Predictors of food web resistance to environmental change

Impacts of environmental change on ecosystems are seemingly highly context-dependent and contingent on details. Here, we provide mathematical and numerical evidence that there is some generality to the resistance of food webs to species loss. To this end, we mathematically analyse simple food web models, run simulations with randomly wired food webs, develop a spatially explicit meta-community model for realistic macroinvertebrate communities inhabiting European streams, and analyse field data. These approaches jointly support three general rules for food web resistance: (1) all else equal, predators are more vulnerable than prey, and predator diversity declines before prey diversity; (2) food webs with higher prey diversity tend to be less resistant to environmental change; and (3) food web resistance decreases with the mean taxon tolerance and reaches a maximum at an optimal ratio of predator to prey tolerance. Our findings elucidate characteristics that make food webs vulnerable, fostering more effective conservation practices.