Ocean Biogeochemistry and Climate Change

Ocean biogeochemistry is a branch of ocean science dealing with the exchange and distribution of chemical elements and compounds among various living and non-living entities within the ocean itself, as well as between the ocean and its neighboring environments (such as the atmosphere and the land). Biogeochemistry also refers to the particular features (such as quantities, rates, space or time) of that exchange and distribution. For instance, currently a quarter of carbon dioxide (CO2) released into the air by burning fossil fuels is absorbed by the ocean, which contains about 50 times more carbon (C) than the atmosphere, thanks to the physical gas transfer, the chemical behavior of CO2 dissolved in water, and the demand for CO2 in photosynthesis that takes place in the ocean. Through its uptake and storage of C, the ocean moderates global warming. Yet, this moderation can easily be changed by the effects of changing climate on the parts of the global ocean system that are involved in C biogeochemistry.

(1). Oceanic Controls on Atmospheric CO2: theories and models.

With the help of complex global climate models, we study the interconnection between the physical climate and various ocean components responsible for carbon uptake and storage (the so-called "carbon pumps").
Because ocean ventilation provides a pathway for atmospheric carbon dioxide—both natural and anthropogenic—to invade a massive volume of the ocean, changes in ventilation are likely to have profound implications for global climate over the 21st century. I am interested in understanding the separate impacts of buoyancy and wind driven changes in ventilation on the oceanic uptake and storage of carbon, with a focus on the critical role of the Southern Ocean. I am also looking for students to work on improving our theoretical understanding of the physical (circulation) and chemical controls of ocean carbon pumps (e.g., Marinov et al. 2008) and anthropogenic carbon uptake of the ocean. A previous project on this topic was funded by NOAA grant NA10OAR4320092 (2010-2013).

Example of theoretical development from Marinov et al. JGR 2008, showing that the more biological carbon is stored in the oceans, the lower atmospheric CO2 is. Circles represent various ocean model simulations, while the lines are theoretical predictions

(2) Deep water formation, convection and links with biogeochemistry

Coupled model simulations show clearly that a major source of uncertainty in IPCC type predictions of future climate is the wide range of predictions of water mass transformations in the Southern Ocean and shelf areas around Antarctica (Friedlingstein et al., 2006, Russell et al., 2006). Southern Hemisphere westerly winds play a crucial role in setting up the stratification and large scale circulation in the Southern Ocean, influencing the amount of upwelling, downwelling and lateral mixing in the region. Recent research shows that in the past 40 years Southern Hemisphere westerlies shifted poleward and the zonal wind speeds increased. Separately, the high latitude ocean has warmed up and freshened. The obvious question to ask is: What is the impact of changing winds on the oceanic solubility and biological “natural” carbon pumps, the net anthropogenic carbon uptake as well as on regional ecology?

Our research to date suggests that deep water formation areas in the Southern Ocean are mostly responsible for setting the air-sea carbon balance on long time scales (of thousands of years). The role of the Southern ocean in the shorter (decadal) time scale climate change is a more recent, and less studied focus area. Some of the deep water formation in the ocean has historically formed in open sea convection areas such as the Weddell Sea. Interesting questions include: What percent of Southern Ocean convection occurs on shelves and what percent occurs in the open ocean? How does deep ocean convection work? What is the role of convection in the exchange of carbon and oxygen between the atmosphere and the ocean and ultimately in the oceanic carbon storage on various timescales?

(3) Ocean Anoxia in the modern and future oceans.

We also want to understand how climate change will affect the availability and distribution of oxygen and nutrients for marine organisms, and what will be the related consequences for ocean ecosystems and fish (who need O2 to breathe). Outstanding questions are whether the extent of anoxia in the tropical oceans will increase as the ocean becomes more stratified over the 21st century, and how short term climate variability combines with long-term climate signals to affect oxygen globally. Cabre et al. (2015; figure below) discusses the present patterns of oxygen in the ocean and future expected changes over the 21st century across the latest generation of climate models.


Spatially averaged distribution of dissolved oxygen across latitudes (horizontal axes) and down the water column (vertical axes, showing depth in meters) in the Pacific between 180°W and 100°W. (a) Observations of dissolved oxygen concentration averaged over time, sourced from the World Ocean Atlas 2009. (b) Dissolved oxygen concentration for the period 1960-1999 derived by averaging the output of several climate models, which were included in the Coupled Model Intercomparison Project Phase 5 (CMPI5). (c) Future change in dissolved oxygen concentration, computed as the difference between CMIP5 multi-model average projections for the period 2060-2099 and the values presented in (b). The predicted 100-year change panel (c) includes diagonal pattern for trends that are consistent across CMIP5 models at the 90% level, and crossed patterns for 80% level.

(4) Paleoclimate

Our research in carbon pumps and oxygen dynamics is directly applicable to the understanding of the glacial-interglacial cycles. I am interested in collaborating with paleoclimatologists to deconvolve the relative roles of changes in ocean biology and stratification in driving glacial-interglacial temperature and CO2 cycles. Interesting questions to pursue include: How have (a) the intensity and spread of deep water formation, as well as high latitude stratification (b) high latitude biogeochemistry and the iron-light colimitation of ecology and (c) the global efficiency of the biological pump - changed on glacial-interglacial scales and how does this parallel the changes that will occur with future climate change? What is the relative impact of each of these mechanisms on atmospheric pCO2 on long paleoclimate timescales?

Ocean Biogeochemistry research team members:

  • Prof Irina Marinov
  • postdoctoral fellow: Dr. Anna Cabré
  • Penn student: Ryan Dungee
  • Collaborator: Dr. Raffaele Bernardello (former postdoc in our group, now at Southampton, UK)
  • Collaborator: Dr. Daniele Bianchi (Assistant Prof, UCLA)
  • Penn student alumni: Karan Sharma, Raghu Palle, Lingbin Cai
  • Representative publications:

    Cabré, A., I. Marinov, R. Bernardello, and D. Bianchi: Oxygen minimum zones in the tropical Pacific across CMIP5 models: mean state differences and climate change trends (2015). Biogeosciences, 12, 5429–5454 doi:10.5194/bg-12-5429-2015.

    Bernardello, R., I. Marinov, JB Palter, ED Galbraith: Impact of Weddell Sea convection on anthropogenic carbon uptake in a coupled model (2014). Geophysical Research Letters.40: 7262–7269 doi: 10.1002/2014GL061313. (PDF)

    Bernardello, R., I. Marinov, JB Palter, JL Sarmiento, ED Galbraith, RD Slater: Response of the Ocean Carbon Pumps to Projected 21st Century Climate Change (2014). Journal of Climate. doi: 10.1175/JCLI-D-13-00343.1 (PDF)

    Moore, C.M., Mills, M.M., Arrigo, K., Berman-Frank, I., Bopp, L., Boyd, P.W., Galbraith, E., Geider, R.J., Guieu, C., Jaccard, S, Jickells, T., La Roche, J., Lenton, T., Mahowold, N., Maranon, E., Marinov, I., Moore, K., Nakatsuka, T., Oschlies, A., Saito, M., Thingstad, F., Tsuda, A., Ulloa, O. and Wallace, D.: Nutrient limitation in the upper ocean: processes, patterns and potential for change (2013). Nature Geosciences. (PDF)

    J. Palter, I. Marinov, N. Gruber, J. L. Sarmiento: Large-scale, persistent nutrient fronts of the world ocean: impacts on biogeochemistry (2013). I.M. Belkin (ed.), Chemical Oceanography of Frontal Zones. Berlin: Springer-Verlag. (PDF)

    Gnandesikan, A., I. Marinov: Export is not enough: Nutrient cycling and carbon sequestration (2008). Marine Ecological Progress Series, invited contribution to the Thematic Section on "Implications of large scale iron fertilization of the oceans", Vol 364:289-294, doi:10.3354/meps/07750. (PDF)

    Marinov, I., M. Folows, A. Gnandesikan, J.L. Sarmiento, and R.D. Slater: How does ocean biology affect atmospheric pCO2? (2008). Theory and Models, JGR Oceans, Vol 113, C07032, doi:10.1029/2007JC004598. (PDF)

    Marinov, I., A. Gnanadesikan, J.L. Sarmiento, R. Toggweiler and B. Mignone: Impact of oceanic circulation on the ocean biological carbon storage and atmospheric pCO2 (2008). Global Biogeochem. Cycles, Vol 22, GB3007, doi:10.1029/2007GB002958. (PDF)

    Marinov, I., A. Gnanadesikan, R. Toggweiler, and J.L. Sarmiento: The Southern Ocean Biogeochemical Divide (2006). Nature (441), doi:10.1038. (PDF| Supplementary)

    Finalized Research Projects:

    Biological-physical controls on the large scale air-sea CO2 flux distributions. The compensation mechanism.

    Irina Marinov with Anand Gnanadesikan, published in Biogeosciences, 2011

    This work explores the impact of circulation changes on the meridional distribution of the steady state air-sea CO2 fluxes. Changes in circulation resulting from modifications in diapycnal mixing or changes in Southern Ocean winds affect both remineralized phosphate and temperature. In the Princeton GCM the biological CO2 flux, which is a function of remineralized phosphate, and the solubility CO2 flux, which is a function of temperature or heat transport, change with diapycnal mixing and Southern Ocean wind magnitude in opposite ways. Thus, while both the biological and solubility air-sea CO2 fluxes vary strongly with diapycnal mixing or wind magnitude, the full (solubility+biological) air-sea CO2 flux shows nearly no variation. If our result holds in the real ocean, it could potentially imply that large changes in the air-sea carbon distribution (due, for example, to the observed increase in Southern Ocean winds) would not be reflected in the observed air-sea CO 2 flux. Since surface restoring of temperature and salinity constraints strongly the solubility air-sea flux, while surface restoring of nutrients constraint the biological air-sea flux, this result highlights the need for using realistic boundary conditions in climate simulations. The validity of this result needs to be tested in more complex models in which surface boundary conditions are allowed to vary in response to circulation changes.

    The Southern Ocean Biogeochemical Divide.

    (I. Marinov, A. Gnanadesikan, R. Toggweiler and J.L. Sarmiento, published in Nature, June 2006)
    (read full text)

    Previous studies have shown that the Southern Ocean is crucial in controlling the atmosphere-ocean balance of carbon dioxide as well as global biological production. Here we demonstrate that two separate regions of the Southern Ocean (the Antarctic and the Subantarctic) control the air-sea carbon dioxide balance and global biological production. The biogeochemical divide between these two regions is likely located in the vicinity of the Polar Front. The large-scale pattern of circulation in the Southern Ocean involves upwelling of deep water, some of which flows to the south to sink as bottom water, and some of which flows to the north to form intermediate and mode waters. We show that the air-sea balance of carbon dioxide is controlled primarily by the biological pump and circulation in the deep-water formation region, whereas global biological productivity is controlled primarily by the biological pump and circulation in the intermediate and mode water formation region. This implies that it may be possible for climate change or human intervention to modify one of these without greatly altering the other.

    Impact of oceanic circulation on biological carbon storage in the ocean and atmospheric pCO2.

    (Marinov, I., A. Gnanadesikan, J.L. Sarmiento, J.R. Toggweiler, M. Follows and B. Mignone, published in GBC, 2008)
    (read full text)

    The atmospheric carbon dioxide partial pressure (pCO2) is set to a large degree by the biological storage of carbon in the deep ocean. A more efficient biological pump results in more carbon storage in the deep ocean and smaller atmospheric pCO2. This study shows that diapycnal mixing, isopycnal mixing and Southern Ocean winds, by changing the Southern Ocean overturning circulation, influence strongly the biological storage of carbon in the ocean and atmospheric pCO2 in a realistic ocean General Circulation Model (GCM). Increased diapycnal mixing and Southern Ocean winds result in less ocean carbon storage and higher atmospheric pCO2, in agreement with earlier box model studies. By contrast, increased isopycnal mixing is shown to increase the storage of carbon in the ocean and decrease atmospheric pCO2.
    Additionally, this paper attempts to clarify a longstanding confusion in the oceanic community about what controls the biological storage of carbon in the ocean. As such, we show conclusively that surface nutrients and biological export production are not good metrics for ocean carbon storage and atmospheric pCO2. By contrast, we show that the fraction of nutrients in the global remineralized pool or the fraction of nutrients in the global preformed pool are excellent indicators for atmospheric pCO2. We develop a simple theory relating global preformed nutrient concentrations and atmospheric pCO2.

    How does atmospheric pCO2 respond to changes in surface nutrients, such as those associated with iron fertilization of the surface ocean?

    Atmospheric pCO2 sensitivity to nutrient depletion in the ocean: theory and models by I. Marinov, M. Follows, A. Gnanadesikan, J.L. Sarmiento and R. Slater, published in JGR Oceans, 2008.
    (read full text)

    Iron fertilization of the HNLC (high-nutrient low-chlorophyll) ocean areas has been proposed as a mechanism to decrease atmospheric CO2 levels through the associated increase in biological production. Increased surface biological production through iron fertilization is one of the leading hypotheses for explaining the lower atmospheric pCO2 observed during glacial times [Martin, 1990]. In the present work we study the impact of increasing biological production on atmospheric pCO2 in the Princeton GCM. The method used is to deplete (i.e., force towards zero) nutrients in large ocean areas and convert them to export production. Here we show how the uptake of atmospheric CO2 following nutrient depletion depends on the region depleted, gas exchange rate, ice, and oceanic circulation. We show that the Southern Ocean (and in particular the Antarctic region south of the Polar Front) is the most important area for CO2 uptake in all models studied. We show that the outcome of nutrient depletion experiments depends critically on the circulation of the ocean, which changes as diapycnal mixing and Southern Ocean winds change. Depleting surface nutrients changes both deep preformed nutrients and the CO2 disequilibrium at the ocean surface, with implications for the total carbon storage in the ocean. We examine how these two effects contribute to changes in atmospheric pCO2 in the context of models with different circulations. In conclusion, we show that ocean physics and the details of the air-sea gas exchange mechanism are crucial in determining the atmospheric pCO2 response to surface nutrient depletion.

    Export is not enough: Nutrient cycling and carbon sequestration.

    Anand Gnanadesikan and Irina Marinov
    published in 2008 in the Marine Ecology Progress Series Theme Section (MEPS-TS) "Implications of large scale iron fertilization of the oceans”

    (read full text)

    The question of whether iron fertilization can yield verifiable carbon sequestration is often cast in terms of whether fertilization results in enhanced particle export. However, models studies show that oceanic carbon storage is only weakly related to global particle export - depending instead on an increase in the carbon associated with the global pool of remineralized nutrients. As a result, local balances are unlikely to describe the global impact of fertilization. Effects that are remote from the fertilization site in time or space, such as reduction in productivity, changes in stoichiometric ratios or changes in the disequilibrium of sinking water can significantly affect the impact of fertilization on atmospheric carbon dioxide. Understanding these effects and constructing models, which accurately represent them, is thus a crucial part of designing large-scale fertilization projects.