Estimations of length-weight relationships and consumption rates of odontocetes in the Mediterranean Sea from stranding data

Stranding data provide fundamental information on biometric traits of cetaceans useful to increase knowledge on ecological traits and their consumption patterns. In this study, the length weight (L -W) relationships through the power regression model (W = a xLb) were calculated for three dolphin species (the striped dolphin, the common bottlenose dolphin and the Risso's dolphin) in several Mediterranean subregions and at the scale of the entire basin. Length (L) and weight (W) data were collected from stranding records during the period from 1983 to 2021 acquired from several databases and the literature. Starting from L -W relationships, a bootstrap method was applied to estimate the mean body weights, the daily ingested biomass (IB) and annual food consumption (AFC) rates of different dolphin species. In particular, four different equations were used to estimate the IB rates. Prey consumption by dolphin species was calculated through AFC rates and the available diet information (expressed in weight fractions) of dolphin species for different Mediterranean subregions. Considering the L -W relationships in the Mediterranean Sea, b coefficient values were equal to 2.578, 2.975 and 2.988 for the striped, the common bottlenose and the Risso's dolphin, respectively. At the Mediterranean scale, the AFC values estimated were 3913 kg (CI 2469-5306) for the Risso's dolphin, 2571 kg (1372-3963) for the common bottlenose dolphin and 1118 kg (531-1570) for the striped dolphin. Prey consumption pattern showed a clear partitioning among the investigated species, where the common bottlenose dolphin exploits neritic demersal and pelagic fishes (e.g. eel fishes, sparids), the striped dolphin exploits mesopelagic fishes and myctophids, and the Risso's dolphin was specialized on bathyal cephalopods of Histioteuthidae family. The results obtained in this study provide new information for the investigated species in several Mediterranean subregions providing a first consistent baseline to support the population dynamics modelling. At the same time, the wide uncertainty ranges of some parameters, as well as the lack of information for some species, stress the necessity of improving the data collection associated to stranding events, especially in the southern Mediterranean areas.