Literature search and inclusion criteria
We conducted a literature search in the ISI Web of Science and Scopus databases (until 28 December 2021), using a combination of different keywords depending on our research questions. To find publications investigating the effects of urbanisation on pollinator diversity, we used the keywords [urban* OR city OR cities OR town] AND [pollinat*] AND [“species richness” OR “species diversity” OR abundance OR density OR Assemblage]. We replaced the last combination with [trait OR phenology OR “body size” OR nest* OR sociality OR diet] to find publications on the effects of urbanisation on pollinator functional traits. To search for studies that investigate the effects of urbanisation on pollination, we changed the last combination to [”pollinat* service” OR ”plant reproduc*” OR seed OR fruit]. This search yielded 719, 708 and 504 publications in Web of Science, and 454, 366 and 348 in Scopus, respectively. In addition, we also surveyed recent reviews (e.g., Fenoglio et al. 2020; Wenzelet al. 2020) for relevant publications.
After removing duplicates, non-English papers and review papers, our survey resulted in a total of 1 205 publications. The publications were filtered by reading the title, abstract and full text according to the following two criteria: (1) studies that reported pollinator abundance and/or richness, phenology, nesting behavior, diet specialization, body size, sociality, fruit set, seed set and visitation rates along urbanisation gradients, or comparisons of urbanisation intensity within urban, urban-rural or urban-natural land use categories; and (2) provided numerical data and reported means, measures of variance and sample sizes for different categories of comparison, or regression or correlation coefficients for urbanisation gradient studies. A total of 133 publications met our criteria (Appendix S2).