Abstract

According to classic theory, species' population dynamics and distributions are less influenced by species interactions under harsh climatic conditions compared to under more benign climatic conditions. In alpine and boreal ecosystems in Fennoscandia, the cyclic dynamics of rodents strongly affect many other species, including ground-nesting birds such as ptarmigan. According to the ‘alternative prey hypothesis’ (APH), the densities of ground-nesting birds and rodents are positively associated due to predator–prey dynamics and prey-switching. However, it remains unclear how the strength of these predator-mediated interactions change along a climatic harshness gradient in comparison with the effects of climatic variation. We built a hierarchical Bayesian model to estimate the sensitivity of ptarmigan populations to interannual variation in climate and rodent occurrence across Norway during 2007–2017. Ptarmigan abundance was positively linked with rodent occurrence, consistent with the APH. Moreover, we found that the link between ptarmigan abundance and rodent dynamics was strongest in colder regions. Our study highlights how species interactions play an important role in population dynamics of species at high latitudes and suggests that they can become even more important in the most climatically harsh regions.

Highlights

  • Climatic variability and species interactions are two key drivers influencing the spatial and temporal patterns in the distribution and abundance of organisms [1,2]

  • We outlined two different hypotheses about how the strength of species interactions might vary along a climatic gradient, and season temperature spatial variation winter temp spring temp temporal variation winter temp spring temp

  • Our results suggest that the impacts of predator-mediated interactions on willow ptarmigan increase along the climate gradient, indicated by more strongly positive impacts of rodent dynamics in colder areas

Read more

Summary

Introduction

Climatic variability and species interactions are two key drivers influencing the spatial and temporal patterns in the distribution and abundance of organisms [1,2]. We included a submodel for rodent occurrence ( presence/absence, modelled as a Bernoulli distribution) within our hierarchical model for ptarmigan This submodel was based on a series of random intercept terms (for survey region [SR], year [Y ], as a factor and their interaction) to flexibly predict the interannual variability in royalsocietypublishing.org/journal/rspb Proc. The model predicted the probability of rodent occurrence at the level of the survey region rather than at the transect-level, which meant smoothing over local reporting and ecological variation This choice was based on the assumptions that many predators (especially birds of prey) forage over reasonable distances and are not likely affected by fine-scale transect-level variation. Code for the fitted model is available in the electronic supplementary material, C

Results
Discussion
Findings
51. Cornulier T et al 2013 Europe-wide dampening of
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call