Tracing and constraining anthropogenic aerosol iron fluxes to the North Atlantic Ocean using iron isotopes

Abstract

Atmospheric dust is an important source of the micronutrient Fe to the oceans. Although relatively insoluble mineral Fe is assumed to be the most important component of dust, a relatively small yet highly soluble anthropogenic component may also be significant. However, quantifying the importance of anthropogenic Fe to the global oceans requires a tracer which can be used to identify and constrain anthropogenic aerosols in situ. Here, we present Fe isotope (δ 56 Fe) data from North Atlantic aerosol samples from the GEOTRACES GA03 section. While soluble aerosol samples collected near the Sahara have near-crustal δ 56 Fe, soluble aerosols from near North America and Europe instead have remarkably fractionated δ 56 Fe values (as light as −1.6‰). Here, we use these observations to fingerprint anthropogenic combustion sources, and to refine aerosol deposition modeling. We show that soluble anthropogenic aerosol Fe flux to the global surface oceans is highly likely to be underestimated, even in the dusty North Atlantic.

Impact of Changes to the Atmospheric Soluble Iron Deposition Flux on Ocean Biogeochemical Cycles in the Anthropocene

“…The approximate factor of 2 increase in PI fire emissions between the two fire models is due in part to different representations of how anthropogenic land use and land cover change over the historical period affect burnt area, and thus emission, as well as LMfire containing an additional emission source from agricultural fires. In the PDHIGHCOMB emission scenario the model simulations are brought into better agreement with recent iron observations (Conway et al, ; Matsui et al, ) by applying a global uniform factor of 5 multiplier to the Luo et al () anthropogenic combustion iron emissions used in the PDBASE simulation, resulting in anthropogenic combustion iron emissions increasing from 0.68 to 3.4 Tg a −1 . Over the Industrial Era, reductions to fire iron emissions may be partially offsetting increases in dust iron emissions, but likely restricted to the Southern Hemisphere (SH) where the dust loadings and anthropogenic combustion emission increases are both smaller than in the Northern Hemisphere (NH).…”

“…To this we added shipping emissions of iron based on the iron content and emissions of particulate matters (Ito et al, ). For the PDHIGHCOMB scenario we applied a global factor of 5 increase to PIBASE anthropogenic combustion emissions, in line with recent observations of their potential magnitude (Conway et al, ; Matsui et al, ). In the FU we assumed that anthropogenic emissions will follow the intermediate emissions scenario “RCP4.5” pathway, and a monthly PD:FU fossil fuel BC emission ratio (Figure S2) was applied to PD anthropogenic combustion iron emissions to estimate their FU flux.…”

“…In the PDHIGHCOMB emission scenario the model simulations are brought into better agreement with recent iron observations (Conway et al, 2019;Matsui et al, 2018) by applying a global uniform factor of 5 multiplier to the Luo et al (2008) anthropogenic combustion iron emissions used in the PDBASE simulation, resulting in anthropogenic combustion iron emissions increasing from 0.68 to 3.4 Tg a −1 . Over the Industrial Era, reductions to fire iron emissions may be partially offsetting increases in dust iron emissions, but likely restricted to the Southern Hemisphere (SH) where the dust loadings and anthropogenic combustion emission increases are both smaller than in the Northern Hemisphere (NH).…”

Improved methodologies for Earth system modelling of atmospheric soluble iron and observation comparisons using the Mechanism of Intermediate complexity for Modelling Iron (MIMI v1.0)

“…The acidic processing of iron contained in aerosol is one pathway through which soluble iron can be liberated from an insoluble form with decreasing pH (Duce and Tindale, 1991;Solmon et al, 2009;Zhu et al, 1997). Organic ligands, in particular oxalate, also increase iron solubility by weakening or cleaving the Fe-O bonds found in iron oxide minerals via complexation (Li et al, 2018;Panias et al, 1996), and in nature this reaction proceeds most rapidly in a slightly acidic aqueous medium, such as cloud droplets (Cornell and Schindler, 1987;Paris et al, 2011;Xu and Gao, 2008). Organic ligand processing has been estimated to increase soluble iron concentrations by up to 75 % more than is achievable with acid processing alone (Ito, 2015;Johnson and Meskhidze, 2013;Myriokefalitakis et al, 2015;Scanza et al, 2018).…”

“…First, we transitioned from a bulk aerosol scheme to a two-moment modal aerosol scheme (Liu et al, 2012), and second, we re-evaluated pyrogenic iron emissions from anthropogenic combustion and fires. The modal aerosol scheme was used to calculate both aerosol mass and number at each time step within an updated global aerosol microphysics model, and both the fire and anthropogenic combustion emissions from Luo et al (2008), which are likely to be underestimated (Conway et al, 2019;Ito et al, 2019;Matsui et al, 2018), were improved upon.…”