Comparaison des inventaires de poisson dans les aires marine protégées de Méditerranée: influence de la surface et l’ ancienneté

Influence of surface area and age Fish are the main fauna compartment concerned with protection measures and management into Marine Protected Areas (MPA). Surveys were conducted in 8 MPAs by various sampling methods (visual surveys by SCUBA diving, anaesthetic catch, poisoning, fishing nets or fishing lines): Port-Cros (France), Lavezzi and Scandola (Corse, France), Monaco, Medes Islands (Spain), Tremiti, Torre Guaceto and Porto Cesareo (Italy). These areas presented similar biotopes (rocky substrata, Posidonia oceanica beds and sandy bottoms) but different ages (1963, Port-Cros to 1997, Porto Cesareo) and sizes (418 ha at Medes Islands to 16600 at Porto Cesareo). Tremiti and Porto Cesareo have a weak protection compared to the other MPAs considered. The first step of this work was to update surveys already published by including new observations and unpublished data. Surveys were then compared using two diversity indexes, Average Taxonomic Distinctness (AvTD) and Variation in Taxonomic Distinctness (VarTD). For each MPA, these indexes have been compared to theoretical values for the Mediterranean Sea. This theoretical value (331 species belonging to 94 families) was calculated from the total number of species reported in the Mediterranean (FishBase) without the lessepsian species, species inhabiting deep waters (more than 100 m depth), and sub-species. A total of 215 (Port-Cros), 198 (Scandola), 191 Lavezzi, 136 (Medes), 108 (Monaco), 105 (Torre Guaceto), 75 (Tremiti and Porto Cesareo) species have been reported in the different MPAs. Number of species found in Port-Cros, Scandola and Lavezzi corresponded to the diversity expected for Mediterranean fish samples: i.e. the diversity expected from a sample of n species (215, 198 or 191 species respectively) drawn at random from the total list (331 species). VarTD of all the well protected zones corresponded also to the Mediterranean expected values. Therefore, well protected areas allowed preservation of taxonomic relationships between species. Spearman non parametric correlations showed that species richness and AvTD were correlated to the age of MPA and not with its size. In conclusion, the implementation of new MPAs is discussed regarding these relationships: the size of the protected area is not as important as an efficient surveillance and a long term protection to ensure a high fish species diversity.