Long-term temperature and sea-level rise stabilization before and beyond 2100: Estimating the additional climate mitigation contribution from China's recent 2060 carbon neutrality pledge

The baseline CO₂ emissions for China and the world from 2015 to 2100 were derived from an NDC pathway developed by Huang et al (2020) using a Computable General Equilibrium model with optimized costs. Note this baseline scenario is already a mitigation pathway globally, which aims at holding a temperature rise under 2 °C by 2100. Thus, it should not be confused with a typical 'baseline' in the literature in which there assumes no or weak climate policy in place, such as SSP3-7.0. In this NDC pathway, CO₂ emission in China peaks in 2030 and continues to decline toward 2100, while the global emission peaks in 2040. Here, we further expand the NDC-related pathway to 2200 by holding the same reduction rate as in 2050–2100 (thick black line in figure 1).

Zoom In Zoom Out Reset image size

The carbon naturality pathways in China were constructed to have two distinct components: reducing gross emissions and increasing the CCUS capacity.

Emission reduction is represented by a power function of time with a reducing rate r as follows:

$$\begin{equation}\begin{array}{*{20}{c}} {{\text{Emission}}\, = \,{\text{Emissio}}{{\text{n}}_{{\text{peak}}}}\, \cdot \,{{\left( {1 - r} \right)}^{t - {t_{{\text{peak}}}}}}} \end{array}\end{equation} \tag{ 1 }$$

where Emission is the emission of China in year t (beginning at t_peak), and Emission_peak and t_peak are the peak emission and corresponding year, respectively.

The growth of CCUS is represented as an S-curve function with a growth rate g, and a cap of CCUS_limit, as follows:

$$\begin{equation}\begin{array}{*{20}{c}} {\frac{{{\text{dCCUS}}}}{{{\text{d}}t}}\, = \,\frac{{\left( {{\text{CCU}}{{\text{S}}_{{\text{limit}}}}\, - \,{\text{CCUS}}} \right)\, \cdot \,{\text{CCUS}}}}{{{\text{CCU}}{{\text{S}}_{{\text{limit}}}}}}\, \cdot \,g} \end{array}\end{equation} \tag{ 2 }$$

where CCUS_limit is set to be 0.5 GtC yr⁻¹ for China according to the layout proposed by Wei et al (2021). As noted previously, the current CCUS capacity is very low at 0.001 GtC (Cai and Li 2020), and that is the initial condition used in this study.

To assess the impact of pathway uncertainty associated with the carbon peaking year (2024–2030) and CNY (2050–2070), a suite of pathways was constructed. The values of r and g for certain CPYs and CNYs are listed in table 1. Because CCUS is limited to 0.5 GtC yr⁻¹, only a sufficiently large r can lead to eventual carbon neutrality. Thus, the value of r was set to 6.0%/year, which is the minimum to realize carbon neutrality in 2060 or later, and 9.0%/year, which is the minimum to realize carbon neutrality in 2050, as well as faster rates, such as 10.5%/year and 12.0%/year, which were used for sensitivity exploration. The corresponding values of g were calculated based on the given r and assumed CPY in order to reach carbon neutrality at certain years. It is reasonable that a larger r would require a smaller g to achieve carbon neutrality at a certain time (e.g. 2060).

International cooperation is required to mitigate climate change. As previously mentioned, several countries raised their own carbon-neutral targets before or shortly after China. As China is one of the world's major emitters, world emissions partially depend on China's emissions. To correlate the world emission response to China's carbon neutrality pathway obtained in the previous section, we fit the world's emissions with China's emissions using all available SSP pathway emission data (Riahi et al 2017) in the low emission range (China's emissions less than 3.5 GtC yr⁻¹) and obtained a quadratic function. The R² of the fitness is 0.94 and the root mean square error is 1.45 GtC yr⁻¹. Therefore, using this function, global emissions (fossil fuel and land use) can be approximately estimated from China's emissions. In addition, global emissions are capped by the baseline that is the NDC pathway as derived from Huang et al (2020).

Considering the technical limitations and the political will to reduce emissions, we considered four cases of decarbonization pathways from 2015 to 2200:

Case A) China-only pathway. The rest of the world stays with the baseline-mitigation emissions, and China's emissions reach carbon neutrality and negative. This is a hypothetical case to quantify the added value of enhanced pledges from China.

Case B) Global zero CO₂ pathway. In addition to Case A, the rest of the world also reduces carbon emissions to zero, but without going negative.

Case C) Global negative CO₂ pathway. In addition to Case B, the rest of the world reaches negative carbon emissions between 2050 and 2070.

Case D) Global zero GHG pathway. This case uses the same carbon emission as Case C, but also considers non-CO₂ GHGs following the RCP4.5 emissions to 2050 (e.g. the peaks of CH₄ and N₂O are 2030 and 2040, respectively), and then declines linearly to zero in 2100, as opposed to the remaining 2.5 GtCeq in RCP4.5.

The MiCES is used to simulate climate change from 1850 to 2200 caused by global GHG and aerosol emissions (Sanderson et al 2017). It is a simplified climate model based on the energy and carbon budget of the Earth's system. In this model, the Earth system is divided into four parts: land, atmosphere, surface ocean, and deep ocean, wherein carbon and heat transfer between these parts can achieve dynamic equilibrium.

In total, the model contains 37 parameters, which can be divided into two categories: heat and carbon transfer parameters and chemical parameters. Herein, we used parameter sets optimized by Chen et al (2020), who investigated the sensitivity of the parameters and found that seven parameters related to heat and carbon transfer were the most sensitive among the 37 parameters. Chen et al (2020) optimized the parameters by fitting the most sensitive parameters with the observed emission and temperature rise within their uncertainty range. For example, the most important parameter, the equilibrium climate sensitivity, starts with the 1 °C–6 °C range and is set to 4 °C in the original model. After the optimization, this parameter is recalibrated to 2.8 °C in the updated model used in this study, which is close to the central values documented in recent IPCC reports. The settings of key parameters are summarized in table 2 for references and reproducibility. Note climate sensitivity is one of the major causes of uncertainty, we use the 2.3 °C−4.7 °C (the 5%–95% bound range) by Sherwood et al (2020) for the uncertainty range calculation.

Table 2. Key parameters in the MiCES.

Parameter	Description and unit	Value
λ	Climate sensitivity K(Wm−2)−1	2.78
${\kappa _{\text{l}}}$	Land surface heat capacity (Ka−1)−1 Wm−2	23.5
${\kappa _{\text{o}}}$	Ocean heat capacity (Ka−1)−1 Wm−2	145.5
Do	Atmosphere–ocean diffusion coefficient Wm−2 K−1	0.19
${\beta _{\text{l}}}$	Biosphere CO2 fertilization parameter Pg ppm−1	2.27
${\beta _{\text{o}}}$	Ocean carbon diffusion parameter Pg ppm−1	4.92
${\gamma _{\text{l}}}$	Biosphere temperature response Pg K−1	−50.8
${\gamma _{\text{o}}}$	Ocean carbon solubility response Pg K−1	−1.52

The historical (up to 2014) global CO₂ emissions come from the PRIMAP-hist dataset (Gütschow et al 2016), which combines several published datasets to provide a GHG emission pathway for both globally and individual countries. The historical CO₂ pathway was merged with constructed CO₂ pathways for 2015–2200 (section 2.2). The radiative effect of non-CO₂ GHGs on the climate was included using an individual chemical module, not the carbon cycle model. We considered other non-CO₂ GHGs listed by the Kyoto Protocol (as well as sulfate that represents the aerosol's cooling effects). We adopt non-CO₂ projections in RCP4.5 (www.iiasa.acat/web-apps/tnt/RcpDb) because the CO₂ emissions in RCP4.5 are the closest to the baseline selected here. These non-CO₂ emissions are considered in all cases.

The SLR was calculated using a semi-empirical statistical model proposed by Vermeer and Rahmstorf (2009):

$$\begin{equation}\begin{array}{*{20}{c}} {\frac{{{\text{dSLR}}}}{{{\text{d}}t}}\, = \,a\, \cdot \,\left( {T\, - \,{T_0}} \right)\, + \,b\, \cdot \,\frac{{{\text{d}}T}}{{{\text{d}}t}}} \end{array}\end{equation} \tag{ 3 }$$

where T and SLR are the temperature rise and SLR at time t, respectively, and T₀ is the base temperature at which sea level is in equilibrium with climate. The coefficient estimates for the model were based on the fit of the observed global temperature and SLR (NOAA 2020, 2021). The 1850–2014 period was considered the historical period used to calibrate the model, generating the optimized coefficients of a = 0.25 cm yr⁻¹ K⁻¹, b = −0.25 cm K⁻¹, and T₀ = −0.43 K. Note that these coefficients are slightly different from those used in Hu et al (2013), which may be because of the different observation data used for training.

Figure 1 shows emissions of different proposed pathways. China's emission pathway decreased from the Chinese NDC baseline (Huang et al 2020) after the assumed CPYs of 2024, 2027, and 2030, and reached neutrality between 2050 and 2070 (figure 1(a)). As the CCUS limit was set to 0.5 GtC yr⁻¹, negative emissions will remain constant when they reach the lower bound. Correspondingly, the world's carbon emissions under three different pathways (figures 1(b)–(d)) have the same peak emissions of 9.3 GtC, 9.5 GtC, and 9.7 GtC in 2024, 2027, and 2030, respectively. Under the only China pathway (as a hypothetical case to isolate the contribution from China's recent pledge alone), 123–160 GtC less carbon will be emitted than that of the baseline NDC during this century (2015–2100). However, the world will not achieve net-zero emissions. Conversely, under Case B (global zero CO₂, figure 1(c)), there is a large decrease in carbon emissions reaching zero around 2050–2070, causing a cumulative difference of 370 GtC to 420 GtC during this century. Under Case C (the negative CO₂ pathway), the emission pathway is the same as in Case B. However, emissions will further decrease to approximately −3 GtC yr⁻¹ after this time (proportional to the imposed cap of 0.5 GtC yr⁻¹ for China), resulting in 30–100 GtC fewer carbon emissions, as compared with Case B (global zero CO₂) during this century.

Based on the simple scaling approach, we estimate that global carbon neutrality will be achieved 2–9 years after China's carbon neutrality, and also, earlier carbon neutrality in China will result in a shorter delay. For example, China reaching carbon neutrality in 2050 will lead the world to neutrality in 2052; however, if China reaches carbon neutrality in 2070, global carbon neutrality will be delayed to 2075–2079. To illustrate the sensitivity of timing of policy implementation, we also find that the uncertainty of CNYs (i.e. 2050–2070) have a stronger impact on the cumulative emission with a 35 GtC difference under the zero CO₂ pathway and with a 130 GtC difference under the negative CO₂ pathway (within this century). In contrast, the near-term CPYs (i.e. 2024–2030) only caused a 40 GtC difference under the negative CO₂ pathway and a 50 GtC difference under the zero CO₂ pathway.

Compared with the global baseline (black line in figure 1(b)), China's carbon neutrality will help reduce the global temperature rise from 2.46 °C (uncertainty range 2.06 °C–4.01 °C) to 2.25 °C–2.30 °C (uncertainty range 1.89 °C–3.75 °C) by 2100 (figure 2(a)). That is, the contribution from China's recent pledge alone will reduce warming by 0.16 °C–0.21 °C, contributing approximately 17%–22% to the Paris Agreement's 1.5 °C goals (relative to the 0.96 °C gaps between the NDC's projection to 1.5 °C). This is consistent with a recent estimate that China achieving carbon neutrality would lower global warming by approximately 0.2 °C–0.3 °C (Climate Action Tracker 2020). A range of benefits comes from the uncertainty associated with the decarbonization pathway, but it can be seen in figure 2(a), an earlier carbon neutralization year would lead to greater avoided warming.

Zoom In Zoom Out Reset image size

With global efforts to achieve carbon neutrality in the 2nd half of this century, all three other cases (global zero CO₂, globally negative CO₂, and global zero GHG) can successfully hold the temperature rise below 2 °C (figures 2(b)–(d)), which is consistent with previous studies (Rogelj et al 2015, 2019, Salvia et al 2021). However, stabilizing the temperature below 1.5 °C is more difficult (horizontal dash line in figure 2), which is discussed in detail next.

Under the global zero CO₂ pathway (Case B in figure 2(b)), the global temperature would slowly decrease after peaking by 1.64 °C–1.78 °C (uncertainty range 1.37 °C–3.03 °C) around 2087 and would remain nearly constant until the end of the century and beyond. Under this pathway, the warming peak year (around 2087) remains the same for all CNYs. That is because CO₂ emissions will continue to be zero after carbon neutrality, so the warming peak is mostly dependent on the non-CO₂ emission and the sensitivity of the earth system. Meanwhile, under the global negative CO₂ pathway (Case C), global temperature will rise, peaking by 1.6 °C–1.8 °C (uncertainty range 1.35 °C–2.92 °C) between 2062 and 2085, which is approximately 15 years later than the CNY, before it will decrease to a rise of 1.5 °C–1.8 °C (uncertainty range 1.21 °C–2.95 °C) in 2100 at a rate of 0.02–0.05 °C/decade. The temperature overshoot is stronger in Case D, under the zero GHG pathway. Specifically, temperatures will reach peaks earlier (at 2060–2067) by 1.6 °C–1.7 °C (uncertainty range 1.34 °C–2.77 °C), in which the peak time varies with CNY, and may achieve a peak temperature before carbon neutrality as non-CO₂ emissions are lower under the zero GHG pathway. After the peak level, the temperature would then decrease to 1.3 °C–1.6 °C (uncertainty range 1.08 °C–2.71 °C) in 2100 and maintain the declining trend into the 22nd century at a rate of 0.07–0.10 °C/decade.

While most of the 15 constructed pathways under the negative CO₂ pathway (Case C) can consistently bend the warming curve past 2100 and eventually bring it back to be under 1.5 °C, to meet the target of limiting the temperature under the 1.5 °C level in 2100, the more ambitious emission ramp-down pathways are required. Among the 15 constructed pathways, only the three most aggressive ones: World's carbon emission peaks in 2024–2027, instead of 2030, and reaches carbon neutrality around 2050; and carbon peaks in 2024 and reaches neutrality before 2058, have the chance to keep the temperature rise under 1.5 °C in 2100 with the aid of coordinated global efforts.

We also emphasize that when supplemented with non-CO₂ GHG neutrality before the end of the century (Case D), which is even more ambitious than cutting the non-CO₂ GHG emissions to 1.4 GtCeq as in SSP1-1.9, it is possible to keep the temperature rise under 1.5 °C in 2100 under more lenient carbon neutrality conditions. Specifically, it would require that World reaches carbon neutrality before 2065 and has an emission peak before 2030, or reaches carbon neutrality in 2070 and has an earlier and lower emission peak in 2024 (figure 2(d)).

In contrast, in the absence of sustained net-negative carbon emissions and zero emissions of non-CO₂ GHGs (Case B), even if the world achieves carbon neutrality as early as 2050, the 1.5 °C targets will barely be achieved (figure 2(b)). This suggests that even with the reduction of human influence, the climate system stays near a steady-state, maintaining the temperature rise for a long time (figure 2(b)), which is consistent with the results of the Zero Emissions Commitment Model Intercomparison Project, which showed that further temperature rise remains close to zero after reaching carbon neutrality (Jones et al 2019, MacDougall et al 2020).

The benefit of pursuing both negative CO₂ emissions and aggressive non-CO₂ GHG cuts does not end in 2100. Figure 2 shows that both the negative CO₂ and zero GHG pathways have cooler temperatures in 2150 than in 2050, while the zero CO₂ pathway remains stable in 2150. Indeed, the projected warming reversal rate in 2100 is 0.02–0.05 °C/decade under the negative CO₂ pathway and 0.07–0.1 °C/decade under the zero GHG pathway, much larger than the 0.003 °C/decade rates in the zero CO₂ pathway (a quasi stabilization).

Figure 3 shows the temperature rises in 2050, 2100, and 2150 depending on different CPYs and CNYs. It can be seen that temperature rise would increase with delayed CPYs and CNYs, especially by 2100 (figure 3). In particular, 20 years earlier in the CNY will lead to a sizable 0.2 °C temperature increase in 2100 under the negative CO₂ and zero GHG pathways and 0.04 °C–0.07 °C under the zero CO₂ pathway. Meanwhile, 6 years of delay in CPY will cause an approximately 0.07 °C–0.09 °C increase in temperature in 2100. This indicates that earlier emission peaks will lead to lower temperature rises, reducing the burden of mitigation later, as suggested by many previous studies (Tong et al 2019, Olhoff and Chistensen 2020).

Zoom In Zoom Out Reset image size

In particular, for the near-term period (period to mid-century), CPYs caused a greater difference in temperature rise than CNYs. We find that a 6 year difference of CPY avoided approximately 0.05 °C–0.07 °C warming in 2050, while a 20 year difference of CNY caused a negligible 0.01 °C. However, in the long-term, CPYs and CNYs caused approximately the same difference in temperature rise in 2100 and 2150, realizing an approximately 0.06 °C–0.09 °C difference for the 6 year range of CPY explored here (2024–2030) and a 0.04 °C–0.08 °C difference for the 20 year range of CNY explored here (2050–2070). Furthermore, CNYs show a greater influence of approximately 0.2 °C–0.3 °C in 2150 in both the negative CO₂ pathway and zero GHG pathway than in the zero CO₂ pathway, which is consistent with the results of the cumulative carbon difference caused by different pathways (section 3.1), because earlier neutrality implies a larger amount of avoided cumulative emission over time.

Even for the pathways that can successfully bend the warming down below 1.5 °C at 2100 or early 22nd century, the number of years exceeding 1.5 °C is another metric to be considered. As in table 3, under both the negative CO₂ case and zero GHG cases, the temperature rise will exceed 1.5 °C at approximately the same time (2043–2046), depending on the CPY. Under the negative CO₂ pathways, there is a 46–104 year period where temperature rise is above 1.5 °C, wherein this period is only 33–68 years under the zero GHG pathway. Further, it can be seen that the CNY has a more influential role, leading to a difference of approximately 40–50 years in years above 1.5 °C under the negative CO₂ pathway and 25 years under the zero GHG pathway. In contrast, CPY only leads to a 15 year difference under the negative CO₂ pathway and a 10 year difference under the zero GHG pathway. Thus, CNYs are the main driver of the number of years exceeding the 1.5 °C. Consistently, for cumulative warming above the 1.5 °C level which can be an analogy to cooling degree days, CNYs are more influential than CPYs, causing an approximately 8 °C·decade difference (between the more aggressive pathways and more relaxed decarbonization pathways), as opposed to a 3 °C·decade difference due to spreads in CPYs.

Looking beyond temperature level, we also note that with efforts to reduce non-CO₂ GHG emissions (Case D), the years above 1.5 °C and cumulative warming above 1.5 °C is reduced to only 60%–70% of that under the negative CO₂ pathway (Case C), further justifying the role of curbing the non-CO₂ GHG emission, beyond the well-demonstrated effects in reducing near-term warming rates (Ocko et al 2021).

While temperature rise can be stabilized at the end of the century, sea level would continue to rise under all pathways. However, different pathways would affect the SLR because SLR is more related to cumulative temperature rise. Our results show that SLR is less sensitive than temperature rise to CPYs and CNYs within the pathways, which may be due to the large heat capacity of the ocean. Specifically, under the baseline pathway, the sea level will rise to 71 cm at the end of the century (relative to 1850) and will rise by 155 cm in 2200 (figure 4). Figure 4 shows the differences caused by the different pathways. Our results (44–45 cm in this century) are between those found by the IPCC (2018) and Rasmussen et al (2018), who stated that the sea level would rise by 40–48 cm under the 1.5 °C pathway. China achieving carbon neutrality between 2050 and 2070 will lead to a 1–2 cm SLR decrease in 2100 and a 6–9 cm decrease in 2200. With global efforts to reach carbon neutrality, SLR decreases can be controlled to more than 5 cm under the other three cases in 2100 but close to 50 cm in 2200. It could also be seen that the difference of SLR from the baseline is more sensitive to CNY than CPY under the negative CO₂ pathway and zero GHG pathway, which is a 10 cm difference caused by CNY and 5 cm by CPY. However, under the zero CO₂ pathway, CPY and CNY cause approximately the same difference.

Zoom In Zoom Out Reset image size

The semi-empirical SLR model used here does not account for potential expected transitions in ice dynamics, which could result in substantially higher SLR (DeConto and Pollard 2016, DeConto et al 2021). Neglecting nonlinear physical processes, feedbacks, and threshold behavior could lead to biased conclusions based on statistical fits. However, statistical modeling of SLR could give cursory yet useful insights into whether SLR can be slowed down, despite land ice dynamics not being fully represented.