It’s not too late – empirics prove potential for fast energy transitions necessary to back off tipping point

A final consideration around feasibility is that the thresholds are not immutable laws—indeed, some historical cases prove that the chosen thresholds can be overcome. Two examples are the surge of gas-fired power in the Netherlands and nuclear power in France, which respectively surged from 5% to 80% and 25% to 75% of the power mix in one decade, surpassing the applied feasibility threshold in this study over tenfold. Since such historical examples of fast transitions are typically driven by public policy and support48, the feasibility analysis can also be interpreted as a mapping exercise of where policy support is strongly needed to overcome existing constraints, which is crucial for achieving stated climate targets as all models and scenarios in this study surpass several feasibility thresholds. Nevertheless, as the results show, this mapping strongly depends on the applied model: deep structural differences between models lead to a wide variety of pathways reaching the same climate targets and, hence, different policy interventions are necessary from different modelling perspectives to make these pathways feasible.

Discussion

Our results suggest that, if announced national near-term (2030) and longer-term (2050–2070) emissions reduction ambitions throughout the world are achieved, global peak temperature increase will stay below 2 °C with ~75% certainty. However, if climate action is not strengthened post-2030, long-term ambitions will not be achievable (long-term ratchet gap) and global temperature increase will be around or above 2 °C by 2100. All applied models agree that—with the current pace of policy implementation—temperature increase will not exceed 2.5 °C over the course of the century but will still have a rising trend thereafter. These results clearly show that climate action and ambitions have notably improved since 2020, when the same set of models and scenario structure projected 2030 emissions to be on average 3 GtCO2 higher (Extended Data Fig. 2) and mean global temperature 0.2–0.3 °C warmer by 2100 18. This improvement, alongside our finding that global emissions are set to peak in the current decade with current NDCs, is in line with earlier post-COP26 assessments22,23,24,27,28,29. The latter assessments, however, also found increasing emissions for current policy scenarios throughout and after the 2020s22,27,28,29, whereas all models in our study find reduction or stabilization of emissions with current policies within this decade. Furthermore, using a diverse set of different IAMs and thus covering a large part of structural uncertainty while reducing undesired model response heterogeneity, our study offers a more robust assessment of emissions and warming implications of post-Glasgow pledges and action, compared to these earlier single-model assessments.

Accounting for post-2030 extrapolations of current policies and NDCs, this study shows substantially lower emissions and end-of-century temperatures than other studies; additionally, our multimodel LTT projections are slightly more optimistic (by ~0.1 °C). An important reason for this difference probably lies in the applied extrapolation method: in contrast to a continued trend in emission intensities used in this study, other studies applied a carbon price equivalent to 2030 action and increasing over time with GDP levels 23,27,28,29, a method leading to more conservative emissions reductions in most models18. There is no straightforward answer on which policy extrapolation method is better: while continuing a trend in emissions reductions may falsely bank on an emissions reduction trend that might not be equally attainable in the future, relying on extrapolated carbon prices may put too much faith in highly uncertain future model assumptions, especially considering that, for example, the decline in costs of low-carbon technologies have traditionally been underestimated in such assumptions12. Another important difference with several other model studies in the literature is that current policies are modelled explicitly (that is, not proxied via carbon prices until 2030; Supplementary Data) and on top of an up-to-date, harmonized set of socioeconomic and technoeconomic assumptions. This contributes to the relatively optimistic assessment of 2030 emissions in the study, as it endogenously assumes ratcheting of emissions reduction targets (by overperforming on these targets), where these are not ambitious.

Our scenario structure reflects three levels of ambition, where for each separate model region emissions are equal or lower with each subsequent level of ambition. Similarly, uncertainty to reach the modelled emissions reduction also increases with each subsequent level of ambition. Hence, despite aggregated national ambitions being in line with a temperature future well below 2 °C, the likelihood of all these national ambitions being achieved should not be overestimated. For each individual country that fails to meet its targets, the probability of higher temperature levels increases. Our feasibility analysis shows that, for most countries, it is far from straightforward to achieve their stated near- and long-term ambitions with the anticipated future demand patterns and technologies. The analysis reveals that there is no ‘free lunch’ in terms of feasibility of mitigation pathways: while different model structures imply heterogenous pathways to decarbonize economies in line with proposed pledges, each of the pathways compatible with a temperature future well below 2 °C faces substantial feasibility challenges, either in socioeconomic, technology scale-up, physical or sustainability dimensions. However, such feasibility challenges should not be interpreted as hard barriers to meet pledged mitigation targets but rather as areas where additional policy support or breakthroughs in technology or consumer behaviour49 may be needed to overcome such challenges.

Our results show that aggregating all national near- and long-term emissions reduction targets is still insufficient to limit global temperature increase to 1.5 °C (long-term ambition gap), which is the highest ambition of the Paris Agreement. Even if the most ambitious scenarios in this study are achieved, the temperature increase may still cause notable and damaging climate impacts50 and be sufficiently strong to activate several climate tipping elements, such as the collapse of the Greenland and West Antarctic ice sheet and the die-off of low-latitude coral reefs51,52. Therefore, even if ambitions have substantially improved in the run-up to and shortly after COP26, either cumulative emissions until 2050 must be reduced more to avoid overshooting the 1.5 °C target, or substantive negative emissions should be achieved after reaching LTTs to reach 1.5 °C by 2100 with a high overshoot (Extended Data Fig. 3). Nevertheless, the results of this study show that, while the focus on further ambition ratcheting should not be lost8, the short- and long-run implementation of the existing set of ambitions is currently the most relevant factor to avoid a climate disaster.