A compilation pipeline for wildlife tracking datasets collected from ground-based and satellite-based telemetry transmission devices

Abstract

Wildlife conservation planning increasingly requires collaboration and integration of research from discrete studies spanning large geographic areas. Tracking datasets are essential for analyzing animal movements and species distributions in relation to environmental conditions and combining them can enable powerful analyses to further aid planning efforts. However, combining datasets necessitates addressing variation in study designs, tracking methodologies, location uncertainty, and data attributes. We outline a compilation pipeline to integrate ground-based and satellite-based telemetry tracking datasets, motivated from our work with greater sage-grouse ( Centrocercus urophasianus ), a highly imperiled species of western North America. Our objective was to create a database with a standardized set of attributes to facilitate filtering locations for spatial analyses. Our pipeline phases are: (1) dataset pre-processing, (2) formatting individual datasets to a common template, (3) dataset binding, (4) error checking, and (5) filtering. Our pipeline includes additional functionality to identify coordinates from recurrently visited locations (e.g., nest sites), which may be of special interest. The final compiled sage-grouse database included nearly 5 million locations collected from 53 datasets and over 19,000 birds tracked from 1980 to 2022, including over 11,000 nest locations. Our error checks flagged 3.9 % of locations as likely errors, predominantly collected from satellite-based telemetry transmissions. We demonstrate the ability of our pipeline to identify nest locations and flag erroneous locations by applying it to simulated tracking datasets. Overall, our workflow offers a transferable approach for researchers aiming to standardize wildlife telemetry datasets and conduct ecological analyses for both individual studies and large-scale collaborations.