Articles | Volume 12, issue 19
Biogeosciences, 12, 5771–5792, 2015
Biogeosciences, 12, 5771–5792, 2015

Research article 12 Oct 2015

Research article | 12 Oct 2015

Modeling the global emission, transport and deposition of trace elements associated with mineral dust

Y. Zhang1,2, N. Mahowald2, R. A. Scanza2, E. Journet3, K. Desboeufs3, S. Albani2, J. F. Kok4, G. Zhuang1, Y. Chen1, D. D. Cohen5, A. Paytan6, M. D. Patey7, E. P. Achterberg7,9, J. P. Engelbrecht8, and K. W. Fomba10 Y. Zhang et al.
  • 1Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Department of Environmental Science and Engineering, Fudan University, Shanghai, China
  • 2Department of Earth and Atmospheric Science, Cornell University, Ithaca, NY, USA
  • 3LISA, UMR CNRS7583, Université Paris-Est Créteil et Université Paris-Diderot, Créteil, France
  • 4Department of Atmospheric and Oceanic Sciences, University of California, Los Angeles, CA, USA
  • 5Australian Nuclear Science and Technology Organization, Locked Bag 2001, Kirrawee DC, NSW, 2232, Australia
  • 6Earth and Planetary Sciences Department, University of California, Santa Cruz, CA 95064, USA
  • 7Ocean and Earth Science, National Oceanography Centre Southampton, University of Southampton, Southampton SO14 3ZH, UK
  • 8Desert Research Institute (DRI), 2215 Raggio Parkway, Reno, Nevada 89512-1095, USA
  • 9GEOMAR, Helmholtz Centre for Ocean Research, 24148 Kiel, Germany
  • 10Leibniz Institute for Tropospheric Research (TROPOS), 04318 Leipzig, Germany

Abstract. Trace element deposition from desert dust has important impacts on ocean primary productivity, the quantification of which could be useful in determining the magnitude and sign of the biogeochemical feedback on radiative forcing. However, the impact of elemental deposition to remote ocean regions is not well understood and is not currently included in global climate models. In this study, emission inventories for eight elements primarily of soil origin, Mg, P, Ca, Mn, Fe, K, Al, and Si are determined based on a global mineral data set and a soil data set. The resulting elemental fractions are used to drive the desert dust model in the Community Earth System Model (CESM) in order to simulate the elemental concentrations of atmospheric dust. Spatial variability of mineral dust elemental fractions is evident on a global scale, particularly for Ca. Simulations of global variations in the Ca / Al ratio, which typically range from around 0.1 to 5.0 in soils, are consistent with observations, suggesting that this ratio is a good signature for dust source regions. The simulated variable fractions of chemical elements are sufficiently different; estimates of deposition should include elemental variations, especially for Ca, Al and Fe. The model results have been evaluated with observations of elemental aerosol concentrations from desert regions and dust events in non-dust regions, providing insights into uncertainties in the modeling approach. The ratios between modeled and observed elemental fractions range from 0.7 to 1.6, except for Mg and Mn (3.4 and 3.5, respectively). Using the soil database improves the correspondence of the spatial heterogeneity in the modeling of several elements (Ca, Al and Fe) compared to observations. Total and soluble dust element fluxes to different ocean basins and ice sheet regions have been estimated, based on the model results. The annual inputs of soluble Mg, P, Ca, Mn, Fe and K associated with dust using the mineral data set are 0.30 Tg, 16.89 Gg, 1.32 Tg, 22.84 Gg, 0.068 Tg, and 0.15 Tg to global oceans and ice sheets.

Short summary
A new technique to determine a size-fractionated global soil elemental emission inventory based on a global soil and mineralogical data set is introduced. Spatial variability of mineral dust elemental fractions (8 elements, e.g., Ca, Fe, Al) is identified on a global scale, particularly for Ca. The Ca/Al ratio ranged between 0.1 and 5.0 and is confirmed as an indicator of dust source regions by a global dust model. Total and soluble dust element fluxes into different ocean basins are estimated.
Final-revised paper