Skip to main content

Ocean carbon cycling since the middle Miocene: Testing the metabolic hypothesis

This project focuses on marine carbon cycling since the Middle Miocene Climate Optimum (MMCO) 15 million years ago.

The ocean biological carbon pump, comprising photosynthesis, food web interactions and gravity, results in the removal of carbon from the surface ocean and its transport through the ocean to the deep (Fig. 1).

Diagram of the Oceanic Food Web
Fig. 1: The ocean biological carbon pump, comprising photosynthesis, food web interactions and gravity, results in the removal of carbon from the surface ocean and its transport through the ocean to the deep. (Image: Office of Biological and Environmental Research of the U.S. Department of Energy Office of Science. (U.S. DOE 2008))

Only about 10% of the carbon exported from the surface ocean makes it to the deep and an even smaller fraction ends up on the sea floor. Biological processes such as photosynthesis and respiration in the euphotic zone, ingestion and respiration/remineralization of sinking organic matter in the mesopelagic zone (Fig. 2), control how much organic carbon reaches the deep ocean and eventually the ocean floor.

Diagram of the simplified ocean biological carbon pump
Fig. 2: The complex food web can be simplified into larger scale processes that result in carbon being removed from the surface ocean. Organic matter (including carbon) in particulate form sinks due to gravity, and some of this ends up buried at the seafloor, forming ocean sediments. (Image: Katherine Crichton)

Ocean temperature appears to significantly affect the rate at which planktonic organisms process carbon (metabolism) with both photosynthesis and respiration occurring faster at warmer temperatures. However, heterotrophic respiration responds twice as fast to temperature changes than photosynthesis. This may potentially cause major changes to the carbon cycle, with more carbon being sequestered when ocean temperatures are cooler and vice versa when warming occurs.

Project hypothesis

Can this temperature dependency of metabolic rates in turn act as a global climate feedback? This intriguing idea is the 'Metabolic Hypothesis' as articulated by Olivarez Lyle and Lyle in 2006.

This project aims to answer this hypothesis by focusing on marine carbon and nutrient cycling since the Middle Miocene Climate Optimum (MMCO), that is, from about 15 million years ago (Ma) to the present – an interval of generally declining temperatures (Fig. 3).

Diagram showing benthic foraminiferal oxygen isotope stack since 18 Ma
Fig. 3: Benthic foraminiferal oxygen isotope stack since 18 Ma (millions of years before present) from Zachos et al. (2008). Trends in the data reflect a combination of global cooling and ice growth since the middle Miocene Climate Optimum at 15 Ma. Red dashed lines show the seven time slices selected for this study.

Proxy data

We are using planktonic foraminifera (calcifying unicellular free-floating protists) as tracers for the biogeochemical cycling of carbon in the upper water column.

We are measuring oxygen and carbon stable isotopes on a range of (living and extinct) depth-stratified, size-constrained planktonic foraminiferal species (e.g. Fig. 4) to reconstruct past water column biogeochemistry and plankton ecology.

Images of a range of (living and extinct) depth-stratified, size-constrained planktonic foraminiferal species
Fig. 4: Middle Miocene planktonic foraminiferal species from two of the studied sites, ODP Site 1138 (Kerguelen Plateau) (top) and DSDP Site 516 (South-West Atlantic) (bottom). (Image: Boscolo Galazzo F.)

Foraminifera are picked from seven time slices spanning from the middle Miocene (15 Ma) up to the Pleistocene, and from ocean sediment samples from a range of latitudinally distributed Integrated Ocean Drilling Program (IODP), Ocean Drilling Project (ODP) and Deep Sea Drilling Project (DSDP) sites (Fig. 5).

Map of DSDP, ODP, IODP sites considered in this study
Fig. 5: Map of Deep Sea Drilling Project (DSDP), Ocean Drilling Project (ODP) and Integrated Ocean Drilling Program (IODP) sites considered in this study.

Surface to depth δ13C and δ18O gradients will provide information about changes in the upper water column cycling of carbon and the evolution of deep niche habitats in planktonic foraminifera. Stable isotopes will be associated with planktonic foraminiferal assemblage data, to monitor changes in faunal diversity through time, and trace element ratios (in partnership with University College London).


We are using and developing an Earth system model incorporating temperature-dependencies in key ocean carbon cycle components to help interpret our data-based results and explore the relationship between water column temperature and carbon cycling.

By combining data and models, we will assess the consequences of temperature-dependent metabolism for past and future carbon cycling.

In order to model the Earth system, there is a balance to be had between the complexity of processes that are represented and the integration level of these processes. Simplified models and Earth system models of intermediate complexity (EMICs) are well suited to paleoclimate studies. We are able to run long, and transient simulations on reasonable timescales such that hypotheses for new processes can be tested.

CGENIE intermediate modelling

In this project we use the cGENIE intermediate complexity Earth system model, (Fig. 6), a 3D dynamic ocean coupled to a 2D atmosphere. The ocean includes a biogeochemical model of marine biota that simulates export production (“Biogem”). More recently, a plankton community model has been coupled to cGENIE (“Ecogem”), and we will use this to consider questions about plankton community changes.

Schematic representation of the cGENIE earth system model
Fig. 6: Schematic representation of the cGENIE earth system model. (Image: (

To simulate the Miocene we are creating a new “Miocene World” for cGENIE. Locations of continents, ocean bathymetry and ocean circulation (that are different to the present day, Fig. 7) need to be validated before we can consider the carbon cycle.

Diagram of global continental and bathymetric configurations for the Miocene and modern world
Fig. 7: Global continental and bathymetric configurations for the Miocene world (top) and the Modern world (bottom). (Image: PALEOMAP project

Measurement data modelling

In the present-day, we have access to much ocean-based data, such as temperature, carbon-13 measurements, pH etc. This provides a means of validating our treatment of processes in the model for the present-day simulation. Paleo-data that reconstructs past climate conditions provides a second means of testing the model for paleo-simulations. At the same time, modelling studies can help to interpret changes seen in the data. Ultimately we can use the model to project possible future changes in the ocean carbon cycle, and to understand past changes.

We can add (and alter the treatment of) processes in the model. In this project, we are interested in metabolism and the system’s response to changes in temperature. The processes of interest include photosynthesis; respiration; remineralisation rates; and dissolved organic carbon cycling.

For the carbon cycle, we are interested in the flux of particulate organic carbon (POC flux) which tells us how much carbon is being transported to the deep from the surface. As an example, temperature dependent remineralisation and uptake results in more of the Particulate Organic Carbon (POC) that is exported at the surface reaching the deep (Fig. 8) in colder regions (and vice versa). In a globally warmer world this may mean that the ocean biological pump becomes less efficient.

Diagram of cGENIE model output for the present-day world
Fig. 8: cGENIE model output for the present-day world. POC flux at (top) 40m depth (export production), and (bottom) at 1040m depth (“sequestration” depth). This simulation includes temperature dependence of nutrient uptake at the surface and nutrient remineralisation through the water column.


U.S. DOE. 2008. Carbon Cycling and Biosequestration: Report from the March 2008 Workshop, DOE/SC-108, U.S. Department of Energy Office of Science. (p. 81) (website)

Zachos, J. C., Dickens, G. R., & Zeebe, R. E. (2008). An early Cenozoic perspective on greenhouse warming and carbon-cycle dynamics. Nature, 451(7176), 279-283.

Olivarez Lyle, A., & Lyle, M. W. (2006). Missing organic carbon in Eocene marine sediments: Is metabolism the biological feedback that maintains end‐member climates? Paleoceanography, 21(2).

F. Boscolo-Galazzo, K. A. Crichton, S. Barker, P. N. Pearson, Temperature dependency of metabolic rates in the upper ocean: A positive feedback to global climate change? Global. Planet. Change 170, 201–212 (2018).

Under review

Boscolo-Galazzo, F., Crichton, K. A., Ridgwell, A., Mawbey, E. M., Wade, B. S., Pearson, P. N.: Temperature controls carbon cycling and biological evolution in the ocean twilight zone, Science, in review, 2020.

Crichton, K. A., Ridgwell, A., Lunt, D. J., Farnsworth, A., and Pearson, P. N.: Data-constrained assessment of ocean circulation changes since the middle Miocene in an Earth system model, Clim. Past Discuss, in review, 2020.

Crichton, K. A., Wilson, J. D., Ridgwell, A., and Pearson, P. N.: Calibration of key temperature-dependent ocean microbial processes in the cGENIE.muffin Earth system model, Geosci. Model Dev. Discuss, in review, 2020.


This project is funded by the Natural Environment Research Council (NERC).

The project team

Principal Investigators

Professor Paul Pearson

Professor Paul Pearson

Honorary Professor

+44 (0)29 2087 4579
Professor Stephen Barker

Professor Stephen Barker

Professor in Earth Science

+44 (0)29 2087 4328

Postdoctoral researchers

  • Dr Flavia Boscolo Galazzo, Department of Earth Science Department, University of Bergen.
  • Dr Katherine Crichton, Postdoctoral research associate, Geography department, Exeter University.
  • Dr Elaine Mawbey, Postdoctoral research associate, British antarctic survey.