A global assessment of climatic niche shifts and human influence in insect invasions

Aim: Non-native invasive insects have major impacts on ecosystem function, agricultural production and human health. To predict the geographical distributions of these species, correlative ecological niche models (ENMs) are typically used. Such methods rely on assumptions of niche conservatism, although there is increasing evidence that many species undergo niche shifts during invasions. The magnitude and direction of niche shifts, however, is likely to vary within and between taxonomic groups, highlighting that an assessment of potential niche shifts in such insects is required. Location: Global. Time period: Current. Major taxa studied: Insects. Methods: We compile a novel database of 22 globally invasive, non-native insect species and test for niche expansion and unfilling across this group. We examine if factors such as the native range size, poleward shifts and human influence may be associated with observed niche changes. Finally, we construct ENMs and examine the reliability of their predictions in light of our niche shift results. Results: Niche expansion was apparent in 12 of the 22 species, suggesting that altered species-climate relationships during invasion is common for this group. Likewise, niche unfilling occurred in 15 of the species. Increasing human disturbance (combining human population, transport networks and land use) explained 40% of observed niche expansions and 54% of incidents of niche unfilling. Niche metrics and ENM performance were sensitive to the choice of background extents. Main conclusions: Many non-native insects expand into new climates in their invasive ranges. The prevalence of niche unfilling across this group suggests climate disequilibrium and the potential for further range expansion. Non-native invasive insects tend to invade areas with similar human disturbance to their native range, and habitat accessibility appears important for these species to achieve their full invasion range potential. Ideally, ENMs should not be used in isolation for this group, but should be coupled with other methods or experiments to test for potential niche change.