Cooperative Institute for Research in Environmental Sciences

How much carbon can we afford to emit?

Myles Allen

Physicist, Climate Diagnostics Group
ClimatePrediction.net, Principal Investigator
Atmospheric, Oceanic and Planetary Physics
University of Oxford, Oxford, England, UK

The 1992 United Nations Framework Convention on Climate Change aimed to achieve "stabilisation of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference in the climate system," accepting that the science available at that time was inadequate to specify what that level was. Despite another decade of intensive research, progress towards an answer has been remarkably limited. In particular, study after study has shown how difficult it is to pin down the risk of a dangerously high long-term equilibrium warming in response to stabilising atmospheric greenhouse gas concentrations at any particular level. One recent example is the climateprediction.net project, using distributed computing resources donated by the general public to perform the world's largest ever climate simulation experiment, which has shown that it is possible to generate full-complexity three-dimensional climate models with a much higher warming response to doubling carbon dioxide than that traditionally regarded as an upper bound.

I will argue that it may always be difficult to pin down the risk of a dangerous level of warming in response to any conceivable "stabilisation scenario" for future emissions, because of uncertainties in both the carbon cycle and the climate system response. In contrast, the maximum warming observed under a "containment scenario", in which we undertake to contain the total amount of carbon released into the atmosphere rather than the emission rate in any particular year, is much better constrained by current observations. For example, past emissions of carbon dioxide amount to approximately 500 gigatonnes of carbon, and preliminary results suggest that releasing 800 more gigatonnes would yield approximately a 20% risk of a carbon-dioxide-induced global warming exceeding 2 degrees, provided emissions cease altogether after that containment target is reached. If we begin from current emissions trends, this risk is relatively insensitive to when we begin emission reductions, provided we succeed in containing the total carbon injection: but, the need for fast (economically dangerous) emission reductions rises rapidly the longer initial reductions are postponed.

About the Lecturer

Myles Allen is a University Lecturer in the Department of Physics, University of Oxford. He leads the Climate Dynamics group and is Principal Investigator of the climateprediction.net project, using distributed computing resources donated by the general public to perform the world's largest climate simulation experiment. After graduating in Physics and Philosophy from Oxford in 1987, he worked for two years in Nairobi, Kenya, including a stint at the Energy Unit of the United Nations Environment Programme where he became interested in the problem of climate change. He returned to Oxford to undertake a Doctorate in Atmospheric, Oceanic and Planetary Physics, researcing internal climate variability. He then moved to the Rutherford Appleton Laboratory in Oxfordshire to work on satellite missions for monitoring global change, and thence to the Massachusetts Institute of Technology on a NOAA Global Change Fellowship to work with Professor Richard Lindzen on the problem of quantifying uncertainty in climate analysis and prediction, which he has focussed on ever since. He contributed to the Third Assessment Report of the Intergovernmental Panel on Climate Change as a Lead Author of the Chapter on detection of change and attribution of causes, and is a Review Editor for the chapter on predictions of global climate change for the IPCC Fourth Assessment. He is married with two children.

Myles Allen