John M. Terhune

Underwater detection of tonal signals between 0.125 and 100kHz by harbor seals (Phoca vitulina)

The underwater hearing sensitivities of two 1-year-old female harbor seals were quantified in a pool built for acoustic research, using a behavioral psychoacoustic technique. The animals were trained to respond when they detected an acoustic signal and not to respond when they did not (go/no-go response). Pure tones (0.125-0.25 kHz) and narrowband frequency modulated (tonal) signals (center frequencies 0.5-100 kHz) of 900 ms duration were tested. Thresholds at each frequency were measured using the up-down staircase method and defined as the stimulus level resulting in a 50% detection rate. The audiograms of the two seals did not differ statistically: both plots showed the typical mammalian U-shape, but with a wide and flat bottom. Maximum sensitivity (54 dB re 1 microPa, rms) occurred at 1 kHz. The frequency range of best hearing (within 10 dB of maximum sensitivity) was from 0.5 to 40 kHz (6(1/3) octaves). Higher hearing thresholds (indicating poorer sensitivity) were observed below 1 and above 40 kHz. Thresholds below 4 kHz were lower than those previously described for harbor seals, which demonstrates the importance of using quiet facilities, built specifically for acoustic research, for hearing studies in marine mammals. The results suggest that under unmasked conditions many anthropogenic noise sources and sounds from conspecifics are audible to harbor seals at greater ranges than formerly believed.

Classification of diverse call types using cluster analysis techniques

ABSTRACT We investigated the problem of categorizing the repertoire of a group of highly varied vocalizations. A set of Weddell seal Leptonychotes weddellim-air calls recorded near Davis, Antarctica, were examined. The repertoire size was estimated by first subjectively assigning each call (based on auditory and spectrographic patterns) to one of a large number of provisional call types. A set of frequency, duration and waveform measurements were made on every call. For each provisional call type, the mean value of each of these measures was calculated. These mean values were used to perform Cluster Analyses of the provisional call types. Beginning with calls clustered closest together, the most similar provisional call types were successively amalgamated until further joining would link two clearly dissimilar vocalizations (rising versus falling frequency sweeps in this case). The Weddell seal repertoire contained 12 call types ranging from long sinusoidal upsweeps to growls. This procedure provides a me...

Hearing threshold shifts and recovery in harbor seals (Phoca vitulina) after octave-band noise exposure at 4 kHz

Safety criteria for underwater sounds from offshore pile driving are needed to protect marine mammals. As a first step toward understanding effects of impulsive sounds, two harbor seals were exposed to octave-band white noise centered at 4 kHz at three mean received sound pressure levels (SPLs; 124, 136, and 148 dB re 1 μPa) at up to six durations (7.5, 15, 30, 60, 120, and 240 min); mean received sound exposure level (SEL) range was 166-190 dB re 1 μPa(2) s. Hearing thresholds were determined before and after exposure. Temporary hearing threshold shifts (TTS) and subsequent recovery were quantified as changes in hearing thresholds at 1-4, 4-8, 8-12, 48, and 96 min after noise exposure in seal 01, and at 12-16, 16-20, 20-24, 60, and 108 min after exposure in seal 02. Maximum TTS (1-4 min after 120 min exposure to 148 dB re 1 μPa; 187 dB SEL) was 10 dB. Recovery occurred within ~60 min. Statistically significant TTSs (>2.5 dB) began to occur at SELs of ~170 (136 SPL, 60 min) and 178 dB re 1 μPa(2) s (148 SPL, 15 min). However, SEL is not an optimal predictor of TTS for long duration, low SPL continuous noise, as duration and SPL play unequal roles in determining induced TTS.

Sound Production by Adult Haddock, Melanogrammus Aeglefinus, in Isolation, Pairs and Trios

Haddock, Melanogrammus aeglefinus, have been previously shown to produce sounds during mating. Several behavioural aspects of sound production of courting haddock were further investigated in relation to sex ratio. We assessed whether (i) single males or females generate sounds when isolated, (ii) sound is produced when one male is present with a female, (iii) sound production becomes altered with the introduction of an additional male, and (iv) sounds are produced independent of egg release. Data were collected from 30 March to 11 June 1999, during the spawning period using small outdoor tanks. Sounds generated by captive males during spawning were categorized as knocks, hums and an intermediate between these two types. Solitary males and females did not produce sounds. Sounds were produced when one male was present with a single female. The knocking call duration increased when a second male was introduced. Sounds produced by males occurred independent of the day of egg release.

Repetition patterns in Weddell seal (Leptonychotes weddellii) underwater multiple element calls.

Many vocalizations produced by Weddell seals (Leptonychotes weddellii) are made up of repeated individual distinct sounds (elements). Patterning of multiple element calls was examined during the breeding season at Casey and Davis, Antarctica. Element and interval durations were measured from 405 calls all > 3 elements in length. The duration of the calls (22+/-16.6 s) did not seem to vary with an increasing number of elements (F4,404=1.83,p = 0.122) because element and interval durations decreased as the number of elements within a call increased. Underwater vocalizations showed seven distinct timing patterns of increasing, decreasing, or constant element and interval durations throughout the calls. One call type occurred with six rhythm patterns, although the majority exhibited only two rhythms. Some call types also displayed steady frequency changes as they progressed. Weddell seal multiple element calls are rhythmically repeated and thus the durations of the elements and intervals within a call occur in a regular manner. Rhythmical repetition used during vocal communication likely enhances the probability of a call being detected and has important implications for the extent to which the seals can successfully transmit information over long distances and during times of high level background noise.

An assessment of the audibility of sound from human transport by breeding Weddell seals (Leptonychotes weddellii)

Anthropogenic noise generated through travel in the Antarctic has the potential to affect the region’s wildlife. Weddell seals (Leptonychotes weddellii) in particular can be exposed to anthropogenic noise because they live under, and breed on, the fast ice on which humans travel. To investigate the potential effects of anthropogenic noise on Weddell seals we developed sound profiles for pedestrian travel, over-snow vehicles, aircraft and watercraft operating at various distances and altitudes from breeding seals. The received 1/3-octave noise levels were then related to an assumed detection threshold for the Weddell seal. We found that most noise levels generated by the pedestrian, quad (4-wheeled, all-terrain vehicle) and Hagglunds (tracked, all-terrain vehicle) were commonly categorised in the inaudible and barely audible range of detection (both in-air and underwater), while noise levels generated by the helicopter, Twin Otter aircraft and Zodiac boat were categorised more commonly in the barely audible and clearly audible range. Experimental underwater recordings of vocal behaviour of Weddell seals exposed to continuous low-amplitude over-snow vehicle noise (i.e. Hagglund operation) were also made. Weddell seals underwater did not alter individual call types in response to low-level Hagglunds noise, but they did decrease their calling rate.

Antimasking aspects of harp seal (Pagophilus groenlandicus) underwater vocalizations

Underwater sounds are very important in social communication of harp seals (Pagophilus groenlandicus) because they are the main means of long- and short-distance communication. Individual harp seals must try to avoid being masked and emit only those calls that will benefit them. Underwater vocalizations of harp seals were recorded during the breeding season. The physical characteristics associated with antimasking attributes of 16 call types were examined. Rising frequency or increasing amplitude within calls were not common. Most of the calls ended abruptly (range 145-966 dB/s), but call onset was more gradual. At high calling rates (95.1-135 calls/min) there were significantly more calls overlapping temporally than at medium (75.1-95 calls/min) or low (35-75 calls/min) calling rates, but even at the highest calling rates, 79.1% of the calls were not overlapped. When 2, 3, or 4 calls overlapped, there were significantly fewer frequency separations of less than 1/3 octave than would be expected by chance. This is important because sounds that are separated by less than 1/3 octave likely mask each other. When 2-4 calls are occurring simultaneously, only 4.5% to 14.2% are masked by virtue of being within 1/3 octave from their nearest neighbor. None of the overlappping calls was of the same type. This suggests that the seals are actively listening to each others calls and are not randomly using the different call types. Harp seals use frequency and temporal separation in conjunction with a wide vocal repertoire to avoid masking each other.

Near-threshold equal-loudness contours for harbor seals (Phoca vitulina) derived from reaction times during underwater audiometry: A preliminary study

Equal-loudness functions describe relationships between the frequencies of sounds and their perceived loudness. This pilot study investigated the possibility of deriving equal-loudness contours based on the assumption that sounds of equal perceived loudness elicit equal reaction times (RTs). During a psychoacoustic underwater hearing study, the responses of two young female harbor seals to tonal signals between 0.125 and 100 kHz were filmed. Frame-by-frame analysis was used to quantify RT (the time between the onset of the sound stimulus and the onset of movement of the seal away from the listening station). Near-threshold equal-latency contours, as surrogates for equal-loudness contours, were estimated from RT-level functions fitted to mean RT data. The closer the received sound pressure level was to the 50% detection hearing threshold, the more slowly the animals reacted to the signal (RT range: 188-982 ms). Equal-latency contours were calculated relative to the RTs shown by each seal at sound levels of 0, 10, and 20 dB above the detection threshold at 1 kHz. Fifty percent detection thresholds are obtained with well-trained subjects actively listening for faint familiar sounds. When calculating audibility ranges of sounds for harbor seals in nature, it may be appropriate to consider levels 20 dB above this threshold.

Stereotyped Calling Patterns of a Male Weddell Seal (Leptonychotes weddellii)

Underwater calling behaviour between breathing bouts of a single adult male Weddell seal (Leptonychotes weddellii) was examined with respect to call type and timing late in the breeding season at Davis Station, Antarctica. Underwater calls and breathing sounds were recorded on 1 and 8 December 1997. Thirty-seven sequences of calls prior to surfacing to breathe and 36 post-submerging sets of calls were analyzed with respect to probability of call type occurrence and timing. Dives were 461 ± 259 s (mean ± S.D.). The seal called every 29.7 ± 56.2 s throughout a dive. The first call after submerging was usually (n = 29 of 36) a low frequency (< 0.8 kHz) growl. Three patterns of three- to five-call type sequences were made following 28 of 36 breathing bouts. Call type patterns after submerging exhibited fewer different sequences than those before surfacing (χ 2 = 61.42, DF = 4, p < 0.000001). The call usage patterns before surfacing were diverse and did not indicate when the seal was going to surface, a time when he would be vulnerable to attack from below. Our findings suggest the hypotheses that territorial male Weddell seals call throughout each dive and use stereotyped call patterns to identify themselves while vocally asserting dominance.

EASILY MEASURED CALL ATTRIBUTES CAN DETECT VOCAL DIFFERENCES BETWEEN WEDDELL SEALS FROM TWO AREAS

ABSTRACT Underwater vocalisations of Weddell seals Leptonychotes weddellii were recorded approximately 1400 km apart at Casey and Davis, Antarctica. Recordings were made during the 1992 and 1997 breeding seasons at Davis and during the 1997 season at Casey. Two observers independently analysed four attributes (start and end frequency, duration and number of elements) of narrow bandwidth calls from each location and time. There were few observer differences when the calls were grouped into four broad types (Trills, Descending Whistles, Ascending Whistles and Mews). Ascending Whistles and Mews were rare at Casey but common at Davis. Descending Whistles occurred significantly more often at Davis. Except for Trills, discriminant function analyses indicated less variation between the call attributes at Davis in 1992 and 1997 than between either of the Davis data sets and that from Casey. Vocalisation differences between Weddell seals from different areas can be detected by measuring common attributes of narrow bandwidth calls.