1. Assessing the efficiency of iron fertilization on atmospheric CO2 using an intermediate complexity ecosystem model of the global ocean Olivier Aumont 1 and Laurent Bopp 2 1 IPSL / LODyC, Paris, France 2 IPSL / LSCE, Gif s/ Yvette, France The Ocean in a High CO2 World

2. Introduction : The HNLC regions The Iron hypothesis Fe

3. The Iron fertilization experiments A: IronexI B: IronexII C: SOIREE D: EisenEx E: SEEDS F: SOFEX G: Planktos H: SERIES A B C D E F G H Main results ChlorophyllFrom a 3 to a 40-fold increase, generally as diatoms pCO 2 A 30 to 90 atm drawdown in surface pCO 2 Export ProductionContrasting results, generally an increase DMSAn increase

4. Mitigation of atmospheric CO2 Large scale iron fertilization Iron fertilization can be used as a means of offsetting the anthropogenic carbone dioxide emission (Martin et al., 1991) Previous estimates (modeling studies) Peng and Broecker, 1991 Joos et al., 1991 Box modelsSouthern Ocean Preindustrial: -17 to -59 atm Anthropogenic : -64 to -107 atm Sarmiento and Orr, 1991OGCM Nutrient restoring Global Ocean Preindustrial: -3 to -72 atm Six and Maier-Reimer, 1993OGCM HAMOCC Southern Ocean Preindustrial: -34 atm Anthropogenic : -50 atm Archer et al., 2000OGCM HAMOCC Global Ocean Preindustrial: -50 atm Ganadesikan et al., 2003OGCM Nutrient restoring Patchy, equatorial Pacific Low efficiency, < 10% of increase in export production as atmospheric CO2

5. Questions Iron fertilization Can the model simulate the main features of the iron fertilization experiments ? What is the spatial and temporal variability of the response to fertilization ? What is the long-term efficiency of the fertilization ? Outline 1. Model description 2. Patchy iron fertilization 3. long-term iron fertilization on the global scale

6. Tools : Tools : Models OPA PISCES PO 4 3- Diatoms MicroZoo P.O.M D.O.M Si Iron Nano-phyto Meso Zoo NO 3 - NH 4 + SmallBig Euphotic Layer (10-200m)

7. Chlorophyll surface concentrations Seawifs (98-03) PISCES June January

8. Iron distribution Annual mean, surface Annual mean, 1000m 0.1 0.5 1 1.5 3 5 0.1 0.5 1 1.5 3 5

9. What limits diatoms growth ? NO 3 + NH 4 PO 4 FeSi

10. Iron Fertilization Patchy Iron Fertilization in the three main HNLC regions Experimental design - Iron concentration set to 2 nM in the mixed layer at day 2 and 5 - The model is integrated for 31 days - Fertilization applied over only one grid box

11. Iron Fertilization : The Southern Ocean (1) 2. Small response ( Chl < 0.7 mg Chl m -3 ) 3. Moderate response (0.7 < Chl< 2.5 mg Chl m -3 ) 4. Strong response (2.5 mg Chl m -3 < Chl ) 1. Blooming conditions (Chl > 1.5 mg Chl m -3 ) 0 2.0 4.0 6.0 0 20 40 60 80 Chla Diatoms relative abundance

12. The Southern Ocean pCO2 ( atm) Export ( ) -100 -60 -20 0 0 40 80 120 2. Si limitation, Si initial < 6 umol L 3. Mixed layer depth > 30 m, macronutrient replete 4. Favorable conditions, strongly iron limited 1. Stratification, ice retreat Why such responses ?

13. The Southern Ocean : Comparison with data pCO2 ( atm) Diatoms relative abundance Chla ( mg Chl m -3 ) SOFEX South SOFEX North SOIREE

14. Seasonal evolution January February July November

15. Iron Fertilization : The equatorial Pacific Diatoms relative abundance Chla ( mg Chl m -3 ) IRONEX II pCO2 ( atm) export (%)

16. Iron Fertilization : everywhere & 50 yr long Changes in Diatoms Relative Abundance +1 Export Production (GtC/yr) 7 8 -10 0 10 20 30 40 50 Years Fe Fertilization Increase of Export Production +50 +5 -5 -50 (gC/m2/yr) +4 -4 +0.2 +1 -0.2 Changes in Chla (mg Chl m -3 )

17. -10 0 10 20 30 40 50 Years Fe Fertilization Atmospheric pCO2 ( atm) Carbon Flux (PgC/yr) 1 0 0 Impact on atmospheric pCO2 Preindustrial conditions -10 -4 atm in 10 yr Fe Fertilization 0.5 0 0 -10 7 8 Atmospheric pCO2 ( atm) Carbon Flux (PgC/yr) Export (PgC/yr) -1.8 atm in 50 yr -10 0 10 20 30 40 50 -8 atm in 50 yr >80% due to Southern Ocean

18. Why such a small efficiency ? Nutrient Limitation of Diatoms Growth NO 3 / NH 4 PO 4 FeSi Control Fe Fert. Light limitation -10 -2 2 1 NO3 ( mol/L) 1 4 10 20

19. Iron Fertilization : Implications for the Sulfur Cycle Changes in Surface DMS Concentrations Export Production Atmospheric pCO2 Carbon Flux -8 ppm in 50 yr -5 -0.5 +0.5 +5 nM 7 8 1 0 0 -10 DMS Flux (TgS/yr) -15 % -10 0 10 20 30 40 50 22 26

20. Conclusions Patchy Iron fertilization : The model roughly captures the main features of in situ iron fertilization experiments, except in the North Pacific. In the Southern Ocean, the response depends highly on the location and the time period of the iron release. Main controlling factors are Si concentrations, the mixed layer depth, and the status of the ecosystem. The favorable season extends from November to March. Large-scale Iron fertilization : Very low efficiency : only 8 ppmv drawdown in atmospheric pCO2 after 50 years. Iron fertilization should be done continuously to keep the additionally stored CO2 within the ocean. Possible drawbacks : N2O production, extension of the anoxic regions, changes in the fisheries, possible decrease in DMS production, …

21. Diatoms relative abundance : vs Data 1 0.8 0.6 0.4 0.2 0 Data from Gregg et al. 2003 JanFebMarAprMayJun JulAugSepOctNovDec 60°S 40°S 20°S 0° 20°N 40°N 60°N 80°N

23. Iron Fertilization : The North Pacific Diatoms relative abundance Chla ( mg Chl m -3 )