Comparing neural networks against click train detectors to reveal temporal trends in passive acoustic sperm whale detections

• Published in: The Journal of the Acoustical Society of America
• Main Authors: Garrobe Fonollosa, Laia, Webber, Thomas, Brotons, José Maria, Cerdà, Margalida, Gillespie, Douglas Michael, Pirotta, Enrico, Rendell, Luke Edward
• Other Authors: University of St Andrews.Scottish Oceans Institute, University of St Andrews.School of Biology, University of St Andrews.Sea Mammal Research Unit, University of St Andrews.Marine Alliance for Science & Technology Scotland, University of St Andrews.Bioacoustics group, University of St Andrews.Sound Tags Group, University of St Andrews.Centre for Research into Ecological & Environmental Modelling, University of St Andrews.Centre for Social Learning & Cognitive Evolution, University of St Andrews.Centre for Biological Diversity, University of St Andrews.Institute of Behavioural and Neural Sciences
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2025
• Subjects: QH301 Biology; DAS; QH301; Calma; Physeter macrocephalus; Sperm whale
• Online Access: https://hdl.handle.net/10023/31210; https://doi.org/10.1121/10.0034602

Funding: The Direcció General de Pesca i Medi Marí del Govern de les Illes Balears - support to obtain data about Sperm Whale presence in Emile Baudot and Mallorca Channel. Additionally, we are grateful to the Fundación Biodiversidad of the Ministerio de Transición Ecológica y el Reto Demográfico for their support of projects CALMA and CALMADOS, which were conducted to assess the temporal presence of cetaceans in the sea mountains Ausiàs March and Monte Olivas. Passive acoustic monitoring (PAM) is an increasingly popular tool to study vocalising species. The amount of data generated by PAM studies calls for robust automatic classifiers. Deep learning (DL) techniques have been proven effective in identifying acoustic signals in challenging datasets, but due to their black-box nature their underlying biases are hard to quantify. This study compares human analyst annotations, a multi-hypothesis tracking (MHT) click train classifier and a DL-based acoustic classifier to classify acoustic recordings based on the presence or absence of sperm whale (Physeter macrocephalus) click trains and study the temporal and spatial distributions of the Mediterranean sperm whale subpopulation around the Balearic Islands. The MHT and DL classifiers showed agreements with human labels of 85.7% and 85.0%, respectively, on data from sites they were trained on, but both saw a drop in performance when deployed on a new site. Agreement rates between classifiers surpassed those between human experts. Modeled seasonal and diel variations in sperm whale detections for both classifiers showed compatible results, revealing an increase in occurrence and diurnal activity during the summer and autumn months. This study highlights the strengths and limitations of two automatic classification algorithms to extract biologically useful information from large acoustic datasets. Peer reviewed