| dc.description.abstract | Oak wilt, caused by Bretziella fagacearum, is expanding beyond its historical U.S. range, and its confirmation in Canada has increased concern about future disease risk and management costs. The disease threatens oak-dominated forests and is difficult to eradicate once established, making early detection, sustained surveillance, and preventive management essential. Oak wilt epidemiology is influenced by climate-sensitive interactions among the host, pathogen, and insect vectors, but the implications of these relationships for future geographic suitability remain poorly resolved. This study first examined regional variation in oak wilt epidemiology and its underlying biotic and abiotic drivers and then evaluated where changing climatic conditions may permit future establishment and expansion.
A PRISMA-guided review of 121 studies revealed two mechanistically distinct epidemics. In the Lake States, seasonal insect transmission and spread through intertree root grafts produce disease-center expansion of 3.6-7.7 m yr-1. In Texas, continuous clonal live-oak root systems and year-round transmission permit expansion approaching 40 m yr-1. Four geographically structured pathogen populations further support regional divergence and help explain the limited transferability of management practices. The synthesis further showed that temperature regulates vector phenology, precipitation and moisture influence fungal mat development, and drought increases host susceptibility, linking regional epidemiology to climate.
Climatic suitability was modeled using 880 confirmed occurrences, 853 spatially stratified background locations, 25 predictors, and seven species distribution algorithms. The four highest-performing models were combined in an AUC-weighted ensemble. Under repeated random 70/30 validation, the ensemble achieved high predictive performance (AUC = 0.940; TSS = 0.802). Independent spatial-block 90/10 validation produced an ensemble AUC of 0.955 and TSS of 0.754, indicating strong geographic discrimination with moderately lower threshold-dependent classification skill. Precipitation, climatic moisture, and relative humidity were the dominant predictors. Historical suitability covered 23.35% of North America and 17.91% of the Asia-Pacific region, including extensive areas beyond the pathogen’s known distribution. Projections under SSP1-2.6, SSP2-4.5, and SSP3-7.0 consistently indicated future expansion. Under SSP3-7.0, by 2071-2100, the suitable area increased to 27.97% in North America and 26.16% in the Asia-Pacific region, with pronounced northward expansion in North America.
By linking climate-sensitive disease mechanisms with regional epidemiology and continental-scale projections, this thesis provides a novel integrated decision framework for surveillance, biosecurity, region-specific intervention, and climate-adaptive oak wilt management. | en_US |
