Abstract
Bats colliding with spinning wind turbine blades result in bat mortality. Carcass surveys at individual wind turbines vary from daily to once a week and from large cleared plots to only the road and pad area. A physics-based model is proposed to guide carcass surveys, for designing curtailment studies to detect treatment fatalities and for improving fatality estimates by accounting for unsearched areas. The model considers the effects of carcass size, weight, and drag, and it accounts for the turbine rotor size and rotation rate to simulate the trajectory of a carcass after it is struck by a wind turbine blade. A carcass parameter is defined as the ratio of drag force to body weight, which accounts for the relative effect of bat biophysical and aerodynamic characteristics. By applying restrictions on carcass survey and turbine yaw data, a limited sample of bat fatalities was obtained, and the analysis revealed that bats fall downwind of wind turbines, indicating wind drift significantly influences carcass fall trajectories. The new ballistics model includes the effect of wind drift on fall trajectory of a carcass. The model was used to investigate the sensitivity of carcass fall trajectories to variability of the input parameters. The tests showed that larger values of the carcass parameter, that is, when drag dominates, such as for small carcasses, resulted in larger downwind drift, whereas large carcasses with smaller carcass parameter values resulted in larger distances within the rotor plane. The relationship of wind speed and RPM was found to influence the carcass downwind distance more compared to the within rotor plane distance. Using carcass survey data, turbine operation data, and wind speed records, for seven bats surveyed the day after colliding with a wind turbine, modelled back-trajectories were used to identify the likely strike location on the rotor. The model can be improved by validating the modelled trajectories with the recorded bat-blade strikes in thermal videos. It should be noted that the findings of the present study are based on the bat fatalities that met strict criteria leading to small sample size and hence requires further evaluation for testing the robustness of the model.