are held in the bill and carried back to the central place to feed the offspring.

References

  1. Stephens, D. W., & Krebs, J. R. (1986). Foraging Theory (synthesizing Orians & Pearson, 1979). Princeton University Press. https://doi.org/10.1515/9780691206790
  2. Kacelnik, A. (Feb 1984). Central place foraging in starlings (Sturnus vulgaris). I. Patch residence time. Journal of Animal Ecology, 53(1), 283โ€“299. https://doi.org/10.2307/4357
  3. Schmid-Hempel, P., Kacelnik, A., & Houston, A. I. (May 1985). Honeybees maximize efficiency by not filling their crop. Behavioral Ecology and Sociobiology, 17(1), 61โ€“66. https://doi.org/10.1007/BF00299430
  4. Schoener, T. W. (Dec 1979). Generality of the size-distance relation in models of optimal feeding. The American Naturalist, 114(6), 902โ€“914. https://doi.org/10.1086/283537
๐Ÿงฌ Support Open Science
Explore interactive science games on LowKey Games or browse scientific apparel & accessories in the CRV Science Shop.
๐Ÿ’ก Spot an inaccuracy or have primary literature to expand this article?
Submit a citation or revision note using your shared CRV Science / LowKey Games profile ID.
Suggest Revision