Mineral aerosol deposition is the dominant source of iron to the open ocean. Soil iron is typically insoluble and understanding the atmospheric processes that convert insoluble iron to the more soluble forms observed over the oceans is crucial. In this paper, we model several proposed processes for the conversion of Fe(III) to Fe(II), and compare with cruise observations. The comparisons show that the model results in similar averaged magnitudes of iron solubility as measured during 8 cruises in 2001–2003. Comparisons show that results of cases including cloud, SO2 and hematite processing are better than the other approaches used using the reaction rates we assume in this paper; unfortunately the reaction rates are not well known, and this hampers our ability to conclusive show one process is more likely than another. The total soluble iron deposited to the global ocean is estimated by the model to range from 0.36 to 1.6 Tg y−1, with 0.88 Tg y−1 being the mean estimate; however there are large uncertainties in these estimates. Comparison shows that the regions with largest differences between the model simulations and observations of iron solubility are in the Southern Atlantic near South America coast and North Atlantic near Spain coast. More observations in these areas or in the South Pacific will help us identify the most important processes. Additionally, laboratory experiments that constrain the reaction rates of different compounds that will result in a net solubilization of iron in aerosols are required to better constrain iron processing in the atmosphere. Additionally, knowing what forms of iron are most bioavailable will assist atmospheric scientists in providing better budgets of iron deposited to the ocean surfaces.

[1] Atmospheric deposition of iron (Fe) is a major source of the micronutrient to the remote ocean. Most studies have focused on the total atmospheric Fe fluxes to the oceans while fewer studies have focused on the chemistry and chemical speciation of atmospheric Fe. This speciation of Fe in the atmosphere is critical to understanding the fraction of Fe that will be labile in surface waters after deposition and consequently has implications for the bioavailability of this atmospherically derived Fe. In this study, 24-hour aerosol samples were collected using a high-volume dichotomous virtual impactor (HVDVI) that collected coarse (D p > 2.5 mm) and fine (D p < 2.5 mm) aerosol fractions on two 90-mm Teflon membrane filters, over the tropical and subtropical North Atlantic Ocean. A sequential aqueous extraction procedure using a pH 4.5 buffer solution and a chemical reductant (hydroxylamine hydrochloride (HA)) was used to measure various labile Fe fractions. The extraction procedure was performed immediately after aerosol sample collection and used time series measurements of Fe(II) using long path length absorbance spectroscopy (LPAS) for analysis of Fe(II). The method measured both the quantities of labile Fe and also the dissolution and reduction kinetics of the labile Fe. Comparisons of HA-reducible Fe and photoreducible Fe concentrations were conducted on board and showed that both reduction processes had similar reduction kinetics and final Fe(II) concentrations during the initial 90 min. The average pseudo-first-order rate constants for the increase in Fe(II) were 0.020 and 0.0076 min À1 for the photoreducible Fe extraction and HA-reduction extraction, respectively. This HA-reducible Fe amount could potentially be used to determine the maximum amount of labile atmospheric Fe that is deposited into the ocean.