Food Webs: Lotka-Volterra Cycles and Trophic Cascades

A network where every edge is a meal

A food web is a directed graph of who eats whom in an ecosystem — plants feeding herbivores, herbivores feeding predators, predators sometimes feeding larger predators. Unlike a simple food chain, real food webs are tangled: most species eat and are eaten by several others, and that redundancy is what determines whether removing one species collapses the system or barely registers.

Lotka-Volterra: the two-species starting point

The classical model of a single predator-prey pair, developed independently by Alfred Lotka and Vito Volterra in the 1920s, is a pair of coupled differential equations:

dx/dt = α·x - β·x·y     (prey growth minus predation)
dy/dt = δ·x·y - γ·y     (predator growth from eating prey, minus death)

where x is prey population, y is predator population, α is the prey's intrinsic growth rate, γ is the predator's death rate, and β, δ scale how often encounters between predator and prey translate into predation and reproduction. This pair of equations, on its own, produces closed cycles: prey boom, predators follow and eventually overshoot, prey crash, predators starve and crash too, and prey recover to start the cycle again — indefinitely, with neither population settling to a fixed point.

Why real food webs rarely oscillate that cleanly

Classical Lotka-Volterra cycles are famously fragile: add any realistic ingredient — a carrying capacity that caps prey growth, a saturating (Type II or III) predation response instead of a linear one, or a third species — and the neat closed orbits typically damp into a stable equilibrium or destabilize into extinction, rather than persisting as perfect cycles. Real ecosystems that do show boom-bust cycles, like the classic Canada lynx and snowshoe hare fur-trapping records, roughly follow the qualitative Lotka-Volterra pattern but never trace the mathematically perfect closed loops the bare two-species equations predict.

What happens when you remove a species

In a network with many species, removing one node can ripple outward in ways that are hard to predict from the local picture alone. A keystone species — one whose ecological effect is disproportionate to its own population size, like a top predator that keeps a herbivore in check — can trigger a trophic cascade when removed: the herbivore below it booms unchecked, over-grazes the plants below that, and the effect keeps propagating down the chain, sometimes reshaping the entire ecosystem's structure. The reintroduction of wolves to Yellowstone in 1995, and the subsequent recovery of aspen and willow stands as elk browsing pressure fell, is the most cited real-world trophic cascade.

Why some webs are more resilient than others

Food webs with more redundancy — several species filling similar ecological roles, so a predator or herbivore has more than one thing it can eat — tend to absorb a single species loss without collapsing, because remaining links can partially compensate. Webs with a narrow "waist" of few dominant connector species are far more fragile: removing one of those hubs can fragment the entire network into disconnected pieces, similar to how removing a highly-connected node collapses connectivity in any other complex network. This is why conservation biology pays special attention to keystone and hub species rather than treating every species loss as equally consequential.

Frequently asked questions

Try it live

Everything above runs in your browser — open Food Web and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.