|dc.description.abstract||Anthropogenic noise in the ocean is steadily increasing as more industrial and recreational activities occur in coastal and offshore areas. Responses of marine mammals to noise vary widely, ranging from temporary behavioural change to permanent physiological damage.
For studies of responses to noise, passive acoustic monitoring offers the dual benefit of recording the noise, and an individuals’ vocal response to it. To be maximally effective, it is first necessary to gather basic information about the whales’ acoustic behaviour. Little is known about the vocal behaviour of southern right whales (SRWs, Eubalaena australis) in the sub-Antarctic Auckland Islands. This habitat is the major calving ground for this species in New Zealand waters. To describe the vocal repertoire, 4355 calls were classified into ten call types, including a long tonal low call (up to 25 s long) which had not been described previously. Random Forest multivariate analysis of 28 measured variables was used to classify calls with a high degree of accuracy (82%). The repertoire encompassed a range of tonal and pulsive sounds similar to those produced by other right whale populations.
For data on right whale presence, vocalisation rate and type, an autonomous recorder was moored at the Auckland Islands for a year. Recordings were made in each month except June, and SRW calls were audible in all months with recordings except January. A total of 35,487 calls was detected, of which upcalls were the most common (11,623) in all months and at all times of the day. Call rate was highest in August (288±5.9 [SE] calls/hour) and July (194±8.3). Vocalisation rates were highest at dusk and night, consistent with the idea that upcalls function primarily as contact calls. Seasonal variation was more important than diel variation for explaining differences in vocalisation rate. An automated detector designed for North Atlantic right whales (NARWs, E. glacialis) detected 80% of upcalls, but false detection rates were high, particularly when call rates were low.
To compare the acoustic environments used by NARWs and SRWs, a long-term statistical analysis of ambient noise was carried out at two calving sites in the Atlantic Ocean and one in the Southern Ocean. SRWs at the Auckland Islands were exposed to far lower levels of ambient noise than NARWs on their respective calving grounds. At low frequencies (40 Hz spectrum level) ambient noise was 20 to 30 dB re 1Pa2/Hz higher in the North Atlantic. This may be reflective of high levels of shipping traffic near major ports in the Atlantic or of species differences.
A detailed comparative study of upcalls revealed that NARWs use significantly higher maximum, minimum and peak frequencies (40%, 30% and 9%, respectively) than SRWs. NARW calls were also longer, although this difference was less pronounced. Upcalls of the two species were easily discriminated via a random generalised linear model, with a very low misclassification rate (<0.01%). The differences in ambient noise levels and upcall characteristics support the evidence from the USA that noise has an important influence on the calling behaviour of right whales. Noise pollution could potentially be of consequence for SRWs in New Zealand, especially as they continue to re-colonise the mainland coast where anthropogenic noise is increasing.