Methodology:

GLS tags were attached to animals' legs during the breeding season or at the end of moult with purpose-designed bands. These were recovered when the birds returned to their colonies to breed. Devices recorded light levels as well as sea surface temperature.

Tracks were analysed using BAStag and SGAT in R where processed files are included in this dataset. Processing estimated coordinate locations at 12h regular intervals.

Data collection:

Positions during the non-breeding season were estimated from Global Location Sensing (GLS) loggers, which allows for the estimation of latitude and longitude from daylight measurements.

Data were downloaded from MigrateTech GLS tags using proprietary MigrateTech software. Subsequently data were decompressed using the BAStrack software 'decompressor'. Clock drift was accounted for using the start time of each tag. Loggers measured light levels every 60 seconds recording the maximum light level in a 10 minute period. Salt-water immersion was assessed and the number of positive tests in each 10 min interval recorded: an immersion value of 0 indicates that the logger was completely dry and 200 that it was completely wet. Processed Estelle data was created from the light data using the BAStag and SGAT packages in R.

Data quality:

Data were corrected for any clock drift over time using BAStrack software 'decompressor'.