Aim We tested whether the geographic variation in the proportion of beta diversity attributed to nestedness or turnover components was explained by the effect of past glaciation events. Specifically, we tested the hypothesis that most of the beta diversity in regions retaining ice until recent periods was due to nestedness. Additionally, we tested whether the variation was influenced by thermal tolerance and the dispersal ability of species.
Location This study analysed data from the New World.
Methods We used presence/absence data for amphibians, birds and mammals of the New World. We calculated beta diversity among each 1°× 1° cell and the adjacent cells using the Sorensen dissimilarity index that expresses the total beta diversity. Furthermore, we partitioned it into turnover and nestedness components. The relative importance of the two latter components was expressed as the proportion of total beta diversity explained by nestedness (βratio). We calculated the correlation between βratio and the time each cell was free of ice since the last glaciation (cell age). To control the effects of spatial autocorrelation, we calculated geographically effective degrees of freedom.
Results The proportion of beta diversity attributed to nestedness was negatively correlated with cell age. Moreover, this effect was stronger for amphibians than mammals, and stronger for mammals than birds.
Main conclusions Our results are in accordance with the hypothesis that the nestedness component of beta diversity is more important in areas affected by glaciations until recent time. The beta diversity in high latitudes is the result of past extinctions and recent recolonization, which result in higher levels of nestedness. This process is more evident for vertebrates with lower dispersal ability and lower temperature tolerance.