The impact of competition on global-change induced range shifts in birds

Abstract

Species do not exist in isolation but among a host of others with which they interact. These interactions can influence species distributions both positively and negatively, at a variety of scales. Understanding how species interactions influence distributions is of particular importance in this current era of rapid anthropogenic change, in which species distributions and assemblages are shifting. Species interactions may influence the ease with which species track their suitable climate niche and colonise new areas, they may facilitate or preclude range expansion. Throughout this thesis, we focus primarily on competitive interactions, more specifically, exploitative competition, interspecific territoriality and hybridisation. We present large-scale, comparative, species-pair analyses across North America and Europe, spanning a large taxonomic breadth of passerines and non-passerines.

We begin by focusing on the predictors of hybridisation, which can have pronounced effects on species coexistence. This is the first time hybridisation has been studied at such a large scale in a species-pair framework, and incorporating phylogenetic relationships. We find strong support for the mistaken identity hypothesis, in that similar, and more closely related species are more likely to hybridise than dissimilar species. We find that as species encounter each other more with greater range and habitat overlap, their likelihood of hybridisation increases. As species ranges change, novel interactions can form, and these interactions can be costly. For instance, hybridisation between natives and non-natives is a widespread conservation problem, and as such, being able to accurately predict interactions between species that don’t currently coexist has important conservation ramifications. We compare the predictive performance of two modelling methods when predicting ‘unknown’ hybridisation events.
Understanding how species interactions have influenced coexistence in the past is vital to understand how they may influence coexistence in the future. We hindcast the effect of competition on historical range dynamics in North America and Europe using two large-scale citizen science datasets (the North American Breeding Bird Survey and the Pan-European Common Bird Monitoring Scheme). We find that interspecifically territorial pairs increased their fine-scale habitat overlap more than non-territorial pairs in North America. We believe this is the result of interspecific territoriality creating spatial segregation within habitats, which subsequently facilitates resource partitioning, allowing similar species to coexist.

We then examine the impact of species interactions on current species distributions using the predictions from Species Distribution Models. We calculate measures of geographic avoidance across North America and Eurasia. Geographic avoidance quantifies exclusion patterns as the proportion of a pair’s predicted range that is realised in overlapping areas compared to the proportion realised in non-overlapping areas. We found that most pairs exhibited some level of avoidance across their ranges. However, we found no evidence for the role of competitive interactions in the observed patterns of avoidance. Finally, we use high-resolution Joint Species Distribution Models across a large spatial extent and attempt to detect the effect of biotic interactions in the residual associations of species pairs. We find that associations between species are overwhelmingly positive and find little evidence that JSDMs can detect species interactions in their residual associations. Instead, we find evidence that unaccounted-for environmental covariates are responsible for observed patterns of association between species. This research greatly expands our knowledge of avian hybridisation and the impact of species interactions on the distributions and coexistence of birds at a variety of geographic scales.