Search for a command to run...
Abstract Literature data on benthic dissolved iron (DFe) fluxes (µmol m −2 d −1 ), bottom water oxygen concentrations (O 2BW , μM), and sedimentary carbon oxidation rates (C OX , mmol m −2 d −1 ) from water depths ranging from 80 to 3700 m were assembled. The data were analyzed with a diagenetic iron model to derive an empirical function for predicting benthic DFe fluxes: where γ (= 170 µmol m −2 d −1 ) is the maximum flux for sediments at steady state located away from river mouths. This simple function unifies previous observations that C OX and O 2BW are important controls on DFe fluxes. Upscaling predicts a global DFe flux from continental margin sediments of 109 ± 55 Gmol yr −1 , of which 72 Gmol yr −1 is contributed by the shelf (<200 m) and 37 Gmol yr −1 by slope sediments (200–2000 m). The predicted deep‐sea flux (>2000 m) of 41 ± 21 Gmol yr −1 is unsupported by empirical data. Previous estimates of benthic DFe fluxes derived using global iron models are far lower (approximately 10–30 Gmol yr −1 ). This can be attributed to (i) inadequate treatment of the role of oxygen on benthic DFe fluxes and (ii) improper consideration of continental shelf processes due to coarse spatial resolution. Globally averaged DFe concentrations in surface waters simulated with the intermediate‐complexity University of Victoria Earth System Climate Model were a factor of 2 higher with the new function. We conclude that (i) the DFe flux from marginal sediments has been underestimated in the marine iron cycle and (ii) iron scavenging in the water column is more intense than currently presumed.