The RCP greenhouse gas concentrations and their extensions from 1765 to 2300

Abstract

We present the greenhouse gas concentrations for the Representative Concentration Pathways (RCPs) and their extensions beyond 2100, the Extended Concentration Pathways (ECPs). These projections include all major anthropogenic greenhouse gases and are a result of a multi-year effort to produce new scenarios for climate change research. We combine a suite of atmospheric concentration observations and emissions estimates for greenhouse gases (GHGs) through the historical period (1750–2005) with harmonized emissions projected by four different Integrated Assessment Models for 2005–2100. As concentrations are somewhat dependent on the future climate itself (due to climate feedbacks in the carbon and other gas cycles), we emulate median response characteristics of models assessed in the IPCC Fourth Assessment Report using the reduced-complexity carbon cycle climate model MAGICC6. Projected ‘best-estimate’ global-mean surface temperature increases (using inter alia a climate sensitivity of 3°C) range from 1.5°C by 2100 for the lowest of the four RCPs, called both RCP3-PD and RCP2.6, to 4.5°C for the highest one, RCP8.5, relative to pre-industrial levels. Beyond 2100, we present the ECPs that are simple extensions of the RCPs, based on the assumption of either smoothly stabilizing concentrations or constant emissions: For example, the lower RCP2.6 pathway represents a strong mitigation scenario and is extended by assuming constant emissions after 2100 (including net negative CO2 emissions), leading to CO2 concentrations returning to 360 ppm by 2300. We also present the GHG concentrations for one supplementary extension, which illustrates the stringent emissions implications of attempting to go back to ECP4.5 concentration levels by 2250 after emissions during the 21st century followed the higher RCP6 scenario. Corresponding radiative forcing values are presented for the RCP and ECPs.

Introduction

A set of scenarios known as Representative Concentration Pathways (RCPs) has been adopted by climate researchers to provide a range of possible futures for the evolution of atmospheric composition (Moss et al. 2008; Moss et al. 2010). These RCPs complement and, for some purposes, are meant to replace earlier scenario-based projections of atmospheric composition, such as those from the Special Report on Emissions Scenarios (SRES; Nakicenovic and Swart 2000). The RCPs are being used to drive climate model simulations planned as part of the World Climate Research Programme’s Fifth Coupled Model Intercomparison Project (CMIP5) (Taylor et al. 2009) and other comparison exercises. The four RCPs are based on multi-gas emission scenarios which were selected from the published literature (Fujino et al. 2006; Smith and Wigley 2006; Clarke et al. 2007; Riahi et al. 2007; van Vuuren et al. 2007; Hijioka et al. 2008; Wise et al. 2009) and updated for release as RCPs (Masui et al. 2011; Riahi et al. 2011; Thomson et al. 2011; van Vuuren et al. 2011b). Because they were produced by four different Integrated Assessment Models (IAMs), there are some inconsistencies in the relationships between emissions and concentrations that could complicate the interpretation of the climatic consequences of the four different scenarios. Furthermore, although concentrations drive traditional coupled atmosphere-ocean climate models, CMIP5 also includes simulations by Earth System Models (ESMs) with a full representation of the carbon cycle. These ESMs are optionally driven by prescribed emissions of carbon dioxide. The CMIP5 exercise, therefore, requires a set of historical and future pathways for both concentrations and emissions (see Appendix 1), ideally produced by a single model. Starting from these standardised concentration datasets, forthcoming CMIP5 intercomparisons will allow our understanding of the relationship between emissions and concentrations to be re-defined.

This study describes how the IAM emissions were processed to produce the RCP GHG concentration values, including the compilation of historical GHG concentrations, the harmonization of emissions towards common 2000–2005 emission levels, the projection of best-estimate future GHG concentrations, and their extension beyond 2100. These concentration pathways lead to radiative forcing values that span a range larger than that of the SRES scenarios. In addition to the central contribution of the IAMs, this process was only possible due to the wide range of contributions from the scientific community, in particular regarding historical emissions, observed concentrations, and emission scenarios for ozone depleting substances (ODSs) (see references in Table 1). An overview of the multi-year process to develop the RCPs can be found in van Vuuren et al. (2011a).

Table 1 Historical mixing ratios of GHG concentrations used to extend RCP concentrations back in time

The harmonized GHG concentration and emissions time series recommended for CMIP5 (Taylor et al. 2009) can be obtained from the RCP database website (RCP Database 2009) available at http://www.iiasa.ac.at/web-apps/tnt/RcpDb and the CMIP5 portal (PCMDI 2009) at http://cmip-pcmdi.llnl.gov/cmip5/. Extended GHG datasets until 2500 (for use in very long-term experiments), and further background information on the generation of the harmonized GHG concentration time series are provided at: http://www.pik-potsdam.de/`mmalte/rcps/.

This paper is structured as follows. First, we discuss the general approach taken to derive GHG concentration data for the RCPs in Section 2.1. Historical GHG concentrations from 1765 to 2005 are discussed in Section 2.2. The harmonization of the emissions from the IAMs is covered in Section 2.3, the assumptions used to calculate concentrations and forcing time series for the RCPs over the 21st century are discussed in Section 2.4. The extension of the RCPs beyond 2100 is discussed in Section 3. Section 4 presents the resulting GHG concentration time series for the RCPs. Section 5 discusses the results, including the inverse emission calculations for the extensions, and Section 6 concludes.

Methods

General approach

Each of the IAM teams can, in principle, provide both emissions and concentration data. However, each of the models uses different historical and base year data for the recent past (years 2000–2005). In order to ensure a smooth transition in the climate model runs from the historical period into the future, a harmonization step for emissions was performed here. Furthermore, it was decided that a single model version of MAGICC (e.g. Wigley and Raper 2001; Wigley et al. 2009; Meinshausen et al. 2011a) should be used to produce a more consistent estimate of concentrations and carbon feedbacks, rather than basing the RCPs on a variety of different model versions. GCAM that produced RCP4.5 (Thomson et al. 2011), for example, uses MAGICC5.3, as does AIM that contributed RCP6 (Masui et al. 2011). MESSAGE that produced RCP8.5 (Riahi et al. 2011) uses an updated version of MAGICC4.2, and IMAGE that produced RCP2.6 (Van Vuuren et al. 2011b)Footnote 1 uses MAGICC6 except for its carbon cycle. Thus, as a second harmonization step, we apply a single climate and carbon cycle treatment, using the latest version 6 of MAGICC (Meinshausen et al. 2011a, b)Footnote 2 to derive concentrations and inverse emissions for the RCPs.

By design, this study is concerned with providing a ‘reference’ starting point for further analysis in model intercomparison exercises rather than providing a detailed uncertainty analysis of the cause-effect chain from emissions to concentrations and global-mean temperatures. We hence limited this documentation to a detailed description of the chosen assumptions in deriving the RCP’s GHG concentrations, which can serve as the starting point for multi-model ensemble analysis in the future. A detailed description of the cause-effect chain as included in MAGICC6 can be found elsewhere (Meinshausen et al. 2011a), including a representation of uncertainties (Wigley and Raper 2002; Meinshausen et al. 2009).

The following subsections describe our four steps to yield harmonized GHG concentrations and emissions for the RCPs from the native output of the four IAM scenarios (see Fig. 1 below).

Fig. 1
figure1

Overview of methods to harmonize emissions, derive GHG concentrations and create the extensions for the RCPs. See text for further details

Historical concentrations

For comparing climate model outcomes with historical climate observations, it is ideal if atmosphere-ocean general circulation models (AOGCMs) are driven with observations of the historical atmospheric composition. Such a comparison can be helpful for assessing the skill of climate models, or to determine the human-contribution to climate change. Building on current literature, and with the help of a number of experts, we compiled a consolidated set of 20th century global and annual mean GHG concentrations. Specifically, we compiled concentrations of carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), eight different hydrofluorocarbons (HFCs, namely HFC-23, HFC-32, HFC-43-10mee, HFC-125, HFC-134a, HFC-143a, HFC-227ea, and HFC-245fa), three perfluorocarbons (PFCs, namely CF4, C2F6, and C6F14), and SF6, as well as concentrations of 16 ODSs (CFC-11, CFC-12, CFC-113, CFC-114, CFC-115, Carbon Tetrachloride, Methyl chloroform, HCFC-22, HCFC-141b, HCFC-142b, Halon-1211, Halon-1202, Halon-1301, Halon-2402, CH3Br, CH3Cl). Building on this database of historical observations, we recommend sets of pre-industrial control run concentrations, depending on whether the historical ‘20th century’ run starts in 1765, or 1850 (or any year in between). Concentrations over this 1765 to 1850 period are constant for the halogenated gases with natural sources, i.e., CF4, CH3Br and CH3Cl, with 35 ppt, 5.8 ppt and 480 ppt, respectively. However, the recommended concentrations for the long-lived GHGs, CO2, CH4, and N2O show a small increase over that period, starting from the levels 278.1 ppm, 721.9 ppb, and 273.0 ppb in 1765 and increasing to 284.7 ppm, 791.0 ppb, 275.4 ppb in 1850, respectively. The concentrations were compiled based upon data available as of mid-2009 (see Table 1).

Harmonization of emissions

The harmonization of GHGs, tropospheric ozone precursors and aerosol emissions to common historical levels is necessary as IAMs do not necessarily start with the same historical emissions inventories, which is a disadvantage for comparisons of the scenarios’ future climate effects. There are several reasons that different IAM scenarios do not share the same historical emissions: besides the uncertainty in the historical record, different IAMs 1) do not include the same set of human activities that lead to emissions, 2) smooth short-term fluctuations differently and/or 3) assume different emissions factors from emissive processes. In addition, the actual ‘real world’ activity levels and emissions factors are inherently uncertain. The most appropriate harmonization method depends on the reasons that underlie the differences in historical emission levels. Given the many different sectors and emissions factors in the IAMs, a simple and transparent approach is followed here.

Reactive gas and aerosol emissions have been harmonized to year 2000 levels in a separate exercise in the RCP creation process (Lamarque et al. 2010; Smith et al. 2011; Granier et al. 2011), namely for sulfate oxides (SOx), carbon monoxide (CO), non-methane volatile organic compounds (NMVOC), nitrogen oxides (NOx), black carbon (BC), organic carbon (OC), ammonia (NH3), and also for methane (CH4) because of its role in atmospheric chemistry. Here, we extend this harmonization of the reactive gas emissions to year 2005 by using the average growth rates in RCP8.5, RCP4.5, and RCP2.6.Footnote 3 This harmonization, therefore, enforces consistency among all RCP scenarios over 2000–2005 period.

For other GHGs, i.e. CO2, N2O, HFCs, PFCs, and SF6, we harmonized global and five-region emission levels using simple scaling routines so that by construction, all four harmonized RCPs share the same 2000 to 2005 emissions data. The five regions are essentially the same as the four SRES regions with the ‘Africa&Latin America (ALM)’ region being split into ‘Middle East and Africa (MAF)’ and ‘Latin America (LAM)’ (see http://www.iiasa.ac.at/web-apps/tnt/RcpDb for a country-by-country definition of these regions). For fossil and industrial CO2 emissions, we used global inventory estimates from Marland et al. (2008) to 2005. Extending the harmonization only to 2005 does not include the substantial emission increase until 2008 or the zero growth rate in 2009 due to the financial crisis (Olivier and Peters 2010). This was a conscious decision, partially because RCPs are not meant to reflect short-term fluctuations. Net land-use CO2 emissions estimated by the IAMs (on average 1.15 GtC in the year 2000) are lower than some other emission estimates, e.g. the 1.41 GtC in year 2000 by Houghton (2008).Footnote 4 To maintain consistency with the underlying land-use patterns (Hurtt et al. 2011), and given the large uncertainty in current global land-use related CO2 emissions of around ±0.5 GtC/yr (DeFries et al. 2002; Canadell et al. 2007), we harmonized the emissions using the IAM average (RCP2.6, RCP4.5 and RCP8.5). For fluorinated gases that are included in the basket of gases controlled under the Kyoto Protocol (HFCs, PFCs, SF6), we used observed concentrations, where available, and derived inverse emission estimates using default lifetime assumptions (Table 2.14 in Forster et al. 2007) within the MAGICC6 coupled gas-cycle climate model. For C6F14, HFC-32, HFC-43-10mee, HFC-227ea, and HFC-245fa, we took available emissions data from either SRES (Nakicenovic and Swart 2000) (HFC-43-10mee), EDGAR4 (EC-JRC and PBL 2009) (C6F14, HFC-227ea) or the non-harmonized RCPs (HFC-32, HFC-245fa). For HFC-245fa, sparse observations exist (Vollmer et al. 2006), pointing to lower, but much faster increasing, emissions than we used here from the original RCP4.5 estimates. This difference might be due to an overestimation of actual emissions by RCPs and/or due to slower release factors in early applications of this foam blowing agent than assumed by the IAMs. For ODSs, we use the emissions that were used to derive, with a box model, the standard WMO (2007) A1 scenario concentrations. Further details are provided in Table 2.

Table 2 Harmonization emission values

We employ a harmonization process whereby the original IAM emission data is adjusted to the common 2000–2005 values and these adjustments are phased out afterwards. Specifically, the longer term RCP emission levels are identical to those of the original IAM emissions from 2050 onwards. In between, from 2005 to 2050, a multiplier, i.e., the ratio between RCP harmonized emission levels and original IAM emissions in 2005, is linearly relaxed back to 1 until 2050 (c.f. Van Vuuren et al. 2008).

Exceptions to this approach are applied for land-use related CO2 emissions and some fluorinated gases. Land-use related CO2 emissions turn negative in some regions, which is why we chose to apply an additive shift of emissions rather than a multiplier. The difference between original IAM emissions and the harmonized levels in 2005 is added to the original IAM data and this offset is linearly reduced to zero by 2030. A scaling factor rather than an offset until 2050 could have resulted in more pronounced negative emissions even in the case of an upward adjustment in 2005.

For fluorinated gases, the SRES scenarios used an external set of emissions (Fenhann 2000). In contrast, the IAMs producing the RCP scenarios now include fluorinated gas emissions in their modeling frameworks. For some fluorinated gas species, updated information on their trends is available, which causes IAMs to project markedly different future emissions compared to SRES. Some IAMs, however, include only a few aggregate fluorinated gas categories. This leads to somewhat artificially narrowed spreads for some fluorinated gas projections.

For HFC-227ea, current emission levels provided in the RCPs were uncertain as emissions for this gas in both RCP4.5 and RCP8.5 were only available as an aggregate together with HFC-125. Large uncertainties in HFC-227ea emissions result from the fact that there were no ambient air measurements that could constrain anthropogenic emission estimates in mid-2009, a fairly unique situation for a GHG - and only recently remedied (Laube et al. 2010). Bottom-up emission estimates (e.g. EDGAR4 data by EC-JRC and PBL 2009) seem to overestimate actual HFC-227ea emissions estimated from ambient measurements (Laube et al. 2010). The overall radiative forcing contribution of both gases, HFC-245fa (see above) and HFC-227ea, is rather small, so any future revisions will likely have a minor effect on aggregate radiative forcing levels. For the RCP4.5 and RCP8.5 scenarios, a simple approximation has been used to estimate future emissions of HFC-227ea and HFC-125. The reported GWP-weighted aggregate emissions (HFC-227ea plus HFC-125) were multiplied by scaling factors from the RCP2.6 scenario, which was the only scenario with separate projections for these two gases. RCP6 emissions for HFC227ea were taken from RCP2.6. EDGAR4 data (EC-JRC and PBL 2009) were used for historical harmonization values for HFC-227ea and a constant scaling factor was applied. Similarly, for HFC-245fa and SF6, a ‘ramped’ scaling until 2050 would have led to a considerable change in growth rates compared to the near-monotonic increase of emissions until 2100 reported for RCP8.5. Thus, a constant scaling factor was applied over time, which led to higher RCP8.5 emissions by 2100 than projected by the original IAM scenario (see Fig. 2).

Fig. 2
figure2

Harmonized emissions under the four RCP scenarios. The non-harmonized scenarios (dashed lines) are in most cases marginally different from the harmonized emissions (solid lines). For illustrative purposes, emissions are weighted with IPCC SAR Global Warming Potentials of a 100-year time horizon (IPCC 1996)

The net effect of the harmonization procedure for the long-lived GHGs and ODSs is in some years moderate, but is generally small and negligible in the long-term. Over all years, the highest upward shift in GWP-weighted (100 year time horizon) (IPCC 1996) aggregate emission levels is 11.5% for the RCP6 scenario, which is due to a substantial upward shift of landuse CO2 emissions from 1.3 to 4.4 GtCO2/yr in year 2005 (+226%). A 4.1% upward adjustment of the RCP2.6 scenario in 2005 resulted as well from the harmonization of landuse CO2 emissions from 2.8 to 4.4 GtCO2/yr in year 2005, in addition to small upward corrections of CH4 and fluorinated gases (see Fig. 2). Harmonization reduced the RCP8.5 emissions slightly, by 2.1% in 2005. The aggregate emission levels of RCP4.5 faced a small upward shift by 2.1% in 2005. By construction, the harmonization procedure had only negligible effects on post-2050 emission levels (<0.3%) given that only a few fluorinated gases were adjusted after 2050.

Note that these induced shifts of emission levels are within the uncertainty of current emission estimates. Nevertheless, due to the cumulative effect on radiative forcing levels for long-lived gases, the harmonization procedure will result in slightly different concentration and temperature projections (which is intended as part of increasing the comparability between the different scenarios). To ensure a smooth transition of concentrations from historical runs for shorter-lived substances, the harmonization is an essential step for consolidating the scenarios from four different IAMs for a climate model inter-comparison. Otherwise, a consistent comparison of the resulting climate consequences across the scenarios would be hindered.

Some other limitations of the harmonized emissions arose. The ODS projections (WMO 2007) do not incorporate the effects of the accelerated HCFC phase-out accepted by the Parties to the Montreal Protocol in 2007. In the absence of specific mitigation policies, this acceleration would be expected to lead to somewhat lower future HCFC emissions (Velders et al. 2007). and higher HFC emissions (Velders et al. 2009). Furthermore, NF3 (as well as several less abundant fluorinated gas compounds) are not included in the RCPs even though they may have small positive radiative forcing effects (Prather and Hsu 2008; Prather and Hsu 2010).

Calculation of GHG concentrations

In most of the experiments for the CMIP5 intercomparison exercise, AOGCMs and ESMs will be driven by historical and future GHG concentrations, not emissions, as shown in Appendix 1 and further described in Taylor et al. (2009). We derive concentrations from the harmonized emissions with a single model, MAGICC6, in order to ensure consistency between and within the different RCPs. Although various versions or parts of MAGICC are used in many IAMs (see above), we chose MAGICC not as a model in its own right, but because of its ability to closely emulate the full range of C4MIP carbon cycle and CMIP3 AOGCMs (Meinshausen et al. 2011a). We now summarize the various gas-cycle parameterizations, before providing more detail on the chosen carbon cycle and climate response settings. Our assumptions regarding non-GHG forcing agents are detailed in Appendix 2.

MAGICC uses gas-cycle parameterizations of different complexity to project concentrations and radiative forcing for CO2, CH4, N2O,3 PFCs, 8 HFCs, SF6 and 16 ODSs (see listing of individual species in Section 2.2). For CO2, MAGICC includes a global carbon cycle model with three land carbon pools, an ocean carbon scheme, and multiple temperature-dependent terrestrial and oceanic fluxes, as well as a parameterization for the CO2 fertilization effect. The model is designed to closely emulate higher complexity carbon cycle models regarding seven aggregated carbon pools and fluxes, as well as atmospheric CO2 concentrations, as described in detail in Meinshausen et al. (2011a). The model does not yet include a nitrogen cycle or interactions between the carbon cycle and nitrogen cycle – reflecting the state of carbon cycle models in C4MIP in 2006 (Friedlingstein et al. 2006). The chosen carbon cycle calibration is further described below.

For projecting CH4 and N2O concentrations, parameterizations from Ehhalt et al. (2001) are used—including simplified temperature-dependent tropospheric OH-chemistry parameterizations. Both CH4 and N2O lifetime includes a dependency of its lifetime on its own abundance. Further details regarding how these processes are treated in MAGICC6 can be found in Meinshausen et al. (2011a). Stratospheric sinks for all fluorinated gases and ODSs are assumed to become enhanced slightly with rising global mean temperatures (15% per degree Celsius warming) leading to shorter lifetimes at higher warming levels—due to a strengthening Brewer-Dobson circulation (Butchart and Scaife 2001). As the lifetime of CH4, tropospheric OH-related sinks of flourinated gases are made dependent on the parameterized changes in the OH abundances.

Previous applications of MAGICC6 have not used a single set of best-estimate parameters to calculate concentrations, radiative forcings and global mean temperatures from prescribed emissions, but instead used an array of parameter sets to project a range of climate responses by emulating CMIP3 GCMs (Meehl et al. 2005) and to project a range of CO2 concentrations by emulating C4MIP carbon cycle models (Friedlingstein et al. 2006) as described in Meinshausen et al. (2011a). Alternatively, historically constrained joint distributions of parameters were used (Meinshausen et al. 2009). To calculate a single set of concentrations for driving climate models (the motivation for the design of the RCPs), we need a single “best” set of model parameters for MAGICC. Many temperature projections will ultimately be produced by the CMIP5 models. However, the importance of having a ‘best-estimate’ future temperature projection within MAGICC6 is that many gases’ concentrations are influenced by temperature or climate feedbacks.

We chose parameters for MAGICC6 that would closely reflect the median of the distribution in global-mean temperature projections that is spanned when emulating the GCMs and carbon cycle models that took part in CMIP3 and C4MIP, respectively. Specifically, we chose an emulation of the C4MIP Bern-CC carbon cycle model (Joos et al. 2001) as ‘best-estimate’ for the carbon cycle behavior. This is both because the Bern-CC model (and earlier versions of this model) has been used for the consolidated concentrations of IPCC SRES scenarios presented in the Third Assessment Report (see Appendix II in Houghton et al. 2001) and because the projected CO2 concentrations from the Bern model are roughly in the middle of the range of C4MIP results (Friedlingstein et al. 2006).

For obtaining a ‘best-estimate’ climate response (which in turn influences concentrations), the intention is to have a set of climate parameters (such as climate sensitivity, vertical ocean diffusivity, etc.) that resembles the median of the AOGCMs that took part in CMIP3. We first emulated 19 of the CMIP3 AOGCMs by using calibrated MAGICC parameters, which span a climate sensitivity range between 1.9 K and 5.7 K (see Table B3 in Meinshausen et al. 2011a). We ran these emulations for the SRES A1B, B1 and A2 and ‘Constant year 2000 concentration’ scenarios, taking into account a complete set of radiative forcing agents (including e.g. indirect aerosol effects). In these emulations, we used the default Bern-CC emulation setting for the MAGICC carbon cycle. We then took the median for each scenario across these 19 global mean temperatures and ocean heat uptake time-series. Thereby, we created pseudo-AOGCM datasets, to which a standard least-squares optimization routine could calibrate a “best” set of 10 climate parameters of MAGICC. This is the same procedure as described in Meinshausen et al.(2011a), except that we fix the climate sensitivity at its best-estimate value of 3 K (Meehl et al. 2007; Knutti and Hegerl 2008). This fixed climate sensitivity is very close to the average of 2.88 K from emulating AOGCM CMIP3 models without a fixed climate sensitivity (see Table 4 in Meinshausen et al. 2011a).

Apart from deriving the default MAGICC climate response parameter set for creating the RCPs, one additional amendment has been implemented in MAGICC6 to serve this RCP process. Inverse emissions are now routinely calculated for all 31 considered GHGs (CO2, CH4, N2O, 3 PFCs, 8 HFCs, SF6 and 16 ODSs). This is of interest when concentration time-series are prescribed for designing the ECPs. For CO2, these inverse emissions can be compared to the allowable emissions derived from ESMs in the course of CMIP5.

Post-2100 extension

The RCP emissions scenarios produced by the IAMs span the period 2005 to 2100. Most IAM research focuses on the 21st century and there is little research addressing comprehensive scenarios beyond 2100. Conducting multi-century climate change analyses, however, requires input data beyond 2100. We designed extensions to the RCPs, the Extended Concentration Pathways (ECPs), using simple rules that have been developed in a series of stakeholder consultations among scientists in the IPCC WG1, WG2 and WG3 communities, including representatives of the IAM groups, representatives of CMIP5, and the wider scientific community. It should be noted that these extensions do not represent fully consistent scenarios, but are simple ‘what-if’ thought experiments that represent the underlying ideas behind each RCP, which are produced for the purposes of providing a common set of input data for long-term model comparison projects. The rules used for the extension of emissions and/or concentrations of the RCPs are shown in Table 3. No explicit assumptions relating to population or economic development have been made for these stylized ‘what-if’ extensions.

Table 3 The RCPs and their simple extension rules beyond 2100 assumed for all GHGs.

Generally, there are three options for the design of a simple extension (with multiple combinations of these possible): I) the forcing and concentrations can simply be kept constant (as done for the SRES scenarios in CMIP3 and assessed by the IPCC Forth Assessment Report, AR4), II) emissions can be adapted over time, e.g. to yield a smooth stabilization at another level than where concentrations are in year 2100, or III) emissions can be kept constant. The two intermediate scenarios RCP4.5 and RCP6 are extended by concentration stabilization, albeit with stabilization achieved in 2150 to avoid discontinuities in the implied emissions. For the lowermost RCP, RCP2.6, emissions were kept constant after 2100. The extension of the highest RCP, RCP8.5, represents a mixture of constant emissions until 2150 and constant concentrations after 2250 (see Table 3). As a result of now including a low mitigation pathway RCP2.6 and due to these extension choices, the ECPs span a much wider range of post-2100 forcing pathways than considered in previous studies, such as CMIP3 assessed in IPCC AR4.

Several alternative extensions were considered for each RCP. For RCP8.5, the full range of possible extensions ranged from a constant forcing that results from simply keeping concentrations constant after 2100 to very high levels that result from assuming that emissions stay constant until 2300. After consultations with the respective expert groups an intermediate extension was selected. This extension avoids a possible discontinuity in emissions trends (that would arise from keeping concentrations constant), and avoids issues of resource availability that a higher extension might raise. Keeping emissions constant would have resulted in CO2 concentration of around 3000 ppm by 2300. The adopted RCP8.5 extension (ECP8.5) leads to a CO2 stabilization after 2250 at roughly 2000 ppm, or more than 7-times pre-industrial CO2 concentrations. The total forcing of this ECP8.5 is hence approximately twice as high as the next highest ECP (ECP6) (see Fig. 4 below). This high forcing for ECP8.5 is significantly above the highest forcing level that was considered in CMIP3 on the basis of IPCC SRES scenarios (i.e., approximately 700 ppm in A1B).

Another alternative extension was considered for RCP2.6, as one may question whether negative emissions can be sustained over very long time periods (in view of finite CO2 storage capacity). An alternative extension, in which emissions would converge back to zero between 2,150 and 2,200, would lead to CO2 concentration of about 380 ppm rather than 360 ppm by 2300. However, in consultation with experts it was concluded that a continuation of the 2100 emission levels cannot be excluded for reasons of physical constraints of sequestration & storage options and would better reflect the character of the RCP2.6 pathway.

In addition to the four standard ECP extensions, we present one supplementary extension, which might be of particular use to investigate irreversibility and the path-dependency of the climate system response to different GHG abundances.Footnote 5 This extension starts from RCP6 in 2100 and merges with the concentrations of the next lower scenario, ECP4.5, by 2250. This is similar to other forcing and temperature overshoot scenarios in the literature (Hare and Meinshausen 2006; Wigley et al. 2007; Lowe et al. 2009). Between 2100 and 2250, we adjusted emissions of this supplementary extension (called SCP6to4.5) by following simple linear and continuous trajectories in order to obtain the desired stabilization concentration level of ECP4.5 by 2250. We chose for transparency linear segments of emissions, i.e., a 50-year long phase of stringent reductions (with an annual reduction of fossil CO2 emissions equivalent to 2.5%/year of 2100 levels) to reach a negative emission floor, which had to be more than twice as negative as under the RCP2.6 scenario (-3.8 GtC/year). Smoother emission trajectories would be possible, although they would imply higher annual reduction rates and/or more negative emission levels than presented here. After 2250, when concentrations are equal by design, the implied inverse CO2 emission levels of the SCP6to4.5 overshoot pathway are consistently lower than those of ECP4.5 (see Fig. 3). This is because of temperature-induced feedbacks and inertia effects in the carbon cycle. In summary, only by assuming a long period of strong reductions and deeply negative CO2 emissions, were we able to ‘make up’ for the higher RCP6 emissions during the 21st century in order to reach ECP4.5 concentration levels by 2250.

Fig. 3
figure3

Emissions for the four RCPs and the supplementary extension SCP6to4.5, which starts from the RCP6 scenario and merges with the ECP4.5 concentrations by 2250. The shaded areas denote times of higher emissions (grey shading) and compensating lower emissions (beige shading)

Similar to this SCP6to4.5 extension, we considered an alternative ECP6 extension with comparatively less stringent post-2100 emission reductions, basically assuming that the RCP2.6 emissions path is followed 100 years later, i.e., in the 22nd century. Specifically, 2020–2100 RCP2.6 emission trajectories were assumed for 2120 to 2200—with linear interpolation between 2100 RCP6 and the extension by 2120. We kept emissions constant after 2200. Following that alternative, radiative forcing would have declined to 4.5 W/m2 only 350 years later, i.e., by 2450.

The conclusion from this alternative RCP6 extension and our SCP6to4.5 supplementary extension is that once high 21st century emissions increase radiative forcing levels to 6 W/m2, it seems very difficult to return to lower levels quickly. Any 4.5 W/m2 overshoot scenario of the sort considered here will imply higher global warming levels for considerable periods of time, i.e., centuries,—and rests on the assumption that stringent post-2100 emission reductions are feasible. A similar conclusion could be drawn for extensions that connect RCP8.5 to ECP6 or RCP4.5 to ECP3-PD long-term concentration levels5.

For the fluorinated gases within the basket of gases that are controlled under the Kyoto Protocol (HFCs, PFCs, and SF6), projections are inherently uncertain, as new applications are constantly being developed (apart from the fact that new fluorinated compounds are designed). For the purpose of designing the SCP6to4.5 extension, a question arises how to bridge the gap between the lower ECP6 fluorinated gas concentration levels towards the higher ECP4.5 ones. Partly because this gap is very small, only 2% in terms of the total radiative forcing difference between ECP6 and ECP4.5, and partly for simplicity, we modify only the emissions of the representative forcing agent HFC-134a in order to ramp up the aggregate forcing from HFCs, PFCs, and SF6.

Results: RCP GHG concentrations

This section presents the resulting GHG concentrations for the RCPs and ECPs, as well as the aggregate total radiative forcing. Total radiative forcing here includes all anthropogenic forcing agents as listed in Table 2.12 in IPCC AR4 WG1 (Forster et al. 2007), including—inter alia—direct and indirect aerosol forcings. See Appendix 2 for assumptions in regard to non-GHG forcings.

In the lowest of the four RCPs, the total radiative forcing peaks at approximately 3 W/m2 and declines thereafter (Fig. 4), (motivating the alternative name for the RCP2.6 pathway, which is RCP3-PD, where PD stands for “Peak&Decline”). CO2 concentrations reach a maximum level of slightly above 440 ppm in the year 2050, and then decline to below today’s levels by 2300 (~360 ppm) (see Table 4). CO2-equivalent concentrations (not shown),Footnote 6 comprising the net effect of all anthropogenic forcing agents (including aerosols), peak at around 460 ppm just before 2050, declining in tandem with CO2 towards 360 ppm by 2300, at which time this scenario’s projection for the net effect of non-CO2 forcing agents is close to zero—similar to the best-estimate non-CO2 forcing estimate for current times (Forster et al. 2007). The aggregate forcing of all long-lived GHGs controlled under the Kyoto-Protocol, expressed as CO2 equivalent, declines from 503 ppm CO2eq in 2050 towards 407 ppm by 2300 in RCP2.6.

Fig. 4
figure4

Total radiative forcing (anthropogenic plus natural) for RCPs,—supporting the original names of the four pathways as there is a close match between peaking, stabilization and 2100 levels for RCP2.6 (called as well RCP3-PD), RCP4.5 & RCP6, as well as RCP8.5, respectively. Note that the stated radiative forcing levels refer to the illustrative default median estimates only. There is substantial uncertainty in current and future radiative forcing levels. Short-term variations in radiative forcing are due to both volcanic forcings in the past (1800–2000) and cyclical solar forcing—assuming a constant 11-year solar cycle (following the CMIP5 recommendation), except at times of stabilization

Table 4 GHG concentrations for pre-industrial, historical, RCP and ECP/SCPs

RCP8.5’s radiative forcing levels by the end of 2100 are around 8.5 W/m2 under our ‘best-estimate’ set of model parameters with forcing levels increasing further thereafter—up to 12 W/m2 by 2250, when concentrations stabilize (Fig. 4). Transient scenarios with such high radiative forcing levels and CO2 concentrations have never before been investigated in model CMIP intercomparison exercises. Across almost all gases, RCP8.5 concentration levels are by far the highest. For example, CH4 concentration stabilizes around 3500 ppb—more than twice as high as the next highest scenarios, RCP4.5 and RCP6, which approximately stabilize at 1,500 ppb (slightly below today’s levels of nearly 1800ppb). The only exceptions are ODS concentrations: RCP4.5, RCP6 and RCP8.5 share the same emission assumptions (WMO 2007, A1 Scenario), but the longer-term ODS concentrations are slightly lower in RCP8.5 due to slight decreases in stratospheric ODS lifetimes via expected changes in stratospheric circulation rates, outweighing initial decreases in tropospheric sinks due to lower OH concentrations (see Section 2.4 and Fig. 5d).

Fig. 5
figure5

GHG concentrations recommended for the CMIP5 climate change research studies. Shown are: a atmospheric CO2; b methane; c nitrous oxide; d (equivalent) CFC-12; and e (equivalent) HFC134a concentrations. Equivalent concentrations are derived so as to equal the aggregate forcing of the represented forcing agents. For CFC-12, those species that are controlled under the Montreal Protocol are aggregated.. For HFC-134a, the gases aggregated are the fluorinated gases controlled under the Kyoto Protocol. Note that this aggregation is based on radiative forcing equivalence, i.e. aggregation is not based on GWPs (annual data for each individual gas is provided on-line)

RCP4.5 and RCP6 are both stabilization scenarios, with constant concentrations after 2150. By stabilizing CO2 concentrations at 543 ppm, RCP4.5 comes very close to a doubling of pre-industrial CO2 concentration (278 ppm)—and is hence only slightly higher than the SRES B1 scenario and its constant extension after 2100 with 540 ppm CO2 (see Bern-CC (reference) case in Appendix II.2.1 in Houghton et al. 2001). The RCP6 scenario stabilizes 200 ppm higher, at 752 ppm CO2 (see Fig. 5).

Discussion

Ensemble results compared to our default concentration and temperature projections

In the above text we selected a specific (‘best-estimate’) set of MAGICC parameters to use in producing a standard set of RCP concentrations. Starting from the harmonized emissions, we can also produce concentrations (and forcing and temperature projections) using 19 individual CMIP3 climate and 9 C4MIP carbon cycle emulations. How does our default set of results compare with the distribution of results from these 171 (=19 × 9) cases?

We perform this comparison using the highest and the lowest RCP scenarios. The results are shown in Fig. 6. Not surprisingly, because the responses to external forcings in all climate models are largely linear, the ‘best-estimate’ results are similar to the median of the individual model results, even in the high forcing RCP8.5 case. The ideal test of our projections, although impractical, would be for the CMIP3 GCM model ensemble to be run again for the RCP8.5 scenario and ECP8.5 extension. Since the post-2100 ECP8.5 forcing case falls well outside the MAGICC SRES calibration range, it would provide a rather stringent test for MAGICC.

Fig. 6
figure6

CO2 concentrations (a), total radiative forcing (natural and anthropogenic) (b) and global mean surface temperatures (c) for RCP2.6 (called as well RCP3-PD) and RCP8.5 (solid lines) compared to the full range of CMIP3 GCM and C4MIP carbon cycle model emulations (shaded areas). The small temporal variations in forcing are caused by the 11-year solar cycle assumption, influencing as well temperature projections

In summary, our default carbon cycle/climate model settings are shown to reflect well the median emulated responses of both the AOGCM and carbon cycle response. The implied relation between emissions and concentrations may, of course, change with the next generation of models. It will be useful to compare the climate and carbon-cycle responses from the next round of climate model experiments with those presented here, which are, necessarily, calibrated to existing model experiments. The allowable CO2 emissions for these pathways could, for example, be lower (or higher) than those found here, if positive carbon cycle feedbacks were higher (or lower) in the CMIP5 generations of ESMs compared with C4MIP models. Similarly, including a nitrogen cycle in the MAGICC and other carbon cycle models will affect future forward and inverse projections (Thornton et al. 2007, 2009).

Inverse emissions for the long-term extension pathways

The extensions of the RCPs and the generation of the supplementary pathway were made using the simple rules given in Table 3. While these extensions are simply thought experiments, we examine here the implied emissions from the SCP6to4.5 scenario to illustrate some key points relevant to future long-term scenarios. The inverse emissions that correspond to this overshoot scenario imply a period of substantial net negative CO2 emissions of -13.9 GtCO2/year—between 2150 and 2230 (see Fig. 3). This value is larger in magnitude than the negative emissions level in the RCP2.6 scenario (−0.93 GtC/yr = −3.41GtCO2/yr), but there are scenarios in the literature that imply similarly negative emission levels already by 2100. For example, Calvin et al. (2009) show scenarios with slightly larger negative emissions levels by 2100 of -10 to -15 GtCO2/year in 2095. In addition to sustained net negative emissions in SCP6to4.5, the reduction rates assumed between 2100 and 2150 are rather substantial, similar to the steepest segments of RCP4.5 and RCP2.6 – but extended over a longer time period.

As highlighted in Fig. 3a, CO2 emissions for the SCP6to4.5 have to be reduced below the RCP4.5 emissions by approximately the same amount as the extra cumulative emissions that were emitted during the 21st century by RCP6. On the timescales of interest here, CO2 does not have finite lifetime (Archer and Brovkin 2008), but is simply being re-distributed between the different active carbon pools. This is largely why cumulative CO2 emissions are a crucial quantity for long-term temperature and atmospheric concentration responses (Kheshgi et al. 2005; Allen et al. 2009; Matthews et al. 2009). For this reason, cumulative emissions for the RCP4.5-ECP4.5 pathway and the RCP6-SCP6to4.5 pathway are roughly of the same size (see Fig. 3b).

For other long-lived GHGs such as N2O, the need to compensate for initially higher emissions with lower emissions later on is still apparent, although cumulative N2O emissions are a bit higher under the SCP6to4.5 pathway compared to RCP4.5-ECP4.5 —even though both paths ultimately stabilize at the same N2O concentration levels by design (see Fig. 3 e&f). For CH4, with its much shorter atmospheric lifetime (Forster et al. 2007), the ultimate concentration depends almost solely on emissions over the final few decades rather than on long-term cumulative emissions.

Limitations

The challenge this study faces is to synthesize gas-cycle and climate response characteristics from a broad body of literature into a single-best set of data and model parameters to produce a common starting point for future model intercomparison exercises. By design, we therefore do not intend to fully represent uncertainties, but rather seek to produce concentration scenarios that reflect ‘middle-of-the-road’ carbon cycle and climate model responses, representative of the IPCC AR4 state of knowledge. While some parts of our approach can thereby be based on recent intercomparison exercises, in particular C4MIP and CMIP3, other assumptions relate to earlier community efforts, such as the OxComp workshop (Ehhalt et al. 2001). Uncertainties pertaining to future OH concentrations, CH4 lifetimes, N2O concentrations or the effect of the inclusion of a nitrogen cycle will result in different ‘reference’ pathway recommendations if this exercise were to be repeated in a couple of years from now. Furthermore, uncertainties arise as well from second-order effects of tropospheric ozone and aerosols, for example. To the extent that our non-GHG modeling assumptions deviate from current generation chemistry modeling results (see e.g. Lamarque in this issue for a comparison of tropospheric ozone levels), indirect effects via global-mean temperature and gas-cycle feedbacks will impact derived GHG concentrations presented here, although this indirect second-order effect is likely limited.

Regarding the RCP extensions, we reiterate that these ECPs are highly stylized and are not the result of detailed analyses of resource limits, but are instead presented in order to provide a range of concentration and forcing pathways for use in climate model experiments. In contrast to the simple stabilization of concentrations after 2100 used for CMIP3, the current ECPs provide a wider range of forcing pathways in which to test long-term model responses.

Conclusions

The historical, 21st century and extended GHG concentration and harmonized emission data presented here are a result of a wide collaboration across scientific communities. In order to obtain a single set of best-estimate projections for future GHG concentrations for the four RCP scenarios, we used an emulation of the median response of both climate and carbon cycle models that took part in the previous climate model inter-comparison exercises CMIP3 (Meehl et al. 2005) and C4MIP (Friedlingstein et al. 2006). The derived concentration results are generated as the starting point for the CMIP5 inter-comparison exercise (Taylor et al. 2009), which will be evaluated in the forthcoming IPCC Fifth Assessment Report. While the inverse emission results here can also provide a yardstick for comparison, it can be expected that the forthcoming CMIP5 generation of ESMs will diagnose ‘allowable’/inverse emissions that differ from the harmonized emissions presented here. This is because of limitations in the extrapolation of previous model's emulation results, and partly because of new process parameterizations within the new generation of climate and carbon cycle models.

While the provision of concentration pathways for use in model inter-comparison exercises is the end point for the RCP scenario production activity, this is still just the beginning of the overall scientific effort to investigate climate change, its impacts and mitigation options under different scenarios (Moss et al. 2010). With respect to the IPCC WG1 community, the harmonized concentrations and emissions data provided here are a starting point for model experiments that will enable a deeper understanding of the earth systems’ response to the anthropogenic perturbations that are driving climate changes.

Notes

  1. 1.

    A note on the naming of the lower RCP pathway: In the literature, this lower pathway is both called RCP3-PD and RCP2.6. “PD” reflects the unique characteristic of this pathway, i.e., its “Peak & Decline” shape—in contrast to other mere stabilization scenarios. Historically, two candidates were discussed for this lower RCP, RCP2.6 and RCP2.9 with RCP2.6 being finally chosen. Hence, both names, RCP3-PD and RCP2.6, can be used interchangeably.

  2. 2.

    A research version of MAGICC6 with RCP default settings used in this study will be available from www.magicc.org for the wider scientific community.

  3. 3.

    RCP6 was harmonized as well, but its values were not used to determine the growth rates (or other averages for harmonization) due to the later finalization date of this data set. Taking into account RCP6 does not substantially change these results relative to other uncertainties.

  4. 4.

    1GtC/yr = 44/12 GtCO2/yr

  5. 5.

    An optional extra extension is provided online as addition to the RCPs and extensions described in this paper, i.e., an extension that brings concentrations back to RCP3-PD levels after 21st century emissions followed RCP4.5. See http://www.pik-potsdam.de/~mmalte/rcps/.

  6. 6.

    available at http://www.pik-potsdam.de/~mmalte/rcps/.

References

  1. Allen MR, Frame DJ, Huntingford C, Jones CD, Lowe JA, Meinshausen M, Meinshausen N (2009) Warming caused by cumulative carbon emissions towards the trillionth tonne. Nature 458(7242):1163

    Article  Google Scholar 

  2. Archer D, Brovkin V (2008) The millennial atmospheric lifetime of anthropogenic CO2. Clim Chang 90(3):283–297

    Article  Google Scholar 

  3. Butchart N, Scaife AA (2001) Removal of chlorofluorocarbons by increased mass exchange between the stratosphere and troposphere in a changing climate. Nature 410(6830):799–802

    Article  Google Scholar 

  4. Butler JH, Battle M, Bender ML, Montzka SA, Clarke AD, Saltzman ES, Sucher CM, Severinghaus JP, Elkins JW (1999) A record of atmospheric halocarbons during the twentieth century from polar firn air. Nature 399(6738):749–755

    Article  Google Scholar 

  5. Calvin KV, Edmonds J, Bond-Lamberty B, Clarke LE, Kim SH, Kyle GP, Smith SJ, Thomson AM, Wise MA (2009) 2.6: Limiting climate change to 450 ppm CO2 Equivalent in the 21st Century. Energy Econ 31:S107–S120

    Article  Google Scholar 

  6. Canadell JG, Le Quere C, Raupach MR, Field CB, Buitenhuis ET, Ciais P, Conway TJ, Gillett NP, Houghton RA, Marland G (2007) Contributions to accelerating atmospheric CO2 growth from economic activity, carbon intensity, and efficiency of natural sinks. Proc Natl Acad Sci USA 104(47):18866–18870

    Article  Google Scholar 

  7. Clarke L, Edmonds J, Jacoby H, Pitcher H, Reilly J, Richels R (2007) Scenarios of the greenhouse gas emission and atmospheric concentrations. Sub-report 2.1A of Synthesis and Assessment Product 2.1 by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research. Washington, USA, Department of Energy, Office of Biological & Environmental Research, p 154

  8. Clerbaux C, Cunnold D, Anderson J, Bernath P, Engel A, Fraser PJ, Mahieu E, Manning AC, Miller J, Montzka SA, Prinn R, Reimann S, Rinsland CP, Simmonds P, Verdonik D, Wuebbles D, Yokouchi Y (2007) Long-lived compounds, Chapter 2. Scientific Assessment of Ozone Depletion: 2006, Global Ozone Research and Monitoring Project - Report No. 50. World Meteorological Organization. Geneva

  9. Conway TJ, Tans PP, Waterman LS, Thoning KW (1994) Evidence for interannual variability of the carbon-cycle from the national-oceanic-and-atmospheric-administration climate-monitoring-and-diagnostics-laboratory global-air-sampling-network. J Geophys Res Atmos 99(D11):22831–22855

    Article  Google Scholar 

  10. Culbertson JA, Prins JM, Grimsrud EP, Rasmussen RA, Khalil MAK, Shearer MJ (2004) Observed trends for CF3-containing compounds in background air at Cape Meares, Oregon, Point Barrow, Alaska, and Palmer Station, Antarctica. Chemosphere 55(8):1109–1119

    Article  Google Scholar 

  11. Daniel JS, Velders GJM, Douglass AR, Forster PMD, Hauglustaine DA, Isaksen ISA, Kuijpers LJM, McCulloch A, Wallington TJ (2007) Halocarbon scenarios, ozone depletion potentials, and global warming potentials, Chapter 8. Scientific Assessment of Ozone Depletion: 2006, Global Ozone Research and Monitoring Project - Report No. 50. World Meteorological Organization, Geneva

  12. Deeds DA, Muhle J, Weiss RF (2008) Tetrafluoromethane in the deep North Pacific Ocean. Geophys Res Lett 35(14)

  13. DeFries RS, Houghton RA, Hansen MC, Field CB, Skole D, Townshend J (2002) Carbon emissions from tropical deforestation and regrowth based on satellite observations for the 1980s and 1990s. Proc Natl Acad Sci USA 99(22):14256–14261

    Article  Google Scholar 

  14. Dlugokencky EJ, Steele LP, Lang PM, Masarie KA (1994) The growth-rate and distribution of atmospheric methane. J Geophys Res Atmos 99(D8):17021–17043

    Article  Google Scholar 

  15. EC-JRC and PBL (2009) European Commission, Joint Research Centre (JRC)/Netherlands Environmental Assessment Agency (PBL)—Emission Database for Global Atmospheric Research (EDGAR), release version 4.0. from http://edgar.jrc.ec.europa.eu

  16. Ehhalt D, Prather MJ, Dentener F, Derwent RG, Dlugokencky E, Holland E, Isaksen ISA, Katima J, Kirchhoff V, Matson P, Midgley P, Wang M (2001) Atmospheric chemistry and greenhouse gases. In: Houghton JT, Ding Y, Griggs DJ et al (eds) Climate change 2001: The scientific basis. Cambridge University Press, Cambridge, p 892

    Google Scholar 

  17. Etheridge DM, Steele LP, Francey RJ, Langenfelds RL (1998a) Atmospheric methane between 1000 AD and present: evidence of anthropogenic emissions and climatic variability. J Geophys Res Atmos 103(D13):15979–15993

    Article  Google Scholar 

  18. Etheridge DM, Steele LP, Langenfelds RL, Francey RJ, Barnola JM, Morgan VI (1998b) Historical CO2 record from the Law Dome DE08, DE08-2, and DSS ice cores. Retrieved May, 2007, from ftp://ftp.ncdc.noaa.gov/pub/data/paleo/icecore/antarctica/law/law_co2.txt

  19. Fenhann J (2000) Industrial non-energy, non-CO2 greenhouse gas emissions. Technol Forecast Soc Chang 63(2–3):313–334

    Article  Google Scholar 

  20. Fluckiger J, Monnin E, Stauffer B, Schwander J, Stocker TF, Chappellaz J, Raynaud D, Barnola JM (2002) High-resolution Holocene N2O ice core record and its relationship with CH4 and CO2. Global Biogeochem Cy 16(1)

  21. Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R, Fahey DW, Haywood J, Lean J, Lowe DC, Myhre G, Nganga J, Prinn R, Raga G, Schulz M, Van Dorland R (2007). Chapter 2: Changes in Atmospheric Constituents and in Radiative Forcing. IPCC Fourth Assessment Report WG 1. IPCC. Cambridge, Cambridge University Press

  22. 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 HD, Raddatz T, Rayner P, Reick C, Roeckner E, Schnitzler K-G, Schnur R, Strassmann K, Weaver K, Yoshikawa C, Zeng N (2006) Climate–carbon cycle feedback analysis: results from the C4MIP model intercomparison. J Clim 19(14):3337–3353

    Article  Google Scholar 

  23. Fujino J, Nair R, Kainuma M, Masui T, Matsuoka Y (2006) Multi-gas mitigation analysis on stabilization scenarios using aim global model. Energ J 343–353

  24. Geller LS, Elkins JW, Lobert JM, Clarke AD, Hurst DF, Butler JH, Myers RC (1997) Tropospheric SF6: observed latitudinal distribution and trends, derived emissions and interhemispheric exchange time. Geophys Res Lett 24(6):675–678

    Article  Google Scholar 

  25. Granier C, Bessagnet B, Bond T, D’Angiola A, van der Gon HG, Frost G, Heil A, Kainuma M, Kaiser J, Kinne S, Klimont Z, Kloster S, Lamarque JF, Liousse C, Matsui T, Meleux F, Mieville A, Ohara T, Riahi K, Schultz M, Smith S, Thomson AM, van Aardenne J, van der Werf G (2011). Evolution of anthropogenic and biomass burning emissions at global and regional scales during the 1980–2010 period. Climatic Change (this issue). doi:10.1007/s10584-011-0154-1

  26. Hare B, Meinshausen M (2006) How much warming are we committed to and how much can be avoided? Clim Chang 75(1):111–149

    Article  Google Scholar 

  27. Hijioka Y, Matsuoka Y, Nishimoto H, Masui M, Kainuma M (2008) Global GHG emissions scenarios under GHG concentration stabilization targets. J Global Environ Eng 13:97–108

    Google Scholar 

  28. Houghton RA (2008) Carbon flux to the atmosphere from land-use changes: 1850–2005. In TRENDS: A compendium of data on global change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., U.S.A. from http://cdiac.ornl.gov/trends/landuse/houghton/houghton.html

  29. Houghton J, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Xiaosu D (eds) (2001) Climate change 2001: The scientific basis; contribution of working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change (IPCC)ge. Cambridge University Press, Cambridge

    Google Scholar 

  30. Hurtt G, Chini L, Frolking S, Betts R, Edmonds J, Feddema J, Fisher G, Goldewijk K, Hibbard KA, Houghton R, Janetos A, Jones C, Kinderman G, Konoshita T, Riahi K, Shevliakova E, Smith S, Stehfest E, Thomson A, Thornton P, van Vuuren DP, Wang Y (2011). Land use Change and earth system dynamics. Climatic Change (this issue). doi:10.1007/s10584-011-0153-2

  31. IPCC (1996) Climate change 1995: The science of climate change. Contribution of WGI to the Second Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge

    Google Scholar 

  32. Joos F, Prentice IC, Sitch S, Meyer R, Hooss G, Plattner G-K, Gerber S, Hasselmann K (2001) Global Warming feedbacks on terrestrial carbon uptake under the Intergovernmental Panel on Climate Change (IPCC) emission scenarios. Global Biogeochem Cycles 15(4):891–907

    Article  Google Scholar 

  33. Keeling CD, Whorf TP (2004) Atmospheric CO2 records from sites in the SIO air sampling network. Retrieved May, 2007, from http://cdiac.esd.ornl.gov/trends/co2/sio-keel.htm

  34. Kheshgi HS, Smith SJ, Edmonds JA (2005) Emissions and atmospheric CO2 stabilization: long-term limits and paths. Mitig Adapt Strateg Glob Chang 10(2):213

    Article  Google Scholar 

  35. Knutti R, Hegerl GC (2008) The equilibrium sensitivity of the Earth’s temperature to radiation changes. Nat Geosci 1(11):735–743

    Article  Google Scholar 

  36. Lamarque JF, Bond TC, Eyring V, Granier C, Heil A, Klimont Z, Lee D, Liousse C, Mieville A, Owen B, Schultz MG, Shindell D, Smith SJ, Stehfest E, Van Aardenne J, Cooper OR, Kainuma M, Mahowald N, McConnell JR, Naik V, Riahi K, van Vuuren DP (2010) Historical (1850–2000) gridded anthropogenic and biomass burning emissions of reactive gases and aerosols: methodology and application. Atmos Chem Phys 10(15):7017

    Article  Google Scholar 

  37. Lamarque JF, Riahi K, Smith S, van Vuuren DP, Vitt F, Meinshausen M (2011) Simulated evolution of the distribution of short-lived greenhouse gases and aerosols using the emissions from the Representative Concentration Pathways. Climatic Change (this issue). doi:10.1007/s10584-011-0155-0

  38. Laube JC, Martinerie P, Witrant E, Blunier T, Schwander J, Brenninkmeijer CAM, Schuck TJ, Bolder M, Rockmann T, van der Veen C, Bonisch H, Engel A, Mills GP, Newland MJ, Oram DE, Reeves CE, Sturges WT (2010) Accelerating growth of HFC-227ea (1,1,1,2,3,3,3-heptafluoropropane) in the atmosphere. Atmos Chem Phys 10(13):5903–5910

    Article  Google Scholar 

  39. Lean JL, Rind DH (2009) How will Earth’s surface temperature change in future decades? Geophys Res Lett 36

  40. Lowe JA, Huntingford C, Raper SCB, Jones CD, Liddicoat SK, Gohar LK (2009) How difficult is it to recover from dangerous levels of global warming? Environ Res Lett 4(1)

  41. Machida T, Nakazawa T, Fujii Y, Aoki S, Watanabe O (1995) Increase in the atmospheric nitrous-oxide concentration during the last 250 years. Geophys Res Lett 22(21):2921–2924

    Article  Google Scholar 

  42. Marland G, Boden TA, Andres RJ (2008) Global, regional, and national fossil fuel CO2 emissions. In Trends: A Compendium of Data on Global Change. O. R. N. L. Carbon Dioxide Information Analysis Center, U.S. Department of Energy. Oak Ridge, Tenn., USA

  43. Masui T, Matsumoto K, Hijioka Y, Kinoshita T, Nozawa T, Ishiwatari S, Kato E, Shukla PR, Yamagata Y, Kainuma M (2011). An Emission Pathway for Stabilization at 6 W/m2 Radiative Forcing. Climatic Change doi:10.1007/s10584-011-0150-5

  44. Matthews HD, Gillett NP, Stott PA, Zickfeld K (2009) The proportionality of global warming to cumulative carbon emissions. Nature 459(7248):829

    Article  Google Scholar 

  45. Meehl GA, Covey C, McAvaney B, Latif M, Stouffer RJ (2005) Overview of coupled model intercomparison project. Bull Am Meteorol Soc (BAMS) 86(89)

  46. Meehl GA, Stocker TF, Collins W, Friedlingstein P, Gaye A, Gregory JM, Kitoh A, Knutti R, Murphy J, Noda A, Raper SCB, Watterson I, Weaver A, Zhao Z-C (2007). Chapter 10: Global climate projections. IPCC Fourth Assessment Report. IPCC. Cambridge, Cambridge University Press

  47. Meinshausen M, Meinshausen N, Hare W, Raper SCB, Frieler K, Knutti R, Frame DJ, Allen MR (2009) Greenhouse-gas emission targets for limiting global warming to 2°C. Nature 458(7242):1158

    Article  Google Scholar 

  48. Meinshausen M, Raper SCB, Wigley TML (2011a) Emulating coupled atmosphere-ocean and carbon cycle models with a simpler model, MAGICC6: Part I—model description and calibration. Atmos Chem Phys 11:1417–1456

    Article  Google Scholar 

  49. Meinshausen M, Wigley TML, Raper SCB (2011b) Emulating atmosphere-ocean and carbon cycle models with a simpler model, MAGICC6: Part 2—Applications. Atmos Chem Phys 11:1457–1471

    Article  Google Scholar 

  50. Metz B, Kuijpers L, Solomon S, Anderson SO, Davidson O, Pons J, de Jager D, Kestin T, Manning M, Meyer L (eds) (2005) IPCC/TEAP special report on safeguarding the ozone layer and the global climate system: Issues related to hydrofluorocarbons and perfluorocarbons. Cambridge University Press, New York

    Google Scholar 

  51. Montzka SA, Butler JH, Myers RC, Thompson TM, Swanson TH, Clarke AD, Lock LT, Elkins JW (1996a) Decline in the tropospheric abundance of halogen from halocarbons: implications for stratospheric ozone depletion. Science 272(5266):1318–1322

    Article  Google Scholar 

  52. Montzka SA, Myers RC, Butler JH, Elkins JW, Lock LT, Clarke AD, Goldstein AH (1996b) Observations of HFC-134a in the remote troposphere. Geophys Res Lett 23(2):169–172

    Article  Google Scholar 

  53. Moss R, Babiker M, Brinkman S., Calvo E, Carter TR, Edmonds J, Elgizouli I, Emori S, Erda L, Hibbard K, Jones R, Kainuma M, Kelleher J, Lamarque J-F, Manning MR, Matthews B, Meehl J, Meyer L, Mitchell JFB, Nakicenovic N, O’Neill B, Pichs R, Riahi K, Rose SK, Runci P, Stouffer RJ, van Vuuren DP, Weyant JP, Wilbanks TJ, van Ypersele JP, Zurek M (2008) Towards new scenarios for analysis of emissions, climate change, impacts, and response strategies: IPCC Expert Meeting Report, 19–21 September 2007. Noordwijkerhout, The Netherlands, pp 155

  54. Moss RH, Edmonds JA, Hibbard KA, Manning MR, Rose SK, van Vuuren DP, Carter TR, Emori S, Kainuma M, Kram T, Meehl GA, Mitchell JFB, Nakicenovic N, Riahi K, Smith SJ, Stouffer RJ, Thomson AM, Weyant JP, Wilbanks TJ (2010) The next generation of scenarios for climate change research and assessment. Nature 463(7282):747–756

    Article  Google Scholar 

  55. Nakicenovic N, Swart R (eds) (2000) IPCC special report on emissions scenarios. Cambridge University Press, Cambridge

    Google Scholar 

  56. Olivier JGJ, Peters JAHW (2010) No growth in total global CO2 emissions in 2009. Bilthoven, Netherlands, Netherlands Environmental Assessment Agency (PBL) 16

  57. Oram DE, Sturges WT, Penkett SA, McCulloch A, Fraser PJ (1998) Growth of fluoroform (CHF3, HFC-23) in the background atmosphere. Geophys Res Lett 25(1):35–38

    Article  Google Scholar 

  58. PCMDI (2009) CMIP5 Coupled Model Intercomparison Project—Overview. from http://cmip-pcmdi.llnl.gov/cmip5/

  59. Peters W, Krol MC, Dlugokencky EJ, Dentener FJ, Bergamaschi P, Dutton G, von Velthoven P, Miller JB, Bruhwiler L, Tans PP (2004) Toward regional-scale modeling using the two-way nested global model TM5: Characterization of transport using SF6. Journal of Geophysical Research-Atmospheres 109(D19)

  60. Prather MJ, Hsu J (2008) NF3, the greenhouse gas missing from Kyoto. Geophys Res Lett 35(12)

  61. Prather MJ, Hsu J (2010) NF3, the greenhouse gas missing from Kyoto (vol 37, L11807, 2010). Geophys Res Lett 37

  62. Prinn RG, Weiss RF, Fraser PJ, Simmonds PG, Cunnold DM, Alyea FN, O’Doherty S, Salameh P, Miller BR, Huang J, Wang RHJ, Hartley DE, Harth C, Steele LP, Sturrock G, Midgley PM, McCulloch A (2000) A history of chemically and radiatively important gases in air deduced from ALE/GAGE/AGAGE. J Geophys Res Atmos 105(D14):17751–17792

    Article  Google Scholar 

  63. RCP Database (2009) RCP Database version 1.0 hosted at IIASA. Retrieved 23 Nov, 2009, from http://www.iiasa.ac.at/web-apps/tnt/RcpDb

  64. Riahi K, Gruebler A, Nakicenovic N (2007) Scenarios of long-term socio-economic and environmental development under climate stabilization. Technol Forecast Soc Chang 74:887–935, Special Issue: Greenhouse Gases - Integrated Assessment

    Article  Google Scholar 

  65. Riahi K, Krey V, Rao S, Chirkov V, Fischer G, Kolp P, Kindermann G, Nakicenovic N, Rafai P (2011). RCP-8.5: Exploring the consequence of high emission trajectories. Climatic Change (this issue). doi:10.1007/s10584-011-0149-y

  66. Smith SJ, Wigley TML (2006) Multi-gas Forcing stabilisation with the MiniCAM. Energ J (Special Issue 3): 373–391

  67. Smith, SJ, van Aardenne J, Klimont Z, Andres R, Volke AC, Delgado Arias S (2011) Anthropogenic Sulfur Dioxide Emissions: 1850–2005 Atmos Chem Phys 11:1101–1116

    Article  Google Scholar 

  68. Taylor K, Stouffer RJ, Meehl GA (2009) A summary of the CMIP5 Experiment Design. from http://cmip-pcmdi.llnl.gov/cmip5/docs/Taylor_CMIP5_design.pdf

  69. Thomson AM, Calvin KV, Smith SJ, Kyle GP, Volke A, Patel P, Delgado-Arias S, Bond-Lamberty B, Wise MA, Clarke LE, Edmonds JA (2011) RCP4.5: A pathway for stabilization of radiative forcing by 2100. Climatic Change (this issue). doi:10.1007/s10584-011-0151-4

  70. Thornton PE, Lamarque JF, Rosenbloom NA, Mahowald NM (2007) Influence of carbon-nitrogen cycle coupling on land model response to CO2 fertilization and climate variability. Global Biogeochem Cycles 21(4)

  71. Thornton PE, Doney SC, Lindsay K, Moore JK, Mahowald N, Randerson JT, Fung I, Lamarque JF, Feddema JJ, Lee YH (2009) Carbon-nitrogen interactions regulate climate-carbon cycle feedbacks: results from an atmosphere-ocean general circulation model. Biogeosciences 6(10):2099–2120

    Article  Google Scholar 

  72. Tomassini L, Reichert P, Knutti R, Stocker TF, Borsuk ME (2007) Robust Bayesian uncertainty analysis of climate system properties using Markov chain Monte Carlo methods. J Clim 20(7):1239–1254

    Article  Google Scholar 

  73. van Vuuren D, den Elzen M, Lucas P, Eickhout B, Strengers B, van Ruijven B, Wonink S, van Houdt R (2007) Stabilizing greenhouse gas concentrations at low levels: an assessment of reduction strategies and costs. Clim Chang 81(2):119

    Article  Google Scholar 

  74. van Vuuren DP, Meinshausen M, Plattner GK, Joos F, Strassmann KM, Smith SJ, Wigley TML, Raper SCB, Riahi K, de la Chesnaye F, den Elzen MGJ, Fujino J, Jiang K, Nakicenovic N, Paltsev S, Reilly JM (2008) Temperature increase of 21st century mitigation scenarios. Proc Natl Acad Sci 105(40):15258–15262

    Article  Google Scholar 

  75. van Vuuren DP, Edmonds J, Kainuma MLT, Riahi K, Thomson A, Matsui T, Hurtt G, Lamarque J-F, Meinshausen M, Smith S, Grainer C, Rose S, Hibbard KA, Nakicenovic N, Krey V, Kram T (2011a). Representative concentration pathways: An overview. Climatic Change (This Issue). doi:10.1007/s10584-011-0148-z

  76. van Vuuren DP, Stehfest E, Den Elzen MGJ, Deetman S, Hof A, Isaac M, Klein Goldewijk K, Kram T, Mendoza Beltran A, Oostenrijk R, Van Vliet J, Van Ruijven B (2011b) RCP2.6: Exploring the possibility to keep global mean temperature change below 2 degree C. Climatic Change (This Issue). doi:10.1007/s10584-011-0152-3

  77. Velders GJM, Andersen SO, Daniel JS, Fahey DW, McFarland M (2007) The importance of the Montreal Protocol in protecting climate. Proc Natl Acad Sci USA 104(12):4814–4819

    Article  Google Scholar 

  78. Velders GJM, Fahey DW, Daniel JS, McFarland M, Andersen SO (2009) The large contribution of projected HFC emissions to future climate forcing. Proc Natl Acad Sci USA 106(27):10949–10954

    Article  Google Scholar 

  79. Vollmer MK, Reimann S, Folini D, Porter LW, Steele LP (2006) First appearance and rapid growth of anthropogenic HFC-245fa (CHF2CH2CF3) in the atmosphere. Geophys Res Lett 33(20)

  80. Wigley TML, Raper SCB (2001) Interpretation of high projections for global-mean warming. Science 293(5529):451–454

    Article  Google Scholar 

  81. Wigley TML, Raper SCB (2002) Reasons for larger warming projections in the IPCC Third Assessment Report. J Clim 15(20):2945–2952

    Article  Google Scholar 

  82. Wigley TML, Richels R, Edmonds J (2007) Overshoot pathways to CO2 stabilization in a multi-gas context. In: Schlesinger M, Kheshgi H, Smithet J et al (eds) Human induced climate change: An interdisciplinary assessment. Cambridge University Press, Cambridge, pp 84–92

    Google Scholar 

  83. Wigley TML, Clarke LE, Edmonds JA, Jacoby HD, Paltsev S, Pitcher H, Reilly JM, Richels R, Sarofim MC, Smith SJ (2009) Uncertainties in climate stabilization. Clim Chang 97(1–2):85–121

    Article  Google Scholar 

  84. Wise MA, Calvin KV, Thomson AM, Clarke LE, Bond-Lamberty B, Sands RD, Smith SJ, Janetos AC, Edmonds JA (2009) Implications of limiting CO2 concentrations for land use and energy. Science 324:1183–1186

    Article  Google Scholar 

  85. WMO (2006) Scientific assessment of ozone depletion: 2006. Global ozone research and monitoring project—Report No. 50. Geneva, Switzerland, World Meteorological Organization: 572

  86. WMO (2007) Scientific assessment of ozone depletion: 2006. Global ozone research and monitoring project—Report No. 50. Geneva, Switzerland, World Meteorological Organization: 572

  87. Worton DR, Sturges WT, Gohar LK, Shine KP, Martinerie P, Oram DE, Humphrey SP, Begley P, Gunn L, Barnola JM, Schwander J, Mulvaney R (2007) Atmospheric trends and radiative forcings of CF4 and C2F6 inferred from firn air. Environ Sci Technol 41(7):2184–2189

    Article  Google Scholar 

Download references

Acknowledgements

We are deeply thankful to all those in the scientific community, that contributed invaluable assistance, datasets, model code and review comments. Without those contributions, creating the harmonized and consolidated GHG concentrations for the RCP scenarios would not have been possible. We are especially indebted to Tom M. L. Wigley and Karl E. Taylor for helpful in-depth comments on an earlier manuscript version. Furthermore, we thank those, who contributed critical data and assistance, namely Toshihiko Masui, Peter Kolp, Volker Krey, A. Mendoza Beltrán, E. Stehfest, Robert Andres, Greg Marland, Richard Houghton, Judith Lean, David Worton, Makiko Sato, Edward Dlugokencky, Patricia Lang, Kenneth A. Masarie, D.M. Etheridge, J.A. Culbertson, G.S. Dutton, T.M. Thompson, J.W. Elkins, B.D. Hall, J. Flückiger, J. van Aardenne, J. Nabel, K. Markmann, J. Rogelj, and Chris Jones and the contributors to data.giss.nasa.gov.

Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Author information

Affiliations

Authors

Corresponding author

Correspondence to Malte Meinshausen.

Appendices

Appendix 1. RCP GHG data within CMIP5

Table 5

Table 5 Contribution of RCP and ECP GHG data presented in this study to the CMIP5 experiments. For a detailed description of the CMIP5 experiments, see Taylor et al. (2009)

Appendix 2. Non-GHG forcing agents

Apart from the harmonized global GHG emissions and concentrations presented in this study, the ESMs and the scientific communities of IPCC WG1 and WG2 will be provided with a series of other input data sets: tropospheric & stratospheric ozone and aerosol abundances (Lamarque et al. 2011), land-use patterns (Hurtt et al. 2011), and solar forcing recommendations (Lean and Rind 2009). For the concentration calculations described here, these non-GHG forcings, including aerosols, tropospheric ozone precursors, solar irradiance and volcanic aerosols, were included since temperature and chemical feedbacks have an influence on atmospheric GHG concentrations and their fluxes (see Methods 2.4). To the extent possible, CMIP5-consistent assumptions in regard to the non-GHG forcings were taken, although it should be noted that our IPCC AR4 based forcing parameterizations of different non-GHG compounds differ from chemistry-climate model runs in relation to tropospheric ozone, for example (Lamarque et al. 2011), and will as well differ from new insights generated by CMIP5 set of models. For the emission-driven ESM runs in CMIP5, some of them will generate CO2 emissions resulting from land-use patterns endogenously, so these emissions will differ from the harmonized IAM emissions used here. For radiative forcing due to solar irradiance changes we followed the CMIP5 recommendation of repeating solar cycle 23 into the future—although we keep solar forcing constant after concentrations are stabilized. Solar irradiance data by Lean and Rind (2009) (see at http://www.geo.fu-berlin.de/en/met/ag/strat/forschung/SOLARIS/Input_data/CMIP5_solar_irradiance.html) is used here as recommended for CMIP5. The irradiance data has been converted into radiative forcing by dividing by 4 (geometrical) and multiplying by 0.7 to take into account albedo. Furthermore, the data is normalized to have an average of zero for the 22 years around 1750.

Concerning volcanic forcings, CMIP5 leaves it to the modeling groups as to how to treat volcanic stratospheric aerosols in the control run and 21st century runs. One problem is that an artificial cooling and reduction of sea level rise will occur in response to the first volcanic events in the historical run (1850), if the control run brought the model in equilibrium without volcanic eruptions. Here, we use a specific setup for volcanic aerosols, which is one—but not the only—option of how GCMs can deal with volcanic forcing for CMIP5. Specifically, we assumed the average volcanic aerosol loadings over the last 100-years (around −0.2 W/m2) to be applied in both the control run and the future runs from 2006 onwards, or equivalently, to shift the volcanic forcing series such that control run and future forcings, as well as the mean over the historical period are zero (see Taylor et al. 2009 for a further discussion of 2011; Meinshausen et al. 2011a). Analogously, GCMs could apply a stratospheric volcanic aerosol loading in their control runs, as well as in the post-2005 projections.

For the historical 20th century run (1765–2005), we derived monthly volcanic radiative forcing from optical thickness of volcanic stratospheric aerosols as used in the NASA GISS model (available from http://data.giss.nasa.gov/), using an optical thickness τ to radiative forcing conversion of −23.5 W/m2/τ. We extended with zero forcing from 2001 to 2005, resulting in a nominal positive forcing of 0.2 W/m2 after being shifted by the 100-year historical mean. Furthermore, we scaled the resulting volcanic forcing by 0.7 in order to obtain a best fit with historical temperature observations using our simple climate model—which compensates for a potential limitation in simple and intermediate complexity models to accurately model responses to volcanic eruptions using the standard forcing assumptions (Tomassini et al. 2007; Meinshausen et al. 2009). Other forcings are assumed according to IPCC AR4, as listed in Table 2.12 in Forster et al. (2007), such as stratospheric water vapour changes due to methane oxidization.

Rights and permissions

Open Access This is an open access article distributed under the terms of the Creative Commons Attribution Noncommercial License (https://creativecommons.org/licenses/by-nc/2.0), which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Reprints and Permissions

About this article

Cite this article

Meinshausen, M., Smith, S.J., Calvin, K. et al. The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109, 213 (2011). https://doi.org/10.1007/s10584-011-0156-z

Download citation

Keywords

  • Carbon Cycle Model
  • Radiative Force Level
  • Overshoot Scenario
  • Anthropogenic Force Agent
  • Inverse Emission