POLLINATOR SPECIES RICHNESS TRENDS: NO DECELERATING DECLINES IN DUTCH BEES AND BUMBLEBEES

Temporal trends (1946-2013) in the species richness of wild bees from the Netherlands are analysed. We apply two methods to estimate richness change which both incorporate models for sampling effects and detection probability. The analysis is repeated for records with specimens deposited in collections, and a subset restricted to spatial grid cells that have been sampled repeatedly across three periods. When fitting non-linear species accumulation curves to species numbers, declines are inferred for bumblebees and at most limited declines for other bees. Capture-recapture analysis applied to species encounter histories infers a constant colonization rate per year and constant (bumblebees) or decreasing (other) local species survival. However, simulations suggest that the method estimates time trends in survival with a negative bias. Species richness trends predicted by the second approach are a 10% reduction in non-Bombus species richness and 29% fewer Bombus species since 1946, comparable to the predictions of the first approach. Neither analysis provides reliable evidence that decelerating declines in species richness occur in these taxa. Therefore we should not infer decelerating declines in pollinator species richness in N-W Europe as previously claimed.