Abstract
Our current knowledge of the chemical composition of ocean dissolved organic matter (DOM) is limited, mainly because of its extreme molecular diversity, low concentration of individual compounds and the elevated ionic strength of ocean waters. As a result, many analytical methods require a previous extraction step. The efficiency and selectivity of the extraction method defines the representativeness of the extracted DOM fraction. Nowadays, the most widespread procedure for concentrating DOM is solid phase extraction (SPE) using styrene divinyl benzene polymer cartridges (PPL). Here, we investigate the effect of SPE-PPL on DOM elemental and optical properties to assess the efficiency and selectivity of this extraction method on water samples from the main intermediate and deep water masses of Arctic, Mediterranean and Antarctic origin present in the Cape Vert Frontal Zone (CVFZ, NW Africa). Furthermore, North and South Atlantic Central waters converge in this area and coastal DOM is injected by the giant upwelling filament of Cape Blanc. On one side, the colored fraction of DOM (CDOM) presented extraction efficiencies comparable to that of the bulk dissolved organic carbon (DOC), but decreased significantly with increasing wavelength, suggesting an affinity of PPL cartridges for low molecular weight organic compounds. While the protein-like fluorescent fraction of DOM (FDOM) was also extracted with the same efficiency than DOC, the extraction efficiency of the humic-like fraction was comparatively much higher. On the other side, dissolved organic nitrogen (DON) extraction efficiencies were about half that of DOC. These contrasting extraction efficiencies of the different DOM pools indicated that the extracts were enriched in N-poor, low molecular weight and recalcitrant DOM, therefore showing less variability than the corresponding bulk DOM. Furthermore, DOC, DON, CDOM and FDOM extracted were not homogeneous through the water column but displayed certain significant differences among water masses in both efficiency and selectivity.