Search

A time-frequency contour extraction and classification algorithm was created to analyze humpback whale vocalizations. The algorithm automatically extracted contours of whale vocalization units by searching for gray-level discontinuities in the spectrogram images. The unit-to-unit similarity was quantified by cross-correlating the contour lines. A library of distinctive humpback units was then generated by applying an unsupervised, cluster-based learning algorithm. The purpose of this study was to provide a fast and automated feature selection tool to describe the vocal signatures of animal groups. This approach could benefit a variety of applications such as species description, identification, and evolution of song structures. The algorithm was tested on humpback whale song data recorded at various locations in Hawaii from 2002 to 2003. Results presented in this paper showed low probability of false alarm (0%–4%) under noisy environments with small boat vessels and snapping shrimp. The classification algorithm was tested on a controlled set of 30 units forming six unit types, and all the units were correctly classified. In a case study on humpback data collected in the Auau Chanel, Hawaii, in 2002, the algorithm extracted 951 units, which were classified into 12 distinctive types.

The relationships between acoustic features of target echoes and the cognitive representations of the target formed by an echolocating dolphin will influence the ease with which the dolphin can recognize a target. A blindfolded Atlantic bottlenose dolphin (Tursiops truncatus) learned to match aspect-dependent three-dimensional targets (such as a cube) at haphazard orientations, although with some difficulty. This task may have been difficult because aspect-dependent targets produce different echoes at different orientations, which required the dolphin to have some capability for object constancy across changes in echo characteristics. Significant target-related differences in echo amplitude, rms bandwidth, and distributions of interhighlight intervals were observed among echoes collected when the dolphin was performing the task. Targets could be classified using a combination of energy flux density and rms bandwidth by a linear discriminant analysis and a nearest centroid classifier. Neither statistical model could classify targets without amplitude information, but the highest accuracy required spectral information as well. This suggests that the dolphin recognized the targets using a multidimensional representation containing amplitude and spectral information and that dolphins can form stable representations of targets regardless of orientation based on varying sensory properties.

Snapping shrimp are among the major contributors of biological noise in shallow bays, harbors, and inlets located in temperate and tropical waters. Snapping shrimp sounds can severely limit the use of underwater acoustics by humans and other animals such as dolphins, whales, and pinnepeds. They produce sounds by rapidly closing their frontal chela, or claws, snapping the two ends together to generate a loud click. The acoustics of the species Synalpheus paraneomeris was studied by measuring the sound produced by individual shrimp housed in a small cage located 1 m from an H‐52 broadband hydrophone. Ten clicks from 40 specimens were digitized at a 1‐MHz sample rate and the data stored on disk. Various acoustic parameters such as source level, spectral content, and center frequency were correlated with claw size and body length. Peak‐top‐peak source levels varied from 183 to 189 dB re: 1 μPa. A typical spectrum had a low‐frequency peak between 2 and 5 kHz and energy extending out to 200 kHz.

Porpoises are well known to emit directional ultrasound beams for detecting and tracking preys; however, how they produce and manipulate directional beams are challenging. Here, we investigated physical mechanism of ultrasound beam formation and control of finless porpoise (N. a. sunameri) by using an integrated scheme of computed tomography, tissue and field measurements, and numerical modeling. The results showed that complex acoustic structures in the porpoise’s forehead contributed to producing directional acoustic field. Furthermore, we demonstrated that the skull, air sacs, connective tissue, muscle, and melon constituted a gradient index (GRIN) structure whose density and sound velocity are positively correlated, and thus regulated the directional beam. The removal or compression deformation of the forehead tissues decentralizes energy and widens sound beam, indicating that the forehead tissues as a tunable natural GRIN material significantly impact beam patterns of the finless porpoise. The result...

Deep diving odontocetesodontocetes feed in the deep ocean on mesopelagic squid and fish. The distribution of these prey are poorly known but does affect the distribution of deep diving odontocete predators. Hence, it is important to collect data on the density and composition of potential prey to identify deep sea habitats that attract deep diving odontocetes. We used a DIDSON imaging sonar to investigate the density and size structure of mesopelagic animals along the Kona coast of Hawaii, where beaked, sperm and pilot whales are regularly seen. Data were collected at multiple stations during three research cruises by lowering the sonar in the water column to sample at different depths and monitoring the mesopelagic biomass with an EK60 38 kHz echosounder. Foraging activity of whales was monitored at few stations using passive acoustic recorders and a towed array of hydrophones. We were able to count and size 7068 pelagic animals. Density and size varied in space, with higher densities and bigger animals ...

On-animal suction cups with embedded hydrophones allow examination of how signals on the forehead of echolocating odontocetes relate to the internal anatomical structure and the transmission beampattern. Risso’s dolphin (Grampus griseus) is an interesting species for this investigation due to the presence of a unique vertical groove in the middle of their forehead. In this study, a linear array of six broadband suction cup hydrophones were attached along the forehead groove of an adult female Risso’s dolphin trained to catch freshly thawed dead squid in front of an eight-element far-field hydrophone array. The animal’s movement was simultaneously observed using an underwater video camera. A total of nine successful prey captures were recorded. During each catch, the animal first emitted regular echolocation clicks, which quickly transitioned into buzzes (clicks with distinctively high repetition rate) until prey capture. The amplitude and relative time of arrival of these signals across all channels were ...

Laguna de Terminos, Campeche, Mexico, is one of the largest coastal lagoons in Mexico where snapping shrimp is the main source of biological sound. We measured the environmental noise from 2004 to 2008 in stations distributed homogeneously in this lagoon during 17 sampling periods using a digital audio tape recorder sampling at 48 kHz with 16 bits for 1 minute at each station. The environmental noise in the recordings was measured with a semi-automatic MATLAB routine designed for this purpose. Results indicate that environmental noise in the lagoon changes from one sampling to another, but it changed drastically after the hurricanes of 2005 in some of the stations. This changes in submarine environmental noise helps us in understanding the effects of natural phenomena in the distribution and abundance of marine wildlife, such as snapping shrimps and bottlenose dolphins. [Work supported by CONACyT-Campeche and PAPIIT&PASPA-UNAM.]

Active acoustics is an important tool to quantitatively assess the densities of pelagic organisms. While it is an important tool, it is hard to determine the organism composition recorded acoustically especially in oligotrophic waters where mixed organism assemblages are common. One such mixed stock in the Hawaii Islands is the Deep Scattering Layers (DSL). These layers are made up of micronekton (small squid, fish, and crustaceans) which may be an important component linking primary producers with higher trophic levels. Trawling can identify the composition in these layers but this tool can be costly and unable to sample as much area as active acoustics. Many organisms in the DSL diurnally migrate to the surface between 0 and 200 m to feed. We surveyed these organisms using active acoustics and trawling collected in June 2013 and March 2014 on the R/V Oscar Elton Sette. We compared the composition of organisms, volume by functional group, and total volume in the trawl with the acoustic backscatter to det...

The echolocation transmission beam pattern of a false killer whale (Pseudorca crassidens) was measured in the vertical and horizontal planes. A vertical array of seven broadband miniature hydrophones was used to measure the beam pattern in the vertical plane and a horizontal array of the same hydrophones was used in the horizontal plane. The measurements were performed in the open waters of Kaneohe Bay, Oahu, Hawaii, while the whale performed a target discrimination task. Four types of signals, characterized by their frequency spectra, were measured. Type‐1 signals had a single low‐frequency peak at 40±9 kHz and a low‐amplitude shoulder at high frequencies. Type‐2 signals had a bimodal frequency characteristic with a primary peak at 46±7 kHz and a secondary peak at 88±13 kHz. Type‐3 signals were also bimodal but with a primary peak at 100±7 kHz and a secondary peak at 49±9 kHz. Type‐4 signals had a single high‐frequency peak at 104±7 kHz. The center frequency of the signals were found to be linearly correlated to the peak‐to‐peak source level, increasing with increasing source level. The major axis of the vertical beam was directed slightly downward between 0 and −5°, in contrast to the +5 to 10° for Tursiops and Delphinapterus. The beam in the horizontal plane was directed forward between 0° and −5°. In both planes, the type‐1 signals had the broadest beam pattern, followed by the type‐2 and type‐3 signals, with the narrowest beam pattern being exhibited by the type‐4 signals. The beam pattern in the horizontal plane was much narrower than the beam pattern in the vertical plane.

Suction cup deployed acoustic tags have been used to study the echolocation behavior of a number of odontocetes, yet the relationship between the projected biosonar signals and tag recordings are not known. Acoustic data obtained from these tags consist of the number of clicks emitted, the depth at which the clicks are emitted and the inter-click intervals during a biosonar search. In order to understand the relationship between the emitted signals detected in the front of a dolphin and the signals detected by a tag, a spherical hydrophone was mounted on a wooden model of a tag and mounted on the back of a dolphin via a suction cup. The dolphin was involved in a biosonar discrimination task and the signals were measured 1 m from its blowhole along the acoustic axis of the beam and the acoustic tag placed in five different locations on the animal’s back. The signals recorded by the tag were complex with the first pulse being a high frequency resonance-like signal followed by clicks resembling the outgoing clicks but experiencing reflective interference. The peak-to-peak amplitude of the signals measured by the tag was between 40 and 50 dB lower than that of the outgoing signal.

Previous studies in the Northwestern Hawaiian Islands (NWHI) focused on shallower communities in and near reefs and did not investigate the organisms living in deeper waters that some apex predators rely on for food, e.g., some odontocetes forage at depths greater than 400 m. To examine the relationship between deep-diving odontocete predators and prey, a Simrad EK60 echosounder operating at 70 kHz collected acoustic abundance throughout the NWHI from May 7 to June 4, 2013. Visual and passive acoustic surveys for marine mammal presence were conducted concurrently with the echosounder. Two broad scattering layers were found, a deep layer from 325 to 670 m and a shallow layer from 0 to 195 m. The highest densities of both deep and shallow scattering organisms were associated with deep slopes of banks and atolls. Beaked and short-finned pilot whale sightings occurred in locations of high scattering density associated with slopes of atolls and banks. It is hypothesized that the high scattering organisms associated with these features are similar to the mesopelagic boundary community found in the Main Hawaiian Islands and support a food web representing the prey of the cetaceans.

Bottlenose dolphin echolocation clicks display a great diversity in temporal and spectral structure, with both unimodal and bimodal spectra observed (Houser et al., JASA, 1999). Wavelet scalograms applied to data collected by the Navy Marine Mammal Program and the Bioacoustic Measuring Tool (BMT) (Martin et al., JASA, 2005) show that echolocation clicks can display two distinct phases: an initial “thump,” followed by an extended “ring” that is adequately modeled by a damped harmonic oscillator. The thump and ring can display either similar or different spectral characteristics, giving rise to a unimodal or bimodal spectrum. A three-mass lumped parameter model, adapted from the speech processing and terrestrial bioacoustics literature, has been used to simulate the oscillation and collision of the dorsal bursae in a dolphin’s nasal passage. The three-mass model reproduces many of the time and frequency domain features of entire click trains as well as individual clicks, including unimodal and bimodal spectra. A key insight of the models is that some slight adhesion between the faces of the colliding bursue seems necessary in order to reproduce the high-frequency click structure. A viscoelastic mucus coating could provide one possible mechanism for this required adhesion force. [Data provided by Steve Martin, NMMF.]

Dolphins have biological sonar abilities that exceed those of any man-made system in an aquatic environment. One problem of particular importance, and for which only limited capabilities exist, is the detection and recognition of targets buried under sediment. This paper reviews dolphin echolocation capabilities and describes a system that uses a dolphin-like signal and biomimetic signal processing mechanisms to emulate the performance of the dolpin on such targets. The system employed a digitized dolphin click with a center frequency of 120 kHz and a 3dB bandwidth of 39 kHz, 50 μs duration. This signal was transmitted through seawater into mud and the echoes reflected from the objects were recorded and digitized. Two spectral estimators were used to extract a time-frequency representation of the echo. One was based on short-time fast Fourier transforms, and the other was based on an autoregressive estimator. The time-frequency representation was then processed by a separate backpropagation neural network for each estimator, designed to derive independent identifications of the targets. These identifications were then combined in a modified probabilistic neural network that used a linear transfer function rather than a binary function for its output. Finally the output of the probabilistic network was processed symbolically by a simple expert system. Three experiments are described in which the system was used to discriminate a small stainless-steel cylinder from cyliners of the same size made of hollow aluminium, foam-filled aluminium, or coral rock embedded in resin. Each of the targets was presented buried in mud at a depth of several centimeters. The system proved highly effectively at recognizing these buried targets.

This letter introduces an algorithm for automatic detection of minke whale boing sounds. This method searches for frequency features of boings without calculating the continuous spectrogram of the data, thereby reducing computational time. The detector has been tested on 8 h of acoustic data recorded at the Station ALOHA Cabled Observatory in March 2007. This dataset was previously analyzed using the cross-correlation detector of xbat and was verified by a human listener, as reported in Oswald et al. [(2011). J. Acoust. Soc. Am. 129, 3353–3360]. A comparison of results indicates that the detector introduced here generates fewer false alarms, and it recognizes low-SNR calls that are missed by xbat.

The overall goal of our research is to understand beaked whale foraging and to learn how to alleviate acoustic encounters between Navy assets and beaked whales and other deep diving odontocetes. Understanding of the characteristics and dynamics of the prey field is critical in understanding the foraging behavior and life cycle of beaked whales. The movement patterns of any animals are strongly affected by the availability of food resources, so in order to understand the foraging behavior of beaked whales, the behavior of the prey, the oceanographic conditions affecting the presence of the prey and how the whales interact with the prey field all needs to be better understood.

High-frequency auditory filter shapes of an Atlantic bottlenose dolphin (Tursiops truncatus) were measured using a notched noise masking source centered on pure tone signals at frequencies of 40, 60, 80 and 100 kHz. A dolphin was trained to swim into a hoop station facing the noise/signal transducer located at a distance of 2 m. The dolphin's masked threshold was determined using an up-down staircase method as the width of the notched noise was randomly varied from 0, 0.2, 04, 0.6, and 0.8 times the test tone frequency. The masked threshold decreased as the width of the notched increased and less noise fell within the auditory filter associated with the test tone. The auditory filter shapes were approximated by fitting a roex (p,r(r)) function to the masked threshold results. A constant-Q value of 8.4 modeled the results within the frequency range of 40 to 100 kHz relatively well. However, between 60 and 100 kHz, the 3 dB bandwidth was relatively similar between 9.5 and 10 kHz, indicating a constant-bandwidth system in this frequency range The mean equivalent rectangular bandwidth calculated from the filter shape was approximately 16.0%, 17.0%, 13.6% and 11.3% of the tone frequencies of 40, 60, 80, and 100 kHz.

The use of remote autonomous passive acoustic recorders (PAR) to determine the distribution of dolphins at a given locations has become very popular. Some investigators are using echolocation clicks to gather information on the presence of dolphins and to identify species. However, in all of these cases, the PAR probably recorded mainly off-axis clicks, even some from behind the animals. Yet there is a very poor understanding of the beam pattern and the click waveform and spectrum from different azimuths around the animal’s body. The beam pattern completely around an echo locating dolphin was measured at 16 different but equally spaced angles in the horizontal plane using an 8-hydrophone array in sequence. Eight channels of data were digitized simultaneously at a sampling rate of 500 kHz. The resulting beam patterns in both planes showed a continuous drop off in sound pressure with azimuth around the animal and reached levels below −50 dB relative to the signal recorded on the beam axis. The signals began to break up into two components at angles greater than ± 45° in the horizontal plane. The center frequency dropped off from its maximum at 0° in a non-uniform matter.

Beluga whales (Delphinapterus leucas) in Cook Inlet, Alaska, are listed as endangered and share habitats with a variety of anthropogenic activities including coastal development, oil and gas exploration, shipping, and military activities. Their population has declined from an estimated 653 animals in 1994 to 321 in 2009 (Hobbs et al. 2009). As a result, there is an urgent need for data that will help regulatory agencies such as the National Oceanic and Atmospheric Administration (NOAA) and Alaska’s Department of Fish and Game (ADF&G) implement effective management and recovery plans. Among the principal types of information needed are quantifiable measures of seasonal presence in the inlet, temporal and spatial patterns of habitat preference, and the occurrence of animals in areas impacted or considered for industrial development.

Marine animals use sound for communication, navigation, predator avoidance, and prey detection. Thus the rise in acoustic energy associated with increasing human activity in the ocean has potential to impact the lives of marine animals. Thirty years ago marine bioacoustics primarily focused on evaluating effects of human-generated sound on hearing and behavior by testing captive animals and visually observing wild animals. Since that time rapidly changing electronic and computing technologies have yielded three tools that revolutionized how bioacousticians study marine animals. These tools are (1) portable systems for measuring electrophysiological auditory evoked potentials, (2) miniaturized tags equipped with positioning sensors and acoustic recording devices for continuous short-term acoustical observation rather than intermittent visual observation, and (3) passive acoustic monitoring (PAM) systems for remote long-term acoustic observations at specific locations. The beauty of these breakthroughs is their direct applicability to wild animals in natural habitats rather than only to animals held in captivity. Hearing capabilities of many wild species including polar bears, beaked whales, and reef fishes have now been assessed by measuring their auditory evoked potentials. Miniaturized acoustic tags temporarily attached to an animal to record its movements and acoustic environment have revealed the acoustic foraging behavior of sperm and beaked whales. Now tags are being adapted to fishes in effort to understand their behavior in the presence of noise. Moving and static PAM systems automatically detect and characterize biological and physical features of an ocean area without adding any acoustic energy to the environment. PAM is becoming a powerful technique for understanding and managing marine habitats. This paper will review the influence of these transformative tools on the knowledge base of marine bioacoustics and elucidation of relationships between marine animals and their acoustic environment, leading to a new, rapidly growing field of marine acoustic ecology.

The School of Ocean and Earth Science Technology (SOEST) of the University of Hawai’i, Honolulu, HI, installed a nearshore coral reef observatory in 2005 to study the geochemistry and physics of bentic processes, wave boundary layer processes, water quality, and sediment pore water processes. Kilo Nalu (“observe the wave” in Hawai’ian) provides underwater nodes for data ­connectivity to the shore via a fiberoptic link and the availability of 24-V DC power so that a suite of observational ­instruments can be used to resolve waves, tides, currents, and nearshore water ­quality. The observatory is in the offshore waters of the Kaka’ako Waterfront Park in Honolulu, HI. In 2008, an ­acoustic-­monitoring system was added to monitor the soundscape for biological entities. The Kilo Nalu acoustic-­monitoring sensor (Fig. 1, red dot) is mounted on the bottom on a coral reef at a depth of 20 m.

A previous comparative analysis of normalized click amplitude spectra from a Tursiops truncatus has shown that those frequencies with the lowest click-to-click variability in spectral content were the frequencies the animal paid attention to during target discrimination tasks. In that case, the dolphin only paid attention to the frequency range between 29–42 kHz which had a significantly higher degree of consistency in spectral content than frequencies above 42 kHz. Here it is shown that despite their morphological and behavioral differences, this same pattern of consistency was used by a Pseudorca crassidens performing a similar discrimination task. This comparison between species provides a foundation for using spectral level variability to determine the frequencies most important for echolocation in rare species and non-captive animals. Such results provide key information for successful management.

The long-term goal of our research is to understand beaked whale foraging process and to learn how to alleviate acoustic encounters between Navy assets and beaked whales and other deep diving odontocetes. The more specific goal of this proposal is to fabricate an integrated instrumentation system that can be used to study foraging behavior of deep diving beaked whales. The most effective manner to minimize acoustic encounters is to have knowledge of the movement patterns of beaked whales in any given body of water. Since the movement patterns of any animals is strongly affected by the availability of food resources, it is critical to understand the foraging behavior of beaked whales, the behavior of the prey, the oceanographic conditions affecting the presence of the prey and how the whales interact with the prey field. Our long-term objectives can be summarized as follows: 1. estimate the three-dimensional spatial extent of potential prey field, 2. collect synoptic data of beaked whale foraging on the prey field, 3. determine the taxa composition of the prey field, 4. estimate the size and density of the micronekton in the prey field, 5. correlate relevant oceanographic parameters with the presence of the prey field, 6. map the spatial and temporal pattern of beaked whales in the study area

Minke whales (Balaenoptera acutorostrata) in the tropical North Pacific are elusive and difficult to detect visually. The recent association of a unique sound called the “boing” to North Pacific minke whales has made it possible to use passive acoustics to investigate the occurrence of this species in Hawaiian waters. One year of recordings (17 February 2007–18 February 2008) made at the Station ALOHA Cabled Observatory were examined to investigate the characteristics of boings and temporal patterns in their occurrence at this site, located 100 km north of Oahu. Characteristics of boings exhibited low variability. Pulse repetition rate and duration measurements matched those for “central” or “Hawaii” boing types. Boings were detected from October until May, with a peak in March. Although no boings were detected from June to September, the absence of boings does not necessarily indicate the absence of minke whales. Significant diel variation in boing rate was not observed. The absence of a diel pattern in boing production suggests that day- or night-time acoustic surveys are equally acceptable methods for studying minke whale occurrence. Future research should include efforts to determine what other sounds are produced by minke whales in this area, and which age/sex classes produce boings.

The waters surrounding the State of Hawaii are high U.S. Navy activity regions and include Pearl Harbor on the island of Oahu, the FORACS (Fleet Operational Readiness and Accuracy Check Site) acoustic range operated by the Hawaii Detachment of NUWC-Newark along the Waianae coast of Oahu, and PMRF (Pacific Missile Range) in the waters off the island of Kauai and the shallow waters of west Maui. Furthermore, the RIMPAC exercises (Rim of Pacific Exercises) that occur on a regular basis involve considerable Naval resources operating close to and within Hawaiian waters. The Pacific Navy is under considerable pressure from environmental groups that have initiated up to five lawsuits to curtail the Navy s use of active sonar for training. Perhaps the best approach in combating the various environmental concerns expressed in the different lawsuits is to gather scientific data and obtain important information on the abundance and distributions of marine mammals in the high Navy activity area of Hawaii. The cost of negative encounters with marine mammals is disproportionately high in terms of negative publicity and lawsuits so it would be prudent to take steps to increase the odds against any encounters. Basic information on the biology, natural history, and behavior of dolphins and whales that frequent waters of high Navy activities are needed. Marine mammal surveys around the Hawaiian Islands have been infrequent and localized. The goal of this project was to perform regular, but relatively inexpensive acoustic surveys in the deep waters between Oahu and Hawaii and Oahu and Kauai on a seasonal basis using the assets of the Young Brothers Shipping Company, Hawaii s largest interisland shipping company. Young Brothers offered to participate by allowing us to board their tugboats and use our acoustic instruments to collect data, on a pro-bono basis. The goal was to establish a robust database of information that currently does not exist in the deep waters between islands.

This paper introduces a novel underwater target classification scheme which recognizes underwater ordnances based on their backscattered Time-Frequency (TF) signatures. The objective is to automatically identify the shape and interior content of sea-disposed underwater munitions and ordnance found in the Hawaiian coastlines. This effort helps in the removal of the above sea-disposals from the ocean floor. A dolphin Sonar is used to generate backscattering signals from the targets. Characteristics of different targets are studied using the Wigner-Ville Distribution (WVD), and are discriminated using a cluster-based classification scheme. This method was tested on the backscattered signals from dummy ordnance measured in a water tank.

Echolocation clicks were recorded from an Atlantic bottlenose dolphin Tursiops truncatus trained to discriminate frequency filtered phantom targets in 1998 and in 2004. These clicks showed consistency within their spectra intensity profiles but only in a certain band of frequencies. In 2004 almost all the clicks were consistent within the 0–42 kHz band regardless of the presented target or the click source level. This region corresponded with previous data showing that in 2004 the dolphin perceived frequencies only from within the 29–42 kHz band during echolocation. Above 42 kHz the consistency was lost. In 1998 the consistent region was found only in the 90–100 kHz band showing a shift had occurred with time. This suggests the dolphin’s echolocation strategy for these discrimination tasks centered on the use of clicks with the same controlled standard frequency content in a certain frequency band to investigate different targets. This consistent region shifted over time to maintain maximum signal to noise ratio of the echoes given certain changing limitations to the echolocation system. The shift in consistency over time indicates these consistent regions were not simply artifacts of click production but rather an active control of frequency content.

Spinner dolphins (Stenella longirostris) exhibit different social behaviours at two regions in the Hawaiian Archipelago: off the high volcanic islands in the SE archipelago they form dynamic groups with ever-changing membership, but in the low carbonate atolls in the NW archipelago they form long-term stable groups. To determine whether these environmental and social differences influence population genetic structure, we surveyed spinner dolphins throughout the Hawaiian Archipelago with mtDNA control region sequences and 10 microsatellite loci (n = 505). F-statistics, Bayesian cluster analyses, and assignment tests revealed population genetic separations between most islands, with less genetic structuring among the NW atolls than among the SE high islands. The populations with the most stable social structure (Midway and Kure Atolls) have the highest gene flow between populations (mtDNA Phi(ST)0.001, P = 0.357; microsatellite F(ST) = -0.001; P = 0.597), and a population with dynamic groups and fluid social structure (the Kona Coast of the island of Hawai'i) has the lowest gene flow (mtDNA 0.042Phi(ST)0.236, P0.05; microsatellite 0.016F(ST)0.040, P0.001). We suggest that gene flow, dispersal, and social structure are influenced by the availability of habitat and resources at each island. Genetic comparisons to a South Pacific location (n = 16) indicate that Hawaiian populations are genetically depauperate and isolated from other Pacific locations (mtDNA 0.216F(ST)0.643, P0.001; microsatellite 0.058F(ST)0.090, P0.001); this isolation may also influence social and genetic structure within Hawai'i. Our results illustrate that genetic and social structure are flexible traits that can vary between even closely-related populations.

Publisher Summary In recent years, a number of hypotheses have been developed to explain differences in sound emission behavior among the different species and families of odontocetes. Most odontocetes emit three fundamental types of sounds: whistles, burst pulses, and echolocation clicks. Dusky dolphins (Lagenorhynchus obscurus) are among this group of odontocetes that emit the three basic types of signals. Acoustic energy propagates in water more efficiently than almost any form of energy. Therefore, acoustics offers the most effective method for an animal to perform various functions such as communication, navigation, obstacle avoidance, and prey avoidance and detection. The dolphins' dependency on acoustics is probably the strongest influence in the evolution of the acoustic processing centers within the dolphin brain. Understanding the acoustics of dusky dolphins will increase the understanding of dusky dolphin behavior and natural history, and may even help to elucidate questions of use of acoustics for different habitats, prey and predator relationships, and related delphinid species.

The axiom that knowledge is power applies directly to the problems experienced by the U.S. Navy in encountering dolphins and whales. These encounters can be avoided if more knowledge and understanding of the behavior, distribution, and movements of these animals. Simply stated, if the Navy had more knowledge of the what, where, when and why of marine mammals in a given body of water, encounters between Naval vessels and marine mammals could be reduced or avoided all together. The ocean is large and the chances of avoiding any interaction with any sizable group of marine mammals are probably much greater than the probability of encountering marine mammals. However, the cost of negative encounters is disproportionately high in terms of negative publicity and law suits so it would be prudent to take steps to increase the odds against any encounters. Therefore, basic information on the biology, natural history, and behavior of cetaceans that frequent waters of high Navy activity are needed to understand ways to avoid encounters. A robust database of this information currently does not exist. There is a higher probability of Naval encounters with marine mammals in Hawaiian waters than in most other regions of the world because of the large number of cetacean species that inhabit or frequent these waters. Approximately 16-20 species of cetaceans can be found in Hawaiian waters. This is a large number of species for such a small geographic area. Knowing what animals are present in a given body of water is important because different species utilize their habitat in different ways. Therefore, it is important to understand the distribution, abundance and movement of dolphins and whales over the day-night cycles and seasonal periods.

The biosonar system of dolphins and porpoises has been studied for about 5 decades and much has been learned [Au, W. W. L. (1993). The Sonar of Dolphins (Springer, New York)]. Most experiments have involved human-made targets; little is known about odontocetes' echolocation of prey. To address this issue, acoustic backscatter from Atlantic cod (Gadus morhua), gray mullet (Chelon labrosus), pollack, (Pollachius pollachius), and sea bass (Dicentrarchus labrax) was measured using simulated biosonar signals of the Atlantic bottlenose dolphin and harbor porpoise. The fish specimens were rotated so that the effects of the fish orientation on the echoes could be determined. Echoes had the highest amplitude and simplest structure when the incident angle was perpendicular to the longitudinal axis of the fish. The complexity of the echoes increased as the aspect angle of the fish moved away from the normal aspect. The echoes in both the time and frequency domains were easily distinguishable among the four species of fish and were generally consistent within species. A cochlear model consisting of a bank of band-passed filters was also used to analyze the echoes. The overall results suggest that there are sufficient acoustic cues available to discriminate between the four species of fish based on the echoes received, independent of aspect angle.

Groups of spinner dolphins have been shown to cooperatively herd small prey. It was hypothesized that the strong group coordination is maintained by acoustic communication, specifically by frequency-modulated whistles. Observations of groups of spinner dolphins foraging at night within a sound-scattering layer were made with a multibeam echosounder while the rates of dolphin sounds were measured using four hydrophones at 6 m depth intervals. Whistles were only detected when dolphins were not foraging and when animals were surfacing. Differences in click rates were found between depths and between different foraging stages but were relatively low when observations indicated that dolphins were actively feeding despite the consistency of these clicks with echolocation signals. Highest click rates occurred within the scattering layer, during transitions between foraging states. This suggests that clicks may be used directly or indirectly to cue group movement during foraging, potentially by detecting other individuals' positions in the group or serving a direct communicative role which would be contrary to the existing assumption that echolocation and communication are compartmentalized. Communicating via clicks would be beneficial as the signal's characteristics minimize the chance of eavesdropping by competing dolphins and large fish. Our results are unable to support the established paradigm for dolphin acoustic communication and suggest an alternate coordination mechanism in foraging spinner dolphins.

The small whales called dolphins, such as the bottlenose dolphin Tursiops truncatus, have sensitive, broadband hearing extending to at least 150 kHz. Using ultra-brief, broadband, intense pulses that may reach 230 dB re 1 μPa at 1 m, this dolphin may have a maximum echolocation range of 100–600 m in the ocean. A narrow elongated cochlea coupled to a hypertrophied central auditory nervous system allows for rapid processing of echoes. All dolphins have relatively large brains. Some, not much larger than human body size, may have brains of 1500 g. Much of the large dolphin brain may be related to the need for rapid processing of echoes in water where sound travel is almost five times as fast as in air.

An echolocating dolphin was trained to discriminate between the material composition of two cylinders oriented at different aspect angles. The two targets for the first task were a hollow aluminum and a hollow stainless steel cylinder of the same dimension. For the second task, the targets were the hollow aluminum cylinder and a cylinder of coral rock aggregates encased in degassed epoxy. The dolphin could discriminate the aluminum and steel cylinders perfectly at aspect angles of 0° and 10° (at 0°, the longitudinal axis of a cylinder was perpendicular to the dolphin). The dolphin had its lowest performance (79% and 81% correct) when the cylinders were at aspect angles of 45° and 90°, respectively. The echoes from the aluminum and steel cylinders were the most similar at 45° and 90° the two angles at which the dolphin’s performance was the lowest. The dolphin performed the second discrimination task between the hollow aluminum and coral rock cylinders almost perfectly, at all aspect angles considered. Differences in the occurrence time and amplitude of highlights along with the duration of the echoes seem to be important cues for discrimination of the cylinders. For a simple discrimination task such as the aluminum and coral rock targets, aspect independent information is available for discrimination at arbitrary aspect angles.

A beluga (Delphinapterus leucas) was trained to detect different length cylinders in front of a clutter screen at five separation distances. The clutter screen consisted of 300, 5.1‐cm‐diam cork spheres located behind the targets. The clutter screen was used in two positions: perpendicular to the body axis of the beluga (90° grazing angle) and rotated 22° (68° grazing angle). Five hollow stainless‐steel cylinders (3.2‐cm‐diameter and 0.38‐cm wall thickness) of different lengths (14, 10, 7, 5, and 3 cm) were used as targets. Detection data were collected on the beluga's performance as a function of the separation between targets and the clutter screen. The beluga's performance was above 80% correct detection for the 14‐ and 10‐cm cylinders as the separation distance decreased from 10.1 to 5.1 cm for both grazing angles. For all targets except the 3‐cm cylinder, the beluga's detection was higher at 0‐cm separation than at 2.5‐cm separation. The beluga's detection performance was higher at 68° than at 90° grazing angle. The target strength of the cylinders and the clutter screen was measured in peak‐to‐peak amplitude and the energy of the incident and reflected echoes. For 0‐cm target‐to‐clutter screen separation at 90° grazing angle, the echo‐to‐reverberation ratio at 50% correct detection was − 1.0 dB, based on the peak‐to‐peak measurements, and − 5.1 dB based on the energy measurement. The results of this experiment suggest that a beluga can detect targets in 3.6 to 5.4 dB more reverberation than previously reported for a bottlenose dolphin (Tursiops truncatus).

Broadband simulated dolphin echolocation signals were used to measure the ex situ backscatter properties of mesopelagic boundary community (MBC) in order to gain a better understanding of the echolocation process of spinner dolphins foraging on the MBC. Subjects were captured by trawling with a 2-m-opening Isaacs-Kidd Midwater Trawl. Backscatter measurements were conducted on the ship in a 2000 L seawater tank with the transducer placed on the bottom pointed upwards. Backscatter measurements were obtained in both the dorsal and lateral aspects for seven myctophids and only in the dorsal aspect for 16 more myctophids, six shrimps, and three squids. The echoes from the myctophids and shrimps usually had two highlights, one from the surface of the animal nearest the transducer and a second probably from the signal propagating through body of the subject and reflecting off the opposite surface of the animal. The squid echoes consisted mainly of a single highlight but sometimes had a low amplitude secondary highlight. The backscatter results were used to estimate the echolocation detection range for spinner dolphins foraging on the mesopelagic boundary community. The results were also compared with multi-frequency volume backscatter of the mesopelagic boundary community sound scattering layer.

This study presents the first target strength measurements of Dosidicus gigas, a large squid that is a key predator, a significant prey, and the target of an important fishery. Target strength of live, tethered squid was related to mantle length with values standardized to the length squared of -62.0, -67.4, -67.9, and -67.6 dB at 38, 70, 120, and 200 kHz, respectively. There were relatively small differences in target strength between dorsal and anterior aspects and none between live and freshly dead squid. Potential scattering mechanisms in squid have been long debated. Here, the reproductive organs had little effect on squid target strength. These data support the hypothesis that the pen may be an important source of squid acoustic scattering. The beak, eyes, and arms, probably via the sucker rings, also play a role in acoustic scattering though their effects were small and frequency specific. An unexpected source of scattering was the cranium of the squid which provided a target strength nearly as high as that of the entire squid though the mechanism remains unclear. Our in situ measurements of the target strength of free-swimming squid support the use of the values presented here in D. gigas assessment studies.

Masked underwater pure‐tone thresholds were obtained for an Atlantic bottlenose dolphin using an up–down staircase method of stimulus presentation and a go/no‐go response paradigm. Two types of masking noise were used: a broadband noise and variable bandwidth noise with sharp low‐ and high‐frequency cutoffs. The animal’s critical ratio was measured at frequencies of 30, 60, 90, 100, 110, 120, and 140 kHz. For frequencies of 100 kHz and below, the critical ratios were similar to those measured by Johnson [J. Acoust. Soc. Am. 44, 965–967 (1968)]. The dolphin’s critical bandwidth at frequencies of 30, 60, and 120 kHz was measured with the variable bandwidth noise. The critical bandwidth was 10.4 dB (11 times) wider than the critical ratio at 30 kHz, 8.2 dB (6.6 times) wider at 60 kHz, and 3.5 dB (2.2 times) wider at 120 kHz.

A computational model of the dolphin auditory system was developed to describe how multiple discrimination cues may be represented and employed during echolocation discrimination tasks. The model consisted of a bank of gammatone filters followed by half-wave rectification and low pass filtering. The output of the model resembles a spectrogram; however, the model reflects temporal and spectral resolving properties of the dolphin auditory system. Model outputs were organized to represent discrimination cues related to spectral, temporal and intensity information. Two empirical experiments, a phase discrimination experiment [Johnson et al., Animal Sonar Processes and Performance (Plenum, New York, 1988)] and a cylinder wall thickness discrimination tasks [Au and Pawolski, J. Comp. Physiol. A 170, 41-47 (1992)] were then simulated. Model performance was compared to dolphin performance. Although multiple discrimination cues were potentially available to the dolphin, simulation results suggest temporal information was used in the former experiment and spectral information in the latter. This model's representation of sound provides a more accurate approximation to what the dolphin may be hearing compared to conventional spectrograms, time-amplitude, or spectral representations.

A dolphin's ability to discriminate targets may depend greatly on the relative amplitudes and the time separations of echo highlights within the received signal. Previous experiments with dolphins varied the physical parameters of targets, but did not fully investigate how changes in these parameters correspond with the scattered acoustic wave forms and the dolphin's subsequent response. This experiment utilizes a phantom echo system to test a dolphin's detection response to relative amplitude differences of secondary and trailing echo highlights and the time separation differences of all the echo highlights both within and outside the animal's integration window. By electronically manipulating the amplitude and temporal separation of the echo highlights, the underlying acoustic classification cues are more efficiently investigated. The animal successfully discriminated between a standard echo signal and one with the secondary highlight amplitude lowered by 7 dB from the standard. Furthermore, the animal successfully discriminated between a standard echo signal and one with the trailing highlight amplitude lowered by 3 dB from the standard and also a standard echo signal and one with a time separation of 150 mus between the secondary and trailing highlights.