Power and Wake Dynamics of Hawkmoth Flight

Sammanfattning: Aerodynamic theory states that the power required to fly is related to flight speed with a ∪-shaped curve. This has been shown in vertebrates, but insects have been proposed to have a flat, or J-shaped power curve. This means that hovering – flying at zero speed – is energy-efficient, which would explain why it is a relatively common flight mode in insects. In this thesis I show that some insects do, in fact, have a ∪-shaped power–speed relationship. I studied the flight of two hawkmoths (M. sexta and M. stellatarum) flying freely in a wind tunnel, and estimated aerodynamic power from the kinetic energy in the wake as well as from downwards velocities, using particle image velocimetry (PIV). In the first study, M.sexta flew at a range of speeds(1-4 ms-1), and the resulting power curve constructed from the data was clearly ∪-shaped. In two separate studies I sampled the wake of M. stellatarum hovering, and flying at 1.5 ms-1, respectively, and could show that hovering was a considerably more expensive mode of flight in this species.With knowledge of the power expenditure during flight one can estimate flight efficiency as well as predict certain characteristic flight speeds, which the animal should choose to optimize its flight depending on the task at hand. PIV is an impractical and expensive technique for estimating power, and so it is common, in e.g. migration studies, to simply use any of the available aerodynamic power models to acquire an estimate. Often these are based in aeroplane theory, and are seldom completely adapted to the special circumstances flyers operating at the scale of insects are exposed to. Here, I compared results from my power estimations with three models to evaluate the efficacy of this approach. I found that, while power in forward-flying M. sexta can be relatively accurately predicted by one model, its sensitivity to certain parameter choices makes predicting optimal flight speeds unreliable. A second model, specifically designed for insect hovering, severely overestimated power in hovering M. stellatarum.The taxa I have studied in this thesis – hawkmoths – are rare among the insects in that they can be easily positioned in a wind tunnel, as the nectar-feeding moths can be trained to feed from an artificial flower. As such, they can be studied when flying freely, but many other species require some form of tethering if we wish to study their flight. In tethered flight, the sensory feedback loop is corrupted, and so it is unclear how naturally a tethered insect flies. I compared kinematics and aerodynamic properties of the wake of M. stellatarum in free-flight as well as tethered, and found that tethered moths increased both their stroke amplitude and stroke plane angle. In addition, they had a less efficient flight.