In recent years, there has been a growing interest in the study of marine mammal vocalisations, particularly to understand their behavioural implications and to explore novel linguistic potentials. Among marine mammals, sperm whales (Physeter macrocephalus) have been identified to possess a highly intricate and diverse vocal repertoire, using sequences of clicks known as codas. In this research, we explore temporal dependencies between the codas within longer bouts as a first step towards understanding their longer vocal exchanges. Our results from analysing simultaneously emitted coda sequences from sperm whale pairs demonstrate that the structure of the next coda produced by the second whale is predictable and influenced by a sequence of up to 10 preceding codas from the first whale. The predictive features include the coda duration, the first and second inter-click-interval, and the duration of the delay between the codas. We explore the level of dependencies between the coda sequences by mutual information analysis. The prediction level is compared for a hidden Markov model, a multi-layer Long-Short-Term Memory (LSTM) and transformers models. Our findings provide insight into the complexity of sperm whale vocalisation interactions.
Anthropogenic soundscape disturbance can constrain acoustic signalling, with potential consequences for reproductive behaviour in wild animals. We used passive acoustic monitoring to investigate how habitat, weather, and human-associated soundscape disturbance are associated with roaring activity of Iberian red deer (Cervus elaphus hispanicus) during the rut on Lous & atilde; Mountain, central Portugal. Twenty-nine AudioMoth recorders collected 150-s recordings every 7.5 min over 14 days, yielding 88,022 manually validated roars. We quantified anthrophony as relative acoustic energy in the 1-2 kHz band and modelled roar counts using a negative binomial generalised linear mixed model. Roaring activity was highest in shrublands and other open habitats, and decreased with increasing wind speed, rainfall and temperature. Importantly, roaring declined with increasing anthrophony and increased with distance from wind turbines, suggesting a possible avoidance of a broader disturbance gradient rather than only a purely acoustic effect. Daily roaring counts also tended to be lower towards the end of the week. Our results highlight that soundscape disturbance associated with human infrastructure and activity may influence both the location and timing of males' vocalisations during the rut, demonstrating the value of soundscape metrics for identifying rutting areas potentially sensitive to disturbance and informing appropriate mitigation strategies.
Passive Acoustic Monitoring (PAM) offers a powerful approach for detecting and assessing the presence of invasive species, thereby supporting the conservation of native ecosystems. In this study, we developed a species-specific classification model using convolutional neural networks (CNNs) to analyse the nocturnal calling activity patterns of Pelophylax nigromaculatus and Dryophytes leopardus in rice paddies, a microhabitat where interactions between translocated and native species are of ecological concern. Despite environmental noise from bird calls and wind, the mel-spectrogram-based model classified anuran vocalisations with high accuracy (88.45%). Misclassifications at dawn were mitigated by limiting the ecological analysis to specific nocturnal periods. The results revealed clearly distinct peak times of nocturnal calling activity between the two species at the study site. These findings demonstrate the usefulness of deep learning for describing fine-scale activity patterns in field-recorded amphibian soundscapes. This study provides methodological insights into the acoustic monitoring of domestically translocated and native species and highlights the potential of bioacoustics approaches for ecological assessment in paddy environments. Future research should focus on refining species classification models and integrating sound-source separation for more accurate species-specific assessments of calling activity patterns.
Anthropogenic noise can vary greatly in several properties, including timing, frequency and amplitude. However, many studies have historically focused on the effects of high amplitude noise, even though 'intermediate' amplitudes of noise may also impact behaviour. In this study, we investigated the effects of low (<= 50 dBA), intermediate (50-60 dBA) and high (>= 60 dBA) amplitudes of anthropogenic noise on the behaviour of wild Western Australian magpies (Gymnorhina tibicen dorsalis). We found that the time magpies spent foraging and vigilant differed significantly under all noise conditions, with greater amplitudes of noise having a larger impact. However, while the vocalisation rate of magpies was significantly reduced under high compared to low noise conditions, there was no difference between vocalisation rates under low and intermediate, or intermediate and high noise conditions. Conversely, while foraging efficiency was significantly reduced under both high and intermediate compared to low noise conditions there was no difference in foraging efficiency between intermediate and high noise conditions suggesting that intermediate levels of noise may impact foraging efficiency in a manner indistinguishable from higher amplitudes. Our results highlight the importance of investigating the impacts of multiple amplitudes of anthropogenic noise on animal behaviour, as different levels of noise may differentially impact wildlife.
Sound pollution affects reef organisms, interfering with various key behaviours such as territoriality and reproduction. Soundscape ecology can reveal the effects of sounds produced by anthropophony on reef ecosystems and guide mitigation strategies. This study provides a baseline assessment and comparison of marine soundscapes in reef environments of Northeastern Brazil, focusing on the influence of sound pollution across different types and seasons of tourism in marine protected areas. Recordings were conducted at distinct sites, covering different exposures, types and seasons of tourism. Sound pollution influenced soundscapes, with higher levels of sound pollution sources being associated with a lower occurrence of biophonic sounds. PSD analysis revealed three main energy peaks across the frequency range studied (0.05 to 10 kHz), but a flattened profile was identified in sites with a higher incidence of sound pollution, resulting in a loss of the sound peak detected at lower frequencies. Sound pollution also affected ACI values, both spatially and temporally. Exposure was the most influential factor, followed by tourism season and tourism type. These results reinforce the role of sound pollution as a significant factor in reef environments, underscoring the urgent need for increased attention, further research, and the development of effective mitigation strategies.
The soundscape is an integral component of natural environments, but it can pose acoustic challenges for wildlife monitoring when undesirable noise interferes with detection. Here, we present a simple approach to reduce the influence of background noise on point count surveys. Using N-mixture models in a Bayesian framework, we estimated detection probability, relative abundance, and population trends for eight bird species and one mammal. Models incorporated acoustic covariates and were fit to 12,048 five-minute point counts from 746 locations from 2019 to 2024. Observers rated ambient noise on a 1-10 scale in all years and used a smartphone app to measure soundscape volume in 2019. Each species was modelled separately for detections within <= 50 m and >50 m from the observer. We hypothesised that both sound metrics would reduce the detection probability, especially for distant detections. Ambient noise was informative in more cases than soundscape volume. Contrary to expectations, soundscape volume more strongly affected detections within 50 m, whereas ambient noise had a greater effect beyond 50 m. Our findings suggest that incorporating soundscape assessments can help correct undercounting bias in relative abundance estimates, improve the comparability of point count surveys, and be easily implemented with minimal logistical burden.
Environmental factors shape animal signals by influencing how sound propagates through natural habitats. The acoustic adaptation hypothesis predicts that signal features promoting effective propagation should be favoured, yet the role of habitat-mediated propagation in shaping song-type prevalence remains unclear. In songbird populations with multiple song types, some are commonly shared while others are rare. We tested the hypothesis that common song types in a population of Bachman's Sparrows possess acoustic features that enhance propagation relative to rare song types. We quantified the acoustic features of common and rare song types and conducted sound propagation experiments. Common song types had shorter inter-syllable gaps, higher peak frequency, broader frequency bandwidth, and higher maximum frequency than rare song types. Consistent with predictions, common song types propagated more effectively for two of five propagation metrics, exhibiting higher amplitude envelope correlation and lower tail-to-signal ratios. Tree density showed a non-significant trend in which common song types exhibited higher amplitude envelope correlation as tree density increased, whereas wind speed and signal height did not influence propagation. Together, these results provide partial support for the hypothesis that propagation efficiency contributes to song-type prevalence, yet social processes during song learning likely play an equal or greater role.
This study investigates the ecology and evolution of dove acoustic communication and applies the findings to reconstruct the acoustic niche of the critically endangered Paraclaravis geoffroyi. We analysed 1187 calls from 14 dove species of the southern Atlantic Forest, using acoustic parameters to perform a Principal Component Analysis (PCA) that illustrates species distributions within acoustic space. A multinomial model evaluated the uniqueness of each species' acoustic niche. Dove species have diverse vocalisations, and the PCA analysis reveals overlap between the calls of ground-dwelling forest species, posing semantic challenges for species recognition. Using the sole available recording of Paraclaravis geoffroyi and the calls of its sister species, P. mondetoura, we simulated its acoustic niche and used it to create new sound recordings using audio editing and music notation tools. Our model achieved 88% accuracy in classifying species, and highlight that most species have distinct calls. P. geoffroyi occupies a unique acoustic space segment. This study offers insights into how environmental adaptation shapes dove acoustic communication and introduces innovative methods to recreate calls of rare or extinct species.
Identifying the signal features that receivers use to respond to the vocalisations of conspecifics is one of the main goals of animal communication studies. In this study, we analysed the temporal structure of alarm vocalisations given by southern house wrens (Troglodytes musculus) when they detect a risk near the nest and evaluated the response of conspecifics to sequences of calls that vary in the organisation of signals in time (rhythmicity). House wrens typically respond to perceived risks by repetitively calling a simple vocalisation. We performed an experiment using a plastic owl model placed at different distances from the nest, and we found that, as risk increased, house wrens called at a higher rate, with less regular and shorter inter-call intervals. We also performed a playback experiment using isochronous, rhythmic, and arrhythmic (random) sequences of a similar number of calls. Receivers did not respond to changes in the rhythmicity of the calling, although they tended to move closer to the speaker when we played back arrhythmic calls. We propose that house wrens may use mainly a simpler temporal calling feature, such as the interval between calls, to assess the perceived risk.
While humpback whale songs are known for their complexity, a quantitative understanding of their temporal structure remains elusive. We apply Multifractal Detrended Fluctuation Analysis (MFDFA) to a large historical dataset of songs spanning four dec-ades (1950s-1990s). Our analysis confirms the songs are robustly multifractal but reveals significant heterogeneity, with distinct sig-natures strongly correlated with the recording year. Notably, 1992 recordings exhibit a markedly different dynamic higher complexity (omega) and strong persistence (alpha(0)>1:0) in stark contrast to the anti- persistent nature (alpha(0)<0:5) of 1950s recordings. These differences are partially confounded by variations in recording quality (SNR), as our characterisation of the ambient background shows it to be a highly persistent, yet strictly monofractal, process. This work thus provides quantitative evidence for long-term shifts in song dynamics, while highlighting the critical challenge of disentangling an anti-persistent, multifractal biological signal from persistent, monofractal environmental noise.
