New publication: model-based clustering of detection functions to reduce uncertainty estimation
Dear MARMAM colleagues, On behalf of my co-authors, I am pleased to share our new publication in Marine Mammal Science: Plard, F. <https://florianeplard.com/> / Araújo, H. E. F. / Astarloa, A. / Saavedra, C. / Vazquez Bonales, J. A. / Authier, M. Using Fusion Effects to Decrease Uncertainty in Distance Sampling Models when Collating Data from Different Surveys <https://onlinelibrary.wiley.com/share/author/DZGJHVJASJKVAQEDEARA?target=10.1111/mms.13104> In this publication, we developed and encoded in our R package AMBIdsm <https://gitlab.univ-lr.fr/pelaverse/AMBIdsm> some functionalities, leveraging the probabilistic programming language Nimble, to perform a joint estimation and clustering of detection functions when analyzing distance sampling data from several surveys over a region. The clustering is achieved over species and survey platform, and can also accommodate important covariates such as Beaufort scale or platform height. This clustering allows to account for between-species and between-survey heterogeneity with data-driven pooling. We show case the gain in esw estimates both a in a simulation study and in a real case study (a part of the CetAMBICion project <https://www.cetambicion-project.eu/?lang=en>) focusing on cetaceans in the Bay of Biscay. If you have any questions about our work, please reach out. Cheers, Matthieu Authier -- Matthieu Authier - Ingénieur de Recherche La Rochelle Université Observatoire Pelagis UAR 3462 CNRS-LRUniv 5 allée de l'océan 17 000 La Rochelle telephone: +00 33 (0)7 77 22 32 67 (remote-working/télé-travail) https://www.observatoire-pelagis.cnrs.fr/pelagis-2/lequipe/matthieu-authier/ Try again. Fail again. Fail better. (Samuel Beckett) Verzögerungen im Betriebsablauf (Deutsche Bahn) 🍍 🦉🐛
participants (1)
-
Matthieu Authier