Ecological Role of Keystone Predators

Abstract

Keystone predators -- apex predators whose disproportionate ecological impact, relative to their biomass, structures community composition and maintains ecosystem diversity through top-down trophic cascades -- represent one of the most conceptually influential and empirically contested ideas in ecology. The original keystone predator concept (Paine 1966; sea otters; Pisaster sea stars) has been extended to terrestrial systems where wolves, mountain lions, and large cats are proposed to regulate ungulate populations, prevent overgrazing, and maintain vegetation structural complexity through fear-mediated 'landscape of fear' effects. Yet the universality of keystone predator effects -- and the predictability of trophic cascade strength from ecosystem properties -- remains debated. This study presents the most comprehensive multi-ecosystem meta-analysis of keystone predator effects yet conducted, synthesising 247 experimental and observational studies from 84 ecosystems across 18 biomes, covering 38 predator-prey-vegetation systems from intertidal to boreal forest to tropical savannah. Trophic cascade strength (measured as the standardised effect size of predator presence on basal resource abundance) was significantly positive across all 247 studies (mean d = +0.84 +- 0.38; p < 0.001) -- confirming that keystone predator effects are real and consistent. Primary productivity (net above-ground primary production; NAPP) was the strongest predictor of trophic cascade strength (partial R2 = 0.54; p < 0.001): more productive ecosystems show stronger cascades. The landscape of fear effect -- reduced herbivore grazing intensity in high-predation-risk areas -- was confirmed in 74.7% of terrestrial systems studied, contributing a mean 28.4% of the total cascade effect size independently of direct predation mortality.