New physical science behind climate change: What does IPCC AR6 tell us?

The UN Panel on Climate Change released its sixth assessment report from Working Group I in August. The new report represents a gold standard for our current understanding of climate science.1IPCCSummary for policymakers.in: Masson-Delmotte V. Zhai P. Pirani A. Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, 2021: 1-41Google Scholar The new physical science assessments are based on improved observational datasets to assess historical warming as well as progress in scientific understanding of the response of the climate system to human-caused greenhouse gas emissions. Relative to AR5, eight aspects of key new findings can be laid out.1We now have a more accurate estimation of historical warming. Recent improvements in global temperature records going back to 1850 have led to a revision of observed historical warming for the AR5 reference period 1986–2005, which in AR6 is higher by 0.08[−0.01 to 0.12]°C than in the AR5. The last decade (2011–2020) was about 1.09[0.95 to 1.20]°C warmer than in 1850–1900.2We now have more evidence on climate attributions. There is increasing evidence from detection and attribution studies supporting that human influence is making extreme climate events, including heat waves, heavy rainfall, and droughts, more frequent and severe. Attribution studies find that emissions of greenhouse gases from human activities are responsible for approximately 1.07°C of warming from 1850–1900 to 2010–2019, with human influence being the main driver of multiple changes we are seeing in the climate.3Constrained climate projections are provided for the first time in the IPCC report. A more sophisticated approach to assess future changes in global surface temperature, ocean warming, and sea-level rise is adopted. Multiple lines of evidence, including climate models, historical data, and observations, are drawn on to provide the most accurate projections to date. The ability to combine these multiple lines of evidence in a robust and quantitative way is a new and encouraging development since the AR5.4We have a new estimation of climate sensitivity with a major reduction in uncertainty. The equilibrium climate sensitivity (ECS) measures the long-term global warming caused by a doubling of CO2 concentration in the atmosphere over that at pre-industrial levels. The new report concludes that it is likely that climate sensitivity lies between 2.5°C and 4°C, with a best estimate of 3°C, compared with between 1.5°C and 4.5°C as assessed in AR5. In concert with the ECS estimation, the transient climate response to cumulative CO2 emissions (TCRE), a metric that measures the transient global average surface temperature change per unit of cumulated CO2 emissions, usually 1,000 PgC, is assessed to be 1.0°C–2.3°C, with a best estimate of 1.65°C. This is a narrower range compared with the AR5 estimate of 0.8°C–2.5°C.5A new concept of Climatic Impact Drivers (CIDs) is introduced to bridge the physical sciences and climate actions. CIDs help to translate physical changes in the climate into what they mean for society and ecosystems. The CID-relevant information is informative for regional climate change adaptation activities and thus can serve as an essential bridge between physical climate science reported by Working Group I and climate change adaptation activities reported by Working Group II. This is a big step toward developing useful and actionable climate science under the umbrella of the IPCC.6The low-likelihood but high-impact (LLHI) events in possible climate futures are explicitly highlighted for the first time in the IPCC report. The LLHI events refer to some potential outcomes of climate change that could have major impacts but are considered unlikely or extremely uncertain. One example of this is collapse of the Antarctic ice sheet, which would lead to a faster than projected sea-level rise. Another example is the impact of volcanoes. Based on paleoclimate and historical evidence, the report stated that it is likely that at least one large explosive volcanic eruption would occur during the 21st century. Such an eruption would reduce global surface temperature and precipitation for one to three years and alter the global monsoon circulation. Assessment of such possible but unlikely outcomes allows a more comprehensive risk assessment.7New climate projection covers the low-end emission for pledged net-zero emission targets for 2050. The main change of AR6 from AR5 in climate projection is the addition of a lower scenario (SSP1-1.9), which includes strong climate and air pollution mitigation. The low-end scenario features net-zero CO2 emissions in the 2050s. As an action of the Paris Agreement, many countries around the world have pledged net-zero emission targets for 2050. The projection under the SSP1-1.9 scenario provides useful information to policymakers for the condition that all countries would be able to fulfill their pledges to reach net-zero CO2 emissions by the middle of the century (Figure 1).8The crossing time of 1.5°C above 1850–1900 levels is projected on the basis of new methods. The report shows that in the next 20 years, the global average temperature is expected to reach or exceed 1.5°C above 1850–1900 levels. This is different to the assessment in the Special Report on 1.5°C of warming (SR1.5), which found that global warming would cross 1.5°C between 2030 and 2052. The difference in the cross-time estimates arises from the fact that SR1.5 and AR6 took two different approaches. AR6 considered a broad set of scenarios and projects a long-term average global warming of 1.5°C, whereas SR1.5 simply looked at a linear extrapolation of warming rates of the recent past. While we highlight these new findings, we should acknowledge that the assessment also calls for further investigations in some key fields toward a better tracking of whether climate is changing and a much better understanding of climate change from past and present to future. From a personal view, more efforts should be devoted to the following fields. Our understanding of past climate changes relies heavily on the observational data. While the report’s conclusion is based on observations that climate change is already affecting every region on Earth, we know that currently there is still limited evidence in many parts of the world due to a lack of observations and research, in particular high mountainous areas such as the Tibetan Plateau, and arid and semi-arid areas.2Chen F.H. Ding L. Piao S.L. et al.The Tibetan Plateau as the engine for Asian environmental change: the Tibetan Plateau Earth system research into a new era.Sci. Bull. 2021; 66: 1263-1266Crossref Scopus (6) Google Scholar New observations including proxy data and paleoclimate record reconstructions are encouraged in such regions where climate records remain sporadic. Climate models are crucial for climate attribution and projection studies. A reliable simulation of precipitation and extremes at a regional scale remains a challenge. Although the performance in simulating the magnitude of precipitation and extremes varies between models, as well as differing regionally, generally higher model resolution is an important step toward a further increase of confidence in simulation. Added values are expected from very-high-resolution convection permitting models (2–5 km),3Zhao Y. Zhou T.J. Li P.X. et al.Added value of a convection permitting model in simulating atmospheric water cycle over the Asian Water Tower.J. Geophys. Res. Atmospheres. 2021; 126 (e2021JD034788)https://doi.org/10.1029/2021JD034788Crossref Scopus (7) Google Scholar which could resolve processes that are not captured in Coupled Model Intercomparison Project (CMIP)-like relatively coarse resolution models. The observational constraints of model projections have been used in AR6 for global surface temperature, ocean heat uptake, and sea-level rise. While this is a big step toward providing a reliable projection, the projections of precipitation, atmospheric circulations, climate extremes, and many other variables are still based on the ensemble of CMIP models. We thus need to develop novel methods by combining multi-model projections with observational constraints to provide robust projections of more climate variables.4Chen X.L. Zhou T.J. Wu P.L. et al.Emergent constraints on future projections of the western north pacific subtropical high.Nat. Commun. 2020; 11: 2802https://doi.org/10.1038/s41467-020-16631-9Crossref PubMed Scopus (23) Google Scholar Accurate estimate of climate sensitivity has been one focus of the IPCC since its establishment. The ECS, as well as TCRE, is crucial for climate projection and the estimate of remaining carbon budgets. While the new report has reduced the uncertainty in both ECS and TCRE estimate, the ranges of uncertainty are still larger than expectation. Narrowing down the uncertainty range of ECS and TCRE estimation should be of the highest priority in climate research, since this is crucial for developing a concrete, science-based strategy for limiting global temperature increase to well below 2°C, aiming for 1.5°C. In addition, in AR6, the single-model initial-condition large ensembles have improved our understanding of the impact of internal variability on the forced changes. While we highlight this as an important amendment to multi-model ensembles of CMIP, we know that hundreds of realizations of possible weather are needed to look at very rare events, and hence more computer resources are needed to support climate sciences in the future. Turning knowledge into action to address the world's sustainable development challenges requires actionable sciences. Although the assessment of the IPCC is policy relevant and does not have a remit to prescribe policies, international projects such Future Earth have called for a strengthened link between science and decision-making processes.5Cheng H. Future Earth and sustainable developments.Innovation. 2020; 1: 100055https://doi.org/10.1016/j.xinn.2020.100055Abstract Full Text Full Text PDF Scopus (33) Google Scholar Using co-design and co-delivery along with independent physical science assessment will help to create a more useful and actionable climate science in the future reports of the IPCC. The authors declare no competing interests.