Search

Automated bioacoustic analysis aids understanding and protection of both marine and terrestrial animals and their habitats across extensive spatiotemporal scales, and typically involves analyzing vast collections of acoustic data. With the advent of deep learning models, classification of important signals from these datasets has markedly improved. These models power critical data analyses for research and decision-making in biodiversity monitoring, animal behaviour studies, and natural resource management. However, deep learning models are often data-hungry and require a significant amount of labeled training data to perform well. While sufficient training data is available for certain taxonomic groups (e.g., common bird species), many classes (such as rare and endangered species, many non-bird taxa, and call-type) lack enough data to train a robust model from scratch. This study investigates the utility of feature embeddings extracted from audio classification models to identify bioacoustic classes other than the ones these models were originally trained on. We evaluate models on diverse datasets, including different bird calls and dialect types, bat calls, marine mammals calls, and amphibians calls. The embeddings extracted from the models trained on bird vocalization data consistently allowed higher quality classification than the embeddings trained on general audio datasets. The results of this study indicate that high-quality feature embeddings from large-scale acoustic bird classifiers can be harnessed for few-shot transfer learning, enabling the learning of new classes from a limited quantity of training data. Our findings reveal the potential for efficient analyses of novel bioacoustic tasks, even in scenarios where available training data is limited to a few samples., (© 2023. The Author(s).)

Understanding relationships between physical drivers and biological response is central to advancing ecological knowledge. Wind is the physical forcing mechanism in coastal upwelling systems, however lags between wind input and biological responses are seldom quantified for marine predators. Lags were examined between wind at an upwelling source, decreased temperatures along the upwelling plume's trajectory, and blue whale occurrence in New Zealand's South Taranaki Bight region (STB). Wind speed and sea surface temperature (SST) were extracted for austral spring-summer months between 2009 and 2019. A hydrophone recorded blue whale vocalizations October 2016-March 2017. Timeseries cross-correlation analyses were conducted between wind speed, SST at different locations along the upwelling plume, and blue whale downswept vocalizations (D calls). Results document increasing lag times (0-2 weeks) between wind speed and SST consistent with the spatial progression of upwelling, culminating with increased D call density at the distal end of the plume three weeks after increased wind speeds at the upwelling source. Lag between wind events and blue whale aggregations (n = 34 aggregations 2013-2019) was 2.09 ± 0.43 weeks. Variation in lag was significantly related to the amount of wind over the preceding 30 days, which likely influences stratification. This study enhances knowledge of physical-biological coupling in upwelling ecosystems and enables improved forecasting of species distribution patterns for dynamic management.

Airports can affect birds by hindering acoustic communication. Here, we investigated the impacts of aircraft events on vocal behavior in wood thrush (Hylocichla mustelina) breeding one mile from an airport in Ithaca, NY, USA. We identified the number of wood thrush songs between 0500 and 0800 h at various distances from the airport and on days with various morning flight schedules. We also analyzed the number of sites from which birds sang during the peak of aircraft events (proxy of number of wood thrush). We found that birds sang more from 0600 to 0640 h when there were aircraft events during this period. This increased vocal behavior is likely explained by increased song output per individual wood thrush, rather than more wood thrush vocalizing. Increased song rate may negatively affect wood thrush fitness through increased energetic demands and/or time tradeoffs with other important behaviors, such as foraging. Identifying the noise thresholds associated with fitness costs (if any) and how different behavioral strategies (i.e. changing the pattern of vocalizations) may allow individuals to evade these costs would be useful for establishing conservation policy in breeding habitats used by passerines, such as the wood thrush.

Early morning calling occurs across diverse taxa, which may be related to optimal conditions for sound transmission. There exists substantial inter- and intra-specific variation in calling time which is influenced by intrinsic, social and/or environmental factors. Here, we investigate environmental predictors of calling in gibbons. We hypothesized that male solos- which occur earlier and tend to be longer than duets-would be more influenced by environmental variables, if earlier, longer calling bouts are energetically costly, and therefore limited by overnight energy expenditure. Our top model for male solo events included amount of rain in the previous 24 hours, and explained 30% of the variance, whereas the top model for duet events (which included presence and amount of rainfall) explained only 5% of the variance. Rain the previous night led to a later start time of male solos (~30 minutes), but our top model for duet start time did not include any reliable predictors. Male solo events appear to be more influenced by environmental factors, and duets may be influenced more by social factors. Our results are in line with previous studies that show that changes in overnight conditions -which may alter energy expenditure -can influence early morning calling behavior.

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.

Investigating long term trends in acoustic communication is essential for understanding the role of sound in social species. Humpback whales are an acoustically plastic species known for producing rapidly-evolving song and a suite of non-song vocalizations ("calls") containing some call types that exhibit short-term stability. By comparing the earliest known acoustic recordings of humpback whales in Southeast Alaska (from the 1970's) with recordings collected in the 1990's, 2000's, and 2010's, we investigated the long-term repertoire stability of calls on Southeast Alaskan foraging grounds. Of the sixteen previously described humpback whale call types produced in Southeast Alaska, twelve were detected in both 1976 and 2012, indicating stability over a 36-year time period; eight call types were present in all four decades and every call type was present in at least three decades. We conclude that the conservation of call types at this temporal scale is indicative of multi-generational persistence and confirms that acoustic communication in humpback whales is comprised of some highly stable call elements in strong contrast to ever-changing song.

Given new distribution patterns of the endangered North Atlantic right whale (NARW; Eubalaena glacialis) population in recent years, an improved understanding of spatio-temporal movements are imperative for the conservation of this species. While so far visual data have provided most information on NARW movements, passive acoustic monitoring (PAM) was used in this study in order to better capture year-round NARW presence. This project used PAM data from 2004 to 2014 collected by 19 organizations throughout the western North Atlantic Ocean. Overall, data from 324 recorders (35,600 days) were processed and analyzed using a classification and detection system. Results highlight almost year-round habitat use of the western North Atlantic Ocean, with a decrease in detections in waters off Cape Hatteras, North Carolina in summer and fall. Data collected post 2010 showed an increased NARW presence in the mid-Atlantic region and a simultaneous decrease in the northern Gulf of Maine. In addition, NARWs were widely distributed across most regions throughout winter months. This study demonstrates that a large-scale analysis of PAM data provides significant value to understanding and tracking shifts in large whale movements over long time scales.