Little consensus has emerged regarding how proximate and ultimate drivers such as productivity, disturbance and temperature may affect species richness and other aspects of biodiversity. Part of the confusion is that most studies examine species richness at a single spatial scale and ignore how the underlying components of species richness can vary with spatial scale. We provide an approach for the measurement of biodiversity that decomposes changes in species rarefaction curves into proximate components attributed to: (a) the species abundance distribution, (b) density of individuals and (c) the spatial arrangement of individuals. We decompose species richness by comparing spatial and nonspatial sample‐ and individual‐based species rarefaction curves that differentially capture the influence of these components to estimate the relative importance of each in driving patterns of species richness change. We tested the validity of our method on simulated data, and we demonstrate it on empirical data on plant species richness in invaded and uninvaded woodlands. We integrated these methods into a new r package (mobr). The metrics that mobr provides will allow ecologists to move beyond comparisons of species richness in response to ecological drivers at a single spatial scale toward a dissection of the proximate components that determine species richness across scales.

Using coveragebased rarefaction to infer non-random species distributions.

Biodiversity metrics often integrate data on the presence and abundance of multiple species. Yet our understanding of covariation between changes to the numbers of individuals, the evenness of species relative abundances, and the total number of species remains limited. Using individual-based rarefaction curves, we show how expected positive relationships among changes in abundance, evenness and richness arise, and how they can break down. We then examined interdependencies between changes in abundance, evenness and richness in more than 1100 assemblages sampled either through time or across space. As predicted, richness changes were greatest when abundance and evenness changed in the same direction, and countervailing changes in abundance and evenness acted to constrain the magnitude of changes in species richness. Site-to-site differences in abundance, evenness, and richness were often decoupled, and pairwise relationships between these components across assemblages were weak. In contrast, changes in species richness and relative

A multiscale framework for disentangling the roles of evenness, density, and aggregation on diversity gradients.