The Severity of Behavioral Changes Observed During Experimental Exposures of Killer (Orcinus orca), Long-Finned Pilot (Globicephala melas), and Sperm (Physeter macrocephalus) Whales to Naval Sonar

Abstract

This study describes behavioral changes of wild cetaceans observed during controlled exposures of naval sonar. In 2006 through 2009, 14 experiments were conducted with killer (n = 4), longfinned pilot (n = 6), and sperm (n = 4) whales. A total of 14 6-7 kHz upsweep, 13 1-2 kHz upsweep, and five 1-2 kHz downsweep sonar exposures, as well as seven Silent vessel control exposure sessions and eight playbacks of killer whale sounds were conducted. Sonar signals were transmitted by a towable source that approached each tagged subject from a starting distance of 6 to 8 km with a ramp up of source levels (from 152 to 158 to a maximum of 198 to 214 dB re: 1 µPa m). This procedure resulted in a gradual escalation of the sonar received level at the whale, measured by towed hydrophones and by tags that record movement and sound (Dtags). Observers tracked the position of each tagged animal and recorded group-level surface behavior. Two expert panels independently scored the severity of diverse behavioral changes observed during each sonar and control exposure, using the 0 to 9 point severity scale of Southall et al. (2007), and then reached consensus with a third-party moderator. The most severe responses scored (i.e., most likely to affect vital rates) included a temporary separation of a calf from its group, cessation of feeding or resting, and avoidance movements that continued after the sonar stopped transmitting. Higher severity scores were more common during sonar exposure than during Silent control sessions. Scored responses started at lower sound pressure levels (SPLs) for killer whales and were more severe during sonar exposures to killer and sperm whales than to longfinned pilot whales. Exposure sessions with the higher source level of 1 to 2 kHz sonar had more changes and a trend for higher maximum severity than 6 to 7 kHz sessions, but the order of the sessions had no effect. This approach is helpful to standardize the description of behavioral changes that occurred during our experiments and to identify and describe the severity of potential responses of free-ranging cetaceans to sonar.

Dose-response relationships for the onset of avoidance of sonar by free-ranging killer whales

“…We conducted experiments exposing free-ranging killer whales to sonar in 2006, 2008, and 2009, and details of each experiment are available in a technical report (Miller et al, 2011) and descriptions of behavioral changes provided in Miller et al (2012). Suction cup tags (Dtag; Johnson and Tyack, 2003) that record sound, depth, three-axis accelerometer, and three-axis magnetometer data were attached to six individuals in four different groups, and subjects were tracked from an observation vessel.…”

“…Therefore, during each exposure session, we gradually increased or "escalated" the sonar dose received by the whale in order to maximize our ability to identify a reaction threshold for each exposure session. Detailed inspection of data from each exposed whale was undertaken on a case-by-case basis to identify whether an avoidance reaction took place, and the precise time when the reaction started to be apparent in the data record (Miller et al, 2012). Different sonar dose measures at the time that avoidance started were considered to be the response threshold for that subject whale for that session, and the observed thresholds were used to estimate parameters in Bayesian dose-response functions.…”

Equal latency contours and auditory weighting functions for the harbour porpoise (Phocoena phocoena)

“…Such extrapolations would greatly increase the applicability of behavioural response thresholds for marine mammals in the wild that were measured during exposure to relatively narrowband sources (e.g. Miller et al, 2012;Tyack et al, 2011). The weighting functions may also enable more accurate estimations of the distances from a variety of sound sources at which physiological responses, such as the onset of TTS, and behavioural responses, such as avoidance of the sound source, occur in marine mammals.…”

The Challenges of Analyzing Behavioral Response Study Data: An Overview of the MOCHA (Multi-study OCean Acoustics Human Effects Analysis) Project

“…Many diverse analyses have been applied to data from each BRS, and we do not attempt to review them all here (although see Tyack et al, 2011;Miller et al, 2012;Curé et al, 2012, DeRuiter et al, 2013. Typically the first stage has been to synthesize the observational data for each individual to determine whether it responded, gauge the magnitude of any response and relate the onset of response to exposure intensity ("dose").…”

“…Miller et al (2012) describe this as expert scoring and outline their scoring method in detail as applied to the 3S data. The ideal scenario, however, is to use quantitative methods to identify points in time where behaviour differs from baseline ("changepoints").…”