Adam S. Frankel

A New Context‐Based Approach to Assess Marine Mammal Behavioral Responses to Anthropogenic Sounds

Acute effects of anthropogenic sounds on marine mammals, such as from military sonars, energy development, and offshore construction, have received considerable international attention from scientists, regulators, and industry. Moreover, there has been increasing recognition and concern about the potential chronic effects of human activities (e.g., shipping). It has been demonstrated that increases in human activity and background noise can alter habitats of marine animals and potentially mask communications for species that rely on sound to mate, feed, avoid predators, and navigate. Without exception, regulatory agencies required to assess and manage the effects of noise on marine mammals have addressed only the acute effects of noise on hearing and behavior. Furthermore, they have relied on a single exposure metric to assess acute effects: the absolute sound level received by the animal. There is compelling evidence that factors other than received sound level, including the activity state of animals exposed to different sounds, the nature and novelty of a sound, and spatial relations between sound source and receiving animals (i.e., the exposure context) strongly affect the probability of a behavioral response. A more comprehensive assessment method is needed that accounts for the fact that multiple contextual factors can affect how animals respond to both acute and chronic noise. We propose a three-part approach. The first includes measurement and evaluation of context-based behavioral responses of marine mammals exposed to various sounds. The second includes new assessment metrics that emphasize relative sound levels (i.e., ratio of signal to background noise and level above hearing threshold). The third considers the effects of chronic and acute noise exposure. All three aspects of sound exposure (context, relative sound level, and chronic noise) mediate behavioral response, and we suggest they be integrated into ecosystem-level management and the spatial planning of human offshore activities.

Quantifying Loss of Acoustic Communication Space for Right Whales in and around a U.S. National Marine Sanctuary

The effects of chronic exposure to increasing levels of human-induced underwater noise on marine animal populations reliant on sound for communication are poorly understood. We sought to further develop methods of quantifying the effects of communication masking associated with human-induced sound on contact-calling North Atlantic right whales (Eubalaena glacialis) in an ecologically relevant area (~10,000 km(2) ) and time period (peak feeding time). We used an array of temporary, bottom-mounted, autonomous acoustic recorders in the Stellwagen Bank National Marine Sanctuary to monitor ambient noise levels, measure levels of sound associated with vessels, and detect and locate calling whales. We related wind speed, as recorded by regional oceanographic buoys, to ambient noise levels. We used vessel-tracking data from the Automatic Identification System to quantify acoustic signatures of large commercial vessels. On the basis of these integrated sound fields, median signal excess (the difference between the signal-to-noise ratio and the assumed recognition differential) for contact-calling right whales was negative (-1 dB) under current ambient noise levels and was further reduced (-2 dB) by the addition of noise from ships. Compared with potential communication space available under historically lower noise conditions, calling right whales may have lost, on average, 63-67% of their communication space. One or more of the 89 calling whales in the study area was exposed to noise levels ≥120 dB re 1 μPa by ships for 20% of the month, and a maximum of 11 whales were exposed to noise at or above this level during a single 10-min period. These results highlight the limitations of exposure-threshold (i.e., dose-response) metrics for assessing chronic anthropogenic noise effects on communication opportunities. Our methods can be used to integrate chronic and wide-ranging noise effects in emerging ocean-planning forums that seek to improve management of cumulative effects of noise on marine species and their habitats.

Depth dependent variation of the echolocation pulse rate of bottlenose dolphins (Tursiops truncatus).

Trained odontocetes appear to have good control over the timing (pulse rate) of their echolocation clicks; however, there is comparatively little information about how free-ranging odontocetes modify their echolocation in relation to their environment. This study investigates echolocation pulse rate in 14 groups of free-ranging bottlenose dolphins (Tursiops truncatus) at a variety of depths (2.4-30.1 m) in the Gulf of Mexico. Linear regression models indicated a significant decrease in mean pulse rate with mean water depth. Pulse rates for most groups were multi-modal. Distance to target estimates were as high as 91.8 m, assuming that echolocation was produced at a maximal rate for the target distance. A 5.29-ms processing lag time was necessary to explain the pulse rate modes observed. Although echolocation is likely reverberation limited, these results support the hypotheses that free-ranging bottlenose dolphins in this area are adapting their echolocation signals for a variety of target detection and ranging purposes, and that the target distance is a function of water depth.

Whistle source levels of free-ranging bottlenose dolphins and Atlantic spotted dolphins in the Gulf of Mexico

Whistles of bottlenose dolphins (Tursiops truncatus) and Atlantic spotted dolphins (Stenella frontalis) in the eastern Gulf of Mexico were recorded and measured with a calibrated towed hydrophone array. Surveys encountered groups of both bottlenose (N = 10) and spotted dolphins (N = 5). Analysis of those data produced 1695 bottlenose dolphin whistles and 1273 spotted dolphin whistles with a high signal-to-noise ratio. Whistle frequency metrics were lower in bottlenose than spotted dolphins, while whistle duration was longer in spotted dolphins, data that may help inform automatic classification algorithms. Source levels were estimated by determining the range and bearing of an individual dolphin from the array and then adding the predicted transmission loss to the calculated received level. The median bottlenose dolphin source level was 138 dB re 1μPa at 1 m with a range of 114-163 dB re 1μPa at 1 m. The median spotted dolphin source level was 138 dB re 1μPa at 1 m with a range of 115-163 dB re 1μPa at 1 m. These source level measurements, in conjunction with estimates of vocalization rates and transmission loss models, can be used to improve passive acoustically determined dolphin abundance estimates in the Gulf of Mexico.

A common sense approach to source metrics.

The analysis and assessment of the impact of anthropogenic sound on the ocean environment require a clear understanding of the spatial, spectral, and temporal properties of the sources that generate the sounds and the animals that are exposed. The mantra “spatial, spectral, and temporal” is one that applies to all acoustic assessment problems and should serve as the underlying basis for the analysis toolbox anyone brings to bear on these issues. Table 1 correlates the salient aspects of the three features of sound production, transmission, and reception with this mantra.

Monitoring the acoustic effects of pile driving for the first offshore wind farm in the United States

The Block Island Wind Farm, the first offshore wind farm in the United States, consists of five turbines in water depths of approximately 30 m. The turbines have a jacket type substructure with piles driven to the bottom to pin the structure to the seabed. A number of acoustic sensors were deployed to monitor the acoustic properties of the pile driving activity. The acoustic sensor systems consisted of an eight element towed hydrophone array, two fixed moorings with four hydrophones each, and a fixed sensor package for measuring particle velocity. The towed array was towed from 1 to 8 km from the pile driving location. The fixed moorings were deployed at 7.5 and 15 km from the pile location. The particle velocity sensor package was deployed at 500 m from the pile driving location. This sensor package consisted of a three-axis geophone on the seabed and a tetrahedral array of four low sensitivity hydrophones at 1 m from the bottom. Data collected on these sensor systems will be presented. Acoustic levels a...

Results and insights from operational acoustic monitoring networks.

Data from two acoustic monitoring networks operating off New England in an area frequented by whales reveal acoustic features of those habitats. These seafloor and moored systems continuously sample the acoustic environment, and resultant data provide mechanisms for mapping, quantifying, and describing the spatio‐spectral‐temporal variability of the acoustic habitat over ecologically meaningful scales. By focusing on species‐specific frequency bands used by fin, humpback, and right whales for long‐range communication, we are beginning to measure the acoustic dynamics of their primary communication channels. Results reveal the extent to which different sources of sound in the ocean, both natural and man‐made, influence the probability of whale communication. In some habitats with high rates of vessel traffic and high levels of vessel noise, the predicted area over which animals can communicate is reduced to a small proportion of what it would be under quiet conditions. The dynamics of this masking effect a...

Dive patterns and source levels of sperm whales off New Zealand

Diving sperm whales off Kaikoura, NZ were recorded using a 4‐element vertical linear hydrophone array deployed from a small boat. The boat was frequently within close proximity to the whale dove. Whales began to click almost immediately after leaving the surface. Dive patterns of whales were determined by estimating the location of each click using an advanced time‐of‐arrival algorithm developed at Cornell University. Analysis of dive patterns showed near vertical descents with speeds ranging from 1–2 m/s. Received click levels were determined and source levels were estimated using a spherical‐spreading loss term, as the ranges to the whales were less than the water depth. Preliminary estimates of peak‐to‐peak source levels range from 191 to 201 dB re: 1 Pa at 1 m (peak‐to‐peak). On one occasion, a sperm whale clicked rapidly at close range (∼8 m) from a hydrophone element. Click rates as high as 22/s suggest that the whale was interrogating the metal weight at the end of the array. The source level of th...

Interindividual variation in the songs of humpback whales.

The function of humpback whale song remains elusive. Humpbacks primarily sing on the wintering grounds. Many hypotheses suggest a male advertisement function. It has been suggested that all whales of a population sing the same version of song, and its structure slowly evolves during and between singing seasons. Field observations of 1989 song indicated substantial variation in the song of humpbacks. The question remained, were the differences between individuals greater than those between successive songs of an individual. The songs of 11 whales were sampled during an 8 day period, to reduce the probability that song evolution would affect the results. Six units from three themes were sampled from four songs from each whale. Unit duration, lowest frequency, bandwidth, frequency of peak amplitude, and source level were measured. A repeated measures ANOVA found significant differences between individual whales and no such differences were found between successive songs of the same individual. These data ind...

Effects of scaled ATOC playbacks on the behavior of humpback whales in Hawaii

A shore‐based team observed humpback whale behavior and movements before, during, and after a playback of the ATOC m‐sequence. Eighty‐four trials were conducted with both ATOC and no‐sound control conditions. Shore observers did not know which sound condition was used. Playbacks originated from a projector suspended beneath a moored boat. The source level was 172 dB re: 1 μPa (60‐ to 90‐Hz band). A second drifting vessel measured received levels and ambient noise. Received levels were measured between 105 and 130 dB. No overt responses were noted during the experiment. Movement and behavioral variables were examined for changes due to playback. Comparison of whale tracks between control and experimental conditions showed no difference in distribution. Preliminary analysis of the data found no changes in the movement variables and a slight change in respiratory behavior. Whales appear to have longer surface intervals during playback then during control trials. The magnitude of this effect on humpbacks appe...