Chemical fate and settling of mineral dust in surface seawater after atmospheric deposition observed from dust seeding experiments in large mesocosms

Abstract. We report here the elemental composition of sinking particles in sediment traps and in the water column following four artificial dust seeding experiments (each representing a flux of 10 g m⁻²). Dry or wet dust deposition were simulated during two large mesocosms field campaigns that took place in the coastal water of Corsica (NW Mediterranean Sea) representative of oligotrophic conditions. The dust additions were carried out with fresh or artificially aged dust (i.e., enriched in nitrate and sulfate by mimicking cloud processing) for various biogeochemical conditions, enabling us to test the effect of these parameters on the chemical composition and settling of dust after deposition. The rates and mechanisms of total mass, particulate organic carbon (POC) and chemical elements (Al, Ba, Ca, Co, Cu, Fe, K, Li, Mg, Mn, Mo, N, Nd, P, S, Sr and Ti) transfer from the mesocosm surface to the sediment traps installed at the base of the mesocosms after dust deposition show that (1) 15% of the initial dust mass was dissolved in the water column in the first 24 h after seeding. Except for Ca, S and N, the elemental composition of dust particles was constant during their settling, showing the relevance of using interelemental ratios, such as Ti/Al as proxy of lithogenic fluxes. (2) Whatever the type of seeding (using fresh dust to simulate dry deposition or artificially aged dust to simulate wet deposition), the particulate phase both in the water column and in the sediment traps was dominated by dust particles. (3) Due to the high Ba content in dust, Ba/Al cannot be used as productivity proxy in the case of high dust input in the sediment traps. Instead, our data suggests that the ratio Co/Al could be a good productivity proxy in this case. (4) After 7 days, between 30 and 68% of added dust was still in suspension in the mesocosms. This difference in the dust settling was directly associated with a difference in POC export, since POC fluxes were highly correlated to dust lithogenic fluxes signifying a ballast effect of dust. The highest fraction of remaining dust in suspension in the mesocosm at the end of the experiment was found inversely correlated to Chl a increase. This suggests that the fertilizing effect of dust on autotrophs organisms, the ballast effect, and POC fluxes are strongly correlated. (5) Our data emphasize a typical mass ratio Lithogenic/POC fluxes around 30 which could be used as reference to estimate the POC export triggered by wet dust deposition event.