A review of progress towards understanding the transient global mean surface temperature response to radiative perturbation
- 1.7k Downloads
The correct understanding of the transient response to external radiative perturbation is important for the interpretation of observed climate change, the prediction of near-future climate change, and committed warming under climate stabilization scenarios, as well as the estimation of equilibrium climate sensitivity based on observation data. It has been known for some time that the radiative damping rate per unit of global mean surface temperature increase varies with time, and this inconstancy affects the transient response. Knowledge of the equilibrium response alone is insufficient, but understanding the transient response of the global mean surface temperature has made rapid progress. The recent progress accompanies the relatively new concept of the efficacies of ocean heat uptake and forcing. The ocean heat uptake efficacy associates the temperature response induced by ocean heat uptake with equilibrium temperature response, and the efficacy of forcing compares the temperature response caused by non-CO2 forcing with that by CO2 forcing.
In this review article, recent studies on these efficacies are discussed, starting from the classical global feedback framework and basis of the transient response. An attempt is made to structure different studies that emphasize different aspects of the transient response and to stress the relevance of those individual studies. The implications on the definition and computation of forcing and on the estimation of the equilibrium response in climate models are also discussed. Along with these discussions, examples are provided with MIROC climate model multi-millennial simulations.
KeywordsTransient climate response Equilibrium climate sensitivity Climate feedback Ocean heat uptake efficacy Efficacy of forcing
A climate system is said to be in equilibrium when it exhibits a statistically stable state for an extended period of time under a given set of boundary conditions. When external radiative perturbation is added to the system (radiative forcing), it begins to reveal different statistical behavior, which includes the mean state. The statistical behavior evolves as time progresses, and the system eventually reaches a new equilibrium that is constrained by the newly imposed boundary conditions. Historically, the equilibrium response has received much attention; the equilibrium climate sensitivity (ECS) is the most notable example (Knutti and Hegerl 2008; Maslin and Austin 2012). The ECS is defined as the globally and annually averaged value of the equilibrium surface temperature response to a doubling of atmospheric CO2 concentration. The ECS is an idealized yet useful concept, because it represents the final boundary of the system, and gives a measure of how much warming would occur in the future even if the current CO2 concentration is sustained and not raised. We note that changes in vegetation and ice-sheet distributions and in atmospheric gas concentrations by a biogeochemical response are excluded from the formal definition of ECS, although the effect of vegetation change is sometimes included in recent model-based estimates. While ice sheets exhibit multiple equilibria (e.g., Abe-Ouchi et al. 2013), the ECS is an inherent property of the climate system, which is independent of the history of radiative forcing.
The transient response, which generally includes the effect of time-varying boundary conditions, is more complicated to characterize and understand than the equilibrium response. In addition, the system does not approach equilibrium at a constant rate even under constant radiative forcing. Nevertheless, the transient response has received more and more attention recently, because it is more relevant to near-future climate change and thus has more practical implications (Allen and Frame 2007). From the viewpoint of climate stabilization scenarios, understanding the transient response is also important for evaluating the magnitude of “committed” warming and the rate of the change accurately (Armour and Roe 2011; Meehl et al. 2005; Plattner et al. 2008; Solomon et al. 2009; Zickfeld et al. 2013). Here, the committed warming is referred to an additional surface warming that is expected to emerge when constant radiative forcing is continuously applied after a stabilization point.
In this review article, we focus on the transient response of the global mean surface temperature to external radiative perturbation, because the magnitudes of many climatic impacts can be scaled proportionally to this global mean variable relatively well (e.g., Ishizaki et al. 2012; Ishizaki et al. 2013; Yoshimori and Abe-Ouchi 2012). One of the motivations for this review arises from the recent, rapid conceptual development on climate feedbacks. Here, climate feedbacks are loosely defined as climate change processes that cause the additional radiative adjustment to the climate system in response to the initial radiative perturbation. For example, previously underappreciated factors such as spatiotemporally varying climate feedbacks related to the ocean heat uptake and the dependence of climate sensitivity on forcing agents have been identified as important. These factors are essential for a quantitative understanding of the transient response. Nevertheless, these factors are rather confusing and not necessarily understood widely by the climate science community. In addition, the recent progress requires further clarification of the definitions of forcing and feedback and discussions on how they are to be computed. It is becoming clear that the omission of such factors would introduce a bias into the estimate of ECS based on observation data. The understanding of the background long-term response to anthropogenic radiative perturbation must also be well established for clarifying observed contemporary climate variations, such as the warming pause or “hiatus” during the early twenty-first century (e.g., Kosaka and Xie 2013; Meehl et al. 2011; Watanabe et al. 2013; Watanabe et al. 2014).
Transient climate response in a global feedback framework
In this section, the transient response of the global mean surface temperature to radiative perturbation is reviewed in a classical global feedback framework. In this framework, the loss of radiative energy from the climate system, i.e., radiative damping, is formulated in proportion to the global mean surface temperature change. In the following, all variables represent global mean values unless noted explicitly. We start with an equilibrium state in which the global mean net radiation at the top of the atmosphere (TOA) is zero.
A classical definition of radiative forcing is instantaneous or stratosphere adjusted forcing in which the radiative perturbation is evaluated at the tropopause instantaneously (on the time scale of radiative transfer) after the perturbation is applied or after the stratospheric temperature responds radiatively to the perturbation. The tropopause is chosen because the troposphere and surface are tightly coupled thermally through radiation and convection. The adjustments of stratospheric temperature are included in the forcing because its response time is on the order of months, and the subsequent forcing thus represents a more effective measure of radiative effect on the troposphere-surface system (Hansen et al. 1997; Stuber et al. 2001). We note that the magnitude of radiative forcing at the TOA is the same as that at the tropopause after the stratospheric adjustments, because any vertical convergence of radiative fluxes in the stratosphere would cause local temperature to adjust (Hansen et al. 1997).
where λ is termed the climate feedback parameter and represents how much energy is lost to space in accordance with the unit increase of the global mean surface temperature T (e.g., Gregory et al. 2015; Gregory et al. 2004; Knutti and Hegerl 2008; Winton et al. 2010). Symbol ∆ denotes the deviation (or anomaly) from the unperturbed climate. Parameter λ must be positive in order for the system to eventually reach a stable state. We note that a “signed” feedback parameter, e.g., Λ ≡ −λ, which has a positive/negative value for the amplification/suppression of initial perturbation (i.e., positive/negative feedback), is also widely used in other literature (e.g., Armour et al. 2013; Boer and Yu 2003; Winton et al. 2013a). N reflects mostly ocean heat uptake with other forms of energy consumption being latent heat for snow/ice melting, for example (Rhein et al. 2013), and eventually diminishes to zero under constant radiative forcing.
When the feedback is distinguished from the forcing as the surface temperature-mediated response of the climate system as in Eq. (1), it is more consistent to include the so-called rapid tropospheric adjustments in the definition of forcing. The troposphere may respond to forcing directly and rapidly (typically, «1 year) without invoking a surface temperature change. Examples of the tropospheric adjustments include cloud response to lower tropospheric warming due to the increase in atmospheric CO2 concentration (cloud adjustments, Gregory and Webb 2008), cloud albedo and lifetime changes due to the increase in aerosols (indirect effect of aerosols), and cloud amount changes due to the increase in light-absorbing aerosols (semi-direct effect of aerosols). The resulting forcing is referred to the stratosphere-troposphere-adjusted radiative forcing or effective radiative forcing (ERF). For details on the mechanism of tropospheric adjustments, readers are referred to Andrews et al. (2012a), Kamae et al. (2015), and Sherwood et al. (2015).
Time scale of the response
where τ ≡ C/λ (Hansen et al. 1981; Hansen et al. 1984; Hansen et al. 1985; Wigley and Schlesinger 1985). Equation (7) means that ΔT reaches 63 % of ΔT eq at t = τ, and the rate depends on the effective heat capacity and the climate feedback parameter. Substitutions of C = 2.1 × 108 or C = 1.5 × 1010 J K−1 m−2, corresponding to a typical ocean mixed-layer depth (50 m) or the average depth of world oceans (3700 m), respectively, and a nominal value of λ = 0.75 W m−2 K−1 (corresponding to the ECS of about 3 °C), yield τ ≅ 5 or τ ≅ 370 years for the two ocean depths, respectively. The value of τ ≅ 5 does not include the effect of the heat exchange between the mixed layer and the deep ocean, and the τ ≅ 370 value assumes that the heat anomaly is communicated instantly throughout the water column. Both cases are unrealistic but give some insight into the response time. In reality, the heat anomaly spreads gradually from the surface to depth, and thus C varies with time. Schwartz (2007, 2008) applied a single-layer model with a similar formulation to estimate the ECS from the observation data, but their use of constant effective heat capacity for the entire climate system on all time scales of response was criticized (Foster et al. 2008; Knutti et al. 2008; Scafetta 2009). Donohoe et al. (2014) reported that the effective heat capacity is equivalent to about 50-m ocean depth during the first decade (consistent with our estimate of 5 years for the mixed-layer ocean) and equivalent to several hundred meters ocean depth after a century following an instantaneous quadrupling of atmospheric CO2 concentration (abrupt4xCO2 experiment) for the CMIP5 AOGCMs.
where γ is the proportionality coefficient. The heat capacity of the atmosphere, land, and other surface components of the climate system is small and implicitly included in the heat capacity of the upper ocean (C), and C is much smaller than the heat capacity of the deep ocean (C D ), i.e., C < <C D . We note that the addition of Eqs. (8) and (9) yields Eq. (1). That is, the surface temperature change and the ocean heat uptake hold a linear relation under constant radiative forcing in the two-layer model. Geoffroy et al. (2013a) applied this model to analyze the CMIP5 AOGCM simulations and showed that the two-layer models tuned to emulate the global mean surface temperature response of AOGCMs in the abrupt4xCO2 experiment can also emulate the result of the slowly increasing CO2 (1pctCO2) experiment reasonably well. This supports the assumption that the two-layer model is effective in capturing the minimal physics of the global mean surface temperature transient response.
Traditionally, one-dimensional box-diffusion ocean models (Hansen et al. 1985; Long and Collins 2013; Siegenthaler and Oeschger 1984; Wigley and Schlesinger 1985) or upwelling-diffusion (UD) ocean models (Andronova and Schlesinger 2001; Baker and Roe 2009; Johansson et al. 2015; Wigley and Raper 2001) have been used. AOGCM studies, however, point out the following limitations of such models: (a) the deep convection at high latitudes identified as an important process in AOGCMs is not properly represented (Gregory 2000); (b) the vertical structure of the heat anomaly stored in the UD ocean models is biased to a shallower depth as compared to an AOGCM (Li et al. 2013); and (c) the total heat anomaly stored in the UD ocean models, which corresponds to the thermosteric sea level rise, is underestimated as compared to an AOGCM (Li et al. 2013).
Kostov et al. (2014) showed that the time scale of the penetration depth of the heat anomaly with respect to the unperturbed climate is positively correlated with the depth and strength of the Atlantic meridional overturning circulation (AMOC) on a century time scale in the eight CMIP5 AOGCMs. This result suggests that the AMOC plays an important role in setting the rate of global ocean heat uptake. In addition to the AMOC, Marshall and Zanna (2014) used a conceptual model to point out the importance of ocean eddy mixing in the Southern Ocean, with larger mixing leading to a faster (yet smaller) global heat uptake response.
Ocean heat uptake efficiency
where ρ ≡ κ + λ is termed the climate resistance (Gregory and Forster 2008; Gregory et al. 2009, 2010). Here, κ reflects primarily oceanic processes and a larger value indicates that more of the heat anomaly is absorbed in the ocean and hence surface warming is suppressed. On the other hand, λ reflects primarily the atmospheric processes (and sea ice-albedo feedback) and the larger value indicates that excessive energy is more effectively lost to space. The resistance ρ depends on κ (oceanic process) and λ (atmospheric process), and as a result of both processes, the surface temperature change remains small when resistance ρ is large.
As indicated by Eq. (14), the ratio of transient warming to equilibrium warming is larger when κ is smaller or λ is larger. We note that a case of large λ is consistent with Eq. (7). Raper et al. (2002) found a negative correlation between κ and λ in nine CMIP2 AOGCMs. They noted that larger surface warming caused by the smaller radiative damping rate of λ accompanies stronger ocean stratification and less heat release from the interior of the ocean through convection at high latitudes. This relation corresponds to a larger net heat uptake of the global ocean, i.e., a larger κ In the unperturbed climate, the AMOC transports heat to the northern high latitudes, where heat is released to the atmosphere. The AMOC acting as a heat sink of the ocean was reported by Knutti and Stocker (2000) and Marshall and Zanna (2014). The negative correlation found by Raper et al. (2002) was not verified in later studies with a larger number of models, however (Geoffroy et al. 2013a; Gregory and Forster 2008; Plattner et al. 2008). We note that Kuhlbrodt and Gregory (2012) emphasized the dominant role of the Southern Ocean and found the importance of eddy-induced transport in the model discrepancies of global ocean heat uptake.
Dufresne and Bony (2008) used Eq. (12) to quantify the relative contribution of radiative forcing, ocean heat uptake, and climate feedbacks to the spread of the TCR as well as the ECS in the CMIP3 AOGCMs. They concluded that cloud feedback is the predominant factor for the spread of both quantities. Boe et al. (2010), on the other hand, found a positive correlation in the CMIP3 AOGCMs between present-day ocean mixed-layer depths and future deep ocean warming and concluded that ocean heat uptake plays a major role in the uncertainty of transient warming, given that deep ocean warming is negatively correlated with surface temperature. Their analysis was based on the correlation of a single factor and so does not compare quantitatively with other factors, such as radiative forcing and climate feedbacks.
Effective climate sensitivity and ocean heat uptake efficacy
While Watterson (2000) showed that ∆T eff is stable over time, numerous studies reported that ∆T eff varies with time or is different from the ECS (e.g., Armour et al. 2013; Bitz et al. 2012; Boer and Yu 2003; Gregory et al. 2004; Kiehl et al. 2006; Li et al. 2013; Long and Collins 2013; Murphy 1995; Senior and Mitchell 2000; Williams et al. 2008; Winton et al. 2010).
In applying Eq. (15), stratosphere adjusted radiative forcing or ERF was used for F. One way to estimate the ERF is the regression method by Gregory et al. (2004), in which the ERF is given by the intercept at ∆T = 0 of the ∆T-N linear regression after a step increase of forcing, if assuming Eq. (1). The blue colors in Fig. 4 represent the result of the abrupt4xCO2 experiment, and the blue crosses represent the global and annual mean anomalies for the initial 150 years. The regression line by Gregory’s method is indicated by a blue line, which gives an ERF of 7.3 W m−2 at the y-intercept. In Fig. 3, the regression line by Gregory’s method is indicated by a blue dashed line, and the ERF and effective climate sensitivity are denoted by F and ∆T eff, respectively.
Williams et al. (2008) proposed effective forcing, in which the feedback parameter remains constant in time. Their effective forcing is different from the ERF by Gregory’s method in that Williams applied linear regression for the stabilization period under constant forcing after 70 years of a monotonic CO2 increase (1pctCO2) experiment and thus included an initial adjustment of the climate system on the decade level. Winton et al. (2010) argued that the decadal scale adjustment clearly contains the oceanic changes, and hence, it is more appropriate to regard the adjustment as a part of the feedback, rather than as a part of the forcing. Winton introduced a new parameter termed ocean heat uptake efficacy (ε), which is described in detail below.
Here, Eq. (20) means that the ocean heat uptake efficacy parameter represents the ratio between feedbacks operating under radiative forcing and ocean heat uptake alone.
The last term on the right side represents the time-varying effect of λ, as explained below.
However, Eqs. (29) and (30) are equivalent (N = H) only when the heat uptake by the upper ocean Eq. (26) is much smaller than that by the deep ocean Eq. (27). We also note that no piecewise behavior is assumed in Eq. (18); the time-varying efficacy parameter (ε) captures the curvature of the slope in Fig. 3. According to Geoffroy et al. (2013a), the model with ocean heat uptake efficacy mimics the CMIP5 AOGCM simulations more closely because it can represent nonlinearity between the surface temperature change and the ocean heat uptake under constant radiative forcing. An example of the nonlinear curvature of the ∆T-N relation in an AOGCM is denoted by the blue and red crosses in Fig. 4. The equilibrium temperature response estimated by Gregory’s method using the initial 20-year (blue circles) or 150-year integration (blue line) in the abrupt4xCO2 experiment tends to be smaller than the “true” equilibrium temperature response obtained from the long millennial integration.
Geoffroy et al. (2012) demonstrated that the spread of the TCR in the CMIP5 AOGCMs is caused by climate feedbacks (λ eq), radiative forcing (F), and ocean heat uptake efficacy (ε 0), in the order of large to small contributions. Their result is consistent with that of Dufresne and Bony (2008) and leads to the conclusion that the uncertainty in cloud feedback is a dominant factor to the uncertainty in both ECS and TCR.
Time dependence of global feedbacks and physical meaning of ocean heat uptake efficacy
Winton et al. (2013b) proposed an indirect radiative influence of the AMOC. They argued that weakening of the AMOC under global warming reduces the heat transport to the northern high latitudes and that corresponds to a shift of ocean heat uptake regions from low to high latitudes. This shift results in the increase of the ocean heat uptake efficacy and slows down the global surface temperature rise. As mentioned already, Kostov et al. (2014), on the other hand, proposed that the stronger AMOC enhances the penetration of the heat anomaly into deeper ocean in the perturbed climate, and the resulting increase in the effective heat capacity of the climate system slows down the global surface temperature rise. Therefore, the quantitative contribution of the AMOC still needs to be established.
Meraner et al. (2013) argued another aspect of inconstancy of the global feedback parameter, in which the parameter becomes less positive and hence the ECS becomes larger in a warmer climate through enhanced water vapor feedback. The state dependency of climate feedbacks has been discussed in many previous studies (e.g., Colman and McAvaney 2009; Colman et al. 1997; Feldl and Roe 2013; Hansen et al. 2005; Jonko et al. 2013; Yoshimori et al. 2011), but the result appears to depend on the model and so the effect of the dependency on the transient response still needs to be established. In a similar context, the possibility of inconstancy of the climate resistance parameter (as well as the global feedback parameter) in the CMIP5 AOGCMs was reported by Gregory et al. (2015), although they also listed errors in the estimated radiative forcing as an alternate or additional possibility.
In summary, time-varying global feedbacks or radiative damping may arise from (a) the state dependency of climate feedbacks, that is, the global feedback parameter depends on the background temperature T; (b) changes in the spatial pattern of the surface temperature anomaly ΔT(r,t) associated with ocean heat uptake and dynamics (ocean heat uptake efficacy) or forcing agent (efficacy of forcing); and (c) the non-local effect of atmospheric dynamics resulting from (b) on the local feedback parameter λ(r,t).
Definitions and estimates of forcing
Implication for the estimate of ECS
As stated in an earlier section, it takes a couple of millennia for a comprehensive climate model or AOGCM to reach a new equilibrium. Therefore, AOGCMs are not routinely run to equilibrium, and the ECS is not obtained precisely by following the definition. Traditionally, atmospheric GCMs coupled to “slab” mixed-layer ocean models (ASGCMs) are used to obtain the ECS. The mixed layer usually has a small yet sufficiently large heat capacity to simulate the annual cycle, and the steady state without considering the change in ocean dynamics is obtained after a few decades of integration. While some ASGCMs exhibit reasonably close ECSs to those obtained by the long integration of AOGCMs, other studies point out the difference (Danabasoglu and Gent 2009; Li et al. 2013; Shell 2013; Stouffer and Manabe 1999; Yokohata et al. 2008). Gregory’s method was adapted by Andrews et al. (2012b) for the CMIP5 AOGCM analysis. It is expected that the linear part of the ΔT-N relation at the later stage of integration (fitting a line to the simulated trajectory closer to the equilibrium point in Figs. 3 and 4) yields an accurate ECS (Armour et al. 2013). Geoffroy et al. (2013a) applied the two-layer energy balance model with the ocean heat uptake efficacy parameter to the CMIP5 AOGCMs and estimated the optimal ECS and other parameters simultaneously. As a consequence, the model captures the nonlinearity of the ΔT-N relations. Although the mean of 16 estimated ECSs differs by only about 8 % compared to that estimated by the ocean heat uptake efficacy parameter of one (Geoffroy et al. 2013b), the difference is more than 20 % for the two AOGCM cases.
The nonlinearity occurring from the time variation of the global climate feedback parameter or the ocean heat uptake efficacy has a very important implication when the ECS is estimated from observation data. A general tendency of the global climate feedback parameter that is decreasing or the ocean heat uptake efficacy that is larger than one indicates that the ECS based on the energy budget may be underestimated when the transient data are used (Geoffroy et al. 2013a; Winton et al. 2010). In other words, the deviation of the effective climate sensitivity from the equilibrium climate sensitivity illustrated in Fig. 3 is not negligible, as pointed out in many studies (e.g., Armour et al. 2013; Rose et al. 2014; Winton et al. 2010). Because the observed data represent the transient stage of the response under increasing greenhouse gas forcing, this deviation is a concern for the observation-based estimate of the ECS (e.g., Forster and Gregory 2006; Gregory et al. 2002; Lewis and Curry 2015; Otto et al. 2013). The inconstancy of the climate resistance parameter yields similar concerns if the ECS estimate is made based on the assumption of its constancy (Gregory et al. 2015; Gregory and Forster 2008). We note that the efficacy of forcing is another important factor that may distort the estimate of the ECS from observations, as pointed out by Kummer and Dessler (2014), Shindell (2014), and Shindell et al. (2015). Marvel et al. (2015) argued that the best estimate of the TCR moves upward from 1.3 to 1.8 °C and the ECS from 2.0 to 2.9–3.0 °C if the efficacy of forcing based on a single model is taken into account. Therefore, even though the time variation of the climate feedback parameter is taken into account, the lack of the forcing efficacy factor (e.g., Masters 2014) may introduce bias in the ECS estimate. So far, the time variation of the forcing efficacy has not been well established. While climate models of reduced complexity enable us to constrain the ECS and TCR with statistical inference by conducting a large ensemble of parameter perturbations, we must not overlook the fact that the effect of the ocean heat uptake efficacy and the forcing efficacy are usually not taken into account (e.g., Aldrin et al. 2012; Skeie et al. 2014). This is not only limited to the ECS estimate based on historical observations but also is closely related to the ECS estimate based on a volcanic event (Merlis et al. 2014) or a past climate (Yoshimori et al. 2009).
Correct understanding of the transient response to radiative perturbation is important for the interpretation of observed climate change, the prediction of near-future climate change, and committed warming under climate stabilization scenarios, as well as the estimation of ECS based on observation data. It has been known for some time that the radiative damping rate per unit of global mean surface temperature increase (i.e., global climate feedback parameter) varies with time. This inconstancy affects the transient global mean surface temperature change, and different studies have emphasized the various different aspects of this inconstancy.
Recent progress has focused not only on the speed of the ocean heat uptake but also the spatial pattern of the ocean heat uptake. The spatial pattern influences the global radiative damping and thus the global climate feedback parameter. These influences arise from the evolving spatial pattern of the surface temperature anomaly and/or changes in the local feedback parameter. The inconstancy of the global climate feedback may be formulated by a single parameter, called ocean heat uptake efficacy. In addition, the dependency of the global climate feedback parameter on the background climate has also been pointed out. The physical mechanisms behind the behavior of the global climate feedback parameter still need to be established. Detailed analysis of the atmosphere-ocean interaction as well as perturbed physics ensemble experiments including the ocean component might be useful for identifying the physical mechanisms (Collins et al. 2006). Ocean heat uptake processes associated with the AMOC and those operating in the Southern Ocean are of great importance in understanding both speed and spatial pattern of ocean heat uptake and require further study.
The efficacy of forcing is another important factor determining the transient response to radiative perturbation. The current understanding of the forcing efficacy is very limited, and its time variation is particularly not well understood. Nevertheless, its inconstancy is presumed to be in the latitudinal distribution of forcing that varies with time.
All of these studies require the estimate of radiative forcing in a consistent manner. As proposed by Radiative Forcing Model Intercomparison Project,1 ERF estimated by using AGCMs with a fixed SST, which does not require extrapolation, would greatly advance the study in a more quantitative way. Because the extrapolation using the regression technique in estimating the ECS would introduce ambiguity, we emphasize the usefulness of the integration of AOGCMs to equilibrium under doubled and quadrupled CO2 forcing as presented in this review article and other studies with different ocean parameterizations (Yamamoto et al. 2015; Yamamoto et al. 2014).
In the fifth intergovernmental panel on climate change (IPCC) assessment report (IPCC, 2013; IPCC-AR5), the lower bound of the ECS estimate was revised from the fourth IPCC assessment report from 2 to 1.5 °C. In addition, the dispersion in the ECS estimate based on the energy budget and the emergent constraint (the empirical relation found in models between the ECS and observable variable) (Fasullo et al. 2015; Klein and Hall 2015) by using observation data precludes the IPCC-AR5 from providing the best estimate. These issues might be resolved at least partially by taking both ocean heat uptake and forcing efficacies into account.
AGCM, atmospheric GCM; AMOC, Atlantic meridional overturning circulation; AOGCM, atmosphere- ocean GCM; ASGCM, atmosphere-slab ocean GCM; CMIP, coupled model intercomparison project; ECS, equilibrium climate sensitivity; ERF, effective radiative forcing; GCM, general circulation model; IPCC, intergovernmental panel on climate change; IPCC-AR5, the fifth assessment report of the IPCC; UD, upwelling-diffusion; SST, sea surface temperature; TCR, transient climate response; TOA, top of the atmosphere
We thank Dr. Kyle Armour and one anonymous reviewer whose suggestions helped to greatly improve the manuscript. Dr. Armour pointed out the piecewise interpretation of Eq. (29) and drew the authors’ attention to the importance of AMOC. Figure 4 was inspired by Dr. Isaac Held’s blog. The authors are also thankful to the MIROC model developing team. The developers of NCL are also appreciated. Simulation data were produced by using the JAMSTEC Earth Simulator and the NIES super computer system. This work was supported by JSPS KAKENHI Grant Number 15K05280.
MY drafted the manuscript. AAO, RO, and HS provided the MIROC simulation data, and MY conducted the analysis. All of the authors made intellectual contributions to the contents and structuring of the manuscript and approved the final manuscript.
The authors declare that they have no competing interests.
- Andrews T, Gregory JM, Webb MJ and Taylor KE (2012b) Forcing, feedbacks and climate sensitivity in CMIP5 coupled atmosphere-ocean climate models. Geophysical Research Letters, 39:doi:Artn L09712 10.1029/2012gl051607Google Scholar
- Armour KC and Roe GH (2011) Climate commitment in an uncertain world. Geophysical Research Letters, 38:doi:Artn L01707 10.1029/2010gl045850Google Scholar
- Boe J, Hall A and Qu X (2010) Deep ocean heat uptake as a major source of spread in transient climate change simulations (vol 37, L17701, 2010). Geophysical Research Letters, 37:doi:Artn L17701 10.1029/2010gl044726Google Scholar
- Boer GJ, Yu B (2003) Climate sensitivity and response. Clim Dyn 20:415–429Google Scholar
- Colman R and McAvaney B (2009) Climate feedbacks under a very broad range of forcing. Geophysical Research Letters, 36:doi:Artn L01702 10.1029/2008gl036268Google Scholar
- Cubasch U, Meehl GA, Boer GJ, Stouffer RJ, Dix M, Noda A, Senior CA, Raper S, Yap KS (2001) Projections of future climate change. In: Houghton JT, Ding Y, Griggs DJ, Noguer M, Van Der Linden PJ, Dai X, Maskell K, Johnson CA (eds) Climate Change 2001: The Physical Science Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 525–582Google Scholar
- Flato G, Marotzke J, Abiodun B, Braconnot P, Chou SC, Collins W, Cox P, Driouech F, Emori S, Eyring V, Forest C, Gleckler P, Guilyardi E, Jakob C, Kattsov V, Reason C, Rummukainen M (2013) Evaluation of climate models. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 741–866Google Scholar
- Foster G, Annan JD, Schmidt GA and Mann ME (2008) Comment on “Heat capacity, time constant, and sensitivity of Earth’s climate system” by S. E. Schwartz. Journal of Geophysical Research, 113:doi:10.1029/2007jd009373Google Scholar
- Geoffroy O, Saint-Martin D and Ribes A (2012) Quantifying the sources of spread in climate change experiments. Geophysical Research Letters, 39:doi:Artn L24703 10.1029/2012gl054172Google Scholar
- Gregory JM and Forster PM (2008) Transient climate response estimated from radiative forcing and observed temperature change. Journal of Geophysical Research-Atmospheres, 113:doi:Artn D23105 10.1029/2008jd010405Google Scholar
- Gregory JM, Ingram WJ, Palmer MA, Jones GS, Stott PA, Thorpe RB, Lowe JA, Johns TC and Williams KD (2004) A new method for diagnosing radiative forcing and climate sensitivity. Geophysical Research Letters, 31:doi:Artn L03205 10.1029/2003gl018747Google Scholar
- Hansen J, Lacis A, Rind D, Russell G, Stone P, Fung I, Ruedy R and Lerner J (1984) Climate sensitivity: analysis of feedback mechanisms. Hansen, J.E. and T. Takahashi (eds.): Climate Processes and Climate Sensitivity, Geophysical Monograph Series, 29, AGU, Washington, D. C., USA, 130-163.Google Scholar
- Hansen J, Sato M, Ruedy R, Nazarenko L, Lacis A, Schmidt GA, Russell G, Aleinov I, Bauer M, Bauer S, Bell N, Cairns B, Canuto V, Chandler M, Cheng Y, Del Genio A, Faluvegi G, Fleming E, Friend A, Hall T, Jackman C, Kelley M, Kiang N, Koch D, Lean J, Lerner J, Lo K, Menon S, Miller R, Minnis P, Novakov T, Oinas V, Perlwitz J, Perlwitz J, Rind D, Romanou A, Shindell D, Stone P, Sun S, Tausnev N, Thresher D, Wielicki B, Wong T, Yao M and Zhang S (2005) Efficacy of climate forcings. Journal of Geophysical Research-Atmospheres, 110:doi:Artn D18104 10.1029/2005jd005776Google Scholar
- IPCC (2013) Climate change 2013: the physical science basis. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, p 1535Google Scholar
- K-1 model developers (2004) K-1 coupled GCM (MIROC) description. Tech. rep., Center for Climate System Research, The University of Tokyo, Tokyo, Japan.Google Scholar
- Knutti R, Krähenmann S, Frame DJ and Allen MR (2008) Comment on “Heat capacity, time constant, and sensitivity of Earth’s climate system” by S. E. Schwartz. Journal of Geophysical Research, 113:doi:10.1029/2007jd009473Google Scholar
- Kuhlbrodt T and Gregory JM (2012) Ocean heat uptake and its consequences for the magnitude of sea level rise and climate change. Geophysical Research Letters, 39:doi:Artn L18608, 10.1029/2012gl052952Google Scholar
- Plattner GK, Knutti R, Joos F, Stocker TF, von Bloh W, Brovkin V, Cameron D, Driesschaert E, Dutkiewicz S, Eby M, Edwards NR, Fichefet T, Hargreaves JC, Jones CD, Loutre MF, Matthews HD, Mouchet A, Muller SA, Nawrath S, Price A, Sokolov A, Strassmann KM, Weaver AJ (2008) Long-term climate commitments projected with climate-carbon cycle models. J Clim 21:2721–2751CrossRefGoogle Scholar
- Rhein M, Rintoul SR, Aoki S, Campos E, Chambers D, Feely RA, Gulev S, Johnson GC, Josey SA, Kostianoy A, Mauritzen C, Roemmich D, Talley LD, Wang F (2013) Observations: ocean. In: Stocker TF, Qin D, Plattner G-K, Tignor M, Allen SK, Boschung J, Nauels A, Xia Y, Bex V, Midgley PM (eds) Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp 255–315Google Scholar
- Schwartz SE (2007) Heat capacity, time constant, and sensitivity of Earth’s climate system. Journal of Geophysical Research-Atmospheres, 112:doi:Artn D24s05 10.1029/2007jd008746Google Scholar
- Schwartz SE (2008) Reply to comments by G. Foster et al., R. Knutti et al., and N. Scafetta on “Heat capacity, time constant, and sensitivity of Earth’s climate system”. Journal of Geophysical Research, 113:doi:10.1029/2008jd009872Google Scholar
- Shine KP, Cook J, Highwood EJ and Joshi MM (2003) An alternative to radiative forcing for estimating the relative importance of climate change mechanisms. Geophysical Research Letters, 30:doi:Artn 2047 10.1029/2003gl018141Google Scholar
- Siegenthaler U, Oeschger H (1984) Transient temperature-changes due to increasing CO2 using simple-models. Ann Glaciol 5:153–159Google Scholar
- Yokohata T, Emori S, Nozawa T, Ogura T, Kawamiya M, Tsushima Y, Suzuki T, Yukimoto S, Abe-Ouchi A, Hasumi H, Sumi A, Kimoto M (2008) Comparison of equilibrium and transient responses to CO2 increase in eight state-of-the-art climate models. Tellus Series a-Dynamic Meteorol Oceanography 60:946–961CrossRefGoogle Scholar
- Zickfeld K, Eby M, Weaver AJ, Alexander K, Crespin E, Edwards NR, Eliseev AV, Feulner G, Fichefet T, Forest CE, Friedlingstein P, Goosse H, Holden PB, Joos F, Kawamiya M, Kicklighter D, Kienert H, Matsumoto K, Mokhov II, Monier E, Olsen SM, Pedersen JOP, Perrette M, Philippon-Berthier G, Ridgwell A, Schlosser A, Von Deimling TS, Shaffer G, Sokolov A, Spahni R, Steinacher M, Tachiiri K, Tokos KS, Yoshimori M, Zeng N, Zhao F (2013) Long-term climate change commitment and reversibility: an EMIC intercomparison. J Clim 26:5782–5809CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.