A multi-model framework to inform the conservation of Chondrichthyes and explore the ecological impacts of their restoration in the Mediterranean Sea

Elasmobranchs (i.e., sharks, rays, and skates) and chimaeras, belong to a group of species called Chondrichthyes, because of their mostly cartilaginous skeleton. In terms of life-history traits, these species are marked by slow growth, late sexual maturity and low fecundity, which make them extremely vulnerable to human pressures, particularly to fishing. Consequently, a reduction in chondrichthyans abundance has been globally observed during the last decades, with many species now being threatened with extinction, especially in the Mediterranean. The effects of such pervasive chondrichthyan loss on marine ecosystem functioning and ecosystem services have not been deeply investigated: few studies highlighted the cascading effects of the loss of top-predator sharks throughout food webs, and ecosystem responses to the loss of meso-predator and demersal species are mostly unexplored. Ecological modelling appears as a strategic tool to systematically screen management and conservation actions and identify the most effective and sustainable strategies. This thesis aims at investigating the status of chondrichthyans populations, by developing single-species tools to inform management and conservation, and at exploring the effects of either the decline and the recovery of such species, while highlighting the methodological limits of species aggregation, by building trophic network models. A surplus production model, was used to investigate the status of the most landed cartilaginous fish in the area of the northern Adriatic Sea. The use of a more sophisticated single species modelling approach, namely Individual-Based Modelling (IBM), was proposed to compare the effects of implementing different management measures. The IBM described the population of smooth-hound sharks (Mustelus spp.) of the n. Adriatic Sea, by integrating bioenergetic theory, density-dependent functions, and intra-specific variability to simulate the main processes of the life cycle of each individual. The model investigated sustainable management scenarios that both favour population growth and sustain the local community which relies on fishing activities. Despite the complexity and the high demand of data of these models, they are useful screening tools to systematically identify and select, within a pool of options, promising management actions which aim toward Chondrichthyes conservation and sustainability of fisheries. To disentangle the role of the functional diversity of chondrichthyans at the ecosystem level, which is often neglected in food web models used for fishery management and conservation, the thesis focused on published models built for Mediterranean sub-basins using the Ecopath with Ecosim approach. All the species with enough available data from the broader sharks and rays functional groups defined by the model authors were extracted and then new models were generated by aggregating species at different levels. Finally, ecological network analysis descriptors and modelled management scenarios (e.g., overfishing) were compared to investigate how the degree of aggregation affects the food web properties and dynamics. The results highlighted the several trophic roles that different chondrichthyans can play, and the different levels at which each species impacts the rest of the trophic network. This result calls for more awareness of the model limitations deriving from food web aggregation driven by data scarcity. Overall, the study highlights advantages and disadvantages of the different modelling approaches that were presented, and of the data that were used. The thesis elucidates the benefits of employing diverse modelling methodologies, demonstrating how their collective application offers a comprehensive perspective on the problem at hand, ultimately yielding scientifically grounded data crucial for informing effective species conservation and fishery management strategies.