Measuring biodiversity across spatial scales in a raised bog using a novel paired‐sample diversity index

Abstract

Summary 1. Biodiversity is an inherently spatial phenomenon. It is determined by environmental heterogeneity and by spatially structured ecological processes such as disturbance, competition and dispersal. However, conventional biodiversity indices are based on discrete samples or pooled sets of samples without sufficient consideration of the spatial relationships between the samples. 2. Here we describe a new method for spatial analysis of species diversity, based on a paired‐sample version of the widely used Gini–Simpson diversity index and its numbers equivalent. The index and its numbers equivalent are plotted as a function of lag distance between two samples along spatial and/or environmental gradients. 3. We demonstrate the potential of this approach by applying it to two transects of fine‐scale (5 × 5 cm quadrat) vegetation data from sites with contrasting hydrology within a raised bog, where the location of each quadrat is accurately recorded and the height of the bog surface above the water table is measured using a terrestrial laser scanner. 4. Both transects have similar alpha‐diversity as measured using the Gini–Simpson index, and the transition between alpha‐ and gamma‐diversity occurs at similar length scales, suggesting that species aggregate at similar scales along both transects. However, the transect from the central bog dome has higher gamma‐diversity than that from the bog margin, and shows more marked significant spatial structure at a length scale of 135–140 cm, corresponding to the typical hummock–hollow microtopography at the site. We show that beta‐diversity at both transects can be attributable to both species clustering along the hydrological gradient, consistent with niche partitioning, as well as independent spatial aggregation of species that is not explained by hydrology. 5. Synthesis. The paired‐sample diversity index described here is a potentially useful tool in detecting and attributing patterns of beta‐diversity along both spatial and environmental gradients.