Source and composition of organic matter in the Bari canyon (Italy): Dense water cascading versus particulate export from the upper ocean

Abstract

Recent studies in the southern Adriatic Sea have shown that dense water cascading is an effective off-shelf transport mechanism. In order to evaluate the relative importance of different processes affecting the down-slope transfer of organic matter (OM), a multi-proxy biogeochemical study was carried out in the southern Adriatic. Three sediment traps were deployed on the southern slope in March 2004 for 1 year at 35 m above the seabed. Surficial sediments were collected on the shelf and slope along a shore-normal transect. Suspended material in the water column at different water depths was collected along nine shore-normal transects. Organic carbon (OC), total nitrogen (TN), carbon isotopic composition (δ 13C and Δ 14C) and CuO reaction products were used to identify the OM sources and to asses the relative importance of vertical and lateral particulate fluxes. Both sinking particulates from the euphotic zone and advected material from the seafloor were collected by the moored instrumentations. During low-mass fluxes, the trapped material exhibited the highest Δ 14C values (from −162.0‰ to +42.3‰) and OC contents (from 1.13% to 2.17%) while the lignin displayed the lowest content (from 0.160 to 0.285 mg 100 mg −1 OC). However, during dense water cascading events, the down-slope particulate flux showed different compositional features. In this period the trapped material displayed the most depleted Δ 14C values (from −275.3‰ to −171.3‰), the lowest OC content (from 0.87% to 1.31%), and the highest lignin content (from 0.230 to 379 mg 100 mg −1 OC). Holocene sea-level rise has drastically reduced sediment availability to the Bari canyon, and this in turn has affected the OC composition of the material funneled toward the deep Adriatic Sea. Parameters based on lignin CuO products suggest that the terrestrial material exported down the slope is very different from riverine inputs and inner shelf sediments. This indicates that neither river floods nor sediment resuspension from the inner shelf contributes directly to the down-slope flux. Rather, the material collected in the sediment traps exhibits an OM composition comparable to the surface sediments collected in the southern outer shelf. This correspondence suggests that resuspension of sediments by bottom currents in the southern outer shelf significantly contribute to down-slope OM transfer. Finally, the presence of lignin phenols in all cups indicates that lateral fluxes were constantly active on the slope throughout the deployment.