Aquatic ecosystems are constantly changing due to anthropic stressors, which can lead to biodiversity loss. Ocean sound is considered an essential ocean variable, with the potential to improve our understanding of its impact on marine life. Fish produce a variety of sounds and their choruses often dominate underwater soundscapes. These sounds have been used to assess communication, behaviour, spawning location, and biodiversity. Artificial intelligence can provide a robust solution to detect and classify fish sounds. However, the main challenge in applying artificial intelligence to recognize fish sounds is the lack of validated sound data for individual species. This review provides an overview of recent publications on the use of machine learning, including deep learning, for fish sound detection, classification, and identification. Key challenges and limitations are discussed, and some points to guide future studies are also provided.

Soundscape ecology has gained prominence in the monitoring of marine ecosystems due to its non-invasive characteristics and spatiotemporal efficiency. However, the development of ecoacoustic indices is a recent field that needs to address many challenges to fulfill its great potential, especially in the context of marine ecology. Here, we reviewed the most recent studies that used ecoacoustic indices in marine ecosystems. The literature search was conducted in the Scopus (Elsevier) database and used the chain referral sampling in the list of references of each publication. In total, we identified 27 publications that used ecoacoustic indices in marine environments such as coral reefs, rocky shores, coastal regions, and offshore regions. A total of four major limitations were identified and addressed, including: the challenge to find adequate acoustic bioindicators; the lack of a universal index or standardized protocol; the issue that most acoustic indices applied to marine environments have been developed to be used in terrestrial environments; and the lack of studies that have tested ecoacoustic indices under different environmental conditions. Once these challenges are addressed, the analysis of marine sound based on the interpretation of ecoacoustic indices has a great potential to become one of the most cost-effective tools for monitoring environments.

This work introduces a description of the complex upwelling regime off the Cabo Frio region in Brazil and shows that ocean modeling, based on the feature-oriented regional modeling system (FORMS) technique, can produce reliable predictions of sound speed fields for the corresponding shallow water environment. This work also shows, through the development of simulations, that the upwelling regime can be responsible for the creation of shadow coastal zones, in which the detection probability is too low for an acoustic source to be detected. The development of the FORMS technique and its validation with real data, for the particular region of coastal upwelling off Cabo Frio, reveals the possibility of a sustainable and reliable forecast system for the corresponding (variable in space and time) underwater acoustic environment.

Sun coral (Tubastraea spp.) is an invasive species that poses a considerable threat to coastal ecosystems. Therefore, early detection is essential for effective monitoring and mitigation of its negative impacts on marine biodiversity. This study presents a novel computer vision approach for automated early detection of invasive Tubastraea species in underwater images. We used the YOLOv8 object detection model, which was trained and validated on a manually annotated dataset augmented with synthetic images. The data augmentation addressed the challenge of limited training data that is prevalent in underwater environments. The model achieved performance metrics (in terms of precision accuracy, recall, mAP50, and F1 score) of over 90% and detected both open and closed coral stage classes. Test phase results were compared with expert validation, demonstrating the model’s effectiveness in rapid detection (16 ms) and its limitations in areas highly covered by Tubastraea. This study demonstrates the potential of deep learning with data augmentation to facilitate the rapid assessment of large image datasets in monitoring sun coral bioinvasion. This approach has the potential to assist managers, taxonomists, and other professionals in the control of invasive alien species.

The study aims to understand the dispersal patterns of non-indigenous Tubastraea spp. (Sun Coral) larvae in the Brazil Current (BC), specifically in the Cape Frio recurrent cyclonic eddy (CFE) scenario. For this, the Regional Ocean Model System was used to simulate the hydrodynamic fields in a high-resolution nested grid, where a model of lagrangian floats, in a good approximation of the larvae properties and considering massive planulation events, was coupled with surface larval release from the Campos Basin area. The simulation was representative of mesoscale features compared to similar studies, ARGO vertical profiles and a py-eddy-track algorithm was used to obtain eddy variables, such as radius, rotational and translational velocities. These parameters are fundamental to access when an eddy tends to trap or not the water, heat and plankton in its interior. CFE turned out to be highly nonlinear, with a strong tendency to trap larvae in its core, acting as a dispersal constrictor when compared with the organisms in the axis of the higher speed of BC. A strong negative correlation (-0.75) was found between the days that larvae were inside the eddy and their distance from the origin. None of the 48,000 larvae released during simulated experiment a 16-day spawning event reached the coast. There are two different patterns for the dispersal, one along the shelf break and another, with higher larval density, off from the 1000 m isobath. The CFE’s presence allows larvae to remain in the same region for longer periods, although in offshore areas. Therefore, as there is considerable availability of fixed substrates on oil rig structures, larvae could settle on them resulting in a possible inter-platforms connectivity between populations of Tubastraea spp. Also, regions in the CFE that present downward vertical velocities (downwelling), may move young larvae to depths of about 60 m suggesting that subsurface colonizations are possible due to specific dynamics of propagating cyclonic eddies. So, identifying the main factors that affect the dispersion of propagules is essential to subsidize management policies for controlling bioinvasion associated with exploitation of hydrocarbon resources in offshore areas.

Maritime shipping has been recognized as the main pathway of species transference throughout the world. Ballast water or fouling on vessels have become the main vectors of aquatic species transportation between coastal marine ecosystems, which can lead to significant social, economic, and ecological negative impacts. Approximately 55% of non-indigenous marine species can be associated with vessel biofouling. However, acting to prevent, detect, manage, and control bioinvasion via biofouling is relatively recent. To improve non-indigenous marine species management, several risk assessment frameworks have been proposed. This research undertook a systematic bibliographic review to investigate whether and how risk assessment has been used to manage marine bioinvasion via biofouling. We analyzed 410 scientific studies published until December 31, 2023, via the Scopus database, filtering them by keywords, title, and This study describes the parameters and methodologies to assess bioinvasion risks. Furthermore, it discusses the incorporation of risk assessments into global, regional, and country-level public policies, the latter based on countries that publish the most on the topic. This research highlights the scarcity of mandatory measures to manage biofouling, as well as the urgent need to standardize risk assessment methodologies and to incorporate the effects of climate change into these assessments. The development of advanced data analysis tools and a unified strategy to effectively address the complex challenge of marine bioinvasions is also suggested.