Using historical data, equilibrium tests and experiments to improve distribution models - a study with ants

Robert R. Dunn, North Carolina State University

Co-investigator: Nathan J. Sanders, University of Tennessee

Collaborator: Matthew J. Fitzpatrick, University of Tennessee

Abstract

Climatic change is predicted to lead to dramatic shifts in the geographic boundaries of distributions of species and the ecological processes those species mediate. Most existing models assume that distributions of species are in equilibrium with climatic variables and that species will track changes in climatic variables perfectly. Empirical studies of range shifts indicate that these assumptions are rarely valid, but few alternatives exist for considering large-spatial scales and multiple species. We propose to integrate species distribution models, tests of distribution equilibrium and experimental manipulations of climate to develop more realistic predictive models for species’ distributions, patterns of diversity and ecosystem process.

Location: We will use historical data for ants from sites throughout eastern North America to predict current distributions of those species, with a resampling of a subset of historical sites for model validation. Data on ant distribution throughout North America will be used to test assumptions of distribution equilibrium. Experimental manipulations will be conducted in North Carolina. Results of these methods will be used to predict future patterns of ant distribution across eastern North America and more generally.

Questions: We address four key questions: 1. Do range shifts conform to model predictions? 2. Are ranges in equilibrium? 3. How do resident and non-resident species respond to local manipulations of temperature? 4. How do we better predict future distributions? We take advantage of the large number of ant species and higher taxa in eastern North America to understand how the answers to these questions vary among taxa.

Methods: We employ a combination of techniques to validate distribution models based on historical data, tests of distribution equilibrium, climate manipulations to compare the relative equilibrium and dispersal ability of different lineages of ants. We use the results of these methods to develop and parameterize separate models of the effects of climatic change on the future distribution of different ant lineages and more than two hundred ant species.

Deliverables: We will increase understanding of how best to incorporate the equilibrium (or disequilibrium) of species’ distributions and varying levels of dispersal into distribution models. Our effort will establish ants as a new model organism for the study of climatic change, but will also generate a general understanding of how to use comparative information about the dispersal ability of taxa to better inform models. Our proposed research will advance modeling approaches associated with focus 2, but the research will also address questions raised in focus 1 and focus 4.