@misc{9240260,
  abstract     = {{Monitoring animal populations has several purposes in ecological sciences, including the study of animal behaviour, population dynamics and biodiversity. The common guillemot, known in North America as the common (or thin-billed) murre (Uria aalge), is an important indicator species for the Baltic Sea ecosystem.

Stora Karlsö, Sweden, is the most prominent breeding location for guillemots in the Baltic Sea. Studies on the species have been conducted on this island for several decades. Current methods for individual identification rely on capturing and ringing the birds and recognition is performed by human observation of ringed birds.

The aim of this thesis is to explore a methodology for acoustic individual identification, leveraging a unique audio dataset of continuous recordings of the colony during the 2025 breeding season. The analysis is focused on the parent-chick interaction taking place right before fledging, when the chick leaves the colony together with its male parent. We develop a semi-automatic approach to extract a labelled dataset from raw data and propose a feature extraction method based on cepstral analysis, inspired by concepts used in speaker recognition. The extracted features are used in a machine learning pipeline to classify individuals.

Results show that there are individual differences in the studied calls of both chicks and adults. According to our experiments, the spectral envelope is a useful feature in acoustic individual identification, yielding high classification accuracy across the three investigated call types. In a classification task including five chicks, we obtained 81% accuracy on a test dataset and for the adults (three individuals) we obtained above 90% accuracy on test data. Moreover, our experiments indicate that the proposed feature extraction method is robust to different acoustical ambient conditions and can classify individuals independently of the call type.}},
  author       = {{Heijmink, Ida}},
  language     = {{eng}},
  note         = {{Student Paper}},
  series       = {{Master's Theses in Mathematical Sciences}},
  title        = {{Acoustic Individual Identification of the Common Guillemot (Uria aalge)}},
  year         = {{2026}},
}