Quantifying Spatial Pattern With Evenness Indices

Abstract

Quantifying the spatial distributions of organisms in simple and meaningful ways is important for understanding the ecology and habitat needs of species subject to anthropogenic disturbances. Most multi‐species conservation and management programs do not yet account for the variation of space‐use patterns exhibited by species preferring the same habitat type. To measure species' space‐use patterns as a step toward determining habitat needs, we suggest using evenness indices. Although commonly used in characterizing communities (i.e., as a measure of species diversity), these indices are suitable for characterizing many other ecological patterns, including patterns of site use by individuals. We investigate the statistical properties of five indices (Camargo's index of evenness, E′; Simpson's index of evenness, E1/D^; Lloyd's index of mean crowding, J; Smith‐Wilson index Evar; and Dispersion index, DL, a variant of the Shannon diversity index) to evaluate their utility for quantifying broad‐scale spatial patterns of migratory shorebirds. We use a Monte Carlo simulation approach to compare these indices for their (1) ability to characterize a wide range of spatial patterns (from even to patchy); (2) ability to discriminate among distributions; and (3) robustness to incomplete sampling. In addition, we compare the ability of these indices to characterize spatial dispersion for four species of shorebirds migrating through the United States: Killdeer (Charadrius vociferus), Semipalmated Plover (C. semipalmatus), Sanderling (Calidris alba), and Red Knot (C. canutus). Overall, we recommend Camargo, Simpson, and Lloyd indices for quantifying spatial dispersion. All three indices gave precise and unbiased estimates once 35 or more sites (3.5% of 1000 simulated sites) were sampled randomly, regardless of the degree of patchiness. Nonrandom sampling resulted in biased estimates until a much greater proportion of sites (at least 30–50%) were sampled, highlighting the importance of site selection in sampling programs. Our analysis of shorebird spatial patterns revealed that evenness indices discriminate well among the species we considered, ranking Killdeer as most dispersed and Red Knot as most aggregated. Our results agree with earlier (independent) assessments of these species' migration strategies and illustrate how simple univariate metrics may be useful tools for characterizing complex spatial patterns.

Spatial trends in community richness, diversity, and evenness across rocky intertidal environmental stress gradients in eastern Canada

“…Simpson's evenness index ranges between 0 and 1 and is relatively unaffected by the rare species in the quadrat and by spatial distribution (Krebs 1999, Payne et al 2005. Data analysis.…”

Species richness and diversity across rocky intertidal elevation gradients in Helgoland: testing predictions from an environmental stress model

“…Simpson's evenness index ranges between 0 and 1 and is relatively unaffected by the rare species in the quadrat and by spatial distribution (Krebs 1999;Payne et al 2005). …”

Measuring homogeneity of ethno-medicinal knowledge related with different corporeal system: A top down approach

“…These previous studies had reported the relative importance value of different arid and semi arid plants, their sensitivity index, their economic assessment and conservation priorities, different ecosystem services (cultural, provisional and regulatory) provided by them, quantification of the types and erosion of cultural transmission, medicinal plants primarily useful for various diseases related to different corporeal system like central nervous, gastro-intestinal, reproductive systems and the market analysis of certain plants in relation to their medicinal properties. Evenness: Evenness based on Shannon and Weaver and Simpson indexes were quantified according to Payne et al, 2005). While Simpson's index of diversity, measures the probability that two individuals selected at random from a sample will belong to same species were quantified according to Mathur (2014a).…”