Dude, where's my mark? Creating robust animal identification schemes informed by communication theory

Abstract The analysis of animal behaviour often requires uniquely marking and tracking many individuals simultaneously for long periods. However, these identifying marks can often be partially removed, erased or obscured, preventing accurate differentiation between individuals. Modern computer-readable tags can mitigate some of these challenges while also creating others, such as not being human-readable. Here, we apply several fundamental principles from the engineering of communications systems to the design of human-readable animal identification schemes that are robust to mark removals. These techniques allow biologists to fully reconstruct the identities of individual animals even as marks degrade over time or become only partially observable due to obstructions. Additionally, we supply an open-source R software package called rabi to make the implementation of these methods painless and accessible to researchers unfamiliar with coding theory. More generally, we hope that exposure to these techniques from engineering will narrow the disciplinary gap, albeit slightly, between experts in animal and technological communication.