Year: 2026
Type: article
Abstract: Many non-human animals produce vocal displays with rhythmic stability. For animals that sing or call underwater, measuring inter-call intervals (ICIs), the time between the starts of consecutive song or call emissions, can be complicated. Sound travels greater distances in the ocean than in air, so underwater recordings often simultaneously capture the vocal displays of multiple individuals. To accurately measure each individual’s ICI, it’s necessary to disentangle their interleaved call sequences. Complex ocean propagation conditions complicate matters further by introducing indirect arrivals into recordings; phantom copies of songs that arise from reflection and refraction. ICI estimations in underwater bioacoustics tend to be constrained to conservative manual processes which are time-consuming, offer substantial room for human error, and limit studies to small sample sizes. We developed an automated ICI estimation method that includes multi-caller separation and multi-path suppression. Our method takes the output of an existing automated call detector and constructs single-caller sequences using TDOA (time difference of arrival) bearing estimation, phase dispersion minimisation, and directed-acyclic longest path optimisation. This pipeline then feeds each single-caller sequence to our ICI estimation algorithm that uses epoch folding and the harmonic $${Z}_{2}^{2}$$ statistic to measure periodicity. We demonstrate the method on a 25-year dataset of audio recordings of the Chagos pygmy blue whale song and show that this population’s ICI is decreasing over time, while individual animals have strong rhythmic consistency within each calling bout. Our new method allows more robust estimation of ICI, with mechanisms for quantifying typical individual rhythmic consistency, and the degree of a population’s ICI stereotypy, facilitating the broader study of long-term change in baleen whale song and other stereotyped animal vocal displays.
Source: Scientific Reports
Authors: Benjamin A. Jancovich, Tracey L. Rogers
Institution: UNSW Sydney
Language: English
Subfield: Ecology
Field: Environmental Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
OpenAlex ID: W7236743933
