Motion-activated camera trap images from the ECN Cairngorm long-term monitoring site, 2010-2022

Data was collected and processed using a standardised methodology included as a supplementary file. Images were collated directly from cameras into yearly named folders by year. Yearly collections were then processed through Microsoft’s object detection AI, MegaDetector (https://github.com/microsoft/CameraTraps/blob/main/megadetector.md), to streamline image content identification. Data extraction then took place using content filters within 'Timelapse: An Image Analyser for Camera Traps' software (https://saul.cpsc.ucalgary.ca/timelapse) with output data exported as a .csv file. Any errors with time and date extracted from image EXIF, were corrected manually during the final processing step. The data was manually checked for errors but has not been through an exhaustive QC process which is recommended as good practice prior to use.