COSTtERA-3

Following the implementation of the COSTtERA and COSTtERA-2 projects, SUBMON is advancing COSTtERA-3 as a new phase of consolidation and progress aimed at deepening knowledge of the bottlenose dolphin and its interaction with trawl fisheries in northern Catalonia.

COSTtERA-3 continues the study of the spatial distribution and behaviour of bottlenose dolphins, as well as their level of dependence on fishing activity and how this evolves over time. However, the project not only maintains the monitoring lines developed in previous phases, but also incorporates new approaches aimed at gaining a better understanding of the species.

Key innovations include the study of body condition using non-invasive techniques, as well as the development of predictive models capable of estimating population responses under different fishing effort and environmental variability scenarios. This approach makes it possible to anticipate potential impacts arising from changes in resource availability and to support decision-making processes in management.

The project integrates advanced technological tools, including drones, underwater cameras (including 360º systems), hydrophones and artificial intelligence, enabling a more comprehensive and precise understanding of the interaction between dolphins and fishing activity.

In addition, COSTtERA-3 strengthens the active involvement of the fishing sector, incorporating its knowledge of the marine environment and promoting a collaborative approach. In doing so, the project lays the foundations for adaptive management based on scientific evidence, aimed at reconciling the conservation of bottlenose dolphins with the sustainability of fishing activities.

Implementation years: 2026-2027.

The study area of the project is the continental shelf located around Cap de Creus, with the geographical boundary of the French border and covering an area of 2,144 km².

In this area, two important protection figures included within the Natura 2000 Network converge: the Site of Community Importance (SCI) “Submarine canyon systems of Cap de Creus” and the Special Protection Area for Birds (SPA) “Empordà Marine Area”.

Boat-based surveys

Year-round distribution

Monitoring of bottlenose dolphin presence and distribution is carried out across different times of the year in order to identify preferential use areas, seasonal changes and their relationship with fishing activity.

Photo-identification (Photo-ID)

Photographic records of observed individuals are collected to enable individual identification and long-term monitoring. This tool allows the estimation of site fidelity, residency patterns and group composition.

Surface behaviour

Standardised data on surface behaviour (feeding, travelling, socialising, vigilance, avoidance, etc.) are collected to identify recurring patterns and interaction contexts.

LiDAR-equipped drones: behaviour and non-invasive biometrics

During surveys, drones equipped with LiDAR technology are used to document surface behaviour and obtain biometric measurements. This makes it possible to assess body condition (size, condition, proportions), explore health indicators and, where possible, detect signals compatible with pregnancy.

Onboard monitoring on trawling vessels

Installation of underwater cameras

Cameras are installed on trawling nets following the protocol developed in the COSTtERA and COSTtERA-2 projects, in order to analyse dolphin behaviour during interactions with fishing activity.

As a technological improvement, 360º cameras are incorporated, significantly expanding the field of view compared to previous studies. This broader coverage reduces blind spots and allows for a better interpretation of animal position, trajectories and manoeuvres around the net, increasing the accuracy of risk assessment.

Acoustics: hydrophone deployment in nets

Hydrophones are installed on the nets to detect the presence of individuals interacting with them, thereby expanding monitoring capacity under low-visibility or operational constraints.

This technique enables continued study of dolphins’ acoustic patterns during interactions, provides information on the level of interaction and allows the quantification of interaction events.

Integration and modelling of ecological, fisheries and environmental variables

Models are developed to estimate the population response of bottlenose dolphins under different scenarios.

Artificial intelligence

The project optimises artificial intelligence algorithms for the automatic detection of bottlenose dolphins in acoustic recordings and underwater video footage obtained during interactions with trawling fisheries, building on advances achieved in COSTtERA-2.

In addition, computer vision models are improved through new convolutional neural network (CNN) architectures. The combined analysis of acoustic data and underwater imagery is also integrated, synchronising both sources to detect behavioural patterns linked to specific acoustic signals.

European Cetacean Society 2026 Annual Conference

The SUBMON team presents part of the work carried out within COSTtERA at the European Cetacean Society (ECS) Annual Conference, held in Dundee, Scotland. You can explore the posters presented here:

• “A Puzzling Calf: First record of an association between a presumed Grampus griseus calf and an adult Tursiops truncatus in the Mediterranean Sea”
• “Automatic image-based detection of bottlenose dolphins in trawl net monitoring cameras”

Explore the COSTtERA project and its different phases