Arthropods in mature eastern hemlock forests: assessing the non-target effects of basal bark imidacloprid application for control of hemlock woolly adelgid on forest floor and canopy arthropod fauna in southwestern Nova Scotia

As the invasive and highly destructive hemlock woolly adelgid (Adelges tsugae) continues its spread northwards into Nova Scotia, the province’s stands of eastern hemlock (Tsuga canadensis) face an ever-mounting threat to their long-term survival. As one of the largest and longest-lived trees occurring in the Canadian Maritimes, eastern hemlock stands comprise a foundational component of one of the region’s climax forest types, providing crucial ecosystem services for innumerable forest flora and fauna. Given the tree’s significance to local ecology, conservation practitioners in the province are implementing measures to control populations of A. tsugae, including application of neonicotinoid insecticides, like imidacloprid. Until recently, imidacloprid treatments to control hemlock woolly adelgid in Nova Scotia were limited to stem injection. To achieve faster and more economical control of A. tsugae, however, some local conservation practitioners are pivoting to basal bark application of the chemical. Spray application of imidacloprid raises concerns about potential harmful impacts to non-target arthropods inhabiting local treated hemlock stands. To address these concerns, we sampled a diverse range of arthropods inhabiting the forest floor and the subcanopy and canopy layers both before and after localized basal bark application of imidacloprid, as well as in nearby untreated stands. For our sampling, we used both passive pitfall trapping and baited Lindgren funnel trapping to sample terrestrial and canopy biodiversity. For both trap types, the most abundant taxa captured (representing 33 arthropod clades) were identified and analyzed for changes in their abundance that were characterised by a treatment and year interaction. While we identified a significant decline in the abundance of one of our target groups linked to imidacloprid treatment, spine-waisted ants (Aphaenogaster spp.), we also identified a significant decline in the abundance of two groups (the ground beetle Synuchus impunctatus and the striped ambrosia beetle Trypodendron lineatum) in plots that did not receive imidacloprid treatment. At a broader family level, several interannual and inter-site differences in taxon diversity and composition not related to imidacloprid treatment were also identified, particularly for the ground beetles (Carabidae), longhorn beetles (Cerambycidae) and true weevils (Curculioindae). While the drivers of this treatment-independent variation are unknown, they may correlate with advanced hemlock canopy decline caused by hemlock woolly adelgid. Further investigation into imidacloprid’s potential to yield non-target effects on local arthropods is warranted, particularly on spine-waisted ants. Further study is also warranted to better understand the ecological implications of eastern hemlock decline and canopy loss driven by hemlock woolly adelgid infestation, especially in a Nova Scotian context.