Nature 2°C paper

Now at

This page provides background information, an Excel tool, presentation slides, the press briefing audio files and some of the data displayed in the figures of the paper, featured on the cover of the 30th April 2009 issue of Nature:

Meinshausen, M., N. Meinshausen, W. Hare, S. C. B. Raper, K. Frieler, R. Knutti, D. J. Frame and M. R. Allen (2009). "Greenhouse-gas emission targets for limiting global warming to 2°C." Nature 458(7242): 1158. (HTML) (PDF) (Supplementary)

In the same issue of Nature, there is a companion paper by Myles Allen and colleagues "Warming caused by cumulative carbon emissions towards the trillionth tonne", a NewsViews piece by Gavin Schmidt and David Archer, an editorial "Time to act", and multiple other essays and opinion pieces (see Furthermore, the authors wrote a commentary "The Exit Strategy", available on Nature Reports Climate Change.

26 Questions & Answers

Background questions and answers on our paper "Greenhouse-gas emission targets for limiting global warming to 2°C". Questions like: What is new about this study?; Will we run out of fossil fuels before we can cause dangerous climate change? How many years at present levels of fossil fuel emissions can we still afford? You specify emissions only until 2050. What happens after 2050?  ... and many more. (PDF, 0.5MB)

The PRIMAP 2°C Check Tool

We provide a simple 2C Check spreadsheet tool (MS Excel) in order to quickly estimate the 2C exceedance probability of your own emission pathway. If you know the cumulative emissions between 2000 and 2050 of your emission pathway, you can query this tool for an illustrative central estimate and the range of 2°C exceedance probabilities. This tool is based on Figure 3 and S1 of our paper. This tool is free of charge, but strictly for non-commercial purposes only. Download here.

Press Briefing and Media coverage

Please find here an audio recording of the press conference of Monday, 27th April 2009 in WMA format (4,2MB). Please see the PIK press release here. 

Media coverage on Nature 2°C article

  • Independent: World will have exceeded 2050 safe carbon emissions limit by 2020, scientists say Source: chaos predicted by CO2 study World will have exceeded 2050 safe carbon ...
    Posted May 7, 2009, 4:15 AM by Malte Meinshausen
  • New Scientist: Humanity's carbon budget set at one trillion tonnes Source: Humanity's carbon budget set at one trillion tonnes If no climate policies are implemented (red ...
    Posted May 7, 2009, 2:06 AM by Malte Meinshausen
  • Reuters: World can burn 25 percent of oil, coal "safely": study Source: can burn 25 percent of oil, coal "safely": study Wed Apr 29, 2009 6:38pm BST  Email | Print | Share | Single Page ...
    Posted May 7, 2009, 2:03 AM by Malte Meinshausen
Showing posts 1 - 3 of 5. View more »

Presentation Slides

Please find here a set of Power Point presentation slides of the figures within the Nature paper and the Supplementary. Please make sure that all figures are appropriately credited with (c) Nature, M. Meinshausen et al. (2009)

Climapoly Board game

The Brazilian newspaper Folha de S.Paulo produced a MONOPOLY-like board game based on our article. Download the do-it-yourself game freely here. All you need is a printer and scissors - and a bit of knowledge in Portuguese.

Simple illustrative Figure

This simple figure illustrates a pathway with a 25% exceedance probability for 2C, assuming one trillion tonnes CO2 emissions (1000 GtCO2) between 2000 and 2050. Download high-resolution vector graphics of this figure. (ZIP Archive, 2.4MB)

Figure caption: Two possible futures: One in which no climate policies are implemented (red), and one with strong action to mitigate emissions (blue). Shown are fossil CO2 emissions (top panel) and corresponding global warming (bottom panel). The shown mitigation pathway limits fossil and land-use related CO2 emissions to 1000 billion tonnes CO2 over the first half of the 21st century with near-zero net emissions thereafter. Greenhouse gas emissions of this pathway in year 2050 are ~70% below 1990 levels. Without climate policies, global warming will cross 2°C by the middle of the century. Strong mitigation actions according to the blue route would limit the risk of exceeding 2°C to 25%. For more details, see Figure 2 in Meinshausen et al. (2009). Credit: M. Meinshausen et al. (2009)

Copyright information: This figure can be freely used as long as the source is stated; licensed under the creative commons attribution license (


Figure 1 | Joint and marginal probability distributions of climate sensitivity and transient climate response (TCR). a, Marginal PDFs of climate sensitivity; b, marginal PDFs of TCR; c, posterior joint distribution constraining model parameters to historical temperatures, ocean heat uptake and radiative forcing under our representative illustrative priors. For comparison, TCR and climate sensitivities are shown in panel c for model versions that yield a close emulation of 19 CMIP3 AOGCMs (white circles).

ASCII data of joint distribution 0.6 MB

Figure 2 | Emissions, concentrations and 21st century global mean temperatures. a, Fossil CO2 emissions for IPCC SRES, EMF-21 scenarios and a selection of EQW pathways analyzed here; b, greenhouse gas emissions, as controlled under the Kyoto Protocol; c, median projections and uncertainties based on our illustrative default case for atmospheric CO2 concentrations for the high SRES A1FI and the low HALVED-BY-2050 scenario, which halves 1990 global Kyoto-gas emissions by 2050; d, total anthropogenic radiative forcing; e, surface air global mean temperature; f, maximum temperature during the 21st century versus cumulative Kyoto-gas emissions for 2000-2049.

Zip Archive of ASCII data 9 MB; All EQW emission pathways (162 MB zipped ASCII)

Figure 3 | The probability of exceeding 2°C warming versus CO2 emitted in the first half of the 21st century. a, Individual scenarios’ probabilities of exceeding 2°C for our illustrative default (small dots) and smoothed (least squares polynomial fits) probabilities for all climate sensitivity distributions (numbered lines, cf. Fig. 1a). The proportion of CMIP3 AOGCMs and C4MIP carbon cycle8 model emulations exceeding 2°C is shown as black dashed line. Coloured areas denote the range of probabilities of staying below 2°C in IPCC AR4 terminology – with the extreme upper distribution (12) being omitted; b, total CO2 emissions already emitted3 between 2000 and 2006 (grey area) and those that could arise from burning available fossil fuel reserves, and from landuse activities between 2006 and 2049 (median and 80% ranges, Methods).

ZIP Archive of ASCII data 0.2 MB | MS Excel 2C Check tool 4.7 MB

Figure S1 | Probabilities of exceeding 2°C for various emissions indicators: a, cumulative Kyoto gas-emissions 2000-2049; b, 2050 year Kyoto-gas emissions: c, 2020 Kyoto-gas emission. Otherwise as Figure 3 in the main part of the paper, In panel c the wide vertical spread of individual scenarios' exceedance probabilities (dots) indicates that 2020 emissions are a relatively poor indicator for maximum warming. This contrasts with 2050 Kyoto-gas emissions (b) for which the narrow vertical spread of individual scenarios' exceedance probabilities suggest that this indicator is well suited for the choosen class of scenarios. The dashed vertical lines in panel b indicate halved 1990 Kyoto-gas emissions and the bold line indicates the respective range of exceedance probabilities derived from our default constraining as well as the emulation of other studies' climate sensivity distributions.

ZIP Archive of ASCII data 0.4 MB | ZIP Archive of xls data 1.1 MB

Figure S1a High Resolution 0.4 MB | Figure S1b High Resolution 0.5 MB | Figure S1c High Resolution 0.5 MB

Slide 9

Figure S2 | Observations and constrained model results for temperatures and ocean heat content changes. a, Modelled average surface temperatures using the illustrative default case and observed2 data with 95% uncertainty range for northern hemisphere ocean; b, northern hemisphere land; c, southern hemisphere ocean; d, suthern hemisphere land; e, the global avergae; f, Modelled and observed 4,5 changes in ocean heat content up to 700m depth. The regional temperatures and the linear trend of ocean heat uptake4 over 1961 to 2003 were used to constrain the climate model parameter space The MAGICC 6.0 results based on illustrative default are shown in blue. The agreement between observed abd modelled ocean heat uptake up to 300m or 3000m depth5 is similar as shown here for 700m.

Figure S3 | Joint distribution of seven key MAGICC parameters, as example of the 82-dimensional parameter space we constrained. a-g, Histograms of parameters are provided in the diagonal, and 2000 randomly drawn parameter sets are indicating the joint distributions (illustrative default) between any two parameters in the off-diagonal scatter plots. Parameter sets that were calibrated to emulate 19 CMIP3 AOGCMs are provided as red dots in the off-diagonal scatter plots of the parameters climate sensitivity (a), ocean vertical diffusivity (b) and equilibrium land-ocean warming ratio (c).

Figure S3 High Resolution 1.8 MB

Figure S6 | Regression of CO2 concentrations in year 2100 to net total radiative forcing. a, CO2 concentrations vs. total radiative forcing for emission scenarios used in this study (small colored dots), and the stabilization scenarios (grey dots) including the linear regression (black line) as shown in Fig. 3.16 of Working Group III, IPCC AR419; b, CO2 concentrations vs. non-CO2 radiative forcing in year 2100, including the regression line from a. The linear regression shown in a is not linear when plotted in panel b due to the non-linear concentration to forcing relationship of CO2. The IPCC AR4 WGIII regression hence includes the implicit assumption of higher non-CO2 forcing, the more CO2 concentrations are reduced below about 475ppm (cf. black bold line bending upwards towards low concentration levels in panel b). Another difference of our study is that – on average – our results suggest lower non-CO2 radiative forcing contributions in 2100 than implied by the WGIII regression line (see text).

Figure S6 High Resolution 1.3 MB