A year above 1.5 °C signals that Earth is most probably within the 20-year period that will reach the Paris Agreement limit

nature climate change Brief Communication https://doi.org/10.1038/s41558-025-02246-9 A year above 1.5 °C signals that Earth is most probably within the 20-year period that will reach the Paris Agreement limit Received: 6 August 2024 Accepted: 14 January 2025 Published online: 10 February 2025 Check for updates Emanuele Bevacqua Jakob Zscheischler 1 , Carl-Friedrich Schleussner 2,3 & 1,4 The temperature goals of the Paris Agreement are measured as 20-year averages exceeding a pre-industrial baseline. The calendar year of 2024 was announced as the first above 1.5 °C relative to pre-industrial levels, but the implications for the corresponding temperature goal are unclear. Here we show that, without very stringent climate mitigation, the first year above 1.5 °C occurs within the first 20-year period with an average warming of 1.5 °C. In 2023, global mean surface air temperature reached 1.43 °C above pre-industrial level (1.32–1.53 °C, likely range)1. This exceeded the best estimate for human-induced warming1 of 1.31 °C (1.1–1.7 °C), indicating that, among other factors, natural variability, including the imprint of an El Niño event, contributed to the observed temperature in 20232–4. The following year, 2024, became the warmest on record globally5–7, and it was announced as the first calendar year above 1.5 °C by several international organizations that independently track the global temperature5,6,8–12. Although some uncertainty across datasets exists7, averaged together, the data indicate a consensus that Earth’s surface air temperature reached 1.55 °C warming in 20246. A single year above 1.5 °C, however, does not mean that the long-term temperature has reached the level referred to in the Paris Agreement, as also highlighted by the Intergovernmental Panel on Climate Change (IPCC)13. The long-term global temperature goal under the United Nations Framework Convention on Climate Change refers to human-induced climate change only14,15, and the Second Periodic Review of the long-term global goal under the Convention has clarified “that [the goal] is assessed over a period of decades”16. Different approaches have been suggested to track progress against the temperature goal13,17,18, but uncertainties in these estimates imply that we will only be able to establish in hindsight whether a certain warming level has been reached with confidence10,19. To account for natural variability when determining humaninduced global warming from the observational record, the IPCC, in its Sixth Assessment Report (AR6), assessed global warming over a 20-year period13. Furthermore, it established global warming levels as the reference point for the assessment of climate impact drivers and extensively documented the climate risks emerging at a 1.5 °C warming level13. Addressing the question of when we will enter a 20-year period with average warming at that level is thus not just an exercise of tracking the global temperature record, but also informs on the onset of a 20-year period where the risks documented in the scientific literature at a 1.5 °C warming level are expected to emerge. This is of direct relevance for climate risk management and adaptation planning. Here, we investigate how single warm years are related to the onset of the 20-year global warming level period by combining observations and climate model simulations of the Coupled Model Intercomparison Project Phase 6 (CMIP6), focusing on models with skill in representing warming trends during 1981–201420. Specifically, we explore the question of whether a single year above 1.5 °C can be seen as an early warning of the world reaching the 1.5 °C long-term warming level. By analysing the already-reached warming levels, observations reveal that the first single year exceeding 0.6 °C, 0.7 °C, 0.8 °C, 0.9 °C and 1.0 °C global warming thresholds have consistently fallen within the first 20-year period in which average temperature reached the same thresholds (Fig. 1a, points falling in the vertical band). This pattern motivates the hypothesis that a similar behaviour may apply to the 1.5 °C threshold. If true, the occurrence of the first single year at 1.5 °C warming would imply that the 20-year period that reaches the Paris Agreement’s lower goal has already started and that the expected impacts at a 1.5 °C warming level will start to emerge. Temperature data from climate models align well with this observed pattern, with the first single year exceeding 0.7–1.0 °C warming typically occurring within the Department of Compound Environmental Risks, Helmholtz Centre for Environmental Research–UFZ, Leipzig, Germany. 2International Institute for Applied Systems Analysis, Laxenburg, Austria. 3Geography Department and IRITHESys Institute, Humboldt-Universität zu Berlin, Berlin, Germany. 4 Department of Hydro Sciences, TUD Dresden University of Technology, Dresden, Germany. e-mail: 1 Nature Climate Change | Volume 15 | March 2025 | 262–265 262 Brief Communication a https://doi.org/10.1038/s41558-025-02246-9 Lag of first warm year from entry time in 20-year period CMIP6 (26 models) 1.5 Observations (consolidated mean) Observations (individual datasets) 1.4 1.3 1.2 Warming level (°C) 1.1 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 −120 −80 −40 0 20 ∆t (years) b Lag of first 1.5 °C year from entry time in 20-year period 96% SSP5-8.5 (25) 100% SSP3-7.0 (24) 100% SSP2-4.5 (26) SSP1-2.6 (26) 73% SSP1-1.9 (12) 75% −20 −15 −10 −5 0 ∆t (years) 5 10 15 20 Fig. 1 | Time lag of the first single year at or above a global warming level from the time of entry in the first 20-year period that reaches the same level. a, The time lag is denoted by Δt, with positive values implying that the first year above a warming level occurred within the 20-year period reaching the same warming level (vertical light orange band). The blue dots indicate different observational datasets. The box plots (showing median and interquartile range) are derived from models under a moderate emissions scenario SSP2-4.5 (black whiskers extend to the most extreme data points within 1.5 times the interquartile range from the box). b, The same as the box plot in a but for 1.5 °C warming under different emission scenarios (here, grey whiskers display the full range if it exceeds black whiskers). The number of models employed (those compatible with observed recent warming trends20) is shown in brackets. The fraction of simulations that fall in the shaded area is also provided. first 20-year period reaching the same warming level (Fig. 1a, box plots). Furthermore, for all considered warming levels, the observed time lag between the first year surpassing a warming threshold (t1yr) and the time of entry in the 20-year period reaching the same threshold (t20yr), that is Δt = t1yr − t20yr, falls inside the climate model-based distribution (Fig. 1a). This underscores the skill of climate models in simulating recent global mean temperature dynamics. Moving t (...truncated)