Because most large, terrestrial mammalian predators have already been lost from more than 95-99% of the contiguous United States and Mexico, many ecological communities are either missing dominant selective forces or have new ones dependent upon humans. Such large-scale manipulations of a key element of most ecosystems offer unique opportunities to investigate how the loss of large carnivores affects communities, including the extent, if any, of interactions at different trophic levels. Here, we demonstrate a cascade of ecological events that were triggered by the local extinction of grizzly bears (Ursus arctos) and wolves (Canis lupus) from the southern Greater Yellowstone Ecosystem. These include (1) the demographic eruption of a large, semi-obligate, riparian-dependent herbivore, the moose (Alces alces), during the past 150 yr; (2) the subsequent alteration of riparian vegetation structure and density by ungulate herbivory; and (3) the coincident reduction of avian neotropical migrants in the impacted willow communities. We contrasted three sites matched hydrologically and ecologically in Grand Teton National Park, Wyoming, USA, where grizzly bears and wolves had been eliminated 60-75 yr ago and moose densities were about five times higher, with those on national forest lands outside the park, where predation by the two large carnivores has been replaced by human hunting and moose densities were lower. Avian species richness and nesting density varied inversely with moose abundance, and two riparian specialists, Gray Catbirds (Dumetella carolinensis) and MacGillivray's Warblers (Oporornis tolmiei), were absent from Park riparian systems where moose densities were high. Our findings not only offer empirical support for the top-down effect of large carnivores in terrestrial communities, but also provide a scientific rationale for restoration options to conserve biological diversity. To predict future impacts, whether overt or subtle, of past management, and to restore biodiversity, more must be known about ecological interactions, including the role of large carnivores. Restoration options with respect to the system that we studied in the southern Greater Yellowstone Ecosystem are simple: (1) do nothing and accept the erosion of biological diversity, (2) replace natural carnivores with human predation, or (3) allow continued dispersal of grizzly bears and wolves into previously occupied, but now vacant, habitat. Although additional science is required to further our understanding of this and other terrestrial systems, a larger conservation challenge remains: to develop public support for ecologically rational conservation options.

Because most large, terrestrial mammalian predators have already been lost from more than 95–99% of the contiguous United States and Mexico, many ecological communities are either missing dominant selective forces or have new ones dependent upon humans. Such large‐scale manipulations of a key element of most ecosystems offer unique opportunities to investigate how the loss of large carnivores affects communities, including the extent, if any, of interactions at different trophic levels. Here, we demonstrate a cascade of ecological events that were triggered by the local extinction of grizzly bears (Ursus arctos) and wolves (Canis lupus) from the southern Greater Yellowstone Ecosystem. These include (1) the demographic eruption of a large, semi‐obligate, riparian‐dependent herbivore, the moose (Alces alces), during the past 150 yr; (2) the subsequent alteration of riparian vegetation structure and density by ungulate herbivory; and (3) the coincident reduction of avian neotropical migrants in the impacted willow communities. We contrasted three sites matched hydrologically and ecologically in Grand Teton National Park, Wyoming, USA, where grizzly bears and wolves had been eliminated 60–75 yr ago and moose densities were about five times higher, with those on national forest lands outside the park, where predation by the two large carnivores has been replaced by human hunting and moose densities were lower. Avian species richness and nesting density varied inversely with moose abundance, and two riparian specialists, Gray Catbirds (Dumetella carolinensis) and MacGillivray's Warblers (Oporornis tolmiei), were absent from Park riparian systems where moose densities were high. Our findings not only offer empirical support for the top‐down effect of large carnivores in terrestrial communities, but also provide a scientific rationale for restoration options to conserve biological diversity. To predict future impacts, whether overt or subtle, of past management, and to restore biodiversity, more must be known about ecological interactions, including the role of large carnivores. Restoration options with respect to the system that we studied in the southern Greater Yellowstone Ecosystem are simple: (1) do nothing and accept the erosion of biological diversity, (2) replace natural carnivores with human predation, or (3) allow continued dispersal of grizzly bears and wolves into previously occupied, but now vacant, habitat. Although additional science is required to further our understanding of this and other terrestrial systems, a larger conservation challenge remains: to develop public support for ecologically rational conservation options.