Trophodynamics of a deep-sea demersal fish assemblage from the bathyal eastern Ionian Sea (Mediterranean Sea)

Daily food consumption of the eight dominant demersal fish species of the bathyal eastern Ionian Sea were determined from field data on four seasonal cruises (April 1999, July August 1999, November 1999 and February 2000). Daily ration (DR) estimates ranged from 0.198 to 4.273% WW/WW. Overall, DR estimates were independent of the model used, and they were comparable to the daily consumption of other deep-sea fauna (e.g. fish and crustaceans). Both sharks studied (Galeus melastomus and Etmopterus spinax) exhibited the highest values of DRs, together with the macrourid Coelorhynchus coelorhynchus in August. Among osteichthyes, DR estimates were related (in a multi-linear regression model) to the nature of their diet (i.e. their trophic level deduced from δ15N isotopic composition, the mean number of prey and trophic diversity). Thus, species feeding at a lower trophic level, ingesting a large number of prey items and with a very diversified diet had higher DR than species from higher trophic level and feeding fewer prey items. By season, the DR of species feeding mainly on mesopelagic prey (Hoplostethus mediterraneus and Helicolenus dactylopterus) were higher in summer, while DR for benthos/suprabenthos feeders (i.e. C. coelorhynchus and Nezumia sclerorhynchus) were higher in spring. Higher food consumption coincides with maximum food availability, both among mesopelagic feeders (higher availability of euphausiids, Pasiphaea sivado and Sergestes arcticus in summer) and among Macrouridae (higher suprabenthos densities in spring). In a tentative estimate the energy intake deduced from diet (i.e. mean energy value of food ingested) was constant in all seasons for each species studied. Results for the energy intake also indicate higher energy intake in the diet of mesopelagic feeders than in the diet of benthic feeders. Overall results are discussed in relation to the deep-sea ecosystem structure and functioning.