A bivariate approach to pheromone-based monitoring is developed for the spruce budworm, Choristoneura fumiferana (Clem.) (Lepidoptera: Tortricidae). The approach uses captures of males at pheromone traps for generation t (♂t) as a transitive term between densities of overwintering larvae in consecutive generations (L2t, L2t+1), based on a large data set including >2,000 observations in the province of Quebec (QC) between the interval 1992 and 2010. Although estimates of L2t and ♂t are autocorrelated to some extent, multi-year assessments of larval densities combined with pheromone trapping are justified by the complementarity (statistical significance) of both L2t and ♂t in predicting L2t+1 for 15 of 18 pairs of 2-yr intervals. Bivariate pheromone-based thresholds (number of males corresponding to specific transitions in larval densities between L2t and L2t+1) are reported for each year. As expected, thresholds for stable populations (L2t = L2t+1) were lower than for populations with positive growth rate (L2t < L2t+1). The thresholds derived in this study have limited heuristic value; however, because they vary greatly from year to year.
Keywords: eruptive population dynamics; forest defoliator; multivariate function; pheromone quality control; pheromone-based threshold.
© Crown copyright 2017.