Observations and modeling the ocean Fe cycle Role

Observations and modeling the ocean Fe cycle: Role in the carbon cycle and state of understanding Ed Boyle Earth, Atmospheric and Planetary Sciences Massachusetts Institute of Technology Cambridge MA 02139

Basic premise: We don’t understand the processes that control the oceanic Fe distribution well enough to design a realistic simulation of iron in the ocean. The task at present is to take simple representations of what we know and see how far these get us - and what they tell us about what observations, experiments, and modeling are needed.

Recent Fe modeling references • Aumont O. and Bopp L. (2006 a) Globalizing results from ocean in situ iron fertilization studies Glob. Biogeochem. Cycles 20, GB 2017, doi: 10. 1029/2005 GB 002591. • Aumont O. and Bopp L. (2006 b) Globalizing results from ocean in situ iron fertilization studies, Glob. Biogeochem. Cyc. 20, GB 2017, doi: 10. 1029/2005 GB 002591. • Christiana J. R. , Verschellb M. A. , Murtuguddec R. , Busalacchib A. J. , and Mc. Claina C. R. (2002) Biogeochemical modelling of the tropical Pacific Ocean. II: Iron biogeochemistry, Deep-Sea Res. II 49, 545565. • Moore J. K. , Doney S. C. , and Lindsay K. (2004) Upper ocean ecosystem dynamics and iron cycling in a global three-dimensional model Glob. Biogeochem. Cyc. 18, GB 4028, doi: 10. 1029/2004 GB 002220. • Moore J. K. , Doney S. C. , Lindsay K. , Mahowald N. , and Michaels A. F. (2006) Nitrogen fixation amplifies the ocean biogeochemical response to decadal timescale variations in mineral dust deposition Tellus B 58, 560– 572 • Moore J. K. and S. Doney. (2007) Iron availability limits the ocean nitrogen inventory stabilizing feedbacks between marine denitrification and nitrogen fixation Glob. Biogeochem. Cyc. 21, GB 2001, doi: 10. 1029/2006 GB 002762. • Patra P. K. , Moore J. K. , Mahowald N. , Uematsu M. , Doney S. C. , and Nakazawa T. (2007) Exploring the sensitivity of interannual basin-scale air-sea CO 2 fluxes to variability in atmospheric dust deposition using ocean carbon cycle models and atmospheric CO 2 inversions J. Geophys. Res. 112, G 02012, doi: 10. 1029/2006 JG 000236. • Tagliabue A. and Arrigo K. R. (2006) Processes governing the supply of iron to phytoplankton in stratified seas J. Geophys. Res. 111, C 06019, doi: 10. 1029/2005 JC 003363. • Tagliabue A. , Bopp L. , and Aumont O. (2007) Ocean biogeochemistry exhibits contrasting responses to a large scale reduction in dust deposition Biogeosciences Discuss. 4(1 -33). • Weber L. , Volker C. , Oschlies A. , and Burchard H. (2007) Iron profiles and speciation of the upper water column at the Bermuda Atlantic time-series Study site: a model based sensitivity study Biogeosciences Discuss. 4(823 -869). • Weber L. , Volker C. , Schartau M. , and Wolf-Gladrow D. A. (2005) Modeling the speciation and biogeochemistry of iron at the Bermuda Atlantic Time-series Study site Glob. Biogeochem. Cyc. 19, GB 1019, doi: 10. 1029/2004 GB 002340.

Some simple Fe model representations • We know the dust flux into the ocean • We know how much iron is released from that dust • Fe dissolution from dust only occurs in the mixed layer • Fe from dust is the only significant source of Fe to the ocean • The Fe: C ratio of phytoplankton is constant and known • Fe ligand concentrations and binding constants are constant throughout the deep ocean • Fe scavenging is simply proportional to free [Fe+++]

Ten top-to-bottom open-ocean iron profiles

Average of data between 2800 and 4000 m: 0. 43 +/- 0. 04 (1 s s. d. ) nmol/kg

Simple representations 1 and 2: • we know the dust flux into the ocean • we know how much iron is released from that dust. In reality: • The dust flux has been estimated from atmospheric dust concentration data at only a few points in the world. Everywhere else, we are extrapolating from intuition, satellite-based column loading estimates, and atmospheric dust models. • There is a large range of estimates for the percentage of Fe released from dust under very different experimental conditions. It is likely that the dust release percentage varies from one place to the other, and under different conditions at the same place.

Global dust fluxes

Fe in the surface waters of the Western North Atlantic

Atlantic Surface Fe N-S Transect

Fe release from Bermuda aerosols Sedwick et al. , in press

Simple representation 3: Fe dissolution from dust only occurs in the mixed layer In reality: Although it is reasonable to presume that a large dissolution flux of Fe is “primed” when the dust falls into the ocean, it is difficult to prove that dissolution does not continue as the dust falls through the depths (or put another way, difficult to quantify how much Fe is released from dust as it falls through the deep ocean).

Simple representation 4: Fe from dust is the only significant source of Fe to the ocean In reality: Oceanic Fe may have significant sources from rivers, continental shelf sediments, continental margin sediments, and hydrothermal vents. These sources have never been properly quantified.

Simple representation 5: The Fe: C ratio of phytoplankton is constant and known In reality: Although Fe is an essential micronutrient, it appears that different organisms have evolved different abilities to survive with different Fe supplies. Open-ocean Antarctic organisms probably survive with the minimum amount of Fe. Organisms under high-dust or coastal Fe inputs may take up more Fe than that, and release more Fe when they sink and regenerate.

Tropical Atlantic

Measures/Landing Atlantic Fe section

Tropical Atlantic Fe maximum occurs within the oxygen minimum

Fe vs. P

Simple representation 6: Fe ligand concentrations and binding constants are constant throughout the deep ocean. In reality: • We have a very limited data base on Fe ligands (note that a titration of a single sample can take about a day)

Trans - North Pacific Ligand Data K. N. Buck, C. I. Measures, W. M. Landing, K. W. Bruland, K. Barbeau, in prep

Simple representation 7: Fe scavenging is simply proportional to free [Fe+++] In reality: Most of the variability of Fe in the deep ocean is seen in the colloidal fraction. It may be that the colloidal fraction is scavenged, and the soluble fraction is (relatively) inert.

Subtropical Atlantic

Western South Atlantic

Fe decreases as NADW moves from the North Atlantic into the South Atlantic: Bridget’s Scavenging Residence Time Estimate: North Atlantic Fe: 0. 67 0. 09 (9) South Atlantic Fe: 0. 47 0. 02 (7) Scavenging Residence Time: (based on Broecker C 14 interpretation => 56 year transit time) 270 140 years

<0. 02 µm Fe is nearly constant in deep waters (0. 2 -0. 4 nmol/kg). Most deep-sea Fe variability is due to changes in colloidal Fe.

<0. 4 µm (DFe) vs 0. 02 -0. 4 µm (CFe) Bergquist et al. (2007) Geochim. Cosmochim. Acta 71: 2960

Parting remarks Temporal variability of Fe in the ocean is important but little understood.

MANTRA project, Sept. /Oct 2004 MP 6 Fe, nmol/kg

BTM Fe data, 1996 -2000

Recent Fe modeling references • Aumont O. and Bopp L. (2006 a) Globalizing results from ocean in situ iron fertilization studies Glob. Biogeochem. Cycles 20, GB 2017, doi: 10. 1029/2005 GB 002591. • Aumont O. and Bopp L. (2006 b) Globalizing results from ocean in situ iron fertilization studies, Glob. Biogeochem. Cyc. 20, GB 2017, doi: 10. 1029/2005 GB 002591. • Christiana J. R. , Verschellb M. A. , Murtuguddec R. , Busalacchib A. J. , and Mc. Claina C. R. (2002) Biogeochemical modelling of the tropical Pacific Ocean. II: Iron biogeochemistry, Deep-Sea Res. II 49, 545565. • Moore J. K. , Doney S. C. , and Lindsay K. (2004) Upper ocean ecosystem dynamics and iron cycling in a global three-dimensional model Glob. Biogeochem. Cyc. 18, GB 4028, doi: 10. 1029/2004 GB 002220. • Moore J. K. , Doney S. C. , Lindsay K. , Mahowald N. , and Michaels A. F. (2006) Nitrogen fixation amplifies the ocean biogeochemical response to decadal timescale variations in mineral dust deposition Tellus B 58, 560 – 572 • Moore J. K. and S. Doney. (2007) Iron availability limits the ocean nitrogen inventory stabilizing feedbacks between marine denitrification and nitrogen fixation Glob. Biogeochem. Cyc. 21, GB 2001, doi: 10. 1029/2006 GB 002762. • Patra P. K. , Moore J. K. , Mahowald N. , Uematsu M. , Doney S. C. , and Nakazawa T. (2007) Exploring the sensitivity of interannual basin-scale air-sea CO 2 fluxes to variability in atmospheric dust deposition using ocean carbon cycle models and atmospheric CO 2 inversions J. Geophys. Res. 112, G 02012, doi: 10. 1029/2006 JG 000236. • Tagliabue A. and Arrigo K. R. (2006) Processes governing the supply of iron to phytoplankton in stratified seas J. Geophys. Res. 111, C 06019, doi: 10. 1029/2005 JC 003363. • Tagliabue A. , Bopp L. , and Aumont O. (2007) Ocean biogeochemistry exhibits contrasting responses to a large scale reduction in dust deposition Biogeosciences Discuss. 4(1 -33). • Weber L. , Volker C. , Oschlies A. , and Burchard H. (2007) Iron profiles and speciation of the upper water column at the Bermuda Atlantic time-series Study site: a model based sensitivity study Biogeosciences Discuss. 4(823 -869). • Weber L. , Volker C. , Schartau M. , and Wolf-Gladrow D. A. (2005) Modeling the speciation and biogeochemistry of iron at the Bermuda Atlantic Time-series Study site Glob. Biogeochem. Cyc. 19, GB 1019, doi: 10. 1029/2004 GB 002340.

Ten top-to-bottom open-ocean iron profiles

24. 5°S Western Atlantic Fe profile

Ten top-to-bottom open-ocean iron profiles

Northern North Atlantic

Average of data between 2800 and 4000 m: 0. 43 +/- 0. 04 (1 s s. d. ) nmol/kg

Martin Fe at PAPA

Tropical Atlantic

Central North Pacific

Northern North Pacific

Simple representation 8: Free Fe 3+ is the only bio-available form of Fe Shaked

Fe in the surface waters of the Western Atlantic Latitude

There are very few deepwater profiles for Fe in the open ocean, yet oceanic Fe limitation is created by century-scale scavenging in the deep sea… • According to prevailing thinking (challenged occasionally by hydrothermal vent and coastal sediment researchers), Fe enters the ocean dominantly in the Northern Hemisphere from eolian deposition, and enters the deep sea dominantly by regeneration from falling biogenic debris. • In the deep sea, Fe is scavenged on a century time scale; when this water upwells to the surface, it is deficient in Fe relative to N, P, Si nutrients. • Atlantic deep waters have the highest levels of Fe. Fe decreases along the conveyor belt to low levels in the Antarctic and South Pacific and then increases a little bit in the far Northern Pacific (probably due to Fe released from continental margin sediments).

Laës (2003) GRL 30: 1902