DOI
https://doi.org/10.25772/GZZP-VF56
Author ORCID Identifier
https://orcid.org/0000-0001-9233-2704
Defense Date
2021
Document Type
Thesis
Degree Name
Master of Science
Department
Biology
First Advisor
Dr. Karen Kester
Second Advisor
Dr. Catherine Hulshof
Abstract
Determining the drivers of species occurrences is a central topic in ecology. Both abiotic and biotic factors are important predictors of species distributions, yet most species distribution models consider abiotic variables only. In this study, the influence of abiotic and biotic factors on current distributions of herbivores, Manduca sexta, and Manduca quinquemaculata, their major cultivated and wild host plants, and a parasitoid, Cotesia congregata, was quantified using an ensemble of three models (generalized linear model, maximum entropy, and random forest). The predictive ability of abiotic-only, biotic-only, and abiotic-biotic models was evaluated. Winter temperatures and precipitation (i.e., precipitation of the driest month and summer precipitation) were important determinants of herbivore distributions, but the two species had different climatic optima. The distribution of the cultivated and wild host plants was largely influenced by elevation, while the probability of occurrence of the parasitoid was primarily driven by summer precipitation. In the biotic-only model, the herbivore distribution was more likely to co-occur with cultivated host plants than wild host plants. The cultivated host plants were the most important factor for predicting herbivore distributions in the abiotic-biotic model. The three models did not differ in predictive ability, which was likely due to data limitations and co-relations between biotic and abiotic variables. Results of this study underscore the need to disentangle relationships between abiotic and biotic predictor variables on species distributions.
Rights
© Emelia Obodum Kusi
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
8-2-2021