Harbor porpoises and the machine: Assisting manual validation of click trains recorded by Cetacean Porpoise Detector
The Journal of the Acoustical Society of America · 01.01.2026 · 159 (1) · 632-646
DOI: 10.1121/10.0042220 ↗Kurz gefasst
Validierungsstudie in der Zeitschrift The Journal of the Acoustical Society of America, online erschienen am 23.01.2026. Verfasst von Marco F W Gauger und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Animal Ecology, Federal Institute of Hydrology, 56068 Koblenz, Rhineland-Palatinate, Germany. Hauptthemen laut MEDLINE: Phocoena, Vocalization, Animal, Echolocation, Acoustics. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 81 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Federal Traffic and Transport Administration.
Abstract (Originalsprache)
Passive acoustic monitoring (PAM) using Cetacean Porpoise Detectors (C-PODs) is a frequently applied method for studying the presence of harbor porpoises. In quiet environments, the KERNO classifier, an algorithm supplied by the manufacturer, can easily detect narrow-band high-frequency click trains emitted by echolocating harbor porpoises. However, precision is low in noisy habitats, as found in a monitoring data set (0.632; Ems, Elbe, and Wismar Bay, Germany, 2018–2023). We validated and labeled a subsample of 235 529 click trains (Elbe and Ems estuaries, 2023–2024) identified by the KERNO classifier and exported their physical characteristics to train a machine learning (ML) model. Extreme gradient boosting performed very well on the testing data (accuracy: 0.985) and the monitoring data set (0.849). The results show that the model could generalize well, beyond the training data. Moreover, this ML tool can reduce the risk of missing out low-precision estimates from random manual validation. The ML tool can complement the validation process, especially if intervals with only one click train are manually validated because false model predictions occurred particularly in these intervals. Hence, this validation tool may significantly improve the workflow in PAM studies using C-PODs, especially in noisy habitats.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Animal Ecology, Federal Institute of Hydrology, 56068 Koblenz, Rhineland-Palatinate, Germany.
ORCID-Profile: Marco F W Gauger, Thomas Taupp
Förderung
- Federal Traffic and Transport Administration
Zitieren
APA
Gauger, M. F. W., & Taupp, T. (2026). Harbor porpoises and the machine: Assisting manual validation of click trains recorded by Cetacean Porpoise Detector. The Journal of the Acoustical Society of America, 159(1), 632-646. https://doi.org/10.1121/10.0042220
Vancouver
Gauger MFW, Taupp T. Harbor porpoises and the machine: Assisting manual validation of click trains recorded by Cetacean Porpoise Detector. The Journal of the Acoustical Society of America. 2026;159(1):632-646. doi:10.1121/10.0042220
BibTeX
@article{Gauger2026,
author = {Gauger, Marco F W and Taupp, Thomas},
title = {Harbor porpoises and the machine: Assisting manual validation of click trains recorded by Cetacean Porpoise Detector},
journal = {The Journal of the Acoustical Society of America},
year = {2026},
volume = {159},
number = {1},
pages = {632--646},
doi = {10.1121/10.0042220}
}
EUVIMED