Doctoral Dissertations
Date of Award
12-2024
Degree Type
Dissertation
Degree Name
Doctor of Philosophy
Major
Entomology, Plant Pathology and Nematology
Major Professor
Jerome F. Grant
Committee Members
Mark T. Windham, Alan S. Windham, Qiusheng Wu, Albert E. Mayfield
Abstract
The invasive pathogen, Harringtonia lauricola, that causes laurel wilt disease has devastated populations Lauraceae plants in the southeastern U.S. This pathogen is transmitted by Xyleborus glabratus Eichhoff, redbay ambrosia beetle. However, other ambrosia beetle species have been confirmed as vectors of the fungus. More recently, H. lauricola has been detected on Lindera benzoin (L.) Blume, northern spicebush. Spicebush is important from an ecological standpoint for several organisms including insects and migratory birds. However, insect diversity and community composition associated with spicebush is poorly known. It is critical to have a basic knowledge of spicebush-associated insects to understand which species may become threatened, endangered, or even locally or regionally extinct if spicebush declines. A multi-year study designed to determine insect diversity and community composition on spicebush was initiated in East Tennessee. A risk assessment was developed to rate spicebush-associated insects for the likelihood of becoming threatened or locally extinct if spicebush declines. A spatiotemporal laurel wilt spread model was also developed to predict the probability of a U.S. county detecting laurel wilt in a future year to determine high-risk areas in which monitoring for laurel wilt should occur. A relatively low insect species richness was collected from spicebush. However, this research found that several insect species likely exhibit a specialized dependence on spicebush. While some species did not directly feed on spicebush, many species were found to shelter on or emerge from stem galls found on spicebush. These observations suggest that the species that will be most affected are native specialist insects. The laurel wilt spread model was able to predict actual detected counties in 2022, 2023, and 2024 with an overall 70.59% accuracy at a threshold for detection of 75%. The mean absolute error (MAE) showed that predicted probabilities of laurel wilt detection were within 21.6% of the actual (100%) detected probabilities in 2022, 2023, and 2024. This model, will inform monitoring efforts that could lead to early detection of laurel wilt. This research will inform protection of spicebush and its associated insects as well as inform future monitoring efforts for laurel wilt.
Recommended Citation
Longmire, Matthew L., "Insect Diversity and Community Composition of Insects Associated with Northern Spicebush and Development of a Spatiotemporal Model to Predict Spread of Laurel Wilt. " PhD diss., University of Tennessee, 2024.
https://trace.tennessee.edu/utk_graddiss/11320