Validating phenology models to optimize management of box tree moth in North America

Box tree moth (BTM), Cydalima perspectalis (Walker) (Lepidoptera: Crambidae), is an invasive species posing a threat to ornamental boxwoods (Buxus spp.) in North America. As the moth continues to spread, there is a need to optimize management strategies for BTM populations in recently invaded areas. Applications of Bacillus thuringiensis kurstaki have proven effective against BTM and application timing, as guided by phenology models, can strengthen efficacy. This study compares two phenology models for predicting the development of BTM in North America: a degree-model with variable biofix dates and a recently developed mechanistic model. While both methods rely on degree days to predict development, the mechanistic model incorporates responses to photoperiod. We compared forecasts from both models, evaluating their ability to predict the appearance and duration of susceptible life stages across regions that were recently invaded by BTM and that span various climatic regimes. Using these findings – and while accounting for the fact that these two phenology models have different levels of complexity and inputs – we aim to clarify potential advantages of each modelling framework for timing management interventions.