{
  "abstract": "Seabirds can be notoriously difficult to monitor using conventional methodology, as many breed on remote islands, are active nocturnally and many breed in underground burrows. Current census methods such as counting burrows and using playbacks are time-consuming and labour-intensive. One novel solution is to use passive acoustic monitoring (PAM) followed by individual recognition and number estimation. Previous studies have successfully achieved individual bird recognition with high accuracy; however, precisely estimating population abundance via their acoustics remains an open problem. To bridge this gap, we tested whether individual Manx shearwaters (Puffinus puffinus) can be identified from acoustic recordings, and whether the numbers of individuals present can be estimated. We collected recordings of single-individual and multiple-individual calls from three colonies in the UK. We developed a pipeline to combine BirdNET, a one-dimensional convolutional neural network, and agglomerative clustering for individual identification and open-set individual counting. Closed-set recognition achieved high accuracy (89.3%) for Manx shearwater vocalisations at the individual level, and recognition accuracy showed no significant difference between sexes. Open-set individual number estimation achieved a mean relative error of 39.1%, and model performance was robust to noise and recording conditions. These findings indicate that vocalisations of both sexes contain individual signatures and demonstrate the feasibility of counting unseen individuals from passive acoustic recordings. Importantly, this study supports population census using PAM in real field scenarios, where unseen individuals are abundant. We provide an easy, non-invasive approach for seabird population monitoring, which is transferable for monitoring other species, and has a high potential for wildlife conservation and ecological research.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": null,
  "author_total": 6,
  "authors": [
    {
      "name": "Zhanyi Lin"
    },
    {
      "name": "Yinka Julianah Adeniji"
    },
    {
      "name": "William D Pearse"
    },
    {
      "name": "Stephanie Harris"
    },
    {
      "name": "Sue Anne Zollinger"
    },
    {
      "name": "Julia Schroeder"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/8f889gj51b2he.bib",
  "cited_by": 0,
  "doi": "10.32942/x25d7z",
  "fulltext_url": null,
  "id": "8f889gj51b2he",
  "issns": [],
  "landing_url": "https://ecoevorxiv.org/repository/view/15160/",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/legalcode",
  "metadata_licence": null,
  "native_subjects": [
    [
      "bepress",
      "Animal Sciences"
    ],
    [
      "bepress",
      "Bioinformatics"
    ],
    [
      "bepress",
      "Ornithology"
    ],
    [
      "openalex",
      "1309|Animal Vocal Communication and Behavior"
    ]
  ],
  "openalex": "W7214225353",
  "pdf_url": "https://ecoevorxiv.org/repository/object/15160/download/26287/",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-24",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/8f889gj51b2he.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:31Z",
      "native_id": "15160",
      "source": "ecoevorxiv",
      "url": "https://ecoevorxiv.org/repository/view/15160/"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "bepress:Bioinformatics",
      "subfield": "bioinformatics"
    }
  ],
  "title": "Individual recognition and number prediction for Manx shearwater (Puffinus puffinus) by passive acoustic monitoring",
  "title_text": "Individual recognition and number prediction for Manx shearwater (Puffinus puffinus) by passive acoustic monitoring",
  "type": "preprint",
  "updated_at": "2026-10-30T12:17:31Z",
  "url": "https://research.draconitic.com/paper/8f889gj51b2he",
  "venue": "EcoEvoRxiv",
  "versions": []
}