Ziphius cavirostris presence relative to the vertical and temporal variability of oceanographic conditions in the Southern California Bight.

The oceanographic conditions of the Southern California Bight (SCB) dictate the distribution and abundance of prey resources and therefore the presence of mobile predators, such as goose-beaked whales (Ziphius cavirostris). Goose-beaked whales are deep-diving odontocetes that spend a majority of their time foraging at depth. Due to their cryptic behavior, little is known about how they respond to seasonal and interannual changes in their environment. This study utilizes passive acoustic data recorded from two sites within the SCB to explore the oceanographic conditions that goose-beaked whales appear to favor. Utilizing optimum multiparameter analysis, modeled temperature and salinity data are used to identify and quantify these source waters: Pacific Subarctic Upper Water (PSUW), Pacific Equatorial Water (PEW), and Eastern North Pacific Central Water (ENPCW). The interannual and seasonal variability in goose-beaked whale presence was related to the variability in El Niño Southern Oscillation events and the fraction and vertical distribution of the three source waters. Goose-beaked whale acoustic presence was highest during the winter and spring and decreased during the late summer and early fall. These seasonal increases occurred at times of increased fractions of PEW in the California Undercurrent and decreased fractions of ENPCW in surface waters. Interannual increases in goose-beaked whale presence occurred during El Niño events. These results establish a baseline understanding of the oceanographic characteristics that correlate with goose-beaked whale presence in the SCB. Furthering our knowledge of this elusive species is key to understanding how anthropogenic activities impact goose-beaked whales.

Machine learning with taxonomic family delimitation aids in the classification of ephemeral beaked whale events in passive acoustic monitoring.

Passive acoustic monitoring is an essential tool for studying beaked whale populations. This approach can monitor elusive and pelagic species, but the volume of data it generates has overwhelmed researchers' ability to quantify species occurrence for effective conservation and management efforts. Automation of data processing is crucial, and machine learning algorithms can rapidly identify species using their sounds. Beaked whale acoustic events, often infrequent and ephemeral, can be missed when co-occurring with signals of more abundant, and acoustically active species that dominate acoustic recordings. Prior efforts on large-scale classification of beaked whale signals with deep neural networks (DNNs) have approached the class as one of many classes, including other odontocete species and anthropogenic signals. That approach tends to miss ephemeral events in favor of more common and dominant classes. Here, we describe a DNN method for improved classification of beaked whale species using an extensive dataset from the western North Atlantic. We demonstrate that by training a DNN to focus on the taxonomic family of beaked whales, ephemeral events were correctly and efficiently identified to species, even with few echolocation clicks. By retrieving ephemeral events, this method can support improved estimation of beaked whale occurrence in regions of high odontocete acoustic activity.

Underwater sound speed profile estimation from vessel traffic recordings and multi-view neural networks.

Sound speed is a critical parameter in ocean acoustic studies, as it determines the propagation and interpretation of recorded sounds. The potential for exploiting oceanic vessel noise as a sound source of opportunity to estimate ocean sound speed profile is investigated. A deep learning-based inversion scheme, relying upon the underwater radiated noise of moving vessels measured by a single hydrophone, is proposed. The dataset used for this study consists of Automatic Identification System data and acoustic recordings of maritime vessels transiting through the Santa Barbara Channel between January 2015 and December 2017. The acoustic recordings and vessel descriptors are used as predictors for regressing sound speed for each meter in the top 200 m of the water column, where sound speeds are most variable. Multiple (typically ranging between 4 and 10) transits were recorded each day; therefore, this dataset provides an opportunity to investigate whether multiple acoustic observations can be leveraged together to improve inversion estimates. The proposed single-transit and multi-transit models resulted in depth-averaged root-mean-square errors of 1.79 and 1.55 m/s, respectively, compared to the seasonal average predictions of 2.80 m/s.

Comparing pre-industrial and modern ocean noise levels in the Santa Barbara Channel.

To understand the extent of anthropogenic noise in the ocean, it is essential to compare the differences between modern noise environments and their pre-industrial equivalents. The Santa Barbara Channel, off the coast of Southern California, is a corridor for the transportation of goods to and from the busiest shipping ports in the Western hemisphere. Commercial ships introduce high levels of underwater noise into the marine environment. To quantify the extent of noise in the region, we modeled pre-industrial ocean noise levels, driven by wind, and modern ocean noise levels, resulting from the presence of both ships and wind. By comparing pre-industrial and modern underwater noise levels, the low-frequency (50 Hz) acoustic environment was found to be degraded by more than 15 dB. These results can be used to identify regions for noise reduction efforts, as well as to model scenarios to identify those with the greatest potential to support marine conservation efforts.

Retrofit-induced changes in the radiated noise and monopole source levels of container ships.

The container shipping line Maersk undertook a Radical Retrofit to improve the energy efficiency of twelve sister container ships. Noise reduction, identified as a potential added benefit of the retrofitting effort, was investigated in this study. A passive acoustic recording dataset from the Santa Barbara Channel off Southern California was used to compile over 100 opportunistic vessel transits of the twelve G-Class container ships, pre- and post-retrofit. Post-retrofit, the G-Class vessels' capacity was increased from ~9,000 twenty-foot equivalent units (TEUs) to ~11,000 TEUs, which required a draft increase of the vessel by 1.5 m on average. The increased vessel draft resulted in higher radiated noise levels (<2 dB) in the mid- and high-frequency bands. Accounting for the Lloyd's mirror (dipole source) effect, the monopole source levels of the post-retrofit ships were found to be significantly lower (>5 dB) than the pre-retrofit ships in the low-frequency band and the reduction was greatest at low speed. Although multiple design changes occurred during retrofitting, the reduction in the low-frequency band most likely results from a reduction in cavitation due to changes in propeller and bow design.

SoundScape learning: An automatic method for separating fish chorus in marine soundscapes.

Marine soundscapes provide the opportunity to non-invasively learn about, monitor, and conserve ecosystems. Some fishes produce sound in chorus, often in association with mating, and there is much to learn about fish choruses and the species producing them. Manually analyzing years of acoustic data is increasingly unfeasible, and is especially challenging with fish chorus, as multiple fish choruses can co-occur in time and frequency and can overlap with vessel noise and other transient sounds. This study proposes an unsupervised automated method, called SoundScape Learning (SSL), to separate fish chorus from soundscape using an integrated technique that makes use of randomized robust principal component analysis (RRPCA), unsupervised clustering, and a neural network. SSL was applied to 14 recording locations off southern and central California and was able to detect a single fish chorus of interest in 5.3 yrs of acoustically diverse soundscapes. Through application of SSL, the chorus of interest was found to be nocturnal, increased in intensity at sunset and sunrise, and was seasonally present from late Spring to late Fall. Further application of SSL will improve understanding of fish behavior, essential habitat, species distribution, and potential human and climate change impacts, and thus allow for protection of vulnerable fish species.

Acoustic localization, validation, and characterization of Rice's whale calls.

The recently named Rice's whale in the Gulf of Mexico is one of the most endangered whales in the world, and improved knowledge of spatiotemporal occurrence patterns is needed to support their recovery and conservation. Passive acoustic monitoring methods for determining spatiotemporal occurrence patterns require identifying the species' call repertoire. Rice's whale call repertoire remains unvalidated though several potential call types have been identified. This study uses sonobuoys and passive acoustic tagging to validate the source of potential call types and to characterize Rice's whale calls. During concurrent visual and acoustic surveys, acoustic-directed approaches were conducted to obtain visual verifications of sources of localized sounds. Of 28 acoustic-directed approaches, 79% led to sightings of balaenopterid whales, of which 10 could be positively identified to species as Rice's whales. Long-moan calls, downsweep sequences, and tonal-sequences are attributed to Rice's whales based on these matches, while anthropogenic sources are ruled out. A potential new call type, the low-frequency downsweep sequence, is characterized from tagged Rice's whale recordings. The validation and characterization of the Rice's whale call repertoire provides foundational information needed to use passive acoustic monitoring for better understanding and conservation of these critically endangered whales.

Echolocation click discrimination for three killer whale ecotypes in the Northeastern Pacific.

Three killer whale ecotypes are found in the Northeastern Pacific: residents, transients, and offshores. These ecotypes can be discriminated in passive acoustic data based on distinct pulsed call repertoires. Killer whale acoustic encounters for which ecotypes were assigned based on pulsed call matching were used to characterize the ecotype-specific echolocation clicks. Recordings were made using seafloor-mounted sensors at shallow (∼120 m) and deep (∼1400 m) monitoring locations off the coast of Washington state. All ecotypes' echolocation clicks were characterized by energy peaks between 12 and 19 kHz, however, resident clicks featured sub peaks at 13.7 and 18.8 kHz, while offshore clicks had a single peak at 14.3 kHz. Transient clicks were rare and were characterized by lower peak frequencies (12.8 kHz). Modal inter-click intervals (ICIs) were consistent but indistinguishable for resident and offshore killer whale encounters at the shallow site (0.21-0.22 s). Offshore ICIs were longer and more variable at the deep site, and no modal ICI was apparent for the transient ecotype. Resident and offshore killer whale ecotype may be identified and distinguished in large passive acoustic datasets based on properties of their echolocation clicks, however, transient echolocation may be unsuitable in isolation as a cue for monitoring applications.

Discriminating and classifying odontocete echolocation clicks in the Hawaiian Islands using machine learning methods.

Passive acoustic monitoring (PAM) has proven a powerful tool for the study of marine mammals, allowing for documentation of biologically relevant factors such as movement patterns or animal behaviors while remaining largely non-invasive and cost effective. From 2008-2019, a set of PAM recordings covering the frequency band of most toothed whale (odontocete) echolocation clicks were collected at sites off the islands of Hawai'i, Kaua'i, and Pearl and Hermes Reef. However, due to the size of this dataset and the complexity of species-level acoustic classification, multi-year, multi-species analyses had not yet been completed. This study shows how a machine learning toolkit can effectively mitigate this problem by detecting and classifying echolocation clicks using a combination of unsupervised clustering methods and human-mediated analyses. Using these methods, it was possible to distill ten unique echolocation click 'types' attributable to regional odontocetes at the genus or species level. In one case, auxiliary sightings and recordings were used to attribute a new click type to the rough-toothed dolphin, Steno bredanensis. Types defined by clustering were then used as input classes in a neural-network based classifier, which was trained, tested, and evaluated on 5-minute binned data segments. Network precision was variable, with lower precision occurring most notably for false killer whales, Pseudorca crassidens, across all sites (35-76%). However, accuracy and recall were high (>96% and >75%, respectively) in all cases except for one type of short-finned pilot whale, Globicephala macrorhynchus, call class at Kaua'i and Pearl and Hermes Reef (recall >66%). These results emphasize the utility of machine learning in analysis of large PAM datasets. The classifier and timeseries developed here will facilitate further analyses of spatiotemporal patterns of included toothed whales. Broader application of these methods may improve the efficiency of global multi-species PAM data processing for echolocation clicks, which is needed as these datasets continue to grow.

Identification of western North Atlantic odontocete echolocation click types using machine learning and spatiotemporal correlates.

A combination of machine learning and expert analyst review was used to detect odontocete echolocation clicks, identify dominant click types, and classify clicks in 32 years of acoustic data collected at 11 autonomous monitoring sites in the western North Atlantic between 2016 and 2019. Previously-described click types for eight known odontocete species or genera were identified in this data set: Blainville's beaked whales (Mesoplodon densirostris), Cuvier's beaked whales (Ziphius cavirostris), Gervais' beaked whales (Mesoplodon europaeus), Sowerby's beaked whales (Mesoplodon bidens), and True's beaked whales (Mesoplodon mirus), Kogia spp., Risso's dolphin (Grampus griseus), and sperm whales (Physeter macrocephalus). Six novel delphinid echolocation click types were identified and named according to their median peak frequencies. Consideration of the spatiotemporal distribution of these unidentified click types, and comparison to historical sighting data, enabled assignment of the probable species identity to three of the six types, and group identity to a fourth type. UD36, UD26, and UD28 were attributed to Risso's dolphin (G. griseus), short-finned pilot whale (G. macrorhynchus), and short-beaked common dolphin (D. delphis), respectively, based on similar regional distributions and seasonal presence patterns. UD19 was attributed to one or more species in the subfamily Globicephalinae based on spectral content and signal timing. UD47 and UD38 represent distinct types for which no clear spatiotemporal match was apparent. This approach leveraged the power of big acoustic and big visual data to add to the catalog of known species-specific acoustic signals and yield new inferences about odontocete spatiotemporal distribution patterns. The tools and call types described here can be used for efficient analysis of other existing and future passive acoustic data sets from this region.

Performance metrics for marine mammal signal detection and classification.

Automatic algorithms for the detection and classification of sound are essential to the analysis of acoustic datasets with long duration. Metrics are needed to assess the performance characteristics of these algorithms. Four metrics for performance evaluation are discussed here: receiver-operating-characteristic (ROC) curves, detection-error-trade-off (DET) curves, precision-recall (PR) curves, and cost curves. These metrics were applied to the generalized power law detector for blue whale D calls [Helble, Ierley, D'Spain, Roch, and Hildebrand (2012). J. Acoust. Soc. Am. 131(4), 2682-2699] and the click-clustering neural-net algorithm for Cuvier's beaked whale echolocation click detection [Frasier, Roch, Soldevilla, Wiggins, Garrison, and Hildebrand (2017). PLoS Comp. Biol. 13(12), e1005823] using data prepared for the 2015 Detection, Classification, Localization and Density Estimation Workshop. Detection class imbalance, particularly the situation of rare occurrence, is common for long-term passive acoustic monitoring datasets and is a factor in the performance of ROC and DET curves with regard to the impact of false positive detections. PR curves overcome this shortcoming when calculated for individual detections and do not rely on the reporting of true negatives. Cost curves provide additional insight on the effective operating range for the detector based on the a priori probability of occurrence. Use of more than a single metric is helpful in understanding the performance of a detection algorithm.

A machine learning pipeline for classification of cetacean echolocation clicks in large underwater acoustic datasets.

Machine learning algorithms, including recent advances in deep learning, are promising for tools for detection and classification of broadband high frequency signals in passive acoustic recordings. However, these methods are generally data-hungry and progress has been limited by challenges related to the lack of labeled datasets adequate for training and testing. Large quantities of known and as yet unidentified broadband signal types mingle in marine recordings, with variability introduced by acoustic propagation, source depths and orientations, and interacting signals. Manual classification of these datasets is unmanageable without an in-depth knowledge of the acoustic context of each recording location. A signal classification pipeline is presented which combines unsupervised and supervised learning phases with opportunities for expert oversight to label signals of interest. The method is illustrated with a case study using unsupervised clustering to identify five toothed whale echolocation click types and two anthropogenic signal categories. These categories are used to train a deep network to classify detected signals in either averaged time bins or as individual detections, in two independent datasets. Bin-level classification achieved higher overall precision (>99%) than click-level classification. However, click-level classification had the advantage of providing a label for every signal, and achieved higher overall recall, with overall precision from 92 to 94%. The results suggest that unsupervised learning is a viable solution for efficiently generating the large, representative training sets needed for applications of deep learning in passive acoustics.

Underwater noise mitigation in the Santa Barbara Channel through incentive-based vessel speed reduction.

Commercial shipping is the dominant source of low-frequency noise in the ocean. It has been shown that the noise radiated by an individual vessel depends upon the vessel's speed. This study quantified the reduction in source levels (SLs) and sound exposure levels (SELs) for ships participating in two variations of a vessel speed reduction (VSR) program. SLs and SELs of individual ships participating in the program between 2014 and 2017 were statistically lower than non-participating ships (p < 0.001). In the 2018 fleet-based program, there were statistical differences between the SLs and SELs of fleets that participated with varying degrees of cooperation. Significant reductions in SL and SEL relied on cooperation of 25% or more in slowing vessel speed. This analysis highlights how slowing vessel speed to 10 knots or less is an effective method in reducing underwater noise emitted from commercial ships.

Using context to train time-domain echolocation click detectors.

This work demonstrates the effectiveness of using humans in the loop processes for constructing large training sets for machine learning tasks. A corpus of over 57 000 toothed whale echolocation clicks was developed by using a permissive energy-based echolocation detector followed by a machine-assisted quality control process that exploits contextual cues. Subsets of these data were used to train feed forward neural networks that detected over 850 000 echolocation clicks that were validated using the same quality control process. It is shown that this network architecture performs well in a variety of contexts and is evaluated against a withheld data set that was collected nearly five years apart from the development data at a location over 600 km distant. The system was capable of finding echolocation bouts that were missed by human analysts, and the patterns of error in the classifier consist primarily of anthropogenic sources that were not included as counter-training examples. In the absence of such events, typical false positive rates are under ten events per hour even at low thresholds.

An empirical model for wind-generated ocean noise.

An empirical model for wind-generated underwater noise is presented that was developed using an extensive dataset of acoustic field recordings and a global wind model. These data encompass more than one hundred years of recording-time and capture high wind events, and were collected both on shallow continental shelves and in open ocean deep-water settings. The model aims to explicitly separate noise generated by wind-related sources from noise produced by anthropogenic sources. Two key wind-related sound-generating mechanisms considered are: surface wave and turbulence interactions, and bubble and bubble cloud oscillations. The model for wind-generated noise shows small frequency dependence (5 dB/decade) at low frequencies (10-100 Hz), and larger frequency dependence (∼15 dB/decade) at higher frequencies (400 Hz-20 kHz). The relationship between noise level and wind speed is linear for low wind speeds (<3.3 m/s) and increases to a higher power law (two or three) at higher wind speeds, suggesting a transition between surface wave/turbulence and bubble source mechanisms. At the highest wind speeds (>15 m/s), noise levels begin to decrease at high frequencies (>10 kHz), likely due to interaction between bubbles and screening of noise radiation in the presence of high-density bubble clouds.

Cetacean distribution models based on visual and passive acoustic data.

Distribution models are needed to understand spatiotemporal patterns in cetacean occurrence and to mitigate anthropogenic impacts. Shipboard line-transect visual surveys are the standard method for estimating abundance and describing the distributions of cetacean populations. Ship-board surveys provide high spatial resolution but lack temporal resolution and seasonal coverage. Stationary passive acoustic monitoring (PAM) employs acoustic sensors to sample point locations nearly continuously, providing high temporal resolution in local habitats across days, seasons and years. To evaluate whether cross-platform data synthesis can improve distribution predictions, models were developed for Cuvier's beaked whales, sperm whales, and Risso's dolphins in the oceanic Gulf of Mexico using two different methods: generalized additive models and neural networks. Neural networks were able to learn unspecified interactions between drivers. Models that incorporated PAM datasets out-performed models trained on visual data alone, and joint models performed best in two out of three cases. The modeling results suggest that, when taken together, multiple species distribution models using a variety of data types may support conservation and management of Gulf of Mexico cetacean populations by improving the understanding of temporal and spatial species distribution trends.

DetEdit: A graphical user interface for annotating and editing events detected in long-term acoustic monitoring data.

Passive acoustic monitoring has become an important data collection method, yielding massive datasets replete with biological, environmental and anthropogenic information. Automated signal detectors and classifiers are needed to identify events within these datasets, such as the presence of species-specific sounds or anthropogenic noise. These automated methods, however, are rarely a complete substitute for expert analyst review. The ability to visualize and annotate acoustic events efficiently can enhance scientific insights from large, previously intractable datasets. A MATLAB-based graphical user interface, called DetEdit, was developed to accelerate the editing and annotating of automated detections from extensive acoustic datasets. This tool is highly-configurable and multipurpose, with uses ranging from annotation and classification of individual signals or signal-clusters and evaluation of signal properties, to identification of false detections and false positive rate estimation. DetEdit allows users to step through acoustic events, displaying a range of signal features, including time series of received levels, long-term spectral averages, time intervals between detections, and scatter plots of peak frequency, RMS, and peak-to-peak received levels. Additionally, it displays either individual, or averaged sound pressure waveforms, and power spectra within each acoustic event. These views simultaneously provide analysts with signal-level detail and encounter-level context. DetEdit creates datasets of signal labels for further analyses, such as training classifiers and quantifying occurrence, abundances, or trends. Although designed for evaluating underwater-recorded odontocete echolocation click detections, DetEdit can be adapted to almost any stereotyped impulsive signal. Our software package complements available tools for the bioacoustic community and is provided open source at https://github.com/MarineBioAcousticsRC/DetEdit.

Automated classification of dolphin echolocation click types from the Gulf of Mexico.

Delphinids produce large numbers of short duration, broadband echolocation clicks which may be useful for species classification in passive acoustic monitoring efforts. A challenge in echolocation click classification is to overcome the many sources of variability to recognize underlying patterns across many detections. An automated unsupervised network-based classification method was developed to simulate the approach a human analyst uses when categorizing click types: Clusters of similar clicks were identified by incorporating multiple click characteristics (spectral shape and inter-click interval distributions) to distinguish within-type from between-type variation, and identify distinct, persistent click types. Once click types were established, an algorithm for classifying novel detections using existing clusters was tested. The automated classification method was applied to a dataset of 52 million clicks detected across five monitoring sites over two years in the Gulf of Mexico (GOM). Seven distinct click types were identified, one of which is known to be associated with an acoustically identifiable delphinid (Risso's dolphin) and six of which are not yet identified. All types occurred at multiple monitoring locations, but the relative occurrence of types varied, particularly between continental shelf and slope locations. Automatically-identified click types from autonomous seafloor recorders without verifiable species identification were compared with clicks detected on sea-surface towed hydrophone arrays in the presence of visually identified delphinid species. These comparisons suggest potential species identities for the animals producing some echolocation click types. The network-based classification method presented here is effective for rapid, unsupervised delphinid click classification across large datasets in which the click types may not be known a priori.

Delphinid echolocation click detection probability on near-seafloor sensors.

The probability of detecting echolocating delphinids on a near-seafloor sensor was estimated using two Monte Carlo simulation methods. One method estimated the probability of detecting a single click (cue counting); the other estimated the probability of detecting a group of delphinids (group counting). Echolocation click beam pattern and source level assumptions strongly influenced detectability predictions by the cue counting model. Group detectability was also influenced by assumptions about group behaviors. Model results were compared to in situ recordings of encounters with Risso's dolphin (Grampus griseus) and presumed pantropical spotted dolphin (Stenella attenuata) from a near-seafloor four-channel tracking sensor deployed in the Gulf of Mexico (25.537°N 84.632°W, depth 1220 m). Horizontal detection range, received level and estimated source level distributions from localized encounters were compared with the model predictions. Agreement between in situ results and model predictions suggests that simulations can be used to estimate detection probabilities when direct distance estimation is not available.

Passive acoustic monitoring of beaked whale densities in the Gulf of Mexico.

Beaked whales are deep diving elusive animals, difficult to census with conventional visual surveys. Methods are presented for the density estimation of beaked whales, using passive acoustic monitoring data collected at sites in the Gulf of Mexico (GOM) from the period during and following the Deepwater Horizon oil spill (2010-2013). Beaked whale species detected include: Gervais' (Mesoplodon europaeus), Cuvier's (Ziphius cavirostris), Blainville's (Mesoplodon densirostris) and an unknown species of Mesoplodon sp. (designated as Beaked Whale Gulf - BWG). For Gervais' and Cuvier's beaked whales, we estimated weekly animal density using two methods, one based on the number of echolocation clicks, and another based on the detection of animal groups during 5 min time-bins. Density estimates derived from these two methods were in good general agreement. At two sites in the western GOM, Gervais' beaked whales were present throughout the monitoring period, but Cuvier's beaked whales were present only seasonally, with periods of low density during the summer and higher density in the winter. At an eastern GOM site, both Gervais' and Cuvier's beaked whales had a high density throughout the monitoring period.