|Landis, Douglas - MICHIGAN STATE UNIV|
|Schemske, Douglas - MICHIGAN STATE UNIV|
|Evans, Jeffrey - MICHIGAN STATE UNIV|
Submitted to: Weed Science Society of America Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: January 16, 2006
Publication Date: February 15, 2006
Citation: Davis, A.S., Landis, D.A., Schemske, D.W., Evans, J.A. 2006. Matrix population models to inform A. petiolata biocontrol agent selection [abstract]. Weed Science Society of America Meeting. 45:36. Technical Abstract: Garlic mustard (Alliaria petiolata (M. Bieb.) Cavara and Grande) invades and degrades woodland habitats in North America. Four species of Ceutorhynchus weevils (Coleoptera: Curculionidae) that attack different stages of A. petiolata are currently undergoing host specificity testing as potential biological control agents. Selection of the best agents depends on both host-specificity and anticipated impacts of each species on A. petiolata population growth rate ('). Matrix population models have proven effective for analyzing impacts of natural enemies on weed demography. Elasticity analysis of a matrix model parameterized from data collected at eight sites in Michigan indicated that changes in the rosette to flowering plant transition, fecundity and seedling survival to the rosette stage consistently had the greatest impacts upon A. petiolata '. These results suggest that rosette-feeders that reduce overwintering survival and seed or stem-petiole feeders that reduce seed output should be particularly effective agents. Simulated feeding impact of two Ceutorhynchus weevils resulted in declining A. petiolata populations at five out of eight locations. Varying demographic parameters within ranges observed in Michigan had large impacts on biocontrol outcomes, indicating that successful containment and reduction of A. petiolata populations may occur under some, but not all, conditions of demographic variability.