Biogeography, macroecology and species' traits mediate competitive interactions in the order Lagomorpha

In addition to abiotic determinants, biotic factors, including competitive, interspecific interactions, limit species' distributions. Environmental changes in human disturbance, land use and climate are predicted to have widespread impacts on interactions between species, especially in the order Lagomorpha due to the higher latitudes and more extreme environmental conditions they occupy. We reviewed the published literature on interspecific interactions in the order Lagomorpha and compared the biogeography, macroecology, phylogeny and traits of species known to interact with those of species with no reported interactions, to investigate how projected future environmental change may affect interactions and potentially alter species' distributions. Thirty-three lagomorph species have competitive interactions reported in the literature; the majority involve hares (Lepus sp.) or the eastern cottontail rabbit (Sylvilagus floridanus). Key regions for interactions are located between 30-50 degrees N of the Equator, and include eastern Asia (southern Russia on the border of Mongolia) and North America (north-western USA). Closely related, large-bodied, similarly sized species occurring in regions of human-modified, typically agricultural landscapes, or at high elevations, are significantly more likely to have reported competitive interactions than other lagomorph species. We identify species' traits associated with competitive interactions, and highlight some potential impacts that future environmental change may have on interspecific interactions. Our approach using bibliometric and biological data is widely applicable, and with relatively straightforward methodologies, can provide insights into interactions between species. Our results have implications for predicting species' responses to global change, and we advise that capturing, parameterizing and incorporating interspecific interactions into analyses (e.g. species distribution modelling) may be more important than suggested by the literature.