Title

FORAGING in FRACTALS? A DEDUCTIVE TEST OF TWO MECHANISTIC COMMUNITY ASSEMBLY MODELS

Date of Award

5-2015

Degree Name

Master of Science (MS)

Department

Biology

Advisor(s)

Mark Ritchie

Subject Categories

Biology | Ecology and Evolutionary Biology | Life Sciences

Abstract

One major goal of community ecology is to understand how basic ecological processes such as selection, dispersal, demographic stochasticity, and speciation affect biodiversity patterns such as Species Abundance Distributions (SADs), Species Area Relationships (SARs), and Abundance Size Distributions (ASDs). Several recent, process-based models have been developed that predict these patterns based on ‘niche’ (e.g. species differences and competition) or ‘neutral’ (e.g. dispersal-limitation and stochasticity) processes. In this paper we conduct a deductive experimental test of SAR and ASD predictions from one ‘neutral’ and one ‘niche’ model, the Unified Neutral Theory of Biodiversity (UNTB) and the Fractal Heterogeneity Theory of Biodiversity (FHTB) respectively, using Auchenorrhyncha insect (i.e., leafhoppers, treehoppers, spittlebugs, and planthoppers) communities in old fields.

We measured Auchenorrhyncha SARs, SADs, and ASDs using sweep sampling of vegetation for three control plots and three plots fertilized with 18g N/m2 of Nitrogen fertilizer (32-0-4). Based on these sampling data we independently estimated all but two FHTB and UNTB model parameters. Each model contains one free parameter, which we estimate by fitting FHTB and UNTB SAD predictions to empirical SADs from control and fertilized plots. We thus avoided as much as possible fitting parameters to our SAR data, which is a common weakness of previous tests. We also compared qualitative model ASD predictions using Auchenorrhyncha body length and mass data.

The results support niche processes being the primary drivers of diversity patterns for Auchenorrhyncha communities in old fields. The observed SARs for both control and fertilized plots do not fall within the 95% confidence intervals of UNTB SAR predictions, but are within the 95% confidence intervals predicted by FHTB. In addition, average species body length and mass negatively scale with species abundance in accordance with predictions of FHTB. UNTB predicts there to be no correlation between abundance and any species trait such as length or mass.

Conclusive tests of ecological theory remain challenging due to the seemingly inescapable presence of free parameters which cannot be independently estimated and must therefore be fit to data. While problematic, the presence of free parameters does not automatically exclude the possibility of successful theory rejection. This study represents one of the most thorough deductive tests of two competing community assembly models to date, and additional deductive tests should be conducted in the future.

Access

Surface provides description only. Full text is available to ProQuest subscribers. Ask your Librarian for assistance.

This document is currently not available here.

Share

COinS