Impacts of dust deposition on dissolved trace metal concentrations (Mn, Al and Fe) during a mesocosm experiment

Abstract

Abstract. The deposition of atmospheric dust is the primary process supplying trace elements abundant in crustal rocks (e.g. Al, Mn and Fe) to the surface ocean. Upon deposition, the residence time in surface waters for each of these elements differs according to their chemical speciation and biological utilization. Presently however their behavior after atmospheric deposition is poorly constrained, principally because of the difficulty in following natural dust events in-situ. In the present work we examined the temporal changes in the biogeochemistry of crustal metals (in particular Al, Mn and Fe) after an artificial dust deposition event. The experiment was contained inside trace metal clean mesocosms (0–12.5 m depths) deployed in the surface waters of the Northwestern Mediterranean, close to the coast of Corsica in the frame of the DUNE project (a DUst experiment in a low Nutrient low chlorophyll Ecosystem). Two consecutive artificial dust deposition events, each mimicking a wet deposition of 10 g m−2 of dust, were performed during the course of this DUNE-2 experiment. The changes in dissolved manganese (dMn), iron (dFe) and aluminium (dAl) concentrations were followed immediately and over the following week and their inventories and loss or dissolution rates were determined. The evolution of the inventories after the two consecutive additions of dust showed distinct behaviors for dMn, dAl and dFe. Even though the mixing conditions differed from one seeding to the other, dMn and dAl showed clear increases directly after both seedings due to dissolution processes. Three days after the dust additions, dAl concentrations decreased as a consequence of scavenging on sinking particles. dAl appeared to be highly affected by the concentrations of biogenic particles, with an order of magnitude difference in its loss rates related to the increase of biomass after the addition of dust. For dFe concentrations, the first dust addition decreased the concentrations through scavenging of the dust particles, whereas the second seeding induced dissolution of Fe from the dust particles. This difference, which might be related to a change in Fe-binding ligand concentration in the mesocosms, highlights the complex processes that control the solubility of Fe. Based on the inventories at the mesocosm scale, the estimations of solubility of metals from dust particles in seawater were 1% for Al and 40% for Mn which were in good agreement with laboratory based estimates. Overall, the trace metal dataset presented here makes a significant contribution to enhancing our knowledge on the processes influencing trace metals release from Saharan dust and the subsequent processes of bio-uptake and scavenging in a low nutrient low chlorophyll area.