Allometric Trophic Network
An Allometric Trophic Network (ATN) is a bioenergetic food-web modeling framework that extends the Yodzis–Innes consumer-resource equations across complex multi-trophic networks by scaling metabolic rates, maximum ingestion rates, and predator–prey handling times with organismal body mass ratios.
An Allometric Trophic Network (ATN) is a mechanistic, bioenergetic modeling framework in theoretical and systems ecology—formulated by Ulrich Brose, Richard J. Williams, and Neo D. Martinez in 2006 as a multi-species network generalization of Peter Yodzis and Stuart Innes's (1992) bioenergetic consumer–resource model—that predicts the biomass dynamics, dynamical stability, and extinction cascades of complex food webs by parameterizing species-specific metabolic rates, maximum consumption rates, and Holling functional response attack and handling parameters as quarter-power allometric scaling functions of adult body mass (Mi) and consumer–resource body-mass ratios (Mi / Mj) (Yodzis & Innes, 1992; Brose et al., 2006; Boit et al., 2012). Within ecology and population dynamics and the study of ecosystems, the Allometric Trophic Network framework resolves Robert M. May's (1972) complexity–stability paradox by demonstrating that real ecological networks—where predators are systematically 10 to 1,000 times larger in body mass than their prey—possess a built-in bioenergetic timescale separation and weak-interaction architecture that stabilizes food webs containing dozens to hundreds of interacting species (Brose et al., 2006).
Theoretical Foundation: The Yodzis–Innes (1992) Bioenergetic Model
Replacing Arbitrary Lotka–Volterra Coefficients with Mass-Specific Bioenergetics
Classical Lotka–Volterra food-web models require S2 pairwise interaction parameters that are virtually impossible to measure empirically for a 50-species community. In 1992, Peter Yodzis and Stuart Innes revolutionized food-web modeling in "Body Size and Consumer-Resource Dynamics" by recognizing that all biological energy fluxes—basal respiration, somatic growth, and maximum ingestion—scale predictably with organismal body mass (M) according to Kleiber's 3/4-power law (where whole-organism metabolic rate scales as M0.75, so mass-specific metabolic and ingestion rates scale as M−0.25; Yodzis & Innes, 1992).
In 2006, Ulrich Brose, Richard J. Williams, and Neo D. Martinez embedded the Yodzis–Innes equations into topological networks generated by the Niche Model (Williams & Martinez, 2000) or empirical lake and terrestrial food webs. In an Allometric Trophic Network of S species, the biomass density Bi (in g C m−2 or μg C L−1) of each autotrophic basal producer i and each heterotrophic consumer i is governed by coupled ordinary differential equations normalized to the intrinsic growth rate r1 of a reference primary producer:
- Basal Primary Producers (dBi / dt):
dBi / dt = ri Gi Bi − ∑j ∈ consumers [ xj yji Fji Bj ] / eji
- Heterotrophic Consumers and Predators (dBi / dt):
dBi / dt = − xi Bi + ∑k ∈ resources xi yik Fik Bi − ∑j ∈ predators [ xj yji Fji Bj ] / eji
where:
- ri = (Mi / Mref)−0.25 is the mass-specific maximum intrinsic growth rate of producer i, and Gi = 1 − (∑ cil Bl) / Ki (or a nutrient-uptake Monod term) is the producer competition/carrying-capacity function.
- xi = ax (Mi / Mref)−0.25 is the mass-specific metabolic maintenance (respiration) rate of consumer i, where the metabolic proportionality constant ax reflects metabolic type (ax ≈ 0.314 for invertebrate ectotherms; ax ≈ 0.88 for vertebrate ectotherms such as fishes; and ax ≈ 2.2–3.5 for endothermic birds and mammals; Yodzis & Innes, 1992; Brose et al., 2006).
- yik is the dimensionless maximum ingestion rate of consumer i relative to its metabolic rate xi (y ≈ 8 for invertebrate predators; y ≈ 4 for vertebrate ectotherms).
- eik is the assimilation efficiency (e ≈ 0.45 for herbivory on vascular plants; e ≈ 0.65–0.85 for carnivory and algivory).
- Fik is the multi-species Holling functional response with Hill exponent q (q = 1 for Type II hyperbolic; q = 2 for Type III sigmoidal; and predator interference coefficient c):
Fik = ( ωik Bkq ) / ( B0,*ik*q + c Bi B0,*ik*q + ∑l ∈ prey(i) ωil Blq )
| Metabolic / Trophic Guild | Mass-Specific Metabolic Constant (ax) | Maximum Ingestion-to-Metabolism Ratio (yi) | Assimilation Efficiency (eik) | Typical Predator-to-Prey Body Mass Ratio (R = Mpred / Mprey) |
|---|---|---|---|---|
| Unicellular Phytoplankton & Algae | Basal producer (ri ∝ Mi−0.25) | Autotrophic nutrient/light uptake | N/A | Consumed by zooplankton (R ≈ 102 – 104) |
| Invertebrate Ectotherm Consumers (Zooplankton, Insects, Spiders, Benthos) | ax = 0.314 | y = 8.0 (High max ingestion relative to low basal metabolism) | 0.45 (Herbivory) / 0.85 (Carnivory) | R ≈ 100.5 – 102 (3× to 100× prey mass; Brose et al., 2006) |
| Vertebrate Ectotherm Consumers (Bony Fishes, Amphibians, Reptiles) | ax = 0.88 | y = 4.0 | 0.45 (Herbivory) / 0.85 (Carnivory) | R ≈ 102 – 104 (100× to 10,000× prey mass) |
| Vertebrate Endotherm Predators (Birds and Mammals) | ax = 2.5 – 3.5 | y = 3.0 – 4.0 (Constrained by high basal thermoregulation) | 0.45 – 0.85 | R ≈ 101 – 103 (except baleen whales R > 107) |
Resolving May's Complexity–Stability Paradox: The Threshold Body-Mass Ratio (Rcrit)
In 1972, Robert M. May proved using random matrix theory that as the number of species (S) and connectance (C) increase in a randomly parameterized food web, the probability of local stability drops precipitously to zero. Why, then, do natural ecosystems containing hundreds of species persist stably for millennia?
By simulating thousands of 20- to 50-species Allometric Trophic Networks across a wide spectrum of average consumer–resource body-mass ratios (R = Mconsumer / Mresource, ranging from 10−2 to 106), Ulrich Brose, Richard Williams, and Neo Martinez (2006) discovered a striking dynamical transition:
- Unstable Oscillations When Predators and Prey Have Equal Body Size (R ≤ 1): When consumers have the same body mass as their prey (R = 100 = 1) or are smaller than their hosts (R < 1, as in parasitoids), consumers and resources operate on the precise same fast metabolic timescale. Strong consumer ingestion rapidly depletes prey biomass, driving large-amplitude Rosenzweig–MacArthur limit cycles that cascade through the network and cause 50% to 75% of species to go extinct!
- Global Network Stabilization Above the Natural Body-Mass Ratio (R ≥ 101 to 102): Compiling a global database of >15,000 empirical trophic links, Brose et al. (2006) showed that in real terrestrial and aquatic food webs, invertebrate predators average ~10 to 100 times heavier than their prey (R ≈ 101–102) and ectothermic vertebrate predators (fishes) average ~100 to 1,000 times heavier (R ≈ 102–103). When these empirical body-mass ratios are entered into the Allometric Trophic Network equations, species persistence jumps from <30% to >85%–95%, and complex food webs become more stable as trophic levels are added!
- Two Biophysical Stabilizing Mechanisms:
- Allometric Timescale Separation: Because per-unit-biomass metabolic and consumption rates scale as M−0.25, a predator that is 10,000 times heavier than its prey (R = 104) turns over its biomass 10 times more slowly (10,000−0.25 = 0.1) than its prey. Fast-reproducing small prey recover from predation perturbations before slow-growing large predators can overshoot carrying capacity.
- Hump-Shaped Foraging Attack Rate (Rall et al., 2012): Biomechanical functional-response meta-analyses by Björn C. Rall, Ulrich Brose, and Owen L. Petchey showed that a predator's attack rate aik is a hump-shaped (Ricker/log-normal) function of the predator–prey mass ratio Mi/Mk: very small prey are unprofitable to pursue, and very large prey escape capture, distributing weak stabilizing links across the network.
Empirical Validation: The Lake Constance Seasonal Plankton Food Web (Boit et al., 2012)
To test whether an Allometric Trophic Network can quantitatively reproduce the real-world seasonal succession of a complex natural ecosystem without curve-fitting arbitrary Lotka–Volterra parameters, Alice Boit, Neo D. Martinez, Richard J. Williams, and Ursula Gaedke (2012) parameterized a 24-guild Allometric Trophic Network (spanning heterotrophic bacteria, 5 phytoplankton size classes, ciliates, rotifers, herbivorous cladocerans Daphnia, carnivorous copepods, and planktivorous fishes covering 10 orders of magnitude in body mass, from 10−13 g to 102 g) against 20 years of weekly empirical biomass measurements in Lake Constance (a deep pre-Alpine lake in Central Europe):
- Driven purely by seasonal solar irradiance, water temperature, mixed-layer phosphorus supply, and allometric body-mass scaling laws, the Lake Constance ATN model accurately reproduced 81% to 92% of the seasonal phase and amplitude trajectories across all 24 guilds—including the spring diatom bloom, the June "clear-water phase" driven by grazing Daphnia, the summer succession to grazing-resistant filamentous cyanobacteria and large dinoflagellates, and late-summer copepod–fish predation dynamics (Boit et al., 2012).
- Today, Allometric Trophic Networks serve as the standard mechanistic engine in Ecopath/Ecosim expansions, global marine fisheries size-spectrum models, and climate-warming food-web forecasts (where Arrhenius exponential warming of metabolic maintenance xi(T) ∝ exp(−Ea / kT) at Ea ≈ 0.65 eV predicts higher starvation risk for large apex ectotherms under warming; Binzer et al., 2012).
Frequently Asked Questions
How does an Allometric Trophic Network solve Robert May's complexity–stability paradox?
Robert May (1972) showed that food webs with random interaction strengths become unstable as species richness increases. Allometric Trophic Networks demonstrate that interaction strengths in real food webs are not random: because predators are typically 10 to 1,000 times larger in body mass than their prey, their mass-specific metabolic and consumption rates (M−0.25) are much lower and slower than the growth rates of their small prey. This allometric hierarchy creates many weak top-down links and fast prey recovery timescales that dampen predator–prey oscillations and stabilize large networks (Brose et al., 2006).
Why does climate warming threaten top predators in Allometric Trophic Networks?
According to the Boltzmann–Arrhenius biochemical kinetics incorporated into thermal ATN models (Binzer et al., 2012), metabolic respiration rates (xi) increase exponentially with temperature as exp(−Ea / kT) (activation energy Ea ≈ 0.65 eV) faster than maximum ingestion rates (yi) and half-saturation densities. As water or air temperature rises, large-bodied apex ectotherms (whose mass-specific ingestion margins are already narrow) experience "metabolic meltdown": their respiratory maintenance demand exceeds their maximum foraging intake, causing top-down extinction cascades from the top of the food web downward.
References
- Yodzis, P., & Innes, S. (1992). "Body size and consumer-resource dynamics." The American Naturalist, 139(6), 1151–1175. https://doi.org/10.1086/285380
- Brose, U., Williams, R. J., & Martinez, N. D. (2006). "Allometric scaling enhances stability in complex food webs." Ecology Letters, 9(11), 1228–1236. https://doi.org/10.1111/j.1461-0248.2006.00978.x
- Boit, A., Martinez, N. D., Williams, R. J., & Gaedke, U. (2012). "Mechanistic theory and modelling of complex food-web dynamics in Lake Constance." Ecology Letters, 15(6), 594–602. https://doi.org/10.1111/j.1461-0248.2012.01777.x
- Rall, B. C., Brose, U., Hartvig, M., Kalinkat, G., Schwarzmüller, F., Vucic-Pestic, O., & Petchey, O. L. (2012). "Universal temperature and body-mass scaling of feeding rates." Philosophical Transactions of the Royal Society B: Biological Sciences, 367(1605), 2923–2934. https://doi.org/10.1098/rstb.2012.0242