Food Web
A food web is a network of feeding relationships within an ecosystem — a directed graph in which nodes represent species or trophic groups and edges represent the transfer of energy and biomass from prey to predator. Unlike a food chain, which traces a single linear path from primary producers to apex predators, a food web captures the branching, reticulate reality of ecological feeding: most species consume multiple prey and are consumed by multiple predators, and the resulting network is typically complex, hierarchical, and robust to the loss of individual connections.
The food web is the oldest and most intuitive of ecological network representations. It predates the mathematical tools of graph theory, but it embodies the same insight: that the structure of interactions among species determines the dynamics, stability, and resilience of the ecosystem as a whole. A food web is not merely a description of who eats whom. It is a map of energy flow, a record of evolutionary history, and a diagnostic tool for predicting the consequences of species loss, invasion, or environmental change.
Structure of Food Webs
Food webs are organized by trophic level: primary producers (plants, algae, chemoautotrophs) convert solar or chemical energy into biomass; primary consumers (herbivores) eat producers; secondary consumers (carnivores) eat herbivores; and tertiary or apex consumers eat other carnivores. This hierarchical structure is not a strict ladder. Omnivory — the consumption of prey from multiple trophic levels — is common, and many species change their trophic position as they grow or as seasonal conditions shift. The 'level' of a species is a statistical property, not a fixed attribute.
The structure of a food web is characterized by several metrics:
Connectance — the fraction of possible feeding links that are realized — typically ranges from 0.03 to 0.3 in published food webs. Low connectance means that most species do not interact directly; high connectance means that interactions are dense. The relationship between connectance and stability was the subject of Robert May's influential 1973 analysis, which suggested that more connected webs are less stable. Subsequent work showed that real food webs are not random networks: they have modular structure, nestedness, and degree distributions that confer stability despite high connectance.
Chain length — the number of trophic links between a basal species and a top predator — is typically 3–6 in most ecosystems. Longer chains are rare because energy transfer efficiency is low (typically 10% or less), and the biomass available to support higher trophic levels diminishes exponentially. This is the ten-percent rule: only about 10% of the energy at one trophic level is converted to biomass at the next. The rule is approximate — aquatic systems sometimes show higher efficiencies — but it sets a hard constraint on food web depth.
Omnivory — the fraction of species that feed on multiple trophic levels — is ubiquitous in real food webs but was historically underrepresented in food web models. Omnivory creates loops in the food web: a species that eats both herbivores and plants is simultaneously a primary and secondary consumer. These loops complicate the dynamics: a perturbation that increases plant biomass can increase omnivore biomass directly (by providing more food) and indirectly (by increasing herbivore biomass, which also increases omnivore food supply). The net effect depends on the relative strengths of the direct and indirect pathways.
Food Webs and Stability
The stability of food webs — their capacity to maintain species composition and biomass distribution in the face of perturbation — has been one of ecology's central puzzles. Robert May's 1973 analysis of randomly assembled food webs suggested that stability decreases with complexity: more species and more connections make the web harder to stabilize. This result seemed to contradict the intuition that diverse ecosystems are more stable, an intuition supported by observations that monocultures are fragile and diverse forests are robust.
The resolution came with the recognition that real food webs are not random. They have:
- Modularity: species cluster into groups that interact intensely within the group and weakly with other groups. Modularity limits the propagation of perturbations: a disturbance that starts in one module is contained within it.
- Nestedness: specialist species feed on subsets of the prey consumed by generalist species. Nestedness provides redundancy: if a specialist is lost, the generalists that shared its prey can partially compensate.
- Degree distributions: most species have few prey and few predators, while a few species (the generalists and the apex predators) have many. This heterogeneity makes the web robust to random species loss (most species are specialists with limited network impact) but vulnerable to targeted loss of generalists or apex predators.
These structural properties mean that the relationship between diversity and stability is not linear. Up to a point, diversity increases stability by providing functional redundancy. Beyond that point, diversity can decrease stability by creating complex interaction loops that amplify perturbations. The optimal diversity depends on the environment: stable environments favor complex, diverse webs; variable environments favor simpler, more modular webs.
Trophic Cascades
A trophic cascade is the propagation of indirect effects through multiple trophic levels of a food web. The classic example is the reintroduction of wolves to Yellowstone National Park: wolf predation reduced elk populations, which released vegetation from grazing pressure, which altered stream geomorphology and increased fish habitat. The cascade is a network phenomenon: it arises because the food web is connected, and a perturbation at one node propagates along the edges to distant nodes.
Not all species removals produce cascades. Whether a perturbation propagates or is absorbed depends on the food web's topology. In a highly connected web, the loss of one predator can be compensated by other predators that share its prey. In a sparsely connected web, the loss of a predator can release its prey from all predation pressure, causing a population explosion that cascades through the web. The empirical observation that marine food webs show stronger cascades than terrestrial webs can be explained by their higher connectance and shorter path lengths: perturbations propagate more efficiently in well-connected networks.
Food Webs and Human Impact
Human activities restructure food webs in ways that are often invisible until they produce catastrophic outcomes. Overfishing removes apex predators, releasing prey populations and triggering cascades that restructure entire marine communities. Habitat fragmentation isolates modules, preventing the migration and recolonization that would maintain network connectivity. Invasive species add new nodes with novel connection patterns, rewiring the web in unpredictable ways. Climate change alters the phenology of species interactions, desynchronizing predator-prey relationships that evolved under stable seasonal cues.
The food web perspective reveals that these impacts are not independent. Overfishing and habitat fragmentation interact: a fragmented web that has lost its apex predators has no capacity to absorb the perturbation caused by an invasive species. Climate change and overfishing interact: a warmed ocean has altered metabolic rates that change the energy transfer efficiencies on which food web structure depends. The management of ecosystems requires understanding not just the direct effects of each human activity but their synergistic effects on food web structure.
The Food Web as a Diagnostic Tool
Beyond its descriptive role, the food web is a practical tool for ecosystem management. By mapping the feeding relationships in an ecosystem, managers can identify the species whose loss would have disproportionate network effects — the keystone species, the hubs, the bridges between modules. They can predict the consequences of species invasions by comparing the invader's trophic position to the positions of native species. They can design restoration strategies that rewire the web in desirable directions: reintroducing apex predators to restore top-down control, adding functional equivalents to replace extinct species, or removing invasive species that have become network hubs.
The food web is not a perfect tool. It captures feeding relationships but misses non-trophic interactions — competition, mutualism, facilitation, ecosystem engineering — that also structure communities. It is typically constructed from dietary data that are incomplete and biased toward conspicuous species. And it is static, while real food webs are constantly rewired by evolution, migration, and behavioral plasticity. But despite these limitations, the food web remains the most powerful framework we have for understanding how species are connected, how energy flows, and how perturbations propagate through ecological communities.