Predictions and retrodictions of the hierarchical representation of habitat in heterogeneous environments

Authors

Jurek Kolasa, McMaster UniversityFollow
Craig R. Allen, USGS Nebraska Cooperative Fish and Wildlife Research Unit, University of NebraskaFollow
Jan Sendzimir, International Institute for Applied Systems Analysis, Laxenburg, AustriaFollow
Craig A. Stow, NOAA – Great Lakes Environmental Research Laboratory, Ann Arbor, MIFollow

Date of this Version

2012

Citation

Ecological Modelling (2012); doi 10.1016/j.ecolmodel.2012.03.030

Comments

This article is a U.S. government work, and is not subject to copyright in the United States.

Abstract

Interaction between habitat and species is central in ecology. Habitat structure may be conceived as being hierarchical, where larger, more diverse, portions or categories contain smaller, more homogeneous portions. When this conceptualization is combined with the observation that species have different abilities to relate to portions of the habitat that differ in their characteristics, a number of known patterns can be derived and new patterns hypothesized. We propose a quantitative form of this habitat–species relationship by considering species abundance to be a function of habitat specialization, habitat fragmentation, amount of habitat, and adult body mass. The model reproduces and explains patterns such as variation in rank–abundance curves, greater variation and extinction probabilities of habitat specialists, discontinuities in traits (abundance, ecological range, pattern of variation, body size) among species sharing a community or area, and triangular distribution of body sizes, among others. The model has affinities to Holling’s textural discontinuity hypothesis and metacommunity theory but differs from both by offering a more general perspective. In support of the model, we illustrate its general potential to capture and explain several empirical observations that historically have been treated independently.