ORCID Identifier(s)

0000-0002-8815-1238

Graduation Semester and Year

2020

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Quantitative Biology

Department

Biology

First Advisor

James P Grover

Abstract

The field of species distribution modeling has expanded rapidly over the past two decades along with increasing computing capability and the introduction of novel analytical techniques. SDMs relate species occurrence to environmental data and are a valuable tool in helping researchers understand species-environment relationships and make projections about future distributions. Model reliability has been shown to be highly dependent upon a range of factors, including statistical techniques, predictor variables, species prevalence, and choice metrics used to assess model accuracy. SDMs developed from ensembles of different models have been shown to improve reliability. Here I develop an ensemble modelling approach for freshwater fish using both statistical and algorithmic techniques and a comprehensive, continental scale dataset. I also evaluate the influence of prevalence and model technique on the most used performance metrics. The results of this ensemble model are then combined into a community consensus model, which is used to identify environmental factors that broadly underly freshwater fish distribution across multiple species. The community consensus model also highlights predictors that are important for rare or specialized species. Lastly, I evaluate the ability of the community consensus model to correctly identify species niche predictors through the use of simulated, virtual species. The results of this dissertation indicate that SDM accuracy should be evaluated using several metrics, and that species prevalence should be considered when selecting both model technique and performance metric. A consensus model can perform as well or even better than individual models and can be used to identify environmental predictors that influence freshwater fish as a group. The consensus approach also highlights predictors that are important for rare species. Finally, the results of a carefully developed community consensus modeling approach are consistent with mechanistic knowledge and are not simply “black boxes”.

Keywords

Species distribution model, Machine learning, Prevalence, Freshwater fish, Consensus model, Ensemble model

Disciplines

Biology | Life Sciences

Comments

Degree granted by The University of Texas at Arlington

30695-2.zip (3892 kB)

Included in

Biology Commons

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.