No abstract

PAMGUARD is open-source, platform-independent software to address the needs of developers and users of Passive Acoustic Monitoring (PAM) systems. For the PAM operator—marine mammal biologist, manager, or mitigator—PAMGUARD provides a flexible and easy-to-use suite of detection, localization, data management, and display modules. These provide a standard interface across different platforms with the flexibility to allow multiple detectors to be added, removed, and configured according to the species of interest and the hardware configuration on a particular project. For developers of PAM systems, an Application Programming Interface (API) has been developed which contains standard classes for the efficient handling of many types of data, interfaces to acquisition hardware and to databases, and a GUI framework for data display. PAMGUARD replicates and exceeds the capabilities of earlier real time monitoring programs such as the IFAW Logger Suite and Ishmael. Ongoing developments include improved real-time location and automated species classification. [PAMGUARD funded by the OGP E&P Sound and Marine Life project.]

Summary1. Acoustic monitoring can be an efficient, cheap, non-invasive alternative to physical trapping of individuals. Spatially explicit capture-recapture (SECR) methods have been proposed to estimate calling animal abundance and density from data collected by a fixed array of microphones. However, these methods make some assumptions that are unlikely to hold in many situations, and the consequences of violating these are yet to be investigated. 2. We generalize existing acoustic SECR methodology, enabling these methods to be used in a much wider variety of situations. We incorporate time-of-arrival (TOA) data collected by the microphone array, increasing the precision of calling animal density estimates. We use our method to estimate calling male density of the Cape Peninsula Moss Frog Arthroleptella lightfooti. 3. Our method gives rise to an estimator of calling animal density that has negligible bias, and 95% confidence intervals with appropriate coverage. We show that using TOA information can substantially improve estimate precision. 4. Our analysis of the A. lightfooti data provides the first statistically rigorous estimate of calling male density for an anuran population using a microphone array. This method fills a methodological gap in the monitoring of frog populations and is applicable to acoustic monitoring of other species that call or vocalize.

This review highlights significant gaps in our knowledge of the effects of seismic air gun noise on marine mammals. Although the characteristics of the seismic signal at different ranges and depths and at higher frequencies are poorly understood, and there are often insufficient data to identify the appropriate acoustic propagation models to apply in particular conditions, these uncertainties are modest compared with those associated with biological factors. Potential biological effects of air gun noise include physical/physiological effects, behavioral disruption, and indirect effects associated with altered prey availability. Physical/physiological effects could include hearing threshold shifts and auditory damage as well as non-auditory disruption, and can be directly caused by sound exposure or the result of behavioral changes in response to sounds, e.g. recent observations suggesting that exposure to loud noise may result in decompression sickness. Direct information on the extent to which seismic pulses could damage hearing are difficult to obtain and as a consequence the impacts on hearing remain poorly known. Behavioral data have been collected for a few species in a limited range of conditions. Responses, including startle and fright, avoidance, and changes in behavior and vocalization patterns, have been observed in baleen whales, odontocetes, and pinnipeds and in some case these have occurred at ranges of tens or hundreds of kilometers. However, behavioral observations are typically variable, some findings are contradictory, and the biological significance of these effects has not been measured. Where feeding, orientation, hazard avoidance, migration or social behavior are altered, it is possible that populations could be adversely affected. There may also be serious long-term consequences due to chronic exposure, and sound could affect marine mammals indirectly by changing the accessibility of their prey species. A precautionary approach to management and regulation must be recommended. While such large degrees of uncertainty remain, this may result in restrictions to operational practices but these could be relaxed if key uncertainties are clarified by appropriate research.

Methods for the fully automatic detection and species classification of odontocete whistles are described. The detector applies a number of noise cancellation techniques to a spectrogram of sound data and then searches for connected regions of data which rise above a pre-determined threshold. When tested on a dataset of recordings which had been carefully annotated by a human operator, the detector was able to detect (recall) 79.6% of human identified sounds that had a signal-to-noise ratio above 10 dB, with 88% of the detections being valid. A significant problem with automatic detectors is that they tend to partially detect whistles or break whistles into several parts. A classifier has been developed specifically to work with fragmented whistle detections. By accumulating statistics over many whistle fragments, correct classification rates of over 94% have been achieved for four species. The success rate is, however, heavily dependent on the number of species included in the classifier mix, with the mean correct classification rate dropping to 58.5% when 12 species were included.

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.

No abstract

Acoustic surveys for sperm whales, using line-transect methodology, were carried out in the Ionian Sea and Straits of Sicily, Mediterranean Sea, in 2003. A total of 17 whales were detected along 3846 km of designed survey track in the Ionian Sea, and no whales along 892 km in the Straits of Sicily. This total was insufficient to estimate a detection function, so further data were obtained from quasi-random passages made elsewhere in the western Mediterranean in the same year. The encounters included several tight aggregations with interanimal spacing less than 1 km, primarily from the western Mediterranean. Including individuals from these aggregations distorted the detection function due to the small sample sizes. No such aggregations were found during formal survey of the two areas of interest, and the aggregations were therefore excluded from detection function estimation. The resultant effective strip half-width was 10.0 km (n=40). On the assumption that g(0)=1, the resulting abundance estimates for the Ionian Sea were 62 (with 95% lognormal confidence limits of [24,165]), and 0 for the Straits of Sicily. The low abundance estimate for the Ionian Sea indicates that careful monitoring of the population is needed in the future. During passages along the Hellenic Trench, that were not part of the designed survey, several sperm whales including two aggregations were detected, suggesting that this may be a higher density area and ought to be considered as a separate stratum when designing future surveys.