This dataset was used to investigate the relationships between wind, waves, and flapping rates in giant petrels (Macronectes spp.) breeding at Bird Island, South Georgia in 2022 and 2023. It is comprised of hourly GPS positions for 36 northern giant petrels (M. halli) and 42 southern giant petrels (M. giganteus) and contains the unique deployment ID assigned to each individual bird, and the sex of each individual bird. Hourly wind speed (km/h) and swell height (m) were extracted from ERA5 hourly data on single levels along the birds' foraging tracks (0.25 x 0.25 degree spatial resolution for wind data, and 0.5 x 0.5 degree spatial resolution for swell data). Wing flaps per hour were calculated using a LULU filter to identify peaks in the z-axis of the accelerometer data. The dataset only contains data for times when the birds were performing commuting flight as predicted by a Hidden Markov Model used on GPS data in order to restrict analyses to in-flight behavior. Data points within 20 km of land around South Georgia were also excluded to remove coastal foraging performed predominantly by male giant petrels. Generalized Additive Mixed Models were constructed to analyze how wind speed and swell height impact flapping rates. 

Please refer to Hallet et al. for specific details about the methods used to create this dataset and to run analyses.