Search

Ecology Letters (2010) 13: 506–515 The translocation of species beyond their native range is a major threat to biodiversity. Invasions by tree-feeding insects attacking native trees and the colonization of introduced trees by native insects result in new insect–tree relationships. To date there is uncertainty about the key factors that influence the outcome of these novel interactions. We report the results of a meta-analysis of 346 pairwise comparisons of forest insect fitness on novel and ancient host tree species from 31 publications. Host specificity of insects and phylogenetic relatedness between ancient and novel host trees emerged as key factors influencing insect fitness. Overall, fitness was significantly lower on novel host species than on ancient hosts. However, in some cases, fitness increased on novel hosts, mainly in polyphagous insects or when close relatives of ancient host trees were colonized. Our synthesis enables greatly improved impact prediction and risk assessment of biological invasions. [ABSTRACT FROM AUTHOR]

Biodiversity loss from plant communities is often acknowledged to affect primary production but little is known about effects on herbivores. We conducted a meta-analysis of a worldwide data set of 119 studies to compare herbivory in single-species and mixed forests. This showed a significant reduction of herbivory in more diverse forests but this varied with the host specificity of insects. In diverse forests, herbivory by oligophagous species was virtually always reduced, whereas the response of polyphagous species was variable. Further analyses revealed that the composition of tree mixtures may be more important than species richness per se because diversity effects on herbivory were greater when mixed forests comprised taxonomically more distant tree species, and when the proportion of non-host trees was greater than that of host trees. These findings provide new support for the role of biodiversity in ecosystem functioning across trophic levels. [ABSTRACT FROM AUTHOR]

Human impacts such as habitat loss, climate change and biological invasions are radically altering biodiversity, with greater effects projected into the future. Evidence suggests human impacts may differ substantially between terrestrial and freshwater ecosystems, but the reasons for these differences are poorly understood. We propose an integrative approach to explain these differences by linking impacts to four fundamental processes that structure communities: dispersal, speciation, species‐level selection and ecological drift. Our goal is to provide process‐based insights into why human impacts, and responses to impacts, may differ across ecosystem types using a mechanistic, eco‐evolutionary comparative framework. To enable these insights, we review and synthesise (i) how the four processes influence diversity and dynamics in terrestrial versus freshwater communities, specifically whether the relative importance of each process differs among ecosystems, and (ii) the pathways by which human impacts can produce divergent responses across ecosystems, due to differences in the strength of processes among ecosystems we identify. Finally, we highlight research gaps and next steps, and discuss how this approach can provide new insights for conservation. By focusing on the processes that shape diversity in communities, we aim to mechanistically link human impacts to ongoing and future changes in ecosystems. [ABSTRACT FROM AUTHOR]