Deconstructing richness patterns by commonness and rarity reveals bioclimatic and spatial effects in black fly metacommunities

Abstract

Deconstructing biological communities by grouping species according to their commonness or rarity might improve our understanding about the processes driving variation in biological communities. Such an approach considers differences among organisms and emergent ecological patterns. In this study, we addressed the relative role of spatial and large-scale bioclimatic variables along a commonness and rarity gradient using Simuliidae (Diptera) species richness. A database of species occurrences at 459 locations in Brazil was used to estimate the distribution of 58 simuliid species. Total species richness at each location was estimated first using all occurrences and then by removing one species at a time, following a commonest to rarest gradient (CtR) and vice-versa (RtC). Partial regression analysis was used to test the influence of sets of bioclimatic (E) and spatial (S) variables for Simuliidae species richness across both CtR and RtC gradients. In the CtR gradient, the pure spatial component alone explained between 40% and 60% of the variation in simuliid richness when the total number of species was greater than ~35. After removal of the 35th most common species, the model fit decreased sharply reaching nearly zero when only rare species were present. Variation explained by the shared component E + S decreased continuously along the CtR gradient. The relative role of predictor variables on the RtC gradient was similar to CtR gradient. However, removing the rare species first did not change which components best explained species richness. Our gradual deconstructive approach revealed that common species contribute more to species richness variation than rare species, and that the role of predictors in explaining this pattern cannot be untangled by analysing richness of rare and common species in a categorical way.