Is it possible to limit global warming to no more than 1.5°C?
Submitted to Climatic Change Letters
Supplementary materials
Ranger1, N., Gohar2, L. K., Lowe2, J. A., Raper3, S.C.B, Bowen1, A. & R.E. Ward1
1 Grantham Research Institute on Climate Change and the Environment, London School of Economics and Political Science, London, UK
2Met Office Hadley Centre, Exeter, UK
3Manchester Metropolitan University (MMU), Manchester, UK
Corresponding author: LK Gohar,
A. The Climate Modelling Approach and its Assumptions
The assessment of climate outcomes used the MAGICC model (version 4.1, Wigley and Raper, 2001) with an updated tuning as used in Lowe et al.(2009). The MAGICC model is a simple climate model; an upwelling diffusion energy-balance model connected to carbon and other gas cycle models, which can be tuned to represent the output of more complex models. The model is computationally fast, so suitable for running large numbers of simulations in order to explore uncertainties. The advantage of the model approach used in this study is that it treats important uncertain climate assumptions, like climate sensitivity, in a probabilistic way. This is a significant advantage over a deterministic approach. However, it does not remove model dependency.
The parameters are chosen such that they are in general consistent with the findings of the IPCC Fourth Assessment Report (AR4) (Solomon et al. 2007). As such, the climate assumptions are likely to lead to more stringent cuts in emissions being needed than predicted by some of the older pre-AR4 studies, because Solomon et al. 2007 increased estimates of the sensitivity of the Earth’s system to human action across a number of variables, including the climate sensitivity. Below we describe our assumptions related to the four main uncertain components/parameters:
Climate Sensitivity: There is much debate over climate sensitivity, which means that it is not currently possible to define such a distribution objectively (see Box 10.2 of Meehl et al. 2007). We use the distribution given by Murphy et al. (2004) as it incorporates information from both models and spatially explicit observations (whereas most studies include only one or the other or only global mean constraints). Comparison with the IPCC conclusions on climate sensitivity suggests that the “most likely” value of Murphy et al. is well aligned with current consensus, but that Murphy et al. under-predicts the likelihood of both low and high temperature outcomes relative to IPCC conclusion (this is unsurprising given that the IPCC conclusions represent a consensus over many studies). This will mean that the uncertainties in temperature given in this study will be narrower than the current consensus.
Ocean Diffusivity: The ocean diffusivity is an important determinant on the rate of warming. A distribution of ocean diffusivity was estimated by fitting a log-normal distribution to the model-derived ocean mixing rates across the range of current global climate models (Table 9.A1, Cubasch et al., 2001).
Carbon Cycle Feedbacks: The uncertainty in the temperature-carbon cycle feedback amplification parameter was derived from the recent model inter-comparison study, C4MIP (Friedlingstein et al. 2006). Each of the model projections are treated with equal probability and no observational constraints are applied.
Assumptions and Representation of the Effect of Aerosols on Climate: the direct and indirect radiative forcing from aerosols is highly uncertain. For example, the IPCC Fourth Assessment Report (IPCC 2007) gave an uncertainty range of -0.9 to -0.1 Wm-2 for direct aerosol forcing and -1.8 to -0.3 Wm-2 for the indirect aerosol forcing in 2005 and rated our “Level of Scientific Understanding” of the forcings as medium/low and low respectively. This uncertainty is represented stochastically in MAGICC and the uncertainty range incorporated is roughly consistent with AR4.
Modelling the historic warming and comparison with observations:
Fig.1 Past warming relative to 1960-1990 mean as simulated by MAGICC 4.1 with the modelling framework used in this study and described in Lowe et al. (2009) (thick black solid line). The observed global annual mean warming as estimated by HadCRUT3v dataset is shown in red (Brohan et al.2006,). The grey line highlights the impact of including the radiative forcing from solar and volcanic contributions taken from the RCP concentration calculation and data group database. The blue line is the warming simulated by MAGICC using the total radiative forcing available from the RCP concentration calculation and data group database.
The simulation of past warming by MAGICC is in close agreement with observed temperature rise in the latter half of the twentieth century (Fig 1). Prior to 1950 the simulated warming is cooler than observations. Further investigation suggests that the anthropogenic aerosol radiative forcing estimated by MAGICC is smaller than other estimates such as those available from the RCP concentration calculation and data group (RCPCCDG) (Meinshausen et al 2011)database and result in lower simulated temperatures for the historical period. However the impact of the lower historic aerosol forcing than observed has little significance on projected median temperatures in the twenty first century (a 1% lower median peak temperature rise seen in the 21stcentury than the standard simulation if including the solar, volcanic and lower aerosol historic radiative forcing in MAGICC).
B. Path Assumptions: Greenhouse Gases
This section describes the greenhouse gas emissions assumptions of the pathways generated for the study. For information on the Set C paths in Section 3.3, refer to Table 1 and van Vuuren et al. (2007).
Set A
Baseline Greenhouse Gas Emissions: The baseline emissions to 2020 are based on the paths described by Meinshausen (2006). Global emissions are assumed to be 37 GtCO2e in 1990, 41 GtCO2e in 2000 and 47 GtCO2e in 2010. Historical baselines are uncertain; for example, the IPCC Fourth Assessment Report estimated an uncertainty of ±4 GtCO2e per year in land-use CO2 emissions alone over the 1990s(Solomon et al. 2007). The historical emissions used here are within the uncertainty range of alternative data sources, such IEA (2009) and WRI (2011).
Multi-gas greenhouse gas emissions assumptions: The time-evolving proportion of long-lived greenhouse gases relative to carbon dioxide (in terms of CO2 equivalent) is assumed to follow the SRES A1B scenario.
The halogen containing compounds controlled by the Montreal Protocol assumptions:
The halogen containing compounds that deplete the ozone layer are assumed to be controlled by the Montreal Protocol and the emissions decrease over time to zero by 2200 for all paths considered in this work.
Set B.
Baseline Greenhouse Gas Emissions: The baseline emissions to 2007 follow the SRES A1B scenario. This scenariowas chosen for the similar (although above) total anthropogenic greenhouse gas emissions to estimates based on observations (Gohar and Lowe 2009).
Multi-gasgreenhouse gas emissions:The method of construction of the pathways is given in detail in Gohar and Lowe (2009). Briefly, CO2 and halogen containing compounds are projected to reduce at a given emissions rate after the peak in emissions in 2014. The emissions of methane are projected to follow the CO2 emissions trend assuming the SRES B1 (Nakicenovic and Swart, 2001) ratio of methane to CO2 emissions. Nitrous oxide emissions are assumed to transition from SRES A1B to SRES B1 emissions by 2027.
The halogen containing compounds controlled by the Montreal Protocol assumptions:
Emissions of the halogen containing compounds that deplete the ozone layer follow the same assumptions as in Set A pathways.
Extended Pathway Comparison
The cumulative emissions in the 21st century for the paths listed in Table 1 in the main text are compared in Fig. 2 illustrating the higher cumulative emissions of the RCP3-PD and the majority of Set B paths than those of Set A. A more detailed illustration of the multi-gas emissions for paths in Set A, Set B and RCP3-PD are shown in Fig3.
Fig 2 The cumulative emissions for the scenarios examined and detailed in Table 1 in the main paper. The light grey shaded bars are those that give at least a 50 per cent chance of global mean temperatures being no more than 1.5°C, with an overshoot of not more than 50 years with the high aerosol scenario only, and those with dark grey shading are the Set A paths that meet this goal with both the high and low aerosol scenario.
Fig.3The emissions for carbon dioxide (a), methane (b), nitrous oxide (c) and the Kyoto fluorinated gases (d) up to 2100 as defined in a selection of paths from Set A, Set B and the RCP3-PD path. The solid curves represent paths that achieve a more than 50 per cent change of limiting warming to no more than 1.5°C after allowing for an overshoot of no more than 50 yeas under a high or low aerosol future aerosol assumption. The dotted curves offer the same level of chance but only for the high future aerosol assumption. The black dashed line describes the RCP3-PD path
C. Future Aerosol EmissionsAssumptions
Projecting anthropogenic aerosol emissions requires one to anticipate trends in fossil-fuel use as well as the aerosol emissions intensity of that activity. We use projections of the aerosol intensity of total carbon dioxide emissions (i.e. the aerosol emissions per gigatonne of carbon dioxide emission) from two sources:
- ‘High’ scenario: based on the IPCC SRES B1 scenario (Nakicenovic et al. 2000)
- ‘Low’ scenario: based on simulations described byMeinshausen (2006)
Both scenarios project a reduction in the aerosol intensity of carbon dioxide emissions over time (consistent with Fisher et al. 2007), but declining at different rates. The aerosol intensities are applied to the projected total carbon emissions for a given path to calculate the overall aerosol emissions.
The Set C RCP3-PD emissions are taken from the RCPCCDG website ( accessed December 2011)
.
Fig.4The aerosol emissions from 1990 to 2100 from a selection of Set A, Set B and the RCP3-PD paths.All Set A and Set B emissions shown are for the high aerosol assumption, unless otherwise stated (A:1 and A:2 low aer) The set C RCP3-PD aerosols are taken from the extended RCPCCDG database (
D. A comparison of the RCP3-PD projected warming to other studies.
The simulations using the RCP3-PD harmonized emissions in this work yields a median warming of no more than 2.1°C in the 21st century and is in close agreement with the Arora et al.(2011) findings of 2.3°C and within the range of 1.3°C to 2.3°C given by emulating the temperature responses of nine CMIP5 GCMs (Good et al, submitted). Our findings are higher than estimates from Meinshausen et al.(2011) and Schewe et al. (2011) which are closer to 1.5°C (see Fig. 5).
Fig. 5 Estimates of peak median warming above pre-industrial levels for the RCP3-PD scenario.
Since our RCP3-PD warming estimate is derived from the greenhouse gas and aerosol emissions trajectories from the RCPCCDG as opposed to the radiative forcing provided on the website we can compare the forcing estimates.If we look at the total anthropogenic radiative forcing projections shown in Fig. 6, we see our total anthropogenic radiative forcing from emissions for RCP3-PD (dashed black line) is a little higher than the MAGICC 6 projections from RCPCCDG (purple, Meinshausen et al., 2011) with the difference due almost entirely to the aerosol contribution (Fig .7). There is uncertainty in estimating the radiative impact of aerosols spanning the chain from aerosol precursors to aerosol radiative forcing estimates. However, looking at Fig. 6 we can see path B:1 from our study has a similar radiative forcing to that of the RCP3-PD from RCPCCDG and we can use this to provide some commentary on the causes of the differences between our study and the study with the greatest chance of keeping warming to 1.5°C. As mentioned before, the B:1 path from our study has a similar integrated total anthropogenic radiative forcing to that of RCP3-PD from RCPCCDG and the projected median peak warmings are1.8°C and slightly above 1.5°C respectively. The other differences in these warmings will be due to the model parameters used in the two studies. If we now compare the path B:1 warming to the RCP3-PD path as simulated from emissions in this study, the median peak warmings are 1.8°C to 2.1°C with the differences due to the total anthropogenic radiative forcing[1] of the two paths (Fig. 7). Therefore, the difference between our warming estimate for the RCP3-PD path and the Meinshausen et al. (2011) estimate can be explained by the difference in the total anthropogenic radiative forcing projections and the model parameters used in the studies.
Fig. 6 The total anthropogenic radiative forcing between 1990 to 2100 for the RCP3-PD, B:1 and B:5 paths as simulated in the main paper from the harmonized greenhouse gas and aerosol emissions and the RCP3-PD total anthropogenic radiative forcing as provided on the RCPCCDG ().
Fig 7 The radiative forcing in 2100 for the non-CO2 gases and constituents for a selection of paths from Set A, Set B and the RCP3-PD path (calculated from the harmonised emissions in our modelling framework).The RCP3-PD* set of radiative forcings are taken from the RCPCCDG (
References
Brohan, P., Kennedy J.J,, Harris, I, Tett S.F.B. and Jones P.D. (2006) Uncertainty estimates in regional and global observed temperature changes: a new dataset from 1850. J. Geophysical Research 111, D12106 doi:10.1029/2005JD006548
Cubasch, U., Meehl, G.A., Boer, G.J. ,Stouffer, R.J., Dix, M., Noda, A., Senior, C.A., Raper, S., Yap K.S., 2001: Projections of future climate change. In: Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change [Houghton, J.T., et al. (eds.)]. CambridgeUniversity Press,Cambridge, United Kingdom and New York, NY, USA, pp. 525–582.
Fisher, B.S., N. Nakicenovic, K. Alfsen, J. Corfee Morlot, F. de la Chesnaye, J.-Ch. Hourcade, K. Jiang, M. Kainuma, E. La Rovere, A. Matysek, A. Rana, K. Riahi, R. Richels, S. Rose, D. van Vuuren, R. Warren, 2007: Issues related to mitigation in the long term context, In: Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Inter-governmental Panel on Climate Change [B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, L.A. Meyer (eds)], Cambridge University Press, Cambridge.
Friedlingstein, P., Cox, P., Betts, R., Bopp, L., von Bloh, W., Brovkin, V., Cadule, P., Doney, S., Eby, M., Fung, I., Bala, G., John, J., Jones, C., Joos, F., Kato, T., Kawamiya, M., Knorr, W., Lindsay, K., Matthews, H. D., Raddatz, T., Rayner, P., Reick, C., Roeckner, E., Schnitzler, K.-G., Schnur, R., Strassmann, K., Weaver, A. J., Yoshikawa, C., Zeng, N., 2006: Climate-carbon cycle feedback analysis: Results from the C4MIP model intercomparison. J. Clim., 19, 3337–3353.
Gohar L.K and Lowe J.A(2009) Summary of the emissions mitigation scenarios: part 1.Work stream 1, Report 2 of the AVOID programme (AV/WS1/D1/R02). Available online at
Good P, Gregory JM, Lowe JA and Andrew T Predicting and understanding CMIP5 representive concentration pathway projections using the response to abrupt CO2 change. Submitted to Climate Dynamics
International Energy Agency 2009:World Energy Outlook 2009,
Le Quéré, C., Raupach, M.R., Canadell, J.G., Marland, G. and others, 2009: Trends in the sources and sinks of carbon dioxide, Nature Geoscience, doi:10.1038/NGE0689,
Lowe, J.A., Huntingford, C., Raper, S.C.B., Jones, C.D., Liddicoat, S.K. and Gohar, L.K. 2009: How difficult is it to recover from dangerous levels of global warming? Environmental Research Letters, v.4, 014019, p.9.
Meehl et al. (2007)Global Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. CambridgeUniversity Press, Cambridge, United Kingdom and New York, NY, USA.
Meinshausen M, Smith SJ, Calvin K, Daniel JS, Kainuma MLT, Larmarque JF, Matsumoto K, Montzka SA, Raper SCB, Riahi K, Thomson A, Velders GJM and van Vuuren DPP (2011) The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109, 213-241
Meinshausen, M., 2006: What does a 2°C target mean for greenhouse gas concentrations? A brief analysis based on multi-gas emission pathways and several climate sensitivity uncertainty estimates. In: Avoiding Dangerous Climate Change [Schellnhuber, H.J., et al. (eds.)]. CambridgeUniversity Press, New York, NY, pp. 265–279
Murphy, J.M., Sexton, D.M.H., Barnett, D.M., Jones, G.S., Webb, M.J., Collins, M., and Stainforth, D.A., 2004: Quantification of modelling uncertainties in a large ensemble of climate change simulations. Nature, 429, 768–772.
Nakićenović, N., and R. Swart (eds.), 2000: Special Report on Emissions Scenarios. A Special Report of Working Group III of the Intergovernmental Panel on Climate Change. CambridgeUniversityPress, Cambridge, United Kingdom and New York, NY, USA, 599 pp
Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.). 2007: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. CambridgeUniversity Press, Cambridge, United Kingdom and New York, NY, USA, 996 pp.
van Vuuren DP , den Elzen MGJ,Lucas PL,Eickhout B, Strengers BJ, van Ruijven B, Wonink S and van Houdt R (2007)Stablizing greenhouse gas concentrations at low levels: an assessment of reduction strategies and costs. Climatic Change 81: 119-159
Wigley, T.M.L. and. Raper, S.C.B. 2001: Interpretation of high projections for global-mean warming. Science, v.293, p.451–454.
World Resources Institute (WRI) 2011. Climate Analysis Indicators Tool (CAIT) Version 8.0. WashingtonDC.
[1] Since both simulations are from the same modelling framework and thus the modelling parameters are the same.