Deep neural networks have advanced the field of detection and classification and allowed for effective identification of signals in challenging data sets. Numerous time-critical conservation needs may benefit from these methods. We developed and empirically studied a variety of deep neural networks to detect the vocalizations of endangered North Atlantic right whales (Eubalaena glacialis). We compared the performance of these deep architectures to that of traditional detection algorithms for the primary vocalization produced by this species, the upcall. We show that deep-learning architectures are capable of producing false-positive rates that are orders of magnitude lower than alternative algorithms while substantially increasing the ability to detect calls. We demonstrate that a deep neural network trained with recordings from a single geographic region recorded over a span of days is capable of generalizing well to data from multiple years and across the species' range, and that the low false positives make the output of the algorithm amenable to quality control for verification. The deep neural networks we developed are relatively easy to implement with existing software, and may provide new insights applicable to the conservation of endangered species.

Several marine autonomous recording units (MARUs) were deployed in northeastern Gulf of Mexico from 2010–2012 to study the acoustic ecology of Bryde's whales (Balaenoptera edeni) following the Deepwater Horizon oil spill. However, the acoustic repertoire of this sub-population is poorly documented, presently limiting the efficacy of acoustic monitoring applications. Numerous stereotyped, low-frequency signals from a putative biological sound source were found throughout the recordings. Sounds fell into three categories distinguished by spectral and temporal properties. Multiple calls overlapped temporally on individual MARUs, suggesting that multiple sources produced these sounds. The basic features are similar to those from other mysticetes, but they differ from any previously published sounds. Since Bryde's whales are the most common mysticete in the Gulf and have previously been observed within the recording area on multiple occasions, it is likely that Bryde's whales are the most probable source of these sounds. These results potentially identify a suite of previously undocumented calls from Bryde's whales, which could facilitate future passive acoustic monitoring efforts to better understand the population dynamics and status of this sub-population.

1. The population of bottlenose dolphins in eastern Scotland has undergone significant range expansion since the 1990s, when a Special Area of Conservation was established for the population.2. Distribution of this population is well described within areas of its range where intensive work has been carried out, such as the inner Moray Firth, St Andrews Bay and the Tay estuary area. However, elsewhere in their range, habitat use is less well understood.3. In this study, a large-scale and long-term passive acoustic array was used to gain a better understanding of bottlenose dolphin habitat use in eastern Scottish waters, complementing and augmenting existing visual surveys. 4. Data from the array were analysed using a three-stage approach. First, acoustic occupancy results were reported; second, temporal trends were modelled; and third, a spatial-temporal-habitat model of acoustic occupancy was created. 5. Results from the acoustic occupancy are in agreement with visual studies that found that areas near known foraging locations were consistently occupied. Results from the temporal trend analysis were inconclusive. Habitat modelling showed that, throughout their range, bottlenose dolphins are most likely to be detected closer to shore, and at a constant distance from shore, in deeper water.

Passive acoustic monitoring is an efficient way to study acoustically active animals but species identification remains a major challenge. C-PODs are popular logging devices that automatically detect odontocete echolocation clicks. However, the accompanying analysis software does not distinguish between delphinid species. Click train features logged by C-PODs were compared to frequency spectra from adjacently deployed continuous recorders. A generalized additive model was then used to categorize C-POD click trains into three groups: broadband click trains, produced by bottlenose dolphin (Tursiops truncatus) or common dolphin (Delphinus delphis), frequency-banded click trains, produced by Risso's (Grampus griseus) or white beaked dolphins (Lagenorhynchus albirostris), and unknown click trains. Incorrect categorization rates for broadband and frequency banded clicks were 0.02 (SD 0.01), but only 30% of the click trains met the categorization threshold. To increase the proportion of categorized click trains, model predictions were pooled within acoustic encounters and a likelihood ratio threshold was used to categorize encounters. This increased the proportion of the click trains meeting either the broadband or frequency banded categorization threshold to 98%. Predicted species distribution at the 30 study sites matched well to visual sighting records from the region.