Urban winners and losers: A meta-analysis of the effects of urbanization on terrestrial arthropods

Background/Question/Methods

Human land use change is one of the leading drivers of biodiversity loss worldwide, and urbanization represents one of the most intense and longest-lasting of these modifications to natural systems. Urbanization gradients typically extend from relatively natural habitats in the rural regions around a city to the highly built city center, encompassing extensive changes to habitat structure, resource availability, and abiotic conditions. The overall impact of urbanization on species richness appears to be negative for many vertebrate communities. Reports of the effects of urbanization on arthropods have been more variable, however, ranging from outbreaks of arthropod pests to local extinctions of native butterfly populations. To understand the overall impacts of urbanization on terrestrial arthropods and to identify factors underlying the variation in species’ responses, we conducted a meta-analysis of the published literature. We compiled a dataset of relevant studies using key word searches on Web of Science, and we supplemented this list with literature cited sections of identified papers and recent reviews. For each paper, we recorded characteristics pertaining to the study (location, city size, study duration), taxon (taxonomic resolution, trophic level, diet breadth, body size), and results.

Results/Conclusions

To date, we have identified 573 studies of the effects of urbanization on terrestrial arthropods, from 60 publications. Five continents are represented in this dataset, although North America and Europe are over-represented. The dataset includes responses of 326 arthropod species in 98 families, from 15 orders. Studies were conducted on individual species densities, assemblage density, and assemblage richness. We calculated the effect size for each study using the log response ratio, LRR, where a positive LRR represents higher densities in the urban core and a negative LRR indicates higher densities in the rural regions. Although the overall mean effect size for the density of individual species was near zero (LRR = -0.21 +/- 0.09 SE), there were clear urban winners (maximum LRR, 6.08) and losers (minimum LRR, -4.82). To understand the extensive variation in effect size among species, we developed a set of candidate linear mixed models, including model terms pertaining to species traits and study characteristics. Taking ground beetles (Coleoptera: Carabidae) as an example, we found that body size and city size were the most important determinants of the urbanization effect on species densities. How other taxa respond is part of our ongoing analysis of this dataset.