COSTtERA-2

Following the successful results of the first project, SUBMON has continued its work with the “COSTtERA-2 Project: Fishing Interaction and Bottlenose Dolphin Behaviour in Trawling Nets of Catalonia“. This new phase of the study aims to deepen the analysis of interactions between bottlenose dolphins and trawling nets in the Catalan region. The project is co-financed by the European Union through the FEMPA (European Maritime, Fisheries and Aquaculture Fund), which allows for expanding research and conservation efforts to ensure the protection of this vulnerable species and the sustainability of fishing activities in the area.

The project includes several study blocks, including surveys of artisanal and trawling fishermen, maritime monitoring campaigns (distribution, behaviour, and photo identification studies of bottlenose dolphins), the installation of cameras and hydrophones on fishing nets, and the development of an Artificial Intelligence (AI) algorithm for analysing images and sound, aiming to enhance the protocol established during the COSTtERA project in 2022.

This approach aims to deepen the understanding of the interaction between bottlenose dolphins and fishing activities in Catalonia by combining standardised methodology (surveys and maritime campaigns) with innovative approaches using underwater cameras, hydrophones, and artificial intelligence algorithms.

Implementation years: 2024-2025.

The project will employ various methodologies to obtain all the necessary information:

Surveys

Surveys of 30% of artisanal and trawling fishermen in Catalonia are carried out, to analyse the degree and type of interaction that occurs in the territory and its impacts on both dolphins and fishermen (estimate of accidental catches and estimate of the damage that these interactions can cause to fishermen). SUBMON has conducted these types of surveys in previous years, and the results serve to analyse the trend of these interactions in the area.

Maritime surveys

The project carries out maritime surveys throughout the year to continue studying the bottlenose dolphin population in northern Catalonia, conducted by SUBMON since 2017, with a particular focus on photo identification, density, and dolphin behaviour.

The data obtained allows for analysing animal density in the area, assessing the degree of dependency on fishing, and comparing the percentage of fishing association and distribution in the area between seasons and years.

Photographs of each sighting are taken and included in the organisation’s photo identification catalogue, which includes more than 600 different bottlenose dolphins.

During the surveys, surface behaviour is recorded when dolphins interact with fishing vessels. The data obtained are analysed to establish repeating patterns that can be related to underwater images and acoustic recordings, thus associating surface behaviours with underwater behaviours and sounds.

Underwater cameras

Underwater cameras are installed on trawling nets to study the underwater behaviour of animals when interacting with fishing nets.

Video cameras are implemented on trawling nets  to study the feeding behaviour of bottlenose dolphins during trawling. The cameras are placed on different vessels, specifically those that have had the most dolphin observations in recent years and that operate at less than 200 metres, which is the range of presence of the target species.

The installation of cameras follows the protocol defined during the COSTtERA project. They are positioned along the net, with special attention to the net’s entrance, to determine, among other things, whether the animals enter the trawling nets in this area and are, therefore, at greater risk of accidental capture.

Hydrophones

Hydrophones are installed on fishing nets  to record the sounds of the animals and detect when dolphins are in the nets. This technique allows for studying the acoustic feeding patterns of dolphins, obtaining information on the degree of interaction, and quantifying the interactions that occur.

Artificial Intelligence

An Artificial Intelligence algorithm is developed to automatically detect the bottlenose dolphin individuals recorded by the hydrophone in collaboration with Keacoustics and the Polytechnic University of Catalonia (UPC).

Data exploration and preprocessing are carried out to evaluate the peculiarities of the recordings. Noise statistics are extracted, and the types of events and their spatio-temporal statistical features, as well as their occurrence, are reviewed.

Furthermore, machine learning models are developed for the automated detection and classification of sounds of interest (clicks, whistles, impulsive sounds), and acoustic patterns of interest that may be associated with feeding events are identified.

Installation of underwater cameras on fishing nets during the COSTtERA project.

The participation of the fishing sector in this project is essential for understanding the interactions with dolphins. Therefore, their support is ensured during the different phases of the project, and dissemination sessions will be organised with the fishing sector to view the recorded images as well as to review the results obtained.

Explore the COSTtERA project and its different phases