An extensive comparison of species-abundance distribution models.

Elita Baldridge,Ethan P White,Xiao Xiao,David J Harris

doi:10.7717/peerj.2823

Elita Baldridge, Ethan P White + Show 2 more

Open Access

https://doi.org/10.7717/peerj.2823

Copy DOI

Abstract

A number of different models have been proposed as descriptions of the species-abundance distribution (SAD). Most evaluations of these models use only one or two models, focus on only a single ecosystem or taxonomic group, or fail to use appropriate statistical methods. We use likelihood and AIC to compare the fit of four of the most widely used models to data on over 16,000 communities from a diverse array of taxonomic groups and ecosystems. Across all datasets combined the log-series, Poisson lognormal, and negative binomial all yield similar overall fits to the data. Therefore, when correcting for differences in the number of parameters the log-series generally provides the best fit to data. Within individual datasets some other distributions performed nearly as well as the log-series even after correcting for the number of parameters. The Zipf distribution is generally a poor characterization of the SAD.

Highlights

The species abundance distribution (SAD) describes the full distribution of commonness and rarity in ecological systems
Data for trees, birds, butterflies and mammals was compiled by White, Thibault & Xiao (2012) from six data sources: the US Forest Service Forest Inventory and Analysis (FIA; USDA Forest Service, 2010), the North American Butterfly Association’s North American Butterfly Count (NABC; North American Butterfly Assoc, 2009), the Mammal Community Database (MCDB; Thibault et al, 2011), Alwyn Gentry’s Forest Transect Data Set (Gentry; Phillips & Miller, 2002), the Audubon Society Christmas Bird Count (CBC; National Audubon Society, 2002), and the US Geological Survey’s North American Breeding Bird Survey (BBS; Pardieck, Ziolkowski Jr & Hudson, 2014)
AICc approximately removes this bias by penalizing models with more degrees of freedom

Summary

Introduction

The species abundance distribution (SAD) describes the full distribution of commonness and rarity in ecological systems. Most comparisons of the different models: (1) use only a small subset of available models (typically two; e.g., McGill, 2003; Volkov et al, 2003; White, Thibault & Xiao, 2012; Connolly et al, 2014); (2) focus on a single ecosystem or taxonomic group (e.g., McGill, 2003; Volkov et al, 2003); or (3) fail to use the most appropriate statistical methods (e.g., Ulrich, Ollik & Ugland, 2010, see Matthews & Whittaker, 2014 for discussion of best statistical methods for fitting SADs) This makes it difficult to draw general conclusions about which, if any, models provide the best empirical fit to species abundance distributions

Methods

Results

Conclusion