A Phosphorus Control on Oceanic Anoxic Events under Greenhouse Climates?

This Project has been filled

Project Background and Rationale:

Episodes of widespread marine oxygen depletion and enhanced organic carbon (OC) burial, which are often associated with mass extinctions, are a common feature of the geological record, as well as being a threat to the modern marine ecosystem as ‘oceanic dead zones’ progressively spread in response to global warming. These events commonly lead to the formation of extensive marine black shales (Figure 1), but the combination of factors that contribute to the initiation, maintainance, and cessation of oceanic anoxic events (OAEs) are poorly understood.

In many cases, enhanced OC burial was likely driven by increased primary productivity stimulated by enhanced availability of nutrients. This drove increased OC respiration, thus forcing water column de-oxygenation. As the ultimate limiting nutrient on geological timescales, the oceanic influx of phosphorus (P) from both terrestrial weathering and recycling from marine sediments is thought to play a central role in initiating and maintaining ocean-wide anoxia. However, the relative importance of P inputs from continental weathering versus internal oceanic recycling, and links to other key elements such as iron (Fe) and sulphur (S), before and during OAEs, remains unclear.

Figure 1. Cretaceous black shales from Wyoming, USA.

In particular, the nature and dynamics of P recycling in the ocean are poorly understood, but must be tightly coupled to redox conditions in the water column and sediments. For example, most bioavailable P is trapped in sediments deposited under oxic conditions, but under anoxic-euxinic (free dissolved H2S) water column conditions, P may be recycled back to surface waters by release from organic matter and Fe oxide minerals, thus stimulating further productivity. However, a third dominant redox state of the ocean – anoxic Fe-rich (ferruginous) conditions – has become recognised over the last few years. Under this little-studied redox state, P may either be trapped in the sediment via uptake by Fe minerals and organic matter, or may be recycled back to the water column, dependent on the chemistry of sediment pore waters during early diagenesis. The spatio-temporal dynamics of these contrasting redox conditions across the shelf and deeper ocean likely control the extent, timing, and persistence of widespread oxygen depletion in the ocean, but have never been studied in detail, and will thus be a key focus of this project.

Aims, Objectives and Key Hypotheses:

This project will test the hypothesis that the continental weathering supply of P to the open ocean is modulated by recycling in shelf seas, with the processing of P on the shelf essentially acting as a ‘gatekeeper’ of P supply to the open ocean (Figure 2). The precise conditions on the shelf controls the behaviour of the P cycle, and hence the spread and intensity of detrimental water column conditions.


Figure 2. Conceptual model of the interaction between weathering, productivity and redox in the ocean. Dashed P-fluxes are dependent on redox state as explained in the cartoon to the right. The sulphate:FeHR (highly reactive Fe) ratio in the ocean likely exerts a control on whether the anoxic ocean becomes euxinic or ferruginous.

Specifically, the project will produce high resolution, multi-proxy records targeting well-preserved, strategically located marine cores from shelf and deeper margin/basin sites, covering two contrasting greenhouse scenarios of differing intensity, to test orbital effects on:

  1. Redox state: A state-of-the-art multiproxy approach will be used to distinguish between key water column redox states linked to potential perturbations of the Fe and S cycles.
  2. Weathering intensity: Elemental ratios will be applied to investigate whether changes in terrestrial weathering intensity drove changes in ocean redox.
  3. Response of the P cycle: P speciation will be utilized in order to understand the ocean biogeochemical response to changes in weathering and ocean redox.

These novel, multi-proxy records will be evaluated via biogeochemical modelling to unravel feedbacks between weathering, redox and P cycling on the shelf and in the deep basins. We anticipate that this combined approach will lead to a step-change in our understanding of some of the most dramatic perturbations to the Earth system of the last 120 million years. This, in turn, will inform on generic processes that likely operated during other episodes of ocean anoxia under past, and potentially future, greenhouse conditions.



Five OAE intervals will be studied (Table 1), each consisting of ~200 samples at 2 cm resolution. When combined with existing data for the onset of OAE 2 at Tarfaya (Poulton et al., 2015), and for OAE 3 at Demerara Rise (März et al., 2008), this will result in comprehensive data sets covering a representative range of shallow and deep water environments at the onset and termination of two contrasting OAEs. Demerara Rise captures nutrient cycling in the deep tropical proto-North Atlantic, while two rather different shelf environments are represented at Tarfaya (adjacent to the open Atlantic) and the Western Interior Seaway (WIS) (extensive epicontinental shelf sea).

Table 1. Core locations, environment and target intervals, including the two existing sites (*) for which redox and P data are available for integration into the current study.

Site Location Environment Target interval
Tarfaya (MPL) N. Atlantic Inner shelf 4 m, onset of OAE 2
Tarfaya (S 57)* N. Atlantic Central shelf 4 m, onset of OAE 2
Demerara Rise (ODP 1258)

Tarfaya (S 57)

Niobrara Fm. (USGS #1 Portland)

N. Atlantic

N. Atlantic


Deep ocean

Central shelf

Epicont. shelf

4 m, onset of OAE 2

4 m, end of OAE 2

4 m, onset of OAE 3

Demerara Rise (ODP 1261)* Equat. Atlantic Deep ocean 1.2 m, full OAE 3


Geochemistry and modelling:

Organic and carbonate C contents (Leco C-S analyser), as well as OC isotopes (to identify the onset and termination of each OAE) will be determined for all samples at. Water column redox will be obtained for all samples using Fe speciation (Poulton and Canfield, 2005) to distinguish euxinic, ferruginous, and oxic water column conditions. This well-calibrated technique (Poulton and Canfield, 2011) quantifies an Fe pool that is highly reactive towards dissolved sulfide (FeHR). Marine sediments deposited under oxic conditions are characterized by FeHR/Fetotal ratios <0.22, whereas ratios >0.38 provide evidence for anoxic conditions. For anoxic samples, the ratio of pyrite Fe to highly reactive Fe (Fepy/FeHR) distinguishes euxinic from ferruginous conditions, whereby ratios >0.7-0.8 indicate euxinia. Trace (ICP-MS) and major elements (ICP-OES) will be determined on all samples after dissolution with HF-HNO3-HClO4. Trace metal ratios (e.g., Mo/Al; U/Al; V/Al; Re/Mo) will be used to provide additional detail on redox state (e.g., strongly euxinic, weakly euxinic, suboxic (e.g., Kendall et al., 2010).

The speciation of P provides critical information on P fluxes and redox-driven recycling (e.g., März et al., 2008). We will apply a revised P speciation technique to all samples (Thompson et al., 2019), which gives Fe-bound P (PFe; P associated with Fe oxides or Fe(II) phosphates), organic P (Porg), authigenic P (Paut; dominantly carbonate fluorapatite) and detrital P (Pdet). The P that was potentially reactive (Preact) is defined as PFe + Porg + Paut. This partitioning provides unprecedented insight into P cycling: P/Al ratios can be compared to average shale to evaluate burial fluxes, while PFe/Ptotal allows identification of P draw-down via Fe minerals under ferruginous conditions. Similarly, Corg/Porg and Corg/Preact ratios inform on repartitioning between phases during deposition and diagenesis, and evaluates the extent of recycling to the water column (e.g., Xiong et al., 2019).

These new data will form the basis for updating an existing box model of P and O2 dynamics across the shelf, open surface ocean and deep ocean, which has previously been applied to OAEs (Tsandev and Slomp, 2009). The model will be used to provide a global context to the geochemical data, to provide a test of the processes that drove, maintained and terminated anoxic conditions in the ocean.


The student will receive training in a wide variety of state-of-the-art experimental and sedimentary geochemical and mineralogical techniques, including techniques that the project supervisors have been personally responsible for developing. In addition, the student will be trained in a wide variety of key transferable skills within the Faculty Graduate School.

Opportunity for Travel:

The student will have an opportunity to collect well preserved drill core samples from collections based in Europe and the US. The student will also be encouraged to present their research at national and international conferences in Europe and North America (for example, the International V.M. Goldschmidt Conference).

References and Further Reading (copies available on request):

Kendall B, Reinhard CT, Lyons TW, Kaufman AJ, Poulton SW, Anbar AD (2010) Pervasive oxygenation along late Archaean ocean margins, Nature Geoscience, 3, 647-652.

März C, Poulton SW, Beckmann B, Küster K, Wagner T, Kasten S (2008) Redox sensitivity of P cycling during marine black shale formation: Dynamics of sulfidic and anoxic, non-sulfidic bottom waters, Geochim. Cosmochim. Acta, 72, 3703-3717.

Poulton SW, Canfield DE (2005) Development of a sequential extraction procedure for iron: Implications for iron partitioning in continentally derived particulates, Chem. Geol., 214, 209-221.

Poulton SW, Canfeld DE (2011) Ferruginous conditions: A dominant feature of the ocean through Earth’s history, Elements, 7, 107-112.

Poulton SW, Henkel S, März C, Urquhart H, Flögel S, Kasten S, Sinninghe Damsté JS, Wagner T (2015) A continental weathering control on orbitally-driven redox-nutrient cycling during Cretaceous Oceanic Anoxic Event 2, Geology, 43, 963-966.

Thompson J, Poulton SW, Guilbaud R, Doyle KA, Reid S, Krom MD (2009) Development of the SEDEX phosphorus speciation method for ancient rocks and modern iron-rich sediments, Chem. Geol., 524, 383-393.

Tsandev I and Slomp CP (2009), Modeling phosphorous cycling and carbon burial during Cretaceous oceanic anoxic events, Earth Planet. Sci. Lett. 286 ,71–79.

Xiong Y, Guilbaud R, Peacock CL, Cox RP, Canfield DE, Krom MD, Poulton SW. 2019. Phosphorus cycling in Lake Cadagno, Switzerland: A low sulfate euxinic ocean analogue. Geochim. Cosmochim. Acta. 251, 116-135