The Biological Pump During the Last Glacial Maximum and Early Deglaciation


An important unanswered question in paleoclimate research is what caused the increase of atmospheric carbon dioxide (CO2) concentrations during the last deglaciation. During the height of the last ice age, the Last Glacial Maximum or LGM roughly 20,000 years ago, climate was cold and CO2 concentrations were low, around 180 parts per million (ppm). Subsequently CO2 concentrations rose to about 280 ppm causing the climate to warm, ice sheets to melt and sea levels to rise. The last deglaciation was the last time in Earth’s history when global climate warmed substantially, comparable to the warming projected for the coming centuries. Currently interactions between climate and the carbon cycle are not well understood, with possible implications for the accuracy of future climate projections. This project will contribute to a better understanding of Earth’s coupled climate-carbon cycle system. Previous research suggests that the carbon that was missing in the atmosphere during the ice age was stored somewhere in the ocean, but at present it is not clear where exactly it was, why it was there, and what the reasons were for its outgassing from the ocean to the atmosphere during the deglaciation. This project will synthesize carbon isotope data from ocean sediments and combine them with detailed model simulations in order to better understand the ocean’s circulation and carbon cycle during the early part of the last deglaciation from 20,000 to 13,000 years before the present. The data synthesis will be accomplished through an international collaborative project (OC3: Ocean Circulation and Carbon Cycling). A newly developed global climate model that includes three-dimensional ocean physics, biogeochemical cycles, isotopes of carbon and nitrogen, and sediments will be constrained by the OC3 carbon isotope synthesis and other existing paleoceanographic datasets such as a recent synthesis of nitrogen isotope data and an ongoing radiocarbon compilation. The goal is to reconstruct quantitatively how ocean carbon storage was affected by different processes, such as the biological pump, sea ice cover, ocean circulation, stratification, iron fertilization, sea level and sediment interactions. The project has the potential to improve our understanding of the deglacial ocean circulation, its carbon cycling, and the rise in atmospheric CO2.

Funded by

the National Science Foundation's Marine Geology and Geophysics Program


Schmittner and Somes (2016) illustrate how carbon and nitrogen isotopes, which provide complementary constraints, can be used together for improved reconstructions of ocean circulation and carbon cycling. Enhanced iron fertlization is required to fit the sediment carbon and nitrogen isotope data. Somes et al. (2017) show that despite large reductions in the input and output of fixed nitrogen to the ocean, its inventory during Last Glacial Maximum (LGM; ~20,000) years ago was not much different from that of today. However, if the phosphorous inventory was higher the nitrogen inventory may also have been higher, but current data do not provide good quantitative constraints on those inventories.

As part of the OC3 project, we have updated the calibration of carbon isotopes δ13Ccib measured in shells of foraminifera (Cibicides) that live on the sea floor. This measurement is used by paleoceanographers to reconstruct δ13CDIC, the isotopic composition of dissolved inorganic carbon (DIC) in bottom water, which includes information on circulation and carbon cycle changes in past oceans (Schmittner et al., 2017). We found relatively small secondary effects from carbonate ion concentrations and pressure on the δ13Ccib, the consideration of which will improve reconstructions.

Muglia et al. (2017) have improved ocean iron cycling in the UVic model and applied it to simulations of the LGM. This paper showed that not only increased iron fluxes from the dustier atmosphere influenced the glacial carbon cycle, but also reduced iron fluxes from sediments due to the lower sea level. A susequent study explored different ocean circulation states and different iron fluxes in LGM simulations (Muglia et al., 2018). It showed that two critical ingredients are required to reproduce carbon and nitrogen isotope reconstructions from sediments: (1) increased iron soluble fluxes in the Southern Ocean and (2) a weaker and shallower Atlantic Meridional Overturning Circulation (AMOC).

Khatiwala et al. (2019) used the best-fitting model from Muglia et al. (2018) for a detailed analysis of the carbon cycle and its effects on atmospheric CO2. It turns out that the full model can explain about three quarters of the observed CO2 drop of ~90 ppm during the LGM. Surprisingly, the largest factor in causing the CO2 decrease was ocean temperatures, which account for about 45 ppm. Increased soluble iron fluxes account for another 25-40 ppm, whereas circulation and sea ice changes have only minor effects.


Water Column δ13CDIC Data Core-Top Sediment δ13CCib Data


  • Khatiwala, S., A. Schmittner and J. Muglia (2019)
    Air-sea disequilibrium enhances ocean carbon storage during glacial periods
    Science Advances, 5(6), doi: 10.1126/sciadv.aaw4981
  • Muschitiello, F., W. J. D’Andrea, A. Schmittner, T. J. Heaton, M. L. Balascio, N. deRoberts, M. W. Caffee, T. E. Woodruff, K. C. Welten, L. C. Skinner, M. H. Simon and T. M. Dokken (2019)
    Deep-water circulation changes lead North Atlantic climate during deglaciation
    Nature Communications, 10(1), 1272, doi: 10.1038/s41467-019-09237-3.
  • Muglia, J. Skinner, L., and A. Schmittner (2018)
    Weak overturning circulation and high Southern Ocean nutrient utilization maximized glacial ocean carbon
    Earth and Planet. Sci. Let., 496, 47-56, doi:10.1016/j.epsl.2018.05.038.
  • Muglia, J. Somes, C. J., Nickelsen, L., and A. Schmittner (2017)
    Combined Effects of Atmospheric and Seafloor Iron Fluxes to the Glacial Ocean
    Paleoceanography, 32(11), 1204-1218, doi:10.1002/2016PA003077.
  • Schmittner, A., H. C. Bostock, O. Cartapanis, W. B. Curry, H. L. Filipsson, E. D. Galbraith, J. Gottschalk, J. C. Herguera, S. Jaccard, L. E. Lisiecki, D. C. Lund, G. Martínez-Méndez, J. Lynch-Stieglitz, A. Mackensen, E. Michel, A. C. Mix, D. W. Oppo, C. D. Peterson, E. L. Sikes, H. J. Spero, and C. Waelbroeck (2017)
    Calibration of the Carbon Isotope Composition (δ13C) of Epibenthic Foraminifera
    Paleoceanography, 32(6), doi:10.1002/2016PA003072.
  • Somes, C. J., Schmittner, A., Muglia, J. and A. Oschlies (2017)
    A three-dimensional model of the marine nitrogen cycle during the Last Glacial Maximum constrained by sedimentary isotopes
    Frontiers in Marine Science, 4, 108, doi:10.3389/fmars.2017.00108.
  • Schmittner, A., and C. J. Somes (2016)
    Complementary Constraints from Carbon (13C) and Nitrogen (15N) Isotopes on the Efficiency of the Glacial Ocean's Soft-Tissue Biological Pump
    Paleoceanography, 31, doi:10.1002/2015PA002905.
  • NSF Public Access Repository:


    last changed Sep. 3, 2019