Estimating wild bee population size with validated distance sampling

Pollinators are declining globally, highlighting the urgent need for accurate methods to monitor their densities and abundances. Distance sampling (DS) is a promising, non-invasive method for estimating insect population abundances, but its estimates have not been compared with independent counts of population size to validate DS efficacy for insects. We addressed this methodological gap on Giannutri Island (Italy), where two strongly declining pollinator populations of Anthophora dispar and Bombus terrestris coexist with managed honey bees (Apis mellifera). DS field surveys, targeting both managed and wild bees, were conducted across 41 transects spanning five land cover types over 12 sampling days. We validated this method by verifying that the number of foraging honey bees estimated with DS is consistent with the literature reported proportion of workers outside the hives (~12.5%–20%) of the known reference population of managed colonies. Using DS, we also verified: (1) a strong consistency in abundance estimates across consecutive days; (2) significant effects of weather conditions on detected abundances and temporal trends matching the known phenology of adults in both wild species; (3) marked differences in densities across land cover types consistent with the known link between pollinator abundances and vegetation structures. In conclusion, DS offers a practical method for obtaining reliable estimates of population size for midsized to large insects when individuals can be readily observed and identified. While its applicability to smaller species and closed habitats may thus be limited, it represents a valuable tool where traditional methods are impractical.