The Paleocene Eocene Thermal Maximum

Background

Of particular interest for evaluating our theoretical understanding of the complex processes that govern the carbon cycle and for testing the sensitivity of coupled climate/biogeochemical models to extreme forcing, is a transient global warming event characterized by a carbon cycle perturbation at the Paleocene-Eocene boundary (55 Mya), referred to as the Paleocene-Eocene Thermal Maximum (PETM). This warming event is regarded as a suitable analog for future climate change and uptake of carbon in the ocean. The PETM was associated with the release of more than 2000 PgC over a period significantly shorter than the residence time of carbon in the ocean (<10 ky; Zachos et al., 2005; Zachos et al., 2007; Zachos et al., 2008). Evidence for the massive release of carbon includes a >3.0‰ negative carbon isotope excursion as recorded in marine and terrestrial fossils (e.g. Koch et al., 1992; Kelly et al., 1996) temporally coupled with a worldwide seafloor carbonate dissolution horizon (e.g., Bralower et al., 1997; Lu et al., 1998; Schmitz et al., 1996; Thomas et al., 1999) as well as shoaling of the lysocline and carbonate compensation depth (CCD), by as much as 2 km in some regions. The climatic impacts of this greenhouse forcing were significant. Sea-surface temperatures (SST) increased by 5°C in the tropics (Tripati and Elderfield, 2004; Zachos et al., 2003), by 6-8°C in the Arctic and sub-Antarctic (Kennett and Stott, 1991; Thomas et al., 1999; Sluijs et al., 2006), and by 4-6°C throughout the deep sea (Tripati and Elderfield, 2005), relative to Paleocene temperatures (see Figure 1). Global humidity and precipitation patterns changed as well (Pagani et al., 2006; Robert and Kennett, 1994; Wing et al., 2005; Brinkhuis et al., 2006; Sluijs et al., 2008), as did deep-sea circulation patterns (Nunes and Norris, 2006; Thomas et al., 2003).

Figure 1: Geographical reconstruction for the PETM from the PALEOMAP Project (www.scotese.com) . Boxes indicate reconstructed surface temperature anomalies for the PETM relative to Paleocene background temperatures based on oxygen isotopes, Mg/Ca ratios and TEX86 (compiled by Appy Sluijs)
Figure 1: Geographical reconstruction for the PETM from the PALEOMAP Project (www.scotese.com) . Boxes indicate reconstructed surface temperature anomalies for the PETM relative to Paleocene background temperatures based on oxygen isotopes, Mg/Ca ratios and TEX86 (compiled by Appy Sluijs)

A second hyperthermal, (roughly 2 m.y. post-PETM, Zachos et al., 2004; Lourens et al., 2005; Westerhold et al., 2005) exhibits many of the characteristics of the PETM including a transient warming, negative carbon isotope excursion, and dissolution horizons. The magnitude of each, however, appears to be half that of the PETM. Nicolo et al. (2007) described a series of four carbonate isotope excursion events in New Zealand during the 2 million years following the PETM, suggesting a mechanism that repeatedly led to the release of large amounts of carbon.

Figure 2. Changes in Earth's temperature and climate across many different time scales (from Zachos et al. [2001]).
Figure 2. Changes in Earth’s temperature and climate across many different time scales (from Zachos et al. [2001]).
The world looked different 55 million years ago. Africa and South America were in closer proximity than today, making the basin of the Atlantic Ocean smaller. An ancient ocean, called the Tethys, separated Europe/Asia from North Africa and the global climate was also much different.

The Paleocene-Eocene Thermal Maximum (PETM) that occurred then was an extreme warm period in Earth’s history. On geologic time scales, the onset of the event was extremely rapid (Figure 1). Earth’s surface temperature rose globally by 5-9 degrees Celsius within a few thousand years, while simultaneously, large amounts of carbon were released into the ocean-atmosphere system.

At present, humans are adding large amounts of greenhouse gases to the atmosphere, which leads to warming of the planet’s surface. The PETM may turn out to be a valuable lesson from the past on climate change and carbon cycling, which helps us to understand the present and develop better predictions for the future.

It has been suggested that catastrophic dissociation of oceanic methane hydrate was responsible for the massive carbon release. We are currently investigating scenarios of carbon release during the PETM using sediment and carbon cycle models. Funding for this project is provided by the National Science Foundation (see Active Research Grants, NSF: EAR0803979).

Objectives and Strategy

The main questions to be addressed in this project are:

  1. What were the mass and rate of the original carbon release? Was the release instantaneous or pulsed?
  2. How did the magnitude and rate of carbon release affect the scale of the changes in the lysocline and CaCO3 accumulation rates?
  3. What were the rates of carbon sequestration and recovery of the carbon cycle and what biogeochemical feedbacks came into play?

Model sensitivity tests are required to simulate the time scale of carbon cycle recovery from rapid injections of CO2. One component of this recovery is weathering of carbonate (short-term) and silicate (long-term) rocks on land. We are carrying out a series of time-slice simulations using the comprehensive climate model CCSM-3 (see below; details in Collins et al., 2006) to predict the oceanic and sedimentary response to a range of carbon addition scenarios. We are assessing which of the scenarios (e.g., clathrate or thermogenic methane vs. fossil carbon – fast vs. slow CO2 input) generate the proper isotopic, climatic, and CCD responses (both regional and global) for the onset, magnitude, and recovery from these events. CCSM-3 will provide improved simulations of ocean biogeochemistry for the initiation and termination of the hyperthermals. The output of these simulations is being used for model-data validation and input for offline three-dimensional carbon cycle modeling studies as well as for process-oriented models in order to investigate effects of different ocean circulation patterns on water column processes, fluxes at the sediment/water interface, and sedimentary composition. The role of geography is being investigated by comparison of PETM results from experiments using different paleogeographic settings with a modern reference run. Sensitivity to changes in the hydrological cycle will be tested, following the work of Bice and Marotzke (2001) that demonstrated a high degree of sensitivity of the ocean circulation to the amounts and location of freshwater discharge to the ocean. Possible switches in the thermohaline circulation related to changes in atmospheric moisture transport could lead to sufficient warming to destabilize seafloor gas hydrates over most of the world ocean to a water depth of at least 1900 m.

The strategy involves integration of the observational database with numerical model results.

Simulations

In March 2007, NCAR CISL (“University Requests for Large Computational Project Support”) granted computing resources of 45,000 GAUs. These resources are being used to carry out the following simulations

  1. Modern climate simulations Ocean model POP (with gx3v5 resolution) coupled on-line with ocean carbon cycle model OCMIP (Doney et al., 2006) forced with heat, freshwater, and momentum fluxes from a validated CCSM-3 experiment (Yeager et al., 2006).

    1.1. Reference run including silica cycle (integration 1000 yrs)
    1.2. Same as 1.1. but with co-limitation of phosphate and iron on silicate production
    1.3. Same as 1.1. but with feedback of the silica cycle on the carbon cycle
    1.4. Same as 1.1. but with sediment model Muds (Archer et al., 2002) and with integration for 10,000 years

  2. PETM climate simulations CCSM-3 (with T31 resolution atmosphere, and gx3v5 resolution in the ocean) coupled on-line with OCMIP and sediment model Muds. An improved marginal sea parameterization and improved geography for the Arctic Ocean is applied.

    2.1. PETM simulations to explore role of geography:
    2.1.1. 8xCO2 with freshwater exchange to the Indian Ocean (1000 yrs integration)
    2.1.2. 8xCO2 with freshwater exchange to the Indian and Atlantic Ocean (500 yrs integration)
    2.1.3. 8x CO2 with freshwater exchange to the Indian and Pacific Ocean (500 yrs integration)
    2.1.4. Same as 2.1.1 but with 1000 m deeper sill at the Panama Isthmus (500 yrs integration)2.2. PETM simulations to explore greenhouse gas forcing and carbon cycle interactions:
    2.2.1. 4xCO2 with freshwater exchange to the Indian Ocean (500 yrs integration)
    2.2.2. Same as 2.1.1. but with 8xCO2 and 8xCH4 (500 yrs integration)

The modeling work is being coordinated with the CCSM Paleoclimate Working Group and builds upon previous simulations with older versions of the CCSM (e.g. Huber et al., 2002; Huber and Caballero, 2003; Huber and Sloan, 2001; Shellito et al., 2003). During the last year, the carbon cycle model Ocean Carbon Model Intercomparison Project (OCMIP) with iron chemistry (Doney et al. (2006) has been coupled on-line to the T31_gx3v5 CCSM-3 for the PETM simulation. This model is currently being extended by a silica cycle using parameterizations similar to Maier-Reimer (1993), Gnanadesikan (1999), Heinze et al. (1999), and Archer et al. (2002). The sensitivity of the opal productivity to co-limitation of phosphate and iron (Yool and Tyrrell, 2005) will be explored in order to better understand changes in the opal deposition during the Eocene. Furthermore, the sedimentary diagenesis model Muds (Archer et al., 2002) has been set up for running on our system. A modern control run with Muds is currently being carried out, and sensitivity experiments with increased temperatures similar to the PETM, increased carbon storage, and increased weathering will be performed later this year.

In Summer 2008, Dr. A. Winguth visited NCAR to improve the PETM CCSM-3 simulation in collaboration with Dr. Shields, Dr. Yeager, and Dr. Kiehl, using the marginal sea parameterization. The Arctic model bathymetry has been improved in order to balance the Arctic freshwater budget for long-term integrations.

The set-up of the modern reference simulation with POP in order to test various geochemical parameterizations, branching out from an existing fully coupled run, has been completed. Work included changes in the data sea-ice component of the branch-out run as well as updates and corrections of forcing files for the active ocean component. The modern reference simulation (experiment 1.1) with POP coupled on-line with OCMIP, including iron and silica cycle, is currently being carried out.

Findings

Modern control experiments

Results for the thermohaline circulation of the model run with the active ocean component POP and data components for atmosphere, land, and sea ice are in good agreement with the fully coupled CCSM3(T31_gx3v5) run (Figure 3). The model also represents well nutrient distributions in the ocean; an example is given in Figure 4 (phosphate concentrations at the surface).

Figure 3. Maximum global overturning circulation and circulation pattern for the fully coupled CCSM3 (on the left) and for the POP only run with prescribed forcing boundary conditions (heat, freshwater and momentum fluxes) (on the right). NH = Northern Hemisphere, SH = Southern Hemisphere.
Figure 3. Maximum global overturning circulation and circulation pattern for the fully coupled CCSM3 (on the left) and for the POP only run with prescribed forcing boundary conditions (heat, freshwater and momentum fluxes) (on the right). NH = Northern Hemisphere, SH = Southern Hemisphere.
Figure 4: Comparison between the model-predicted (left) and the World Ocean Atlas 2005 (Garcia et al., 2006; right) phosphate concentration at the surface of the ocean.
Figure 4: Comparison between the model-predicted (left) and the World Ocean Atlas 2005 (Garcia et al., 2006; right) phosphate concentration at the surface of the ocean.

 

Improvement of geochemical tracers

A silica cycle has been implemented into the OCMIP version of Doney et al. (2006) for paleoclimatic simulations (M.S. Student M. Franklin). In the modern ocean, opal production dominates the calcareous production in large areas. For the PETM, significant siliceous deposits have been recovered near New Zealand (e.g. Hollis et al., 2005) and in the North Sea (e.g. Gradstein et al., 1994), but opal producers were not as dominant as in the modern ocean. The model is initialized with observed annual mean silicate concentrations from the World Ocean Atlas 2005 (Garcia et al., 2006). The silicate uptake JSi in the euphotic zone z < zcomp is taken from:

JSi= FI · FN · max(1,zvol/zcomp)/τ        (1)

with the light limitation term FI from Doney et al. (2006). FN is a Michaelis-Menton Kinetics limiting term:

FN = SiO4/(SiO4 + KSi)

with a half-saturation constant Si = 8.2 nmol L-1 (Heinze et al., 1999). The dissolution of opal in subsurface water masses is temperature-dependent and taken from Gnanadesikan (1999). Preliminary results indicate that the prediction of the opal production is in reasonable agreement with the observational evidence and previous model studies (Heinze et al., 2003; Figure 9).

Figure 5: Implemented silica cycle into POP/OCMIP for paleoclimatic studies. Shown here is the export production of opal in mol m-2 yr-1 using eq. (1). For comparison with coarse -resolution models see Fig. 2a in Heinze et al. (2003)
Figure 5: Implemented silica cycle into POP/OCMIP for paleoclimatic studies. Shown here is the export production of opal in mol m-2 yr-1 using eq. (1). For comparison with coarse -resolution models see Fig. 2a in Heinze et al. (2003)

PETM Experiments

The Paleocene-Eocene CCSM experiment with 8xCO2 has been integrated for 670 years including the OCMIP-type carbon cycle. Surface temperatures indicate a much lower pole-to-equator temperature gradient than in the modern climate with moderate sea surface temperatures of 10 to 16°C in high latitudes and 30 to 34°C in low latitudes (Figure 8). Simulated temperatures are underestimated for the tropics (~5°C) and the high latitudes (~8°C) (compare to Figure 1, “Research and Education Activities”), therefore we have designed a future experiment with higher greenhouse gas levels (8xCH4) and improved polar cloud parameterization in collaboration with NCAR.

Figure 6: Sea surface temperature and horizontal velocities from the CCSM3 8xCO2 PETM experiment in °C.
Figure 6: Sea surface temperature and horizontal velocities from the CCSM3 8xCO2 PETM experiment in °C.

The meridional overturning circulation at the PETM appears more symmetric than the modern one due to deep water formation in the North Pacific and Southern Ocean (Figure 9, top). The Southern Ocean circulation cell is substantially reduced. Moreover, the juvenile Atlantic limits penetration of water masses from the Southern Ocean. For the PETM simulation, global poleward heat transport is enhanced in the southern hemisphere mid-latitudes because of a narrow juvenile Atlantic during the PETM (Figure 9, bottom). In contrast, in the modern South Atlantic the heat transport is directed equatorward.

Figure 7: Global meridional overturning circulation in Sv for the 8xCO2 PETM simulation (upper left) and for the modern (upper right). Poleward heat transport is shown in the lower part of the figure (PETM: blue solid; modern: black dashed).
Figure 7: Global meridional overturning circulation in Sv for the 8xCO2 PETM simulation (upper left) and for the modern (upper right). Poleward heat transport is shown in the lower part of the figure (PETM: blue solid; modern: black dashed).

Distribution of biogeochemical tracers reflects changes in the circulation and biogeochemical sources and sinks. Export production is highly correlated with Ekman-induced upwelling and with convective overturning of nutrient-rich water masses (Figure 10). Similar to the modern equatorial Pacific but with a larger east-west extent, a strong equatorial upwelling in the Eocene tropical Pacific enriches nutrients and stimulates marine productivity in the euphotic zone. In the high latitudes, a deep mixed-layer in the western Eocene North Pacific and around East Australia leads to strong marine productivity. The simulated δ13C gradients in the Atlantic (Figure 11) correlate well with the CIE reconstructions from Nunes and Norris (2006). Nunes and Norris (2006) inferred from stable carbon and oxygen isotope data that the rapid rise of global temperatures during the PETM may have led to drastic changes in the ocean chemistry and circulation. The east-to-west gradient in the central Pacific is in reasonable agreement with the reconstruction from the sedimentary record.

Figure 8: Left: Export production of particulate organic matter in moles m-2 yr-1. Right: Vertical (contours 10-4 cm s-1) and horizontal (vector, cm) velocities at the base of the euphotic zone.
Figure 8: Left: Export production of particulate organic matter in moles m-2 yr-1. Right: Vertical (contours 10-4 cm s-1) and horizontal (vector, cm) velocities at the base of the euphotic zone.
Figure 9: Deep-sea δ13C in 1000 m depth inferred from phosphate concentration using the modern deep-sea relationship of δ13C=2.7-1.1 PO4 (Broecker and Maier-Reimer, 1992). The north-to-south gradients in the Atlantic are about 0.8‰, which matches reasonably well the gradient of 0.7‰ of Nunes and Norris (2006) for the carbon isotope excursion (CIE) at the PETM.
Figure 9: Deep-sea δ13C in 1000 m depth inferred from phosphate concentration using the modern deep-sea relationship of δ13C=2.7-1.1 PO4 (Broecker and Maier-Reimer, 1992). The north-to-south gradients in the Atlantic are about 0.8‰, which matches reasonably well the gradient of 0.7‰ of Nunes and Norris (2006) for the carbon isotope excursion (CIE) at the PETM.

The following main conclusions can be drawn from this study:

  1. The simulated Atlantic deep-sea circulation indicates a southward transport into the Southern Ocean in agreement with paleoproxies.
  2. Deepwater formation in the North Pacific and Southern Ocean is associated with high productivity.
  3. Simulations with even higher greenhouse gas concentrations are required to yield a better match with the reconstructed temperatures for the PETM.

We are planning to carry out additional experiments with lower greenhouse gas concentrations (4xCO2) to investigate the response of the geochemistry to changes in greenhouse warming and circulation.

Project Training and Development

Morgan Franklin
Morgan Franklin has been trained in setting up and running CCSM. She has acquired skills in Fortran programming and has enhanced her academic abilities through literature search and studies. She has familiarized herself with the silica cycle and possibilities of its parameterization.

Dierk Polzin
Dierk Polzin has explored new graphics programs and has enhanced his skills in input data preparation for the CCSM.

Cornelia Winguth
Cornelia Winguth has been trained in setting up and running CCSM and in setting up and running the Muds model. She has adapted and applied NCAR plot routines in order to visualize experiment results. She has enhanced her Fortran programming skills and acquired more knowledge on the carbon cycle and its feedback mechanisms. Insights gained from this work are used in the undergraduate classes taught by Cornelia Winguth at the University of Texas at Arlington.

Jesse Cope
Jesse Cope has been trained in setting up and running CCSM. He has acquired skills in Fortran programming and has enhanced his academic abilities through literature search and studies. He has familiarized himself with various possible paleogeographic settings at the time of the PETM and their possible impact on ocean circulation.

Teresa Sykes
Teresa Sykes has been trained in setting up and running the Muds model. She has acquired skills in Fortran programming and has enhanced her academic abilities through literature search and studies. She has familiarized herself with principles of CCD/lysocline changes and weathering.

Vinit Asher
Vinit Asher has been trained in processing CCSM simulation results with graphical software (e.g. NCL graphics and NetCDF), in programming in UNIX and Fortran, in data management, in running Muds, and has enhanced his skills in website design.

Ashley Wright
Ashley Wright has enhanced her skills in literature research and in graphical software package applications (such as Adobe).

Contributions Within Discipline

The ocean is the largest sink for anthropogenic CO2 and has absorbed nearly 130 PgC of the 380 PgC emitted to the atmosphere since the onset of the Industrial Revolution (Sabine et al., 2004; Feely et al., 2004). If emissions continue to rise unabated for the next three centuries, an additional 4000 PgC or more will be input to the atmosphere and ocean. As a consequence, significant climatic changes might occur and the surface ocean may undergo acidification to the extent that corals and other calcifying organisms will be unable to precipitate their skeletons (Zachos et al., 2008).

Ancient global warming events provide a unique opportunity to gain insight into the long-term impacts of rapidly rising CO2 levels on modern climate, ocean carbonate chemistry, and biota. They will also make it possible to identify potential non-linear feedbacks, and test climate and biogeochemical model sensitivity. Such information is essential to providing scientific leaders and policy makers with a better sense of the consequences of unabated anthropogenic CO2 emissions for global climate, ocean carbon chemistry and marine food chains. The early Eocene hyperthermals represent natural carbon cycle experiments involving rapid input of carbon to the atmosphere/ocean on scales comparable to the modern anthropogenic forcing.

We are carrying out a series of time-slice simulations using the comprehensive climate model CCSM-3 including a carbon cycle model as well as a sedimentary diagenesis model in order to predict the oceanic and sedimentary response to a range of carbon addition scenarios. Our work includes simulations for the Paleocene-Eocene Thermal Maximum (PETM; 55 Mya) as well as the first quantitative analysis of the ELMO event (53 Mya). We will assess which of the different possible scenarios (e.g., clathrate or thermogenic methane vs. fossil carbon – fast vs. slow CO2 input) generate the proper isotopic, climatic, and CCD responses (both regional and global) for the onset, magnitude, and recovery from these events.

Contributions to Human Resurce Development

The project provides research opportunities for graduate students Morgan Franklin, Jesse Cope, Teresa Sykes, and Vinit Asher who are gaining skills in the application of a global climate system model, of a global sediment diagenesis model, in model development, and in climate research. These skills will be beneficial for their further career in geosciences.

Cornelia Winguth has obtained skills in the application of a global climate system model and in model development. She will apply these skills in future projects. The knowledge gained about the PETM hyperthermals and about the global carbon cycle will be beneficial for her teaching.

Outreach Activities

Arne Winguth has presented research results at UNT Health Science Center, Fort Worth, during the National Public Health Week on April 7, 2008.

Cornelia Winguth has taught geology to first graders at Little Elementary School, Arlington, Texas.

Personnel

Dr. Arne Winguth, PhD – Principal Investigator @ UTA

Dr. Cornelia Winguth, PhD – Faculty Research Associate @ UTA

Morgan Franklin, B.S., Graduate Student – Silica Cycle @ UW-Madison

Jesse Cope, B.S., Graduate Student – Bathymetry @ UTA

Ashley Wright, Undergraduate Student – Oxygen Isotopes @ UTA

Vinit Asher, B.S., Graduate Student – Technical Assistant & Web Content Developer @ UTA

Organizational Partners

University of California-Santa Cruz
Lead institution for collaborative project, observational group (J. Zachos).

The Pennsylvania State University
Collaborating institution, modeling and observational group (T. Bralower and L.R. Kump).

University of Hawaii
Collaborating institution, modeling group (R.E. Zeebe).

Purdue University
Collaborating institution, modeling group (G.J. Bowen).

Williams College
Collaborating institution, observational group (H.M. Stoll).

Yale University
Collaborating institution, observational group (M. Pagani).

California Institute of Technology
Collaborating institution, observational group (K.A. Farley).

National Center For Atmospheric Research
Use of NCAR computing resources, collaboration on CCSM experiments and model development (J. Kiehl, C. Shields, K. Lindsay, S. Yaeger, B. Kauffman, N. Mahowald).

University of Chicago
Collaboration on Muds model development and implementation into CCSM (D. Archer).

University of Texas at Arlington
Collaboration on paleogeography for the PETM (C. Scotese).

University of Northern Colorado
Collaboration on CCSM experiments for the PETM (L. Shellito).

Meetings and Presentations

Modeling strategies and first results have been presented at the following meetings:

  • NCAR Workshop on PETM Data-Model Integration in Santa Fe, New Mexico, May 31-June 01, 2007 by A. Winguth et al.: “Changes in the Marine Carbon Cycle During the PETM: A Model Study”.
  • 12th Annual CCSM Workshop in Breckenridge, Colorado, June 19-21, 2007, by A. Winguth et al.: “Dynamics of Carbon Release and Sequestration During two Early Eocene Hyperthermals”.
  • Project meeting “Santa Cruz PE Carbon Workshop”, December 7-8, 2007, by A. Winguth et al.: “Marine Productivity Changes t the PETM – A Model Study”.
  • AGU Fall Meeting in San Francisco, December 10-14, 2007, by C. Winguth et al.: “Modeling Biogeochemical Responses to Massive Carbon Releases at the Paleocene-Eocene Thermal Maximum”.
  • EGU General Assembly in Vienna, April 13-18, 2008, by A. Winguth et al.: “Modeling of Changes in Marine Carbon Uptake During Warm Climates”.
  • 13th Annual CCSM Workshop in Breckenridge, Colorado, June 17-19, 2008, by A. Winguth et al. (presented by T. Sykes): “Carbon Cycle Changes During the Paleocene-Eocene Thermal Maximum: A CCSM Sensitivity Study”.
  • In addition, A. Winguth convened the session “Warm Climates in the Past” (CL36) at the EGU General Assembly in Vienna, April 13-18, 2008.

Acknowledgements

This research at the University of Texas at Arlington is supported by the National Science Foundation(NSF EAR0803979).

References

Archer, D.E., J.L. Morford, and S.R. Emerson, 2002. A model of suboxic sedimentary diagenesis suitable for automatic tuning and gridded global domains. Global Biogeochem. Cycles 16,doi:10.1029/2000GB001288

Bice, K.L., and J. Marotzke, 2001. Numerical evidence against reversed thermohaline circulation in the warm Paleocene/Eocene ocean. J. Geophys. Res. 106 (11), 529– 11,542, 2001.

Bralower, T.J., D.J. Thomas, J.C. Zachos, M.M. Hirschmann, U. Rohl, H. Sigurdsson, E. Thomas, and D.L. Whitney, 1997. High-resolution records of the late Paleocene thermal maximum and circum-Caribbean volcanism: Is there a causal link? Geology 25, 963-966.

Brinkhuis, H., and 22 others, 2006. Episodic fresh surface waters in the Eocene Artic Ocean. Nature 441, 606-609. Collins, W.D., and 14 others, 2006. The Community Climate System Model Version 3 (CCSM3). J. Climate 19, 2122-2143.

Doney, S.C., K. Lindsay, I. Fung and J. John, 2006. Natural variability in a stable 1000 year coupled climate-carbon cycle simulation. J. Climate 19, 3033-3054.

Gnanadesikan, A., 1999. A global model of silicon cycling: Sensitivity to eddy parameterization and dissolution. Global Biogeochemical Cycles 13, 199-220.

Huber, M., and L.C. Sloan, 2001. Heat transport, deep waters, and thermal gradients: Coupled simulation of an Eocene greenhouse climate. Geophys. Res. Lett. 28, 3481-3484.

Huber, M., and R. Caballero, 2003. Eocene El Niño: Evidence for robust tropical dynamics in the “hothouse”. Science 299, 877-881.

Huber, B.T., R.D. Norris, and K.G. MacLeod, 2002. Deep-sea paleotemperature record of extreme warmth during the Cretaceous. Geology 30, 123-126.

Kelly, D.C., T.J. Bralower, J.C. Zachos, I.P. Silva, and E. Thomas, 1996. Rapid diversification of planktonic foraminifers in the tropical Pacific (ODP Site 865) during the Late Paleocene Thermal Maximum. Geology 24, 423-426.

Kennett, J.P., and L.D. Stott, 1991. Abrupt deep-sea warming, palaeoceanographic changes and benthic extinctions at the end of the Palaeocene. Nature 353, 225-229.

Koch, P.L., J.C. Zachos, and P.D. Gingerich, 1992. Correlation between isotope records in marine and continental carbon reservoirs near the Palaeocene-Eocene boundary. Nature 358, 319-322.

Lourens, L.J., A. Sluijs, D. Kroon, J.C. Zachos, E. Thomas, U. Rohl, J. Bowles, and I. Raffi, 2005. Astronomical pacing of late Palaeocene to early Eocene global warming events. Nature 435, 1083-1087.

Lu, G.Y., T. Adatte, G. Keller, and N. Ortiz, 1998. Abrupt climatic, oceanographic and ecologic changes near the Paleocene-Eocene transition in the deep Tethys basin: The Alademilla section, southern Spain. Eclogae Geologicae Helvetiae 91, 293-306.

Maier-Reimer, E., 1993. Geochemical cycles in an ocean general circulation model: Preindustrial tracer distributions. Global Biogeochemical Cycles 7(3), 645-677.

Nunes, F., and R.D. Norris, 2006. Abrupt reversal in ocean overturining during the Paleocene/Eocene warm period. Nature 439, 60-63.

Pagani, M., N. Pedentchouk, M. Huber, A. Sluijs, S. Schouten, H. Brinkhuis, J.S. Sinninghe Damsté, G.R. Dickens, and the Expedition 302 Scientists, 2006. Arctic hydrology during global warming at the Paleocene/Eocene thermal maximum. Nature 442, 671-675.

Robert, C., and J.P. Kennett, 1994. Antarctic subtropical humid episode at the Paleocene-Eocene boundary – clay-mineral evidence. Geology 22, 211-214.

Schmitz, B., R.P. Speijer, and M.P. Aubry, 1996. Latest Paleocene benthic extinction event on the southern Tethyan shelf (Egypt): Foraminiferal stable isotopic (delta C- 13,delta O-18) records. Geology 24, 347-350.

Shellito, C.J., L.C. Sloan, and M. Huber, 2003. Climate model sensitivity to atmosphericc CO2 levels in the Early-Middle Paleogene. Paleogeogr., Paleoclimatol., Paleoecol. 193, 113-123.

Thomas, D.J., T.J. Bralower, and C.E. Jones, 2003. Neodymium isotopic reconstruction of late Paleocene-early Eocene thermohaline circulation. Earth and Planetary Science Letters 209, 309-322.

Thomas, D.J., T.J. Bralower, and J.C. Zachos, 1999. New evidence for subtropical warming during the late Paleocene thermal maximum: Stable isotopes from Deep Sea Drilling Project Site 527, Walvis Ridge. Paleoceanography 14, 561-570.

Tripati, A.K., and H. Elderfield, 2004. Abrupt hydrographic changes in the equatorial Pacific and subtropical Atlantic from foraminiferal Mg/Ca indicate greenhouse origin for the thermal maximum at the Paleocene-Eocene Boundary. Geochemistry Geophysics Geosystems 5, 2003GC000631.

Westerhold, T., U. Röhl, J. Laskar, I. Raffi, J. Bowles, L.J. Lourens, and J.C. Zachos, 2005. New high-resolution chronology from the first complete late Paleocene – early Eocene marine records from Walvis Ridge: duration of Chron 24r and new constraints on the timing of early Eocene global warming events. Geologic Society of America Ann. Meeting, Salt Lake City.

Wing, S.L., G.J. Harrington, F.A. Smith, J.I. Bloch, D.M. Boyer, and K.H. Freeman, 2005.Transient floral change and rapid global warming at the Paleocene-Eocene boundary. Science 10, 993-996.

Zeebe, R. E., and J. C. Zachos. Reversed deep-sea carbonate ion basin-gradient during Paleocene-Eocene Thermal Maximum. Paleoceanography, 22, PA3201, doi:10.1029/2006PA001395, 2007.

Zachos, J. C., Pagani, M., Sloan, L., Thomas, E., and Billups, K., 2001, Trends, rythyms, and aberrations in global climate 65 Ma to Present. Science, v. 292, p.686. Zachos, J.C., D. Kroon, P. Blum, et al., 2004. Early Cenozoic Extreme Climates: The Walvis Ridge Transect. Proc. ODP, Initial Reports: College Station.

Zachos, J.C., M.W. Wara, S. Bohaty, M.L. Delaney, M.R. Petrizzo, A. Brill, T.J. Bralower, and I. Premoli-Silva, 2003. A transient rise in tropical sea surface temperature during the Paleocene-Eocene Thermal Maximum. Science 302, 1551-1554.

Zachos, J.C., and 11 others, 2005. Rapid acidification of the ocean during the Paeocene-Eocene thermal maximum. Science 308, 1611-1615.