In the framework of mapping underwater environments, Posidonia Oceanica (PO) is a fundamental indicator for environmental monitoring and related studies and analyses for climate change. The POSEIDON project aims to study, develop and test new methodologies to provide knowledge, as well as ultra-high-resolution (UHR) and beyond ultra-high-resolution (BUHR) mapping and monitoring data for the conservation of the PO and related meadows and banquettes in the Mediterranean area. The project is funded by the National Recovery and Resilience Plan and stems from the recommendation of international organizations (European Environmental Directives, Mediterranean PO Network, International Partnership for Blue Carbon, Intergovernmental Oceanographic Commission of UNESCO in the Save the Wave project). Starting from new integrated survey methods and legacy data, the PO, ichthyofauna and seagrass banquettes will be mapped to understand their evolution and extract indicators and useful trends. Furthermore, the produced information will be stored and shared in a webGIS and a dedicated app to prevent unexpected damages during sailing or other activities. POSEIDON will use imaging sensors (RGB and multi/hyperspectral cameras) to extract information. With specific lighting equipment for significant water depths, these sensors will be conducted (by scuba divers or Autonomous Underwater Vehicles) next to the object to acquire BUHR data. Finally, to monitor the fauna (fish, other species, invertebrates), terrestrial and underwater videogrammetry will be applied using mono and stereo-cameras with object detection algorithms.

In the context of increasing pressure on marine species, effective and automated methodologies for biodiversity monitoring are essential, particularly in sensitive ecosystems such as the Mediterranean Sea. This study presents an integrated deep learning approach for multi-object tracking and species recognition from underwater videos, aiming to automate and improve fish population censuses within Marine areas with a focus in Marine Protected Areas in Sardinia. Various detection models, including DeepFins, DeepEcomar, YOLOv8, and YOLOE, and tracking algorithms such as DeepSort, ByteTrack, and SAMURAI were evaluated. According to the achieved tests the YOLOE model, carefully trained on the Mediterranean-specific SardinIA dataset, demonstrated the best detection performance, while DeepSort proved most effective in maintaining individual identities across complex scenarios. The AI based achieved results compared with traditional visual census methods (underwater visual census, UVC and diver operated video census, DOVC), showing high accuracy in total abundance estimation and good agreement for dominant species. These findings suggest that deep learning techniques offer a promising, scalable solution for marine biodiversity monitoring, although challenges remain in species-level classification. In the present paper the following methodologies and the achieved results are reported.

In this study, we aimed to investigate the presence of an alteration in the feeding ecology of fish after the deployment of anti-trawling reefs at the Punta Tramontana site (Sardinia, Italy). To achieve this aim, we examined prey in the stomach contents of two target species of fish: Diplodus annularis and Mullus surmuletus. The samples were obtained from fishing activities carried out over one year from June 2017 to October 2018 at two impact sites, selected by the presence of artificial reefs, and two control sites. The results showed that installing these artificial devices increased the food spectrum availability of the target species and induced two different ecological feeding behaviors, probably derived from their different ecologies. D. annularis changed its alimentary strategy from generalist at control sites to specialist at impact sites, focusing its diet on the crustacean Gnathia maxillaris. Mullus surmuletus was not affected by the presence of the barriers, which constitute an additional site where the fish feed during their foraging activities. In conclusion, these anti-trawling reefs, in addition to the purpose for which they are designed (in this case, avoiding illegal trawling), did not produce any negative environmental impact on surrounding marine biota.