Numerical response of Trypodendron lineatum to hemlock looper defoliation in boreal balsam fir forest

The hemlock looper is one of the most damaging insect defoliators in Canada, often killing trees after only one or two years of heavy defoliation. A study on the short-term succession of saproxylic beetles after a hemlock looper outbreak highlighted the dominance of the striped ambrosia beetle in boreal balsam fir stands heavily defoliated by this looper. In the present study, we measured the numerical response of this ambrosia beetle to balsam fir defoliation. Our objectives were to determine 1) the defoliation threshold triggering massive colonization of balsam fir trees by the striped ambrosia beetle, using emergence traps placed around tree trunks and 2) if the upcoming colonization could be forecasted with multidirectional flight-interception traps or trunk-window traps. Results showed that colonization only began when balsam fir trees reached 95 % defoliation. Trees reaching this threshold had 54 % probability of being massively colonized as indicated by the emergence of hundreds of beetles from single trees over the summer. The probability of massive colonization reached 61 % when defoliation was total. Both, the trunk-window trap and the multidirectional flight-interception trap, proved reliable in predicting striped ambrosia beetle attacks. Either trap captured this beetle in all plots where at least one tree had been massively colonized. These traps could thus be helpful for monitoring populations of the striped ambrosia beetle, predicting tree colonization and planning salvage logging plans. • Hemlock looper feeding on balsam firs does not trigger immediate secondary attacks. • Fully defoliated firs during the first summer will be avoided by secondary beetles. • 2 years after heavy defoliation, only Trypodendron lineatum colonizes fir boles. • Interception traps are efficient to monitor T. lineatum signs of colonization. • 95 % defoliation threshold and traps may help managers optimizing salvage logging.