Spatio-temporal variation in click production rates of beaked whales: Implications for passive acoustic density estimation

• Published in: The Journal of the Acoustical Society of America Vol. 141; no. 3; p. 1962
• Main Authors: Warren, Victoria E, Marques, Tiago A, Harris, Danielle, Thomas, Len, Tyack, Peter L, Aguilar de Soto, Natacha, Hickmott, Leigh S, Johnson, Mark P
• Format: Journal Article
• Language: English
• Published: United States 01.03.2017
• Subjects: Acoustics; Animals; Population Density; Signal Processing, Computer-Assisted; Sound Spectrography; Time Factors; Vocalization, Animal - classification; Whales - classification; Whales - physiology
• ISSN: 1520-8524
• DOI: 10.1121/1.4978439

Passive acoustic monitoring has become an increasingly prevalent tool for estimating density of marine mammals, such as beaked whales, which vocalize often but are difficult to survey visually. Counts of acoustic cues (e.g., vocalizations), when corrected for detection probability, can be translated into animal density estimates by applying an individual cue production rate multiplier. It is essential to understand variation in these rates to avoid biased estimates. The most direct way to measure cue production rate is with animal-mounted acoustic recorders. This study utilized data from sound recording tags deployed on Blainville's (Mesoplodon densirostris, 19 deployments) and Cuvier's (Ziphius cavirostris, 16 deployments) beaked whales, in two locations per species, to explore spatial and temporal variation in click production rates. No spatial or temporal variation was detected within the average click production rate of Blainville's beaked whales when calculated over dive cycles (including silent periods between dives); however, spatial variation was detected when averaged only over vocal periods. Cuvier's beaked whales exhibited significant spatial and temporal variation in click production rates within vocal periods and when silent periods were included. This evidence of variation emphasizes the need to utilize appropriate cue production rates when estimating density from passive acoustic data.