Location: Chemistry ResearchTitle: Predicting county-level southern pine beetle outbreaks from neighborhood patterns) Author
Submitted to: Environmental Entomology
Publication Type: Peer reviewed journal
Publication Acceptance Date: 1/31/2011
Publication Date: 7/25/2011
Citation: Duehl, A.J., Bishir, J., Hain, F.P. 2011. Predicting county-level southern pine beetle outbreaks from neighborhood patterns. Environmental Entomology. 40(2):273-280. Interpretive Summary: The southern pine beetle is the most important pest of southern pine forests. The states with southern pine management and beetle activity have kept infestation records for the past half century. By looking at the patterns of infestation year by year we determined what infestation patterns are important to future infestations. Before working with the Center for Medical, Agricultural and Veterinary Entomology in Gainesville, FL, scientist Adrian Duehl worked with other researchers at NC State University to show that county patterns, particularly the infestation status of a county can aid in predicting future infestations. In the southern pine beetle system, the best predictor of future infestations in a county is the infestation status of that county in the previous year. Counties that were infested last year are likely to be infested again in future years. At the county level a history of infestation always increases infestation risk. Future research on this topic should add in other predictive variables, such as landscape structure and environmental conditions to improve the predictive power of the model.
Technical Abstract: The southern pine beetle (Dendroctonus frontalis, Coleoptera: Curculionidae) is the most destructive insect in southern forests. States have kept county-level records on the locations of beetle outbreaks for the past forty-eight years. In this study, we seek to determine how accurately patterns of county-level infestations in preceding years can predict infestation occurrence in the current year. We consider a variety of methods for predicting infestations, including quantification either of the exact locations of infested counties during one or two preceding years, or of the infestation intensity in these years. The methods have similar predictive abilities, but the simpler methods perform somewhat better than the more complex ones. The factors most correlated with infestations in future years are infestation in the current year and also the number of surrounding counties that are infested. We conclude that historic county patterns can help predict the probability of future infestations in the region, but that use of county-level patterns alone leaves much of the year-to-year variability unexplained.