Short-term responses of sperm whales Physeter macrocephalus to the attachment of suction cup tags

• Published in: Marine Ecology Progress Series
• Main Authors: Warren, Victoria E., Miller, Patrick J.O., Tyack, Peter L.
• Format: Article in Journal/Newspaper
• Language: English
• Published: 2020
• Subjects: Biotelemetry; Diving behaviour; Physeter macrocephalus; Sperm whale; Suction cup tag; Tagging effect
• Online Access: https://research-portal.st-andrews.ac.uk/en/researchoutput/shortterm-responses-of-sperm-whales-physeter-macrocephalus-to-the-attachment-of-suction-cup-tags(e1fceee8-a78e-4476-a275-d5c045ae6bff).html; https://doi.org/10.3354/meps13344

Animal-mounted data logging devices are used to study the behaviour, physiology, and ecology of free-ranging marine mammals, as well as their reactions to controlled exposures. It is important to consider whether collected data are representative of natural behaviour or biased by responses to tagging. In species with stereotypical diving behaviour, tagging responses can be quantified by identifying anomalous dives. Data from 36 suction cup tag deployments on sperm whales Physeter macrocephalus from 4 locations were analysed to consider whether tagging effects were evident within 5 dive parameters: Maximum dive depth, dive duration, descent speed, depth difference between start of clicking and first prey capture attempt, and buzz rate. Linear mixed models were generated for each response parameter and covariates for dive index were added to assess whether model fit improved when the order of dives was taken into account. Time-decaying tagging effects were noted in maximum dive depth (first dives were 25% shallower than average) and buzz rate (first dives contained 34% fewer buzzes per minute than average). In the Azores, the first 3 dives subsequent to tag attachment featured faster descent speeds than average. The whales were likely responding to the cumulative 'dose' of research activity at the surface: Multiple boat approaches, tag placement, and general disturbance. Disturbance should be minimised during tagging, and the extent and duration of responses should be quantified. Modelling of quantified tagging responses could enable correction of these responses in tag data.