Automatic grunt detector and recognizer for Atlantic cod (Gadus morhua) (PDF)

Urazghildiiev, I.R. and S.M. Van Parijs

Journal of the Acoustical Society of America 139: 2532-2540 (2016)

DOI: 10.1121/1.4948569

Northwest Atlantic cod (Gadus morhua) have been heavily overfished in recent years and have not yet recovered. Passive acoustic technology offers a new approach to identify the spatial location of spawning fish, as well as their seasonal and long term persistence in an area. To date, the lack of a species-specific detector has made searching for Atlantic cod grunts in large amounts of passive acoustic data cumbersome. To address this problem, an automatic grunt detection and recognition algorithm that processes yearlong passive acoustic data recordings was designed. The proposed technique is a two-stage hypothesis testing algorithm that includes detecting and recognizing all grunt-like sounds. Test results demonstrated that the algorithm provided a detection probability of 0.93 for grunts with a signal-to-noise ratio (SNR) higher than 10 dB, and a detection probability of 0.8 for grunts with the SNR ranging from 3 to 10 dB. This detector is being used to identify cod in current and historical data from U.S. waters. Its use has significantly reduced the time required to find and validate the presence of cod grunts.

Behavioural responses of western gray whales to a 4-D seismic survey off northeastern Sakhalin Island, Russia (PDF)

Gailey, G., O. Sychenko, T. McDonald, R. Racca, A. Rutenko, and K. Bröker

Endangered Species Research 30: 53-71 (2016)

DOI: 10.3354/esr00713

A seismic survey was conducted off the northeastern coast of Sakhalin Island, Russia in 2010. The survey area was adjacent to the only known near-shore feeding ground of the Critically Endangered population of western gray whales Eschrichtius robustus in the western Pacific south of the Aleutian Islands. This study examined the effectiveness of efforts to minimize the behavioural responses of the whales to vessel proximity and sound during the survey. Two shore-based behavioural observation teams monitored whale movements and respirations pre-, during and post-seismic survey. Theodolite tracking and focal-animal follow methods were used to collect behavioural data. Mixed linear models were used to examine deviations from ‘normal’ patterns in 10 movement and 7 respiration response variables in relation to vessel proximity, vessel/whale relative orientations and 8 received sound metrics to examine if seismic survey sound and/or vessel activity influenced the whales’ behaviour. Behavioural state and water depth were the best ‘natural’ predictors of whale movements and respiration. After considering natural variation, none of the response variables were significantly associated with seismic survey or vessel sounds. A whale’s distance from shore and its orientation relative to the closest vessel were found to be significantly influenced by vessel proximity, which suggested some non-sound related disturbance. The lack of evidence that the whales responded to seismic survey sound and vessel traffic by changing either their movement or respiration patterns could indicate that the current mitigation strategy is effective. However, power analyses suggest that our sample sizes were too small to detect subtle to moderate changes in gray whale behaviour.

Bowhead whale localization using asynchronous hydrophones in the Chukchi Sea (PDF)

Warner, G.A., S.E. Dosso, D.E. Hannay, and J. Dettmer

Journal of the Acoustical Society of America 140: 20-34 (2016)

DOI: 10.1121/1.4954755

This paper estimates bowhead whale locations and uncertainties using non-linear Bayesian inversion of their modally-dispersed calls recorded on asynchronous recorders in the Chukchi Sea, Alaska. Bowhead calls were recorded on a cluster of 7 asynchronous ocean-bottom hydrophones that were separated by 0.5–9.2 km. A warping time-frequency analysis is used to extract relative mode arrival times as a function of frequency for nine frequency-modulated whale calls that dispersed in the shallow water environment. Each call was recorded on multiple hydrophones and the mode arrival times are inverted for: the whale location in the horizontal plane, source instantaneous frequency (IF), water sound-speed profile, seabed geoacoustic parameters, relative recorder clock drifts, and residual error standard deviations, all with estimated uncertainties. A simulation study shows that accurate prior environmental knowledge is not required for accurate localization as long as the inversion treats the environment as unknown. Joint inversion of multiple recorded calls is shown to substantially reduce uncertainties in location, source IF, and relative clock drift. Whale location uncertainties are estimated to be 30–160 m and relative clock drift uncertainties are 3–26 ms.

COMPILE—A generic benchmark case for predictions of marine pile-driving noise

Lippert, S., M. Nijhof, T. Lippert, D. Wilkes, A. Gavrilov, K. Heitmann, M. Ruhnau, O. von Estorff, A. Schäfke, … A. MacGillivray, … M. Wood, et al.

IEEE J. Ocean. Eng. 41: 1061-1071 (2016)

DOI: 10.1109/JOE.2016.2524738

The prediction of underwater noise emissions from impact pile driving during near-shore and offshore construction activities and its potential effect on the marine environment has been a major field of research for several years. A number of different modeling approaches have been suggested recently to predict the radiated sound pressure at different distances and depths from a driven pile. As there are no closed-form analytical solutions for this complex class of problems and for a lack of publicly available measurement data, the need for a benchmark case arises to compare the different approaches. Such a benchmark case was set up by the Institute of Modelling and Computation, Hamburg University of Technology (Hamburg, Germany) and the Organisation for Applied Scientific Research (TNO, The Hague, The Netherlands). Research groups from all over the world, who are involved in modeling sound emissions from offshore pile driving, were invited to contribute to the first so-called COMPILE (a portmanteau combining computation, comparison, and pile) workshop in Hamburg in June 2014. In this paper, the benchmark case is presented, alongside an overview of the seven models and the associated results contributed by the research groups from six different countries. The modeling results from the workshop are discussed, exhibiting a remarkable consistency in the provided levels out to several tens of kilometers. Additionally, possible future benchmark case extensions are proposed.

Gray whale densities during a seismic survey off Sakhalin Island, Russia (PDF)

Muir, J.E., L. Ainsworth, R. Racca, Y. Bychkov, G. Gailey, V. Vladimirov, S. Starodymov, and K. Bröker

Endangered Species Research 29: 211-227 (2016)

DOI: 10.3354/esr00709

Some whale populations that were severely reduced by commercial whaling have shown strong recovery since becoming protected, while others remain depleted and of high conservation concern. Small populations are particularly susceptible to anthropogenic threats, including acoustic disturbance from industrial activities such as seismic surveys. Here, we investigated if sound exposure from a 16 d seismic survey displaced gray whales Eschrichtius robustus from their coastal feeding area off northeastern Sakhalin Island, Russia. We conducted multiple shore-based surveys per day, weather permitting, and created daily 1 km2 density surfaces that provided snapshots of gray whale distribution throughout the seismic activity. A Bayesian spatio-temporal analysis was used to examine possible effects of characteristics of sound exposure from seismic airguns on gray whale occupancy and abundance. Models suggested highest occupancy in areas with moderate sound exposure. Slightly decreased densities were associated with sound exposure when the pattern for the previous 3 d was high sound on Day 2 and low sound on Days 1 and 3. Our findings should be interpreted with caution, given the low number of positive densities. This was due to success of the primary mitigation measure, which was to conduct the seismic survey as early in the feeding season as possible when few gray whales would be present. It is also possible that observed differences in occupancy and density reflect changes in prey availability rather than noise. Prey distribution and abundance data were unavailable for our study, and this important covariate could not be included in models.

Modeling the aggregated exposure and responses of bowhead whales Balaena mysticetus to multiple sources of anthropogenic underwater sound (PDF)

Ellison, W.T., R. Racca, C.W. Clark, B. Streever, A.S. Frankel, E. Fleishman, R. Angliss, J. Berger, D. Ketten, et al.

Endangered Species Research 30: 95-108 (2016)

DOI: 10.3354/esr00727

Potential responses of marine mammals to anthropogenic underwater sound are usually assessed by researchers and regulators on the basis of exposure to a single, relatively loud sound source. However, marine mammals typically receive sounds from multiple, dynamic sources. We developed a method to aggregate modeled sounds from multiple sources and estimate the sound levels received by individuals. To illustrate the method, we modeled the sound fields of 9 sources associated with oil development and estimated the sound received over 47 d by a population of 10000 simulated bowhead whales Balaena mysticetus on their annual migration through the Alaskan Beaufort Sea. Empirical data were sufficient to parameterize simulations of the distribution of individual whales over time and their range of movement patterns. We ran 2 simulations to estimate the sound exposure history and distances traveled by bowhead whales: one in which they could change their movement paths (avert) in response to set levels of sound and one in which they could not avert. When animals could not avert, about 2% of the simulated population was exposed to root mean square (rms) sound pressure levels (SPL) ≥180 dB re 1 µPa, a level that regulators in the U.S. often associate with injury. When animals could avert from sound levels that regulators often associate with behavioral disturbance (rms SPL >160 dB re 1 µPa), <1% of the simulated population was exposed to levels associated with injury. Nevertheless, many simulated bowhead whales received sound levels considerably above ambient throughout their migration. Our method enables estimates of the aggregated level of sound to which populations are exposed over extensive areas and time periods.

Quantitative classification of harbor seal breeding calls in Georgia Strait, Canada (PDF)

Nikolich, K., H. Frouin-Mouy, and A. Acevedo-Gutiérrez

Journal of the Acoustical Society of America 140: 1300-1308 (2016)

DOI: 10.1121/1.4961008

During breeding season, male harbor seals (Phoca vitulina) produce underwater calls used in sexual competition and advertisement. Call characteristics vary among populations, and within-population differences are thought to represent individual variation. However, vocalizations have not been described for several populations of this widely-distributed and genetically diverse species. This study describes the vocal repertoire of harbor seals from British Columbia, Canada. Underwater recordings were made near Hornby Island during the summer of 2014 using a single hydrophone. A wide variability was detected in breeding vocalizations within this single breeding site. Four candidate call types were identified, containing six subtypes. Linear discriminant analysis showed 88% agreement with subjective classification of call types, and 74% agreement for call subtypes. Classification tree analysis gave a 92% agreement with candidate call types, with all splits made on the basis of call duration. Differences in duration may have reflected individual differences among seals. This study suggests that the vocal repertoire of harbor seals in this area comprises a vocal continuum rather than discrete call types. Further work with the ability to localize calls may help to determine whether this complexity represents variability due to propagation conditions, animal orientation, or differences among individual seals.

Short- and long-term monitoring of underwater sound levels in the Hudson River (New York, USA) (PDF)

Martin, S.B. and A.N. Popper

Journal of the Acoustical Society of America 139: 1886-1897 (2016)

DOI: 10.1121/1.4944876

There is a growing body of research on natural and man-made sounds that create aquatic soundscapes. Less is known about the soundscapes of shallow waters, such as in harbors, rivers, and lakes. Knowledge of soundscapes is needed as a baseline against which to determine the changes in noise levels resulting from human activities. To provide baseline data for the Hudson River at the site of the Tappan Zee Bridge, 12 acoustic data loggers were deployed for a 24-h period at ranges of 0–3000 m from the bridge, and four of the data loggers were re-deployed for three months of continuous recording. Results demonstrate that this region of the river is relatively quiet compared to open ocean conditions and other large river systems. Moreover, the soundscape had temporal and spatial diversity. The temporal patterns of underwater noise from the bridge change with the cadence of human activity. Bridge noise (e.g., road traffic) was only detected within 300 m; farther from the bridge, boating activity increased sound levels during the day, and especially on the weekend. Results also suggest that recording near the river bottom produced lower pseudo-noise levels than previous studies that recorded in the river water column.

The combined use of visual and acoustic data collection techniques for winter killer whale (Orcinus orca) observations (PDF)

Burham, R., R. Palm, D. Duffus, X. Mouy, and A. Riera

Global Ecology and Conservation 8: 24-30 (2016)

DOI: 10.1016/j.gecco.2016.08.001

Observations of cetaceans during the winter are difficult, if not impossible in some locations, yet their presence, habitat use, and behaviour during this period are important for conservation and management. Typically, observations come from vessel surveys, with citizen science networks increasingly adding significant sighting data. In compliment to this, acoustic data collection systems can be deployed to collect information remotely over long periods, and in almost any conditions. Here we describe how the combination of these data collection techniques works to fill knowledge gaps, with data from a well-established citizen science network, and a single passive acoustic monitoring (PAM) recorder integrated to identify killer whale presence during winter months in Clayoquot Sound, on the west coast of Vancouver Island.

Together these data show the overwinter use of Clayoquot Sound by killer whales is greater than previously thought. During the study period, February 21 to April 25, 2015, the citizen science network noted 14 visual encounters ranging from Amphitrite Point to Hot Spring Cove, Vancouver Island. The PAM recorded 17 acoustic encounters within the 10 km detection radius of the recorder, deployed off Siwash Point, Flores Island. This included 15 encounters not recorded by the visual network. Both resident and Bigg’s (transient) transient whale groups were recorded, although analysis of vocalizations determined that the majority of the encounters recorded acoustically were of northern resident killer whales. This may be a function of life history, with Bigg’s killer whales typically noted to be less acoustically active, or could represent greater site use by this group. This first use of acoustic monitoring over the winter, complemented with visual data, can establish a better understanding of year-round use of this area by killer whales and has broader application to other sites.

The under-ice soundscape in Great Slave Lake near the city of Yellowknife, Northwest Territories, Canada (PDF)

Martin, S.B. and P.A. Cott

Journal of Great Lakes Research 42: 248-255 (2016)

DOI: 10.1016/j.jglr.2015.09.012

Most recent research and monitoring of under-water “soundscapes” has focused on marine systems in open water conditions. Here we present the first long-term assessment of the diel and seasonal patterns of a fresh-water aquatic soundscape under-ice cover. Acoustic data recorded in Yellowknife Bay, Great Slave Lake in Canada's Northwest Territories, measured the under-ice soundscape near an ice road and airport. From December to late January, the soundscape consisted of geophony from ice cracking and anthrophony from snowmobiles, aircraft, and road vehicles. In late January, burbot spawning calls began and added a localized biophony source to the soundscape that increased the total sound pressure level due to an increase in sound levels in the 10–425 Hz frequency band. The median 1 min root-mean-square sound pressure level (rms SPL) in the period without burbot biophony was 90.3 dB re 1 μPa. The measured hourly rms SPL was negatively correlated with air temperature in the 200–800 Hz band but positively correlated with average hourly wind speed in the 800–8000 Hz band. The nightly mean rms SPL was 88 dB re 1 μPa and increased to 96 dB re 1 μPa in late afternoon. This diel cycle had a strong positive correlation with the number of minutes per hour where ice-road vehicles were detected. Further work is recommended to quantify the soundscape in deep-water areas of large lakes and to include particle motion. Such information will enable the assessment of cumulative impacts of anthrophony and geophony on aquatic biota.

Underwater acoustic behavior of bearded seals (Erignathus barbatus) in the northeastern Chukchi Sea, 2007–2010 (PDF)

Frouin-Mouy, H., X. Mouy, B. Martin, and D. Hannay

Marine Mammal Science 32: 141-160 (2016)

DOI: 10.1111/mms.12246

Bearded seal (Erignathus barbatus) calls were recorded using autonomous passive acoustic recorders deployed in the northeastern Chukchi Sea between October 2007 and October 2010. Continuous acoustic data were acquired during summer (August to mid-October), and overwinter data (mid-October through July) were acquired on a duty cycle of 40/48 min every 4 h. We investigated the spatio-temporal distribution and acoustic behavior of vocalizing bearded seals in this multiyear data set. Peaks in calling occurred in spring, coinciding with the mating period, and calls stopped abruptly in late June/early July. Fewer calls were detected in summer, and the vocal presence of seals increased with the formation of pack ice in winter. Vocal activity was higher at night than during the day, with a peak around 0400 (AKST). Monthly patterns in proportional use of each call type and call duration were examined for the first time. The proportion and duration of AL1(T) and AL2(T) call types increased during the mating period, suggesting that males advertise their breeding condition by producing those specific longer trills. The observed seasonal and diel trends were consistent between years. These results improve our understanding of occurrence and acoustic behavior of bearded seals across the northeastern Chukchi Sea.