Acoustic behavior of beaked whales, with implications for acoustic monitoring

Beaked whales (Cetacea: Ziphiidea) of the genera Ziphius and Mesoplodon have been reported to mass strand during naval sonar exercises, and the ability to mitigate such negative effects is of increasing importance. Beaked whales dive for long periods and am. difficult to sight at the surface. We discuss new results suggesting that passive acoustic monitoring for beaked whales holds promise for detecting their presence. A new non-invasive acoustic recording tag was attached to beaked whales of two species, 3 Mesoplodon densirostris and 7 Ziphius cavirostris. High-frequency clicks were recorded for about 30 min every 2-4 hours during deep dives of all tagged whales. The tagged whales only clicked at depths below 200 in, down to a maximum depth near 2000 in. During each foraging dive, both species produced thousands of directional, ultrasonic clicks with no significant energy below 20 kHz. The clicks are produced in trains with mean interclick intervals near 0.4 s. Individual clicks are frequency modulated pulses with durations of 200-300 mu sec and center frequencies around 40 kHz, -10 dB bandwidths of 20-30 kHz. The sound beam for Ziphius is narrow with an estimated directionality index of more than 25 dB, source levels up to 214 dBpp re 1 mu Pa at 1 m. Given knowledge of the source level, frequency range, and directionality of beaked whale clicks, along with the precise orientation and location of beaked whales clicking during foraging dives, it is possible to model the "acoustic footprint" of foraging dives. These data can be used to estimate the probability of detecting the presence of beaked whales by listening. Critical parameters include the density of sensors and the listening time needed to achieve adequate detection probability. Using the observed dive behavior of K, tagged beaked whales of genera Ziphius and Mesoplodon, we worked with the NATO Undersea Research Center to develop a simulation of detector performance for arrays of hydrophones at variable depths between 100 and 1400 in. We demonstrate that for continuous listening for at least the 2-4 hour interval between the start of deep dives, high detection probability could be achieved with acoustic sensors spaced a few km apart. The greatest problem for beaked whale detectors is likely to be false alarms from biological and physical acoustic transients (such as dolphin clicks), and monitoring efforts will need to trade off probability of false alarm against maximizing detection probability. While results may be somewhat poorer in realistic noise conditions and with practical detectors, there appears to be considerable potential for the acoustic detection of beaked