Search

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.

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.

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.

A concern for applications of machine learning techniques to bioacoustics is whether or not classifiers learn the categories for which they were trained. Unfortunately, information such as characteristics of specific recording equipment or noise environments can also be learned. This question is examined in the context of identifying delphinid species by their echolocation clicks. To reduce the ambiguity between species classification performance and other confounding factors, species whose clicks can be readily distinguished were used in this study: Pacific white-sided and Risso's dolphins. A subset of data from autonomous acoustic recorders located at seven sites in the Southern California Bight collected between 2006 and 2012 was selected. Cepstral-based features were extracted for each echolocation click and Gaussian mixture models were used to classify groups of 100 clicks. One hundred Monte-Carlo three-fold experiments were conducted to examine classification performance where fold composition was determined by acoustic encounter, recorder characteristics, or recording site. The error rate increased from 6.1% when grouped by acoustic encounter to 18.1%, 46.2%, and 33.2% for grouping by equipment, equipment category, and site, respectively. A noise compensation technique reduced error for these grouping schemes to 2.7%, 4.4%, 6.7%, and 11.4%, respectively, a reduction in error rate of 56%-86%.

At least ten species of beaked whales inhabit the North Pacific, but little is known about their abundance, ecology, and behavior, as they are elusive and difficult to distinguish visually at sea. Six of these species produce known species-specific frequency modulated (FM) echolocation pulses: Baird's, Blainville's, Cuvier's, Deraniyagala's, Longman's, and Stejneger's beaked whales. Additionally, one described FM pulse (BWC) from Cross Seamount, Hawai'i, and three unknown FM pulse types (BW40, BW43, BW70) have been identified from almost 11 cumulative years of autonomous recordings at 24 sites throughout the North Pacific. Most sites had a dominant FM pulse type with other types being either absent or limited. There was not a strong seasonal influence on the occurrence of these signals at any site, but longer time series may reveal smaller, consistent fluctuations. Only the species producing BWC signals, detected throughout the Pacific Islands region, consistently showed a diel cycle with nocturnal foraging. By comparing stranding and sighting information with acoustic findings, we hypothesize that BWC signals are produced by ginkgo-toothed beaked whales. BW43 signal encounters were restricted to Southern California and may be produced by Perrin's beaked whale, known only from Californian waters. BW70 signals were detected in the southern Gulf of California, which is prime habitat for Pygmy beaked whales. Hubb's beaked whale may have produced the BW40 signals encountered off central and southern California; however, these signals were also recorded off Pearl and Hermes Reef and Wake Atoll, which are well south of their known range.

To study delphinid near surface movements and behavior, two L-shaped hydrophone arrays and one vertical hydrophone line array were deployed at shallow depths (<125 m) from the floating instrument platform R/P FLIP, moored northwest of San Clemente Island in the Southern California Bight. A three-dimensional propagation-model based passive acoustic tracking method was developed and used to track a group of five offshore killer whales (Orcinus orca) using their emitted clicks. In addition, killer whale pulsed calls and high-frequency modulated (HFM) signals were localized using other standard techniques. Based on these tracks sound source levels for the killer whales were estimated. The peak to peak source levels for echolocation clicks vary between 170-205 dB re 1 μPa @ 1 m, for HFM calls between 185-193 dB re 1 μPa @ 1 m, and for pulsed calls between 146-158 dB re 1 μPa @ 1 m.

Beaked whale echolocation signals are mostly frequency-modulated (FM) upsweep pulses and appear to be species specific. Evolutionary processes of niche separation may have driven differentiation of beaked whale signals used for spatial orientation and foraging. FM pulses of eight species of beaked whales were identified, as well as five distinct pulse types of unknown species, but presumed to be from beaked whales. Current evidence suggests these five distinct but unidentified FM pulse types are also species-specific and are each produced by a separate species. There may be a relationship between adult body length and center frequency with smaller whales producing higher frequency signals. This could be due to anatomical and physiological restraints or it could be an evolutionary adaption for detection of smaller prey for smaller whales with higher resolution using higher frequencies. The disadvantage of higher frequencies is a shorter detection range. Whales echolocating with the highest frequencies, or broadband, likely lower source level signals also use a higher repetition rate, which might compensate for the shorter detection range. Habitat modeling with acoustic detections should give further insights into how niches and prey may have shaped species-specific FM pulse types.

This study presents a system for classifying echolocation clicks of six species of odontocetes in the Southern California Bight: Visually confirmed bottlenose dolphins, short- and long-beaked common dolphins, Pacific white-sided dolphins, Risso's dolphins, and presumed Cuvier's beaked whales. Echolocation clicks are represented by cepstral feature vectors that are classified by Gaussian mixture models. A randomized cross-validation experiment is designed to provide conditions similar to those found in a field-deployed system. To prevent matched conditions from inappropriately lowering the error rate, echolocation clicks associated with a single sighting are never split across the training and test data. Sightings are randomly permuted before assignment to folds in the experiment. This allows different combinations of the training and test data to be used while keeping data from each sighting entirely in the training or test set. The system achieves a mean error rate of 22% across 100 randomized three-fold cross-validation experiments. Four of the six species had mean error rates lower than the overall mean, with the presumed Cuvier's beaked whale clicks showing the best performance (<2% error rate). Long-beaked common and bottlenose dolphins proved the most difficult to classify, with mean error rates of 53% and 68%, respectively.

Spectral parameters were used to discriminate between echolocation clicks produced by three dolphin species at Palmyra Atoll: melon-headed whales (Peponocephala electra), bottlenose dolphins (Tursiops truncatus) and Gray's spinner dolphins (Stenella longirostris longirostris). Single species acoustic behavior during daytime observations was recorded with a towed hydrophone array sampling at 192 and 480 kHz. Additionally, an autonomous, bottom moored High-frequency Acoustic Recording Package (HARP) collected acoustic data with a sampling rate of 200 kHz. Melon-headed whale echolocation clicks had the lowest peak and center frequencies, spinner dolphins had the highest frequencies and bottlenose dolphins were nested in between these two species. Frequency differences were significant. Temporal parameters were not well suited for classification. Feature differences were enhanced by reducing variability within a set of single clicks by calculating mean spectra for groups of clicks. Median peak frequencies of averaged clicks (group size 50) of melon-headed whales ranged between 24.4 and 29.7 kHz, of bottlenose dolphins between 26.7 and 36.7 kHz, and of spinner dolphins between 33.8 and 36.0 kHz. Discriminant function analysis showed the ability to correctly discriminate between 93% of melon-headed whales, 75% of spinner dolphins and 54% of bottlenose dolphins.

Acoustic recordings from Palmyra Atoll, northern Line Islands, central Pacific, showed upsweep frequency modulated pulses reminiscent of those produced by beaked whales. These signals had higher frequencies, broader bandwidths, longer pulse durations and shorter inter-pulse intervals than previously described pulses of Blainville's, Cuvier's and Gervais' beaked whales [Zimmer et al. (2005). J. Acoust. Soc. Am. 117, 3919-3927; Johnson et al. (2006). J. Exp. Biol. 209, 5038-5050; Gillespie et al. (2009). J. Acoust. Soc. Am. 125, 3428-3433]. They were distinctly different temporally and spectrally from the unknown beaked whale at Cross Seamount, HI [McDonald et al. (2009). J. Acoust. Soc. Am. 125, 624-627]. Genetics on beaked whale specimens found at Palmyra Atoll suggest the presence of a poorly known beaked whale species. Mesoplodon sp. might be the source of the FM pulses described in this paper. The Palmyra Atoll FM pulse peak frequency was at 44 kHz with a -10 dB bandwidth of 26 kHz. Mean pulse duration was 355 mus and inter-pulse interval was 225 ms, with a bimodal distribution. Buzz sequences were detected with inter-pulse intervals below 20 ms and unmodulated spectra, with about 20 dB lower amplitude than prior FM pulses. These clicks had a 39 kHz bandwidth (-10 dB), peak frequency at 37 kHz, click duration 155 mus, and inter-click interval between 4 and 10 ms.

The spectral and temporal properties of echolocation clicks and the use of clicks for species classification are investigated for five species of free-ranging dolphins found offshore of southern California: short-beaked common (Delphinus delphis), long-beaked common (D. capensis), Risso's (Grampus griseus), Pacific white-sided (Lagenorhynchus obliquidens), and bottlenose (Tursiops truncatus) dolphins. Spectral properties are compared among the five species and unique spectral peak and notch patterns are described for two species. The spectral peak mean values from Pacific white-sided dolphin clicks are 22.2, 26.6, 33.7, and 37.3 kHz and from Risso's dolphins are 22.4, 25.5, 30.5, and 38.8 kHz. The spectral notch mean values from Pacific white-sided dolphin clicks are 19.0, 24.5, and 29.7 kHz and from Risso's dolphins are 19.6, 27.7, and 35.9 kHz. Analysis of variance analyses indicate that spectral peaks and notches within the frequency band 24-35 kHz are distinct between the two species and exhibit low variation within each species. Post hoc tests divide Pacific white-sided dolphin recordings into two distinct subsets containing different click types, which are hypothesized to represent the different populations that occur within the region. Bottlenose and common dolphin clicks do not show consistent patterns of spectral peaks or notches within the frequency band examined (1-100 kHz).

A method for the automatic classification of free-ranging delphinid vocalizations is presented. The vocalizations of short-beaked and long-beaked common (Delphinus delphis and Delphinus capensis), Pacific white-sided (Lagenorhynchus obliquidens), and bottlenose (Tursiops truncatus) dolphins were recorded in a pelagic environment of the Southern California Bight and the Gulf of California over a period of 4 years. Cepstral feature vectors are extracted from call data which contain simultaneous overlapping whistles, burst-pulses, and clicks from a single species. These features are grouped into multisecond segments. A portion of the data is used to train Gaussian mixture models of varying orders for each species. The remaining call data are used to test the performance of the models. Species are predicted based upon probabilistic measures of model similarity with test segment groups having durations between 1 and 25 s. For this data set, 256 mixture Gaussian mixture models and segments of at least 10 s of call data resulted in the best classification results. The classifier predicts the species of groups with 67%-75% accuracy depending upon the partitioning of the training and test data.

Many terrestrial and marine species have a diel activity pattern, and their acoustic signaling follows their current behavioral state. Whistles and echolocation clicks on long-term recordings produced by melon-headed whales (Peponocephala electra) at Palmyra Atoll indicated that these signals were used selectively during different phases of the day, strengthening the idea of nighttime foraging and daytime resting with afternoon socializing for this species. Spectral features of their echolocation clicks changed from day to night, shifting the median center frequency up. Additionally, click received levels increased with increasing ambient noise during both day and night. Ambient noise over a wide frequency band was on average higher at night. The diel adjustment of click features might be a reaction to acoustic masking caused by these nighttime sounds. Similar adaptations have been documented for numerous taxa in response to noise. Or it could be, unrelated, an increase in biosonar source levels and with it a shift in center frequency to enhance detection distances during foraging at night. Call modifications in intensity, directionality, frequency, and duration according to echolocation task are well established for bats. This finding indicates that melon-headed whales have flexibility in their acoustic behavior, and they collectively and repeatedly adapt their signals from day- to nighttime circumstances.