Temperature and Morphology Affect the Performance and Cost of Flight in Spruce Budworm Females.

Dispersal is a key process in the spatial and temporal dynamics of insect populations. Dispersal depends on the flight performance of individual insects, which is influenced by their environment, morphology, and physiological state. Temperature affects flight performance and costs through its effect on the metabolism of ectotherms. It can also induce developmental changes in flight-related traits that influence flight biomechanics and insect size, affecting the resources available for this activity. We thus need to understand how temperature during flight modulates flight performance and costs, but also how morphology affects them. Using flight mills, models were developed to describe how flight performance and costs of individual spruce budworm females varied over a range of temperatures (10°C-35°C). Variations of barometric pressure and morphological traits were also incorporated in these models. Flight propensity dropped below 20°C, and when female mass increased relative to wing area, suggesting that there is a wing load beyond which flight probability decreases. Speed, duration, and distance of flight decreased as temperature deviated from 23°C, while wingbeat frequency increased consistently with temperature. Females with long and broad wings had greater flight endurance. Mass loss and use of water and fuel (lipids and carbohydrates) increased with temperature, while the remaining lipids were not affected. As a result, female budworms allocated a daily energy budget to flight, which was proportional to their mass. Large females also benefited from an energy saving due to their mass during flight. Water loss was proportional to female mass but decreased with distance traveled, reaching hydric quasi-homoeostasis at distances greater than 10 km. Our approach reveals the underlying mechanisms of flight and highlights the factors that influence the landing process after dispersal in the spruce budworm. The relationships presented in this study can help develop more realistic models of dispersal for this boreal forest pest.