How ancient warm periods can help predict future climate change

Several more decades of increased carbon dioxide emissions could lead to melting ice sheets, mass extinctions and extreme weather becoming the norm. We can’t yet be certain of the exact impacts, but we can look to the past to predict the future.

We could start with the last time Earth experienced CO2 levels comparable to those expected in the near future, a period 56m to 34m years ago known as the Eocene.

The Eocene began as a period of extreme warmth around 10m years after the final dinosaurs died. Alligators lived in the Canadian Arctic while palm trees grew along the East Antarctic coastline. Over time, the planet gradually cooled, until the Eocene was brought to a close with the formation of a large ice sheet on Antarctica.

During the Eocene, carbon dioxide (CO2) concentrations in the atmosphere were much higher than today, with estimates usually ranging between 700 and 1,400 parts per million (ppm). As these values are similar to those anticipated by the end of this century (420 to 935ppm), scientists are increasingly using the Eocene to help predict future climate change.

We’re particularly interested in the link between carbon dioxide levels and global temperature, often referred to as “equilibrium climate sensitivity” – the temperature change that results from a doubling of atmospheric CO2, once fast climate feedbacks (such as water vapour, clouds and sea ice) have had time to act.

To investigate climate sensitivity during the Eocene we generated new estimates of CO2 throughout the period. Our study, written with colleagues from the Universities of Bristol, Cardiff and Southampton, is published in Nature.

Reconstruction of the 40m year old planktonic foraminifer Acarinina mcgowrani.
Richard Bizley ( and Paul Pearson, Cardiff University, CC BY

As we can’t directly measure the Eocene’s carbon dioxide levels, we have to use “proxies” preserved within sedimentary rocks. Our study utilises planktonic foraminifera, tiny marine organisms which record the chemical composition of seawater in their shells. From these fossils we can figure out the acidity level of the ocean they lived in, which is in turn affected by the concentration of atmospheric CO2.

We found that CO2 levels approximately halved during the Eocene, from around 1,400ppm to roughly 770ppm, which explains most of the sea surface cooling that occurred during the period. This supports previously unsubstantiated theories that carbon dioxide was responsible for the extreme warmth of the early Eocene and that its decline was responsible for the subsequent cooling.

We then estimated global mean temperatures during the Eocene (again from proxies such as fossilised leaves or marine microfossils) and accounted for changes in vegetation, the position of the continents, and the lack of ice sheets. This yields a climate sensitivity value of 2.1°C to 4.6°C per doubling of CO2. This is similar to that predicted for our own warm future (1.5 to 4.5°C per doubling of CO2).
Our work reinforces previous findings which looked at sensitivity in more recent time intervals. It also gives us confidence that our Eocene-like future is well mapped out by current climate models.

Fossil foraminifera from Tanzania – their intricate shells capture details of the ocean 33-50m years ago.
Paul Pearson, Cardiff University, CC BY

Rich Pancost, a paleoclimate expert and co-author on both studies, explains: “Most importantly, the collective research into Earth history reveals that the climate can and has changed. And consequently, there is little doubt from our history that transforming fossil carbon underground into carbon dioxide in the air – as we are doing today – will significantly affect the climate we experience for the foreseeable future.”

Our work also has implications for other elements of the climate system. Specifically, what is the impact of higher CO2 and a warmer climate upon the water cycle? A recent study investigating environmental change during the early Eocene – the warmest interval of the past 65m years – found an increase in global precipitation and evaporation rates and an increase in heat transport from the equator to the poles. The latter is consistent with leaf fossil evidence from the Arctic which suggests that high precipitation rates were common.

However, changes in the water cycle are likely to vary between regions. For example, low to mid latitudes likely became drier overall, but with more intense, seasonal rainfall events. Although very few studies have investigated the water cycle of the Eocene, understanding how this operates during past warm climates could provide insights into the mechanisms which will govern future changes.
The Conversation
This blog was written by Cabot Institute member Gordon Inglis, Postdoctoral Research Associate in Organic Geochemistry, University of Bristol and Eleni Anagnostou, Postdoctoral Research Fellow, Ocean and Earth Science, University of Southampton

This article was originally published on The Conversation. Read the original article.

Uncertain World: Understanding past and future sea level rise

A recent study published in Science Advances suggests that if we burn all attainable fossil fuels (up to 12,000 gigatonnes of carbon), the Antarctic ice sheet is likely to become almost ice-free within 10,000 years. However, what does this mean in terms of sea level rise? To illustrate this we have designed an infographic which shows the likely extent of sea level rise under a range of different scenarios. We have chosen to use the Wills Memorial Building as an example and assume, for the purpose of this exercise, that it resides at sea level (Figure 1).

1) Sea level rise over the next century:

The most recent report by the Intergovernmental Panel on Climate Change (IPCC AR5) indicates that if we continue emitting greenhouse gases under business-as-usual scenarios (i.e. no reduction in emissions), it is likely that global mean sea level will rise between 0.52 and 0.98 m by the year 2100. If we are more optimistic, and we allow greenhouse gas emissions to peak in 2040 and decline thereafter, the range of likely global mean sea level rise is lower, but not insignificant (0.36 to 0.71 m). Both of these estimates are illustrated below and shown alongside the Wills Memorial Building.

Figure 1: An infographic showing the approximate height of sea level rise depending upon a range of different scenarios (Fretwell et al., 2013; IPCC AR5). This assumes the Wills Memorial Building resides at sea level

Although ~30 to 100 cm of sea level rise may seem insignificant, it is worth considering what this means for other regions. For example, “…since 80% of its 1,200 islands are no more than 1m above sea level“, sea level rise has the potential to impact up to 360,000 citizens and lead to widespread migration.

The reason that scientists provide a range of values for sea level rise is that the climate system is very complex. For example, under low emissions scenarios, there is expected to be an increase in moisture content around Antarctica, leading to increased snowfall along the ice sheet margins. However, under higher emissions scenarios, ice sheet discharge overcompensates for an increase in snowfall, leading to a net sea level rise.

2) Sea level rise over 10,000 years:

The variations between these two emission scenarios are less important when looking over longer timescales. Winklemann et al. (2015) have recently simulated changes in the Antarctic ice sheet over the next 10,000 years using a combination of climate and ice sheet models. From these experiments, it is clear that ice loss is driven by two key feedback mechanisms. The first begins with warming and subsequent retreat of the grounding line (Figure 2). The grounding line is the region where ice transitions from a grounded ice sheet to a freely-floating ice shelf. When the grounding line retreats to a point where the ice sheet falls below sea level, then ice sheets can become unstable.

Figure 2: A schematic of an ice sheet showing the position of the grounding line (bottom right). Image credit:

Winklemann et al. (2015) argue that the West Antarctic Ice Sheet (WAIS) becomes unstable when cumulative carbon emissions reach 600 to 800 gigatonnes of carbon (this is equivalent to a 2 degree rise in temperature by 2100). If this part of the Antarctic Ice Sheet becomes unstable, we can expect ~4 m of global sea level rise (Figure 1).Once a specific temperature is reached, a second feedback then kicks in. This destabilises the rest of the Antarctic ice sheet via the so-called surface elevation feedback. On the timescale of 10,000 years this will eventually lead to an almost ice-free Antarctica (Winklemann et al. 2015).

Figure 3: Predicted ice-sheet loss on Antarctica under different carbon emission pathways (Winkelmann et al., 2015: Science Advances).

3) Sea level rise over millions of years:

Palaeoclimatologists can provide insights into the fate of ice sheets over longer timescales. For example, the last time Antarctica was ice-free was during the early Eocene (~56 to 48 million years ago). During this interval, carbon dioxide concentrations were much higher and allowed the development of lush, tropical rainforests along the ancient coastline (Figure 4). Gradual cooling over millions of years eventually culminated in the sudden and rapid establishment of ice-sheets on Antarctica. This occurred ~34 million years ago and was likely driven by a reduction in carbon dioxide (and perhaps some other feedback mechanisms). Although Antarctica has fluctuated in size since then, it has never been completely ice-free since the Eocene. However, under rising carbon emissions, we are rapidly returning to a world that has not been seen for at least 34 million years.

Figure 4: This may be what the East Antarctic coastline looked like during the early Eocene (Pross et al., 2012).

Further reading:

  • Fretwell et al. 2013. Bedmap2: improved ice bed, surface and thickness datasets for Antarctica. The Cryosphere. v. 7.
  • Winkelmann et al. 2015 Combustion of available fossil fuel resources sufficient to eliminate the Antarctic Ice Sheet. Science Advances, v.1.
  • Bamber et al., 2009. Reassessment of the Potential Sea-Level Rise from a Collapse of the West Antarctic Ice Sheet. Science. v. 324
  • Church et al. 2013.  Sea Level Change. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA (see Chapter 13; Table 13.5, p. 1182 for 21st Century sea-level rise estimates).
  • Pross et al., 2012. Persistent near-tropical warmth on the Antarctic continent during the early Eocene epoch. Nature. v. 488.

n.b. As with the IPCC, we occasionally use the following terms to indicate the assessed likelihood of an outcome or a result. These are noted in italics: Virtually certain 99–100% probability, Very likely 90–100%, Likely 66–100%, About as likely as not 33–66%, Unlikely 0–33%, Very unlikely 0–10%, Exceptionally unlikely 0–1%.


Correction: the original post incorrectly stated that “… more than 80% of the Maldives lie one metre below sea level”. This has since been amended. Thanks to @radicalrodent for spotting this.

This blog was written by Gordon Inglis (@climategordon), a palaeoclimatologist working in the Organic Geochemistry Unit within the School of Chemistry. The infographic was created by Catherine McIntyre (@cathmci), an organic geochemistry PhD student working in same group.

Insights from the Natural Systems and Processes Poster Session

The Natural Systems and Processes Poster Session (NSPPS) is a University-wide poster session for postgraduate students within the Faculty of Science aimed at increasing inter-departmental connections within a relaxed and informal environment. This year’s event, which was hosted within the Great Hall of the Wills Memorial Building, was attended by ~90 PhD students from a wide variety of disciplines and hundreds more visitors came from across the University to view the posters. Most participants were interested in tackling the challenges of uncertain environmental change with an emphasis upon climate change, natural hazards and human impacts on the environment.

The Natural Systems and Processes Poster Session 2015 in the Great Hall
in the Wills Memorial Building (Image credit: D. Naafs)
Adam McAleer, a final year PhD student working in the Department of Earth Sciences, is interested in measuring the flux of greenhouse gases from restored peatlands within Exmoor National Park. The Exmoor Mires Project seeks to raise water levels via blocking of old agricultural drains in order to re-saturate the peatlands and recover its peat-forming biogeochemistry. This will potentially lead the mires to become carbon dioxide sinks and methane sources. As wetter plants were found to have a strong association to higher methane emissions, certain plant species have the potential to be used as a proxy for methane fluxes and restoration success. Mark Lunt, a third year PhD student working within the Atmospheric Chemistry Research Group, is interested in the fate of other greenhouse gases, such as hydrofluorocarbons (HFCs). Hydrofluorocarbons are organic compounds that contain fluorine and hydrogen atoms and are used as refrigerants, aerosol propellants, solvents, and fire retardants in the place of chloroflourocarbons (CFCs). Although HFCs do not harm the ozone layer, they can contribute to global warming. In developing countries, demand for HFCs are increasing rapidly; as a result, both the USA and China have agreed to begin work on phasing out hydroflourocarbons.

Felipe (left) discussing his research to staff and students  (Image credit: D. Naafs)
Catherine McIntyre (1st year) and John Pemberton (1st year), based within the Organic Geochemistry Unit, presented work from the NERC-funded DOMAINE project. This project aims to look at dissolved organic matter (DOM) in freshwater ecosystems and public water supplies and will focus upon the fate of carbon, nitrogen and phosphorus. Phosphorus, for example, is used to make fertilisers and can be incorporated into lakes and streams via terrestrial run-off. As phosphorus is a key limiting nutrient, it can also stimulate algal blooms and lead to eutrophication (i.e. oxygen starvation). Indeed, the global phosphorus cycle has already been highly perturbed, as shown below. As very little is known about organic phosphorus, the DOMAIN project will investigate this further using via high-resolution molecular techniques.

Four of the nine planetary boundaries  have now been crossed (Steffen et al., 2015; Science)
Other students are using the past to explore the future. Matt Carmichael, a final year PhD based within the School of Chemistry, is interested in understanding how the hydrological cycle varied during past warm climates. Of particular interest is the early Eocene (~48 to 56 million years ago), an interval characterised by high atmospheric carbon dioxide, high sea surface temperatures and the absence of continental ice sheets. However, the impact of these changes on the wider Earth system, especially those related to precipitation patterns, vegetation and biogeochemical cycles, remain poorly understood. This is achieved using climate models which can simulate changes in the atmosphere and the ocean during the Eocene. Future climatic change will also have a profound effect upon the hydrological cycle with the potential to make floods and droughts more extreme.

How the East Antarctic coastline might have looked during the early Eocene (Pross et al., 2012; Nature)

Collectively, the NSPPS highlights the wide variety of research undertaken with the Faculty of Science and is a great opportunity for PhD students to present their research in a relaxed setting.

This blog was written by Gordon Inglis (@climategordon) a final year PhD student within the School of Chemistry. Additional thanks to Adam McAleer, Matt Carmichael, Mark Lunt, Catherine McIntyre and John Pemberton whose work is highlighted here. 

Do we care too much about nature?

Over 80% of British adults believe that the natural environment should be protected at all costs. Yet, a recent report suggests that “government progress on commitments to the natural environment has been largely static” (1). Indeed, the budget for DEFRA, the Department for Environment, Food and Rural Affairs, has been slashed by 10% (£37m) and a reduction in green levies is likely as the government attempts to reduce domestic energy bills.

Has the government lost interest in the environment? Or do we care too much about nature?
To discuss this further, the Cabot Institute hosted a public recording of BBC Radio 4′s Shared Planeta show which explores the complex relationship between the human populations and wildlife. John Burton, CEO of the World Land Trust (WLT), was the first panellist and is a well known journalist and conservationist who has raised £19m for nature conservation in Africa, Asia and Central and Southern America. He believes that we should think about policy on “the life scale of an oak tree” and that further measures are required to protect the environment, both at home and abroad. The second panellist, Hannah Stoddart, is the head of the economic justice policy team at Oxfam GB and believes that fairer redistribution of wealth is more important than wildlife conservation.
Do we care about nature?
A new report, by the Environmental Funders Network, suggests that one in ten UK adults are now a member or supporter of Britain’s environmental and conservation groups (2). This equates to nearly 4.5 million people, with 81 organisations protecting species and 78 working on climate change. Although 44% of funding is allocated to biodiversity and nature protection, only 7.3% of total funds have been allocated to the climate and the atmosphere. This suggests we are more interested in ‘traditional’ environmental issues than climate change. A recent research project by the RSPB indicates that four out of five UK children are no longer connected with nature (3). Dr Mike Clarke, the chief executive of the RSPB, explains that “…nature is in trouble, and children’s connection to nature is closely linked to this”. At a time where UK species are in decline, are we doing enough to engage young people in the natural world?
An alternative to conservation
Both John Burton and Hannah Stoddart agree that nature is important and that conservation can help protect endangered landscapes. However, many conservation sites are maintained in ”favourable condition”. In other words, they are kept in the condition they were found when designated as conversation sites. A alternative concept, known as rewilding, attempts to reverse the destruction of nature by standing back and allowing nature to control its own destiny.
Currently, farmers have to prevent the development of foreign or exotic vegetation on their land. This results in the development of bare land, lacking in biodiversity. Removal of the ‘agricultural condition’ rule and the introduction of rewilding may allow this land to flourish once again. George Monbiot, author of Feral, is particularly interested in the reintroduction of megafauna, large animals that existed at the end of the last glacial period (>11ka) (4). It seems hard to believe, but over ten thousand years ago, elephants, rhinoceri and camels roamed Europe while other animals, such as bison, wolves and wildcats, were particularly widespread throughout the UK.
Indeed, the re-introduction of missing species can have a profound effect on wildlife. In 1995, grey wolves were reintroducedto Yellowstone National Park for the first time in 50 years (5). The elk population, who were now at risk of predation by wolves, began to redistribute. This allowed willow and aspen trees to flourish and increased the habitat for certain bird species, small mammals, beavers, and moose. This effect, known as a trophic cascade, suggests that careful reintroduction of megafauna into the wild can allow ecosystems to flourish. However, rewilding can backfire. In 2008, endangered Mallorcan toads were reintroduced into the natural population but were infected with Batrachochytrium dendrobatidis, a well-known fungus that can threaten amphibians (6). As a result, the Mallorcan toads are now in danger of being wiped out once again. Despite this, I believe that rewilding in the UK is feasible and could allow the public, especially children, to reconnect with nature in new and exciting ways.
  1. Nature Check 2013.
  2. Passionate Collaboraton.
  3. RSPB Connecting with Nature.
  4. Monbiot, G. Feral: searching for enchantment on the frontiers of rewilding. Allen Lane.
  5. Ripple et al,. 2001. Trophic cascades among wolves, elk and aspen on Yellowstone National Parks’s northern range.Biological Conservation102. 227-234
  6. Walker et al, 2008. Invasive pathogens threaten species recovery programs. Current Biology18. R853-R854

Paul F. Hoffman visits the University of Bristol


Paul F Hoffman of Harvard

On the 24th and 25th of September, Professor Paul F Hoffman of Harvard University (USA) kindly offered to visit the University of Bristol for two days. Fresh from fieldwork in Namibia, Paul agreed to give two talks: one upon Cryogenian glaciations and another upon the interaction of climate scientists and geologists.

Snowball Earth – Image from COSMOS

Paul is perhaps most well known for his part in the development of the Snowball Earth theory, suggesting that during the Cryogenian (850 to 635 million years ago) ice covered the entire globe, from the poles to the tropics. This theory is based upon multiple strands of evidence including palaeomagnetics, sedimentology, isotopic analysis and numerical modelling. Paul succinctly summarised these ideas while also discussing some new results published in Science two years ago. The authors of this paper suggest that during the breakup of Rodinia, a proterozoic supercontinent, the eruption of the Franklin Large Igneous Province (LIP) in Canada (716Ma) may have produced a climatic state more susceptible to glaciation. Although there have been many critics of Snowball Earth, it seems Paul remains loyal to the theory.  A wine reception was held afterwards within the School of Geography and allowed for further discussion amongst staff and students.

Paul gave a second talk on 25th September to a selection of PhDs and PDRAs who attend the Climate Journal Club (see below for details). Paul chose to give a more anecdotal, but nonetheless interesting, talk on the co-evolution of climate scientists and geologists during the last 250 years. His talk focused upon the development of a theory: from indifference to hysteria, followed by rejection and then finally acceptance. I asked him where Snowball Earth stands. He replied that it was somewhere in between hysteria and rejection!

Maybe in 50 years time we will know whether Paul was right all along…

For more details, see the following references:

Hoffman, P.F., et al (1998) A neoproterozoic Snowball Earth. Science, 281, 1342
MacDonald, F.A., et al (2010) Calibrating the Crypogenian. Nature, 327, 1241

This blog was written by Gordon Inglis who runs the Climate Journal Club at the University of Bristol.

For more details on attending the Climate Journal Club (bimonthly event designed to allow PhD and PDRAs to discuss a selection of climate-themed paper), please email