Skip to main content
Dryad

Data from: Effects of predator recovery and climate change on the long-term demography of a flagship herbivore

Loonam, Kenneth1; Brown, Casey2; Rowland, Mary3; Wisdom, Michael3; Clark, Darren2; Levi, Taal1

Published Sep 03, 2026 on Dryad. https://doi.org/10.5061/dryad.msbcc2gds

Data files

Sep 03, 2026 version files 16.17 KB

Click names to download individual files

Abstract

Large herbivores play important roles in ecosystems across the globe, but limitations to study and inference have left gaps in predictive theories of their population dynamics, particularly following carnivore recoveries. Studies using spatial variation have explored the broad patterns of top-down and bottom-up forces in large herbivores. However, temporal changes in a population provide the most direct evidence for predicting future effects, and long-term studies spanning variation in top-down and bottom-up forces remain rare.

We used 36 years of data on elk (Cervus canadensis) in the western United States in an integrated population model to test the effects of top-down (Puma concolor recovery) and bottom-up (drought) forces on pregnancy rates and age-based survival across a range of elk densities in the context of human harvest. We then quantified the proportion of variance in population growth attributable to those forces relative to the influences of harvest, age-structure, and unmodelled variation in demographic rates.

Both top-down and bottom-up forces had measurable effects on elk population dynamics, with puma recovery reducing survival in the first two years of life and drought reducing pregnancy probability and survival at all ages. We also found negative density dependence in pregnancy, consistent with bottom-up control, and positive density dependence in survival of young, consistent with predator swamping. Harvest explained the greatest proportion of variation in population growth (26%). Of the non-harvest parameters, drought severity had the largest effect, explaining as much variation (16%) as all of the unmodelled changes in demographic rates. Puma recovery explained 13% of the variation, and density dependence explained 8%.

Without more long-term studies spanning sufficient ecological variation, the predictive abilities of large herbivore ecology will remain limited. With ongoing anthropogenic changes to the forces acting on large herbivores around the world, improving predictions will be invaluable to conservation.