Aaron N. Rice

Acoustic communication is a central behaviour in vertebrates. Vocalizations across species reflect adaptations to different ecological and social contexts, but are also influenced by evolutionary history and functional constraints. As an iconic soniferous taxon, baleen whales produce vocalizations exhibiting a wide variety of properties. These signalling behaviours may correlate with non-acoustic behaviours, including anti-predator defence (e.g. ‘flight’ or ‘fight’ behaviours). We compiled acoustic features from 1500 sounds from the literature for all baleen whale species. The acoustic space, considered here as the two-dimensional space of frequency vs. time occupied by a species’ vocalizations, was quantified, revealing that humpback whales consistently occupy the largest acoustic space, while balaenopterids occupy smaller acoustic areas. Comparative methods, including Phylogenetic Generalized Least Squares were applied to investigate the relationship between acoustic traits and body length while accounting for evolutionary history. Body length significantly explained the limits of frequency range, but had no explanatory power for the other traits when phylogeny was considered. Residual variation in acoustic traits appeared largely independent of phylogeny, suggesting that factors beyond body length and shared ancestry contribute to the observed behavioural, ecological, and acoustic diversity. It highlights the complex interplay between morphology, evolution, and acoustic trait variation in baleen whales.

Summary: Many protected marine mammals use sound as a primary sensory modality for communication and foraging, so acoustic disruptions can be costly. To investigate the behavioural responses of a marine top predator, the common bottlenose dolphin ( Tursiops truncatus ), to underwater sound levels, we used a non-invasive method of localizing dolphin pods using their whistles detected on a fixed array. Additionally, for the first time, we tracked them over time, accounting for acoustic localization error using a state–space movement model. We investigated movement parameters (speed and path straightness) and the animals’ spatial distribution in relation to broadband (10–7080 Hz) and mid-frequency (2820–7080 Hz) ambient sound levels. Dolphins utilized different areas under elevated broadband and mid-frequency ambient sound levels, but the novel analysis of travel speed and directness of dolphin pod movement did not show significant variations. Bottlenose dolphins in the Mid-Atlantic Bight probably were not habituated to elevated ambient sound levels, as evidenced by their altered habitat use. Altered distributions in response to changes in the acoustic environment could impact the foraging success, health and vital rates of this protected species.

Coral reef monitoring needs scalable, non-invasive tools to complement resource-intensive traditional survey methods. Passive Acoustic Monitoring (PAM) offers a promising supplement, but its effectiveness is limited by the difficulty of attributing recorded sounds to species outside of previously well-characterized taxa. Using Omnidirectional Underwater Passive Acoustic Cameras (UPAC-360), we identified sounds from 31 reef fish species across 14 families on the Kona coast of Hawai’i Island, including 13 not previously documented as soniferous. By releasing video and audio specimens, we have created the largest open-access collection of in-situ reef fish sounds to date for the Pacific. A subset of acoustically distinctive taxa—such as Hawaiian Dascyllus ( Dascyllus albisella ), Lei Triggerfish ( Sufflamen bursa ), soldierfishes ( Myripristis spp. ), wrasses, and herbivorous grazers—were identifiable in PAM recordings through manual acoustic and spectrogram review. Through identifying particular sounds linked to species with different ecological roles, these sounds have the potential to serve as indicators of reef function to increase the information and value coming from PAM surveys of Hawaiian and Pacific coral reefs.

Substantial knowledge gaps exist regarding the migratory patterns of many whale species. Recent meta-analyses of passive acoustic sensor networks have provided critical information on the distribution of several baleen whale species across the western North Atlantic Ocean. However, of the baleen whales known to regularly inhabit this region, knowledge of common minke whales Balaenoptera acutorostrata remains scarce, particularly at lower latitudes. A large passive acoustic monitoring (PAM) data set from the Wider Caribbean Region, comprising 92 PAM recorder deployments and over 11900 recording days between 2001 and 2025, was analyzed using a convolutional neural network-based minke whale pulse train detector. Results provide novel insights into the spatial and temporal distribution of the common minke whale in the tropics and subtropics of the western North Atlantic Ocean. Our findings reveal pronounced spatial and seasonal occurrence patterns, with detections concentrated between 14° and 18°N during winter months (December-March). This includes the first acoustic evidence of common minke whales in the US Virgin Islands, Dominican Republic, Montserrat, and Guadeloupe. Summer detections at some sites suggest the presence of non-migratory individuals, challenging previous assumptions that their presence at lower northern latitudes is limited to winter months. Despite considerable recording effort in the Gulf of Mexico, only one pulse train was detected, providing limited evidence of the species in this region. Results from this broad-scale analysis refine the known winter range and potential breeding grounds of North Atlantic common minke whales and underscore the value of PAM in data-poor and logistically challenging environments.

Motivation: Fish sounds are integral to a variety of ecological functions, including reproduction, predator–prey interactions and recruitment, with ever‐growing interest in their relationships to anthropogenic impacts and applications for passive acoustic monitoring. Until recently, however, fish sound production data were often not easily accessible, limiting research, management and public awareness. FishSounds.net launched in 2021 to compile and disseminate global fish sound production information and recordings. Here, we describe the subsequent FishSounds version releases 2.0, 2.1, 2.2, 3.0 and 3.1 (cumulatively referred to as Versions 2 and 3). We updated the core dataset to include any fish species studied for sound production up until the year 2023. We added over 1000 new fish sound recordings, collated from FishSounds contributors or Cornell University's Macaulay Library. Connections with FishBase and the World Register of Marine Species were strengthened to improve the species information provided on FishSounds and facilitate data sharing. We also created several interactive visualisation tools, including a dendrogram and map view, to allow users to explore trends in known fish sound production. These updates have made FishSounds now the largest catalogue of fish sound production knowledge, utilised by over 17,000 users annually and featuring 1252 fish species studied across 1013 references as well as 1304 recordings. Main Types of Variables Contained: Fish sound production information compiled from the scientific literature, representative fish sound recordings with associated metadata and supporting images and species data drawn from other repositories. Spatial Location and Grain: Global. Time Period and Grain: 1874–2023. Major Taxa and Level of Measurement: Fishes (Agnatha, Chondrichthyes, Sarcopterygii, Actinopterygii). All sound production information and most recordings have species‐level identification, with any others identified to the lowest possible taxonomic level. Software Format: The complete database is presented on FishSounds.net , with versioned image, audio, tabular and text files in a Borealis data repository.

Coral reef soundscapes hold an untapped wealth of biodiversity information. While they are easy to record and filled with snapping shrimp and fish sounds, they are difficult to decipher because we know little about which sounds are made by which species. With identified fish sounds, acoustic monitoring can directly inform biodiversity metrics, detect indicator or invasive species, identify behavioural events and estimate abundance at temporal and spatial scales that are impossible with methods like eDNA or visual surveys. The missing link, knowing which sounds come from which species, is exceedingly difficult to establish with fish, especially on a species‐rich coral reef. Using a novel method to visualize in situ underwater sound, we have developed a technique that combines visualizations of spatial audio with concentric 360° video recordings, a combination not previously accomplished underwater. We have identified and assigned the most extensive collection of natural sounds of fishes. Further, we demonstrate that these species identifications can be used to decipher a passive acoustic monitor recording. We have collected our identified recordings into a growing open‐access resource to catalyse passive acoustic monitoring research, enabling a species‐specific resolution of coral reef soundscape dynamics and providing critical validated information for developing machine‐learning models to analyse an ever‐expanding collection of long‐term recordings.

Effective recovery plans for endangered species rely on insights into species’ ecology to identify risks and develop population recovery strategies. Data gaps pose challenges for many species of conservation concern, particularly those with cryptic behaviors or that occupy difficult-to-access habitats. Sounds produced by these species offer an effective means of observing many such marine and aquatic species, and for this reason, passive acoustic monitoring has emerged as an important study and assessment approach in marine systems. This approach is only just beginning to be applied for aquatic species monitoring in freshwater habitats. Atlantic sturgeon Acipenser oxyrinchus, a species of conservation concern along the US East Coast, remains poorly understood due to persistent data gaps despite years of conservation efforts. While sounds have been described for other sturgeons, sounds from Atlantic sturgeon have not yet been reported. Here, we characterized acoustic cues associated with Atlantic sturgeon in the Hudson River, New York, USA, and identified a low-frequency (44 Hz peak frequency) signal strongly correlated with the occurrence of telemetry-tagged adults which enter the river to spawn. We corroborated these efforts with recordings of captive Atlantic sturgeon, in which we detected the same sound type during a spawning period. Our findings provide an opportunity to develop passive acoustic monitoring strategies for Atlantic sturgeon, offering a non-invasive tool for understanding the spatiotemporal distribution of spawning activity across their range. We demonstrate potential applications of passive acoustic monitoring to inform sturgeon conservation and management, including characterizing habitat use, identifying cross-species interactions, and providing abundance indices.

Machine learning has the potential to revolutionize passive acoustic monitoring (PAM) for ecological assessments. However, high annotation and computing costs limit the field’s adoption. Generalizable pretrained networks can overcome these costs, but high-quality pretraining requires vast annotated libraries, limiting their current development to data-rich bird taxa. Here, we identify the optimum pretraining strategy for data-deficient domains, using tropical reefs as a representative case study. We assembled ReefSet, an annotated library of 57 000 reef sounds taken across 16 datasets, though still modest in scale compared to annotated bird libraries. We performed multiple pretraining experiments and found that pretraining on a library of bird audio 50 times the size of ReefSet provides notably superior generalizability on held-out reef datasets, with a mean area under the receiver operating characteristic curve (AUC-ROC) of 0.881 (±0.11), compared to pretraining on ReefSet itself or unrelated audio, with a mean AUC-ROC of 0.724 (±0.05) and 0.834 (±0.05), respectively. However, our key findings show that cross-domain mixing, where bird, reef and unrelated audio are combined during pretraining, provides superior transfer learning performance, with an AUC-ROC of 0.933 (±0.02). SurfPerch, our optimum pretrained network, provides a strong foundation for automated analysis of tropical reef and related PAM data with minimal annotation and computing costs. This article is part of the theme issue ‘Acoustic monitoring for tropical ecology and conservation’.

Marine passive acoustic monitoring (PAM) has produced petabytes of data that are used by researchers, resource managers, industry, and regulators to understand how marine animals use sound and the impacts of anthropogenic noise on species and ecosystems throughout the global ocean. These big data provide unprecedented opportunities to study underwater soundscapes and marine ecology but also enormous challenges to efficiently extract information. To address these challenges, a U.S. federally funded and led Sound Cooperative ( SoundCoop ) project built community-focused cyberinfrastructure to promote improved, scalable and sustainable processing and access of marine PAM data for management, science, industry and military applications. Driven by cross-institutional participation representing a diversity of data collection methods and conditions, the SoundCoop project established guidance for standardized processing of ocean sound level metrics using freeware software toolkits and developed core tools and processes that support open science. Four examples of comparative analyses that connect disparate PAM monitoring efforts, and integrate non-acoustic data illustrate how comparable, interoperable sound level metrics support a more coherent and synoptic perspective on global ocean soundscapes using methods that current and future PAM projects can leverage. Such a framework around PAM big data offers the opportunity to revolutionize large-scale marine ecology and oceanography in similar ways to other transformative approaches for understanding environmental or ecological patterns and processes at global scales.