Final Government Distribution Technical Summary IPCC AR6 WGI 1 Table of Content 2 3 Introduction ..................................................................................................................................................... 4 4 Box TS.1: Core Concepts Central to This Report .................................................................................. 5 5 TS.1 A Changing Climate .......................................................................................................................... 11 6 TS.1.1 Context of a Changing Climate ........................................................................................... 11 7 Box TS.2: Paleoclimate ........................................................................................................................... 12 8 TS.1.2 Progress in Climate Science ................................................................................................ 15 9 TS.1.2.1 Observation-based products and their assessments ............................................................. 15 10 TS.1.2.2 Climate Model Performance ................................................................................................ 16 11 TS.1.2.3 Understanding Climate Variability and Emerging Changes................................................ 18 12 TS.1.2.4 Understanding of Human Influence..................................................................................... 20 13 TS.1.3 Assessing Future Climate Change ....................................................................................... 20 14 TS.1.3.1 Climate Change Scenarios ................................................................................................... 21 15 TS.1.3.2 Global Warming Levels and Cumulative CO2 Emissions ................................................... 23 16 TS.1.4 From Global to Regional Climate Information for Impact and Risk Assessment ............... 25 17 Cross-Section Box TS.1: Global Surface Temperature Change ............................................................... 27 18 TS.2 Large-scale Climate Change: Mean Climate, Variability and Extremes ..................................... 31 19 TS.2.1 Changes Across the Global Climate System ....................................................................... 31 20 TS.2.2 Changes in the Drivers of the Climate System .................................................................... 35 21 TS.2.3 Upper Air Temperatures and Atmospheric Circulation....................................................... 37 22 Box TS.3: Low-Likelihood, High-Warming Storylines ........................................................................ 38 23 TS.2.4 The Ocean ............................................................................................................................ 40 24 TS.2.5 The Cryosphere.................................................................................................................... 42 25 Box TS.4: Sea Level ................................................................................................................................. 44 26 Box TS.5: The Carbon Cycle .................................................................................................................. 46 27 TS.2.6 Land Climate, Including Biosphere and Extremes .............................................................. 48 28 Box TS.6: Water Cycle ............................................................................................................................ 50 29 Infographic TS.1: Climate Futures. ............................................................................................................. 52 30 TS.3 Understanding the Climate System Response and Implications for Limiting Global Warming 55 31 TS.3.1 Radiative Forcing and Energy Budget ................................................................................. 55 32 TS.3.2 Climate Sensitivity and Earth-System Feedbacks ............................................................... 57 33 TS.3.2.1 Equilibrium Climate Sensitivity, Transient Climate Response, and Transient Climate 34 Response to Cumulative Carbon-dioxide Emissions............................................................. 57 35 TS.3.2.2 Earth System Feedbacks ...................................................................................................... 59 36 TS.3.3 Temperature Stabilization, Net Zero Emissions and Mitigation ......................................... 61 37 TS.3.3.1 Remaining Carbon Budgets and Temperature Stabilization................................................ 61 38 TS.3.3.2 Carbon Dioxide Removal .................................................................................................... 64 39 TS.3.3.3 Relating Different Forcing Agents ...................................................................................... 66 40 Box TS.7: Climate and Air Quality Responses to Short-lived Climate Forcers in Shared Do Not Cite, Quote or Distribute TS-2 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Socioeconomic Pathways ....................................................................................................... 68 2 Box TS.8: Earth System Response to Solar Radiation Modification .................................................. 69 3 Box TS.9: Irreversibility, Tipping Points and Abrupt Changes ......................................................... 71 4 TS.4 Regional Climate Change ................................................................................................................. 72 5 TS.4.1 Generation and Communication of Regional Climate Change Information ....................... 72 6 TS.4.1.1 Sources and Methodologies for Generating Regional Climate Information ....................... 73 7 Box TS.10: Event Attribution................................................................................................................... 73 8 TS.4.1.2 Regional Climate Information Distillation and Climate Services ....................................... 75 9 Box TS.11: Climate Services ..................................................................................................................... 76 10 Box TS.12: Multiple Lines of Evidence for Assessing Regional Climate Change and the Interactive 11 Atlas 76 12 TS.4.2 Drivers of Regional Climate Variability and Change.......................................................... 77 13 TS.4.2.1 Regional Fingerprints of Anthropogenic and Natural Forcing ............................................ 78 14 TS.4.2.2 Modes of Variability and Regional Teleconnections .......................................................... 78 15 TS.4.2.3 Interplay Between Drivers of Climate Variability and Change at Regional Scales ............ 81 16 Box TS.13: Monsoons ................................................................................................................................ 83 17 TS.4.3 Regional Climate Change and Implications for Climate Extremes and Climatic Impact- 18 Drivers 85 19 TS.4.3.1 Common Regional Changes in Climatic Impact-Drivers .................................................... 87 20 TS.4.3.2 Region-by-Region Changes in Climatic Impact-Drivers..................................................... 89 21 TS.4.3.2.1 Africa ............................................................................................................................... 90 22 TS.4.3.2.2 Asia .................................................................................................................................. 91 23 TS.4.3.2.3 Australasia ....................................................................................................................... 93 24 TS.4.3.2.4 Central and South America .............................................................................................. 94 25 TS.4.3.2.5 Europe.............................................................................................................................. 95 26 TS.4.3.2.6 North America ................................................................................................................. 95 27 TS.4.3.2.7 Small Islands.................................................................................................................... 96 28 TS.4.3.2.8 Polar ................................................................................................................................. 97 29 TS.4.3.2.9 Ocean ............................................................................................................................... 98 30 TS.4.3.2.10 Other Typological Domains ............................................................................................ 99 31 Box TS.14: Urban Areas ........................................................................................................................... 99 32 Figures .......................................................................................................................................................... 101 33 34 Do Not Cite, Quote or Distribute TS-3 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Introduction 2 3 The Working Group I (WGI) contribution to the Intergovernmental Panel on Climate Change Sixth 4 Assessment Report (AR6) assess the physical science basis of climate change. As part of that contribution, 5 this Technical Summary (TS) is designed to bridge between the comprehensive assessment of the WGI 6 Chapters and its Summary for Policymakers (SPM). It is primarily built from the Executive Summaries of 7 the individual chapters and atlas and provides a synthesis of key findings based on multiple lines of evidence 8 (e.g., analyses of observations, models, paleoclimate information and understanding of physical, chemical 9 and biological processes and components of the climate system). All the findings and figures here are 10 supported by and traceable to the underlying chapters, with relevant chapter sections indicated in curly 11 brackets. 12 13 Throughout this Technical Summary, key assessment findings are reported using the IPCC calibrated 14 uncertainty language (Chapter 1, Box 1.1). Two calibrated approaches are used to communicate the degree 15 of certainty in key findings, which are based on author teams’ evaluations of underlying scientific 16 understanding: 17 18 (1) Confidence1 is a qualitative measure of the validity of a finding, based on the type, amount, quality 19 and consistency of evidence (e.g., data, mechanistic understanding, theory, models, expert judgment) 20 and the degree of agreement; and 21 (2) Likelihood2 provides a quantified measure of confidence in a finding expressed probabilistically 22 (e.g., based on statistical analysis of observations or model results, or both, and expert judgement by 23 the author team or from a formal quantitative survey of expert views, or both). 24 25 Where there is sufficient scientific confidence, findings can also be formulated as statements of fact without 26 uncertainty qualifiers. Throughout IPCC reports, the calibrated language is clearly identified by being 27 typeset in italics. 28 29 The context and progress in climate science (TS.1) is followed by a Cross-Section Box TS.1 on global 30 surface temperature change. TS.2 provides information about past and future large-scale changes in all 31 components of the climate system. TS.3 summarises knowledge and understanding of climate forcings, 32 feedbacks and responses. Infographic TS.1 uses a storyline approach to integrate findings on possible climate 33 futures. Finally, TS.4 provides a synthesis of climate information at regional scales.3 The list of acronyms 34 used in the WGI Report is in Annex VIII. 35 36 The AR6 WGI report promotes best practices in traceability and reproducibility, including through adoption 37 of the Findable, Accessible, Interoperable, and Reusable (FAIR) principles for scientific data. Each chapter 38 has a data table (in its Supplementary Material) documenting the input data and code used to generate its 39 figures and tables. In addition, a collection of data and code from the report has been made freely-available 40 online via long-term archives. ([URL to access WGI data to be added by 30 June]) 41 1 In this Technical Summary, the following summary terms are used to describe the available evidence: limited, medium, or robust; and for the degree of agreement: low, medium, or high. A level of confidence is expressed using five qualifiers: very low, low, medium, high, and very high, and typeset in italics, e.g., medium confidence. For a given evidence and agreement statement, different confidence levels can be assigned, but increasing levels of evidence and degrees of agreement are correlated with increasing confidence (see Chapter 1, Box 1.1 for more details). 2 In this Technical Summary, the following terms have been used to indicate the assessed likelihood of an outcome or a result: virtually certain 99– 100% probability, very likely 90–100%, likely 66–100%, about as likely as not 33–66%, unlikely 0–33%, very unlikely 0–10%, exceptionally unlikely 0–1%. Additional terms (extremely likely: 95–100%, more likely than not >50–100%, and extremely unlikely 0–5%) may also be used when appropriate. Assessed likelihood is typeset in italics, e.g., very likely (see Chapter 1, Box 1.1 for more details). Throughout the WGI report and unless stated otherwise, uncertainty is quantified using 90% uncertainty intervals. The 90% uncertainty interval, reported in square brackets [x to y], is estimated to have a 90% likelihood of covering the value that is being estimated. The range encompasses the median value, and there is an estimated 10% combined likelihood of the value being below the lower end of the range (x) and above its upper end (y). Often the distribution will be considered symmetric about the corresponding best estimate, but this is not always the case. In this report, an assessed 90% uncertainty interval is referred to as a ‘very likely range’. Similarly, an assessed 66% uncertainty interval is referred to as a ‘likely range’. 3 The regional trackback matrices that provide the location of the assessment findings synthesized in TS.4 are in the Supplementary Material (SM) for Chapter 10. Do Not Cite, Quote or Distribute TS-4 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 These FAIR principles are central to the WGI Interactive Atlas, an online tool that complements the WGI 2 Report by providing flexible spatial and temporal analyses of past, observed and projected climate change 3 information. ([URL to access WGI data to be added by 30 June]). 4 5 Regarding the representation of robustness and uncertainty in maps, the method chosen for the AR64 differs 6 from the method used in the Sixth Assessment Report (AR5). This choice is based on new research in the 7 visualization of uncertainty and on user surveys. 8 9 10 [START BOX TS.1 HERE] 11 12 Box TS.1: Core Concepts Central to This Report 13 14 This box provides short descriptions of key concepts which are relevant to the AR6 WGI assessment, with a 15 focus on their use in the Technical Summary and the Summary for Policymakers. The Glossary (Annex VII) 16 includes more information on these concepts along with definitions of many other important terms and 17 concepts used in this Report. 18 19 Characteristics of Climate Change Assessment 20 21 Global warming: Global warming refers to the change of global surface temperature relative to a baseline 22 depending upon the application. Specific global warming levels, such as 1.5°C, 2°C, 3°C or 4°C, are defined 23 as changes in global surface temperature relative to the years 1850–1900 as the baseline (the earliest period of 24 reliable observations with sufficient geographic coverage). They are used to assess and communicate 25 information about global and regional changes, linking to scenarios and used as a common basis for WGII and 26 WGIII assessments. (TS.1.3, Cross-Section Box TS.1) {1.4.1, 1.6.2, 4.6.1, Cross-Chapter Boxes 1.5, 2.3, 11.1, 27 and 12.1, Atlas.3-Atlas.11, Glossary} 28 29 Emergence: Emergence refers to the experience or appearance of novel conditions of a particular climate 30 variable in a given region. This concept is often expressed as the ratio of the change in a climate variable 31 relative to the amplitude of natural variations of that variable (often termed a ‘signal-to-noise’ ratio, with 32 emergence occurring at a defined threshold of this ratio). Emergence can be expressed in terms of a time or a 33 global warming level at which the novel conditions appear and can be estimated using observations or model 34 simulations. (TS.1.2.3, TS.4.2) {1.4.2, FAQ 1.2, 7.5.5, 10.3, 10.4, 12.5.2, Cross-Chapter Box Atlas.1, 35 Glossary} 36 37 Cumulative carbon dioxide (CO2) emissions: The total net amount of CO2 emitted into the atmosphere as a 38 result of human activities. Given the nearly linear relationship between cumulative CO 2 emissions and 39 increases in global surface temperature, cumulative CO2 emissions are relevant for understanding how past 40 and future CO2 emissions affect global surface temperature. A related term – remaining carbon budget – is 41 used to describe the total net amount of CO2 that could be released in the future by human activities while 42 keeping global warming to a specific global warming level, such as 1.5°C, taking into account the warming 43 contribution from non-CO2 forcers as well. The remaining carbon budget is expressed from a recent specified 44 date, while the total carbon budget is expressed starting from the pre-industrial period. (TS.1.3, TS.3.4) {1.6.3, 45 5.5, Glossary} 46 47 Net zero CO2 emissions: A condition that occurs when the amount of CO2 emitted into the atmosphere by 48 human activities equals the amount of CO2 removed from the atmosphere by human activities over a specified 49 period of time. Net negative CO2 emissions occur when anthropogenic removals exceed anthropogenic 50 emissions. (TS.3.3) {Box 1.4, Glossary} 4 The AR6 figures follow either one of the following approaches. For observations, the absence of ‘x’ symbols shows areas with statistical significance (while the presence of ‘x’ indicates non-significance). For model projections, the method offers two approaches with varying complexity. In the simple approach, high agreement (≥80%) is indicated with no overlay, and diagonal lines (///) shows low agreement (<80%); In the advanced approach, areas with no overlay display robust signal (≥66% of models show change greater than the variability threshold and ≥80% of all models agree on the sign of change), reverse diagonal lines (\\\) shows no robust signal, and crossed lines show conflicting signals (i.e., significant change but low agreement). Cross-Chapter Box Atlas.1 provides more information on the AR6 method for visualizing robustness and uncertainty on maps. Do Not Cite, Quote or Distribute TS-5 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Human Influence on the Climate System 2 3 Earth’s energy imbalance: In a stable climate, the amount of energy that the Earth receives from the Sun is 4 approximately in balance with the amount of energy that is lost to space in the form of reflected sunlight and 5 thermal radiation. ‘Climate drivers’, such as an increase in greenhouse gases or aerosols, interfere with this 6 balance, causing the system to either gain or lose energy. The strength of a climate driver is quantified by its 7 effective radiative forcing (ERF), measured in W m-2. Positive ERF leads to warming and negative ERF leads 8 to cooling. That warming or cooling in turn can change the energy imbalance through many positive 9 (amplifying) or negative (dampening) climate feedbacks. (TS.2.2, TS.3.1, TS.3.2) {2.2.8, 7.2, 7.3, 7.4, Box 10 7.1, Box 7.2, Glossary} 11 12 Attribution: Attribution is the process of evaluating the relative contributions of multiple causal factors to an 13 observed change in climate variables (e.g., global surface temperature, global mean sea level change), or to 14 the occurrence of extreme weather or climate-related events. Attributed causal factors include human activities 15 (such as increases in greenhouse gas concentration and aerosols, or land-use change) or natural external drivers 16 (solar and volcanic influences), and in some cases internal variability. (TS.1.2.4, TS.2, Box TS.10) {Cross- 17 Working Group Box: Attribution, 3.5, 3.8, 10.4, 11.2.4, Glossary} 18 19 Committed change, long-term commitment: Changes in the climate system, resulting from past, present and 20 future human activities, which will continue long into the future (centuries to millennia) even with strong 21 reductions in greenhouse gas emissions. Some aspects of the climate system, including the terrestrial 22 biosphere, deep ocean and the cryosphere, respond much more slowly than surface temperatures to changes in 23 greenhouse gas concentrations. As a result, there are already substantial committed changes associated with 24 past greenhouse gas emissions. For example, global mean sea level will continue to rise for thousands of years, 25 even if future CO2 emissions are reduced to net zero and global warming halted, as excess energy due to past 26 emissions continues to propagate into the deep ocean and as glaciers and ice sheets continue to melt. (TS.2.1, 27 Box TS.4, Box TS.9) {1.2.1, 1.3, Box 1.2, Cross-Chapter Box 5.3} 28 29 Climate Information for Regional Climate Change and Risk Assessment 30 31 Distillation: The process of synthesizing information about climate change from multiple lines of evidence 32 obtained from a variety of sources, taking into account user context and values. It leads to an increase in the 33 usability, usefulness, and relevance of climate information, enhances stakeholder trust, and expands the 34 foundation of evidence used in climate services. It is particularly relevant in the context of co-producing 35 regional-scale climate information to support decision-making. (TS.4.1, Box TS.11) {10.1, 10.5, 12.6} 36 37 (Climate change) risk: The concept of risk is a key aspect of how the IPCC assesses and communicates to 38 decision-makers about the potential for adverse consequences for human or ecological systems, recognising 39 the diversity of values and objectives associated with such systems. In the context of climate change, risks can 40 arise from potential impacts of climate change as well as human responses to climate change. WGI contributes 41 to the common IPCC risk framing through the assessment of relevant climate information, including climatic 42 impact-drivers and low-likelihood, high impact outcomes. (TS.1.4, TS.4.1, Box TS.4) {Cross-Chapter Boxes 43 1.3 and 12.1, Glossary} 44 45 Climatic impact-drivers: Physical climate system conditions (e.g., means, events, extremes) that can be 46 directly connected with having impacts on human or ecological systems are described as ‘climatic impact- 47 drivers’ (CIDs) without anticipating whether their impacts are detrimental (i.e., as for hazards in the context 48 of climate change risks) or provide potential opportunities. A range of indices may capture the sector- or 49 application-relevant characteristics of a climatic impact-driver and can reflect exceedances of identified 50 tolerance thresholds. (TS.1.4, TS.4.3) {12.1-12.3, FAQ12.1, Glossary} 51 52 Storylines: The term storyline is used both in connection to scenarios (related to a future trajectory of emissions 53 or socio-economic developments) or to describe plausible trajectories of weather and climate conditions or 54 events, especially those related to high levels of risk. Physical climate storylines are introduced in AR6 to 55 explore uncertainties in climate change and natural climate variability, to develop and communicate integrated Do Not Cite, Quote or Distribute TS-6 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 and context-relevant regional climate information, and to address issues with deep uncertainty5, including low- 2 likelihood, high-impact outcomes. (TS.1.4, Box TS.3, Infographic TS.1) {1.4.4, Box 10.2, Glossary} 3 4 Low-likelihood, high impact outcomes: Events whose probability of occurrence is low or not well known (as 5 in the context of deep uncertainty) but whose potential impacts on society and ecosystems could be high. To 6 better inform risk assessment and decision-making, such low-likelihood outcomes are considered if they are 7 associated with very large consequences and may therefore constitute material risks, even though those 8 consequences do not necessarily represent the most likely outcome. (TS.1.4, Box TS.3, Figure TS.6) {1.4.4, 9 4.8, Cross Chapter Box 1.3, Glossary} 10 11 [END BOX TS.1 HERE] 12 13 14 As part of the AR6 cycle, the IPCC produced three Special Reports in 2018 and 2019: the Special Report on 15 Global Warming of 1.5°C (SR1.5), the Special Report on Oceans and Cryosphere in a Changing Climate 16 (SROCC), and the Special Report on Climate Change and Land (SRCCL). 17 18 The AR6 WGI Report provides a full and comprehensive assessment of the physical science basis of climate 19 change that builds on the previous assessments and these Special Reports and consider new information and 20 knowledge from the recent scientific literature6, including longer observational datasets, new scenarios and 21 model results. 22 23 The structure of the AR6 WGI report is designed to enhance the visibility of knowledge developments and to 24 facilitate the integration of multiple lines of evidence, thereby improving confidence in findings. The Report 25 has been peer-reviewed by the scientific community and governments (Annex X provides the Expert 26 Reviewer list). The substantive introduction provided by Chapter 1 is followed by a first set of chapters 27 dedicated to large-scale climate knowledge (Chapters 2–4), which encompasses observations and 28 paleoclimate evidence, causes of observed changes, and projections, and are complemented by Chapter 11 29 for large-scale changes in extremes. The second set of chapters (Chapters 5–9) is orientated around the 30 understanding of key climate system components and processes, including the global cycles of carbon, 31 energy and water; short-lived climate forcers and their link to air quality; the ocean, cryosphere and sea level 32 change. The last set of chapters (Chapters 10–12 and the Atlas) is dedicated to the assessment and distillation 33 of regional climate information from multiple lines of evidence at sub-continental to local scales (including 34 urban climate), with a focus on recent and projected regional changes in mean climate, extremes, and 35 climatic impact-drivers. The new online Interactive Atlas allows users to interact in a flexible manner 36 through maps, time series and summary statistics with climate information for a set of updated WGI 37 reference regions. The Report also includes 34 Frequently Asked Questions and answers for the general 38 public. [URL to access FAQs to be added by 30 June] 39 40 Together, this Technical Summary and the underlying chapters aim at providing a comprehensive picture of 41 knowledge progress since the WGI AR5. Multiple lines of scientific evidence confirm that the climate is 42 changing due to human influence. Important advances in the ability to understand past, present, and possible 43 future changes should result in better-informed decision-making. 44 45 Some of the new results and main updates to key findings in AR6 WGI compared to AR5, SR1.5, SRCCL, 46 and SROCC are summarized below. Relevant Technical Summary sections with further details are shown in 5 Although not a core concept of the WGI Report, deep uncertainty is used in the Technical Summary in the following sense: ‘A situation of deep uncertainty exists when experts or stakeholders do not know or cannot agree on: (1) appropriate conceptual models that describe relationships among key driving forces in a system; (2) the probability distributions used to represent uncertainty about key variables and parameters; and/or (3) how to weigh and value desirable alternative outcomes’ (Lempert et al., 2003). Lempert, R. J., Popper, S. W., and Bankes, S. C. (2003). Shaping the next one hundred years: New methods for quantitative long-term strategy analysis (MR-1626-RPC). Santa Monica, CA: The RAND Pardee Center. 6 The assessment covers scientific literature accepted for publication by 31 January 2021. Do Not Cite, Quote or Distribute TS-7 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 parenthesis after each bullet point. 2 3 Selected Updates and/or New Results since AR5 4 5 ● Human influence7 on the climate system is now an established fact: The Fourth Assessment 6 Report (AR4) stated in 2007 that ‘warming of the climate system is unequivocal’, and the AR5 7 stated in 2013 that ‘human influence on the climate system is clear’. Combined evidence from across 8 the climate system strengthens this finding. It is unequivocal that the increase of CO2, methane 9 (CH4) and nitrous oxide (N2O) in the atmosphere over the industrial era is the result of human 10 activities and that human influence is the principal driver of many changes observed across the 11 atmosphere, ocean, cryosphere and biosphere. (TS.1.2, TS.2.1) 12 ● Observed global warming to date: A combination of improved observational records and a series 13 of very warm years since AR5 have resulted in a substantial increase in the estimated level of global 14 warming to date. The contribution of changes in observational understanding alone between AR5 15 and AR6 leads to an increase of about 0.1°C in the estimated warming since 1850–1900. For the 16 decade 2011–2020, the increase in global surface temperature since 1850–1900 is assessed to be 17 1.09 [0.95 to 1.20] °C.8 Estimates of crossing times of global warming levels and estimates of 18 remaining carbon budgets are updated accordingly. (TS.1.2, Cross-Section Box TS.1) 19 ● Paleoclimate evidence: The AR5 assessed that many of the changes observed since the 1950s are 20 unprecedented over decades to millennia. Updated paleoclimate evidence strengthens this 21 assessment; over the past several decades, key indicators of the climate system are increasingly at 22 levels unseen in centuries to millennia and are changing at rates unprecedented in at least the last 23 2000 years. (Box TS.2, TS.2) 24 ● Updated assessment of recent warming: The AR5 reported a smaller rate of increase in global 25 mean surface temperature over the period 1998–2012 than the rate calculated since 1951. Based on 26 updated observational datasets showing a larger trend over 1998–2012 than earlier estimates, there is 27 now high confidence that the observed 1998–2012 global surface temperature trend is consistent 28 with ensembles of climate model simulations, and there is now very high confidence that the slower 29 rate of global surface temperature increase observed over this period was a temporary event induced 30 by internal and naturally forced variability that partly offset the anthropogenic surface warming 31 trend over this period, while heat uptake continued to increase in the ocean. Since 2012, strong 32 warming has been observed, with the past five years (2016–2020) being the hottest five-year period 33 in the instrumental record since at least 1850 (high confidence). (TS.1.2, Cross-Section Box TS.1) 34 ● Magnitude of climate system response: In this Report, it has been possible to reduce the long- 35 standing uncertainty ranges for metrics that quantify the response of the climate system to radiative 36 forcing, such as the equilibrium climate sensitivity (ECS) and the transient climate response (TCR), 37 due to substantial advances (e.g., a 50% reduction in the uncertainty range of cloud feedbacks) and 38 improved integration of multiple lines of evidence, including paleoclimate information. Improved 39 quantification of effective radiative forcing, the climate system radiative response, and the observed 40 energy increase in the Earth system over the past five decades demonstrate improved consistency 41 between independent estimates of climate drivers, the combined climate feedbacks, and the observed 42 energy increase relative to AR5. (TS.3.2) 43 ● Improved constraints on projections of future climate change: For the first time in an IPCC 44 report, the assessed future change in global surface temperature is consistently constructed by 45 combining scenario-based projections (which the AR5 focused on) with observational constraints 46 based on past simulations of warming as well as the updated assessment of ECS and TCR. In 47 addition, initialized forecasts have been used for the period 2019–2018. The inclusion of these lines 48 of evidence reduces the assessed uncertainty for each scenario. (TS.1.3, Cross-Section Box TS.1) 7 Human influence on the climate system refers to human-driven activities that lead to changes in the climate system due to perturbations of the Earth’s energy budget (also called anthropogenic forcing). Human influence results from emissions of greenhouse gases, aerosols and tropospheric ozone precursors, ozone-depleting substances, and land-use change. 8 Throughout the WGI report and unless stated otherwise, uncertainty is quantified using 90% uncertainty intervals. The 90% uncertainty interval, reported in square brackets [x to y], is estimated to have a 90% likelihood of covering the value that is being estimated. The range encompasses the median value and there is an estimated 10% combined likelihood of the value being below the lower end of the range (x) and above its upper end (y). Often the distribution will be considered symmetric about the corresponding best estimate, but this is not always the case. In this report, an assessed 90% uncertainty interval is referred to as a ‘very likely range’. Similarly, an assessed 66% uncertainty interval is referred to as a ‘likely range’ Do Not Cite, Quote or Distribute TS-8 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 ● Air quality: The AR5 assessed that projections of air quality are driven primarily by precursor 2 emissions, including methane. New scenarios explore a diversity of future options in air pollution 3 management. The AR6 WGI reports rapid recent shifts in the geographical distribution of some of 4 these precursor emissions, confirms the AR5 finding, and shows higher warming effects of short- 5 lived climate forcers in scenarios with the highest air pollution. (TS.1.3, TS.2.2, Box TS.7) 6 ● Effects of short-lived climate forcers on global warming: The AR5 assessed the radiative forcing 7 for emitted compounds. The AR6 has extended this by assessing the emission-based effective 8 radiative forcings (ERFs) also accounting for aerosol–cloud interactions. The best estimates of ERF 9 attributed to sulphur dioxide (SO2) and CH4 emissions are substantially greater than in AR5, while 10 that of black carbon is substantially reduced. The magnitude of uncertainty in the ERF due to black 11 carbon emissions has also been reduced relative to AR5. 12 ● Global water cycle: The AR5 assessed that anthropogenic influences have likely affected the global 13 water cycle since 1960. The dedicated chapter in the AR6 WGI (Chapter 8) concludes with high 14 confidence that human-caused climate change has driven detectable changes in the global water 15 cycle since the mid-20th century, with a better understanding of the response to aerosol and 16 greenhouse gas changes. The AR6 WGI further projects with high confidence an increase in the 17 variability of the water cycle in most regions of the world and under all emissions scenarios (Box 18 TS.6) 19 ● Extreme events: The AR5 assessed that human influence had been detected in changes in some 20 climate extremes. A dedicated chapter in the AR6 (Chapter 11) concludes that it is now an 21 established fact that human-induced greenhouse gas emissions have led to an increased frequency 22 and/or intensity of some weather and climate extremes since 1850, in particular for temperature 23 extremes. Evidence of observed changes and attribution to human influence has strengthened for 24 several types of extremes since AR5, in particular for extreme precipitation, droughts, tropical 25 cyclones and compound extremes (including fire weather). (TS.1.2, TS.2.1) 26 27 Selected Updates and/or New Results Since AR5 and SR1.5 28 29 ● Timing of crossing 1.5°C global warming: Slightly different approaches are used in SR1.5 and in 30 this Report. SR1.5 assessed a likely range of 2030 to 2052 for reaching a global warming level of 31 1.5°C (for a 30-year period), assuming a continued, constant rate of warming. In AR6, combining 32 the larger estimate of global warming to date and the assessed climate response to all considered 33 scenarios, the central estimate of crossing 1.5°C of global warming (for a 20-year period) occurs in 34 the early 2030s, ten years earlier than the midpoint of the likely range assessed in the SR1.5, 35 assuming no major volcanic eruption. (TS.1.3, Cross-Section Box TS.1) 36 ● Remaining carbon budgets: The AR5 had assessed the transient climate response to cumulative 37 emissions of CO2 to be likely in the range of 0.8°C to 2.5°C per 1000 GtC (1 GtC = 1 PgC = 3.667 38 GtCO2), and this was also used in SR1.5. The assessment in AR6, based on multiple lines of 39 evidence, leads to a narrower likely range of 1.0°C–2.3°C per 1000 GtC. This has been incorporated 40 in updated estimates of remaining carbon budgets (see TS.3.3.1), together with methodological 41 improvements and recent observations. (TS.1.3, TS.3.3) 42 ● Effect of short-lived climate forcers on global warming in coming decades: The SR1.5 stated 43 that reductions in emissions of cooling aerosols partially offset greenhouse gas mitigation effects for 44 two to three decades in pathways limiting global warming to 1.5°C. The AR6 assessment updates the 45 AR5 assessment of the net cooling effect of aerosols and confirms that changes in short-lived 46 climate forcers will very likely cause further warming in the next two decades across all scenarios 47 (TS.1.3, Box TS.7) 48 ● COVID-19: Temporary emission reductions in 2020 associated with COVID-19 containment led to 49 small and positive net radiative effect (warming influence). However, global and regional climate 50 responses to this forcing are undetectable above internal climate variability due to the temporary 51 nature of emission reductions. (TS.3.3) 52 53 Selected Updates and/or New Results Since AR5, SRCCL and SROCC 54 55 ● Atmospheric concentration of methane: SRCCL reported a resumption of atmospheric methane Do Not Cite, Quote or Distribute TS-9 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 concentration growth since 2007. WGI AR6 reports a faster growth over 2014–2019 and assesses 2 growth since 2007 to be largely driven by emissions from the fossil fuels and agriculture (dominated 3 by livestock) sectors. (TS.2.2) 4 ● Land and ocean carbon sinks: SRCCL assessed that the persistence of the land carbon sink is 5 uncertain due to climate change. WGI AR6 finds that land and ocean carbon sinks are projected to 6 continue to grow until 2100 with increasing atmospheric concentrations of CO2, but the fraction of 7 emissions taken up by land and ocean is expected to decline as the CO2 concentration increases, with 8 a much larger uncertainty range for the land sink. AR5, SR1.5 and SRCCL assessed carbon dioxide 9 removal options and scenarios. WGI AR6 finds that the carbon cycle response is asymmetric for 10 pulse emissions or removals, which means that CO2 emissions would be more effective at raising 11 atmospheric CO2 than CO2 removals are at lowering atmospheric CO2. (TS.3.3, Box TS.5) 12 ● Ocean stratification increase9: Refined analyses of available observations in the AR6 lead to a 13 reassessment of the rate of increase of the global stratification in the upper 200 m to be double that 14 estimated in SROCC from 1970 to 2018. (TS.2.4) 15 ● Projected ocean oxygen loss: Future subsurface oxygen decline in new projections assessed in WGI 16 AR6 is substantially greater in 2080–2099 than assessed in SROCC. (TS.2.4) 17 ● Ice loss from glaciers and ice sheets: since SROCC, globally resolved glacier changes have 18 improved estimates of glacier mass loss over the past 20 years, and estimates of the Greenland and 19 Antarctic Ice Sheet loss have been extended to 2020. (TS.2.5) 20 ● Observed global mean sea level change: new observation-based estimates published since SROCC 21 lead to an assessed sea level rise estimate from 1901 to 2018 that is now consistent with the sum of 22 individual components and consistent with closure of the global energy budget. (Box TS.4) 23 ● Projected global mean sea level change: AR6 projections of global mean sea level are based on 24 projections from ocean thermal expansion and land ice contribution estimates, which are consistent 25 with the assessed equilibrium climate sensitivity and assessed changes in global surface temperature. 26 They are underpinned by new land ice model intercomparisons and consideration of processes 27 associated with low confidence to characterise the deep uncertainty in future ice loss from 28 Antarctica. AR6 projections based on new models and methods are broadly consistent with SROCC 29 findings. (Box TS.4) 30 31 9 Increased stratification reduces the vertical exchange of heat, salinity, oxygen, carbon, and nutrients. Stratification is an important indicator for ocean circulation. Do Not Cite, Quote or Distribute TS-10 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.1 A Changing Climate 2 3 This section introduces the assessment of the physical science basis of climate change in the AR6 and 4 presents the climate context in which this assessment takes place, recent progress in climate science and the 5 relevance of global and regional climate information for impact and risk assessments. The future emissions 6 scenarios and global warming levels, used to integrate assessments across WGI AR6, are introduced and 7 their applications for future climate projections are briefly addressed. Paleoclimate science provides a long- 8 term context for observed climate change of the past 150 years and the projected changes in the 21st century 9 and beyond (Box TS.2). The assessment of past, current and future global surface temperature changes 10 relative to the standard baselines and reference periods10 used throughout this Report is summarized in 11 Cross-Section Box TS.1. 12 13 14 TS.1.1 Context of a Changing Climate 15 16 AR6 WGI assesses new scientific evidence relevant for a world whose climate system is rapidly changing, 17 overwhelmingly due to human influence. The five IPCC assessment cycles since 1990 have comprehensively 18 and consistently laid out the rapidly accumulating evidence of a changing climate system, with the Fourth 19 Assessment Report (AR4) in 2007 being the first to conclude that warming of the climate system is 20 unequivocal. Sustained changes have been documented in all major elements of the climate system: the 21 atmosphere, land, cryosphere, biosphere and ocean (TS.2). Multiple lines of evidence indicate the recent 22 large-scale climatic changes are unprecedented in a multi-millennial context, and that they represent a 23 millennial-scale commitment for the slow-responding elements of the climate system, resulting in continued 24 worldwide loss of ice, increase in ocean heat content, sea level rise and deep ocean acidification (Box TS.2; 25 Section TS.2). {1.2.1, 1.3, Box 1.2, 2.2, 2.3, Figure 2.34, 5.1, 5.3, 9.2, 9.4-9.6, Appendix 1.A} 26 27 Earth’s climate system has evolved over many millions of years, and evidence from natural archives 28 provides a long-term perspective on observed and projected changes over the coming centuries. These 29 reconstructions of past climate also show that atmospheric CO2 concentrations and global surface 30 temperature are strongly coupled (Figure TS.1), based on evidence from a variety of proxy records over 31 multiple time scales (Box TS.2; TS.2). Levels of global warming (see Core Concepts Box) that have not 32 been seen in millions of years could be reached by 2300, depending on the emissions pathway that is 33 followed (TS.1.3). For example, there is medium confidence that, by 2300, an intermediate scenario11 used in 34 the report leads to global surface temperatures of 2.3°C–4.6°C higher than 1850–1900, similar to the mid- 35 Pliocene Warm Period (2.5°C–4°C), about 3.2 million years ago, whereas the high CO2 emissions scenario 36 SSP5-8.5 leads to temperatures of 6.6°C–14.1°C by 2300, which overlaps with the Early Eocene Climate 37 Optimum (10°C–18°C), about 50 million years ago. {Cross-Chapter Box 2.1 and 2.4, 2.3.1, 4.3.1.1, 4.7.1.2, 38 7.4.4.1} 39 40 41 [START FIGURE TS.1 HERE] 42 43 Figure TS.1: Changes in atmospheric CO2 and global surface temperature (relative to 1850-1900) from the 44 deep past to the next 300 years. The intent is to show that CO2 and temperature covary, both in the 45 past and into the future, and that projected CO2 and temperatures are similar to those only from many 46 millions of years ago. CO2 concentrations from millions of years ago are reconstructed from multiple 47 proxy records (grey dots are data from 2.2.3.1, Figure 2.3 shown with cubic-spline fit). CO2 levels for 48 the last 800,000 years through the mid-20th century are from air trapped in polar ice; recent values are 10 Several baselines or reference periods are used consistently throughout AR6 WGI. Baseline refers to a period against which anomalies (i.e., differences from the average value for the baseline period) are calculated. Examples include the 1750 baseline (used for anthropogenic radiative forcings), the 1850–1900 baseline (an approximation for pre-industrial global surface temperature from which global warming levels are calculated) and the 1995–2014 baseline (used for many climate model projections). A reference period indicates a time period over which various statistics are calculated (e.g., the near-term reference period, 2021–2040). Paleo reference periods are listed in Box TS.2. {1.4.1, Cross-Chapter Box 1.2 and 2.1} 11 Please refer to Section TS.1.3.1 for an overview of the climate change scenarios used in this report. Do Not Cite, Quote or Distribute TS-11 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 from direct air measurements (Figure TS.9). {1.2.1.2, 2.2.3, Figures 1.5, 2.4, 2.5} Global surface 2 temperature prior to 1850 is estimated from marine oxygen isotopes, one of multiple sources of 3 evidence used to assess paleo temperatures in this report. {2.3.1.1.1, Cross-Chapter Box 2.1, Figure 1} 4 Temperature of the past 170 years is the AR6 assessed mean (Cross-Section Box TS.1). {2.3.1.1} CO2 5 levels and global surface temperature change for the future are shown for three SSP scenarios (TS.1.3) 6 through 2300 CE, using Earth System Model emulators calibrated to the assessed global surface 7 temperatures. {4.7.1, Cross-Chapter Box 7.1} Their smooth trajectories do not account for inter- 8 annual to inter-decadal variability, including transient response to potential volcanic eruptions. 9 {Cross-Chapter Box 4.1} Global maps for two paleo reference periods are based on CMIP6 and pre- 10 CMIP6 multi-model means, with site-level proxy data for comparison (squares and circles are marine 11 and terrestrial, respectively) (Box TS.2). {Cross-Chapter Box 2.1, Figure 7.13} The map for 2020 is 12 an estimate of the total observed warming since 1850-1900. {Figure 1.14} Global maps at right show 13 two SSP scenarios at 2100 (2081-2100) {4.5.1} and at 2300 (2281-2300; map from CMIP6 models; 14 temperature assessed in 4.7.1). A brief account of the major climate forcings associated with past 15 global temperature changes is in Cross-Chapter Box 2.1. 16 17 [END FIGURE TS.1 HERE] 18 19 20 Understanding of the climate system’s fundamental elements is robust and well established. Scientists in the 21 19th century identified the major natural factors influencing the climate system. They also hypothesized the 22 potential for anthropogenic climate change due to carbon dioxide (CO2) emitted by combustion of fossil 23 fuels (petroleum, coal, natural gas). The principal natural drivers of climate change, including changes in 24 incoming solar radiation, volcanic activity, orbital cycles, and changes in global biogeochemical cycles, have 25 been studied systematically since the early 20th century. Other major anthropogenic drivers, such as 26 atmospheric aerosols (fine solid particles or liquid droplets), land-use change and non-CO2 greenhouse 27 gases, were identified by the 1970s. Since systematic scientific assessments began in the 1970s, the influence 28 of human activity on the warming of the climate system has evolved from theory to established fact (see also 29 TS.2). The evidence for human influence on recent climate change strengthened from the IPCC First 30 Assessment Report in 1990 to the IPCC Fifth Assessment Report in 2013/14, and is now even stronger in 31 this assessment (TS.1.2.4, TS.2). Changes across a greater number of climate system components, including 32 changes in regional climate and extremes can now be attributed to human influence (see TS.2 and TS.4). 33 {1.3.1 - 1.3.5, 3.1, 11.2, 11.9} 34 35 36 [START BOX TS.2 HERE] 37 38 Box TS.2: Paleoclimate 39 40 Paleoclimate evidence is integrated within multiple lines of evidence across the WGI report to more fully 41 understand the climate system. Paleo evidence extends instrument-based observations of climate variables 42 and climate drivers back in time, providing the long-term context needed to gauge the extent to which recent 43 and potential future changes are unusual (TS.2, Figure TS.1). Pre-industrial climate states complement 44 evidence from climate model projections by providing real-world examples of climate characteristics for past 45 global warming levels, with empirical evidence for how the slow-responding components of the climate 46 system operate over centuries to millennia – the time scale for committed climate change (Core Concepts 47 Box, Box TS.4, Box TS.9). Information about the state of the climate system during well-described 48 paleoclimate reference periods helps narrow the uncertainty range in the overall assessment of Earth’s 49 sensitivity to climate forcing (TS.3.2.1). {Cross-Chapter Box 2.1, FAQ 1.3, FAQ 2.1} 50 51 Paleoclimate reference periods. Over the long evolution of the Earth’s climate, several periods have 52 received extensive research attention as examples of distinct climate states and rapid climate transitions (Box 53 TS.2, Figure 1). These paleoclimate reference periods represent the present geological era (Cenozoic; past 65 54 million years) and are used across chapters to help structure the assessment of climate changes prior to 55 industrialization. Cross-Chapter Box 2.1 describes the reference periods, along with a brief account of their 56 climate forcings, and lists where each is discussed in other chapters. Cross-Chapter Box 2.4 summarizes Do Not Cite, Quote or Distribute TS-12 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 information on one of the reference periods, the mid-Pliocene Warm Period. The Interactive Atlas includes 2 model output from the World Climate Research Programme Coupled Model Intercomparison Project Phase 6 3 (CMIP6) for four of the paleoclimate reference periods. 4 5 6 [START BOX TS.2, FIGURE 1 HERE] 7 8 Box TS.2, Figure 1: Paleoclimate and recent reference periods, with selected key indicators. The intent of this 9 figure is to list the paleoclimate reference periods used in the WGI report, to summarize three key 10 global climate indicators, and compare CO2 with global temperature over multiple periods. (a) 11 Three large-scale climate indicators (atmospheric CO2, global surface temperature relative to 12 1850-1900, and global mean sea level relative to 1900), based on assessments in Chapter 2, with 13 confidence levels ranging from low to very high. (b) Comparison between global surface 14 temperature (relative to 1850-1900) and atmospheric CO2 concentration for multiple reference 15 periods (mid-points with 5–95% ranges). {2.2.3, 2.3.1.1, 2.3.3.3, Figure 2.34} 16 17 [END BOX TS.2, FIGURE 1 HERE] 18 19 20 Paleoclimate models and reconstructions. Climate models that target paleoclimate reference periods have 21 been featured by the IPCC since the First Assessment Report. Under the framework of CMIP6-PMIP4 22 (Paleoclimate Modelling Intercomparison Project), new protocols for model intercomparisons have been 23 developed for multiple paleoclimate reference periods. These modelling efforts have led to improved 24 understanding of the climate response to different external forcings, including changes in Earth’s orbital and 25 plate movements, solar irradiance, volcanism, ice-sheet size, and atmospheric greenhouse gases. Likewise, 26 quantitative reconstructions of climate variables from proxy records that are compared with paleoclimate 27 simulations have improved as the number of study sites and variety of proxy types have expanded, and as 28 records have been compiled into new regional and global datasets. {1.3.2, 1.5.1, Cross-Chapter Boxes 2.1 29 and 2.4} 30 31 Global surface temperature. Since AR5, updated climate forcings, improved models, new understanding of 32 the strengths and weaknesses of a growing array of proxy records, better chronologies, and more robust 33 proxy data products have led to better agreement between models and reconstructions. For global surface 34 temperature, the mid-point of the AR6-assessed range and the median of the model-simulated temperatures 35 differ by an average of 0.5°C across five reference periods; they overlap within their 90% ranges in four of 36 five cases, which together span from about 6°C [5–7]°C colder during the Last Glacial Maximum to about 37 14 [10 to 18] °C warmer during the Early Eocene, relative to 1850–1900 (Box TS.2, Figure 2a). Changes in 38 temperature by latitude in response to multiple forcings show that polar amplification (stronger warming at 39 high latitudes than the global average) is a prominent feature of the climate system across multiple climate 40 states, and the ability of models to simulate this polar amplification in past warm climates has improved 41 since AR5 (high confidence). Over the past millennium, and especially since about 1300 CE, simulated 42 global surface temperature anomalies are well within the uncertainty of reconstructions (medium 43 confidence), except for some short periods immediately following large volcanic eruptions, for which 44 different forcing datasets disagree (Box TS.2, Figure 2b). {2.3.1.1, 3.3.3.1, 3.8.2.1, 7.4.4.1.2} 45 46 47 [START BOX TS.2, FIGURE 2 HERE] 48 49 Box TS.2, Figure 2: Global surface temperature as estimated from proxy records (reconstructed) and climate 50 models (simulated). The intent of this figure is to show the agreement between observations and 51 models of global temperatures during paleo reference periods. (a) For individual paleoclimate 52 reference periods. (b) For the last millennium, with instrumental temperature (AR6 assessed 53 mean,10-year smoothed). Model uncertainties in (a) and (b) are 5-95% ranges of multi-model 54 ensemble means; reconstructed uncertainties are 5-95% ranges (medium confidence) of (a) 55 midpoints and (b) multi-method ensemble median. {2.3.1.1, Figure 2.34, Figure 3.2c, Figure 3.44} 56 Do Not Cite, Quote or Distribute TS-13 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [END BOX TS.2, FIGURE 2 HERE] 2 3 4 Equilibrium climate sensitivity. Paleoclimate data provide evidence to estimate equilibrium climate 5 sensitivity (ECS12) (TS.3.2.1). In AR6, refinements in paleo data for paleoclimate reference periods indicate 6 that ECS is very likely greater than 1.5°C and likely less than 4.5°C, which is largely consistent with other 7 lines of evidence and helps narrow the uncertainty range of the overall assessment of ECS. Some of the 8 CMIP6 climate models that have either high (> 5°C) or low (< 2°C) ECS also simulate past global surface 9 temperature changes outside the range of proxy-based reconstructions for the coldest and warmest reference 10 periods. Since AR5, independent lines of evidence, including proxy records from past warm periods and 11 glacial-interglacial cycles, indicate that sensitivity to forcing increases as temperature increases (TS.3.2.2). 12 {7.4.3.2, 7.5.3, 7.5.6, Table 7.11} 13 14 Water cycle. New hydroclimate reconstructions and model-data comparisons have improved the 15 understanding of the causes and effects of long-term changes in atmospheric and ocean circulation, including 16 monsoon variability and modes of variability (Box TS.13, TS.4.2). Climate models are able to reproduce 17 decadal drought variability on large regional scales, including the severity, persistence and spatial extent of 18 past megadroughts known from proxy records (medium confidence). Some long-standing discrepancies 19 remain, however, such as the magnitude of African monsoon precipitation during the early Holocene, 20 suggesting continuing knowledge gaps. Paleoclimate evidence shows that, in relatively high CO2 climates 21 such as the Pliocene, Walker circulation over the equatorial Pacific Ocean weakens, supporting the high 22 confidence model projections of weakened Walker cells by the end of the 21st century. {3.3.2, 8.3.1.6, 23 8.4.1.6, 8.5.2.1, 9.2} 24 25 Sea level and ice sheets. Although past and future global warming differ in their forcings, evidence from 26 paleoclimate records and modelling show that ice sheet mass and global mean sea level (GMSL) responded 27 dynamically over multiple millennia (high confidence). This evidence helps to constrain estimates of the 28 committed GMSL response to global warming (Box TS.4). For example, under a past global warming levels 29 of around 2.5°C–4°C relative to 1850–1900, like during the mid-Pliocene Warm Period, sea level was 5–25 30 m higher than 1900 (medium confidence); under past global warming levels of 10°C–18°C, like during the 31 Early Eocene, the planet was essentially ice free (high confidence). Constraints from these past warm 32 periods, combined with physical understanding, glaciology and modelling, indicate a committed long- 33 term GMSL rise over 10,000 years, reaching about 8–13 m for sustained peak global warming of 2°C and to 34 28–37 m for 5°C, which exceeds the AR5 estimate. {2.3.3.3, 9.4.1.4, 9.4.2.6, 9.6.2, 9.6.3.5} 35 36 Ocean. Since AR5, better integration of paleo-oceanographic data with modelling along with higher- 37 resolution analyses of transient changes have improved understanding of long-term ocean processes. Low- 38 latitude sea-surface temperatures at the Last Glacial Maximum cooled more than previously inferred, 39 resolving some inconsistencies noted in AR5. This paleo context supports the assessment that ongoing 40 increase in ocean heat content (OHC) represents a long-term commitment (see Core Concepts Box), 41 essentially irreversible on human time scales (high confidence). Estimates of past global OHC variations 42 generally track those of sea surface temperatures around Antarctica, underscoring the importance of 43 Southern Ocean processes in regulating deep-ocean temperatures. Paleoclimate data, along with other 44 evidence of glacial-interglacial changes, show that Antarctic Circumpolar flow strengthened and that 45 ventilation of Antarctic Bottom Water accelerated during warming intervals, facilitating release of CO2 46 stored in the deep ocean to the atmosphere. Paleo evidence suggests significant reduction of deep-ocean 47 ventilation associated with meltwater input during times of peak warmth. {2.3.1.1, 2.3.3.1, 9.2.2, 9.2.3.2} 48 49 Carbon cycle. Past climate states were associated with substantial differences in the inventories of the 50 various carbon reservoirs, including the atmosphere (TS.2.2). Since AR5, the quantification of carbon stocks 51 has improved due to the development of novel sedimentary proxies and stable-isotope analyses of air trapped 52 in polar ice. Terrestrial carbon storage decreased markedly during the Last Glacial Maximum by 300–600 12 In this report, equilibrium climate sensitivity is defined as the equilibrium (steady state) change in the surface temperature following a doubling of the atmospheric carbon dioxide (CO2) concentration from pre-industrial conditions. Do Not Cite, Quote or Distribute TS-14 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 PgC, possibly by 850 PgC when accounting for interactions with the lithosphere and ocean sediments, a 2 larger reduction than previously estimated, owing to colder and drier climate. At the same time, the storage 3 of remineralised carbon in the ocean interior increased by as much as 750–950 PgC, sufficient to balance the 4 removal of carbon from the atmosphere (200 PgC) and terrestrial biosphere reservoirs combined (high 5 confidence). {5.1.2.2} 6 7 [END BOX TS.2 HERE] 8 9 10 TS.1.2 Progress in Climate Science 11 12 TS.1.2.1 Observation-based products and their assessments 13 14 Observational capabilities have continued to improve and expand overall since AR5, enabling improved 15 consistency between independent estimates of climate drivers, the combined climate feedbacks, and the 16 observed energy and sea level increase. Satellite climate records and improved reanalyses are used as an 17 additional line of evidence for assessing changes at the global and regional scales. However, there have also 18 been reductions in some observational data coverage or continuity and limited access to data resulting from 19 data policy issues. Natural archives of past climate, such as tropical glaciers, have also been subject to losses 20 (in part due to anthropogenic climate change). {1.5.1, 1.5.2, 10.2.2} 21 22 Earth system observations are an essential driver of progress in our understanding of climate change. 23 Overall, capabilities to observe the physical climate system have continued to improve and expand. 24 Improvements are particularly evident in ocean observing networks and remote sensing systems. Records 25 from several recently instigated satellite measurement techniques are now long enough to be relevant for 26 climate assessments. For example, globally distributed, high-vertical-resolution profiles of temperature and 27 humidity in the upper troposphere and stratosphere can be obtained from the early 2000s using global 28 navigation satellite systems, leading to updated estimates of recent atmospheric warming. Improved 29 measurements of ocean heat content, warming of the land surface, ice sheet mass loss, and sea level changes 30 allow a better closure of the global energy and sea level budgets relative to AR5. For surface and balloon- 31 based networks, apparent regional data reductions result from a combination of data policy issues, data 32 curation/provision challenges, and real cessation of observations, and are to an extent counter-balanced by 33 improvements elsewhere. Limited observational records of extreme events and spatial data gaps currently 34 limit the assessment of some observed regional climate change. {1.5.1, 2.3.2, 7.2.2, Box 7.2, Cross-Chapter 35 Box 9.1, 9.6.1, 10.2.2, 10.6, 11.2, 12.4} 36 37 New paleoclimate reconstructions from natural archives have enabled more robust reconstructions of the 38 spatial and temporal patterns of past climate changes over multiple time scales (Box TS.2). However, 39 paleoclimate archives, such as tropical glaciers and modern natural archives used for calibration (e.g., corals 40 and trees), are rapidly disappearing owing to a host of pressures, including increasing temperatures (high 41 confidence). Substantial quantities of past instrumental observations of weather and other climate variables, 42 over both land and ocean, which could fill gaps in existing datasets, remain un-digitized or inaccessible. 43 These include measurements of temperature (air and sea surface), rainfall, surface pressure, wind strength 44 and direction, sunshine amount and many other variables dating back into the 19th century. {1.5.1} 45 46 Reanalyses combine observations and models (e.g., a numerical weather prediction model) using data 47 assimilation techniques to provide a spatially complete, dynamically consistent estimate of multiple variables 48 describing the evolving climate state. Since AR5, new reanalyses have been developed for the atmosphere 49 and the ocean with various combinations of increased resolution, extended records, more consistent data 50 assimilation and larger availability of uncertainty estimates. Limitations remain, for example, in how 51 reanalyses represent global-scale changes to the water cycle. Regional reanalyses use high-resolution, 52 limited-area models constrained by regional observations and with boundary conditions from global 53 reanalyses. There is high confidence that regional reanalyses better represent the frequencies of extremes and 54 variability in precipitation, surface air temperature and surface wind than global reanalyses and provide Do Not Cite, Quote or Distribute TS-15 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 estimates that are more consistent with independent observations than dynamical downscaling approaches. 2 {1.5.2, 10.2.1.2, Annex I} 3 4 5 TS.1.2.2 Climate Model Performance 6 7 Developments in the latest generation CMIP6 climate and Earth system models, including new and better 8 representation of physical, chemical and biological processes, as well as higher resolution, have improved 9 the simulation of the recent mean climate of most large-scale indicators of climate change (high confidence, 10 Figure TS.2) and many other aspects across the Earth system. Projections of the increase in global surface 11 temperature, the pattern of warming, and global mean sea level rise from previous IPCC Assessment Reports 12 and other studies are broadly consistent with subsequent observations, especially when accounting for the 13 difference in radiative forcing scenarios used for making projections and the radiative forcings that actually 14 occurred. While past warming is well simulated by the new generation of models, some individual models 15 simulate past surface warming that is either below or above that observed. The information about how well 16 models simulate past warming, as well as other insights from observations and theory, are used to assess 17 projections of global warming. (see Cross-Section Box TS.1). Increasing horizontal resolution in global 18 climate models improves the representation of small-scale features and the statistics of daily precipitation 19 (high confidence). Earth system models, which include additional biogeochemical feedbacks, often perform 20 as well as their lower-complexity global climate model counterparts, which do not account for these 21 additional feedbacks (medium confidence). {1.3.6, 1.5.3, 3.1, 3.5.1, 3.8.2, 4.3.1, 4.3.4, 7.5, 8.5.1, 9.6.3.1} 22 23 24 Climate model simulations coordinated and collected as part of the World Climate Research Programme’s 25 Coupled Model Intercomparison Project Phase 6 (CMIP6), complemented by a range of results from the 26 previous phase (CMIP5), constitute a key line of evidence supporting this Report. The latest generation of 27 CMIP6 models have an improved representation of physical processes relative to previous generations, and a 28 wider range of Earth system models now represent biogeochemical cycles. Higher-resolution models that 29 better capture smaller-scale processes are also increasingly becoming available for climate change research 30 (Figure TS.2, Panels a and b). Results from coordinated regional climate modelling initiatives, such as the 31 Coordinated Regional Climate Downscaling Experiment (CORDEX) complement and add value to the 32 CMIP global models, particularly in complex topography zones, coastal areas and small islands, as well as 33 for extremes. {1.5.3, 1.5.4, 6.2.1.2, 6.3.6, 8.5.1, 10.3.3, Atlas.1.4} 34 35 Projections of the increase in global surface temperature and the pattern of warming from previous IPCC 36 Assessment Reports and other studies are broadly consistent with subsequent observations (limited evidence, 37 high agreement), especially when accounting for the difference in radiative forcing scenarios used for 38 making projections and the radiative forcings that actually occurred. The AR5 and the SROCC projections of 39 GMSL for the 2007–2018 period have been shown to be consistent with observed trends in GMSL and 40 regional weighted mean tide gauges. {1.3.6, 9.6.3.1, Figure TS.3} 41 42 For most large-scale indicators of climate change, the simulated recent mean climate from CMIP6 models 43 underpinning this assessment have improved compared to the CMIP5 models used in AR5 (high confidence). 44 This is evident from the performance of 18 simulated atmospheric and land large-scale indicators of climate 45 change between the three generations of models (CMIP3, CMIP5, and CMIP6) when benchmarked against 46 reanalysis and observational data (Figure TS.2, Panel c). Earth system models, characterized by additional 47 biogeochemical feedbacks, often perform at least as well as related, more constrained, lower-complexity 48 models lacking these feedbacks (medium confidence). {3.8.2, 10.3.3.3} 49 The CMIP6 multi-model mean global surface temperature change from 1850–1900 to 2010–2019 is close to 50 the best estimate of the observed warming. However, some CMIP6 models simulate a warming that is below 51 or above the assessed very likely range. The CMIP6 models also reproduce surface temperature variations 52 over the past millennium, including the cooling that follows periods of intense volcanism (medium 53 confidence). For upper air temperature, an overestimation of the upper tropical troposphere warming by Do Not Cite, Quote or Distribute TS-16 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 about 0.1°C per decade between 1979 and 2014 persists in most CMIP5 and CMIP6 models (medium 2 confidence), whereas the differences between simulated and improved satellite-derived estimates of change 3 in global mean temperature through the depth of the stratosphere have decreased. {3.3.1} 4 Some CMIP6 models demonstrate an improvement in how clouds are represented. CMIP5 models 5 commonly displayed a negative shortwave cloud radiative effect that was too weak in the present climate. 6 These errors have been reduced, especially over the Southern Ocean, due to a more realistic simulation of 7 supercooled liquid droplets with sufficient numbers and an associated increase in the cloud optical depth. 8 Because a negative cloud optical depth feedback in response to surface warming results from ‘brightening’ 9 of clouds via active phase change from ice to liquid cloud particles (increasing their shortwave cloud 10 radiative effect), the extratropical cloud shortwave feedback in CMIP6 models tends to be less negative, 11 leading to a better agreement with observational estimates (medium confidence). CMIP6 models generally 12 represent more processes that drive aerosol–cloud interactions than the previous generation of climate 13 models, but there is only medium confidence that those enhancements improve their fitness-for-purpose of 14 simulating radiative forcing of aerosol–cloud interactions. {7.4.2, FAQ 7.2, 6.4} 15 16 CMIP6 models still have deficiencies in simulating precipitation patterns, particularly in the tropical ocean. 17 Increasing horizontal resolution in global climate models improves the representation of small-scale features 18 and the statistics of daily precipitation (high confidence). There is high confidence that high-resolution 19 global, regional and hydrological models provide a better representation of land surfaces, including 20 topography, vegetation and land-use change, which can improve the accuracy of simulations of regional 21 changes in the terrestrial water cycle. {3.3.2, 8.5.1, 10.3.3, 11.2.3} 22 23 There is high confidence that climate models can reproduce the recent observed mean state and overall 24 warming of temperature extremes globally and in most regions, although the magnitude of the trends may 25 differ. There is high confidence in the ability of models to capture the large-scale spatial distribution of 26 precipitation extremes over land. The overall performance of CMIP6 models in simulating the intensity and 27 frequency of extreme precipitation is similar to that of CMIP5 models (high confidence). {Cross-Chapter 28 Box 3.2, 11.3.3, 11.4.3} 29 30 The structure and magnitude of multi-model mean ocean temperature biases have not changed substantially 31 between CMIP5 and CMIP6 (medium confidence). Since AR5, there is improved consistency between recent 32 observed estimates and model simulations of changes in upper (<700 m) ocean heat content. The mean zonal 33 and overturning circulations of the Southern Ocean and the mean overturning circulation of the North 34 Atlantic (AMOC) are broadly reproduced by CMIP5 and CMIP6 models. {3.5.1, 3.5.4, 9.2.3, 9.3.2, 9.4.2} 35 36 CMIP6 models better simulate the sensitivity of Arctic sea ice area to anthropogenic CO2 emissions, and 37 thus better capture the time evolution of the satellite-observed Arctic sea ice loss (high confidence). The 38 ability to model ice-sheet processes has improved substantially since AR5. As a consequence, we have 39 medium confidence in the representation of key processes related to surface-mass balance and retreat of the 40 grounding-line (the junction between a grounded ice sheet and an ice shelf, where the ice starts to float) in 41 the absence of instabilities. However, there remains low confidence in simulations of ice-sheet instabilities, 42 ice-shelf disintegration and basal melting owing to their high sensitivity to both uncertain oceanic forcing 43 and uncertain boundary conditions and parameters. {1.5.3, 2.3.2, 3.4.1, 3.4.2, 3.8.2, 9.3.1, 9.3.2, 9.4.1, 9.4.2} 44 45 CMIP6 models are able to reproduce most aspects of the spatial structure and variance of the El Niño- 46 Southern Oscillation (ENSO) and Indian Ocean Basin and Dipole modes of variability (medium confidence). 47 However, despite a slight improvement in CMIP6, some underlying processes are still poorly represented. 48 Models reproduce observed spatial features and variance of the Southern Annular Mode (SAM) and 49 Northern Annular Modes (NAM) very well (high confidence). The summertime SAM trend is well captured, 50 with CMIP6 models outperforming CMIP5 models (medium confidence). By contrast, the cause of the 51 Northern Annular Mode (NAM) trend towards its positive phase is not well understood. In the Tropical 52 Atlantic basin, which contains the Atlantic Zonal and Meridional modes, major biases in modelled mean 53 state and variability remain. Model performance is limited in reproducing sea surface temperature anomalies 54 for decadal modes of variability, despite improvements from CMIP5 to CMIP6 (medium confidence) (see Do Not Cite, Quote or Distribute TS-17 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 also TS.1.4.2.2, Table TS.4). {3.7.3-3.7.7} 2 3 Earth system models simulate globally averaged land carbon sinks within the range of observation-based 4 estimates (high confidence), but global-scale agreement masks large regional disagreements. There is also 5 high confidence that the ESMs simulate the weakening of the global net flux of CO2 into the ocean during 6 the 1990s, as well as the strengthening of the flux from 2000. {3.6.1, 3.6.2} 7 8 Two important quantities used to estimate how the climate system responds to changes in GHG 9 concentrations are the equilibrium climate sensitivity (ECS) and transient climate response (TCR13). The 10 CMIP6 ensemble has broader ranges of ECS and TCR values than CMIP5 (see TS.3.2 for the assessed 11 range). These higher sensitivity values can, in some models, be traced to changes in extratropical cloud 12 feedbacks (medium confidence). To combine evidence from CMIP6 models and independent assessments of 13 ECS and TCR, various emulators are used throughout the report. Emulators are a broad class of simple 14 climate models or statistical methods that reproduce the behaviour of complex ESMs to represent key 15 characteristics of the climate system, such as global surface temperature and sea level projections. The main 16 application of emulators in AR6 is to extrapolate insights from ESMs and observational constraints to 17 produce projections from a larger set of emissions scenarios, which is achieved due to their computational 18 efficiency. These emulated projections are also used for scenario classification in WGIII. {Box 4.1, 4.3.4., 19 7.4.2, 7.5.6, CCB7.1, FAQ7.2} 20 21 22 [START FIGURE TS.2 HERE] 23 24 Figure TS.2: Progress in climate models. The intent is to show present improvements in climate models in 25 resolution, complexity and representation of key variables. (a) Evolution of model horizontal 26 resolution and vertical levels (based on Figure 1.19); (b) Evolution of inclusion of processes and 27 resolution from CMIP Phase 3 (CMIP3) to CMIP6 (Annex II). (c) Centred pattern correlations 28 between models and observations for the annual mean climatology over the period 1980–1999. 29 Results are shown for individual CMIP3 (cyan), CMIP5 (blue) and CMIP6 (red) models (one 30 ensemble member is used) as short lines, along with the corresponding ensemble averages (long 31 lines). The correlations are shown between the models and the primary reference observational data 32 set (from left to right: ERA5, GPCP-SG, CERES-EBAF, CERES-EBAF, CERES-EBAF, CERES- 33 EBAF, JR-55, ERA5, ERA5, ERA5, ERA5, ERA5, ERA5, AIRS, ERA5, ESACCI-Soilmoisture, 34 LAI3g, MTE). In addition, the correlation between the primary reference and additional observational 35 data sets (from left to right: NCEP, GHCN, -, -, -, -, ERA5, HadISST, NCEP, NCEP, NCEP, NCEP, 36 NCEP, NCEP, ERA5, NCEP, -, -, FLUXCOM) are shown (solid grey circles) if available. To ensure a 37 fair comparison across a range of model resolutions, the pattern correlations are computed after 38 regridding all datasets to a resolution of 4º in longitude and 5º in latitude. (Expanded from Figure 39 3.43; produced with ESMValTool version 2). 40 41 [END FIGURE TS.2 HERE] 42 43 44 TS.1.2.3 Understanding Climate Variability and Emerging Changes 45 46 Observed changes in climate are unequivocal at the global scale and are increasingly apparent on regional 47 and local spatial scales. Both the rate of long-term change and the amplitude of year-to-year variations differ 48 between regions and across climate variables, thus influencing when changes emerge or become apparent 49 compared to natural variations (see Core Concepts Box). The signal of temperature change has emerged 50 more clearly in tropical regions, where year-to-year variations tend to be small over land, than in regions 51 with greater warming but larger year-to-year variations (high confidence) (Figure TS.3). Long-term changes 52 in other variables have emerged in many regions, such as for some weather and climate extremes and Arctic 53 sea ice area. {1.4.2, Cross-Chapter Box 3.1, 9.3.1, 11.3.2, 12.5.2} 13 In this report, transient climate response is defined as the surface temperature response for the hypothetical scenario in which atmospheric carbon dioxide (CO2) increases at 1% yr-1 from pre-industrial to the time of a doubling of atmospheric CO2 concentration. Do Not Cite, Quote or Distribute TS-18 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Observational datasets have been extended and improved since AR5, providing stronger evidence that the 2 climate is changing and allowing better estimates of natural climate variability on decadal time scales. There 3 is very high confidence that the slower rate of global surface temperature change observed over 1998–2012 4 compared to 1951–2012 was temporary, and was, with high confidence, induced by internal variability 5 (particularly Pacific Decadal Variability) and variations in solar irradiance and volcanic forcing that partly 6 offset the anthropogenic warming over this period. Global ocean heat content continued to increase 7 throughout this period, indicating continuous warming of the entire climate system (very high confidence). 8 Hot extremes also continued to increase during this period over land (high confidence). Even in a continually 9 warming climate, periods of reduced and increased trends in global surface temperature at decadal time 10 scales will continue to occur in the 21st century (very high confidence). {Cross-Chapter Box 3.1, 3.3.1, 3.5.1, 11 4.6.2, 11.3.2} 12 13 Since AR5, the increased use of ‘large ensembles’, or multiple simulations with the same climate model but 14 using different initial conditions, supports improved understanding of the relative roles of internal variability 15 and forced change in the climate system. Simulations and understanding of modes of climate variability, 16 including teleconnections, have improved since AR5 (medium confidence), and larger ensembles allow a 17 better quantification of uncertainty in projections due to internal climate variability. {1.4.2, 1.5.3, 1.5.4, 4.2, 18 Box 4.1, 4.4.1, 8.5.2, 10.3.4, 10.4} 19 20 Changes in regional climate can be detected even though natural climate variations can temporarily increase 21 or obscure anthropogenic climate change on decadal time scales. While anthropogenic forcing has 22 contributed to multi-decadal mean precipitation changes in several regions, internal variability can delay 23 emergence of the anthropogenic signal in long-term precipitation changes in many land regions (high 24 confidence). {10.4.1, 10.4.2, 10.4.3} 25 26 Mean temperatures and heat extremes have emerged above natural variability in almost all land regions with 27 high confidence. Changes in temperature-related variables, such as regional temperatures, growing season 28 length, extreme heat and frost, have already occurred, and there is medium confidence that many of these 29 changes are attributable to human activities. Several impact-relevant changes have not yet emerged from the 30 natural variability but will emerge sooner or later in this century depending on the emissions scenario (high 31 confidence). Ocean acidification and deoxygenation have already emerged over most of the global open 32 ocean, as has a reduction in Arctic sea ice (high confidence). {9.3.1, 9.6.4, 11.2, 11.3, 12.4, 12.5, Atlas.3- 33 Atlas.11} 34 35 36 [START FIGURE TS.3 HERE] 37 38 Figure TS.3: Emergence of changes in temperature over the historical period. The intent of this figure is to 39 show how observed changes in temperature have emerged and that the emergence pattern agrees with 40 model simulations. The observed change in temperature at a global warming level of 1°C (top map), 41 and the signal-to-noise ratio (the change in temperature at a global warming level of 1°C, divided by 42 the size of year-to-year variations, bottom map) using data from Berkeley Earth. The right panels 43 show the zonal means of the maps and include data from different observational datasets (red) and the 44 CMIP6 simulations (black, including the 5-95% range) processed in the same way as the observations. 45 {1.4.2, 10.4.3} 46 47 [END FIGURE TS.3 HERE] 48 49 50 51 52 53 54 55 Do Not Cite, Quote or Distribute TS-19 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.1.2.4 Understanding of Human Influence 2 3 The evidence for human influence on recent climate change has strengthened progressively from the IPCC 4 Second Assessment Report to the AR5 and is even stronger in this assessment, including for regional scales 5 and for extremes. Human influence in the IPCC context refers to the human activities that lead to or 6 contribute to a climate response, such as the human-induced emissions of greenhouse gases that 7 subsequently alter the atmosphere’s radiative properties, resulting in warming of the climate system. Other 8 human activities influencing climate include the emission of aerosols and other short-lived climate forcers, 9 and land-use change such as urbanisation. Progress in our understanding of human influence is gained from 10 longer observational datasets, improved paleoclimate information, a stronger warming signal since AR5, and 11 improvements in climate models, physical understanding and attribution techniques (see Core Concepts 12 Box). Since AR5, the attribution to human influence has become possible across a wider range of climate 13 variables and climatic impact-drivers (CIDs, see Core Concepts Box). New techniques and analyses drawing 14 on several lines of evidence have provided greater confidence in attributing changes in regional weather and 15 climate extremes to human influence (high confidence). {1.3, 1.5.1, Appendix 1A, 3.1-3.8, 5.2, 6.4.2, 7.3.5, 16 7.4.4, 8.3.1, 10.4, Cross-Chapter Box 10.3, 11.2-11.9, 12.4, TS.3} 17 18 Combining the evidence from across the climate system increases the level of confidence in the attribution of 19 observed climate change to human influence and reduces the uncertainties associated with assessments based 20 on single variables. 21 22 Since AR5, the accumulation of energy in the Earth system has become established as a robust measure of 23 the rate of global climate change on interannual-to-decadal time scales. The rate of accumulation of energy is 24 equivalent to the Earth’s energy imbalance and can be quantified by changes in the global energy inventory 25 for all components of the climate system, including global ocean heat uptake, warming of the atmosphere, 26 warming of the land, and melting of ice. Compared to changes in global surface temperature, the Earth's 27 energy imbalance (see Core Concepts Box) exhibits less variability, enabling more accurate identification 28 and estimation of trends. 29 30 Identifying the human-induced components contributing to the energy budget provides an implicit estimate 31 of the human influence on global climate change. {Cross-Working Group Box: Attribution, 3.8, 7.2.2, Box 32 7.2 and Cross Chapter Box 9.1, TS.2, TS.3.1} 33 34 Regional climate changes can be moderated or amplified by regional forcing from land use and land-cover 35 changes or from aerosol concentrations and other short-lived climate forcers (SLCFs). For example, the 36 difference in observed warming trends between cities and their surroundings can partly be attributed to 37 urbanization (very high confidence). While established attribution techniques provide confidence in our 38 assessment of human influence on large-scale climate changes (as described in TS.2), new techniques 39 developed since AR5, including attribution of individual events, have provided greater confidence in 40 attributing changes in climate extremes to climate change. Multiple attribution approaches support the 41 contribution of human influence to several regional multi-decadal mean precipitation changes (high 42 confidence). Understanding about past and future changes in weather and climate extremes has increased due 43 to better observation-based datasets, physical understanding of processes, an increasing proportion of 44 scientific literature combining different lines of evidence, and improved accessibility to different types of 45 climate models (high confidence) (see TS.2, TS.4). {Cross-Working Group Box: Attribution, 1.5, 3.2, 3.5, 46 5.2, 6.4.3, 8.3, 9.6, 10.1, 10.2, 10.3.3, 10.4.1, 10.4.2, 10.4.3, 10.5, 10.6, Cross-Chapter Box 10.3, Box 10.3, 47 11.1.6, 11.2-11.9, 12.4}. 48 49 50 TS.1.3 Assessing Future Climate Change 51 52 Various frameworks can be used to assess future climatic changes and to synthesize knowledge across 53 climate change assessment in WGI, WGII and WGIII. These frameworks include: (i) scenarios, (ii) global 54 warming levels and (iii) cumulative CO2 emissions (see Core Concepts Box). The latter two offer scenario- Do Not Cite, Quote or Distribute TS-20 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 and path-independent approaches to assess future projections. Additional choices, for instance with regard to 2 common reference periods and time windows for which changes are assessed, can further help to facilitate 3 integration across the WGI report and across the whole AR6 (see TS.1.1). {1.4.1, 1.6, Cross-Chapter Box 4 1.4, 4.2.2, 4.2.4, Cross-Chapter Box 11.1} 5 6 7 TS.1.3.1 Climate Change Scenarios 8 9 A core set of five scenarios based on the Shared Socioeconomic Pathways (SSPs) are used consistently 10 across this report: SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5. These scenarios cover a broader 11 range of greenhouse gas and air pollutant futures than assessed in earlier WGI reports, and they include high- 12 CO2 emissions pathways without climate change mitigation as well as new low-CO2 emissions pathways 13 (Figure TS.4). In these scenarios, differences in air pollution control and variations in climate change 14 mitigation stringency strongly affect anthropogenic emission trajectories of SLCFs. Modelling studies 15 relying on the Representative Concentration Pathways (RCPs) used in the AR5 complement the assessment 16 based on SSP scenarios, for example at the regional scale. A comparison of simulations from CMIP5 using 17 the RCPs with SSP-based simulations from CMIP6 shows that about half of the increase in simulated 18 warming in CMIP6 versus CMIP5 arises because higher climate sensitivity is more prevalent in CMIP6 19 model versions; the other half arises from higher radiative forcing in nominally corresponding scenarios 20 (e.g., RCP8.5 and SSP5-8.5; medium confidence). The feasibility or likelihood of individual scenarios is not 21 part of this assessment, which focuses on the climate response to a large range of emissions scenarios. 22 {1.5.4, 1.6, Cross-Chapter Box 1.4, 4.2, 4.3, 4.6, 6.6, 6.7, Cross-Chapter Box 7.1, Atlas.2.1} 23 24 Climate change projections with climate models require information about future emissions or 25 concentrations of greenhouse gases, aerosols, ozone depleting substances, and land use over time (Figure 26 TS.4). This information can be provided by scenarios, which are internally consistent projections of these 27 quantities based on assumptions of how socio-economic systems could evolve over the 21st century. 28 Emissions from natural sources, such as the ocean and the land biosphere, are usually assumed to be 29 constant, or to evolve in response to changes in anthropogenic forcings or to projected climate change. 30 Natural forcings, such as past changes in solar irradiance and historical volcanic eruptions, are represented in 31 model simulations covering the historical era. Future simulations assessed in this report account for projected 32 changes in solar irradiance and for the long-term mean background forcing from volcanoes, but not for 33 individual volcanic eruptions. Scenarios have a long history in IPCC as a method for systematically 34 examining possible futures and allow to follow the cause-effect chain: from anthropogenic emissions, to 35 changes in atmospheric concentrations, to changes in the Earth’ energy balance (‘forcing’), to changes in 36 global climate and ultimately regional climate and climatic impact-drivers (Figure TS.4; see TS.2; 37 Infographic TS.1). {1.5.4, 1.6.1, 4.2.2, 4.4.4, Cross-Chapter Box 4.1, 11.1} 38 39 40 [START FIGURE TS.4 HERE] 41 42 Figure TS.4: The climate change cause-effect chain: from anthropogenic emissions, to changes in atmospheric 43 concentration, to changes in the Earth’s energy balance (‘forcing’), to changes to changes in global 44 climate and ultimately regional climate and climatic impact-drivers. Shown is the core set of five SSP 45 scenarios as well as emission and concentration ranges for the previous RCP scenarios in year 2100; 46 CO2 emissions (GtCO2 yr–1), panel top left; CH4 emissions (middle) and SO2, NOx emissions (all in 47 Mt yr–1), top right; concentrations of atmospheric CO2 (ppm) and CH4 (ppb), second row left and 48 right; effective radiative forcing for both anthropogenic and natural forcings (W m–2), third row; 49 changes in global surface air temperature (°C) relative to 1850–1900, fourth row; maps of projected 50 temperature change (°C) (left) and changes in annual-mean precipitation (%) (right) at GWL 2°C 51 relative to 1850–1900 (see also Figure TS.5), bottom row. Carbon cycle and non-CO2 biogeochemical 52 feedbacks will also influence the ultimate response to anthropogenic emissions (arrows on the left). 53 {1.6.1, Cross-Chapter Box 1.4, 4.2.2, 4.3.1, 4.6.1, 4.6.2} 54 55 [END FIGURE TS.4 HERE] Do Not Cite, Quote or Distribute TS-21 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 The uncertainty in climate change projections that results from assessing alternative socio-economic futures, 2 the so-called scenario uncertainty, is explored through the use of scenario sets. Designed to span a wide 3 range of possible future conditions, these scenarios do not intend to match how events actually unfold in the 4 future, and they do not account for impacts of climate change on the socio-economic pathways. Besides 5 scenario uncertainty, climate change projections are also subject to climate response uncertainty (i.e., the 6 uncertainty related to our understanding of the key physical processes and structural uncertainties in climate 7 models) and irreducible and intrinsic uncertainties related to internal variability. Depending on the spatial 8 and temporal scales of the projection, and on the variable of interest, the relative importance of these 9 different uncertainties may vary substantially. {1.4.3, 1.6, 4.2.5, Box 4.1, 8.5.1} 10 11 In this report, a core set of five scenarios is used to explore climate change over the 21st century and beyond 12 (Section TS.2). They are labelled SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.514, and span a wide 13 range of radiative forcing levels in 2100. Scenarios in AR6 cover a broader range of emissions futures than 14 considered in AR5, including high CO2 emissions scenarios without climate change mitigation as well as a 15 low CO2 emissions scenario reaching net zero CO2 emissions (see Core Concepts Box) around mid-century. 16 These SSP scenarios offer unprecedented detail of input data for ESM simulations and allow for a more 17 comprehensive assessment of climate drivers and responses, in particular because some aspects, such as the 18 temporal evolution of pollutants, emissions or changes in land use and land cover, span a broader range in 19 the SSP scenarios than in the RCPs used in the AR5. Modelling studies utilizing the RCPs complement the 20 assessment based on SSP scenarios, for example, at the regional scale (Section TS.4). Scenario extensions 21 are based on assumptions about the post-2100 evolution of emissions or of radiative forcing that is 22 independent from the modelling of socio-economic dynamics, which does not extend beyond 2100. To 23 explore specific dimensions, such as air pollution or temporary overshoot of a given warming level, scenario 24 variants are used in addition to the core set. {1.6.1, Cross-Chapter Box 1.4, 4.2.2, 4.2.6, 4.7.1, Cross-Chapter 25 Box 7.1} 26 27 SSP1-1.9 represents the low end of future emissions pathways, leading to warming below 1.5°C in 2100 and 28 limited temperature overshoot over the course of the 21st century (see Figure TS.6). At the opposite end of 29 the range, SSP5-8.5 represents the very high warming end of future emissions pathways from the literature. 30 SSP3-7.0 has overall lower GHG emissions than SSP5-8.5 but, for example, CO2 emissions still almost 31 double by 2100 compared to today’s levels. SSP2-4.5 and SSP1-2.6 represent scenarios with stronger 32 climate change mitigation and thus lower GHG emissions. SSP1-2.6 was designed to limit warming to below 33 2°C. Infographic TS.1 presents a narrative depiction of SSP-related climate futures. No likelihood is attached 34 to the scenarios assessed in this report, and the feasibility of specific scenarios in relation to current trends is 35 best informed by the WGIII contribution to AR6. In the scenario literature, the plausibility of some scenarios 36 with high CO2 emissions, such as RCP8.5 or SSP5-8.5, has been debated in light of recent developments in 37 the energy sector. However, climate projections from these scenarios can still be valuable because the 38 concentration-levels reached in RCP8.5 or SSP5-8.5 and corresponding simulated climate futures cannot be 39 ruled out. That is because of uncertainty in carbon-cycle feedbacks which in nominally lower emissions 40 trajectories can result in projected concentrations that are higher than the central concentration-levels 41 typically used to drive model projections {1.6.1; Cross-Chapter Box 1.4; 4.2.2, 5.4; SROCC; Chapter 3 in 42 WGIII}. 43 44 The socio-economic narratives underlying SSP-based scenarios differ in their assumed level of air pollution 45 control. Together with variations in climate change mitigation stringency, this difference strongly affects 46 anthropogenic emission trajectories of SLCFs, some of which are also air pollutants. SSP1 and SSP5 assume 47 strong pollution control, projecting a decline of global emissions of ozone precursors (except methane) and 48 of aerosols and most of their precursors in the mid- to long term. The reductions due to air pollution controls 49 are further strengthened in scenarios that assume a marked decarbonization, such as SSP1-1.9 or SSP1-2.6. 50 SSP2-4.5 is a medium pollution-control scenario with air pollutant emissions following current trends, and 51 SSP3-7.0 is a weak pollution-control scenario with strong increases in emissions of air pollutants over the 52 21st century. Methane emissions in SSP-based scenarios vary with the overall climate change mitigation 14 Throughout this report, scenarios are referred to as SSPx-y, where “SSPx” refers to the Shared Socioeconomic Pathway or “SSP” describing the socio-economic trends underlying the scenario and “y” refers to the approximate target level of radiative forcing (in W m-2) resulting from the scenario in the year 2100. Do Not Cite, Quote or Distribute TS-22 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 stringency, declining rapidly in SSP1-1.9 and SSP1-2.6 but declining only after 2070 in SSP5-8.5. SSP 2 trajectories span a wider range of air pollutant emissions than considered in the RCP scenarios (see Figure 3 TS.4), reflecting the potential for large regional differences in their assumed pollution policies. Their effects 4 on climate and air pollution are assessed in Box TS.7. {4.4.4, 6.6.1, Figure 6.4; 6.7.1, Figure 6.19} 5 6 Since the RCPs are also labelled by the level of radiative forcing they reach in 2100, they can in principle be 7 related to the core set of AR6 scenarios (Figure TS.4). However, the RCPs and SSP-based scenarios are not 8 directly comparable. First, the gas-to-gas compositions differ; for example, the SSP5-8.5 scenario has higher 9 CO2 but lower methane concentrations compared to RCP8.5. Second, the projected 21st-century trajectories 10 may differ, even if they result in the same radiative forcing by 2100. Third, the overall effective radiative 11 forcing (see Core Concepts Box) may differ, and tends to be higher for the SSPs compared to RCPs that 12 share the same nominal stratospheric-temperature-adjusted radiative forcing label. Comparing the 13 differences between CMIP5 and CMIP6 projections (Cross-Section Box TS.1) that were driven by RCPs and 14 SSP-based scenarios, respectively, indicates that about half of the difference in simulated warming arises 15 because of higher climate sensitivity being more prevalent in CMIP6 model versions; the remainder arises 16 from higher ERF in nominally corresponding scenarios (e.g., RCP8.5 and SSP5-8.5; medium confidence) 17 (see TS.1.2.2). In SSP1-2.6 and SSP2-4.5, changes in effective radiative forcing also explain about half of 18 the changes in the range of warming (medium confidence). For SSP5-8.5, higher climate sensitivity is the 19 primary reason behind the upper end of the CMIP6-projected warming being higher than for RCP8.5 in 20 CMIP5 (medium confidence). Note that AR6 uses multiple lines of evidence beyond CMIP6 results to assess 21 global surface temperature under various scenarios (see Cross-Section Box TS.1 for the detailed assessment). 22 {1.6, 4.2.2, 4.6.2.2, Cross-Chapter Box 7.1} 23 24 Earth system models can be driven by anthropogenic CO2 emission (‘emissions-driven’ runs), in which case 25 atmospheric CO2 concentration is a projected variable; or by prescribed time-varying atmospheric 26 concentrations (‘concentration-driven’ runs). In emissions-driven runs, changes in climate feed back on the 27 carbon cycle and interactively modify the projected CO2 concentration in each ESM, thus adding the 28 uncertainty in the carbon cycle response to climate change to the projections. Concentration-driven 29 simulations are based on a central estimate of carbon cycle feedbacks, while emissions-driven simulations 30 help quantify the role of feedback uncertainty. The differences in the few ESMs for which both emission and 31 concentration-driven runs were available for the same scenario are small and do not affect the assessment of 32 global surface temperature projections discussed in Cross-Section Box TS.1 and Section TS.2 (high 33 confidence). By the end of the 21st century, emission-driven simulations are on average around 0.1°C cooler 34 than concentration-driven runs reflecting the generally lower CO2 concentrations simulated by the emission 35 driven ESMs, and a spread about 0.1°C greater reflecting the range of simulated CO2 concentrations. 36 However, these carbon cycle–climate feedbacks do affect the transient climate response to cumulative CO2 37 emissions (TCRE15), and their quantification is crucial for the assessment of remaining carbon budgets 38 consistent with global warming levels simulated by ESMs (see TS.3). {1.6.1, Cross-Chapter Box 1.4, 4.2, 39 4.3.1, 5.4.5, Cross-Chapter Box 7.1} 40 41 42 TS.1.3.2 Global Warming Levels and Cumulative CO2 Emissions 43 44 Quantifying geographical response patterns of climate change at various global warming levels (GWLs), 45 such as 1.5 or 2°C above the 1850–1900 period, is useful for characterizing changes in mean climate, 46 extremes and climatic impact-drivers. GWLs are used in this report as a dimension of integration 47 independent of the timing when the warming level is reached and of the emissions scenario that led to the 48 warming. For many climate variables the response pattern for a given GWL is consistent across different 49 scenarios. However, this is not the case for slowly responding processes, such as ice sheet and glacier mass 50 loss, deep ocean warming, and the related sea level rise. The response of these variables depends on the time 51 it takes to reach the GWL, differs if the warming is reached in a transient warming state or after a temporary 52 overshoot of the warming level, and will continue to evolve, over centuries to millennia, even after global 53 warming has stabilized. Different GWLs correspond closely to specific cumulative CO2 emissions due to 15 The transient surface temperature change per unit of cumulative CO 2 emissions, usually 1000 GtC; Do Not Cite, Quote or Distribute TS-23 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 their near-linear relationship with global surface temperature. This report uses 1.0°C, 1.5°C, 2.0°C, 3.0°C, 2 and 4.0°C above 1850–1900 conditions as a primary set of GWLs. {1.6.2, 4.2.4, 4.6.1, 5.5, Cross-Chapter 3 Box 11.1, Cross-chapter Box 12.1} 4 5 For many indicators of climate change, such as seasonal and annual mean and extreme surface air 6 temperatures and precipitation, the geographical patterns of changes are well estimated by the level of global 7 surface warming, independently of the details of the emission pathways that caused the warming, or the time 8 at which the level of warming is attained. GWLs, defined as a global surface temperature increase of, for 9 example, 1.5°C or 2°C relative to the mean of 1850–1900, are therefore a useful way to integrate climate 10 information independently of specific scenarios or time periods. {1.6.2, 4.2.4, 4.6.1, 11.2.4, Cross-Chapter 11 Box 11.1} 12 13 The use of GWLs allows disentangling the contribution of changes in global warming from regional aspects 14 of the climate response, as scenario differences in response patterns at a given GWL are often smaller than 15 model uncertainty and internal variability. The relationship between the GWL and response patterns is often 16 linear, but integration of information can also be done for non-linear changes, like the frequency of heat 17 extremes. The requirement is that the relationship to the GWL is broadly independent of the scenario and 18 relative contribution of radiative forcing agents. {1.6, 11.2.4, Cross-Chapter Box 11.1} 19 20 The GWL approach to integration of climate information also has some limitations. Variables that are quick 21 to respond to warming, like temperature and precipitation, including extremes, sea ice area, permafrost and 22 snow cover, show little scenario dependence for a given GWL, whereas slow-responding variables such as 23 glacier and ice sheet mass, warming of the deep ocean and their contributions to sea level rise, have 24 substantial dependency on the trajectory of warming taken to reach the GWL. A given GWL can also be 25 reached for different balances between anthropogenic forcing agents, such as long-lived greenhouse gas and 26 SLCF emissions, and the response patterns may depend on this balance. Finally, there is a difference in the 27 response even for temperature-related variables if a GWL is reached in a rapidly warming transient state or 28 in an equilibrium state when the land–sea warming contrast is less pronounced. In this Report the climate 29 responses at different GWLs are calculated based on climate model projections for the 21st century (see 30 Figure TS.5), which are mostly not in equilibrium. The SSP1-1.9 scenario allows assessing the response to a 31 GWL of about 1.5°C after a (relatively) short-term stabilization by the end of the 21st century. {4.6.2, 32 9.3.1.1, 9.5.2.3, 9.5.3.3, 11.2.4, Cross-Chapter Box 11.1, Cross-chapter Box 12.1} 33 34 Global warming levels are highly relevant as a dimension of integration across scientific disciplines and 35 socio-economic actors and are motivated by the long-term goal in the Paris Agreement of ‘holding the 36 increase in the global average temperature to well below 2°C above pre-industrial levels and to pursue efforts 37 to limit the temperature increase to 1.5°C above pre-industrial levels’. The evolution of aggregated impacts 38 with temperature levels has also been widely used and embedded in the WGII assessment. This includes the 39 ‘Reasons for Concern’ (RFC) and other ‘burning ember’ diagrams in IPCC WGII. The RFC framework has 40 been further expanded in the SR1.5, the SROCC and SRCCL by explicitly looking at the differential impacts 41 between half-degree GWLs and the evolution of risk for different socio-economic assumptions. {1.4.4, 1.6.2, 42 11.2.4, 12.5.2, Cross-chapter Box 11.1, Cross-Chapter Box 12.1} 43 44 SR1.5 concluded, ‘climate models project robust differences in regional climate characteristics between 45 present-day and global warming of 1.5°C, and between 1.5°C and 2°C’. This report adopts a set of common 46 GWLs across which climate projections, impacts, adaptation challenges and climate change mitigation 47 challenges can be integrated, within and across the three WGs, relative to 1850–1900. The core set of GWLs 48 in this Report are 1.0°C (close to present day conditions), 1.5°C, 2.0°C, 3.0°C, and 4.0°C. {1.4, 1.6.2, Cross- 49 Chapter Box 1.2, Table 1.5, Cross-chapter Box 11.1} 50 51 Connecting Scenarios and Global Warming Levels 52 53 In this report, scenario-based climate projections are translated into GWLs by aggregating the ESM model 54 response at specific GWL across scenarios (see Figure TS.5 and Figure TS.6). The climate response pattern 55 for the 20-year period around when individual simulations reach a given GWL are averaged across all Do Not Cite, Quote or Distribute TS-24 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 models and scenarios that reach that GWL. The best estimate and likely range of the timing of when a certain 2 GWL is reached under a particular scenario (or ‘GWL-crossing time’), however, is based not only on CMIP6 3 output, but on a combined assessment taking into account the observed warming to date, CMIP6 output and 4 additional lines of evidence (see Cross-Section Box TS.1). {4.3.4, Cross-Chapter Box 11.1, Atlas.2, 5 Interactive Atlas} 6 7 8 [START FIGURE TS.5 HERE] 9 10 Figure TS.5: How scenarios are linked to global warming levels (GWLs), and examples of the evolution of 11 patterns of change with global warming levels. Left: Illustrative example of GWLs defined as 12 global surface temperature response to anthropogenic emissions in unconstrained CMIP6 simulations, 13 for two illustrative scenarios (SSP1-2.6 and SSP3-7.0). The time when a given simulation reaches a 14 GWL, e.g., +2°C, relative to 1850-1900 is taken as the time when the central year of a 20-year 15 running mean first reaches that level of warming. See the dots for +2°C, and how not all simulations 16 reach all levels of warming. The assessment of the timing when a GWL is reached takes into account 17 additional lines of evidence and is discussed in Cross-Section Box TS.1. Right: Multi-model, multi- 18 simulation average response patterns of change in near-surface air temperature, precipitation 19 (expressed as percentage change) and soil moisture (expressed in standard deviations of interannual 20 variability), for three GWLs. The number to the top right of the panels shows the number of model 21 simulations averaged across including all models that reach the corresponding GWL in any of the 5 22 SSPs. See TS.2 for discussion. (See also Cross-Chapter Box 11.1) 23 24 [END FIGURE TS.5 HERE] 25 26 27 Global warming levels are closely related to cumulative CO2 (and in some cases CO2-equivalent) emissions. 28 This report confirms the assessment of the WGI AR5 and SR1.5 that a near-linear relationship exists 29 between cumulative CO2 emissions and the resulting increase in global surface temperature (Section TS.3.2). 30 This implies that continued CO2 emissions will cause further warming and associated changes in all 31 components of the climate system. For declining cumulative CO2 emissions (i.e., if negative net emissions 32 are achieved), the relationship is less strong for some components, such as the hydrological cycle. The WGI 33 report uses cumulative CO2 emissions to compare climate response across scenarios and provides a link to 34 the emission pathways assessment in WGIII. The advantage of using cumulative CO2 emissions is that it is 35 an inherent emissions scenario characteristic rather than an outcome of the scenario-based projections, where 36 uncertainties in the cause–effect chain from emissions to temperature change are important (Figure TS.4), for 37 example, the uncertainty in effective radiative forcing (ERF) and transient climate response (TCR). 38 Cumulative CO2 emissions can also provide a link to the assessments of mitigation options. Cumulative CO2 39 emissions do not carry information about non-CO2 emissions, although these can be included with specific 40 emission metrics to estimate CO2-equivalent emissions. (TS.3.3) {1.3.2, 1.6, 4.6.2, 5.5, 7.6, 8.4.3} 41 42 43 TS.1.4 From Global to Regional Climate Information for Impact and Risk Assessment 44 45 The WGI AR6 has an expanded focus on regional information supported by the increased availability of 46 coordinated regional climate model ensemble projections and improvements in the sophistication and 47 resolution of global and regional climate models (high confidence). Multiple lines of evidence can be used to 48 construct climate information on a global to regional scale and can be further distilled in a co-production 49 process to meet user needs (high confidence). To better support risk assessment, a common risk framework 50 across all three Working Groups has been implemented in AR6, and low-likelihood but high-impact 51 outcomes are explicitly addressed in WGI by using physical climate storylines (see Core Concepts Box). 52 Climatic impact-drivers are physical climate system conditions (e.g., means, events, extremes) that affect an 53 element of society or ecosystems. They are the WGI contribution to the risk framing without anticipating 54 whether their impact provides potential opportunities or is detrimental (i.e., as for hazards). Many global and 55 regional climatic impact-drivers have a direct relation to global warming levels (high confidence). {1.4.4, 56 1.5.2-1.5.4, 4.8, 10.1, 10.5.1, 11.2.4, 11.9, 12.1-12.3, 12.6, Atlas.1.3.3-1.3.4, Atlas.1.4, Atlas.1.4.4, Boxes Do Not Cite, Quote or Distribute TS-25 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 10.2 and 11.2, Cross-Chapter Boxes 1.3, 10.3, 11.1, 12.1 and 12.2} 2 3 Climate change is a global phenomenon, but manifests differently in different regions. The impacts of 4 climate change are generally experienced at local, national and regional scales, and these are also the scales 5 at which decisions are typically made. Robust climate change information is increasingly available at 6 regional scales for impact and risk assessments. Depending on the climate information context, geographical 7 regions in AR6 may refer to larger areas, such as sub-continents and oceanic regions, or to typological 8 regions, such as monsoon regions, coastlines, mountain ranges or cities, as used in TS.4. A new set of 9 standard AR6 WGI reference regions has also been included in this report (Fig. TS.6 bottom panels). {1.4.5, 10 10.1, 11.9, 12.1-12.4, Atlas.1.3.3-1.3.4} 11 12 Global and regional climate models are important sources of climate information at the regional scale. Since 13 AR5, a more comprehensive assessment of past and future evolution of a range of climate variables on a 14 regional scale has been enabled by the increased availability of coordinated ensemble regional climate model 15 projections and improvements in the level of sophistication and resolution of global and regional climate 16 models. This has been complemented by observational, attribution and sectoral-vulnerability studies 17 informing, for instance, about impact-relevant tolerance thresholds. {10.3.3, 11.9, 12.1, 12.3, 12.6, Atlas.3- 18 Atlas.11, Interactive Atlas} 19 20 Multiple lines of evidence derived from observations, model simulations and other approaches can be used to 21 construct climate information on a regional scale as described in detail in TS.4.1.1 and TS.4.1.2. Depending 22 on the phenomena and specific context, these sources and methodologies include theoretical understanding 23 of the relevant processes, drivers and feedbacks of climate at regional scale, trends in observed data from 24 multiple datasets, and the attribution of these trends to specific drivers. Furthermore, simulations from 25 different model types (including global and regional climate models, emulators, statistical downscaling 26 methods, etc.) and experiments (e.g., CMIP, CORDEX, and large ensembles of single-model simulations 27 with different initial conditions), attribution methodologies as well as other relevant local knowledge (e.g., 28 indigenous knowledge) are utilized. (see Box TS.11). {1.5.3,1.5.4, Cross-Chapter Box 7.1, 10.2,10.3-10.6, 29 11.2, Atlas.1.4, Cross-Chapter Box 10.3} 30 31 From the multiple lines of evidence, climate information can be distilled in a co-production process that 32 involves users, related stakeholders and producers of climate information, considering the specific context of 33 the question at stake, the underlying values, and the challenge of communicating across different 34 communities. The co-production process is an essential part of climate services, which are discussed in 35 TS.4.1.2. {10.5, 12.6, Cross-Chapter Box 12.2} 36 37 38 [START FIGURE TS.6 HERE] 39 40 Figure TS.6: A graphical abstract for key aspects of the Technical Summary related to observed and 41 projected changes in global surface temperature and associated regional changes in climatic 42 impact-drivers relevant for impact and risk assessment. Top left: a schematic representation of the 43 likelihood for equilibrium climate sensitivity (ECS), consistent with the AR6 assessment (see Chapter 44 7; TS.3). ECS values above 5°C and below 2°C are termed low-likelihood high warming (LLHW) and 45 low-likelihood low warming, respectively. Top right: Observed (see Cross-Section Box TS.1) and 46 projected global surface temperature changes, shown as global warming levels (GWL) relative to 47 1850-1900, using the assessed 95% (top), 50% (middle) and 5% (bottom) likelihood time series (see 48 Chapter 4; TS.2). Bottom panels show maps of CMIP6 median projections of two climatic impact- 49 drivers (CIDs) at three different GWLs (columns for 1.5, 2 and 4°C) for the AR6 land regions (see 50 Chapters 1, 10, Atlas; TS.4). The heat warning index is the number of days per year averaged across 51 each region at which a heat warning for human health at level ‘danger’ would be issued according to 52 the U.S. National Oceanic and Atmospheric Administration (NOAA) (NOAA HI41, see Chapter 12 53 and Technical Annex VI). The maps of extreme rainfall changes show the percentage change in the 54 amount of rain falling on the wettest day of a year (Rx1day, relative to 1995-2014, see Chapter 11) 55 averaged across each region when the respective GWL is reached. Additional CIDs are discussed in 56 TS.4. Do Not Cite, Quote or Distribute TS-26 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [END FIGURE TS.6 HERE] 2 3 4 With the aim of informing decision-making at local or regional scales, a common risk framework has been 5 implemented in AR6. Methodologies have been developed to construct more impact- and risk-relevant 6 climate change information tailored to regions and stakeholders. Physical storyline approaches are used in 7 order to build climate information based on multiple lines of evidence, and which can explicitly address 8 physically plausible, but low-likelihood, high-impact outcomes and uncertainties related to climate 9 variability for consideration in risk assessments (see Figure TS.6). {4.8, 12.1-12.3, 12.6, Box 9.4, 10.5, Box 10 10.2, Box 11.2, Cross-Chapter Box 1.3, Glossary} 11 12 The climatic impact-driver framework developed in AR6 supports an assessment of changing climate 13 conditions that are relevant for sectoral impacts and risk assessment. Climatic impact-drivers (CIDs) are 14 physical climate system conditions (e.g., means, extremes, events) that affect an element of society or 15 ecosystems and are thus a potential priority for providing climate information. For instance, the heat index 16 used by the U.S. National Oceanic and Atmospheric Administration (NOAA HI) for issuing heat warnings is 17 a CID index that can be associated with adverse human health impacts due to heat stress (see Figure TS.6). 18 Depending on system tolerance, CIDs and their changes can be detrimental (i.e., hazards in the risk framing), 19 beneficial, neutral, or a mixture of each across interacting system elements, regions and sectors (aligning 20 with WGII Sectoral Chapters 2–8). Each sector is affected by multiple CIDs, and each CID affects multiple 21 sectors. Climate change has already altered CID profiles and resulted in shifting magnitude, frequency, 22 duration, seasonality and spatial extent of associated indices (high confidence) (see regional details in 23 TS.4.3). {12.1-12.4, Table 12.1, Table 12.2, Technical Annex VI} 24 25 Many global- and regional-scale CIDs, including extremes, have a direct relation to global warming levels 26 (GWLs) and can thus inform the hazard component of ‘Representative Key Risks’ and ‘Reasons for 27 Concern’ assessed by AR6 WGII. These include heat, cold, wet and dry hazards, both mean and extremes; 28 cryospheric hazards (snow cover, ice extent, permafrost) and oceanic hazards (marine heatwaves) (high 29 confidence) (Figure TS.6). Establishing links between specific GWLs with tipping points and irreversible 30 behaviour is challenging due to model uncertainties and lack of observations, but their occurrence cannot be 31 excluded, and their likelihood of occurrence generally increases at greater warming levels (Box TS.1, TS.9). 32 {11.2.4, Box 11.2, Cross-Chapter Boxes 11.1 and 12.1} 33 34 35 [START CROSS-SECTION BOX TS.1 HERE] 36 37 Cross-Section Box TS.1: Global Surface Temperature Change 38 39 This box synthesizes the outcomes of the assessment of past, current and future global surface temperature. 40 Global mean surface temperature (GMST) and global surface air temperature (GSAT) are the two primary 41 metrics of global surface temperature used to estimate global warming in IPCC reports. GMST merges sea 42 surface temperatures (SSTs) over the ocean and 2 m air temperature over land and sea ice areas and is used 43 in most paleo, historical and present-day observational estimates. The GSAT metric is 2 m air temperature 44 over all surfaces and is the diagnostic generally used from climate models. Changes in GMST and GSAT 45 over time differ by at most 10% in either direction (high confidence), but conflicting lines of evidence from 46 models and direct observations, combined with limitations in theoretical understanding, lead to low 47 confidence in the sign of any difference in long-term trend. Therefore, long-term changes in GMST/GSAT 48 are presently assessed to be identical, with expanded uncertainty in GSAT estimates. Hence the term global 49 surface temperature is used in reference to both quantities in the text of the TS and SPM. {Cross-Chapter 50 Box 2.3} 51 52 Global surface temperature has increased by 1.09 [0.95 to 1.20] °C from 1850–1900 to 2011–2020, and the 53 last decade was more likely than not warmer than any multi-centennial period after the Last Interglacial, 54 roughly 125,000 years ago. The likely range of human-induced warming in global surface temperature in 55 2010–2019 relative to 1850–1900 is 1.07 [0.8 to 1.3] °C, encompassing the observed warming, while the Do Not Cite, Quote or Distribute TS-27 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 change attributable to natural forcing is only –0.1°C to +0.1°C. Compared to 1850–1900, average global 2 surface temperature over the period 2081–2100 is very likely to be higher by 1.0°C–1.8°C in the low CO2 3 emissions scenario SSP1-1.9 and by 3.3°C–5.7°C in the high CO2 emissions scenario SSP5-8.5. In all 4 scenarios assessed here except SSP5-8.5, the central estimate of 20-year averaged global surface warming 5 crossing the 1.5°C level lies in the early 2030s, which is about ten years earlier than the midpoint of the 6 likely range (2030–2052) assessed in the SR1.5. It is more likely than not that under SSP1-1.9, global surface 7 temperature relative to 1850–1900 will remain below 1.6°C throughout the 21st century, implying a potential 8 temporary overshoot of 1.5°C global warming of no more than 0.1°C. Global surface temperature in any 9 individual year, could exceed 1.5°C relative to 1850–1900 by 2030 with a likelihood between 40% and 60% 10 across the scenarios considered here (medium confidence). A 2°C increase in global surface temperature 11 relative to 1850–1900 will be crossed under SSP5-8.5 but is extremely unlikely to be crossed under SSP1- 12 1.9. Periods of reduced and increased GSAT trends at decadal time scales will continue to occur in the 21st 13 century (very high confidence). The effect of strong mitigation on 20-year global surface temperature trends 14 would be likely to emerge during the near term (2021–2040), assuming no major volcanic eruptions occur. 15 (Figure TS.8; Cross-Section Box TS.1, Figure 1), {2.3, 3.3, 4.3, 4.4, 4.6, 7.3}. 16 17 Surface Temperature History 18 19 Dataset innovations, particularly more comprehensive representation of polar regions, and the availability of 20 new datasets have led to an assessment of increased global surface temperature change relative to the directly 21 equivalent estimates reported in AR5. The contribution of changes in observational understanding alone 22 between AR5 and AR6 in assessing temperature changes from 1850–1900 to 1986–2005 is estimated at 0.08 23 [–0.01 to 0.12] °C. From 1850–1900 to 1995–2014, global surface temperature increased by 0.85 [0.69 to 24 0.95] °C, and to the most recent decade (2011–2020) by 1.09 [0.95 to 1.20] °C. Each of the last four decades 25 has in turn been warmer than any decade that preceded it since 1850. Temperatures have increased faster 26 over land than over the ocean since 1850–1900, with warming to 2011–2020 of 1.59 [1.34 to 1.83] °C over 27 land and 0.88 [0.68 to 1.01] °C over the ocean. {2.3.1, Cross-Chapter Box 2.3} 28 29 Global surface temperature during the period 1850–1900 is used as an approximation for pre-industrial 30 conditions for consistency with AR5 and AR6 Special Reports, whilst recognizing that radiative forcings 31 have a baseline of 1750 for the start of anthropogenic influences. It is likely that there was a net 32 anthropogenic forcing of 0.0–0.3 Wm-2 in 1850–1900 relative to 1750 (medium confidence), and from the 33 period around 1750 to 1850–1900 there was a change in global surface temperature of around 0.1°C (likely 34 range –0.1 to +0.3°C, medium confidence), with an anthropogenic component of 0.0°C–0.2°C (likely range, 35 medium confidence). {Cross-Chapter Box 1.2, 7.3.5}. 36 37 Global surface temperature has evolved over geological time (Figure TS.1, Box TS.2). Beginning 38 approximately 6500 years ago, global surface temperature generally decreased, culminating in the coldest 39 multi-century interval of the post-glacial period (since roughly 7000 years ago), which occurred between 40 around 1450 and 1850 (high confidence). Over the last 50 years, global surface temperature has increased at 41 an observed rate unprecedented in at least the last two thousand years (medium confidence), and it is more 42 likely than not that no multi-centennial period after the Last Interglacial (roughly 125,000 years ago) was 43 warmer globally than the most recent decade (Cross-Section Box TS.1, Figure 1). During the mid-Pliocene 44 Warm Period, around 3.3–3.0 million years ago, global surface temperature was 2.5°C–4°C warmer, and 45 during the Last Interglacial, it was 0.5°C–1.5°C warmer than 1850–1900 (medium confidence). {2.3.1, 46 Cross-Chapter Box 2.1 and 2.4} 47 48 49 [START CROSS-SECTION BOX TS.1, FIGURE 1 HERE] 50 51 Cross-Section Box TS.1, Figure 1: Earth’s surface temperature history and future with key findings annotated 52 within each panel. The intent of this figure is to show global surface temperature 53 observed changes from the Holocene to now, and projected changes. (a) Global 54 surface temperature over the Holocene divided into three time scales. (i) 12,000 to 55 1000 years ago (10,000 BCE to 1000 CE) in 100-year time steps, (ii) 1000 to 1900 Do Not Cite, Quote or Distribute TS-28 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 CE, 10-year smooth, and (iii) 1900 to 2020 CE (mean of four datasets in panel 2 (c)). Median of the multi-method reconstruction (bold lines), with 5% and 95% 3 percentiles of the ensemble members (thin lines). Vertical bars are 5th to 95th 4 percentile ranges of estimated global surface temperature for the Last Interglacial 5 and mid Holocene (medium confidence) (Section 2.3.1.1). All temperatures 6 relative to 1850–1900. (b) Spatially resolved trends (°C per decade) for (upper 7 map) HadCRUTv5 over 1981–2020, and (lower map, total change) multi-model 8 mean projected changes from 1995–2014 to 2081–2010 in the SST3-7.0 scenario. 9 Observed trends have been calculated where data are present in both the first and 10 last decade and for at least 70% of all years within the period using OLS. 11 Significance is assessed with AR(1) correction and denoted by stippling. Hatched 12 areas in the lower map show areas of conflicting model evidence on significance 13 of changes. (c) Temperature from instrumental data for 1850–2020, including 14 annually resolved averages for the four global surface temperature datasets 15 assessed in Section 2.3.1.1.3 (see text for references). The grey shading shows the 16 uncertainty associated with the HadCRUTv5 estimate. All temperatures relative to 17 the 1850–1900 reference period. (d) Recent past and 2015–2050 evolution of 18 annual mean global surface temperature change relative to 1850–1900, from 19 HadCRUTv5 (black), CMIP6 historical simulations (up to 2014, in grey, ensemble 20 mean solid, 5% and 95% percentiles dashed, individual models thin), and CMIP6 21 projections under scenario SSP2-4.5, from four models that have an equilibrium 22 climate sensitivity near the assessed central value (thick yellow). Solid thin 23 coloured lines show the assessed central estimate of 20-year change in global 24 surface temperature for 2015–2050 under three scenarios, and dashed thin 25 coloured lines the corresponding 5% and 95% quantiles. (e) Assessed projected 26 change in 20-year running mean global surface temperature for five scenarios 27 (central estimate solid, very likely range shaded for SSP1-2.6 and SSP3-7.0), 28 relative to 1995–2014 (left y-axis) and 1850–1900 (right y-axis). The y-axis on the 29 right-hand side is shifted upward by 0.85°C, the central estimate of the observed 30 warming for 1995–2014, relative to 1850–1900. The right y-axis in (e) is the same 31 as the y-axis in (d). 32 33 [END CROSS-SECTION BOX TS.1, FIGURE 1 HERE] 34 35 36 Current Warming 37 38 There is very high confidence that the CMIP6 model ensemble reproduces observed global surface 39 temperature trends and variability since 1850 with errors small enough for detection and attribution of 40 human-induced warming. The CMIP6 multi-model mean global surface warming between 1850–1900 and 41 2010–2019 is close to the best estimate of observed warming, though some CMIP6 models simulate a 42 warming that is outside the assessed very likely observed range. {3.3.1} 43 44 The likely range of human-induced change in global surface temperature in 2010–201916 relative to 1850– 45 1900 is 1.07 [0.8 to 1.3] °C (Figure Cross-Section Box TS.1, Figure 1), encompassing the observed warming 46 for that period of 1.06 [0.88 to 1.21] °C, while change attributable to natural forcing is only –0.1 to +0.1°C. 47 This assessment is consistent with an estimate of the human-induced global surface temperature rise based 48 on assessed ranges of perturbations to the top of the atmosphere (effective radiative forcing), and metrics of 49 feedbacks of the climate response (equilibrium climate sensitivity and the transient climate response). Over 50 the same period, well-mixed greenhouse gas forcing likely warmed global surface temperature by 1.0°C to 51 2.0°C, while aerosols and other anthropogenic forcings likely cooled global surface temperature by 0.0°C to 52 0.8°C. {2.3.1, 3.3.1, 7.3.5, Cross-Chapter Box 7.1} 53 54 The observed slower global surface temperature increase (relative to preceding and following periods) in the 55 1998–2012 period, sometimes referred to as ‘the hiatus’, was temporary (very high confidence). The increase 16 Assessment of human-induced warming took place before 2020 data were available and hence concludes in 2019. Do Not Cite, Quote or Distribute TS-29 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 in global surface temperature during the 1998–2012 period is also greater in the data sets used in the AR6 2 assessment than in those available at the time of AR5. Using these updated observational data sets and a like- 3 for-like consistent comparison of simulated and observed global surface temperature, all observed estimates 4 of the 1998–2012 trend lie within the very likely range of CMIP6 trends. Since 2012, global surface 5 temperature has warmed strongly, with the past five years (2016–2020) being the hottest five-year period 6 between 1850 and 2020 (high confidence). {2.3.1, 3.3.1, 3.5.1, Cross-Chapter Box 3.1} 7 8 Future Changes in Global Surface Temperature 9 10 The AR6 assessment of future change in global surface temperature is, for the first time in an IPCC report, 11 explicitly constructed by combining new projections for the SSP scenarios with observational constraints 12 based on past simulated warming as well as the AR6-updated assessment of equilibrium climate sensitivity 13 and transient climate response. Climate forecasts initialized from the observed climate state have in addition 14 been used for the period 2019–2028. The inclusion of additional lines of evidence has reduced the assessed 15 uncertainty ranges for each scenario (Cross-Section Box TS.1, Figure 1). {4.3.1, 4.3.4, Box 4.1, 7.5} 16 17 During the near term (2021–2040), a 1.5°C increase in global surface temperature, relative to 1850–1900, is 18 very likely to occur in scenario SSP5-8.5, likely to occur in scenarios SSP2-4.5 and SSP3-7.0, and more likely 19 than not to occur in scenarios SSP1-1.9 and SSP1-2.6. The time of crossing a warming level here is defined 20 here as the midpoint of the first 20-year period during which the average global surface temperature exceeds 21 the level. In all scenarios assessed here except SSP5-8.5, the central estimate of crossing the 1.5°C level lies 22 in the early 2030s. This is about ten years earlier than the midpoint of the likely range (2030–2052) assessed 23 in the SR1.5, which assumed continuation of the then-current warming rate; this rate has been confirmed in 24 the AR6. Roughly half of the ten-year difference arises from a larger historical warming diagnosed in AR6. 25 The other half arises because for central estimates of climate sensitivity, most scenarios show stronger 26 warming over the near term than was estimated as ‘current’ in SR1.5 (medium confidence). (Cross-Section 27 Box TS.1, Table 1) {2.3.1, Cross-Chapter Box 2.3, 3.3.1, 4.3.4, BOX 4.1} 28 29 It is more likely than not that under SSP1-1.9, global surface temperature relative to 1850–1900 will remain 30 below 1.6°C throughout the 21st century, implying a potential temporary overshoot of 1.5°C global warming 31 of no more than 0.1°C. If climate sensitivity lies near the lower end of the assessed very likely range, 32 crossing the 1.5°C warming level is avoided in scenarios SSP1-1.9 and SSP1-2.6 (medium confidence). 33 Global surface temperature in any individual year, in contrast to the 20-year average, could by 2030 exceed 34 1.5°C relative to 1850–1900 with a likelihood between 40% and 60%, across the scenarios considered here 35 (medium confidence). (Cross-Section Box TS.1, Table 1) {4.3.4, 4.4.1, BOX 4.1, 7.5} 36 37 During the 21st century, a 2°C increase in global surface temperature relative to 1850–1900 will be crossed 38 under SSP5-8.5 and SSP3-7.0, will extremely likely be crossed under SSP2-4.5, but is unlikely to be crossed 39 under SSP1-2.6 and extremely unlikely to be crossed under SSP1-1.9. For the mid-term period 2041–2060, 40 this 2°C global warming level is very likely to be crossed under SSP5-8.5, likely to be crossed under SSP3- 41 7.0, and more likely than not to be crossed under SSP2-4.5. (Cross-Section Box TS.1, Table 1) {4.3.4} 42 43 Events of reduced and increased global surface temperature trends at decadal timescales will continue to 44 occur in the 21st century, but will not affect the centennial warming (very high confidence). If strong 45 mitigation is applied from 2020 onward as reflected in SSP1-1.9, its effect on 20-year trends in global 46 surface temperature would likely emerge during the near term (2021–2040), measured against an assumed 47 non-mitigation scenario such as SSP3-7.0 and SSP5-8.5. All statements about crossing the 1.5°C level 48 assume that no major volcanic eruption occurs during the near term (Cross-Section Box TS.1, Table 1). 49 {2.3.1, Cross-chapter Box 2.3, 4.3.4, 4.4.1, 4.6.3, Box 4.1} 50 51 Compared to 1850–1900, average global surface temperature over the period 2081–2100 is very likely to be 52 higher by 1.0°C–1.8°C in the low CO2 emissions scenario SSP1-1.9 and by 3.3°C–5.7°C in the high CO2 53 emissions scenario SSP5-8.5. For the scenarios SSP1-2.6, SSP2-4.5, and SSP3-7.0, the corresponding very 54 likely ranges are 1.3°C–2.4°C, 2.1°C–3.5°C, and 2.8°C–4.6°C, respectively. The uncertainty ranges for the 55 period 2081–2100 continue to be dominated by the uncertainty in equilibrium climate sensitivity and Do Not Cite, Quote or Distribute TS-30 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 transient climate response (very high confidence) (Cross-Section Box TS.1, Table 1). {4.3.1, 4.3.4, 4.4.1, 2 7.5} 3 4 The CMIP6 models project a wider range of global surface temperature change than the assessed range (high 5 confidence); furthermore, the CMIP6 global surface temperature increase tends to be larger than in CMIP5 6 (very high confidence). {4.3.1, 4.3.4, 4.6.2, 7.5.6} 7 8 9 [START CROSS-SECTION BOX TS.1, TABLE 1 HERE] 10 11 Cross-Section Box TS.1, Table 1: Assessment results for 20-year averaged change in global surface temperature 12 based on multiple lines of evidence. The change is displayed in °C relative to the 13 1850–1900 reference period for selected time periods, and as the first 20-year period 14 during which the average global surface temperature change exceeds the specified level 15 relative to the period 1850–1900. The entries give both the central estimate and, in 16 parentheses, the very likely (5–95%) range. An entry n.c. means that the global warming 17 level is not crossed during the period 2021–2100. 18 SSP1-1.9 SSP1-2.6 SSP2-4.5 SSP3-7.0 SSP5-8.5 Near term, 2021– 1.5 (1.2, 1.7) 1.5 (1.2, 1.8) 1.5 (1.2, 1.8) 1.5 (1.2, 1.8) 1.6 (1.3, 1.9) 2040 Mid-term, 2041– 1.6 (1.2, 2.0) 1.7 (1.3, 2.2) 2.0 (1.6, 2.5) 2.1 (1.7, 2.6) 2.4 (1.9, 3.0) 2060 Long term, 2081– 1.4 (1.0, 1.8) 1.8 (1.3, 2.4) 2.7 (2.1, 3.5) 3.6 (2.8, 4.6) 4.4 (3.3, 5.7) 2100 2025–2044 2023–2042 2021–2040 2021–2040 2018–2037 1.5°C (2013–2032, n.c.) (2012–2031, n.c.) (2012–2031, (2013–2032, (2011–2030, 2037–2056) 2033–2052) 2029–2048) n.c. n.c. 2043–2062 2037–2056 2032–2051 2°C (n.c., n.c.) (2031–2050, n.c.) (2028–2047, (2026–2045, (2023–2042, 2075–2094) 2053–2072) 2044–2063) n.c. n.c. n.c. 2066–2085 2055–2074 3°C (n.c., n.c.) (n.c., n.c.) (2061–2080, n.c.) (2050–2069, n.c.) (2042–2061, 2074–2093) n.c. n.c. n.c. n.c. 2075–2094 4°C (n.c., n.c.) (n.c., n.c.) (n.c., n.c.) (2070–2089, n.c.) (2058–2077, n.c.) 19 20 21 [END CROSS-SECTION BOX TS.1, TABLE 1 HERE] 22 23 [END CROSS-SECTION BOX TS.1 HERE] 24 25 26 TS.2 Large-scale Climate Change: Mean Climate, Variability and Extremes 27 28 This section summarizes knowledge about observed and projected large-scale climate change (including 29 variability and extremes), drivers and attribution of observed changes to human activities. It describes 30 observed and projected large-scale changes associated with major components of the climate system: 31 atmosphere, ocean (including sea level change), land, biosphere and cryosphere, and the carbon, energy and 32 water cycles. In each subsection, reconstructed past, observed and attributed recent, and projected near- and 33 long-term changes to mean climate, variability and extremes are presented, where possible, in an integrated 34 way. See TS.1.3.1 for information on the scenarios used for projections. 35 36 37 TS.2.1 Changes Across the Global Climate System 38 39 In addition to global surface temperature (Cross-Section Box TS.1), a wide range of indicators across all 40 components of the climate system is changing rapidly (Figure TS.7), with many at levels unseen in Do Not Cite, Quote or Distribute TS-31 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 millennia. The observed changes provide a coherent picture of a warming world, many aspects of which 2 have now been formally attributed to human influences, and human influence on the climate system as a 3 whole is assessed as unequivocal for the first time in IPCC assessment reports (Table TS.1, Figure TS.7). It 4 is virtually certain that global surface temperature rise and associated changes can be limited through rapid 5 and substantial reductions in global GHG emissions. Continued GHG emissions greatly increase the 6 likelihood of potentially irreversible changes in the global climate system (Box TS.3), in particular with 7 respect to the contribution of ice sheets to global sea level change (high confidence). {2.3, 3.8, 4.3, 4.6, 4.7, 8 7.2-7.4, 9.2-9.6, CCB 7.1} 9 10 Earth system model (ESM) simulations of the historical period since 1850 are only able to reproduce the 11 observed changes in key climate indicators when anthropogenic forcings are included (Figure TS.7). Taken 12 together with numerous formal attribution studies across an even broader range of indicators and theoretical 13 understanding, this underpins the unequivocal attribution of observed warming of the climate system to 14 human influence (Table TS.1). {2.3, 3.8} 15 16 17 [START FIGURE TS.7 HERE] 18 19 Figure TS.7: Simulated and observed changes compared to the 1995–2014 average in key large-scale 20 indicators of climate change across the climate system, for continents, ocean basins and globally 21 up to 2014. Black lines show observations, orange lines and shading show the multi-model mean and 22 5–95th percentile ranges for CMIP6 historical simulations including anthropogenic and natural 23 forcing, and green lines and shading show corresponding ensemble means and 5-95th percentile ranges 24 for CMIP6 natural-only simulations. Observations after 2014 (including, for example, a strong 25 subsequent decrease of Antarctic sea-ice area that leads to no significant overall trend since 1979) are 26 not shown because the CMIP6 historical simulations end in 2014. A 3-year running mean smoothing 27 has been applied to all observational time series. {3.8, Figure 3.41} 28 29 [END FIGURE TS.7 HERE] 30 31 32 [START TABLE TS.1 HERE] 33 34 Table TS.1: Assessment of observed changes in large-scale indicators of mean climate across climate system 35 components, and their attribution to human influence. The colour coding indicates the assessed 36 confidence in / likelihood of the human contribution as a driver or main driver (specified in that case) 37 where available (see colour key). Otherwise, explanatory text is provided in cells with white 38 background. The relevant chapter section for more detailed information is listed in each table cell. 39 40 Change in indicator Observed change assessment Human contribution assessment Atmosphere and water cycle Likely range of human contribution Warming of global mean surface air {2.3.1, CCB2.3} (0.8°C–1.3°C) encompasses observed temperature since 1850-1900 warming (0.9°C–1.2°C) {3.3.1} Warming of the troposphere since 1979 {2.3.1} Main driver {3.3.1} Cooling of the lower stratosphere Since mid-20th century {2.3.1} Main driver 1979 - mid-1990s {3.3.1} Large-scale precipitation and upper {2.3.1} {3.3.2, 3.3.3} troposphere humidity changes since 1979 Expansion of the zonal mean Hadley {2.3.1} Southern Hemisphere {3.3.3} Circulation since the 1980s Ocean Ocean heat content increase since the {2.3.3, 2.3.4, 9.2.1, CCB 9.1} Main driver {3.5.1} 1970s Salinity changes since the mid-20th {2.3.3, 2.3.4, 9.2.2} {3.5.2} century Global mean sea level rise since 1970 {2.3.3, 9.6.1} Main driver {3.5.3} Do Not Cite, Quote or Distribute TS-32 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI Cryosphere Arctic sea ice loss since 1979 {2.3.2, 9.3.1} Main driver {3.4.1} Reduction in Northern Hemisphere spring {2.3.2, 9.5.3} {3.4.2} snow cover since 1950 Greenland Ice Sheet mass loss since {2.3.2, 9.4.1} {3.4.3} 1990s Limited evidence & medium agreement Antarctic Ice Sheet mass loss since 1990s {2.3.2, 9.4.2} {3.4.3} Retreat of glaciers {2.3.2, 9.5.1} Main driver {3.4.3} Carbon cycle Increased amplitude of the seasonal cycle {2.3.4} Main driver {3.6.1} of atmospheric CO2 since the early 1960s Acidification of the global surface ocean {SROCC, 5.3.2, CCB 5.3} Main driver {3.6.2} Land climate (extremes, see Table TS.12) Mean 2-m land warming since 1850–1900 (about 40% larger than global mean {2.3.1} Main driver {3.3.1} warming) Synthesis Warming of the global climate system {2.3.5} {3.8.1} since preindustrial times 1 see text medium likely / high very extremely virtually fact description confidence confidence likely likely certain 2 3 [END TABLE TS.1 HERE] 4 5 6 Future climate change across a range of atmospheric, cryospheric, oceanic and biospheric indicators depends 7 upon future emissions pathways. Outcomes for a broad range of indicators increasingly diverge through the 8 21st century across the different SSPs (TS.1.3.1; Figure TS.8). Due to the slow response of the deep ocean 9 and ice sheets, this divergence continues long after 2100, and 21st century emissions choices will have 10 implications for GMSL rise for centuries to millennia. Furthermore, it is likely that at least one large volcanic 11 eruption will occur during the 21st century. Such an eruption would reduce global surface temperature for 12 several years, decrease land precipitation, alter monsoon circulation and modify extreme precipitation, at 13 both global and regional scales. {4.3, 4.7, 9.4, 9.6, Cross-Chapter Box 4.1} 14 15 16 [START FIGURE TS.8 HERE] 17 18 Figure TS.8: Observed, simulated and projected changes compared to the 1995–2014 average in 4 key 19 indicators of the climate system through to 2100 differentiated by SSP scenario pathway. Past 20 simulations are based on the CMIP6 multi-model ensemble. Future projections are based on the 21 assessed ranges based upon multiple lines of evidence for (a) global surface temperature (Cross- 22 Section Box TS.1) and (b) global ocean heat content and the associated thermosteric sea level 23 contribution to Global Mean Sea Level (GMSL) change (right-hand axis) using a climate model 24 emulator (Cross-Chapter Box 7.1), and CMIP6 simulations for (c) Arctic September sea ice and (d) 25 Global land precipitation. SSP1-1.9 and SSP1-2.6 projections show that reduced GHG emissions lead 26 to a stabilization of global surface temperature, Arctic sea ice area and global land precipitation over 27 the 21st century. SSP1-2.6 shows that emissions reductions have the potential to substantially reduce 28 the increase in ocean heat content and thermosteric sea level rise over the 21st century but that some 29 increase is unavoidable. {4.3, 9.3, 9.6, Figure 4.2, Figure 9.6} 30 31 [END FIGURE TS.8 HERE] 32 33 34 Observational records show changes in a wide range of climate extremes that have been linked to human 35 influence on the climate system (Table TS.2). In many cases, the frequency and intensity of future changes in 36 extremes can be directly linked to the magnitude of future projected warming. Changes in extremes are 37 widespread over land since the 1950s, including a virtually certain global increase in extreme air Do Not Cite, Quote or Distribute TS-33 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 temperatures and a likely intensification in global-scale extreme precipitation. It is extremely likely that 2 human influence is the main contributor to the observed increase (decrease) in the likelihood and severity of 3 hot (cold) extremes (Table TS.2). The frequency of extreme temperature and precipitation events in the 4 current climate will change with warming, with warm extremes becoming more frequent (virtually certain), 5 cold extremes becoming less frequent (extremely likely) and precipitation extremes becoming more frequent 6 in most locations (very likely). {9.6.4, 11.2, 11,3, 11.4, 11.6, 11.7, 11.8, 11.9, Box 9.2} 7 8 9 [START TABLE TS.2 HERE] 10 11 Table TS.2: Summary table on observed changes in extremes, their attribution since 1950 (except where stated 12 otherwise), and projected changes at +1.5°C, +2°C and +4°C of global warming, on global and 13 continental scale. Warm/hot extremes refer to warmer and/or more frequent hot days and nights and 14 warm spells/heat waves, over most land areas. Cold extremes refer to warmer and/or fewer cold days 15 and nights, and cold spells/cold waves, over most land areas. Drought events are relative to a 16 predominant fraction of land area. For tropical cyclones, observed changes and attribution refer to 17 categories 3-5, while projected changes refer to categories 4-5. Tables 11.1 and 11.2 are more detailed 18 versions of this table, containing in particular information on regional scales. In general, higher 19 warming levels imply stronger projected changes also for indicators where the confidence level does 20 not depend on the warming level and the table does not explicitly quantify the global sensitivity. {9.6, 21 Box 9.2, 11.3, 11.7} 22 Observed (since Projected at GWL (°C) Change in indicator Attributed (since 1950) 1950) +1.5 +2 +4 Warm/hot extremes: ✔︎ Frequency or intensity ⬆︎ ⬆︎ ⬆︎ ⬆︎ Main driver Cold extremes: ✔︎ Frequency or intensity ⬇ ⬇ ⬇ ⬇ Main driver ⬆︎ ✔︎ ⬆︎ ⬆︎ ⬆︎ Heavy precipitation Over majority of land Main driver of the events: Frequency, regions with good observed intensification in most land intensity and/or amount in most land regions observational of heavy precipitation regions coverage in land regions ⬆︎ ⬆︎ ⬆︎ Agricultural and ⬆︎ ✔︎ ecological droughts: for predominant for predominant for predominant for predominant for predominant fraction of land fraction of land fraction of land Intensity and/or duration fraction of land area fraction of land area area area area Precipitation associated ⬆︎ ⬆︎ ⬆︎ with tropical cyclones ⬆︎ ✔︎ Rate +11% Rate +14% Rate +28% Tropical cyclones: ⬆︎ ⬆︎ ⬆︎ Proportion of intense ⬆︎ ✔︎ cyclones +10% +13% +20% Compound events: co- ⬆︎ ✔︎ ⬆︎ occurrent heat waves and droughts (Frequency) (Frequency) (Frequency and intensity increases with warming) Marine heatwaves: ⬆︎ ✔︎ ⬆︎ Intensity & frequency (since 1900) (since 2006) Strongest in tropical and Arctic Ocean Extreme sea levels: ⬆︎ ⬆︎ ✔︎ Frequency (since 1960) (Scenario-based assessment for 21st century) 23 medium likely / high extremely virtually very likely confidence confidence likely certain 24 25 26 27 [END TABLE TS.2 HERE] 28 Do Not Cite, Quote or Distribute TS-34 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.2.2 Changes in the Drivers of the Climate System 2 3 Since 1750, changes in the drivers of the climate system are dominated by the warming influence of 4 increases in atmospheric GHG concentrations and a cooling influence from aerosols, both resulting from 5 human activities. In comparison there has been negligible long-term influence from solar activity and 6 volcanoes. Concentrations of CO2, CH4, and N2O have increased to levels unprecedented in at least 800,000 7 years, and there is high confidence that current CO2 concentrations have not been experienced for at least 2 8 million years. Global mean concentrations of anthropogenic aerosols peaked in the late 20th century and 9 have slowly declined since in northern mid-latitudes, although they continue to increase in South Asia and 10 East Africa (high confidence). The total anthropogenic effective radiative forcing (ERF) in 2019, relative to 11 1750, was 2.72 [1.96 to 3.48] W m−2 (medium confidence) and has likely been growing at an increasing rate 12 since the 1970s. {2.2, 6.4, 7.2, 7.3} 13 14 Solar activity since 1900 was high but not exceptional compared to the past 9000 years (high confidence). 15 The average magnitude and variability of volcanic aerosols since 1900 has not been unusual compared to at 16 least the past 2500 years (medium confidence). However, sporadic strong volcanic eruptions can lead to 17 temporary drops in global surface temperature lasting 2–5 years. {2.2.1, 2.2.2, 2.2.8, CCB4.1} 18 19 Atmospheric CO2 concentrations have changed substantially over millions of years (Figure TS.1). Current 20 levels of atmospheric CO2 have not been experienced for at least 2 million years (high confidence, Figure 21 TS.9a). Over 1750–2019, CO2 increased by 131.6 ± 2.9 ppm (47.3%). The centennial rate of change of CO2 22 since 1850 has no precedent in at least the past 800,000 years (Figure TS.9), and the fastest rates of change 23 over the last 56 million years were at least a factor of 4 lower (low confidence) than over 1900–2019. Several 24 networks of high-accuracy surface observations show that concentrations of CO2 have exceeded 400 ppm, 25 reaching 409.9 (± 0.3) ppm in 2019 (Figure TS.9c). The ERF from CO2 in 2019 (relative to 1750) was 2.16 26 Wm-2. {2.2.3, 5.1.2, 5.2.1, 7.3, Box TS.5} 27 28 By 2019, concentrations of CH4 reached 1866.3 (± 3.3) ppb (Figure TS.9c). The increase since 1750 of 1137 29 ± 10 ppb (157.8%) far exceeds the range over multiple glacial-interglacial transitions of the past 800,000 30 years (high confidence). In the 1990s, CH4 concentrations plateaued, but started to increase again around 31 2007 at an average rate of 7.6 ± 2.7 ppb yr-1 (2010–2019; high confidence). There is high confidence that this 32 recent growth is largely driven by emissions from fossil fuel exploitation, livestock, and waste, with ENSO 33 driving multi-annual variability of wetland and biomass burning emissions. In 2019, ERF from CH4 was 0.54 34 Wm-2. {2.2.3, 5.2.2, 7.3} 35 36 Since 1750, N2O increased by 62.0 ± 6.0 ppb, reaching a level of 332.1 (± 0.4) ppb in 2019. The increase 37 since 1750 is of comparable magnitude to glacial-interglacial fluctuations of the past 800,000 years (Figure 38 TS.9c). N2O concentration trends since 1980 are largely driven by a 30% increase in emissions from the 39 expansion and intensification of global agriculture (high confidence). By 2019 its ERF was 0.21 Wm-2. 40 {2.2.3, 5.2.3} 41 42 43 [START FIGURE TS.9 HERE] 44 45 Figure TS.9: Changes in well-mixed greenhouse gas (WMGHG) concentrations and Effective Radiative 46 Forcing. a) Changes in CO2 from proxy records over the past 3.5 million years; b) Changes in all 47 three WMGHGs from ice core records over the Common Era; c) directly observed WMGHG changes 48 since the mid-20th century; d) Evolution of ERF and components since 1750. Further details on data 49 sources and processing are available in the associated FAIR data table. {2.2, Figures 2.3, 2.4 and 50 2.10} 51 52 [END FIGURE TS.9 HERE] 53 54 55 Do Not Cite, Quote or Distribute TS-35 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Halogenated gases consist of chlorofluorocarbons (CFCs), hydrochlorofluorocarbons (HCFCs), 2 hydrofluorocarbons (HFCs) and other gases, many of which can deplete stratospheric ozone and warm the 3 atmosphere. In response to controls on production and consumption mandated by the Montreal Protocol on 4 Substances that Deplete the Ozone Layer and its amendments, the atmospheric abundances of most CFCs 5 have continued to decline since AR5. Abundances of of HFCs, which are replacements for CFCs and 6 HCFCs, are increasing (high confidence), though increases of the major HCFCs have slowed in recent years. 7 The ERF from halogenated components in 2019 was 0.41 Wm-2. {2.2.4, 6.3.4, 7.3.2} 8 9 Tropospheric aerosols mainly act to cool the climate system, directly by reflecting solar radiation, and 10 indirectly through enhancing cloud reflectance. Ice cores show increases in aerosols across the Northern 11 Hemisphere mid-latitudes since 1700, and reductions since the late 20th century (high confidence). Aerosol 12 optical depth (AOD), derived from satellite- and ground-based radiometers, has decreased since 2000 over 13 the mid-latitude continents of both hemispheres, but increased over South Asia and East Africa (high 14 confidence). Trends in AOD are more pronounced from sub-micrometre aerosols for which the 15 anthropogenic contribution is particularly large. Global carbonaceous aerosol budgets and trends remain 16 poorly characterised due to limited observations, but black carbon (BC), a warming aerosol component, is 17 declining in several regions of the Northern Hemisphere (low confidence). Total aerosol ERF in 2019, 18 relative to 1750, is −1.1 [−1.7 to −0.4] W m−2 (medium confidence), and more likely than not became less 19 negative since the late 20th century, with low confidence in the magnitude of post-2014 changes due to 20 conflicting evidence (TS.3.1). {2.2.6, 6.2.1, 6.3.5, 6.4.1, 7.3.3} 21 22 There is high confidence that tropospheric ozone has been increasing from 1750 in response to 23 anthropogenic changes in ozone precursor emissions (nitrogen oxides, carbon monoxide, non-methane 24 volatile organic compounds, and methane), but with medium confidence in the magnitude of this change, due 25 to limited observational evidence and knowledge gaps. Since the mid-20th century, tropospheric ozone 26 surface concentrations have increased by 30–70% across the Northern Hemisphere (medium confidence); 27 since the mid-1990s free tropospheric ozone has increased by 2–7% per decade in most northern mid-latitude 28 regions, and 2–12% per decade in sampled tropical regions. Future changes in surface ozone concentrations 29 will be primarily driven by changes in precursor emissions rather than climate change (high confidence). 30 Stratospheric ozone has declined between 60˚S–60˚N by 2.2% from 1964–1980 to 2014–2017 (high 31 confidence), with the largest declines during 1980–1995. The strongest loss of stratospheric ozone continues 32 to occur in austral spring over Antarctica (ozone hole), with emergent signs of recovery after 2000. The 33 1750–2019 ERF for total (stratospheric and tropospheric) ozone is 0.47 [0.24 to 0.71] W m−2, which is 34 dominated by tropospheric ozone changes. {2.2.5; 6.3.2., 7.3.2, 7.3.5} 35 36 The global mean abundance of hydroxyl (OH) radical, or ‘oxidising capacity’, chemically regulates the 37 lifetimes of many short-lived climate forcers (SLCFs), and therefore the radiative forcing of methane, ozone, 38 secondary aerosols and many halogenated species. Model estimates suggest no significant change in 39 oxidising capacity from 1850 to 1980 (low confidence). Increases of about 9% over 1980–2014 computed by 40 ESMs and carbon cycle models are not confirmed by observationally constrained inverse models, rendering 41 an overall medium confidence in stable OH or positive trends since the 1980s, and implying that OH is not 42 the primary driver of recent observed growth in CH4. {6.3.6, CCB5.2} 43 44 Land use and land-cover change exert biophysical and biogeochemical effects. There is medium confidence 45 that the biophysical effects of land-use change since 1750, most notably the increase in global albedo, have 46 had an overall cooling on climate, whereas biogeochemical effects (i.e., changes in GHG and volatile 47 organic compound emissions or sinks) led to net warming. Overall land use and land cover ERF is estimated 48 at -0.2 [-0.3 to -0.1] W m−2. {2.2.7, 7.3.4, SRCCL 2.5} 49 50 The total anthropogenic ERF (Figure TS.9) in 2019 relative to 1750 was 2.72 [1.96 to 3.48] W m−2 (Figure 51 TS.9), dominated by GHGs (positive ERF) and partially offset by aerosols (negative ERF). The rate of 52 change of ERF likely has increased since the 1970s, mainly due to growing CO2 concentrations and less 53 negative aerosol ERF. {2.2.8, 7.3} 54 Do Not Cite, Quote or Distribute TS-36 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.2.3 Upper Air Temperatures and Atmospheric Circulation 2 3 The effects of human-induced climate change have been clearly identified in observations of atmospheric 4 temperature and some aspects of circulation, and these effects are likely to intensify in the future. 5 Tropospheric warming and stratospheric cooling are virtually certain to continue with continued net 6 emissions of greenhouse gases. Several aspects of the atmospheric circulation have likely changed since the 7 mid-20th century, and human influence has likely contributed to the observed poleward expansion of the 8 Southern Hemisphere Hadley Cell and very likely contributed to the observed poleward shift of the Southern 9 Hemisphere extratropical jet in summer. It is likely that the mid-latitude jet will shift poleward and 10 strengthen, accompanied by a strengthening of the storm track in the Southern Hemisphere by 2100 under 11 the high CO2 emissions scenarios. It is likely that the proportion of intense tropical cyclones has increased 12 over the last four decades and that this cannot be explained entirely by natural variability. There is low 13 confidence in observed recent changes in the total number of extratropical cyclones over both hemispheres. 14 The proportion of tropical cyclones which are intense is expected to increase (high confidence) but the total 15 global number of tropical cyclones is expected to decrease or remain unchanged (medium confidence). {2.3, 16 3.3, 4.3, 4.4, 4.5, 8.3, 8.4, 11.7} 17 18 The troposphere has warmed since at least the 1950s, and it is virtually certain that the stratosphere has 19 cooled. It is very likely that human-induced increases in GHGs were the main driver of tropospheric warming 20 since 1979. It is extremely likely that anthropogenic forcing, both from increases in GHG concentration and 21 depletion of stratospheric ozone due to ozone-depleting substances, was the main driver of upper 22 stratospheric cooling since 1979. It is very likely that global mean stratospheric cooling will be larger for 23 scenarios with higher atmospheric CO2 concentrations. In the tropics, since at least 2001 (when new 24 techniques permit more robust quantification), the upper troposphere has warmed faster than the near-surface 25 (medium confidence) (Figure TS.10). There is medium confidence that most CMIP5 and CMIP6 models 26 overestimate the observed warming in the upper tropical troposphere over the period 1979–2014, in part 27 because they overestimate tropical SST warming. It is likely that future tropical upper tropospheric warming 28 will be larger than at the tropical surface. {2.3.1, 3.3.1, 4.5.1} 29 30 31 [START FIGURE TS.10 HERE] 32 33 Figure TS.10: Observed and projected upper air temperature and circulation changes. Upper panels: Left: 34 Zonal cross-section of temperature trends for 2002-2019 in the upper troposphere region for the ROM 35 SAF radio-occultation dataset. Middle: Change in the annual and zonal mean atmospheric temperature 36 (°C) in 2081-2100 in SSP1-2.6 relative to 1995-2014 for 36 CMIP6 models. Right: the same in SSP3- 37 7.0 for 32 models. Lower panels: Left: Long-term mean (thin black colour) and linear trend (colour) 38 of zonal mean DJF zonal winds for ERA5. Middle: multi-model mean change in annual and zonal 39 mean wind (m s-1) in 2081-2100 in SSP1-2.6 relative to 1995-2014 based on 34 CMIP6 models. The 40 1995-2014 climatology is shown in contours with spacing 10 m s -1. Right: the same for SSP3-7.0 for 41 31 models. {2.3.1, 4.5.1, Figures 2.12, 2.18, and 4.26} 42 43 [END FIGURE TS.10 HERE] 44 45 46 The Hadley Circulation has likely widened since at least the 1980s, predominantly in the Northern 47 Hemisphere, although there is only medium confidence in the extent of the changes. This has been 48 accompanied by a strengthening of the Hadley Circulation in the Northern Hemisphere (medium confidence). 49 It is likely that human influence has contributed to the poleward expansion of the zonal mean Hadley cell in 50 the Southern Hemisphere since the 1980s, which is projected to further expand with global warming (high 51 confidence). There is medium confidence that the observed poleward expansion in the Northern Hemisphere 52 is within the range of internal variability. {2.3.1, 3.3.3, 8.4.3} 53 54 Since the 1970s near-surface average winds have likely weakened over land. Over the ocean, near-surface 55 average winds likely strengthened over 1980–2000, but divergent estimates lead to low confidence thereafter. Do Not Cite, Quote or Distribute TS-37 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Extratropical storm tracks have likely shifted poleward since the 1980s. There is low confidence in projected 2 poleward shifts of the Northern Hemisphere mid-latitude jet and storm tracks due to large internal variability 3 and structural uncertainty in model simulations. There is medium confidence in a projected decrease of 4 frequency of atmospheric blocking over Greenland and the North Pacific in boreal winter in 2081–2100 5 under the SSP3-7.0 and SSP5-8.5 scenarios. There is high confidence that Southern Hemisphere storm tracks 6 and associated precipitation have migrated polewards over recent decades, especially in the austral summer 7 and autumn, associated with a trend towards more positive phases of the Southern Annular Mode (SAM) 8 (TS.4.2.2) and the strengthening and southward shift of the Southern Hemisphere extratropical jet in austral 9 summer. In the long term (2081–2100), the Southern Hemisphere mid-latitude jet is likely to shift poleward 10 and strengthen under SSP5-8.5 scenario relative to 1995–2014, accompanied by an increase in the SAM 11 (TS.4.2.2). It is likely that wind speeds associated with extratropical cyclones will strengthen in the Southern 12 Hemisphere storm track for SSP5-8.5. There is low confidence in the potential role of Arctic warming and 13 sea ice loss on historical or projected mid-latitude atmospheric variability. {2.3.1, 3.3.3, 3.7.2, 4.3.3, 4.4.3, 14 4.5.1, 4.5.3, 8.2.2, 8.3.2, Cross-Chapter Box 10.1} 15 16 It is likely that the proportion of major (Category 3–5) tropical cyclones (TCs) and the frequency of rapid TC 17 intensification events have increased over the past four decades. The average location of peak TC wind- 18 intensity has very likely migrated poleward in the western North Pacific Ocean since the 1940s, and TC 19 forward translation speed has likely slowed over the contiguous USA since 1900. It is likely that the 20 poleward migration of TCs in the western North Pacific and the global increase in TC intensity rates cannot 21 be explained entirely by natural variability. There is high confidence that average peak TC wind speeds and 22 the proportion of Category 4–5 TCs will increase with warming and that peak winds of the most intense TCs 23 will increase. There is medium confidence that the average location where TCs reach their maximum wind- 24 intensity will migrate poleward in the western North Pacific Ocean, while the total global frequency of TC 25 formation will decrease or remain unchanged with increasing global warming (medium confidence). {11.7} 26 27 There is low confidence in observed recent changes in the total number of extratropical cyclones over both 28 hemispheres. There is also low confidence in past-century trends in the number and intensity of the strongest 29 extratropical cyclones over the Northern Hemisphere due to the large interannual-to-decadal variability and 30 temporal and spatial heterogeneities in the volume and type of assimilated data in atmospheric reanalyses, 31 particularly before the satellite era. Over the Southern Hemisphere, it is likely that the number of 32 extratropical cyclones with low central pressures (<980 hPa) has increased since 1979. The frequency of 33 intense extratropical cyclones is projected to decrease (medium confidence). Projected changes in the 34 intensity depend on the resolution of climate models (medium confidence). There is medium confidence that 35 wind speeds associated with extratropical cyclones will change following changes in the storm tracks. {2.3.1, 36 3.3.3, 4.5.1, 4.5.3, 8.3.2, 8.4.2, 11.7.2, 11.7.2} 37 38 39 [START BOX TS.3 HERE] 40 41 Box TS.3: Low-Likelihood, High-Warming Storylines 42 43 Future global warming exceeding the assessed very likely range cannot be ruled out and is potentially 44 associated with the highest risks for society and ecosystems. Such low-likelihood, high-warming storylines 45 tend to exhibit substantially greater changes in the intensity of regional drying and wetting than the multi- 46 model mean. Even at levels of warming within the very likely range, global and regional low-likelihood 47 outcomes might occur, such as large precipitation changes, additional sea level rise associated with 48 collapsing ice sheets, or abrupt ocean circulation changes. While there is medium confidence that the Atlantic 49 Meridional Overturning circulation (AMOC) will not experience an abrupt collapse before 2100, if it were to 50 occur, it would very likely cause abrupt shifts in regional weather patterns and water cycle. The probability 51 of these low-likelihood outcomes increases with higher global warming levels. If the real-world climate 52 sensitivity lies at the high end of the assessed range, then global and regional changes substantially outside 53 the very likely range projections occur for a given emissions scenario. With increasing global warming, some 54 very rare extremes and some compound events (multivariate or concurrent extremes) with low likelihood in 55 past and current climate will become more frequent, and there is a higher chance that events unprecedented Do Not Cite, Quote or Distribute TS-38 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 in the observational record occur (high confidence). Finally, low likelihood, high-impact outcomes may also 2 arise from a series of very large volcanic eruptions that could substantially alter the 21st century climate 3 trajectory compared to SSP-based Earth system model projections. {Cross-Chapter Box 4.1, 4.3, 4.4, 4.8, 4 7.3, 7.4, 7.5, 8.6, 9.2, 9.6, Box TS.4, Box 9.4, Box 11.2, CCB 12.1} 5 6 Previous IPCC reports largely focused their assessment on the projected very likely range of future surface 7 warming and associated climate change. However, a comprehensive risk assessment also requires 8 considering the potentially larger changes in the physical climate system that are unlikely or very unlikely but 9 possible and potentially associated with the highest risks for society and ecosystems (Figure TS.6). Since 10 AR5, the development of physical climate storylines of high warming has emerged as a useful approach for 11 exploring the future risk space that lies outside of the IPCC very likely range projections. {4.8} 12 13 Uncertainty in the true values of equilibrium climate sensitivity (ECS) and transient climate response (TCR) 14 dominate uncertainty in projections of future warming under moderate to strong emissions scenarios 15 (TS.3.2). A real-world ECS higher than the assessed very likely range (2°C–5°C) would require a strong 16 historical aerosol cooling and/or a trend towards stronger warming from positive feedbacks linked to changes 17 in SST patterns (pattern effects), combined with a strong positive cloud feedback and substantial biases in 18 paleoclimate reconstructions – each of which is assessed as either unlikely or very unlikely, but not ruled out. 19 Since CMIP6 contains several ESMs that exceed the upper bound of the assessed very likely range in future 20 surface warming, these models can be used to develop low-likelihood, high warming storylines to explore 21 risks and vulnerabilities, even in the absence of a quantitative assessment of likelihood. {4.3.4, 4.8, 7.3.2, 22 7.4.4, 7.5.2, 7.5.5, 7.5.7} 23 24 CMIP6 models with surface warming outside, or close to, the upper bound of the very likely range exhibit 25 patterns of large widespread temperature and precipitation changes that differ substantially from the multi- 26 model mean in all scenarios. For SSP5-8.5, the high-warming models exhibit widespread warming of more 27 than 6°C over most extratropical land regions and parts of the Amazon. In the Arctic, annual mean 28 temperatures increase by more than 10°C relative to present-day, corresponding to about 30% more than the 29 best estimate of warming. Even for SSP1-2.6, high-warming models show on average 2°C–3°C warming 30 relative to present-day conditions over much of Eurasia and North America (about 40% more than the best 31 estimate of warming) and more than 4°C warming relative to the present over the Arctic (Box TS.3, Figure 32 1) in 2081–2100. Such a high global warming storyline would imply that the remaining carbon budget 33 consistent with a 2°C warming is smaller than the assessed very likely range. Put another way, even if a 34 carbon budget that likely limits warming to 2°C is met, a low-likelihood, high-warming storyline would 35 result in warming of 2.5°C or more. {4.8} 36 37 CMIP6 models with global warming close to the upper bound of the assessed very likely warming range tend 38 to exhibit greater changes in the intensity of regional drying and wetting than the multi-model mean. 39 Furthermore, these model projections show a larger area of drying and tend to show a larger fraction of 40 strong precipitation increases than the multi-model mean. However, regional precipitation changes arise 41 from both thermodynamic and dynamic processes so that the most pronounced global warming levels are not 42 necessarily associated with the strongest precipitation response. Abrupt human-caused changes to the water 43 cycle cannot be ruled out. Positive land surface feedbacks, involving vegetation and dust, can contribute to 44 abrupt changes in aridity, but there is only low confidence that such changes will occur during the 21st 45 century. Continued Amazon deforestation, combined with a warming climate, raises the probability that this 46 ecosystem will cross a tipping point into a dry state during the 21st century (low confidence). {4.8, 8.6.2, 47 Box TS.9} 48 49 While there is medium confidence that the projected decline in the Atlantic Meridional Overturning 50 Circulation (AMOC) (TS.2.4) will not involve an abrupt collapse before 2100, such a collapse might be 51 triggered by an unexpected meltwater influx from the Greenland Ice Sheet. If an AMOC collapse were to 52 occur, it would very likely cause abrupt shifts in the weather patterns and water cycle, such as a southward 53 shift in the tropical rain belt, and could result in weakening of the African and Asian monsoons and 54 strengthening of Southern Hemisphere monsoons. {4.7.2, 8.6.1, 9.2.3, Box TS.9, Box TS.13} 55 Do Not Cite, Quote or Distribute TS-39 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Very rare extremes and compound or concurrent events, such as the 2018 concurrent heatwaves across the 2 Northern Hemisphere, are often associated with large impacts. The changing climate state is already altering 3 the likelihood of extreme events, such as decadal droughts and extreme sea levels, and will continue to do so 4 under future warming. Compound events and concurrent extremes contribute to increasing probability of 5 low-likelihood, high-impact outcomes, and will become more frequent with increasing global warming (high 6 confidence). Higher warming levels increase the likelihood of events unprecedented in the observational 7 record. {Box 11.2, 9.6.4} 8 9 Finally, low likelihood storylines need not necessarily relate solely to the human-induced changes in climate. 10 A low-likelihood, high-impact outcome, consistent with historical precedent in the past 2,500 years, would 11 be to see several large volcanic eruptions that could greatly alter the 21st century climate trajectory 12 compared to SSP-based Earth system model projections. {Cross-Chapter Box 4.1} 13 14 15 [START BOX TS.3, FIGURE 1 HERE] 16 17 Box TS.3, Figure 1: High-warming storylines. (a) CMIP6 multi-model mean linearly scaled to the assessed best 18 global surface temperature estimate for SSP1-2.6 in 2081–2100 relative to 1995–2014, (b) mean 19 across five high-warming models with global surface temperature changes warming nearest to the 20 upper bound of the assessed very likely range, (c) mean across five very high-warming models 21 with global surface temperature changes warming higher than the assessed very likely. (d-f) Same 22 as (a-c) but for SSP5-8.5. Note the different colour bars in (a-c) and (d-f). {4.7, Figure 4.41} 23 24 [END BOX TS.3, FIGURE 1 HERE] 25 26 [END BOX TS.3 HERE] 27 28 29 TS.2.4 The Ocean 30 31 Observations, models and paleo-evidence indicate that recently observed changes in the ocean are 32 unprecedented for centuries to millennia (high confidence). Over the past four to six decades, it is virtually 33 certain that the global ocean has warmed, with human influence extremely likely the main driver since the 34 1970s, making climate change irreversible over centuries to millennia (medium confidence). It is virtually 35 certain that upper ocean salinity contrasts have increased since the 1950s and extremely likely that human 36 influence has contributed. It is virtually certain that upper ocean stratification has increased since 1970 and 37 that sea water pH has declined globally over the last 40 years, with human influence the main driver of the 38 observed ocean acidification (virtually certain). There is high confidence that marine heatwaves have 39 become more frequent in the 20th century, and most of them have been attributed to anthropogenic warming 40 since 2006 (very likely). There is high confidence that oxygen levels have dropped in many regions since the 41 mid 20th century and that the geographic range of many marine organisms has changed over the last two 42 decades. The amount of ocean warming observed since 1971 will likely at least double by 2100 under a low 43 warming scenario (SSP1-2.6) and will increase by 4-8 times under a high warming scenario (SSP5-8.5). 44 Stratification (virtually certain), acidification (virtually certain), deoxygenation (high confidence) and 45 marine heatwave frequency (high confidence) will continue to increase in the 21st century. While there is 46 low confidence in 20th century AMOC change, it is very likely that AMOC will decline over the 21st century 47 (Figure TS.11). {2.3, 3.5, 3.6, 4.3, 5.3, 7.2, 9.2, 9.4, 9.6, Box 9.2, 12.4} 48 49 It is virtually certain that the global ocean has warmed since at least 1971, representing about 90% of the 50 increase in the global energy inventory (TS.3.1). The ocean is currently warming faster than at any other 51 time since at least the last deglacial transition (medium confidence), with warming extending to depths well 52 below 2000 m (very high confidence). It is extremely likely that human influence was the main driver of 53 ocean warming. Ocean warming will continue over the 21st century (virtually certain), and will likely 54 continue until at least to 2300 even for low CO2 emissions scenarios. Ocean warming is irreversible over 55 centuries to millennia (medium confidence), but the magnitude of warming is scenario-dependent from about Do Not Cite, Quote or Distribute TS-40 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 the mid-21st century (medium confidence). The warming will not be globally uniform, with heat primarily 2 stored in Southern Ocean water-masses and weaker warming in the subpolar North Atlantic (high 3 confidence). Limitations in the understanding of feedback mechanisms limit our confidence in future ocean 4 warming close to Antarctica and how this will affect sea ice and ice shelves. {2.3.3, 3.5.1, 4.7.2, 7.2.2, 9.2.2, 5 9.2.3, 9.2.4, 9.3.2, 9.6.1; Cross-Chapter Box 9.1} 6 7 Global mean SST has increased since the beginning of the 20th century by 0.88 [0.68 to 1.01] °C, and it is 8 virtually certain it will continue to increase throughout the 21st century with increasing hazards to marine 9 ecosystems (medium confidence). Marine heatwaves have become more frequent over the 20th century (high 10 confidence), approximately doubling in frequency (high confidence) and becoming more intense and longer 11 since the 1980s (medium confidence). Most of the marine heatwaves over 2006–2015 have been attributed to 12 anthropogenic warming (very likely). Marine heatwaves will continue to increase in frequency, with a likely 13 global increase of 2–9 times in 2081–2100 compared to 1995–2014 under SSP1-2.6, and 3–15 times under 14 SSP5-8.5 (Figure TS.11a), with the largest changes in the tropical and Arctic ocean. {2.3.1, 9.2.1; Box 9.2; 15 12.4.8; Cross-Chapter Box 2.3} 16 17 Observed upper ocean stratification (0–200 m) has increased globally since at least 1970 (virtually certain). 18 Based on recent refined analyses of the available observations, there is high confidence that it increased by 19 4.9 ±1.5% from 1970–2018, which is about twice as much as assessed in the SROCC, and will continue to 20 increase throughout the 21st century at a rate depending on the emissions scenario (virtually certain). {2.3.3, 21 9.2.1} 22 23 It is virtually certain that since 1950 near-surface high-salinity regions have become more saline, while low- 24 salinity regions have become fresher, with medium confidence that this is linked to an intensification of the 25 hydrological cycle (Box TS.6). It is extremely likely that human influence has contributed to this salinity 26 change and that the large-scale pattern will grow in amplitude over the 21st century (medium confidence). 27 {2.3.3, 3.5.2, 9.2.2, 12.4.8} 28 29 The AMOC was relatively stable during the past 8,000 years (medium confidence). There is low confidence 30 in the quantification of AMOC changes in the 20th century because of low agreement in quantitative 31 reconstructed and simulated trends, missing key processes in both models and measurements used for 32 formulating proxies, and new model evaluations. Direct observational records since the mid-2000s are too 33 short to determine the relative contributions of internal variability, natural forcing and anthropogenic forcing 34 to AMOC change (high confidence). An AMOC decline over the 21st century is very likely for all SSP 35 scenarios; a possible abrupt decline is assessed further in Box TS.3 (Figure TS.11b). {2.3.3, 3.5.4, 4.3.2, 36 8.6.1, 9.2.3, Cross-Chapter Box 12.3} 37 38 There is high confidence that many ocean currents will change in the 21st century in response to changes of 39 wind stress. There is low confidence in 21st century change of Southern Ocean circulation, despite high 40 confidence that it is sensitive to changes in wind patterns and increased ice-shelf melt. Western boundary 41 currents and subtropical gyres have shifted poleward since 1993 (medium confidence). Subtropical gyres, the 42 East Australian Current Extension, the Agulhas Current, and the Brazil Current are projected to intensify in 43 the 21st century in response to changes in wind stress, while the Gulf Stream and the Indonesian 44 Throughflow are projected to weaken (medium confidence). All of the four main eastern boundary upwelling 45 systems are projected to weaken at low latitudes and intensify at high latitudes in the 21st century (high 46 confidence). {2.3.3, 9.2.3} 47 48 It is virtually certain that surface pH has declined globally over the last 40 years and that the main driver is 49 uptake of anthropogenic CO2. Ocean acidification and associated reductions in the saturation state of calcium 50 carbonate – a constituent of skeletons or shells of a variety of marine organisms – is expected to increase in 51 the 21st century under all emissions scenarios (high confidence). There is very high confidence that present- 52 day surface pH values are unprecedented for at least 26,000 years and current rates of pH change are 53 unprecedented since at least that time. Over the past 2–3 decades, a pH decline in the ocean interior has been 54 observed in all ocean basins (high confidence) (Figure TS.11d). {2.3.3, 2.3.4, 3.6.2, 4.3.2, 5.3.2, 5.3.3, 5.6.3, 55 12.4.8} Do Not Cite, Quote or Distribute TS-41 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Open-ocean deoxygenation and expansion of oxygen minimum zones has been observed in many areas of 2 the global ocean since the mid 20th century (high confidence), in part due to human influence (medium 3 confidence). Deoxygenation is projected to continue to increase with ocean warming (high confidence) 4 (Figure TS.11c). Higher climate sensitivity and reduced ocean ventilation in CMIP6 compared to CMIP5 5 results in substantially greater projections of subsurface (100–600 m) oxygen decline than reported in 6 SROCC for the period 2080–2099. {2.3.3, 2.3.4, Cross-Chapter Box 2.4, 3.6.2, 5.3.3, 12.4.8} 7 8 Over at least the last two decades, the geographic range of many marine organisms has shifted towards the 9 poles and towards greater depths (high confidence), indicative of shifts towards cooler waters. The range of a 10 smaller subset of organisms has shifted equatorward and to shallower depths (high confidence). Phenological 11 metrics associated with the life cycles of many organisms have also changed over the last two decades or 12 longer (high confidence). Since the changes in the geographical range of organisms and their phenological 13 metrics have been observed to differ with species and location, there is the possibility of disruption to major 14 marine ecosystems. {2.3.4} 15 16 17 [START FIGURE TS.11 HERE] 18 19 Figure TS.11: Past and future ocean and ice sheet changes. Observed and simulated historical changes and 20 projected future changes under varying greenhouse gas emissions scenarios. Simulated and projected 21 ocean changes are shown as CMIP6 ensemble mean, and 5-95% range (shading) is provided for 22 scenario SSP1-2.6 and SSP3-7.0 (except in panel a where range provided for scenario SSP1-2.6 and 23 SSP5-8.5). Mean and 5-95% range in 2100 are shown as vertical bars on the right-hand side of each 24 panel. (a) Change in multiplication factor in surface ocean marine heatwave days relative to 1995- 25 2014 (defined as days exceeding the 99th percentile in SST from 1995-2014 distribution). Assessed 26 observational change span 1982-2019 from AVHRR satellite SST. (b) AMOC transport relative to 27 1995-2014 (defined as maximum transport at 26°N). Assessed observational change spans 2004-2018 28 from the RAPID array smoothed with a 12-month running mean (shading around the mean shows the 29 12-month running standard deviation around the mean). (c) Global mean percent change in ocean 30 oxygen (100–600 m depth), relative to 1995-2014. Assessed observational trends and very likely range 31 are from the SROCC assessment, and spans 1970–2010 centered on 2005. (d) Global mean surface 32 pH. Assessed observational change span 1985-2019, from the CMEMS SOCAT-based reconstruction 33 (shading around the global mean shows the 90% confidence interval). (e), (f): Ice sheet mass changes. 34 Projected ice sheet changes are shown as median, 5-95% range (light shading), and 17-83% range 35 (dark shading) of cumulative mass loss and sea level equivalent from ISMIP6 emulation under SSP1- 36 26 and SSP5-85 (shading and bold line), with individual emulated projections as thin lines. Median 37 (dot), 17-83% range (thick vertical bar), and 5-95% range (thin vertical bar) in 2100 are shown as 38 vertical bars on the right-hand side of each panel, from ISMIP6, ISMIP6 emulation, and LARMIP-2. 39 Observation-based estimates: For Greenland (e), for 1972-2018 (Mouginot), for 1992-2016 (Bamber), 40 for 1992-2020 (IMBIE) and total estimated mass loss range for 1840-1972 (Box). For Antarctica (f), 41 estimates based on satellite data combined with simulated surface mass balance and glacial isostatic 42 adjustment for 1992-2020 (IMBIE), 1992-2016 (Bamber), and 1979-2017 (Rignot). Left inset maps: 43 mean Greenland elevation changes 2010-2017 derived from CryoSat-2 radar altimetry (e) and mean 44 Antarctica elevation changes 1978-2017 derived from restored analog radar records (f). Right inset 45 maps: ISMIP6 model mean (2093- 2100) projected changes under the MIROC5 climate model for the 46 RCP8.5 scenario. {Box 9.2, 2.3.3, 2.3.4, 3.5.4, 4.3.2, 5.3.2, 5.3.3, 5.6.3, 9.2.3, 9.4.1, 9.4.2, Box 9.2 47 Figure 1, Figure 9.10, Figure 9.17, Figure 9.18} 48 49 [END FIGURE TS.11 HERE] 50 51 52 TS.2.5 The Cryosphere 53 54 Over recent decades, widespread loss of snow and ice has been observed, and several elements of the 55 cryosphere are now in states unseen in centuries (high confidence). Human influence was very likely the 56 main driver of observed reductions in Arctic sea ice since the late 1970s (with late-summer sea ice loss likely 57 unprecedented for at least 1000 years) and the widespread retreat of glaciers (unprecedented in at least the 58 last 2,000 years, medium confidence). Furthermore, human influence very likely contributed to the observed Do Not Cite, Quote or Distribute TS-42 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Northern Hemisphere spring snow cover decrease since 1950. By contrast, Antarctic sea ice area experienced 2 no significant net change since 1979, and there is only low confidence in its projected changes. The Arctic 3 Ocean is projected to become practically sea ice-free in late summer under high CO2 emissions scenarios by 4 the end of the 21st century (high confidence). It is virtually certain that further warming will lead to further 5 reductions of Northern Hemisphere snow cover, and there is high confidence that this is also the case for 6 near-surface permafrost volume. Glaciers will continue to lose mass at least for several decades even if 7 global temperature is stabilized (very high confidence), and mass loss over the 21st century is virtually 8 certain for the Greenland Ice Sheet and likely for the Antarctic Ice Sheet. Deep uncertainty persists with 9 respect to the possible evolution of the Antarctic Ice Sheet within the 21st century and beyond, in particular 10 due to the potential instability of the West Antarctic Ice Sheet. {2.3, 3.4, 4.3, 8.3, 9.3-9.6, Box 9.4, 12.4} 11 12 Current Arctic sea ice coverage levels (both annual and late summer) are at their lowest since at least 1850 13 (high confidence), and for late summer for the past 1,000 years (medium confidence). Since the late 1970s, 14 Arctic sea ice area and thickness have decreased in both summer and winter, with sea ice becoming younger, 15 thinner and more dynamic (very high confidence). It is very likely that anthropogenic forcing, mainly due to 16 greenhouse gas increases, was the main driver of this loss, although new evidence suggests that 17 anthropogenic aerosol forcing has offset part of the greenhouse gas–induced losses since the 1950s (medium 18 confidence). The annual Arctic sea ice area minimum will likely fall below 1 million km2 at least once before 19 2050 under all assessed SSP scenarios. This practically sea ice-free state will become the norm for late 20 summer by the end of the 21st century in high CO2 emissions scenarios (high confidence). Arctic summer 21 sea ice varies approximately linearly with global surface temperature, implying that there is no tipping point 22 and observed/projected losses are potentially reversible (high confidence). {2.3.2, 3.4.1, 4.3.2, 9.3.1, 12.4.9} 23 24 For Antarctic sea ice, there is no significant trend in satellite-observed sea ice area from 1979 to 2020 in both 25 winter and summer, due to regionally opposing trends and large internal variability. Due to mismatches 26 between model simulations and observations, combined with a lack of understanding of reasons for 27 substantial inter-model spread, there is low confidence in model projections of future Antarctic sea ice 28 changes, particularly at the regional level. {2.3.2, 3.4.1, 9.3.2} 29 30 In permafrost regions, increases in ground temperatures in the upper 30 m over the past three to four decades 31 have been widespread (high confidence). For each additional 1°C of warming (up to 4°C above the 1850– 32 1900 level), the global volume of perennially frozen ground to 3 m below the surface is projected to decrease 33 by about 25% relative to the present volume (medium confidence). However, these decreases may be 34 underestimated due to an incomplete representation of relevant physical processes in ESMs (low confidence). 35 Seasonal snow cover is treated in TS.2.6. {2.3.2, 9.5.2, 12.4.9} 36 37 There is very high confidence that, with few exceptions, glaciers have retreated since the second half of the 38 19th century; this behaviour is unprecedented in at least the last 2,000 years (medium confidence). Mountain 39 glaciers very likely contributed 67.2 [41.8 to 92.6] mm to the observed GMSL change between 1901 and 40 2018. This retreat has occurred at increased rates since the 1990s, with human influences very likely being 41 the main driver. Under RCP2.6 and RCP8.5, respectively, glaciers are projected to lose 18% ± 13% and 36% 42 ± 20% of their current mass over the 21st century (medium confidence). {2.3.2, 3.4.3, 9.5.1, 9.6.1} 43 44 The Greenland Ice Sheet was smaller than at present during the Last Interglacial period (roughly 125,000 45 years ago) and the mid-Holocene (roughly 6,000 years ago) (high confidence). After reaching a recent 46 maximum ice mass at some point between 1450 and 1850, the ice sheet retreated overall, with some decades 47 likely close to equilibrium (i.e., mass loss approximately equalling mass gained). It is virtually certain that 48 the Greenland Ice Sheet has lost mass since the 1990s, with human influence a contributing factor (medium 49 confidence). There is high confidence that annual mass changes have been consistently negative since the 50 early 2000s. Over the period 1992–2020, Greenland likely lost 4890 ± 460 Gt of ice, contributing 13.5 ± 1.3 51 mm to global mean sea level rise. There is high confidence that Greenland ice mass losses are increasingly 52 dominated by surface melting and runoff, with large interannual variability arising from changes in surface 53 mass balance. Projections of future Greenland ice-mass loss (Box TS.4, Table 1; Figure TS.11e) are 54 dominated by increased surface melt under all emissions scenarios (high confidence). Potential irreversible 55 long-term loss of the Greenland Ice Sheet, and of parts of the Antarctic Ice Sheet, is assessed in Box TS.9. Do Not Cite, Quote or Distribute TS-43 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 {2.3.2, 3.4.3, 9.4.1, 9.4.2, 9.6.3, Atlas.11.2} 2 3 It is likely that the Antarctic Ice Sheet has lost 2670 ± 530 Gt, contributing 7.4 ± 1.5 mm to global mean sea 4 level rise over 1992–2020. The total Antarctic ice mass losses were dominated by the West Antarctic Ice 5 Sheet, with combined West Antarctic and Peninsula annual loss rates increasing since about 2000 (very high 6 confidence). Furthermore, it is very likely that parts of the East Antarctic Ice Sheet have lost mass since 7 1979. Since the 1970s, snowfall has likely increased over the western Antarctic Peninsula and eastern West 8 Antarctica, with large spatial and interannual variability over the rest of Antarctica. Mass losses from West 9 Antarctic outlet glaciers, mainly induced by ice shelf basal melt (high confidence), outpace mass gain from 10 increased snow accumulation on the continent (very high confidence). However, there is only limited 11 evidence, with medium agreement, of anthropogenic forcing of the observed Antarctic mass loss since 1992 12 (with low confidence in process attribution). Increasing mass loss from ice shelves and inland discharge will 13 likely continue to outpace increasing snowfall over the 21st century (Figure TS.11f). Deep uncertainty 14 persists with respect to the possible evolution of the Antarctic Ice Sheet along high-end mass-loss storylines 15 within the 21st century and beyond, primarily related to the abrupt and widespread onset of Marine Ice Sheet 16 Instability and Marine Ice Cliff Instability. {2.3.2, 3.4.3, 9.4.2, 9.6.3, Box 9.4, Atlas.11.1, Box TS.3, Box 17 TS.4} 18 19 20 [START BOX TS.4 HERE] 21 22 Box TS.4: Sea Level 23 24 Global mean sea level (GMSL) increased by 0.20 [0.15 to 0.25] m over the period 1901 to 2018 with a rate 25 of rise that has accelerated since the 1960s to 3.7 [3.2 to 4.2] mm yr-1 for the period 2006–2018 (high 26 confidence). Human activity was very likely the main driver of observed GMSL rise since 1970, and new 27 observational evidence leads to an assessed sea level rise over the period 1901 to 2018 that is consistent with 28 the sum of individual components contributing to sea level rise, including expansion due to ocean warming 29 and melting of glaciers and ice sheets (high confidence). It is virtually certain that GMSL will continue to 30 rise over the 21st century in response to continued warming of the climate system (Box TS.4, Figure 1). Sea 31 level responds to GHG emissions more slowly than global surface temperature, leading to weaker scenario 32 dependence over the 21st century than for global surface temperature (high confidence). This slow response 33 also leads to long-term committed sea level rise, associated with ongoing ocean heat uptake and the slow 34 adjustment of the ice sheets, that will continue over the centuries and millennia following cessation of 35 emissions (high confidence). By 2100, GMSL is projected to rise by 0.28–0.55 m (likely range) under SSP1- 36 1.9 and 0.63–1.02 m (likely range) under SSP5-8.5 relative to the 1995–2014 average (medium confidence). 37 Under the higher CO2 emissions scenarios, there is deep uncertainty in sea level projections for 2100 and 38 beyond associated with the ice-sheet responses to warming. In a low-likelihood, high-impact storyline and a 39 high CO2 emissions scenario, ice-sheet processes characterized by deep uncertainty could drive GMSL rise 40 up to about 5 m by 2150. Given the long-term commitment, uncertainty in the timing of reaching different 41 GMSL rise levels is an important consideration for adaptation planning. {2.3, 3.4, 3.5, 9.6, Box 9.4, Box 42 TS.9 Cross-Chapter Box 9.1, Table 9.5} 43 44 GMSL change is driven by warming or cooling of the ocean (and the associated expansion/contraction) and 45 changes in the amount of ice and water stored on land. Paleo-evidence shows that GMSL has been about 70 46 m higher and 130 m lower than present within the past 55 million years and was likely 5 to 10 m higher 47 during the Last Interglacial (Box TS.2, Figure 1). Sea level observations show that GMSL rose by 0.20 [0.15 48 to 0.25] m over the period 1901–2018 at an average rate of 1.7 [1.3 to 2.2] mm yr-1. New analyses and paleo- 49 evidence since the AR5 show this rate is very likely faster than during any century over at least the last three 50 millennia (high confidence). Since the AR5, there is strengthened evidence for an increase in the rate of 51 GMSL rise since the mid-20th century, with an average rate of 2.3 [1.6–3.1] mm yr-1 over the period 1971– 52 2018 increasing to 3.7 [3.2–4.2] mm yr-1 for the period 2006–2018 (high confidence). {2.3.3.3, 9.6.1, 9.6.2} 53 54 GMSL will continue to rise throughout the 21st century (Box TS.4, Figure 1a). Considering only those 55 processes in whose projections we have at least medium confidence, relative to the period 1995–2014, Do Not Cite, Quote or Distribute TS-44 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 GMSL is projected to rise between 0.18 m (0.15–0.23 m, likely range) (SSP1-1.9) and 0.23 m (0.20–0.30 m, 2 likely range) (SSP5-8.5) by 2050. By 2100, the projected rise is between 0.38 m (0.28–0.55 m, likely range) 3 (SSP1-1.9) and 0.77 m (0.63–1.02 m, likely range) (SSP5-8.5) (Table 9.9). The methods, models and 4 scenarios used for sea level projections in the AR6 are updated from those employed by the SROCC, with 5 contributions informed by the latest model projections described in the ocean and cryosphere sections 6 (TS.2.4 and TS.2.5). Despite these differences, the sea level projections are broadly consistent with those of 7 the SROCC. {4.3.2, 9.6.3} 8 9 Importantly, likely range projections do not include those ice-sheet-related processes whose quantification is 10 highly uncertain or that are characterized by deep uncertainty. Higher amounts of GMSL rise before 2100 11 could be caused by earlier-than-projected disintegration of marine ice shelves, the abrupt, widespread onset 12 of Marine Ice Sheet Instability (MISI) and Marine Ice Cliff Instability (MICI) around Antarctica, and faster- 13 than-projected changes in the surface mass balance and dynamical ice loss from Greenland (Box TS.4, 14 Figure 1). In a low-likelihood, high-impact storyline and a high CO2 emissions scenario, such processes 15 could in combination contribute more than one additional meter of sea level rise by 2100. {Box TS.3, 4.3.2, 16 9.6.3, Box 9.4} 17 18 Beyond 2100, GMSL will continue to rise for centuries to millennia due to continuing deep ocean heat 19 uptake and mass loss from ice sheets, and will remain elevated for thousands of years (high confidence). By 20 2150, considering only those processes in whose projections we have at least medium confidence and 21 assuming no acceleration in ice-mass flux after 2100, GMSL is projected to rise between 0.6 m (0.4–0.9 m, 22 likely range) (SSP1-1.9) and 1.4 m (1.0–1.9 m, likely range) (SSP5-8.5), relative to the period 1995–2014 23 based on the SSP scenario extensions. Under high CO2 emissions, processes in which there is low 24 confidence, such as Marine Ice Cliff Instability (MICI), could drive GMSL rise up to about 5 m by 2150 25 (Box TS.4, Figure 1a). By 2300, GMSL will rise 0.3–3.1 m under low CO2 emissions (SSP1-2.6) (low 26 confidence). Under high CO2 emissions (SSP5-8.5), projected GMSL rise is between 1.7 and 6.8 m by 2300 27 in the absence of MICI and by up to 16 m considering MICI (low confidence). Over 2000 years, there is 28 medium agreement and limited evidence that committed GMSL rise is projected to be about 2–3 m with 29 1.5°C peak warming, 2–6 m with 2°C of peak warming, 4–10 m with 3°C of peak warming, 12–16 m with 30 4°C of peak warming, and 19–22 m with 5°C of peak warming. {TS.1.3.1, 9.6.3} 31 32 Looking at uncertainty in time provides an alternative perspective on uncertainty in future sea-level rise (Box 33 TS.4, Figure 1c). For example, considering only medium confidence processes, GMSL rise is likely to 34 exceed 0.5 m between about 2080 and 2170 under SSP1-2.6 and between about 2070 and 2090 under SSP5- 35 8.5. Given the long-term commitment, uncertainty in the timing of reaching different levels of GMSL rise is 36 an important consideration for adaptation planning. {9.6.3} 37 38 At regional scales, additional processes come into play that modify the local sea level change relative to 39 GMSL, including vertical land motion; ocean circulation and density changes and gravitational, rotational, 40 and deformational effects arising from the redistribution of water and ice mass between land and the ocean. 41 These processes give rise to a spatial pattern that tends to increase sea level rise at the low latitudes and 42 reduce sea-level rise at high latitudes. However, over the 21st century, the majority of coastal locations have 43 a median projected regional sea level rise within ± 20% of the projected GMSL change (medium 44 confidence). Further details on regional sea level change and extremes are provided in TS.4. {9.6.3} 45 46 47 [START BOX TS.4, FIGURE 1 HERE] 48 49 Box TS.4, Figure 1: Global mean sea level change on different time scales and under different scenarios. (a) 50 GMSL change from 1900 to 2150, observed (1900–2018) and projected under the SSP scenarios 51 (2000–2150), relative to a 1995–2014 baseline. Solid lines show median projections. Shaded 52 regions show likely ranges for SSP1-2.6 and SSP3-7.0. Dotted and dashed lines show respectively 53 the 83rd and 95th percentile low-confidence projections for SSP5-8.5. Bars at right 54 show likely ranges for SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5 in 2150. Lightly 55 shaded thick/thin bars show 17th–83rd/5th–95th percentile low-confidence ranges in 2150 for 56 SSP1-2.6 and SSP5-8.5, based upon projection methods incorporating structured expert judgement Do Not Cite, Quote or Distribute TS-45 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 and Marine Ice Cliff Instability. Low-confidence range for SSP5-8.5 in 2150 extends to 4.8/5.4 m 2 at the 83rd/95th percentile. (b) GMSL change on 100- (blue), 2,000- (green) and 10,000-year 3 (magenta) time scales as a function of global surface temperature, relative to 1850–1900. For 100- 4 year projections, GMSL is projected for the year 2100, relative to a 1995-2014 baseline, and 5 temperature anomalies are average values over 2081–2100. For longer-term commitments, 6 warming is indexed by peak warming above 1850–1900 reached after cessation of emissions. 7 Shaded regions show paleo-constraints on global surface temperature and GMSL for the Last 8 Interglacial and mid-Pliocene Warm Period. Lightly shaded thick/thin blue bars show 17th– 9 83rd/5th–95th percentile low-confidence ranges for SSP1-2.6 and SSP5-8.5 in 2100, plotted at 2°C 10 and 5°C. (c) Timing of exceedance of GMSL thresholds of 0.5, 1.0, 1.5 and 2.0 m, under different 11 SSPs. Lightly shaded thick/thin bars show 1th7–83rd/5th–95th percentile low-confidence ranges 12 for SSP1-2.6 and SSP5-8.5. 13 14 [END BOX TS.4, FIGURE 1 HERE] 15 16 [END BOX TS.4 HERE] 17 18 19 [START BOX TS.5 HERE] 20 21 Box TS.5: The Carbon Cycle 22 23 The continued growth of atmospheric CO2 concentrations over the industrial era is unequivocally due to 24 emissions from human activities. Ocean and land carbon sinks slow the rise of CO2 in the atmosphere. 25 Projections show that while land and ocean sinks absorb more CO2 under high emissions scenarios than low 26 emissions scenarios, the fraction of emissions removed from the atmosphere by natural sinks decreases with 27 higher concentrations (high confidence). Projected ocean and land sinks show similar responses for a given 28 scenario, but the land sink has a much higher inter-annual variability and wider model spread. The slowed 29 growth rates of the carbon sinks projected for the second half of this century are linked to strengthening 30 carbon–climate feedbacks and stabilization of atmospheric CO2 under medium-to-no-mitigation and high- 31 mitigation scenarios respectively. {5.2, 5.4} 32 33 Carbon sinks for anthropogenic CO2 are associated with mainly physical ocean and biospheric land processes 34 that drive the exchange of carbon between multiple land, ocean and atmospheric reservoirs. These exchanges 35 are driven by increasing atmospheric CO2, but modulated by changes in climate (Box TS.5, Figure 1 c,d). 36 The Northern and Southern Hemispheres dominate the land and ocean sinks, respectively (Box TS.5, Figure 37 1). Ocean circulation and thermodynamic processes also play a critical role in coupling the global carbon and 38 energy (heat) cycles. There is high confidence that this ocean carbon–heat nexus is an important basis for one 39 of the most important carbon–climate metrics, the transient climate response to cumulative CO2 emissions 40 (TCRE) (TS.3.2.1) used to determine the remaining carbon budget. {5.1, 5.2, 5.5, 9.2, Cross-Chapter Box 41 5.3} 42 43 Based on multiple lines of evidence using interhemispheric gradients of CO2 concentrations, isotopes, and 44 inventory data, it is unequivocal that the growth in CO2 in the atmosphere since 1750 (see TS.2.2) is due to 45 the direct emissions from human activities. The combustion of fossil fuels and land-use change for the period 46 1750–2019 resulted in the release of 700 ± 75 PgC (likely range, 1 PgC = 1015 g of carbon) to the 47 atmosphere, of which about 41% ± 11% remains in the atmosphere today (high confidence). Of the total 48 anthropogenic CO2 emissions, the combustion of fossil fuels was responsible for about 64% ± 15%, growing 49 to an 86% ± 14% contribution over the past 10 years. The remainder resulted from land-use change. During 50 the last decade (2010–2019), average annual anthropogenic CO2 emissions reached the highest levels in 51 human history at 10.9 ± 0.9 PgC yr-1 (high confidence). Of these emissions, 46% accumulated in the 52 atmosphere (5.1 ± 0.02 PgC yr-1), 23% (2.5 ± 0.6 PgC yr-1) was taken up by the ocean and 31% (3.4 ± 0.9 53 PgC yr-1) was removed by terrestrial ecosystems (high confidence). {5.2.1, 5.2.2, 5.2.3} 54 55 The ocean (high confidence) and land (medium confidence) sinks of CO2 have increased with anthropogenic 56 emissions over the past six decades (Box TS.5, Figure 1). This coherence between emissions and the growth Do Not Cite, Quote or Distribute TS-46 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 in ocean and land sinks has resulted in the airborne fraction of anthropogenic CO2 remaining at 44%±10% 2 over the past 60 years (high confidence). Interannual and decadal variability of the ocean and land sinks 3 indicate that they are sensitive to changes in the growth rate of emissions as well as climate variability, and 4 therefore also sensitive to climate change (high confidence). {5.2.1} 5 6 The land CO2 sink is driven by carbon uptake by vegetation, with large interannual variability, for example, 7 linked to ENSO. Since the 1980s, carbon fertilization from rising atmospheric CO2 has increased the strength 8 of the net land CO2 sink (medium confidence). During the historical period, the growth of the ocean sink has 9 been primarily determined by the growth rate of atmospheric CO2. However, there is medium confidence that 10 changes to physical and chemical processes in the ocean and in the land biosphere, which govern carbon 11 feedbacks, are already modifying the characteristics of variability, particularly the seasonal cycle of CO2 in 12 both the ocean and land. However, changes to the multi-decadal trends in the sinks have not yet been 13 observed. {2.3.4, 3.6.1, 5.2.1, TS.2.3, TS.2.6} 14 15 In AR6, ESM projections are assessed with CO2 concentrations by 2100 from about 400 ppm (SSP1-1.9) to 16 above 1100 ppm (SSP5-8.5). Most simulations are performed with prescribed atmospheric CO2 17 concentrations, which already account for a central estimate of climate–carbon feedback effects. CO2 18 emissions-driven simulations account for uncertainty in these feedbacks, but do not significantly change the 19 projected global surface temperature changes (high confidence). Although land and ocean sinks absorb more 20 CO2 under high emissions than low emissions scenarios, the fraction of emissions removed from the 21 atmosphere decreases (high confidence). This means that the more CO2 that is emitted, the less efficient the 22 ocean and land sinks become (high confidence), an effect which compensates for the logarithmic relationship 23 between CO2 and its radiative forcing, which means that for each unit increase in additional atmospheric CO2 24 the effect on global temperature decreases. (Box TS.5, Figure 1f,g). {4.3.1, 5.4.5, 5.5.1.2} 25 26 Ocean and land sinks show similar responses for a given scenario, but the land sink has a much higher inter- 27 annual variability and wider model spread. Under SSP3-7.0 and SSP5-8.5, the initial growth of both sinks in 28 response to increasing atmospheric concentrations of CO2 is subsequently limited by emerging carbon– 29 climate feedbacks (high confidence) (Box TS.5, Figure 1f). Projections show that the ocean and land sinks 30 will stop growing from the second part of the 21st century under all emissions scenarios, but with different 31 drivers for different emissions scenarios. Under SSP3-7.0 and SSP5-8.5, the weakening growth rate of the 32 ocean CO2 sink in the second half of the century is primarily linked to the strengthening positive feedback 33 from reduced carbonate buffering capacity, ocean warming and altered ocean circulation (e.g., AMOC 34 changes). In contrast, for SSP1-1.9, SSP1-2.6 and SSP2-4.5, the weakening growth rate of the ocean carbon 35 sink is a response to the stabilizing or declining atmospheric CO2 concentrations. Under high CO2 emissions 36 scenarios, it is very likely that the land carbon sink will grow more slowly due to warming and drying from 37 the mid 21st century, but it is very unlikely that it will switch from being a sink to a source before 2100. 38 Climate change alone is expected to increase land carbon accumulation in the high latitudes (not including 39 permafrost, which is assessed in TS2.5 and TS.3.2.2), but also to lead to a counteracting loss of land carbon 40 in the tropics (medium confidence). ESM projections show that the overall uncertainty of atmospheric CO2 41 by 2100 is still dominated by the emission pathway, but carbon–climate feedbacks (see TS.3.3.2) are 42 important, with increasing uncertainties in high emissions pathways (Box TS.5, Figure 1e). {4.3.2, 5.4.1, 43 5.4.2, 5.4.4, 5.4.5, 11.6, 11.9, Cross-Chapter Box 5.1, Cross-Chapter Box 5.3, TS.3.3.2} 44 45 Under three SSP scenarios with long-term extensions until 2300 (SSP5-8.5, SSP5-3.4-OS, SSP1-2.6), ESMs 46 project a change of the land from a sink to a source (medium confidence). The scenarios make simplified 47 assumptions about emissions reductions, with SSP1-2.6 and SSP5-3.4-OS reaching about 400 ppm by 2300, 48 while SSP5-8.5 exceeds 2000 ppm. Under high emissions the transition is warming-driven, whereas it is 49 linked to the decline in atmospheric CO2 under net negative CO2 emissions. The ocean remains a sink 50 throughout the period to 2300 except under very large net negative emissions. The response of the natural 51 aspects of the carbon cycle to carbon dioxide removal is further developed in TS.3.3.2. {5.4.9, TS.3.3.2; 52 Figure Box TS.5f} 53 54 Do Not Cite, Quote or Distribute TS-47 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [START BOX TS.5, FIGURE 1 HERE] 2 3 Box TS.5, Figure 1: Carbon cycle processes and projections. Carbon cycle response to forcings. The figure shows 4 changes in carbon storage in response to elevated CO2 (a, b) and the response to climate warming 5 (c, d). Maps show spatial patterns of changes in carbon uptake during simulations with 1% per 6 year increase in CO2 {section 5.4.5.5}, and zonal mean plots show distribution of carbon changes 7 is dominated by the land (green lines) in the tropics and northern hemisphere and ocean (blue 8 lines) in the southern hemisphere. Hatching indicates regions where fewer than 80% of models 9 agree on the sign of response. (e) Future CO 2 projections: projected CO2 concentrations in the SSP 10 scenarios in response to anthropogenic emissions, results from coupled ESMs for SSP5-8.5 and 11 from the MAGICC7 emulator for other scenarios {section 4.3.1}. (f) Future carbon fluxes: 12 projected combined land and ocean fluxes (positive downward) up to 2100 for the SSP scenarios, 13 and extended to 2300 for available scenarios, 5-95% uncertainty plumes shown for SSP1-2.6 and 14 SSP3-7.0 {section 5.4.5.4, 5.4.10}. The numbers near the top show the number of model 15 simulations used. (g) Sink fraction: the fraction of cumulative emissions of CO 2 removed by land 16 and ocean sinks. The sink fraction is smaller under conditions of higher emissions. {5.4.5, 5.5.1; 17 Figure 5.27; Figure 4.31; Figure 5.25; Figure 5.30; Figure 5.31} 18 19 [END BOX TS.5, FIGURE 1 HERE] 20 21 22 23 [END BOX TS.5 HERE] 24 25 26 TS.2.6 Land Climate, Including Biosphere and Extremes 27 28 Land surface air temperatures have risen faster than the global surface temperature since the 1850s, and it is 29 virtually certain that this differential warming will persist into the future. It is virtually certain that the 30 frequency and intensity of hot extremes and the intensity and duration of heat waves have increased since 31 1950 and will further increase in the future even if global warming is stabilized at 1.5°C. The frequency and 32 intensity of heavy precipitation events have increased over a majority of those land regions with good 33 observational coverage (high confidence) and will extremely likely increase over most continents with 34 additional global warming. Over the past half century, key aspects of the biosphere have changed in ways 35 that are consistent with large-scale warming: climate zones have shifted poleward, and the growing season 36 length in the Northern Hemisphere extratropics has increased (high confidence). The amplitude of the 37 seasonal cycle of atmospheric CO2 poleward of 45°N has increased since the 1960s (very high confidence), 38 with increasing productivity of the land biosphere due to the increasing atmospheric CO2 concentration as 39 the main driver (medium confidence). Global-scale vegetation greenness has increased since the 1980s (high 40 confidence). {2.3, 3.6, 4.3, 4.5, 5.2, 11.3, 11.4, 11.9, 12.4} 41 42 Observed temperatures over land have increased by 1.61 [1.34–1.83] °C between the period 1850–1900 and 43 2011–2020. Warming of the land is about 45% larger than for global surface temperature, and about 80% 44 larger than warming of the ocean surface. Warming of the land surface during the period 1971–2018 45 contributed about 5% of the increase in the global energy inventory (TS.3.1), nearly twice the estimate in 46 AR5 (high confidence). It is virtually certain that the average surface warming over land will continue to be 47 higher than over the ocean throughout the 21st century. The warming pattern will likely vary seasonally, with 48 northern high latitudes warming more during winter than summer (medium confidence). {2.3.1, 4.3.1, 4.5.1, 49 Box 7.2, 7.2.2, Cross-Chapter Box 9.1, 11.3, Atlas 11.2} 50 51 The frequency and intensity of hot extremes (warm days and nights) and the intensity and duration of 52 heatwaves have increased globally and in most regions since 1950, while the frequency and intensity of cold 53 extremes have decreased (virtually certain). There is high confidence that the increases in frequency and 54 severity of hot extremes are due to human-induced climate change. Some recent extreme events would have 55 been extremely unlikely to occur without human influence on the climate system. It is virtually certain that 56 the further changes in hot and cold extremes will occur throughout the 21st century in nearly all inhabited Do Not Cite, Quote or Distribute TS-48 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 regions, even if global warming is stabilized at 1.5°C (Table TS.2, Figure TS.12a). {1.3, Cross-Chapter Box 2 3.2, 11.1.4, 11.3.2, 11.3.4, 11.3.5, 11.9, 12.4} 3 4 Greater warming over land alters key water cycle characteristics (Box TS.6). The rate of change in mean 5 precipitation and runoff, and their variability, increases with global warming (Figure TS.12e,f). The majority 6 of the land area has experienced decreases in available water during dry seasons due to the overall increase 7 in evapotranspiration (medium confidence). The land area affected by increasing drought frequency and 8 severity will expand with increasing global warming (high confidence; Figure TS.12c). There is low 9 confidence that the increase of plant water-use efficiency due to higher atmospheric CO2 concentration 10 alleviates extreme agricultural and ecological droughts in conditions characterized by limited soil moisture 11 and increased atmospheric evaporative demand. {2.3.1, CCB 5.1, 8.2.3, 8.4.1, Box 11.1, 11.2.4, 11.4, 11.6} 12 13 Northern Hemisphere spring snow cover has decreased since at least 1978 (very high confidence), and there 14 is high confidence that trends in snow cover loss extend back to 1950. It is very likely that human influence 15 contributed to these reductions. Earlier onset of snowmelt has contributed to seasonally dependent changes 16 in streamflow (high confidence). A further decrease of Northern Hemisphere seasonal snow cover extent is 17 virtually certain under further global warming (Figure TS.12d). {2.3.2, 3.4.2, 8.3.2. 9.5.3, 12.4, Atlas 8.2, 18 9.2, 11.2} 19 20 The frequency and intensity of heavy precipitation events have increased over a majority of land regions 21 with good observational coverage since 1950 (high confidence, Box TS.6, Table TS.2). Human influence is 22 likely the main driver of this change (Table TS.2). It is extremely likely that on most continents heavy 23 precipitation will become more frequent and more intense with additional global warming (Table TS.2, 24 Figure TS.12 b). The projected increase in heavy precipitation extremes translates to an increase in the 25 frequency and magnitude of pluvial floods (high confidence) (Table TS.2). {Cross-Chapter Box 3.2, 8.4.1, 26 11.4.2, 11.4.4, 11.5.5, 12.4} 27 28 The probability of compound extreme events has likely increased due to human-induced climate change. 29 Concurrent heat waves and droughts have become more frequent over the last century, and this trend will 30 continue with higher global warming (high confidence). The probability of compound flooding (storm surge, 31 extreme rainfall and/or river flow) has increased in some locations, and will continue to increase due to both 32 sea level rise and increases in heavy precipitation, including changes in precipitation intensity associated 33 with tropical cyclones (high confidence). {11.8.1, 11.8.2, 11.8.3} 34 35 Changes in key aspects of the terrestrial biosphere, such as an increase of the growing season length in much 36 of the Northern Hemisphere extratropics since the mid-20th century (high confidence), are consistent with 37 large-scale warming. At the same time an increase in the amplitude of the seasonal cycle of atmospheric CO2 38 beyond 45°N since the early 1960s (high confidence) and a global-scale increase in vegetation greenness of 39 the terrestrial surface since the early 1980s (high confidence) have been observed. Increasing atmospheric 40 CO2, warming at high latitudes and land management interventions have contributed to the observed 41 greening trend, but there is low confidence in their relative roles. There is medium confidence that increased 42 plant growth associated with CO2 fertilization is the main driver of the observed increase in amplitude of the 43 seasonal cycle of atmospheric CO2 in the Northern hemisphere. Reactive nitrogen, ozone and aerosols affect 44 terrestrial vegetation and carbon cycle through deposition and effects on large-scale radiation (high 45 confidence), but the magnitude of these effects on the land carbon sink, ecosystem productivity and indirect 46 CO2 forcing remains uncertain. {2.3.4, 3.6.1, 5.2.1, 6.4.5, 12.3.7, 12.4} 47 48 Over the last century, there has been a poleward and upslope shift in the distribution of many land species 49 (very high confidence) as well as increases in species turnover within many ecosystems (high confidence). 50 There is high confidence that the geographical distribution of climate zones has shifted in many parts of the 51 world in the last half century. SRCCL concluded that continued warming will exacerbate desertification 52 processes (medium confidence) and ecosystems will become increasingly exposed to climates beyond those 53 that they are currently adapted to (high confidence). There is medium confidence that climate change will 54 increase disturbance by, for example, fire and tree mortality across several ecosystems. Increases are 55 projected in drought, aridity, and fire weather in some regions (TS.4.3; high confidence). There is low Do Not Cite, Quote or Distribute TS-49 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 confidence in the magnitude of these changes, but the probability of crossing uncertain regional thresholds 2 (e.g., fires, forest dieback) increases with further warming (high confidence). The response of 3 biogeochemical cycles to the anthropogenic perturbation can be abrupt at regional scales, and irreversible on 4 decadal to century time scales (high confidence). {2.3.4, 5.4.3, 5.4.9, 11.6, 11.8, 12.5, SRCCL 2.2, SRCCL 5 2.5, SR1.5 3.4} 6 7 8 [START FIGURE TS.12 HERE] 9 10 Figure TS.12: Land-related changes relative to the 1850-1900 as a function of global warming levels. a) 11 Changes in the frequency (left scale) and intensity (in °C, right scale) of daily hot extremes occurring 12 every 10- and 50-years; b) as a), but for daily heavy precipitation extremes, with intensity change in 13 %; c) Changes in 10-year droughts aggregated over drought-prone regions (CNA, NCA, SCA, NSA, 14 SAM, SWS, SSA, MED, WSAF, ESAF, MDG, SAU, and EAU; for definitions of these regions, see 15 Atlas.2), with drought intensity (right scale) represented by the change of annual mean soil moisture, 16 normalized with respect to interannual variability; d) Changes in Northern Hemisphere spring (March- 17 April-May) snow cover extent relative to 1850-1900; e,f) Relative change (%) in annual mean of total 18 precipitable water (grey line), precipitation (red solid lines), runoff (blue solid lines) and in standard 19 deviation (i.e. variability) of precipitation (red dashed lines) and runoff (blue dashed lines) averaged 20 over (e) tropical and (f) extratropical land as function of global warming levels. CMIP6 models that 21 reached a 5°C warming level in the 21st century in SSP5-8.5 above the 1850-1900 average have been 22 used. Precipitation and runoff variability are estimated by respective standard deviation after 23 removing linear trends. Error bars show the 17-83% confidence interval for the warmest +5°C global 24 warming level. {Figures 11.6, 11.7, 11.12, 11.15, 11.18, 9.24, 8.16, Atlas.2} 25 26 [END FIGURE TS.12 HERE] 27 28 29 [START BOX TS.6 HERE] 30 31 Box TS.6: Water Cycle 32 33 Human-caused climate change has driven detectable changes in the global water cycle since the mid-20th 34 century (high confidence), and it is projected to cause substantial further changes at both global and regional 35 scales (high confidence). Global land precipitation has likely increased since 1950, with a faster increase 36 since the 1980s (medium confidence). Atmospheric water vapour has increased throughout the troposphere 37 since at least the 1980s (likely). Annual global land precipitation will increase over the 21st century as global 38 surface temperature increases (high confidence). Human influence has been detected in amplified surface 39 salinity and precipitation minus evaporation (P-E) patterns over the ocean (high confidence). The severity of 40 very wet and very dry events increase in a warming climate (high confidence), but changes in atmospheric 41 circulation patterns affect where and how often these extremes occur. Water cycle variability and related 42 extremes are projected to increase faster than mean changes in most regions of the world and under all 43 emission scenarios (high confidence). Over the 21st century, the total land area subject to drought will 44 increase and droughts will become more frequent and severe (high confidence). Near-term projected changes 45 in precipitation are uncertain mainly because of internal variability, model uncertainty and uncertainty in 46 forcings from natural and anthropogenic aerosols (medium confidence). Over the 21st century and beyond, 47 abrupt human-caused changes to the water cycle cannot be excluded (medium confidence). {2.3, 3.3, 4.3, 4.4, 48 4.5, 4.6, 8.2, 8.3, 8.4, 8.5, 8.6, 11.4, 11.6, 11.9} 49 50 There is high confidence that the global water cycle has intensified since at least 1980 expressed by, for 51 example, increased atmospheric moisture fluxes and amplified precipitation minus evaporation patterns. 52 Global land precipitation has likely increased since 1950, with a faster increase since the 1980s (medium 53 confidence), and a likely human contribution to patterns of change, particularly for increases in high-latitude 54 precipitation over the Northern Hemisphere. Increases in global mean precipitation are determined by a 55 robust response to global surface temperature (very likely 2–3% per °C) that is partly offset by fast 56 atmospheric adjustments to atmospheric heating by GHGs and aerosols (TS.3.2.2). The overall effect of Do Not Cite, Quote or Distribute TS-50 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 anthropogenic aerosols is to reduce global precipitation through surface radiative cooling effects (high 2 confidence). Over much of the 20th century, opposing effects of GHGs and aerosols on precipitation have 3 been observed for some regional monsoons (high confidence) (Box TS.13). Global annual precipitation over 4 land is projected to increase on average by 2.4% (–0.2% to 4.7% likely range) under SSP1-1.9, 4.6% (1.5% 5 to 8.3% likely range) under SSP2-4.5, and 8.3% (0.9% to 12.9% likely range) under SSP5-8.5 by 2081–2100 6 relative to 1995–2014 (Box TS.6, Figure 1). Inter-model differences and internal variability contribute to a 7 substantial range in projections of large-scale and regional water cycle changes (high confidence). The 8 occurrence of volcanic eruptions can alter the water cycle for several years (high confidence). Projected 9 patterns of precipitation change exhibit substantial regional differences and seasonal contrast as global 10 surface temperature increases over the 21st century (Box TS.6, Figure 1). {2.3.1, 3.3.2, 3.3.3, 3.5.2, 4.3.1, 11 4.4.1, 4.5.1, 4.6.1, Cross-Chapter Box 4.1, 8.2.1, 8.2.2, 8.2.3, Box 8.1, 8.3.2.4, 8.4.1, 8.5.2, 10.4.2} 12 13 Global total column water vapour content has very likely increased since the 1980s, and it is likely that 14 human influence has contributed to tropical upper tropospheric moistening. Near-surface specific humidity 15 has increased over the ocean (likely) and land (very likely) since at least the 1970s, with a detectable human 16 influence (medium confidence). Human influence has been detected in amplified surface salinity and 17 precipitation minus evaporation (P-E) patterns over the ocean (high confidence). It is virtually certain that 18 evaporation will increase over the ocean, and very likely that evapotranspiration will increase over land, with 19 regional variations under future surface warming (Box TS.6, Figure 1). There is high confidence that 20 projected increases in precipitation amount and intensity will be associated with increased runoff in northern 21 high latitudes (Box TS.6, Figure 1). In response to cryosphere changes (TS.2.5), there have been changes in 22 streamflow seasonality, including an earlier occurrence of peak streamflow in high-latitude and mountain 23 catchments (high confidence). Projected runoff (Box TS.6, Figure 1c) is typically decreased by contributions 24 from small glaciers because of glacier mass loss, while runoff from larger glaciers will generally increase 25 with increasing global warming levels until their mass becomes depleted (high confidence). {2.3.1, 3.3.2, 26 3.3.3, 3.5.2, 8.2.3, 8.4.1, 11.5} 27 28 Warming over land drives an increase in atmospheric evaporative demand and in the severity of drought 29 events (high confidence). Greater warming over land than over the ocean alters atmospheric circulation 30 patterns and reduces continental near-surface relative humidity, which contributes to regional drying (high 31 confidence). A very likely decrease in relative humidity has occurred over much of the global land area since 32 2000. Projected increases in evapotranspiration due to growing atmospheric water demand will decrease soil 33 moisture (Box TS.6, Figure 1) over the Mediterranean region, southwestern North America, South Africa, 34 southwestern South America and southwestern Australia (high confidence). Some tropical regions are also 35 projected to experience enhanced aridity, including the Amazon basin and Central America (high 36 confidence). The total land area subject to increasing drought frequency and severity will expand (high 37 confidence), and in the Mediterranean, southwestern South America, and western North America, future 38 aridification will far exceed the magnitude of change seen in the last millennium (high confidence). {4.5.1, 39 8.2.2, 8.2.3, 8.4.1, Box 8.2, 11.6, 11.9} 40 41 Land-use change and water extraction for irrigation have influenced local and regional responses in the water 42 cycle (high confidence). Large-scale deforestation likely decreases evapotranspiration and precipitation and 43 increases runoff over the deforested regions relative to the regional effects of climate change (medium 44 confidence). Urbanization increases local precipitation (medium confidence) and runoff intensity (high 45 confidence) (Box TS.14:). Increased precipitation intensities have enhanced groundwater recharge, most 46 notably in tropical regions (medium confidence). There is high confidence that groundwater depletion has 47 occurred since at least the start of the 21st century, as a consequence of groundwater withdrawals for 48 irrigation in agricultural areas in drylands. {8.2.3, 8.3.1, 11.1.6, 11.4, 11.6, FAQ 8.1} 49 50 51 [START BOX TS.6, FIGURE 1 HERE] 52 53 Box TS.6, Figure 1: Projected water cycle changes. Long-term (2081–2100) projected annual mean changes (%) 54 relative to present-day (1995–2014) in the SSP2-4.5 emissions scenario for (a) precipitation, (b) 55 surface evapotranspiration, (c) total runoff and (d) surface soil moisture. Top-right panel numbers Do Not Cite, Quote or Distribute TS-51 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 indicate the number of CMIP6 models used for estimating the ensemble mean. For other scenarios, 2 please refer to relevant figures in Chapter 8. Uncertainty is represented using the simple approach: 3 No overlay indicates regions with high model agreement, where ≥80% of models agree on sign of 4 change; diagonal lines indicate regions with low model agreement, where <80% of models agree 5 on sign of change. For more information on the simple approach, please refer to the Cross-Chapter 6 Box Atlas.1. {8.4.1, Figures 8.14, 8.17, 8.18, 8.19} 7 8 [END BOX TS.6, FIGURE 1 HERE] 9 10 11 Water cycle variability and related extremes are projected to increase faster than mean changes in most 12 regions of the world and under all emissions scenarios (high confidence). A warmer climate increases 13 moisture transport into weather systems, which intensifies wet seasons and events (high confidence). The 14 magnitudes of projected precipitation increases and related extreme events depend on model resolution and 15 the representation of convective processes (high confidence). Increases in near-surface atmospheric moisture 16 capacity of about 7% per 1ºC of warming lead to a similar response in the intensification of heavy 17 precipitation from sub-daily up to seasonal time scales, increasing the severity of flood hazards (high 18 confidence). The average and maximum rain-rates associated with tropical and extratropical cyclones, 19 atmospheric rivers and severe convective storms will therefore also increase with future warming (high 20 confidence). For some regions, there is medium confidence that peak tropical cyclone rain-rates will increase 21 by more than 7% per 1°C of warming due to increased low-level moisture convergence caused by increases 22 in wind intensity. In the tropics year-round and in the summer season elsewhere, interannual variability of 23 precipitation and runoff over land is projected to increase at a faster rate than changes in seasonal mean 24 precipitation (Figure TS.12e,f) (medium confidence). Sub-seasonal precipitation variability is also projected 25 to increase with fewer rainy days but increased daily mean precipitation intensity over many land regions 26 (high confidence). {4.5.3, 8.2.3, 8.4.1, 8.4.2, 8.5.1, 8.5.2, 11.4,11.5, 11.7, 11.9} 27 28 [END BOX TS.6 HERE] 29 30 31 [START INFOGRAPHIC TS.1 HERE] 32 33 Infographic TS.1: Climate Futures. 34 (top left) Annual emissions of CO2 for the five core Shared Socio-economic Pathway 35 (SSP) scenarios (very low: SSP1-1.9, low: SSP1-2.6, medium: SSP2-4.5, high: SSP3-7.0, 36 very high SSP5-8.5). (bottom left) Projected warming for each of these emissions 37 scenarios. (top right) Response of some selected climate variables to 4 levels of global 38 warming (°C). (bottom right) The long-term effect of each global warming level on sea 39 level. See TS.1.3.1 for more detail on the SSP climate change scenarios. 40 41 Do Not Cite, Quote or Distribute TS-52 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 a) 2 3 4 Do Not Cite, Quote or Distribute TS-53 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 b) 2 3 4 [END INFOGRAPHIC TS.1 HERE] 5 6 7 8 Do Not Cite, Quote or Distribute TS-54 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.3 Understanding the Climate System Response and Implications for Limiting Global Warming 2 3 This section summarizes advances in our knowledge of the Earth’s energy budget, including the time 4 evolution of forcings and climate feedbacks that lead to the climate system responses summarized in TS.2. It 5 assesses advances since AR5 and SR1.5 in the estimation of remaining carbon budgets, the Earth system 6 response to carbon dioxide removal and quantification of metrics that allow comparison of the relative 7 effects of different forcing agents. The section also highlights future climate and air pollution responses due 8 to projected changes in short-lived climate forcers (SLCFs), the state of understanding of the climate 9 response to potential interventions related to solar radiation modification (SRM), and irreversibility, tipping 10 points and abrupt changes in the climate system. 11 12 13 TS.3.1 Radiative Forcing and Energy Budget 14 15 Since AR5, the accumulation of energy in the Earth system, quantified by observations of warming of the 16 ocean, atmosphere, land and melting of ice, has become established as a robust measure of the rate of global 17 climate change on interannual-to-decadal time scales. Compared to changes in global surface temperature, 18 the increase in the global energy inventory exhibits less variability, and thus better indicates underlying 19 climate trends. The global energy inventory increased by 435 [325 to 545] Zettajoules (ZJ, equal to 1021 20 Joules) for the period 1971–2018 and 153 [100 to 206] ZJ for the period 2006–2018 (Figure TS.13), with 21 more than 90% accounted for by ocean warming. To put these numbers in context, the 2006–2018 average 22 Earth system heating is equivalent to approximately 20 times the rate of global energy consumption in 2018. 23 The accumulation of energy is driven by a positive total anthropogenic effective radiative forcing (ERF) 24 relative to 1750. The best estimate ERF of 2.72 W m−2 has increased by 0.43 Wm-2 relative to that given in 25 AR5 (for 1750–2014) due to an increase in the greenhouse gas ERF that is partly compensated by a more 26 negative aerosol ERF compared to AR5. The greenhouse gas ERF has been revised due to changes in 27 atmospheric concentrations and updates to forcing efficiencies, while the revision to aerosol ERF is due to 28 increased understanding of aerosol–cloud interactions and is supported by improved agreement between 29 different lines of evidence. Improved quantification of ERF, the climate system radiative response and the 30 observed energy increase in the Earth system for the period 1971–2018 demonstrate improved closure of the 31 global energy budget (i.e., the extent to which the sum of the integrated forcing and the integrated radiative 32 response equals the energy gain of the Earth system) compared to AR5. (high confidence) {7.2.2, Box 7.2, 33 Table 7.1, 7.3.5, 7.5.2, FAQ7.1} 34 35 36 [START FIGURE TS.13 HERE] 37 38 Figure TS.13: Estimates of the net cumulative energy change (ZJ = 10 21 Joules) for the period 1971–2018 39 associated with: (a) observations of changes in the Global Energy Inventory (b) Integrated 40 Radiative Forcing; (c) Integrated Radiative Response. The intent is to show assessed changes in 41 energy budget and ERFs. Black dotted lines indicate the central estimate with likely and very likely 42 ranges as indicated in the legend. The grey dotted lines indicate the energy change associated with an 43 estimated 1850-1900 Earth energy imbalance of 0.2 W m-2 (panel a) and an illustration of an assumed 44 pattern effect of –0.5 W m–2°C–1 (panel c). Background grey lines indicate equivalent heating rates in 45 W m–2 per unit area of Earth’s surface. Panels (d) and (e) show the breakdown of components, as 46 indicated in the legend, for the Global Energy Inventory and Integrated Radiative Forcing, 47 respectively. Panel (f) shows the Global Energy Budget assessed for the period 1971–2018, i.e. the 48 consistency between the change in the Global Energy Inventory relative to 1850-1900 and the implied 49 energy change from Integrated Radiative Forcing plus Integrated Radiative Response under a number 50 of different assumptions, as indicated in the figure legend, including assumptions of correlated and 51 uncorrelated uncertainties in Forcing plus Response. Shading represents the very likely range for 52 observed energy change relative to 1850-1900 and likely range for all other quantities. Forcing and 53 Response timeseries are expressed relative to a baseline period of 1850–1900. {Box 7.2 Figure 1} 54 55 [END FIGURE TS.13 HERE] 56 Do Not Cite, Quote or Distribute TS-55 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 The global energy inventory change for the period 1971–2018 corresponds to an Earth energy imbalance 2 (Box TS.1) of 0.57 [0.43 to 0.72] W m-2, increasing to 0.79 [0.52 to 1.06] W m-2 for the period 2006–2018. 3 Ocean heat uptake is by far the largest contribution and accounts for 91% of the total energy change. Land 4 warming, melting of ice and warming of the atmosphere account for about 5%, 3% and 1% of the total 5 change, respectively. More comprehensive analysis of inventory components, cross-validation of satellite 6 and in situ-based estimates of the global energy imbalance and closure of the global sea level budget have 7 led to a strengthened assessment relative to AR5. (high confidence) {Box 7.2, 7.2.2, Table 7.1, 7.5.2.3, 8 Cross-Chapter Box 9.1, 9.6.1, Table 9.5} 9 10 As in AR5, the perturbations to the Earth’s top-of-atmosphere energy budget are quantified using ERFs (also 11 see TS.2.2). These include any consequent adjustments to the climate system (e.g., from changes in 12 atmospheric temperatures, clouds, and water vapour as shown in Figure TS.14), but exclude any surface 13 temperature response. Since the AR5, ERFs have been estimated for a larger number of forcing agents and 14 shown to be more closely related to the temperature response than the stratospheric-temperature adjusted 15 radiative forcing. (high confidence) {7.3.1} 16 17 18 [START FIGURE TS.14 HERE] 19 20 Figure TS.14: Schematic representation of changes in the top-of atmosphere (TOA) radiation budget following 21 a perturbation. The intent of the figure is to illustrate the concept of adjustments in the climate 22 system following a perturbation in the radiation budget. The baseline TOA energy budget (a) 23 responds instantaneously to perturbations (b), leading to adjustments in the atmospheric meteorology 24 and composition, and land surface that are independent of changes in surface temperature (c). Surface 25 temperature changes (here using an increase as an example) lead to physical, biogeophysical and 26 biogeochemical feedback processes (d). Long term feedback processes, such as those involving ice 27 sheets, are not shown here. {adapted from Chapter 7 Figure 7.2, FAQ 7.2 Figure 1, and Figure 8.3} 28 29 [END FIGURE TS.14 HERE] 30 31 32 Improved quantification of ERF, the climate system radiative response, and the observed energy increase in 33 the Earth system for the period 1971–2018, demonstrate improved closure of the global energy budget 34 relative to AR5 (Figure TS.13). Combining the likely range of ERF over this period with the central estimate 35 of radiative response gives an expected energy gain of 340 [47 to 662] ZJ. Both estimates are consistent with 36 an independent observation-based assessment of the global energy increase of 284 [96 to 471] ZJ, (very 37 likely range) expressed relative to the estimated 1850–1900 Earth energy imbalance. (high confidence) 38 {7.2.2, Box 7.2, 7.3.5} 39 40 The assessed greenhouse gas ERF over the 1750–2019 period (TS.2.2) has increased by 0.59 W m−2 over 41 AR5 estimates for 1750–2011. This increase includes +0.34 W m-2 from increases in atmospheric 42 concentrations of well-mixed greenhouse gases (including halogenated species) since 2011, +0.15 W m-2 43 from upwards revisions of their radiative efficiencies and +0.10 W m-2 from re-evaluation of the ozone and 44 stratospheric water vapour ERF. {7.3.2, 7.3.4, 7.3.5} 45 46 For CO2, CH4, N2O, and chlorofluorocarbons, there is now evidence to quantify the effect on ERF of 47 tropospheric adjustments. The assessed ERF for a doubling of CO2 compared to 1750 levels (3.9 ± 0.5 48 Wm‑2) is larger than in AR5. For CO2, the adjustments include the physiological effects on vegetation. The 49 reactive well-mixed greenhouse gases (methane, nitrous oxide, and halocarbons) cause additional chemical 50 adjustments to the atmosphere through changes in ozone and aerosols (Figure TS.15a). The ERF due to 51 methane emissions is 1.21 [0.90 to 1.51] W m–2, of which 0.33 [0.25 to 0.41] W m–2 is attributed to chemical 52 adjustments mainly via ozone. These chemical adjustments also affect the emission metrics (Section 53 TS.3.3.3). SO2 emission changes make the dominant contribution to the ERF from aerosol–cloud interactions 54 (high confidence). Over the 1750–2019 period, the contributions from the emitted compounds to global 55 surface temperature changes broadly match their contributions to the ERF (high confidence) (Figure 56 TS.15b). Since a peak in emissions-induced SO2 ERF has already occurred recently (TS.2.2) and since there Do Not Cite, Quote or Distribute TS-56 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 is a delay in the full global surface temperature response owing to the thermal inertia in the climate system, 2 changes in SO2 emissions have a slightly larger contribution to global surface temperature change compared 3 with changes in CO2 emissions, relative to their respective contributions to ERF. {6.4.2, 7.3.2} 4 5 Aerosols contributed an ERF of –1.3 [–2.0 to –0.6] W m–2 over the period 1750 to 2014 (medium 6 confidence). The ERF due to aerosol–cloud interactions (ERFaci) contributes most to the magnitude of the 7 total aerosol ERF (high confidence), and is assessed to be –1.0 [–1.7 to –0.3] W m–2 (medium confidence), 8 with the remainder due to aerosol–radiation interactions (ERFari), assessed to be –0.3 [–0.6 to 0.0] W m–2 9 (medium confidence). There has been an increase in the estimated magnitude – but a reduction in the 10 uncertainty – of the total aerosol ERF relative to AR5, supported by a combination of increased process- 11 understanding and progress in modelling and observational analyses (Figure TS.15c). Effective radiative 12 forcing estimates from these separate lines of evidence are now consistent with each other, in contrast to 13 AR5, and support the assessment that it is virtually certain that the total aerosol ERF is negative. Compared 14 to AR5, the assessed magnitude of ERFaci has increased, while that of ERFari has decreased. {7.3.3, 7.3.5} 15 16 17 [START FIGURE TS.15 HERE] 18 19 Figure TS.15: Contribution to ERF and b) global surface temperature change from component emissions 20 between 1750 to 2019 based on CMIP6 models and c) net aerosol effective radiative forcing 21 (ERF) from different lines of evidence. The intent of the figure is to show advances since AR5 in the 22 understanding of a) aerosol ERF from different lines of evidence as assessed in Chapter 7, b) 23 emissions-based ERF and c) global surface temperature response for SLCFs as estimated in Chapter 24 6. In panel a), ERFs for well-mixed greenhouse gases (WMGHGs) are from the analytical formulae. 25 ERFs for other components are multi-model means based on ESM simulations that quantify the effect 26 of individual components. The derived emission-based ERFs are rescaled to match the concentration- 27 based ERFs in Figure 7.6. Error bars are 5-95% and for the ERF account for uncertainty in radiative 28 efficiencies and multi-model error in the means. In panel b), the global mean temperature response is 29 calculated from the ERF time series using an impulse response function. In panel c), the AR6 30 assessment is based on energy balance constraints, observational evidence from satellite retrievals, 31 and climate model-based evidence. For each line of evidence, the assessed best-estimate contributions 32 from ERF due to ERFari and ERFaci are shown with darker and paler shading, respectively. Estimates 33 from individual CMIP5 and CMIP6 models are depicted by blue and red crosses, respectively. The 34 observational assessment for ERFari is taken from the instantaneous forcing due to aerosol-radiation 35 interactions (IRFari). Uncertainty ranges are given in black bars for the total aerosol ERF and depict 36 very likely ranges. {Sections 7.3.3, 6.4.2, Cross-Chapter Box 7.1, Figures 6.12, 7.5 ; Table 7.8} 37 38 [END FIGURE TS.15 HERE] 39 40 41 TS.3.2 Climate Sensitivity and Earth-System Feedbacks 42 43 TS.3.2.1 Equilibrium Climate Sensitivity, Transient Climate Response, and Transient Climate Response to 44 Cumulative Carbon-dioxide Emissions 45 46 Since AR5, substantial quantitative progress has been made in combining new evidence of Earth’s climate 47 sensitivity, with improvements in the understanding and quantification of Earth’s energy imbalance, the 48 instrumental record of global surface temperature change, paleoclimate change from proxy records, climate 49 feedbacks and their dependence on time scale and climate state. A key advance is the broad agreement across 50 these multiple lines of evidence, supporting a best estimate of equilibrium climate sensitivity (ECS) of 3°C, 51 with a very likely range of 2°C to 5°C. The likely range of 2.5°C to 4°C is narrower than the AR5 likely range 52 of 1.5°C to 4.5°C. {7.4, 7.5} 53 54 Constraints on ECS and TCR (see Glossary) are based on four main lines of evidence: feedback process 55 understanding, climate change and variability seen within the instrumental record, paleoclimate evidence, 56 and so-called ‘emergent constraints’, whereby a relationship between an observable quantity and either ECS Do Not Cite, Quote or Distribute TS-57 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 or TCR established within an ensemble of models is combined with observations to derive a constraint on 2 ECS or TCR. In reports up to and including the IPCC third assessment report, ECS and TCR derived directly 3 from ESMs were the primary line of evidence. However, since AR4, historical warming and paleoclimates 4 provided useful additional evidence (Figure TS.16a). AR6 differs from previous reports in not directly using 5 climate model estimates of ECS and TCR in the assessed ranges of climate sensitivity. {1.5, 7.5} 6 7 It is now clear that when estimating ECS and TCR, the dependence of feedbacks on time scales and the 8 climate state must be accounted for. Feedback processes are expected to become more positive overall (more 9 amplifying of global surface temperature changes) on multi-decadal time scales as the spatial pattern of 10 surface warming evolves and global surface temperature increases, leading to an ECS that is higher than was 11 inferred in AR5 based on warming over the instrumental record (high confidence). Historical surface 12 temperature change since 1870 has shown relatively little warming in several key regions of positive 13 feedbacks, including the eastern equatorial Pacific Ocean and the Southern Ocean, while showing greater 14 warming in key regions of negative feedbacks, including the western Pacific warm pool. Based on process 15 understanding, climate modelling, and paleoclimate reconstructions of past warm periods, it is expected that 16 future warming will become enhanced over the eastern Pacific Ocean (medium confidence) and Southern 17 Ocean (high confidence) on centennial time scales. This new understanding, along with updated estimates of 18 historical temperature change, ERF, and energy imbalance, reconciles previously disparate ECS estimates. 19 {7.4.4, 7.5.2, 7.5.3,} 20 21 The AR6 best estimate of ECS is 3°C, the likely range is 2.5 to 4°C and the very likely range is 2 to 5°C. 22 There is a high level of agreement among the four main lines of evidence listed above (Figure TS.16b), and 23 altogether it is virtually certain that ECS is larger than 1.5°C, but currently it is not possible to rule out ECS 24 values above 5°C. Therefore, the 5°C upper end of the very likely range is assessed with medium confidence 25 and the other bounds with high confidence. {7.5.5} 26 27 Based on process understanding, warming over the instrumental record, and emergent constraints, the best 28 estimate of TCR is 1.8°C, the likely range is 1.4° to 2.2°C and the very likely range is 1.2° to 2.4°C. There is 29 a high level of agreement among the different lines of evidence (Figure TS.16c) (high confidence). {7.5.5} 30 31 On average, CMIP6 models have higher mean ECS and TCR values than the CMIP5 generation of models 32 and also have higher mean values and wider spreads than the assessed best estimates and very likely ranges 33 within this Report. These higher mean ECS and TCR values can, in some models, be traced to changes in 34 extratropical cloud feedbacks (medium confidence). The broader ECS and TCR ranges from CMIP6 also 35 lead the models to project a range of future warming that is wider than the assessed future warming range, 36 which is based on multiple lines of evidence (Cross-Section Box TS.1). However, some of the high- 37 sensitivity CMIP6 models (TS.1.2.2) are less consistent with observed recent changes in global warming and 38 with paleoclimate proxy records than models with ECS within the very likely range. Similarly, some of the 39 low-sensitivity models are less consistent with the paleoclimate data. The CMIP6 models with the highest 40 ECS and TCRs values provide insights into low-likelihood, high-impact futures, which cannot be excluded 41 based on currently available evidence (Cross-Section Box TS.1). {4.3.1, 4.3.4, 7.4.2, 7.5.6} 42 43 Uncertainties regarding the true value of ECS and TCR are the dominant source of uncertainty in global 44 temperature projections over the 21st century under moderate to high GHG concentrations scenarios. For 45 scenarios that reach net zero CO2 emissions (TS.3.3), the uncertainty in the ERF values of aerosol and other 46 SLCFs contribute substantial uncertainty in projected temperature. Global ocean heat uptake is a smaller 47 source of uncertainty in centennial warming. {7.5.7} 48 49 50 [START FIGURE TS.16 HERE] 51 52 Figure TS.16: a) Evolution of equilibrium climate sensitivity (ECS) assessments from the Charney Report 53 through a succession of IPCC Assessment Reports to AR6, and lines of evidence and combined 54 assessment for (b) ECS and (c) transient climate response (TCR) in AR6. In panel (a), the lines of 55 evidence considered are listed below each assessment. Best estimates are marked by horizontal bars, Do Not Cite, Quote or Distribute TS-58 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 likely ranges by vertical bars, and very likely ranges by dotted vertical bars. In panel (b) and (c), 2 assessed ranges are taken from Tables 7.13 and 7.14 for ECS and TCR respectively. Note that for the 3 ECS assessment based on both the instrumental record and paleoclimates, limits (i.e., one-sided 4 distributions) are given, which have twice the probability of being outside the maximum/minimum 5 value at a given end, compared to ranges (i.e., two tailed distributions) which are given for the other 6 lines of evidence. For example, the extremely likely limit of greater than 95% probability corresponds 7 to one side of the very likely (5% to 95%) range. Best estimates are given as either a single number or 8 by a range represented by grey box. CMIP6 ESM values are not directly used as a line of evidence but 9 are presented on the Figure for comparison. {Sections 1.5, 7.5; Tables 7.13, 7.14; Figures 7.18} 10 11 [END FIGURE TS.16 HERE] 12 13 The transient climate response to cumulative CO2 emissions (TCRE) is the ratio between globally averaged 14 surface temperature increase and cumulative CO2 emissions (see Glossary). This report reaffirms with high 15 confidence the finding of AR5 that there is a near-linear relationship between cumulative CO2 emissions and 16 the increase in global average temperature caused by CO2 over the course of this century for global warming 17 levels up to at least 2°C relative to 1850–1900. The TCRE falls likely in the 1.0°C–2.3°C per 1000 PgC 18 range, with a best estimate of 1.65°C per 1000 PgC. This range is about 15% narrower than the 0.8°–2.5°C 19 per 1000 PgC assessment of the AR5 because of a better integration of evidence across chapters, in particular 20 the assessment of TCR. Beyond this century, there is low confidence that the TCRE alone remains an 21 accurate predictor of temperature changes in scenarios of very low or net negative CO2 emissions because of 22 uncertain Earth system feedbacks that can result in further changes in temperature or a path dependency of 23 warming as a function of cumulative CO2 emissions. {5.4, 5.5.1, 4.6.2} 24 25 26 TS.3.2.2 Earth System Feedbacks 27 28 The combined effect of all climate feedback processes is to amplify the climate response to forcing (virtually 29 certain). While major advances in the understanding of cloud processes have increased the level of 30 confidence and decreased the uncertainty range for the cloud feedback by about 50% compared to AR5, 31 clouds remain the largest contribution to overall uncertainty in climate feedbacks (high confidence). 32 Uncertainties in the ECS and other climate sensitivity metrics, such as the transient climate response (TCR) 33 and the transient climate response to cumulative CO2 emissions (TCRE), are the dominant source of 34 uncertainty in global temperature projections over the 21st century under moderate to high GHG emissions 35 scenarios. CMIP6 models have higher mean values and wider spreads in ECS and TCR than the assessed 36 best estimates and very likely ranges within this Report, leading the models to project a range of future 37 warming that is wider than the assessed future warming range (TS.2.2). {7.1, 7.4.2, 7.5} 38 39 Earth system feedbacks can be categorised into three broad groups: physical feedbacks, biogeophysical and 40 biogeochemical feedbacks, and feedbacks associated with ice sheets. In previous assessments, the ECS has 41 been associated with a distinct set of physical feedbacks (Planck response, water vapour, lapse rate, surface 42 albedo, and cloud feedbacks). In this assessment, a more general definition of ECS is adopted whereby all 43 biogeophysical and biogeochemical feedbacks that do not affect the atmospheric concentration of CO2 are 44 included. These include changes in natural methane emissions, natural aerosol emissions, nitrous oxide, 45 ozone, and vegetation, which all act on time scales of years to decades and are therefore relevant for 46 temperature change over the 21st century. Because the total biogeophysical and non-CO2 biogeochemical 47 feedback is assessed to have a central value that is near zero (low confidence), including it does not affect 48 the assessed ECS but does contribute to the net feedback uncertainty. The biogeochemical feedbacks that 49 affect the atmospheric concentration of CO2 are not included because ECS is defined as the response to a 50 sustained doubling of CO2. Moreover, the long-term feedbacks associated with ice sheets are not included in 51 the ECS owing to their long time scales of adjustment. {5.4, 6.4, 7.4, 7.5, Box 7.1} 52 53 The net effect of changes in clouds in response to global warming is to amplify human-induced warming, 54 that is, the net cloud feedback is positive (high confidence). Compared to AR5, major advances in the 55 understanding of cloud processes have increased the level of confidence and decreased the uncertainty range Do Not Cite, Quote or Distribute TS-59 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 in the cloud feedback by about 50% (Figure TS.17a). An assessment of the low-altitude cloud feedback over 2 the subtropical ocean, which was previously the major source of uncertainty in the net cloud feedback, is 3 improved owing to a combined use of climate model simulations, satellite observations, and explicit 4 simulations of clouds, altogether leading to strong evidence that this type of cloud amplifies global warming. 5 The net cloud feedback is assessed to be +0.42 [–0.10 to 0.94] W m–2 °C–1. A net negative cloud feedback is 6 very unlikely. The CMIP5 and CMIP6 ranges of cloud feedback are similar to this assessed range, with 7 CMIP6 having a slightly more positive median cloud feedback (high confidence). The surface albedo 8 feedback and combined water vapour-lapse rate feedback are positive (Figure TS.17a), with high confidence 9 in the estimated value of each based on multiple lines of evidence, including observations, models, and 10 theory. {7.4.2, Figure 7.14, Table 7.10, Box TS.6} 11 12 Natural sources and sinks of non-CO2 greenhouse gases such as methane (CH4) and nitrous oxide (N2O) 13 respond both directly and indirectly to atmospheric CO2 concentration and climate change, and thereby give 14 rise to additional biogeochemical feedbacks in the climate system. Many of these feedbacks are only 15 partially understood and are not yet fully included in ESMs. There is medium confidence that the net 16 response of natural ocean and land CH4 and N2O sources to future warming will be increased emissions, but 17 the magnitude and timing of the responses of each individual process is known with low confidence. {5.4.7} 18 19 Non-CO2 biogeochemical feedbacks induced from changes in emissions, abundances or lifetimes of SLCFs 20 mediated by natural processes or atmospheric chemistry are assessed to decrease ECS (Figure TS.17b). 21 These non-CO2 biogeochemical feedbacks are estimated from ESMs, which since AR5 have advanced to 22 include a consistent representation of biogeochemical cycles and atmospheric chemistry. However, process- 23 level understanding of many biogeochemical feedbacks involving SLCFs, particularly natural emissions, is 24 still emerging, resulting in low confidence in the magnitude and sign of the feedbacks. The central estimate 25 of the total biogeophysical and non-CO2 biogeochemical feedback is assessed to be −0.01 [–0.27 to +0.25] 26 W m–2 °C–1 (Figure TS.17a). {5.4.7, 5.4.8, 6.2.2, 6.4.5, 7.4, Table 7.10} 27 28 The combined effect of all known radiative feedbacks (physical, biogeophysical, and non-CO2 29 biogeochemical) is to amplify the base climate response (in the absence of feedbacks), also known as the 30 Planck temperature response17 (virtually certain). Combining these feedbacks with the Planck response, the 31 net climate feedback parameter is assessed to be –1.16 [–1.81 to –0.51] W m–2 °C–1, which is slightly less 32 negative than that inferred from the overall ECS assessment. The combined water vapour and lapse rate 33 feedback makes the largest single contribution to global warming, whereas the cloud feedback remains the 34 largest contribution to overall uncertainty. Due to the state-dependence of feedbacks, as evidenced from 35 paleoclimate observations and from models, the net feedback parameter will increase (become less negative) 36 as global temperature increases. Furthermore, on long time scales the ice sheet feedback parameter is very 37 likely positive, promoting additional warming on millennial timescales as ice sheets come into equilibrium 38 with the forcing. (high confidence) {7.4.2, 7.4.3, Figure 7.14, Table 7.10} 39 40 The carbon cycle provides for additional feedbacks on climate owing to the sensitivity of land–atmosphere 41 and ocean–atmosphere carbon fluxes and storage to changes in climate and in atmospheric CO2 (Figure 42 TS.17c). Because of the time scales associated with land and ocean carbon uptake, these feedbacks are 43 known to be scenario dependent. Feedback estimates deviate from linearity in scenarios of stabilizing or 44 reducing concentrations. With high confidence, increased atmospheric CO2 will lead to increased land and 45 ocean carbon uptake, acting as a negative feedback on climate change. It is likely that a warmer climate will 46 lead to reduced land and ocean carbon uptake, acting as a positive feedback. {4.3.2, 5.4.1-5, Box TS.4} 47 48 Thawing terrestrial permafrost will lead to carbon release (high confidence), but there is low confidence in 49 the timing, magnitude and the relative roles of CO2 versus CH4 as feedback processes. An ensemble of 50 models projects CO2 release from permafrost to be 3–41 PgC per 1ºC of global warming by 2100, leading to 51 warming strong enough that it must be included in estimates of the remaining carbon budget but weaker than 52 the warming from fossil fuel burning. However, the incomplete representation of important processes, such 17 For reference, the Planck temperature response for a doubling of atmospheric CO 2 is approximately 1.2°C at equilibrium. Do Not Cite, Quote or Distribute TS-60 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 as abrupt thaw, combined with weak observational constraints, only allow low confidence in both the 2 magnitude of these estimates and in how linearly proportional this feedback is to the amount of global 3 warming. There is emerging evidence that permafrost thaw and thermokarst give rise to increased CH4 and 4 N2O emissions, which leads to the combined radiative forcing from permafrost thaw being larger than from 5 CO2 emissions only. However, the quantitative understanding of these additional feedbacks is low, 6 particularly for N2O. These feedbacks, as well as potential additional carbon losses due to climate-induced 7 fire feedback are not routinely included in Earth System models. {Box 5.1, 5.4.3, 5.4.7, 5.4.8, Box TS.9} 8 9 10 [START FIGURE TS.17 HERE] 11 12 Figure TS.17: An overview of physical and biogeochemical feedbacks in the climate system. The intent is to 13 summarize assessed estimates of physical, biogeophysical and biogeochemical feedbacks on global 14 temperature based on Chapters 5, 6 and 7. (a) Synthesis of physical, biogeophysical and non-CO2 15 biogeochemical feedbacks that are included in the definition of ECS assessed in this Technical 16 Summary. These feedbacks have been assessed using multiple lines of evidence including 17 observations, models and theory. The net feedback is the sum of the Planck response, water vapour 18 and lapse rate, surface albedo, cloud, and biogeophysical and non-CO2 biogeochemical feedbacks. 19 Bars denote the mean feedback values and uncertainties represent very likely ranges; (b) Estimated 20 values of individual biogeophysical and non-CO2 biogeochemical feedbacks. The atmospheric 21 methane lifetime and other non-CO2 biogeochemical feedbacks have been calculated using global 22 Earth System Model simulations from AerChemMIP, while the CH 4 and N2O source responses to 23 climate have been assessed for the year 2100 using a range of modelling approaches using simplified 24 radiative forcing equations. The estimates represent the mean and 5-95% range. The level of 25 confidence in these estimates is low owing to the large model spread. (c) carbon-cycle feedbacks as 26 simulated by models participating in the C4MIP of CMIP6. An independent estimate of the additional 27 positive carbon-cycle climate feedbacks from permafrost thaw, which is not considered in most 28 C4MIP models, is added. The estimates represent the mean and 5-95% range. Note that these 29 feedbacks act through modifying the atmospheric concentration of CO 2 and thus are not included in 30 the definition of ECS, which assumes a doubling of CO 2 but are included in the definition and 31 assessed range of TCRE. {Sections Box 5.1, 5.4.7, 5.4.8, 6.4.5, 7.4.2, Figure 5.29, Tables 6.9, 7.10} 32 33 [END FIGURE TS.17 HERE] 34 35 36 TS.3.3 Temperature Stabilization, Net Zero Emissions and Mitigation 37 38 TS.3.3.1 Remaining Carbon Budgets and Temperature Stabilization 39 40 The near-linear relationship between cumulative CO2 emissions and maximum global surface temperature 41 increase caused by CO2 implies that stabilizing human-induced global temperature increase at any level 42 requires net anthropogenic CO2 emissions to become zero. This near-linear relationship further implies that 43 mitigation requirements for limiting warming to specific levels can be quantified in terms of a carbon budget 44 (high confidence). Remaining carbon budget estimates have been updated since the AR5 with 45 methodological improvements, resulting in larger estimates that are consistent with SR1.5. Several factors, 46 including estimates of historical warming, future emissions from thawing permafrost, variations in projected 47 non-CO2 warming, and the global surface temperature change after cessation of CO2 emissions, affect the 48 exact value of carbon budgets (high confidence). {1.3.5, Box 1.2, 4.7.1, 5.5} 49 50 Limiting further climate change would require substantial and sustained reductions of GHG emissions. 51 Without net zero CO2 emissions, and a decrease in the net non-CO2 forcing (or sufficient net negative CO2 52 emissions to offset any further warming from net non-CO2 forcing), the climate system will continue to 53 warm. There is high confidence that mitigation requirements for limiting warming to specific levels over this 54 century can be estimated using a carbon budget that relates cumulative CO2 emissions to global mean 55 temperature increase (Figure TS.18, Table TS.3). For the period 1850–2019, a total of 2390 ± 240 GtCO2 of 56 anthropogenic CO2 has been emitted. Remaining carbon budgets (starting from 1 January 2020) for limiting Do Not Cite, Quote or Distribute TS-61 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 warming to 1.5°C, 1.7°C, and 2.0°C are estimated at 500 GtCO2, 850 GtCO2 and 1350 GtCO2, respectively, 2 based on the 50th percentile of TCRE. For the 67th percentile, the respective values are 400 GtCO2, 700 3 GtCO2 and 1150 GtCO2. The remaining carbon budget estimates for different temperature limits assume that 4 non-CO2 emissions are mitigated consistent with the median reductions found in scenarios in the literature as 5 assessed in SR1.5, but they may vary by an estimated ±220 GtCO2 depending on how deeply future non-CO2 6 emissions are assumed to be reduced (Table TS.3). {5.5.2, 5.6, Box 5.2, 7.6} 7 8 9 [START FIGURE TS.18 HERE] 10 11 Figure TS.18: Illustration of (a) relationship between cumulative emissions of CO 2 and global mean surface air 12 temperature increase and (b) the assessment of the remaining carbon budget from its 13 constituting components based on multiple lines of evidence. Carbon budgets consistent with 14 various levels of additional warming are provided in Table 5.8 and should not be read from the 15 illustrations in either panel. In panel (a) thin black line shows historical CO 2 emissions together with 16 the assessed global surface temperature increase from 1850-1900 as assessed in Chapter 2 (Box 2.3). 17 The orange-brown range with its central line shows the estimated human-induced share of historical 18 warming. The vertical orange-brown line shows the assessed range of historical human-induced 19 warming for the 2010–2019 period relative to 1850-1900 (Chapter 3). The grey cone shows the 20 assessed likely range for the transient climate response to cumulative emissions of carbon dioxide 21 (TCRE) (Section 5.5.1.4), starting from 2015. Thin coloured lines show CMIP6 simulations for the 22 five scenarios of the WGI core set (SSP1-1.9, green; SSP1-2.6, blue; SSP2-4.5, yellow; SSP3-7.0, red; 23 SSP5-8.5, maroon), starting from 2015 and until 2100. Diagnosed carbon emissions are 24 complemented with estimated land-use change emissions for each respective scenario. Coloured areas 25 show the Chapter 4 assessed very likely range of global surface temperature projections and thick 26 coloured central lines the median estimate, for each respective scenario, relative to the original 27 scenario emissions. For panel (b), the remaining allowable warming is estimated by combining the 28 global warming limit of interest with the assessed historical human induced warming (Section 29 5.5.2.2.2), the assessed future potential non-CO2 warming contribution (Section 5.5.2.2.3) and the 30 ZEC (Section 5.5.2.2.4). The remaining allowable warming (vertical blue bar) is subsequently 31 combined with the assessed TCRE (Section 5.5.1.4 and 5.5.2.2.1) and contribution of unrepresented 32 Earth system feedbacks (Section 5.5.2.2.5) to provide an assessed estimate of the remaining carbon 33 budget (horizontal blue bar, Table 5.8). Note that contributions in panel (b) are illustrative and are not 34 to scale. For example, the central ZEC estimate was assessed to be zero. {Box 2.3; Sections 5.2.1, 35 5.2.2; Figure 5.31} 36 37 [END FIGURE TS.18 HERE] 38 39 40 [START TABLE TS.3 HERE] 41 42 Table TS.3: Estimates of remaining carbon budgets and their uncertainties. Assessed estimates are provided 43 for additional human-induced warming, expressed as global surface temperature, since the recent past 44 (2010–2019), which likely amounted to 0.8° to 1.3°C with a best estimate of 1.07°C relative to 1850– 45 1900. Historical CO2 emissions between 1850 and 2014 have been estimated at about 2180 ± 240 46 GtCO2 (1-sigma range), while since 1 January 2015, an additional 210 GtCO 2 has been emitted until 47 the end of 2019. GtCO2 values to the nearest 50. {Table 5.8, Table 3.1, Table 5.1, Table 5.7, 5.5.1, 48 5.5.2, Box 5.2} 49 50 51 52 53 54 55 56 57 58 Do Not Cite, Quote or Distribute TS-62 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI Global Global Estimated remaining carbon budgets Scenario Geophysical uncertainties*(4) surface surface starting from 1 January 2020 and subject to variations and variation temperat temperat uncertainties quantified in the columns on the right ure ure change change since since 2010– 1850– 2019 1900 *(1) °C °C Percentiles of TCRE*(2) Non-CO2 Non-CO2 Historical ZEC Recent GtCO2 scenario forcing and temperature uncertai emissions variation *(3) response uncertainty*( nty uncertainty uncertainty 1) *(5) 17th 33rd 50th 67th 83rd GtCO2 GtCO2 GtCO2 GtCO2 GtCO2 0.43 1.5 900 650 500 400 300 Values can Values can 0.53 1.6 1200 850 650 550 400 vary by at vary by at least least ±220 due to 0.63 1.7 1450 1050 850 700 550 ±220 due to uncertainty in 0.73 1.8 1750 1250 1000 850 650 choices the warming ±550 ±420 ±20 0.83 1.9 2000 1450 1200 1000 800 related to non- response to CO2 future 0.93 2 2300 1700 1350 1150 900 emissions non-CO2 mitigation emissions *(1) Human-induced global surface temperature increase between 1850–1900 and 2010–2019 is assessed at 0.8–1.3°C (likely range; Cross-Section Box TS.1) with a best estimate of 1.07°C. Combined with a central estimate of TCRE (1.65 °C EgC -1) this uncertainty in isolation results in a potential variation of remaining carbon budgets of ±550 GtCO2, which, however, is not independent of the assessed uncertainty of TCRE and thus not fully additional. *(2) TCRE: transient climate response to cumulative emissions of carbon dioxide, assessed to fall likely between 1.0–2.3°C EgC-1 with a normal distribution, from which the percentiles are taken. Additional Earth system feedbacks are included in the remaining carbon budget estimates as discussed in Section 5.5.2.2.5. *(3) Estimates assume that non-CO2 emissions are mitigated consistent with the median reductions found in scenarios in the literature as assessed in SR1.5. Non- CO2 scenario variations indicate how much remaining carbon budget estimates vary due to different scenario assumptions related to the future evolution of non-CO2 emissions in mitigation scenarios from SR1.5 that reach net zero CO2 emissions. This variation is additional to the uncertainty in TCRE. The WGIII Contribution to AR6 will reassess the potential for non-CO2 mitigation based on literature since the SR1.5. *(4) Geophysical uncertainties reported in these columns and TCRE uncertainty are not statistically independent, as uncertainty in TCRE depends on uncertainty in the assessment of historical temperature, non-CO2 versus CO2 forcing and uncertainty in emissions estimates. These estimates cannot be formally combined and these uncertainty variations are not directly additional to the spread of remaining carbon budgets due to TCRE uncertainty reported in columns 3 to 7. *(5) Recent emissions uncertainty reflects the ±10% uncertainty in the historical CO 2 emissions estimate since 1 January 2015. 1 2 [END TABLE TS.3 HERE] 3 4 5 There is high confidence that several factors, including estimates of historical warming, future emissions 6 from thawing permafrost, and variations in projected non-CO2 warming, affect the value of carbon budgets 7 but do not change the conclusion that global CO2 emissions would need to decline to net zero to halt global 8 warming. Estimates may vary by ±220 GtCO2 depending on the level of non-CO2 emissions at the time 9 global anthropogenic CO2 emissions reach net zero levels. This variation is referred to as non-CO2 scenario 10 uncertainty and will be further assessed in the AR6 Working Group III Contribution. Geophysical 11 uncertainties surrounding the climate response to these non-CO2 emissions result in an additional uncertainty 12 of at least ±220 GtCO2, and uncertainties in the level of historical warming result in a ±550 GtCO2 13 uncertainty. {5.4, 5.5.2} 14 15 Methodological improvements and new evidence result in updated remaining carbon budget estimates. The 16 assessment in AR6 applies the same methodological improvements as in SR1.5, which uses a recent 17 observed baseline for historic temperature change and cumulative emissions. Changes compared to SR1.5 18 are therefore small: the assessment of new evidence results in updated median remaining carbon budget 19 estimates for limiting warming to 1.5°C and 2°C being the same and about 60 GtCO2 smaller, respectively, 20 after accounting for emissions since SR1.5. Meanwhile, remaining carbon budgets for limiting warming to 21 1.5°C would be about 300–350 GtCO2 larger if evidence and methods available at the time of the AR5 22 would be used. If a specific remaining carbon budget is exceeded, this results in a lower probability of 23 keeping warming below a specified temperature level and higher irreversible global warming over decades to 24 centuries, or alternatively a need for net negative CO2 emissions or further reductions in non-CO2 25 greenhouse gases after net zero CO2 is achieved to return warming to lower levels in the long term. {5.5.2, 26 5.6, Box 5.2} 27 28 Based on idealized model simulations that explore the climate response once CO2 emissions have been 29 brought to zero, the magnitude of the zero CO2 emissions commitment (ZEC, see Glossary) is assessed to be Do Not Cite, Quote or Distribute TS-63 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 likely smaller than 0.3°C for time scales of about half a century and cumulative CO2 emissions broadly 2 consistent with global warming of 2°C. However, there is low confidence about its sign on time scales of 3 about half a century. For lower cumulative CO2 emissions, the range would be smaller yet with equal 4 uncertainty about the sign. If the ZEC is positive on decadal time scales, additional warming leads to a 5 reduction in the estimates of remaining carbon budgets, and vice versa if it is negative. {4.7.1, 5.5.2} 6 7 Permafrost thaw is included in estimates together with other feedbacks that are often not captured by models. 8 Limitations in modelling studies combined with weak observational constraints only allow low confidence in 9 the magnitude of these estimates (TS.3.2.2). Despite the large uncertainties surrounding the quantification of 10 the effect of additional Earth system feedback processes, such as emissions from wetlands and permafrost 11 thaw, these feedbacks represent identified additional risk factors that scale with additional warming and 12 mostly increase the challenge of limiting warming to specific temperature levels. These uncertainties do not 13 change the basic conclusion that global CO2 emissions would need to decline to net zero to halt global 14 warming. {Box 5.1, 5.4.8, 5.5.2} 15 TS.3.3.2 Carbon Dioxide Removal 16 17 Deliberate carbon dioxide removal (CDR) from the atmosphere has the potential to compensate for residual 18 CO2 emissions to reach net zero CO2 emissions or to generate net negative CO2 emissions. In the same way 19 that part of current anthropogenic net CO2 emissions are taken up by land and ocean carbon stores, net CO2 20 removal will be partially counteracted by CO2 release from these stores (very high confidence). Asymmetry 21 in the carbon cycle response to simultaneous CO2 emissions and removals implies that a larger amount of 22 CO2 would need to be removed to compensate for an emission of a given magnitude to attain the same 23 change in atmospheric CO2 (medium confidence). CDR methods have wide-ranging side-effects that can 24 either weaken or strengthen the carbon sequestration and cooling potential of these methods and affect the 25 achievement of sustainable development goals (high confidence). {4.6.3, 5.6} 26 27 Carbon dioxide removal (CDR) refers to anthropogenic activities that deliberately remove CO2 from the 28 atmosphere and durably store it in geological, terrestrial or ocean reservoirs, or in products. Carbon dioxide 29 is removed from the atmosphere by enhancing biological or geochemical carbon sinks or by direct capture of 30 CO2 from air. Emission pathways that limit global warming to 1.5°C or 2°C typically assume the use of 31 CDR approaches in combination with GHG emissions reductions. CDR approaches could be used to 32 compensate for residual emissions from sectors that are difficult or costly to decarbonize. CDR could also be 33 implemented at a large scale to generate global net negative CO2 emissions (i.e., anthropogenic CO2 34 removals exceeding anthropogenic emissions), which could compensate for earlier emissions as a way to 35 meet long-term climate stabilization goals after a temperature overshoot. This Report assesses the effects of 36 CDR on the carbon cycle and climate. Co-benefits and trade-offs for biodiversity, water and food production 37 are briefly discussed for completeness, but a comprehensive assessment of the ecological and socio- 38 economic dimensions of CDR options is left to the WGII and WGIII reports. {4.6.3, 5.6} 39 40 CDR methods have the potential to sequester CO2 from the atmosphere (high confidence). In the same way 41 part of current anthropogenic net CO2 emissions are taken up by land and ocean carbon stores, net CO2 42 removal will be partially counteracted by CO2 release from these stores, such that the amount of CO2 43 sequestered by CDR will not result in an equivalent drop in atmospheric CO2 (very high confidence). The 44 fraction of CO2 removed from the atmosphere that is not replaced by CO2 released from carbon stores, a 45 measure of CDR effectiveness, decreases slightly with increasing amounts of removal (medium confidence) 46 and decreases strongly if CDR is applied at lower atmospheric CO2 concentrations (medium confidence). The 47 reduction in global surface temperature is approximately linearly related to cumulative CO2 removal (high 48 confidence). Because of this near-linear relationship, the amount of cooling per unit CO2 removed is 49 approximately independent of the rate and amount of removal (medium confidence). {4.6.3, 5.6.2.1, Figure 50 5.32, Figure 5.34} 51 52 Due to non-linearities in the climate system, the century-scale climate–carbon cycle response to a CO2 53 removal from the atmosphere is not always equal and opposite to its response to a simultaneous CO2 54 emission (medium confidence). For CO2 emissions of 100 PgC released from a pre-industrial from a state in Do Not Cite, Quote or Distribute TS-64 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 equilibrium with pre-industrial atmospheric CO2 levels, CMIP6 models simulate that 27± 6% (mean ± 1 2 standard deviation) of emissions remain in the atmosphere 80–100 years after the emissions, whereas for 3 removals of 100 PgC only 23 ± 6% of removals remain out of the atmosphere. This asymmetry implies that 4 an extra amount of CDR is required to compensate for a positive emission of a given magnitude to attain the 5 same change in atmospheric CO2. Due to limited agreement between models, there is low confidence in the 6 sign of the asymmetry of the temperature response to CO2 emissions and removals. {4.6.3, 5.6.2.1, Figure 7 5.35} 8 9 Simulations with ESMs indicate that under scenarios where CO2 emissions gradually decline, reach net zero 10 and become net negative during the 21st century (e.g., SSP1–2.6), land and ocean carbon sinks begin to 11 weaken in response to declining atmospheric CO2 concentrations, and the land sink eventually turns into a 12 source (Figure TS.19). This sink-to-source transition occurs decades to a few centuries after CO2 emissions 13 become net negative. The ocean remains a sink of CO2 for centuries after emissions become net negative. 14 Under scenarios with large net negative CO2 emissions (e.g., SSP5–3.4–OS) and rapidly declining CO2 15 concentrations, the land source is larger than for SSP1–2.6 and the ocean also switches to a source. While the 16 general response is robust across models, there is low confidence in the timing of the sink-to-source 17 transition and the magnitude of the CO2 source in scenarios with net negative CO2 emissions. Carbon dioxide 18 removal could reverse some aspects climate change if CO2 emissions become net negative, but some changes 19 would continue in their current direction for decades to millennia. For instance, sea level rise due to ocean 20 thermal expansion would not reverse for several centuries to millennia (high confidence) (Box TS.4). {4.6.3, 21 5.4.10, 5.6.2.1, Figure 5.30, Figure 5.33} 22 23 24 [START FIGURE TS.19 HERE] 25 26 Figure TS.19: Carbon sink response in a scenario with net CO2 removal from the atmosphere. The intent of this 27 figure is to show how atmospheric CO2 evolves under negative emissions and its dependence on the 28 negative emissions technologies. It also shows the evolution of the ocean and land sinks. Shown are 29 CO2 flux components from concentration-driven Earth system model simulations during different 30 emission stages of SSP1–2.6 and its long-term extension. (a) Large net positive CO2 emissions, (b) 31 small net positive CO2 emissions, (c) – (d) net negative CO2 emissions, (e) net zero CO2 emissions. 32 Positive flux components act to raise the atmospheric CO 2 concentration, whereas negative 33 components act to lower the CO2 concentration. Net CO2 emissions, land and ocean CO2 fluxes 34 represent the multi-model mean and standard deviation (error bar) of four ESMs (CanESM5, 35 UKESM1, CESM2-WACCM, IPSL-CM6a-LR) and one EMIC (UVic ESCM). Net CO2 emissions are 36 calculated from concentration-driven Earth system model simulations as the residual from the rate of 37 increase in atmospheric CO2 and land and ocean CO2 fluxes. Fluxes are accumulated over each 50- 38 year period and converted to concentration units (ppm). {5.6.2.1, Figure 5.33} 39 40 [END FIGURE TS.19 HERE] 41 42 43 CDR methods have a range of side effects that can either weaken or strengthen the carbon sequestration and 44 cooling potential of these methods and affect the achievement of sustainable development goals (high 45 confidence). Biophysical and biogeochemical side-effects of CDR methods are associated with changes in 46 surface albedo, the water cycle, emissions of CH4 and N2O, ocean acidification and marine ecosystem 47 productivity (high confidence). These side-effects and associated Earth system feedbacks can decrease 48 carbon uptake and/or change local and regional climate and in turn limit the CO2 sequestration and cooling 49 potential of specific CDR methods (medium confidence). Deployment of CDR, particularly on land, can also 50 affect water quality and quantity, food production and biodiversity (high confidence). These effects are often 51 highly dependent on local context, management regime, prior land use, and scale (high confidence). The 52 largest co-benefits are obtained with methods that seek to restore natural ecosystems or improve soil carbon 53 sequestration (medium confidence). The climate and biogeochemical effects of terminating CDR are 54 expected to be small for most CDR methods (medium confidence). {4.6.3, 5.6.2.2, 8.4.3, 8.6.3, Figure 5.36} Do Not Cite, Quote or Distribute TS-65 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.3.3.3 Relating Different Forcing Agents 2 3 When including other GHGs, the choice of emission metric affects the quantification of net zero GHG 4 emissions and their resulting temperature outcome (high confidence). Reaching and sustaining net zero GHG 5 emissions typically leads to a peak and decline in temperatures when quantified with the global warming 6 potential over a 100-year period (GWP-100). Carbon-cycle responses are more robustly accounted for in 7 emission metrics compared to AR5 (high confidence). New emission metric approaches can be used to 8 generate equivalent cumulative emissions of CO2 for short-lived greenhouse gases based on their rate of 9 emissions. {7.6.2} 10 11 Over 10- to 20-year time scales, the temperature response to a single year’s worth of current emissions of 12 SLCFs is at least as large as that of CO2, but because the effect of SLCFs decays rapidly over the first few 13 decades after emission, the net long-term temperature response to a single year’s worth of emissions is 14 predominantly determined by cumulative CO2 emissions. Emission reductions in 2020 associated with 15 COVID-19 containment led to small and positive global ERF; however, global and regional climate 16 responses to the forcing are undetectable above internal variability due to the temporary nature of emission 17 reductions. {6.6, Cross-Chapter Box 6.1} 18 19 The relative climate effects of different forcing agents are typically quantified using emission metrics that 20 compare the effects of an idealised pulse of 1 kg of some climate forcing agent against a reference climate 21 forcing agent, almost always CO2. The two most prominent pulse emissions metrics are the global warming 22 potential (GWP) and global temperature change potential (GTP) (see Glossary). The climate responses to 23 CO2 emissions by convention include the effects of warming on the carbon cycle, so for consistency these 24 need to be determined also for non-CO2 emissions. The methodology for doing this has been placed on a 25 more robust scientific footing compared to AR5 (high confidence). Methane from fossil fuel sources has 26 slightly higher emission metric values than those from biogenic sources since it leads to additional fossil CO2 27 in the atmosphere (high confidence). Updates to the chemical adjustments for methane and nitrous oxide 28 emissions (Section TS.3.1) and revisions in their lifetimes result in emission metrics for GWP and GTP that 29 are slightly lower than in AR5 (medium confidence). Emission metrics for the entire suite of GHGs assessed 30 in the AR6 have been calculated for various time horizons. {7.6.1, Table 7.15, Table 7.SM.7} 31 32 New emission metric approaches, such as GWP* and Combined-GTP (CGTP), relate changes in the 33 emission rate of short-lived greenhouse gases to equivalent cumulative emissions of CO2 (cumulative CO2-e 34 emissions). Global surface temperature response from aggregated emissions of short-lived greenhouse gases 35 over time is determined by multiplying these cumulative CO2-e by TCRE (see TS.3.2.1). When GHGs are 36 aggregated using standard metrics such as GWP or GTP, cumulative CO2-e emissions are not necessarily 37 proportional to future global surface temperature outcomes (high confidence) {Box 7.3, 7.6.1} 38 39 Emission metrics are needed to aggregate baskets of gases to determine net zero GHG emissions. Generally, 40 achieving net zero CO2 emissions and declining non-CO2 radiative forcing would halt human-induced 41 warming. Reaching net zero GHG emissions quantified by GWP-100 typically leads to declining 42 temperatures after net zero GHGs emissions are achieved if the basket includes short-lived gases, such as 43 methane. Net zero GHG emissions defined by CGTP or GWP* imply net zero CO2 and other long-lived 44 GHG emissions and constant (CGTP) or gradually declining (GWP*) emissions of short-lived gases. The 45 warming evolution resulting from net zero GHG emissions defined in this way corresponds approximately to 46 reaching net zero CO2 emissions, and would thus not lead to declining temperatures after net zero GHG 47 emissions are achieved but to an approximate temperature stabilization (high confidence). The choice of 48 emission metric hence affects the quantification of net zero GHG emissions, and therefore the resulting 49 temperature outcome of reaching and sustaining net zero GHG emissions levels (high confidence). {7.6.1.4, 50 7.6.2, 7.6.3} 51 52 As pointed out in AR5, ultimately, it is a matter for policymakers to decide which emission metric is most 53 applicable to their needs. This Report does not recommend the use of any specific emission metric as the 54 most appropriate metric depends on the policy goal and context (see Chapter 7, Section 7.6). A detailed Do Not Cite, Quote or Distribute TS-66 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 assessment of GHG metrics to support climate change mitigation and associated policy contexts is provided 2 in the WGIII contribution to the AR6. 3 4 The global surface temperature response following a climate change mitigation measure that affects 5 emissions of both short- and long-lived climate forcers depends on their lifetimes, their ERFs, how fast and 6 for how long the emissions are reduced, and the thermal inertia in the climate system. Mitigation, relying on 7 emission reductions and implemented through new legislation or technology standards, implies that 8 emissions reductions occur year after year. Global temperature response to a year’s worth of current 9 emissions from different sectors informs about the mitigation potential (Figure TS.20). Over 10- to 20-year 10 time scales, the influence of SLCFs is at least as large as that of CO2, with sectors producing the largest 11 warming being fossil fuel production and distribution, agriculture, and waste management. Because the 12 effect of the SLCFs decays rapidly over the first few decades after emission, the net long-term temperature 13 effect from a single year’s worth of current emissions is predominantly determined by CO2. Fossil fuel 14 combustion for energy, industry and land transportation are the largest contributing sectors on a 100-year 15 time scale (high confidence). Current emissions of CO2, N2O and SLCFs from East Asia and North America 16 are the largest regional contributors to additional net future warming on both short (medium confidence) and 17 long time scales (10 and 100 year) (high confidence). {6.6.1, 6.6.2, Figure 6.16} 18 19 COVID-19 restrictions led to detectable reductions in global anthropogenic NOx (about 35% in April 2020) 20 and fossil CO2 (7%, with estimates ranging from 5.8% to 13.0%) emissions, driven largely by reduced 21 emissions from the transportation sector (medium confidence). There is high confidence that, with the 22 exception of surface ozone, reductions in pollutant precursors contributed to temporarily improved air 23 quality in most regions of the world. However, these reductions were lower than that would be expected 24 from sustained implementation of policies addressing air quality and climate change (medium confidence). 25 Overall, the net global ERF from COVID-19 containment was likely small and positive for 2020 (with a 26 temporary peak value less than 0.2 W m–2), thus temporarily adding to the total anthropogenic climate 27 influence, with positive forcing (warming influence) from aerosol changes dominating over negative 28 forcings (cooling influence) from CO2, NOx and contrail cirrus changes. Consistent with this small net 29 radiative forcing, and against a large component of internal variability, Earth system models show no 30 detectable effect on global or regional surface temperature or precipitation (high confidence). {Cross Chapter 31 Box 6.1} 32 33 34 [START FIGURE TS.20 HERE] 35 36 Figure TS.20: Global surface temperature change 10 and 100 years after a one-year pulse of present-day 37 emissions. This figure shows the sectoral contribution to present-day climate change by specific 38 climate forcers including CO2 as well as SLCFs. The temperature response is broken down by 39 individual species and shown for total anthropogenic emissions (top), and sectoral emissions on 10- 40 year (left) and 100-year time scales (right). Sectors are sorted by (high-to-low) net temperature effect 41 on the 10-year time scale. Error bars in the top panel show the 5-95% range in net temperature effect 42 due to uncertainty in radiative forcing only (calculated using a Monte Carlo approach and best 43 estimate uncertainties from the literature). Emissions for 2014 are from the CMIP6 emissions dataset, 44 except for HFCs and aviation H2O which rely on other datasets (see Section 6.6.2 for more details). 45 CO2 emissions are excluded from open biomass burning and residential biofuel use. {6.6.2, Figure 46 6.16} 47 48 [END FIGURE TS.20 HERE] 49 50 51 52 53 54 55 56 Do Not Cite, Quote or Distribute TS-67 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [START BOX TS.7 HERE] 2 3 Box TS.7: Climate and Air Quality Responses to Short-lived Climate Forcers in Shared 4 Socioeconomic Pathways 5 6 Future changes in emissions of short-lived climate forcers (SLCFs) are expected to cause an additional 7 global mean warming with a large diversity in the end-of-century response across the WGI core set of SSPs, 8 depending upon the level of climate change and air pollution mitigation. This additional warming is either 9 due to reductions in cooling aerosols for air pollution regulation or due to increases in methane, ozone and 10 HFCs. This additional warming is stable after 2040 in SSPs associated with lower global air pollution as long 11 as methane emissions are also mitigated, but the overall warming induced by SLCF changes is higher in 12 scenarios in which air quality continues to deteriorate (induced by growing fossil fuel use and limited air 13 pollution control) (high confidence). Sustained methane mitigation reduces global surface ozone, 14 contributing to air quality improvements and also reduces surface temperature in the longer term, but only 15 sustained CO2 emission reductions allow long-term climate stabilization (high confidence). Future changes 16 in air quality (near-surface ozone and particulate matter, or PM) at global and local scales are predominantly 17 driven by changes in ozone and aerosol precursor emissions rather than climate (high confidence). Air 18 quality improvements driven by rapid decarbonization strategies, as in SSP1-1.9 and SSP1-2.6, are not 19 sufficient in the near term to achieve air quality guidelines set by the World Health Organization in some 20 highly polluted regions (high confidence). Additional policies (e.g., access to clean energy, waste 21 management) envisaged to attain United Nations Sustainable Development Goals bring complementary 22 SLCF reduction. {4,4,4, 6.6.3, 6.7.3, BOX TS.7 Figure 1, Box 6.2} 23 24 The net effect of SLCF emissions changes on temperature will depend on how emissions of warming and 25 cooling SLCFs will evolve in the future. The magnitude of the cooling effect of aerosols remains the largest 26 uncertainty in the effect of SLCFs in future climate projections. Since the SLCFs have undergone large 27 changes over the past two decades, the temperature and air pollution responses are estimated relative to year 28 2019 instead of 1995–2014. 29 30 Temperature Response 31 In the next two decades, it is very likely that SLCF emission changes will cause a warming relative to 2019, 32 across the WGI core set of SSPs (see TS.1.3.1), in addition to the warming from long-lived GHGs. The net 33 effect of SLFC and HFC changes in global surface temperature across the SSPs is a likely warming of 34 0.06°C–0.35°C in 2040 relative to 2019. This near-term global mean warming linked to SLCFs is quite 35 similar in magnitude across the SSPs due to competing effects of warming (methane, ozone) and cooling 36 (aerosols) forcers (Box TS.7, Figure 1). There is greater diversity in the end-of-century response among the 37 scenarios. SLCF changes in scenarios with no climate change mitigation (SSP3-7.0 and SSP5-8.5) will cause 38 a warming in the likely range of 0.4°C–0.9°C in 2100 relative to 2019 due to increases in methane, 39 tropospheric ozone and HFC levels. For the stringent climate change and pollution mitigation scenarios 40 (SSP1-1.9 and SSP1-2.6), the cooling from reductions in methane, ozone and HFCs partially balances the 41 warming from reduced aerosols, primarily sulphate, and the overall SLCF effect is a likely increase in global 42 surface temperature of 0.0°C–0.3°C in 2100, relative to 2019. With intermediate climate change and air 43 pollution mitigations, SLCFs in SSP2-4.5 add a likely warming of 0.2°C–0.5°C to global surface temperature 44 change in 2100, with the largest warming resulting from reductions in aerosols. {4.4.4, 6.7.3} 45 46 Assuming implementation and efficient enforcement of both the Kigali Amendment to the Montreal Protocol 47 on Ozone Depleting Substances and current national plans limit emissions (as in SSP1-2.6), the effects of 48 HFCs on global surface temperature, relative to 2019, would remain below +0. 02°C from 2050 onwards 49 versus about +0.04°C–0.08°C in 2050 and +0.1°C–0.3°C in 2100 considering only national HFC regulations 50 decided prior to the Kigali Amendment (as in SSP5-8.5) (medium confidence). {6.6.3, 6.7.3} 51 52 Air Quality Responses 53 Air pollution projections range from strong reductions in global surface ozone and PM (e.g., SSP1-2.6, with 54 stringent mitigation of both air pollution and climate change) to no improvement and even degradation (e.g., Do Not Cite, Quote or Distribute TS-68 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 SSP3-7.0 without climate change mitigation and with only weak air pollution control) (high confidence). 2 Under the SSP3-7.0 scenario, PM levels are projected to increase until 2050 over large parts of Asia, and 3 surface ozone pollution is projected to worsen over all continental areas through 2100 (high confidence). In 4 SSP5-8.5, a scenario without climate change mitigation but with stringent air pollution control, PM levels 5 decline through 2100, but high methane levels hamper the decline in global surface ozone at least until 2080 6 (high confidence). {6.7.1} 7 8 9 [START BOX TS.7, FIGURE 1 HERE] 10 11 Box TS.7, Figure 1: Effects of short-lived climate forcers (SLCFs) on global surface temperature and air 12 pollution across the WGI core set of Shared Socio-Economic Pathways (SSPs). The intent of 13 this figure is to show the climate and air quality (surface ozone and PM 2.5) response to SLCFs in 14 the SSP scenarios for near and long-term. Effects of net aerosols, tropospheric ozone, 15 hydrofluorocarbons (HFCs) (with lifetimes less than 50 years), and methane are compared with 16 those of total anthropogenic forcing for 2040 and 2100 relative to year 2019. The global surface 17 temperature changes are based on historical and future evolution of Effective Radiative Forcing 18 (ERF) as assessed in chapter 7 of this report. The temperature responses to the ERFs are calculated 19 with a common impulse response function (RT) for the climate response, consistent with the metric 20 calculations in Chapter 7 (Box 7.1). The RT has an equilibrium climate sensitivity of 3.0°C for a 21 doubling of atmospheric CO2 concentration (feedback parameter of -1.31 W m-2 °C-1). The 22 scenario total (grey bar) includes all anthropogenic forcings (long- and short-lived climate forcers, 23 and land use changes). Uncertainties are 5-95% ranges. The global changes in air pollutant 24 concentrations (ozone and PM2.5) are based on multimodel CMIP6 simulations and represent 25 changes in 5-year mean surface continental concentrations for 2040 and 2098 relative to 2019. 26 Uncertainty bars represent inter-model ±1 standard deviation. {6.7.2, 6.7.3, Figure 6.24} 27 28 [END BOX TS.7, FIGURE 1 HERE] 29 30 [END BOX TS.7 HERE] 31 32 33 [START BOX TS.8 HERE] 34 35 Box TS.8: Earth System Response to Solar Radiation Modification 36 37 Since AR5, further modelling work has been conducted on aerosol-based solar radiation modification (SRM) 38 options such as stratospheric aerosol injection, marine cloud brightening, and cirrus cloud thinning18 and 39 their climate and biogeochemical effects. These investigations have consistently shown that SRM could 40 offset some of the effects of increasing GHGs on global and regional climate, including the carbon and water 41 cycles (high confidence). However, there would be substantial residual or overcompensating climate change 42 at the regional scales and seasonal time scales (high confidence), and large uncertainties associated with 43 aerosol–cloud–radiation interactions persist. The cooling caused by SRM would increase the global land and 44 ocean CO2 sinks (medium confidence), but this would not stop CO2 from increasing in the atmosphere or 45 affect the resulting ocean acidification under continued anthropogenic emissions (high confidence). It is 46 likely that abrupt water cycle changes will occur if SRM techniques are implemented rapidly. A sudden and 47 sustained termination of SRM in a high CO2 emissions scenario would cause rapid climate change (high 48 confidence). However, a gradual phase-out of SRM combined with emission reduction and CDR would 49 avoid these termination effects (medium confidence). {4.6.3, 5.6.3. 6.4.6, 8.6.3}. 50 51 Solar radiation modification (SRM) refers to deliberate large-scale climate intervention options that are 52 studied as potential supplements to deep mitigation, for example, in scenarios that overshoot climate 53 stabilization goals. SRM options aim to offset some of the warming effects of GHG emissions by 18 Although cirrus cloud thinning aims to cool the planet by increasing longwave emissions to space, it is included in the portfolio of SRM options for consistency with AR5 and SR1.5. {4.6.3.3} Do Not Cite, Quote or Distribute TS-69 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 modification of the Earth’s shortwave radiation budget. Following SR1.5, the SRM assessed in this report 2 also includes some options, such as cirrus cloud thinning, that alter the longwave radiation budget. SRM 3 contrasts with climate change mitigation activities, such as emission reductions and CDR, as it introduces a 4 ‘mask’ to the climate change problem by altering the Earth’s radiation budget, rather than attempting to 5 address the root cause of the problem, which is the increase in GHGs in the atmosphere. By masking only 6 the climate effects of GHG emissions, SRM does not address other issues related to atmospheric CO2 7 increase, such as ocean acidification. The WGI report assesses physical understanding of the Earth system 8 response to proposed SRM and is based primarily on idealized climate model simulations. There are 9 important other considerations, such as risk to human and natural systems, perceptions, ethics, cost, 10 governance, and trans-boundary issues and their relationship to the United Nations Sustainable Development 11 Goals—issues that the WGII (Chapter 16) and WGIII (Chapter 14) reports address. {4.6.3} 12 13 SRM options include those that increase surface albedo, brighten marine clouds by increasing the amount of 14 cloud condensation nuclei, or reduce the optical depth of cirrus clouds by seeding them with ice nucleating 15 particles. However, the most commonly studied approaches attempt to mimic the cooling effects of major 16 volcanic eruptions by injecting reflective aerosols (e.g., sulphates) or their precursors (e.g., sulphur dioxide) 17 into the stratosphere. {4.6.3, 5.6.3, 6.4.6} 18 19 SRM could offset some effects of greenhouse gas-induced warming on global and regional climate, but there 20 would be substantial residual and overcompensating climate change at the regional scale and seasonal time 21 scales (high confidence). Since the AR5, more modelling work has been conducted with more sophisticated 22 treatment of aerosol-based SRM approaches, but the uncertainties in cloud–aerosol–radiation interactions are 23 still large (high confidence). Modelling studies suggest that it is possible to stabilize multiple large-scale 24 temperature indicators simultaneously by tailoring the deployment strategy of SRM options (medium 25 confidence) but with large residual or overcompensating regional and seasonal climate changes. {4.6.3} 26 27 SRM approaches targeting shortwave radiation are likely to reduce global mean precipitation, relative to 28 future CO2 emissions scenarios, if all global mean warming is offset. In contrast, cirrus cloud thinning, 29 targeting longwave radiation, is expected to cause an increase in global mean precipitation (medium 30 confidence). If shortwave approaches are used to offset global mean warming, the magnitude of reduction in 31 regional precipitation minus evapotranspiration (P-E) (Box TS.5), which is more relevant to freshwater 32 availability, is smaller than precipitation decrease because of simultaneous reductions in both precipitation 33 and evapotranspiration (medium confidence). {4.6.3, 8.2.1, 8.6.3}. 34 35 If SRM is used to cool the planet, it would cause a reduction in plant and soil respiration and slow the 36 reduction of ocean carbon uptake due to warming (medium confidence). The result would be an enhancement 37 of the global land and ocean CO2 sinks (medium confidence) and a slight reduction in atmospheric CO2 38 concentration relative to unmitigated climate change. However, SRM would not stop CO2 from increasing in 39 the atmosphere or affect the resulting ocean acidification under continued anthropogenic emissions (high 40 confidence). {5.6.3} 41 42 The effect of stratospheric aerosol injection on global temperature and precipitation is projected by models to 43 be detectable after one to two decades, which is similar to the time scale for the emergence of the benefits of 44 emissions reductions. A sudden and sustained termination of SRM in a high GHG emissions scenario would 45 cause rapid climate change and a reversal of the SRM effects on the carbon sinks (high confidence). It is also 46 likely that a termination of strong SRM would drive abrupt changes in the water cycle globally and 47 regionally, especially in the tropical regions by shifting the ITCZ and Hadley cells. At the regional scale, 48 non-linear responses cannot be excluded, due to changes in evapotranspiration. However, a gradual phase- 49 out of SRM combined with emissions reductions and CDR would avoid larger rates of changes (medium 50 confidence). {4.6.3, 5.6.3, 8.6.3}. 51 52 [END BOX TS.8 HERE] 53 54 55 Do Not Cite, Quote or Distribute TS-70 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [START BOX TS.9 HERE] 2 3 Box TS.9: Irreversibility, Tipping Points and Abrupt Changes 4 5 The present rate of response of many aspects of the climate system are proportionate to the rate of recent 6 temperature change, but some aspects may respond disproportionately. Some climate system components are 7 slow to respond, such as the deep ocean overturning circulation and the ice sheets. It is virtually certain that 8 irreversible, committed change is already underway for the slow-to-respond processes as they come into 9 adjustment for past and present emissions. The paleoclimate record indicates that tipping elements exist in 10 the climate system where processes undergo sudden shifts toward a different sensitivity to forcing, such as 11 during a major deglaciation, where one degree of temperature change might correspond to a large or small 12 ice sheet mass loss during different stages. For global climate indicators, evidence for abrupt change is 13 limited, but deep ocean warming, acidification and sea level rise are committed to ongoing change for 14 millennia after global surface temperatures initially stabilize and are irreversible on human time scales (very 15 high confidence). At the regional scale, abrupt responses, tipping points and even reversals in the direction of 16 change cannot be excluded (high confidence). Some regional abrupt changes and tipping points could have 17 severe local impacts, such as unprecedented weather, extreme temperatures and increased frequency of 18 droughts and forest fires. Models that exhibit such tipping points are characterised by abrupt changes once 19 the threshold is crossed, and even a return to pre-threshold surface temperatures or to atmospheric carbon 20 dioxide concentrations, does not guarantee that the tipping elements return to their pre-threshold state. 21 Monitoring and early warning systems are being put into place to observe tipping elements in the climate 22 system. {Box TS.2, Box TS.4, TS 3.3.2, 1.3, 1.4.4, 1.5, 4.3.2, Table 4.10, 5.3.4, 5.4.9, 7.5.3, 9.2.2, 9.2.4, 23 9.4.1, 9.4.2, 9.6.3, Cross-chapter Box 12.1} 24 25 Understanding of multi-decadal reversibility (i.e., the system returns to the previous climate state within 26 multiple decades after the radiative forcing is removed) has improved since AR5 for many atmospheric, land 27 surface and sea ice climate metrics following sea surface temperature recovery. Some processes suspected of 28 having tipping points, such as the Atlantic Meridional Overturning Circulation (AMOC), have been found to 29 often undergo recovery after temperature stabilization with a time delay (low confidence). However, 30 substantial irreversibility is further substantiated for some cryosphere changes, ocean warming, sea level rise, 31 and ocean acidification. {4.7.2, 5.3.3, 5.4.9, 9.2.2, 9.2.4, 9.4.1, 9.4.2, 9.6.3} 32 33 Some climate system components are slow to respond, such as the deep ocean overturning circulation and 34 the ice sheets. It is likely that under stabilization of global warming at 1.5°C, 2.0°C, or 3.0°C relative to 35 1850–1900, the AMOC will continue to weaken for several decades by about 15%, 20% and 30% of its 36 strength and then recover to pre-decline values over several centuries (medium confidence). At sustained 37 warming levels between 2°C and 3°C, there is limited evidence that the Greenland and West Antarctic Ice 38 Sheets will be lost almost completely and irreversibly over multiple millennia; both the probability of their 39 complete loss and the rate of mass loss increases with higher surface temperatures (high confidence). At 40 sustained warming levels between 3°C and 5°C, near-complete loss of the Greenland Ice Sheet and complete 41 loss of the West Antarctic Ice Sheet is projected to occur irreversibly over multiple millennia (medium 42 confidence); with substantial parts or all of Wilkes Subglacial Basin in East Antarctica lost over multiple 43 millennia (low confidence). Early-warning signals of accelerated sea-level-rise from Antarctica, could 44 possibly be observed within the next few decades. For other hazards (e.g., ice sheet behaviour, glacier mass 45 loss and global mean sea level change, coastal floods, coastal erosion, air pollution, and ocean acidification) 46 the time and/or scenario dimensions remain critical, and a simple and robust relationship with global 47 warming level cannot be established (high confidence). {4.3.2, 4.7.2, 5.4.3, 5.4.5, 5.4.8, 8.6, 9.2, 9.4, Box 48 9.3, Cross-Chapter Box 12.1} 49 50 For global climate indicators, evidence for abrupt change is limited. For global warming up to 2°C above 51 1850-1900 levels, paleoclimate records do not indicate abrupt changes in the carbon cycle (low confidence). 52 Despite the wide range of model responses, uncertainty in atmospheric CO2 by 2100 is dominated by future 53 anthropogenic emissions rather than uncertainties related to carbon-climate feedbacks (high confidence). 54 There is no evidence of abrupt change in climate projections of global temperature for the next century: there 55 is a near-linear relationship between cumulative CO2 emissions and maximum global mean surface air Do Not Cite, Quote or Distribute TS-71 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 temperature increase caused by CO2 over the course of this century for global warming levels up to at least 2 2°C relative to 1850-1900. The increase in global ocean heat content (TS2.4) will likely continue until at 3 least 2300 even for low-emission scenarios, and global mean sea level rise will continue to rise for centuries 4 to millennia following cessation of emissions (Box TS.4) due to continuing deep ocean heat uptake and mass 5 loss of the Greenland and Antarctic Ice Sheets (high confidence). {2.2.3; Cross-Chapter Box 2.1; 5.1.1; 5.4; 6 Cross-Chapter Box 5.1; Figures 5.3, 5.4, 5.25, 5.26; 9.2.2; 9.2.4} 7 8 The response of biogeochemical cycles to anthropogenic perturbations can be abrupt at regional scales and 9 irreversible on decadal to century time scales (high confidence). The probability of crossing uncertain 10 regional thresholds increases with climate change (high confidence). It is very unlikely that gas clathrates 11 (mostly methane) in deeper terrestrial permafrost and subsea clathrates will lead to a detectable departure 12 from the emissions trajectory during this century. Possible abrupt changes and tipping points in 13 biogeochemical cycles lead to additional uncertainty in 21st century atmospheric GHG concentrations, but 14 future anthropogenic emissions remain the dominant uncertainty (high confidence). There is potential for 15 abrupt water cycle changes in some high-emission scenarios, but there is no overall consistency regarding 16 the magnitude and timing of such changes. Positive land surface feedbacks, including vegetation, dust, and 17 snow, can contribute to abrupt changes in aridity, but there is only low confidence that such changes will 18 occur during the 21st century. Continued Amazon deforestation, combined with a warming climate, raises 19 the probability that this ecosystem will cross a tipping point into a dry state during the 21st century (low 20 confidence). {TS3.2.2, 5.4.3, 5.4.5, 5.4.8, 5.4.9, 8.6.2, 8.6.3, Cross-chapter Box 12.1} 21 22 [END BOX TS.9 HERE] 23 24 25 TS.4 Regional Climate Change 26 27 This section focuses on how to generate regional climate change information and its relevance for climate 28 services; the drivers of regional climate variability and change, and how they are being affected by 29 anthropogenic factors; and observed, attributed and projected changes in climate, including extreme events 30 and climatic impact-drivers (CID), across all regions of the world. There is a small set of CID changes common 31 to all land or ocean regions and a specific set of changes from a broader range of CIDs seen in each region. 32 This regional diversity results from regional climate being determined by a complex interplay between the 33 seasonal-to-multidecadal variation of large-scale modes of climate variability, external natural and 34 anthropogenic forcings, local climate processes and related feedbacks. 35 36 37 TS.4.1 Generation and Communication of Regional Climate Change Information 38 39 Climate change information at regional scale is generated using a range of data sources and methodologies. 40 Multi-model ensembles and models with a range of resolutions are important data sources, and discarding 41 models that fundamentally misrepresent relevant processes improves the credibility of ensemble information 42 related to these processes. A key methodology is distillation, combining lines of evidence and accounting for 43 stakeholder context and values, which helps ensure the information is relevant, useful and trusted for decision 44 making (see Core Concepts Box) (high confidence). Since the AR5, physical climate storylines have emerged 45 as a complementary approach to ensemble projections to generate more accessible climate information and 46 promote a more comprehensive treatment of risk. They have been used as part of the distillation process within 47 climate services to generate the required context-relevant, credible and trusted climate information. Since AR5, 48 climate change information produced for climate services has increased significantly due to scientific and 49 technological advancements and growing user awareness, requirements, and demand (very high confidence). 50 The decision-making context, level of user engagement and co-production between scientists, practitioners 51 and users are important determinants of the type of climate service developed and its utility in supporting 52 adaptation, mitigation and risk management decisions. {Box TS.1, 10.3, Cross-Chapter Box 10.3, 10.6, 12.6, 53 Cross-chapter Box 12.2, Box TS.3} 54 Do Not Cite, Quote or Distribute TS-72 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.4.1.1 Sources and Methodologies for Generating Regional Climate Information 2 3 Climate change information at regional scale is generated using a range of data sources and methodologies 4 (TS.1.4). Understanding observed regional climate change and variability is based on the availability and 5 analysis of multiple observational datasets that are suitable for evaluating the phenomena of interest (e.g., 6 extreme events), including accounting for observational uncertainty. These datasets are combined with climate 7 model simulations of observed changes and events to attribute causes of those changes and events to large- 8 and regional-scale anthropogenic and natural drivers and to assess the performance of the models. Future 9 simulations with many climate models (multi-model ensembles) are then used to generate and quantify ranges 10 of projected regional climate responses. Discarding models that fundamentally misrepresent relevant processes 11 improves the credibility of regional climate information generated from these ensembles (high confidence). 12 However, multi-model mean and ensemble spread are not a full measure of the range of projection uncertainty 13 and are not sufficient to characterize low-likelihood, high-impact changes (Box TS.3) or situations where 14 different models simulate substantially different or even opposite changes (high confidence). Large single- 15 model ensembles are now available and provide a more comprehensive spectrum of possible changes 16 associated with internal variability (high confidence). {1.5.1, 1.5.4, 10.2, 10.3.3, 10.3.4, 10.4.1, 10.6.2, 11.2, 17 Box 11.2, Cross-chapter Box 11.1, 12.4, Atlas.1.4.1, TS.1.2.1, TS.1.2.3, TS.4.2} 18 19 Depending on the region of interest, representing regionally important forcings (e.g., aerosols, land-use change 20 and ozone concentrations) and feedbacks (e.g., between snow and albedo, soil-moisture and temperature, soil- 21 moisture and precipitation) in climate models is a prerequisite for them to reproduce past regional trends to 22 underpin the reliability of future projections (medium confidence). In some cases, even the sign of a projected 23 change in regional climate cannot be trusted if relevant regional processes are not represented, for example, 24 for variables such as precipitation and wind speed (medium confidence). In some regions, either geographical 25 (e.g., Central Africa, Antarctica) or typological (e.g., mountainous areas, Small Islands and cities), and for 26 certain phenomena, fewer observational records are available or accessible, which limits the assessment of 27 regional climate change in these cases. {1.5.1, 1.5.3, 1.5.4, 8.5.1, 10.2, 10.3.3, 10.4.1, 11.1.6, 11.2, 12.4, 28 Atlas.8.3, Atlas.11.1.5, Cross-Chapter Box Atlas.2, TS.1.2.2} 29 30 Methodologies such as statistical downscaling, bias adjustment and weather generators are beneficial as an 31 interface between climate model projections and impact modelling and for deriving user-relevant indicators 32 (high confidence). However, the performance of these techniques depends on that of the driving climate model: 33 in particular, bias adjustment cannot overcome all consequences of unresolved or strongly misrepresented 34 physical processes such as large-scale circulation biases or local feedbacks (medium confidence). {10.3.3, 35 Cross-Chapter Box 10.2, 12.2, Atlas.2.2} 36 37 38 [START BOX TS.10 HERE] 39 40 Box TS.10: Event Attribution 41 42 The attribution of observed changes in extremes to human influence (including greenhouse gas and aerosol 43 emissions and land-use changes) has substantially advanced since AR5, in particular for extreme 44 precipitation, droughts, tropical cyclones, and compound extremes (high confidence). There is limited 45 evidence for windstorms and convective storms. Some recent hot extreme events would have been extremely 46 unlikely to occur without human influence on the climate system. (TS.1) {Cross-Working Group Box: 47 Attribution, 11.2, 11.3, 11.4, 11.6, 11.7, 11.8} 48 49 Since AR5, the attribution of extreme weather events has emerged as a growing field of climate research 50 with an increasing body of literature. It provides evidence that greenhouse gases and other external forcings 51 have affected individual extreme weather events by disentangling anthropogenic drivers from natural 52 variability. Event attribution is now an important line of evidence for assessing changes in extremes on 53 regional scales. {Cross-Working Group Box: Attribution, TS.1, 11.1.4} 54 Do Not Cite, Quote or Distribute TS-73 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 The regional extremes and events that have been studied are geographically uneven. A few events, for 2 example, extreme rainfall events in the UK, heat waves in Australia, or Hurricane Harvey that hit Texas in 3 2017, have been heavily studied. Many highly impactful extreme weather events have not been studied in the 4 event attribution framework, particularly in the developing world where studies are generally lacking. This is 5 due to various reasons, including lack of observational data, lack of reliable climate models, and lack of 6 scientific capacity. While the events that have been studied are not representative of all extreme events that 7 have occurred, and results from these studies may also be subject to selection bias, the large number of event 8 attribution studies provide evidence that changes in the properties of these local and individual events are in 9 line with expected consequences of human influence on the climate and can be attributed to external drivers. 10 {Cross-Working Group Box: Attribution, TS.4.1, 11.1.4, 11.2.2} 11 12 It is very likely that human influence is the main contributor to the observed increase in the intensity and 13 frequency of hot extremes and the observed decrease in the intensity and frequency of cold extremes on 14 continental scales. Some specific recent hot extreme events would have been extremely unlikely to occur 15 without human influence on the climate system. Changes in aerosol concentrations have likely slowed the 16 increase in hot extremes in some regions in particular from 1950–1980. No-till farming, irrigation, and crop 17 expansion have similarly attenuated increases in summer hot extremes in some regions, such as central North 18 America (medium confidence). {11.3.4} 19 20 Human influence has contributed to the intensification of heavy precipitation in three continents where 21 observational data are most abundant, including North America, Europe and Asia (high confidence). On 22 regional scales, evidence of human influence on extreme precipitation is limited, but new evidence from 23 attributing individual heavy precipitation events found that human influence was a significant driver of the 24 events. {11.4.4} 25 26 There is low confidence that human influence has affected trends in meteorological droughts in most regions, 27 but medium confidence that they have contributed to the severity of some specific events. There is medium 28 confidence that human-induced climate change has contributed to increasing trends in the probability or 29 intensity of recent agricultural and ecological droughts, leading to an increase of the affected land area. 30 {11.6.4} 31 32 Event attribution studies of specific strong tropical cyclones provide limited evidence for anthropogenic 33 effects on tropical cyclone intensifications so far, but high confidence for increases in precipitation. There is 34 high confidence that anthropogenic climate change contributed to extreme rainfall amounts during Hurricane 35 Harvey (2017) and other intense tropical cyclones. {11.7.3} 36 37 The number of evident attribution studies on compound events is limited. There is medium confidence that 38 weather conditions that promote wildfires have become more probable in southern Europe, northern Eurasia, 39 the USA, and Australia over the last century. In Australia a number of event attribution studies show that 40 there is medium confidence of increase in fire weather conditions due to human influence. {11.8.3, 12.4.3.2} 41 42 43 [START BOX TS.10, FIGURE 1 HERE] 44 45 Box TS.10, Figure 1: Synthesis of assessed observed changes and human influence of hot extremes (panel a), 46 heavy precipitation (panel b) and agricultural and ecological drought (panel c) for the 47 IPCC AR6 regions (displayed as hexagons). The colours in each panel represent the four 48 outcomes of the assessment on the observed changes: In Panel a): red – at least medium 49 confidence in an observed increase in hot extremes; blue – at least medium confidence in an 50 observed decrease in hot extremes; white – no significant change in hot extremes is observed for 51 the region as a whole; grey – the evidence in this region is insufficient (because of a lack of data 52 and/or literature) to make an assessment for the region as a whole. In panel b): green – at least 53 medium confidence in an observed increase in heavy rainfall; yellow - at least medium 54 confidence in an observed decrease in heavy rainfall; white – no significant change in heavy 55 rainfall is observed for the region as a whole; grey – the evidence in this region is insufficient Do Not Cite, Quote or Distribute TS-74 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 (because of a lack of data and/or literature) to make an assessment for the region as a whole. In 2 panel c): yellow – at least medium confidence in an observed increase in agricultural and 3 ecological drought; green - at least medium confidence in an observed decrease in agricultural 4 and ecological drought; white – no significant change in agricultural and ecological drought is 5 observed for the region as a whole; grey – the evidence in this region is insufficient (because of a 6 lack of data and/or literature) to make an assessment for the region as a whole. Each panel 7 represents in addition the synthesis of assessment of the human influence on the observed 8 changes based on available trend detection and attribution and event attribution scientific 9 publications. The level of confidence is indicated by a number of dots: high confidence (three 10 dots), medium confidence (two dots), low confidence (one dot), and when no assessment is 11 possible, due to insufficient evidence for the specific region (horizontal bar). For hot extremes, 12 the evidence is mostly drawn from changes in metrics based on daily maximum temperatures, 13 regional studies using other metrics (heatwave duration, frequency and intensity) are used in 14 addition {11.9.2}. For heavy precipitation, the evidence is mostly drawn from changes in metrics 15 based on one-day or five-day precipitation amounts using global and regional studies {11.9.3}. 16 Agricultural and ecological droughts are assessed based on observed and projected changes in 17 total column soil moisture, complemented by evidence on changes in surface soil moisture, 18 water-balance (precipitation minus evapotranspiration) and metrics driven by precipitation and 19 atmospheric evaporative demand. {11.9.3} All assessments are made for each AR6 region as a 20 whole and for the timeframe from 1950 to present thus, more local or assessment made on 21 shorter time scales might differ from what is shown in the figure. {11.9, Table TS.5}. 22 23 [END BOX TS.10, FIGURE 1 HERE] 24 25 [END BOX TS.10 HERE] 26 27 28 TS.4.1.2 Regional Climate Information Distillation and Climate Services 29 30 The construction of regional climate information involves people with a variety of backgrounds, from various 31 disciplines, who have different sets of experiences, capabilities and values. The process of synthesizing climate 32 information from different lines of evidence from a number of sources, taking into account the context of a 33 user vulnerable to climate variability and change and the values of all relevant actors, is called distillation. 34 Distillation is conditioned by the sources available, the actors involved and the context, which all depend 35 heavily on the regions considered, and framed by the question being addressed. Distilling regional climate 36 information from multiple lines of evidence and taking the user context into account increases fitness, 37 usefulness, relevance and trust in that information for use in climate services (Box TS.11) and decision-making 38 (high confidence). {1.2.3, 10.1.4, 10.5, Cross-Chapter Box 10.3, 12.6} 39 40 The distillation process can vary substantially, as it needs to consider multiple lines of evidence on all 41 physically plausible outcomes (especially when they are contrasting) relevant to a specific decision required 42 in response to a changing climate. Confidence in the distilled regional climate information is enhanced when 43 there is agreement across multiple lines of evidence, so the outcome can be limited if these are inconsistent or 44 contradictory. For example, in the Mediterranean region the agreement between different lines of evidence 45 such as observations, projections by regional and global models, and understanding of the underlying 46 mechanisms provides high confidence in summer warming that exceeds the global average (see Box TS.12). 47 In a less clear-cut case for Cape Town, despite consistency among global model future projections, there is 48 medium confidence in a projected future drier climate due to the lack of consistency in links between increasing 49 greenhouse gases, changes in a key mode of variability (the Southern Annular Mode) and drought in Cape 50 Town among different observation periods and in model simulations. {10.5.3, 10.6, 10.6.2, 10.6.4, Cross- 51 Chapter Box 10.3, 12.4} 52 53 Since the AR5, physical climate storyline approaches have emerged as a complementary instrument to provide 54 a different perspective on or additional climate information, to facilitate communication of the information or 55 provide a more flexible consideration of risk. Storylines that condition climatic events and processes on a set 56 of plausible but distinct large-scale climatic changes enable the exploration of uncertainties in regional climate 57 projections. For example, they can explicitly address low-likelihood, high-impact outcomes, which would be Do Not Cite, Quote or Distribute TS-75 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 less emphasized in a probabilistic approach, and can be embedded in a user’s risk landscape, taking account 2 of socio-economic factors as well as physical climate changes. Storylines can also be used to communicate 3 climate information by narrative elements describing and contextualising the main climatological features and 4 the relevant consequences in the user context and, as such, can be used as part of a climate information 5 distillation process. {1.4.4., Box 10.2, Box 11.2, 11.2, Cross-chapter Box 12.2} 6 7 8 [START BOX TS.11 HERE] 9 10 Box TS.11: Climate Services 11 12 Climate services involve providing climate information to assist decision-making, for example, about how 13 extreme rainfall will change to inform improvements in urban drainage. Since AR5, there has been a significant 14 increase in the range and diversity of climate service activities (very high confidence). The level of user- 15 engagement, co-design and co-production are factors determining the utility of climate services, while resource 16 limitations for these activities constrain their full potential. {12.6, Cross-Chapter Box 12.2} 17 18 Climate services include engagement from users and providers and an effective access mechanism; they are 19 responsive to user needs and based on integrating scientifically credible information and relevant expertise. 20 Climate services are being developed across regions, sectors, time scales and users-groups and include a range 21 of knowledge brokerage and integration activities. These involve identifying knowledge needs; compiling, 22 translating and disseminating knowledge; coordinating networks and building capacity through informed 23 decision-making; analysis, evaluation and development of policy; and personal consultation. 24 25 Since AR5, climate change information produced in climate service contexts has increased significantly due 26 to scientific and technological advancements and growing user awareness, requirements, and demand (very 27 high confidence). Climate services are growing rapidly and are highly diverse in their practices and products. 28 The decision-making context, level of user engagement and co-production between scientists, practitioners 29 and intended users are important determinants of the type of climate service developed and their utility for 30 supporting adaptation, mitigation and risk management decisions. They require different types of user– 31 producer engagement depending on what the service aims to deliver (high confidence), and these fall into three 32 broad categories: website-based services, interactive group activities and focused relationships. 33 34 Realization of the full potential of climate services is often hindered by limited resources for the co-design and 35 co-production process, including sustained engagement between scientists, service providers and users (high 36 confidence). Further challenges relate to the development and provision of climate services, generation of 37 climate service products, communication with users, and evaluation of their quality and socio-economic 38 benefit. {TS.4.1, 1.2.3, 10.5.4, 12.6, Cross-Chapter Box 12.2, Glossary} 39 40 [END BOX TS.11 HERE] 41 42 43 [START BOX TS.12 HERE] 44 45 Box TS.12: Multiple Lines of Evidence for Assessing Regional Climate Change and the Interactive 46 Atlas 47 48 A key novel element in the AR6 is the WGI Atlas, which includes the Interactive Atlas. The Interactive Atlas 49 provides the ability to explore much of the observational and climate model data used as lines of evidence in 50 this assessment to generate regional climate information. {10.6.4, Atlas.2, Interactive Atlas} 51 52 A significant innovation in the AR6 WGI report is the Atlas. Part of its remit is to provide region-by-region 53 assessment on changes in mean climate and to link with other WGI chapters to generate climate change 54 information for the regions. An important component is the new online interactive tool, the Interactive Atlas, Do Not Cite, Quote or Distribute TS-76 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 with flexible spatial and temporal analyses of much of the observed, simulated past and projected future 2 climate change data underpinning the WGI assessment. This includes the ability to generate global maps and 3 a number of regionally aggregated products (time series, scatter plots, tables, etc.) for a range of observations 4 and ensemble climate change projections of variables (such as changes in the CIDs summarized in Table TS.5) 5 from CMIP5, CORDEX and CMIP6. The data can be displayed and summarized under a range of SSP-RCP 6 scenarios and future time slices and also for different global warming levels, relative to several different 7 baseline periods. The maps and various statistics can be generated for annual mean trends and changes or for 8 any user-specified season. A new set of WGI reference regions is used for the regional summary statistics and 9 applied widely throughout the report (with the regions available, along with aggregated datasets and the code 10 to generate these at the ATLAS GitHub: https://github.com/IPCC-WG1/Atlas). 11 12 Box TS.12, Figure 1 shows how the Interactive Atlas products, together with other lines of evidence, can be 13 used to generate climate information for an illustrative example of the Mediterranean summer warming. The 14 lines of evidence include the understanding of relevant mechanisms, dynamic and thermodynamic processes 15 and the effect of aerosols in this case (Box TS.12, Figure 1a), trends in observational datasets (which can have 16 different spatial and temporal coverage – Box TS.12, Figure 1b, c), attribution of these trends and temperature 17 projections from global and regional climate models at different resolutions, including single-model initial- 18 condition large ensembles (SMILEs; Box TS.12, Figure 1d, e). Taken together, this evidence shows there is 19 high confidence that the projected Mediterranean summer temperature increase will be larger than in the global 20 mean with consistent results from CMIP5 and CMIP6 (Box TS.12, Figure 1e). However, CMIP6 results project 21 both more pronounced warming than CMIP5 for a given emissions scenario and time period and a greater 22 range of changes (Box TS.12, Figure 1d). {10.6.4, Atlas.2, Interactive Atlas} 23 24 25 [START BOX TS.12, FIGURE 1 HERE] 26 27 Box TS.12, Figure 1: Example of generating regional climate information from multiple lines of evidence for the 28 case of Mediterranean summer warming, with indication of the information available from 29 the Interactive Atlas. (a) Mechanisms and feedbacks involved in enhanced Mediterranean 30 summer warming. (b) Locations of observing stations from different datasets. (c) Distribution of 31 1960‒2014 summer temperature trends (°C per decade) for observations (black crosses), CMIP5 32 (blue circles), CMIP6 (red circles), HighResMIP (orange circles), CORDEX EUR-44 (light blue 33 circles), CORDEX EUR-11 (green circles), and selected single Model Initial-condition Large 34 Ensembles - SMILEs (grey boxplots, MIROC6, CSIRO-Mk3-6-0, MPI-ESM and d4PDF). (d) 35 Time series of area averaged (25°N‒50°N, 10°W‒40°E) land point summer temperature 36 anomalies (°C, baseline period is 1995–2014): the boxplot shows long term (2081–2100) 37 temperature changes of different CMIP6 scenarios in respect to the baseline period. (e) Projected 38 Mediterranean summer warming in comparison to global annual mean warming of CMIP5 39 (RCP2.6, RCP4.5, RCP6.0 and RCP8.5) and CMIP6 (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5- 40 8.5) ensemble means (lines) and spread (shading). {Figure 10.20, Figure 10.21, Figure Atlas.8} 41 42 [END BOX TS.12, FIGURE 1 HERE] 43 44 [END BOX TS.12 HERE] 45 46 47 TS.4.2 Drivers of Regional Climate Variability and Change 48 49 Anthropogenic forcing, including GHGs and aerosols but also regional land use and irrigation have all affected 50 observed regional climate changes (high confidence) and will continue to do so in the future (high confidence), 51 with various degrees of influence and response times, depending on warming levels, the nature of the forcing 52 and the relative importance of internal variability. Since the late 19th century, major modes of variability 53 (MoVs) exhibited fluctuations in frequency and magnitude at multi-decadal time scales, but no sustained trends 54 outside the range of internal variability (Table TS.4). An exception is the Southern Annular Mode (SAM), 55 which has become systematically more positive (high confidence) and is projected to be more positive in all 56 seasons, except for December-January-February (DJF), in high CO2 emissions scenarios (high confidence). Do Not Cite, Quote or Distribute TS-77 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 The influence of stratospheric ozone forcing on the SAM trend has been reduced since the early 2000s 2 compared to earlier decades, contributing to the weakening of its positive trend as observed over 2000–2019 3 (medium confidence). In the near-term, projected changes in most of the MoVs and related teleconnections 4 will likely be dominated by internal variability. In the long-term, it is very likely that the precipitation variance 5 related to El Niño–Southern Oscillation will increase. Physical climate storylines, including the complex 6 interplay between climate drivers, MoVs, and local and remote forcing, increase confidence in the 7 understanding and use of observed and projected regional changes. {2.4, 3.7, 4.3, 4.4, 4.5, 6.4, 8.3, 8.4, 10.3, 8 10.4, 11.3} 9 10 11 TS.4.2.1 Regional Fingerprints of Anthropogenic and Natural Forcing 12 13 While anthropogenic forcing has contributed to multi-decadal mean precipitation changes in several regions, 14 internal variability can delay emergence of the anthropogenic signal in long-term precipitation changes in 15 many land regions (high confidence). At the regional scale, the effect of human-induced GHG forcing on 16 extreme temperature is moderated or amplified by soil moisture feedback, snow/ice-albedo feedback, regional 17 forcing from land-use/land-cover changes, forcing from aerosol concentrations, or decadal/multi-decadal 18 natural variability. Changes in local and remote aerosol forcings lead to south-north gradients of the effective 19 radiative forcing (ERF) (hemispherical asymmetry). Along latitudes, it is more uniform with strong 20 amplification of the temperature response towards the Arctic (medium confidence). The decrease of SO2 21 emissions since the 1980s reduces the damping effect of aerosols, leading to a faster increase in surface air 22 temperature that is most pronounced at mid- and high latitudes of the Northern Hemisphere, where the largest 23 emission reductions have taken place (medium confidence). {1.3, 3.4.1, 6.3.4, 6.4.1, 6.4.3, 8.3.1, 8.3.2, Box 24 8.1, 10.4.2, 10.6, 11.1.6, 11.3} 25 26 Multidecadal dimming and brightening trends in incoming solar radiation at the Earth’s surface occurred at 27 widespread locations (high confidence). Multi-decadal variation in anthropogenic aerosol emissions are 28 thought to be a major contributor (medium confidence), but multi-decadal variability in cloudiness may also 29 have played a role. Volcanic eruptions affect regional climate through their spatially heterogeneous effect on 30 the radiative budget as well as through triggering dynamical responses by favouring a given phase from some 31 MoVs, for instance. {1.4.1, Cross-Chapter Box 1.2, 2.2.1, 2.2.2, 3.7.1, 3.7.3, 4.3.1, 4.4.1, 4.4.4, Cross-Chapter 32 Box 4.1, 7.2.2, 8.5.2, 10.1.4, 11.1.6, 11.3.1} 33 34 Historical urbanization affects the observed warming trends in cities and their surroundings (very high 35 confidence). Future urbanization will amplify the projected air temperature under different background 36 climates, with a strong effect on minimum temperatures that could be as large as the global warming signal 37 (very high confidence). Irrigation and crop expansion have attenuated increases in summer hot extremes in 38 some regions, such as central North America (medium confidence). {Box 10.3, 11.1.6, 11.3, Box TS.6, Box 39 TS.14} 40 41 42 TS.4.2.2 Modes of Variability and Regional Teleconnections 43 44 Modes of Variability (MoVs) (Annex IV, Table TS.4) have existed for millennia or longer (high confidence), 45 but there is low confidence in detailed reconstructions of most of them prior to direct instrumental records. 46 MoVs are treated as a main source of uncertainties associated with internal dynamics, as they can either 47 accentuate or dampen, even mask, the anthropogenically-forced responses. {2.4, 8.5.2, 10.4, 10.6, 11.1.5, 48 Atlas.3.1} 49 50 Since the late 19th century, major MoVs (Table TS.4) show no sustained trends, exhibiting fluctuations in 51 frequency and magnitude at multi-decadal time scales, except for the Southern Annular Mode (SAM), which 52 has become systematically more positive (high confidence) (Table TS.4). It is very likely that human influence 53 has contributed to this trend from the 1970s to the 1990s, and to the associated strengthening and southward 54 shift of the Southern Hemispheric extratropical jet in austral summer. The influence of stratospheric ozone 55 forcing on the SAM trend has been reduced since the early 2000s compared to earlier decades, contributing to Do Not Cite, Quote or Distribute TS-78 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 the weakening of its positive trend observed over 2000–2019 (medium confidence). By contrast, the cause of 2 the Northern Annular Mode (NAM) trend toward its positive phase since the 1960s and associated northward 3 shifts of Northern Hemispheric extratropical jet and storm track in boreal winter is not well understood. The 4 evaluation of model performance on simulating MoVs is assessed in TS.1.2.2. {2.3.3, 2.4, 3.3.3, 3.7.1, 3.7.2} 5 6 In the near term, the forced change in SAM in austral summer is likely to be weaker than observed during the 7 late 20th century under all five SSPs assessed. This is because of the opposing influence in the near- to mid- 8 term from stratospheric ozone recovery and increases in other greenhouse gases on the Southern Hemisphere 9 summertime mid-latitude circulation (high confidence). In the near term, forced changes in the SAM in austral 10 summer are therefore likely to be smaller than changes due to natural internal variability. In the long-term 11 (2081–2100) under the SSP5-8.5 scenario, the SAM index is likely to increase in all seasons relative to 1995– 12 2014. The CMIP6 multi-model ensemble projects a long-term (2081–2100) increase in the boreal wintertime 13 NAM index under SSP3-7.0 and SSP5-8.5, but regional associated changes may deviate from a simple shift in 14 the mid-latitude circulation due to a modified teleconnection resulting from interaction with a modified mean 15 background state. {4.3.3, 4.4.3, 4.5.1, 4.5.3, 8.4.2} 16 17 Human influence has not affected the principal tropical modes of interannual climate variability (Table TS.4) 18 and their associated regional teleconnections beyond the range of internal variability (high confidence). It is 19 virtually certain that the ENSO will remain the dominant mode of interannual variability in a warmer world. 20 There is no consensus from models for a systematic change in amplitude of ENSO SST variability over the 21 21st century in any of the SSP scenarios assessed (medium confidence). However, it is very likely that rainfall 22 variability related to ENSO will be enhanced significantly by the latter half of the 21st century in the SSP2- 23 4.5, SSP3-7.0, and SSP5-8.5 scenarios, regardless of the amplitude changes in SST variability related to the 24 mode It is very likely that rainfall variability related to changes in the strength and spatial extent of ENSO 25 teleconnections will lead to significant changes at regional scale. {3.7.3, 3.7.4, 3.7.5, 4.3.3, 4.5.3, 8.4.2, 10.3.3} 26 27 Modes of decadal and multidecadal variability over the Pacific and Atlantic Ocean exhibit no significant 28 changes in variance over the period of observational records (high confidence). There is medium confidence 29 that anthropogenic and volcanic aerosols contributed to observed temporal evolution in the Atlantic 30 Multidecadal Variability (AMV) and associated regional teleconnections, especially since the 1960s, but there 31 is low confidence in the magnitude of this influence and the relative contributions of natural and anthropogenic 32 forcings. Internal variability is the main driver of Pacific Decadal Variability (PDV) observed since the start 33 of the instrumental records (high confidence), despite some modelling evidence for potential external 34 influence. There is medium confidence that the AMV will undergo a shift towards a negative phase in the near- 35 term. {2.4, 3.7.6, 3.7.7, 8.5.2, 4.4.3} 36 37 38 [START TABLE TS.4 HERE] 39 40 Table TS.4: Summary of the assessments on modes of variability (MoVs) and associated teleconnections. (a) 41 Assessments on observed changes since the start of instrumental records, CMIP5 and CMIP6 model 42 performance, human influence on the observed changes, and near-term (2021–2040) and mid- to long- 43 term (2041–2100) changes. Curves schematically illustrate the assessed overall changes, with the 44 horizontal axis indicating time, and are not intended to precisely represent the time evolution. (b) 45 Fraction of surface air temperature (SAT) and precipitation (pr) variance explained at interannual 46 timescale by each MoV for each AR6 region (numbers in each cell; in percent). Values correspond to 47 the average of significant explained variance fractions based on HadCRUT, GISTEMP, BerkeleyEarth 48 and CRU-TS (for SAT) and GPCC and CRU-TS (for precipitation). Significance is tested based on F- 49 statistics at the 95% level confidence, and a slash indicates that the value is not significant in more than 50 half of the available data sets. The colour scale corresponds to the sign and values of the explained 51 variance as shown at the bottom. The corresponding anomaly maps are shown in Annex IV. DJF: 52 December-January-February. MAM: March-April-May. JJA: June-July-August. SON: September- 53 October-November. In (b), Northern Annular Mode (NAM) and El Niño–Southern Oscillation (ENSO) 54 teleconnections are evaluated for 1959–2019, Southern Annular Mode (SAM) for 1979–2019, Indian 55 Ocean Basin (IOB), Indian Ocean Dipole (IOD), Atlantic Zonal Mode (AZM) and Atlantic Meridional Do Not Cite, Quote or Distribute TS-79 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Mode (AMM) for 1958–2019, and Pacific Decadal Variability (PDV) and Atlantic Multidecadal 2 Variability (AMV) for 1900–2019. All data are linearly detrended prior to computation. {2.4, 3.7, 4.3.3, 3 4.4.3, 4.5.3, Table Atlas.1, Annex IV, TS.1.2.2} 4 5 (a) Assessments on MoV (a) Assessments on MoV NAM SAM ENSO IOB IOD AZM AMM PDV AMV Within proxy- Within proxy- Dominated Dominated inferred inferred Limited Limited by by Past changes since the variability variability evidence evidence multidecadal multidecadal start of observations range range fluctuations fluctuations {2.4.1.1} {2.4.1.2} {2.4.2} {2.4.3} {2.4.3} {2.4.4} {2.4.4} {2.4.5} {2.4.6} High High Medium Medium Medium Low Low Medium Medium CMIP5 and CMIP6 performance performance performance performance performance performance performance performance performance model performance {3.7.1} {3.7.2} {3.7.3} {3.7.4} {3.7.4} {3.7.5} {3.7.5} {3.7.6} {3.7.7} Contributed Contributed No robust through GHG Low No robust Not No robust No robust Not Human influence on the evidence (all seasons) agreement evidence detected evidence evidence detected through aerosols observed changes &ozone (DJF) {3.7.1} {3.7.2} {3.7.3} {3.7.4} {3.7.4} {3.7.5} {3.7.5} {3.7.6} {3.7.7} Internal Internal Near-term future No robust No robust No robust No robust Limited variability variability evidence evidence evidence evidence evidence dominates dominates changes (2021–2040) {4.4.3.1} {4.4.3.1} {4.4.3.2} {4.4.3.3} {4.4.3.3} {4.4.3.4} {4.4.3.4} {4.4.3.5} {4.4.3.6} Mid-to-long-term No robust No robust No robust No changes evidence evidence evidence future changes (2041– Internal 2100) variability dominates {4.3.3.1; {4.3.3.1; {4.3.3.2; {4.5.3.3} {4.5.3.3} {4.5.3.4} {4.5.3.4} {4.5.3.5} {4.5.3.6} 4.5.3.1} 4.5.3.1} 4.5.3.2} 6 low confidence medium confidence high confidence more likely than not likely very likely 7 8 (b) Regional climate anomalies associated with MoV 9 (b) Regional climate anomalies associated with MoV Mode NAM SAM ENSO IOB IOD AZM AMM PDV AMV Season DJF DJF DJF MAM SON JJA JJA annual annual Variable SAT pr SAT pr SAT pr SAT pr SAT pr SAT pr SAT pr SAT pr SAT pr Sahara 58 14 10 19 12 9 12 25 Western Africa 25 15 45 21 10 6 6 23 Central Africa 19 8 10 14 50 13 10 14 11 Africa North Eastern Africa 19 7 14 36 32 7 7 South Eastern Africa 14 22 36 57 10 4 9 West Southern Africa 49 26 27 16 8 4 12 5 East Southern Africa 13 75 34 35 7 4 6 Madagascar 24 24 7 11 10 9 5 West Siberia 45 7 9 11 East Siberia 52 3 11 Russian Far East 8 10 11 6 5 5 West Central Asia 15 21 4 East Central Asia 38 Asia Tibetan Plateau 15 15 7 11 6 5 9 East Asia 7 20 23 9 9 13 South Asia 9 12 8 8 5 Southeast Asia 39 31 73 6 48 5 12 7 Arabian Peninsula 32 10 24 20 5 13 7 Northern Australia 21 13 38 19 7 7 7 alasia Austr Central Australia 14 21 12 18 22 20 7 7 6 5 East Australia 22 20 11 18 9 8 7 7 8 Do Not Cite, Quote or Distribute TS-80 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI Southern Australia 11 23 40 8 3 New Zealand 16 Southern Central America 21 16 33 10 11 17 6 6 7 Central & South America Northwestern South America 7 14 16 82 17 54 18 13 16 7 8 Northern South America 7 56 58 61 22 17 24 9 12 7 Northeastern South America 25 58 19 9 12 8 South American Monsoon 54 31 22 7 6 7 Southwestern South America 10 16 14 17 10 16 8 Southeastern South America 21 13 21 10 12 5 6 Southern South America 23 13 7 9 Mediterranean 28 58 7 19 Europe Western & Central Europe 28 18 13 10 4 8 Eastern Europe 35 7 Northern Europe 53 32 6 North Central America 10 26 13 27 18 7 12 15 12 6 19 North America Western North America 4 6 5 Central North America 17 12 17 8 3 9 6 Eastern North America 12 11 9 4 9 4 Northeastern North America 18 26 8 10 9 4 Northwestern North America 14 10 8 17 8 4 Caribbean 10 15 18 26 8 10 17 12 7 5 Islands Small Pacific Greenland/Iceland 42 8 7 44 Polar Terrestrial Russian Arctic 25 10 6 11 8 Regions West Antarctica 8 21 East Antarctica 38 1 colder warmer drier wetter Not significant in >50% 2 of available data sets 3 40 30 20 0 20 30 40 Data unavailable in 4 40 [END 30 20 TABLE 0 TS.4 20 30 HERE] 40 Precipitation anomalies and explained variance (%) >50% of data sets Temperature anomalies and explained variance (%) 5 6 7 TS.4.2.3 Interplay Between Drivers of Climate Variability and Change at Regional Scales 8 9 Anthropogenic forcing has been a major driver of regional mean temperature change since 1950 in many 10 subcontinental regions of the world (virtually certain). At regional scales, internal variability is stronger, and 11 uncertainties in observations, models and external forcing are all larger than at the global scale, hindering a 12 robust assessment of the relative contributions of greenhouse gases, stratospheric ozone, and different aerosol 13 species in most of the cases. Multiple lines of evidence, combining multi-model ensemble global projections 14 with those coming from single-model initial-condition large ensembles, show that internal variability is largely 15 contributing to the delayed or absent emergence of the anthropogenic signal in long-term regional mean 16 precipitation changes (high confidence). Internal variability in ocean dynamics dominates regional patterns on 17 annual to decadal time scales (high confidence). The anthropogenic signal in regional sea level change will 18 emerge in most regions by 2100 (medium confidence). {9.2.4, 9.6.1, 10.4.1, 10.4.2, 10.4.3} 19 20 Regional climate change is subject to the complex interplay between multiple external forcings and internal 21 variability. Time evolution of mechanisms operating at different time scales can modify the amplitude of the 22 regional-scale response of temperature, and both the amplitude and sign of the response of precipitation, to 23 anthropogenic forcing (high confidence). These mechanisms include non-linear temperature, precipitation and 24 soil moisture feedbacks, slow and fast responses of SST patterns and atmospheric circulation changes to 25 increasing GHGs. Land use and aerosol forcings and land–atmosphere feedback play important roles in 26 modulating regional changes, for instance in weather and climate extremes (high confidence). These can also 27 lead to a higher warming of extreme temperatures compared to mean temperature (high confidence), and 28 possibly cooling in some regions (medium confidence). The soil moisture–temperature feedback was shown 29 to be relevant for past and present-day heat waves based on observations and model simulations. {10.4.3, 30 11.1.6, 11.3.1} Do Not Cite, Quote or Distribute TS-81 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 Southeastern South America (SES) is one of the AR6 WGI reference regions (outlined with black thick contour 3 in Figure TS.21a) and it is used here as an illustrative example of the interplay between drivers of climate 4 variability and change at regional scale. Austral summer (DJF) precipitation positive trends have been 5 observed over the region during 1950–2014. Drivers of this change include MoVs, such as AMV, ENSO, and 6 PDV, as well as external forcing, like GHG increases and ozone depletion together with aerosols (as illustrated 7 in Figure TS.21a). MoVs and external forcing collectively affect climate phenomena, such as the Hadley cell 8 width and strength, Rossby waves activity emerging from the large-scale tropical SST anomalies, and the 9 Southern Hemisphere polar vortex, which are relevant for the region. In fact, local changes over SES in terms 10 of moisture convergence, ascending motion and storm-track locations depend on these climate phenomena, 11 and they are overall responsible for the observed precipitation trends. Projections suggest continuing positive 12 trends in rainfall over SES in the near-term in response to GHG emissions scenarios. Multi-model mean and 13 ensemble spread are not sufficient to characterise situations where different models simulate substantially 14 different or even opposite changes (high confidence). In such cases, physical climate storylines addressing 15 possible outcomes for climate phenomena shown to play a role in the variability of the region of interest can 16 aid the interpretation of projection uncertainties. In addition, single-model initial-condition large ensembles of 17 many realisations of internal variability are required to separate internal variability from forced changes (high 18 confidence) and to partition the different sources of uncertainties as a function of future assessed periods. 19 {10.3.4, 10.4.2, Figure 10.12a} 20 21 22 [START FIGURE TS.21 HERE] 23 24 Figure TS.21: Example of the interplay between drivers of climate variability and change at regional scale to 25 understand past and projected changes. The figure intent is to show an illustrative pathway for 26 understanding past, and anticipating future, climate change at regional scale in the presence of 27 uncertainties. (a) Identification of the climate drivers and their influences on climate phenomena 28 contributing through teleconnection to Southeastern South America summer (DJF) precipitation 29 variability and trends observed over 1950–2014. Drivers (red squares) include MoVs as well as 30 external forcing. Observed precipitation linear trend from GPCC is shown on continents (green-brown 31 colour bar in mm month–1 per decade) and the SES AR6 WGI reference region is outlined with the 32 thick black contour. Climate phenomena leading to local effect on SES are schematically presented 33 (blue ovals). (b) Time series of decadal precipitation anomalies for DJF SES simulated from seven 34 large ensembles of historical + RCP8.5 simulations over 1950–2100. Shading corresponds to the 5th– 35 95th range of climate outcomes given from each large ensemble for precipitation (in mm/month) and 36 thick coloured lines stand for their respective ensemble mean. The thick timeseries in white 37 corresponds to the multi-model multi-member ensemble mean with model contribution being 38 weighted according to their ensemble size. GPCC observation is shown in the light black line with 39 squares over 1950–2014 and the 1995–2014 baseline period has been retained for calculation of 40 anomalies in all datasets. (c) Quantification of the respective weight (in percent) between the 41 individual sources of uncertainties (internal in grey, model in magenta and scenario in green) at near- 42 term, mid-term and long-term temporal windows defined in AR6 and highlighted in (b) for SES DJF 43 precipitation. All computations are done with respect to 1995–2014, taken as the reference period and 44 the scenario uncertainty is estimated from CMIP5 using the same set of models as for the large 45 ensembles that have run different RCP scenarios. {Figure 10.12a} 46 47 [END FIGURE TS.21 HERE] 48 49 50 51 52 53 54 55 56 57 Do Not Cite, Quote or Distribute TS-82 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [START BOX TS.13 HERE] 2 3 Box TS.13: Monsoons 4 5 Global land monsoon precipitation decreased from the 1950s to the 1980s, partly due to anthropogenic 6 aerosols, but has increased since then in response to GHG forcing and large-scale multi-decadal variability 7 (medium confidence). Northern Hemispheric anthropogenic aerosols weakened the regional monsoon 8 circulations in South Asia, East Asia and West Africa during the second half of the 20th century, thereby 9 offsetting the expected strengthening of monsoon precipitation in response to GHG-induced warming (high 10 confidence). During the 21st century, global land monsoon precipitation is projected to increase in response to 11 GHG warming in all time horizons and scenarios (high confidence). Over South and Southeast Asia, East Asia 12 and the central Sahel, monsoon precipitation is projected to increase, whereas over North America and the far 13 western Sahel it is projected to decrease (medium confidence). There is low confidence in projected 14 precipitation changes in the South American and Australian-Maritime Continent monsoons. At global and 15 regional scales, near-term monsoons changes will be dominated by the effects of internal variability (medium 16 confidence). {2.3, Cross-Chapter Box 2.4, 3.3, 4.4, 4.5, 8.2, 8.3, 8.4, 8.5, Box 8.1, Box 8.2, 10.6} 17 18 Global Monsoon 19 20 Paleoclimate records indicate that during warm climates, like the mid-Pliocene Warm Period, monsoon 21 systems were stronger (medium confidence). In the instrumental records, global summer monsoon precipitation 22 intensity has likely increased since the 1980s, dominated by Northern Hemisphere summer trends and large 23 multi-decadal variability. Contrary to the expected increase of precipitation under global warming, the 24 Northern Hemisphere monsoon regions experienced declining precipitation from the 1950s to 1980s, which is 25 partly attributable to the influence of anthropogenic aerosols (medium confidence) (Box TS.13, Figure 1). 26 {2.3.1, Cross-Chapter Box 2.4, 3.3.2, 3.3.3} 27 28 With continued global warming, it is likely that global land monsoon precipitation will increase during this 29 century (Box TS.13, Figure 1), particularly in the Northern Hemisphere, although the monsoon circulation is 30 projected to weaken. A slowdown of the tropical circulation with global warming can partly offset the 31 warming-induced strengthening of precipitation in monsoon regions (high confidence). In the near term, global 32 monsoon changes are likely to be dominated by the effects of internal variability and model uncertainties 33 (medium confidence). In the long term, global monsoon rainfall change will feature a robust north-south 34 asymmetry characterized by a greater increase in the Northern Hemisphere than in the Southern Hemisphere 35 and an east-west asymmetry characterized by enhanced Asian-African monsoons and a weakened North 36 American monsoon (medium confidence). {4.4.1, 4.5.1, 8.4.1} 37 38 Regional Monsoons 39 40 Paleoclimate reconstructions indicate stronger monsoons in the Northern Hemisphere but weaker ones in the 41 Southern Hemisphere during warm periods, particularly for the South and Southeast Asian, East Asian, North 42 and South American monsoons, with the opposite occurring during cold periods (medium confidence). It is 43 very likely that Northern Hemispheric anthropogenic aerosols weakened the regional monsoon circulations in 44 South Asia, East Asia and West Africa during the second half of the 20th century, thereby offsetting the 45 expected strengthening of monsoon precipitation in response to GHG-induced warming (Box TS.13, Figure 46 1). Multiple lines of evidence explain this contrast over South Asia, with the observed trends dominated by 47 the effects of aerosols, while future projections are mostly driven by GHG increases. The recent partial 48 recovery and enhanced intensity of monsoon precipitation over West Africa is related to the growing influence 49 of GHGs with an additional contribution due to the reduced cooling effect of anthropogenic aerosols, emitted 50 largely from North America and Europe (medium confidence). For other regional monsoons, that is, North and 51 South America and Australia, there is low confidence in the attribution of recent changes in precipitation (Box 52 TS.13, Figure 1) and winds. {2.3.1, 8.3.1, 8.3.2, Box 8.1, 10.6.3} 53 54 Projections of regional monsoons during the 21st century indicate contrasting (region-dependent) and Do Not Cite, Quote or Distribute TS-83 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 uncertain precipitation and circulation changes. The annual contrast between the wettest and driest month of 2 the year is likely to increase by 3–5% per degree Celsius in most monsoon regions in terms of precipitation, 3 precipitation minus evaporation, and runoff (medium confidence). For the North American monsoon, 4 projections indicate a decrease in precipitation, whereas increased monsoon rainfall is projected over South 5 and Southeast Asia and over East Asia (medium confidence) (Box TS.13, Figure 1). West African monsoon 6 precipitation is projected to increase over the central Sahel and decrease over the far western Sahel (medium 7 confidence). There is low confidence in projected precipitation changes in the South American and Australian- 8 Maritime Continent regional monsoons (for both magnitude or sign) (Box TS.13, Figure 1). There is medium 9 confidence that the monsoon season will be delayed in the Sahel and high confidence that it will be delayed in 10 North and South America. {8.2.2, 8.4.2.4, Box 8.2} 11 12 Building the Assessment from Multiple Lines of Evidence 13 14 Large natural variability of monsoon precipitation across different time scales, found in both paleoclimate 15 reconstructions and instrumental measurements, poses an inherent challenge for robust quantification of future 16 changes in precipitation at regional and smaller spatial scales. At both global and regional scales, there is 17 medium confidence that internal variability contributes the largest uncertainty related to projected changes, at 18 least in the near term (2021–2040). A collapse of the Atlantic Multidecadal Overturning Circulation (AMOC) 19 could weaken the African and Asian monsoons but strengthen the Southern Hemisphere monsoons (high 20 confidence). {4.4.4, Cross-Chapter Box 4.1, 4.5.1, 8.5.2, 8.6.1, 9.2.3, 10.6.3} 21 22 Overall, long-term (2081–2100) future changes in regional monsoons like the South and Southeast Asian 23 monsoon are generally consistent across global (including high-resolution) and regional climate models and 24 supported by theoretical arguments. Uncertainties in simulating the observed characteristics of regional 25 monsoon precipitation are related to varying complexities of regional monsoon processes and their responses 26 to external forcing, internal variability, and deficiencies in representing monsoon warm rain processes, 27 organized tropical convection, heavy orographic rainfall and cloud–aerosol interactions. {8.3.2, 8.5.1, 10.3.3, 28 10.6.3} 29 30 31 [START BOX TS.13, FIGURE 1 HERE] 32 33 Box TS.13, Figure 1: Global and regional monsoons: past trends and projected changes. The intent of this figure 34 is to show changes in precipitation over regional monsoon domains in terms of observed past 35 trends, how greenhouse gases and aerosols relate to these changes, and in terms of future 36 projections in one intermediate emission scenario in the near, medium and long terms. (a) Global 37 (black contour) and regional monsoons (colour shaded) domains. The global monsoon (GM) is 38 defined as the area with local summer-minus-winter precipitation rate exceeding 2.5 mm day –1 39 (see Annex V). The regional monsoon domains are defined based on published literature and 40 expert judgement (see Annex V), and also accounting for the fact that the climatological summer 41 monsoon rainy season varies across the individual regions. Assessed regional monsoons are 42 South and Southeast Asia (SAsiaM, Jun-Jul-Aug-Sep), East Asia (EAsiaM, Jun-Jul-Aug), West 43 Africa (WAfriM, Jun-Jul-Aug-Sep), North America (NAmerM, Jul-Aug-Sep), South America 44 (SAmerM, Dec-Jan-Feb), Australia and Maritime Continent Monsoon (AusMCM, Dec-Jan- 45 Feb). Equatorial South America (EqSAmer) and South Africa (SAfri) regions are also shown, as 46 they receive unimodal summer seasonal rainfall although their qualification as monsoons is 47 subject to discussion. (b) Global and regional monsoons precipitation trends based on DAMIP 48 CMIP6 simulations with both natural and anthropogenic (ALL), GHG only (GHG), aerosols 49 only (AER) and natural only (NAT) radiative forcing. Weighted ensemble means are based on 50 nine CMIP6 models contributing to the MIP (with at least 3 members). Observed trends 51 computed from CRU GPCP, and APHRO (only for SAsiaM and EAsiaM) datasets are shown as 52 well. (c) Percentage change in projected seasonal mean precipitation over global and regional 53 monsoons domain in the near-term (2021–2040), mid-term (2041–2060), and long-term (2081– 54 2100) under SSP2-4.5 based on 24 CMIP6 models. {Figure 8.11, Figure 8.22} 55 56 [END BOX TS.13, FIGURE 1 HERE] 57 Do Not Cite, Quote or Distribute TS-84 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [END BOX TS.13 HERE] 2 3 4 TS.4.3 Regional Climate Change and Implications for Climate Extremes and Climatic Impact-Drivers 5 6 Current climate in all regions is already distinct from the climate of the early or mid-20th century with respect 7 to several climatic impact-drivers (CIDs), resulting in shifting magnitude, frequency, duration, seasonality, 8 and spatial extent of associated climate indices (high confidence). It is very likely that mean temperatures have 9 increased in all land regions and will continue to increase at rates greater than the global average (high 10 confidence). The frequency of heat and cold extremes have increased and decreased, respectively. These 11 changes are attributed to human influence in almost all regions (medium to high confidence) and will continue 12 through the 21st century (high confidence). In particular, extreme heat would exceed critical thresholds for 13 health, agriculture and other sectors more frequently by the mid 21st century with 2°C of global warming (high 14 confidence). Relative sea level rise is very likely to virtually certain (depending on the region) to continue 15 during the 21st century, contributing to increased coastal flooding in low-lying areas (high confidence) and 16 coastal erosion along most sandy coasts (high confidence). Sea level will continue to rise beyond 2100 (high 17 confidence). Every region of the world will experience concurrent changes in multiple CIDs by mid-century 18 (high confidence). Even for the current climate, climate change-induced shifts in CID distributions and event 19 probabilities, some of which have occurred over recent decades, are relevant for risk assessments.{Box TS.4, 20 11.9, 12.1, 12.2, 12.4, 12.5, Atlas.3–Atlas.11} 21 22 An overview of changes in regional CIDs (introduced in TS.1) is given in Table TS.5, which summarizes 23 multiple lines of evidence on regional climate change derived from observed trends, attribution of these trends 24 and future projections. The level of confidence and the amplitude in the projected direction of change in CIDs 25 at a given time-horizon depends on climate change mitigation efforts over the 21st century. It is evident from 26 Table TS.5 that many heat, cold, snow and ice, coastal and oceanic CID changes are projected with high 27 confidence in most regions starting from a global warming level (GWL) of 2°C, indicating worldwide 28 challenges. Changes in many other regional CIDs have higher confidence later in the 21st century or at higher 29 GWLs (high confidence), and another small subset are projected with high confidence for the 1.5°C GWL. 30 This section focuses on the 2°C GWL and mid-century time period because the signal emerges from natural 31 variability for a wider range of CIDs at this higher warming level. Figure TS.22 shows the geographical 32 location of regions belonging to one of five groups characterized by a specific combination of changing 33 climatic impact-drivers (CIDs). {Cross-Chapter Box 10.3, Box 11.1, 10.5, 11.1, 11.9, 12.1, 12.2, 12.4, 12.5} 34 35 36 [START TABLE TS.5 HERE] 37 38 Table TS.5: Summary of confidence for climatic impact-driver changes in each AR6 WG I reference region 39 (illustrated in Figure TS.25) across multiple lines of evidence: observed, attributed and projected 40 directional changes. The colours represent their projected aggregate characteristic changes for the 41 mid-21st century, considering scenarios RCP4.5, SSP2-4.5, SRES A1B, or above (RCP6.0, RCP8.5, 42 SSP3-7.0, SSP5-8.5, SRES A2), which approximately encompasses global warming levels of 2.0°C to 43 2.4°C. Arrows indicate medium to high confidence trends derived from observations, and asterisks 44 indicate medium and high confidence in attribution of observed changes. (North Africa is not an AR6 45 WGI reference region, but assessment here is based upon the African portion of the Mediterranean 46 reference region). {Tables 12.3–12.11 and Tables 11.4–11.21} 47 48 Do Not Cite, Quote or Distribute TS-85 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Do Not Cite, Quote or Distribute TS-86 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 [END TABLE TS.5 HERE] 2 3 4 [START FIGURE TS.22 HERE] 5 6 Figure TS.22: (Panel a): shows the geographical location of regions belonging to one of five groups characterized by 7 a specific combination of changing climatic impact-drivers (CIDs). The five groups are represented by 8 the five different colours, and the CID combinations associated with each group are represented in the 9 corresponding ‘fingerprint’ and text below the map. Each fingerprint comprises a set of CIDs 10 projected to change with high confidence in every region in the group, and a second set of CIDs, one 11 or more of which are projected to change in each region with high or medium confidence. The CID 12 combinations follow a progression from those becoming hotter and drier (group 1) to those becoming 13 hotter and wetter (group 5). In between (groups 2–-4), the CIDs that change include some becoming 14 drier and some wetter and always include a set of CIDs which are getting hotter. Tropical cyclones 15 and severe wind CID changes are represented on the map with black dots in the regions affected. 16 Regions affected by coastal CID changes are described by text on the map. The five groups are chosen 17 to provide a reasonable level of detail for each region-specific detail whilst not overwhelming the map 18 with a full summary all aspects of the assessment, which is available in Table TS.5. [Placeholder: This 19 summary is also represented visually in the Interactive Atlas.] The CID changes summarized in the 20 figure represent high and medium confidence changes projected if a level of 2°C of global warming is 21 attained around 2050. The bar chart in panel b) shows the numbers of regions where each CID is 22 increasing or decreasing with medium or high confidence for all land regions reported in the map of 23 panel a) and for the ocean regions. The regions coloured in the map comprise the WG I AR6 reference 24 regions, which include inhabited land areas and an additional non-continuous Pacific Islands region 25 labelled PAC. Definitions of the acronyms of the other regions are provided in Atlas.1 and the 26 Interactive Atlas. {Table TS.5, Figure TS.24} 27 28 [END FIGURE TS.22 HERE] 29 30 31 TS.4.3.1 Common Regional Changes in Climatic Impact-Drivers 32 33 Heat and cold: Changes in temperature-related CIDs such as mean temperatures, growing season length, and 34 extreme heat and frost have already occurred (high confidence), and many of these changes have been 35 attributed to human activities (medium confidence). Over all land regions with sufficient data (i.e., all except 36 Antarctica), observed changes in temperature have already clearly emerged outside the range of internal 37 variability, relative to 1850–1900 (Figure TS.24). In tropical regions, recent past temperature distributions 38 have already shifted to a range different to that of the early 20th century (high confidence) (TS.1.2.4). Most 39 land areas have very likely warmed by at least 0.1°C per decade since 1960, and faster in recent decades. On 40 regional-to-continental scales, trends of increased frequency of hot extremes and decrease of cold extremes 41 are generally consistent with the global-scale trends in mean temperature (high confidence). In a few regions, 42 trends are difficult to assess due to limited data availability. {2.3.1.1, 11.3, 11.9, 12.4, Atlas.3.1, Interactive 43 Atlas} 44 45 46 [START FIGURE TS.23 HERE] 47 48 Figure TS.23: Time period during which the signals of temperature change in observed data aggregated over 49 the reference regions emerged from the noise of annual variability in the respective aggregated 50 data, using a signal-to-noise ratio of 2 as the threshold for emergence. Emergence time is 51 calculated for two global observational datasets, (a) Berkeley Earth and (b) CRUTEM5. Regions in 52 the CRUTEM5 map are shaded grey when data are available over less than 50% of the area of the 53 region. {Figure Atlas.11, TS.1.2.4} 54 55 [END FIGURE TS.23 HERE] 56 57 Do Not Cite, Quote or Distribute TS-87 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Warming trends observed in recent decades are projected to continue over the 21st century and over most land 2 regions at a rate higher than the global average (high confidence). For given global warming levels, model 3 projections from CMIP6 show future regional warming changes that are similar to those projected by CMIP5. 4 However, projected regional warming in CMIP6 for given time periods and emissions scenarios has a wider 5 range with a higher upper limit compared to CMIP5 because of the higher climate sensitivity in some CMIP6 6 models and differences in the forcings. {Atlas} 7 8 Under RCP8.5/SSP5-8.5, it is likely that most land areas will experience further warming of at least 4°C 9 compared to a 1995–2014 baseline by the end of the 21st century, and in some areas significantly more. At 10 increasing warming levels, extreme heat will exceed critical thresholds for health, agriculture and other sectors 11 more frequently (high confidence), and it is likely that cold spells will become less frequent towards the end 12 of the century. For example, by the end of the 21st century, dangerous humid heat thresholds, such as the 13 NOAA Heat Index (HI) of 41°C, will be exceeded much more frequently under the SSP5-8.5 scenario than 14 under SSP1-2.6 and will affect many regions (high confidence). In many tropical regions, the number of days 15 per year where a HI of 41°C is exceeded would increase by more than 100 days relative to the recent past 16 under SSP5-8.5, while this increase will be limited to less than 50 days under SSP1-2.6 (high confidence) 17 (Figure TS.6). The number of days per year where temperature exceeds 35°C would increase by more than 150 18 days in many tropical areas by end of century for SSP5-8.5 scenario, such as the Amazon basin and South East 19 Asia under SSP5-8.5, while it is expected to increase by less than 60 days in these areas under SSP1-2.6 (except 20 for the Amazon Basin) (high confidence) (Figure TS.24). {4.6.1, 11.3, 11.9, 12.4, 12.5.2, Atlas} 21 22 23 [START FIGURE TS.24 HERE] 24 25 Figure TS.24: Projected change in the mean number of days per year with maximum temperature exceeding 26 35°C for CMIP5 (first column), CMIP6 (second column) and CORDEX (thirth column). The 27 map shows the median change in the number of days per year between the mid-century (2041–2060) 28 or end-century (2081–2100) and historical (1995–2014) periods for the CMIP5 and 29 CORDEX RCP8.5 and RCP2.6 and CMIP6 SSP5-8.5 and SSP1-2.6 scenarios ensembles. Stippling 30 indicates areas where less than 80% of the models agree on the sign of change. {Interactive Atlas} 31 32 [END FIGURE TS.24 HERE] 33 34 35 Wet and dry: Compared to the global scale, precipitation internal variability is stronger at the regional scale 36 while uncertainties in observations, models and external forcing are all larger. However, GHG forcing has 37 driven increased contrasts in precipitation amounts between wet and dry seasons and weather regimes over 38 tropical land areas (medium confidence), with a detectable precipitation increase in the northern high latitudes 39 (high confidence) (Box TS.6). The frequency and intensity of heavy precipitation events have increased over 40 a majority of land regions with good observational coverage (high confidence). A majority of land areas have 41 experienced decreases in available water in dry seasons due to human-induced climate change associated with 42 changes in evapotranspiration (medium confidence). Global hydrological models project a larger fraction of 43 land areas to be affected by an increase rather than by a decrease in river floods (medium confidence). Extreme 44 precipitation and pluvial flooding will increase in many regions around the world on almost all continents 45 (high confidence), but regional changes in river floods are more uncertain than changes in pluvial floods 46 because complex hydrological processes, including land cover and human water management are involved. 47 {Box 8.2, 8.2.2.1, 8.3.1, 10.4.1, 11.5, 11.6, 11.9, 12.4, 12.5.1, Atlas.3.1, Interactive Atlas} 48 49 Wind: Observed mean wind speed is decreasing over most land areas where observational coverage is high 50 (medium confidence). It is likely that the global proportion of major tropical cyclone (TC) intensities 51 (Categories 3–5) over the past four decades has increased. The proportion of intense TCs, average peak TC 52 wind speeds, and peak wind speeds of the most intense TCs will increase on the global scale with increasing 53 global warming (high confidence). {11.7.1} 54 55 Snow and ice: Many aspects of the cryosphere either have seen significant changes in the recent past or will 56 see them during the 21st century (high confidence). Glaciers will continue to shrink and permafrost to thaw in Do Not Cite, Quote or Distribute TS-88 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 all regions where they are present (high confidence). Also, it is virtually certain that snow cover will experience 2 a decline over most land regions during the 21st century, in terms of water equivalent, extent and annual 3 duration. There is high confidence that the global warming-induced earlier onset of spring snowmelt and 4 increased melting of glaciers have already contributed to seasonal changes in streamflow in high-latitude and 5 low-elevation mountain catchments. Nevertheless, it is very likely that some high-latitude regions will 6 experience an increase in winter snow water equivalent due to the effect of increased snowfall prevailing over 7 warming-induced increased snowmelt. {TS.2.5, Box 8.2, 8.2.2.1, 8.3.1, 9.4, 9.5.1, 9.5.2, 12.4, Atlas} 8 9 Coastal and oceanic: There is high confidence that SST will increase in all oceanic regions, excepting the 10 North Atlantic. Regional sea-level change has been the main driver of changes in extreme sea levels across the 11 quasi-global tide gauge network over the 20th century (high confidence). With the exception of a few regions 12 with substantial land uplift, relative sea-level rise is very likely to virtually certain (depending on the region) 13 to continue along the 21st century, contributing to increased coastal flooding in low-lying areas (high 14 confidence) and coastal erosion along most sandy coasts (high confidence) over the 21st century. In the open 15 ocean, acidification, changes in sea ice and deoxygenation have already emerged in many areas (high 16 confidence). Marine heatwaves (MWHs) are also expected to increase around the globe over the 21st century 17 (high confidence). {TS.2.4, Box 9.2, 9.2.1.1, 9.6, 9.6.4, 9.6.4.2, 12.4} 18 19 Other variables and concurrent CID changes: It is virtually certain that atmospheric CO2 and oceanic pH 20 will increase in all climate scenarios, until net zero CO2 emissions are achieved (TS.2.2). In nearly all regions, 21 there is low confidence in changes in hail, ice storms, severe storms, dust storms, heavy snowfall, and 22 avalanches, although this does not indicate that these CIDs will not be affected by climate change. For such 23 CIDs, observations are often short-term or lack homogeneity, and models often do not have sufficient 24 resolution or accurate parametrizations to adequately simulate them over climate change time scales. The 25 probability of compound events has increased in the past due to human-induced climate change and will likely 26 continue to increase with further global warming, including for concurrent heat waves and droughts, compound 27 flooding and the possibility of connected sectors experiencing multiple regional extreme events at the same 28 time (for example, in multiple breadbaskets) (high confidence). {5.3.4.2, 11.8, Box 11.3, Box 11.4, 12.4} 29 30 31 TS.4.3.2 Region-by-Region Changes in Climatic Impact-Drivers 32 33 This section provides a continental synthesis of changes in CIDs, some examples of which are presented in 34 Figure TS.25. 35 36 With 2°C global warming, and as early as the mid-21st century, a wide range of CIDs, particularly related to 37 the water cycle and storms, are expected to show simultaneous region-specific changes relative to recent past 38 with high or medium confidence. In a number of regions (Southern Africa, the Mediterranean, North Central 39 America, Western North America, the Amazon regions, South Western South America, and Australia), 40 increases in one or more of drought, aridity and fire weather (high confidence) will affect a wide range of 41 sectors, including agriculture, forestry, health and ecosystems. In another group of regions (Northwestern, 42 Central and Eastern North America, Arctic regions, Northwestern South America, Northern and Central 43 Western Europe, Siberia, Central, South and East Asia, Southern Australia and New Zealand), decreases in 44 snow and ice or increases in pluvial/river flooding (high confidence) will affect sectors such as winter tourism, 45 energy production, river transportation, and infrastructure. {11.9, 12.3, 12.4, 12.5, Table 12.2} 46 47 48 [START FIGURE TS.25 HERE] 49 50 Figure TS.25: Distribution of projected changes in selected climatic impact-driver indices for selected regions 51 for CMIP6, CMIP5 and CORDEX model ensembles. Different indices are shown for different 52 region: for east Europe and north Asia the mean number of days per year with maximum temperature 53 exceeding 35°C, for Central America and the Caribbean, and the Arabian peninsula, western, southern 54 and easten Asia the mean number of days per year with the NOAA Heat Index exceeding 41°C, for 55 Australasia, East Asia and Russia far East the average shoreline position change, for South America, Do Not Cite, Quote or Distribute TS-89 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Europe and Africa the mean change in 1-in-100-years river discharge per unit catchment area (m3 s–1 2 km–2), and for North America the median change in the number of days with snow water equivalent 3 (SWE) over 100 mm. For each box plot the changes or the climatological values are reported respect 4 to, or compared to, the recent past (1995–2014) period for 1.5 oC, 2oC and 4oC global warming levels 5 and for mid-century (2041–2060) or end-century (2081–2100) periods for the CMIP5 and 6 CORDEX RCP8.5 and RCP2.6 and CMIP6 SSP5-8.5 and SSP1-2.6 scenarios ensembles {Figure 7 12.5, Figure 12.6, Figure 12.9, Figure 12.SM.1, Figure 12.SM.2, Figure 12.SM.6}. 8 [END FIGURE TS.25 HERE] 9 10 11 TS.4.3.2.1 Africa 12 13 Additional regional changes in Africa, besides those described in TS.4.3.1, include a projected decrease in 14 total precipitation in the northernmost and southernmost regions (high confidence), with West and East Africa 15 each having a west-to-east pattern of decreasing-to-increasing precipitation (medium confidence). Increases in 16 heavy precipitation that can lead to pluvial floods (high confidence) are projected for most African regions, 17 even as increasing dry CIDs (aridity, hydrological, agricultural and ecological droughts, fire weather) are 18 projected in the western part of West Africa, North Africa, South Africa and the Mediterranean regions 19 (medium to high confidence). {8.4, 11.3, 11.6, 11.9, 12.4, Atlas.4} 20 21 In addition to the main changes summarized above and in TS.4.3.1, additional details per CID are given below. 22 23 Heat and cold: Observed and projected increases in mean temperature and a shift toward heat extreme 24 characteristics are broadly similar to the generic pattern described in TS.4.3.1. {2.3.1.1.2, 11.3, 11.9, 12.4.1.1, 25 Atlas.4.2, Atlas.4.4} 26 27 Wet and dry: Mean precipitation changes have been observed over Africa, but the historical trends are not 28 spatially coherent (high confidence). Northern East Africa, Eastern Southern Africa and Central Africa have 29 experienced a decline in rainfall since about 1980 and parts of West Africa an increase (high confidence). 30 Increases in the frequency and/or the intensity of heavy rainfall have been observed in Eastern and Western 31 Southern Africa, and the eastern Mediterranean region (medium confidence). Increasing trends in river flood 32 occurrences can be identified beyond 1980 in Eastern and Western Southern Africa (medium confidence) and 33 West Africa (high confidence). However, Northern Africa and Western Southern Africa are likely to have a 34 reduction in precipitation. Over West Africa, rainfall is projected to decrease in the Western Sahel subregion 35 and increase along the Guinea Coast subregion (medium confidence). Rainfall is likely to reduce over the 36 western part of East Africa but increase in the eastern part of the region (medium confidence). {8.3.1.6, 11.4, 37 11.9, 12.4.1.2, Atlas.4.2, Atlas.4.5, Interactive Atlas} 38 39 Precipitation declines and aridity trends in West Africa, Central Africa, Southern Africa and the Mediterranean 40 co-occur with trends towards increased agricultural and ecological droughts in the same regions (medium 41 confidence). Trends towards increased hydrological droughts have been observed in the Mediterranean (high 42 confidence) and West Africa (medium confidence). These trends correspond with projected regional increases 43 in aridity and fire weather conditions (high confidence). {8.3.1.6, 8.4.1.6, 11.6, 11.9, 12.4.1.2} 44 45 Wind: Mean wind, extreme winds and the wind energy potential in North Africa and the Mediterranean are 46 projected to decrease across all scenarios (high confidence). Over West Africa and South Africa, a future 47 significant increase in wind speed and wind energy potential is projected (medium confidence). There is a 48 projected decrease in the frequency of tropical cyclones making landfall over Madagascar, Eastern Southern 49 Africa and East Africa (medium confidence). {12.4.1.3} 50 51 Snow and ice: There is high confidence that African glaciers and snow have very significantly decreased in 52 the last decades and that this trend will continue in the 21st century. {12.4.1.4} 53 54 Coastal and oceanic: Relative sea level has increased at a higher rate than GMSL around Africa over the last 55 3 decades. The present day 1-in-100-years Extreme Total Water Level (ETWL) is between 0.1 m and 1.2 m Do Not Cite, Quote or Distribute TS-90 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 around Africa, with values around 1 m or above along the South West, South East and Central East coasts. 2 Satellite-derived shoreline retreat rates up to 1 m yr–1 have been observed around the continent from 1984 to 3 2015, except in Southeastern Africa, which has experienced a shoreline progradation (growth) rate of 0.1 m 4 yr–1 over the same period. {12.4.1.5} 5 TS.4.3.2.2 Asia 6 Due to the high climatological and geographical heterogeneity of Asia, the assessment findings below are 7 summarised over five sub-continental areas comprising one or more of the AR6 WG I reference regions (Box 8 TS.12): East Asia (EAS+ECA), North Asia (WSB+ESB+RFE), South Asia (SAS+TIB), Southeast Asia (SEA) 9 and Southwest Asia (ARP+WCA). 10 11 Additional regional changes in Asia, besides those features described in TS.4.3.1, include historical trends of 12 annual precipitation that show considerable regional differences (high confidence). East Asian Monsoon 13 precipitation has changed, with drying in the north and wetting in the south since the 1950s, and annual mean 14 precipitation totals very likely have increased over most territories of North Asia since the mid-1970s (high 15 confidence). South Asian summer monsoon precipitation decreased over several areas since the mid-20th 16 century (high confidence) but is likely to increase during the 21st century, with enhanced interannual 17 variability. 18 19 Increases in precipitation and rivers floods are projected over much of Asia: in the annual mean precipitation 20 in East, North, South and Southeast Asia (high confidence); for extremes in East, South, West Central, North 21 and Southeast Asia (high confidence) and Arabian Peninsula (medium confidence); and for river floods in East, 22 South and Southeast Asia and East Siberia (medium confidence). Aridity in East and West Central Asia is 23 projected to increase, especially beyond the middle of the 21st century and global warming levels beyond 2℃, 24 (medium confidence). Fire weather seasons are projected to lengthen and intensify everywhere except 25 Southeast Asia, Tibetan Plateau and Arabian Peninsula (medium confidence). 26 27 There is a large uncertainty in the future continuation of observed decreasing trends in surface wind speeds in 28 Asia (high confidence), with medium confidence that mean wind speeds will decrease in North Asia, East Asia 29 and Tibetan Plateau and that tropical cyclones will have decreasing frequency and increasing intensity overall 30 in Southeast and East Asia. 31 32 Over North Asia, increases in permafrost temperature and its thawing have been observed over recent decades 33 (high confidence). Future projections indicate continuing decline in seasonal snow duration, glacial mass, and 34 permafrost area by mid-century (high confidence). Snow-covered areas and snow volumes will decrease in 35 most regions of the Hindu Kush Himalaya during the 21st century and snowline elevations will rise (high 36 confidence) and glacier volumes are likely to decline with greater mass loss in higher CO2 emissions scenarios. 37 Heavy snowfall is increasing in East Asia and North Asia (medium confidence) but with limited evidence on 38 future changes in hail and snow avalanches. 39 40 {2.3, 8.3, 8.4, 9.5, 9.6, 10.6, Box 10.4, 11.4, 11.5, 11.7, 11.9, 12.4.2, Atlas.3.1, Atlas.5, Atlas.5.2, Atlas.5.3, 41 Atlas.5.4, Atlas.5.5, Box TS.13} 42 43 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 44 45 Heat and cold: Over all regions of Asia, observed and projected increases in mean temperature and a shift 46 toward heat extreme characteristics are broadly similar to the generic pattern described in TS.4.3.1. Over 47 Southeast Asia annual mean surface temperature will likely increase by a slightly smaller amount than the 48 global average. {Atlas.5.4.4} 49 50 Wet and dry: Over East Asia, historical trends of annual precipitation show considerable regional differences 51 but with increases over northwest China and South Korea (high confidence). Daily precipitation extremes have 52 increased over part of the region (high confidence). Extreme hydrological drought frequency has increased in 53 a region extending from southwest to northeast China, with projected increases of agricultural and ecological 54 drought for 4°C GWL and fire weather for 2°C and above (medium confidence). {8.3.2, 8.4.2, 11.4.4, 11.4.5, Do Not Cite, Quote or Distribute TS-91 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 11.9, 12.4.2.2, Atlas.5.1.2} 2 3 Over North Asia, annual mean precipitation totals have very likely increased, causing more intense flooding 4 events, and there is medium confidence that the number of dry days has decreased. Concurrently total soil 5 moisture is projected to decline extensively (medium confidence). {8.3.1.3, 8.4.1.6, 11.4.5, 11.5.2, 11.5.5, 6 12.4.2.2, Atlas.5.2.2} 7 8 Over South Asia, the summer monsoon precipitation decreased over several areas since the mid-20th century 9 (high confidence), while it increased in parts of the western HKH and decreased over eastern-central HKH 10 (medium confidence) (Box TS.12). The frequency of heavy precipitation and flood events has increased over 11 several areas during the last few decades (medium confidence). {8.3.1.3, 8.3.2.4.1, 8.4.1.5, 8.4.2.4.1, 10.6.3.3, 12 10.6.3.5, 10.6.3.6, 10.6.3.8, Cross-Chapter Box 10.4, 11.4.1, 11.4.2, 11.4.5, 11.5.5, 12.4.2.2, Box 10.4, Atlas 13 5.3.2} 14 15 Over Southeast Asia, mean precipitation trends are not spatially coherent or consistent across datasets and 16 seasons (high confidence). Most of the region has experienced an increase in rainfall intensity but with a 17 reduced number of wet days (medium confidence). Rainfall is projected to increase in the northern parts of 18 Southeast Asia and decrease in areas in the Maritime Continent (medium confidence). {8.4.1, 11.4.2, 11.5.5, 19 11.9, 12.4.2.2, Atlas.3.1, Atlas.5.4.2, Atlas.5.4.4} 20 21 Over Southwest Asia, an observed annual precipitation decline over the Arabian Peninsula since the 1980s of 22 6.3 mm per decade is contrasted with observed increases between 1.3 mm and 4.8 mm per decade during 23 1960–2013 over the elevated part of eastern West Central Asia (very high confidence), along with an increase 24 of the frequency and intensity of extreme precipitation. {Figure 8.19, Figure 8.20, 8.3.1.6, 8.4.1.6, Table 25 11.2A, 11.9, 12.4.2.2, Atlas.5.5} 26 27 Wind: Over East Asia, the terrestrial near-surface wind speed has decreased and is projected to decrease 28 further in the future (medium confidence). Since the mid 1980’s, there has been an increase in the number and 29 intensification rate of intense TCs (medium confidence), with a significant northwestward shift in tracks and a 30 northward shift in their average latitude, increasing exposure over East China, the Korean Peninsula and the 31 Japanese Archipelago (medium confidence). {11.7.1, 12.4.2.3} 32 33 Over North Asia there is medium confidence for a decreasing trend in wind speed during 1979–2018 and for 34 projected continuing decreases of terrestrial near-surface wind speed (medium confidence). {2.3.1.4.4, 35 12.4.2.3} 36 37 Over Southeast Asia, although there is no significant long-term trend in the number of TC, fewer but more 38 extreme TCs have affected the Philippines during 1951–2013. {11.7.4, 12.4.2.4} 39 40 Snow and ice: Over East Asia, decreases have been observed in the frequency and increases in the mean 41 intensity of snowfall in north-western, north-eastern and south-eastern China and the eastern Tibetan Plateau 42 since the 1960s. Heavy snowfall is projected to occur more frequently in some parts of Japan (medium 43 confidence). {12.4.2.4, Atlas.5.1.2} 44 45 Over North Asia, seasonal snow duration and extent have decreased in recent decades (high confidence), and 46 maximum snow depth likely has increased since the mid-1970s, particularly over south of the Russian-Far- 47 East. {2.3.2.5, 8.3.1.7.2, 9.5, 12.4.2.4, Atlas.5.2, Atlas.5.4} 48 49 Over South Asia, snow-cover has reduced over most of the HKH since the early 21st century, and glaciers 50 have thinned, retreated, and lost mass since the 1970s (high confidence) although the Karakoram glaciers have 51 either slightly gained mass or are in an approximately balanced state (medium confidence). {8.3.1.7.1, Cross- 52 Chapter Box 10.4} 53 54 Over Southwest Asia, mountain permafrost degradation at high altitudes has increased the instability of 55 mountain slopes in the past decade (medium confidence). More than 60% of glacier mass in the Caucasus is Do Not Cite, Quote or Distribute TS-92 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 projected to disappear under RCP8.5 emissions by the end of the 21st century (medium confidence). {9.5.1, 2 9.5.3, 12.4.2.4} 3 4 Coastal and oceanic: Over the last three decades, relative sea level has increased at a rate higher than GMSL 5 around Asia (high confidence). Gross coastal area loss and shoreline retreat has been observed over 1984– 6 2015, but with localized shoreline progradation in the Russian Far East, East and Southeast Asia. {12.4.2.5} 7 8 Projections show that regional-mean sea level continues to rise (high confidence), ranging from 0.4–0.5 m 9 under SSP1-2.6 to 0.8–1.0 m under SSP5-8.5 for 2081–2100 relative to 1995–2014 (median values). This will 10 contribute to more frequent coastal flooding and higher Extreme Total Water Level (ETWL) in low-lying areas 11 and coastal erosion along sandy beaches (high confidence). There is high confidence that compound effects of 12 climate change, land subsidence, and human factors will lead to higher flood levels and prolonged inundation 13 in the Mekong Delta and other Asian coasts. {9.6.1, 9.6.3, 12.4.2.5} 14 15 16 TS.4.3.2.3 Australasia 17 18 Additional regional changes in Australasia, besides those features described in TS.4.3.1, include a significant 19 decrease in April to October rainfall in southwest Western Australia, observed from 1910 to 2019 and 20 attributable to human influence (high confidence), which is very likely to continue in future. Agricultural and 21 ecological and hydrological droughts have increased over southern Australia (medium confidence), and 22 meteorological droughts have decreased over northern and central Australia (medium confidence). Relative 23 sea level has increased over the period 1993–2018 at a rate higher than GMSL around Australasia (high 24 confidence). Sandy shorelines have retreated around the region, except in southern Australia, where a shoreline 25 progradation rate of 0.1 m yr–1 has been observed. In the future, heavy precipitation and pluvial flooding are 26 very likely to increase over northern Australia and central Australia, and they are likely to increase elsewhere 27 in Australasia for global warming levels (GWLs) exceeding 2°C and with medium confidence for a 2°C GWL. 28 Agricultural and ecological droughts are projected to increase in southern and eastern Australia (medium 29 confidence) for a 2°C GWL. Fire weather is projected to increase throughout Australia (high confidence) and 30 New Zealand (medium confidence). Snowfall is expected to decrease throughout the region at high altitudes 31 in both Australia (high confidence) and New Zealand (medium confidence), with glaciers receding in New 32 Zealand (high confidence). {11.4, Table 11.6, 12.3, 12.4.3, Atlas.6.4, Atlas.6.5} 33 34 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 35 36 Heat and cold: Observed and projected increases in mean temperature and a shift toward heat extreme 37 characteristics are broadly similar to the generic pattern described in TS.4.3.1. {11.9, 12.4.3.1, Atlas.6} 38 39 Wet and dry: There is medium confidence that heavy precipitation has increased in northern Australia since 40 1950. Annual mean precipitation is projected to increase in the south and west of New Zealand (medium 41 confidence) and is projected to decrease in southwest western Australia (high confidence), eastern Australia 42 (medium confidence), and in the north and east of New Zealand (medium confidence) for a GWL of 2°C. There 43 is medium confidence that river flooding will increase in New Zealand and Australia, with higher increases in 44 northern Australia. Aridity is projected to increase with medium confidence in southern Australia (high 45 confidence in southwest Western Australia), eastern Australia (medium confidence) and in the north and east 46 of New Zealand (medium confidence) for GWLs around 2°C. {11.4, Table 11.6, 11.9, 12.4.3.2, Atlas.6.2} 47 48 Wind: Mean wind speeds are projected to increase in parts of northeastern Australia (medium confidence) by 49 the end of the 21st century, under high CO2 emissions scenarios. TCs in north eastern and north Australia are 50 projected to decrease in number (high confidence) but increase in intensity except for ‘east coast lows’ (low 51 confidence). {12.4.3.3} 52 53 Snow and ice: Observations in Australia show that the snow season length has decreased by 5% in the last 54 five decades. Furthermore, the date of peak snowfall in Australia has advanced by 11 days over the last 5 Do Not Cite, Quote or Distribute TS-93 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 decades. Glacier ice volume in New Zealand has decreased by 33% from 1977 to 2018. {12.4.3.4, Atlas.6.2} 2 3 Coastal and oceanic: Observed changes in marine heat waves (MHWs) over the 20th century in the region 4 show an increase in their occurrence frequency, except along the south east coast of New Zealand, an increase 5 in duration per event, and the total number of MHW days per decade, with the change being stronger in the 6 Tasman Sea than elsewhere. The present day 1-in-100-year ETWL is between 0.5–2.5 m around most of 7 Australia, except the northwestern coast where 1-in-100-year ETWL can be as high as 6 m–7 m. {Box 9.1, 8 12.3.1.5, 12.4.3.5} 9 10 11 TS.4.3.2.4 Central and South America 12 13 Additional regional changes in Central and South America, besides those features described in TS.4.3.1, 14 include increases in mean and extreme precipitation in Southeastern South America since the 1960s (high 15 confidence) (TS.4.2.3). Decreasing trends in mean precipitation and increasing trends in agricultural and 16 ecological drought are observed over Northeastern South America (medium confidence). The intensity and 17 frequency of extreme precipitation and pluvial floods is projected to increase over Southeastern South 18 America, Southern South America, Northern South America, South American Monsoon and Northeastern 19 South America (medium confidence) for a 2°C GWL and above. Increases of agricultural and ecological 20 drought are projected in South America Monsoon and Southern South America, and fire weather is projected 21 to increase over several regions (Northern South America, the South American Monsoon, Northeastern South 22 America and Southwestern South America) (high confidence). {8.3, 8.4, 11.3, 11.4, 11.9, Table 11.13, Table 23 11.14, Table 11.15, 12.4.4.2, Atlas.7.1, Atlas.7.2} 24 25 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 26 27 Heat and cold: Observed and projected increases in mean temperature and a shift toward heat extreme 28 characteristics are broadly similar to the generic pattern described in TS.4.3.1. {11.3.2, 11.3.5, Table 11.13, 29 12.4.4.1, Atlas.7.1.2, Atlas.7.2.2, Atlas.7.2.4} 30 31 Wet and dry: Mean precipitation is projected to change in a dipole pattern with increases in Northwestern 32 and Southeastern South America and decreases in Northeastern and Southwestern South America (high 33 confidence) with further decreases in Northern South America and South Central America (medium 34 confidence). In Northern South America and South Central America, aridity and agricultural and ecological 35 droughts are increasing with medium confidence. Fire weather is projected to increase over Southern Central 36 America and Southern South America with medium confidence. {8.3.1.3, 8.4.2.4.5, 11.4.2, 11.9, Table 11.14, 37 Table 11.15, 12.4.4.2, Atlas.7.2.2, Atlas.7.2.4} 38 39 Wind: Climate projections indicate an increase in mean wind speed and in wind power potential over the 40 Amazonian region (Northern South America, South American Monsoon, Northeastern South America) 41 (medium confidence). {12.4.4.3} 42 43 Snow and ice: Glacier volume loss and permafrost thawing will likely continue in the Andes Cordillera under 44 all climate scenarios, causing important reductions in river flow and potentially high-magnitude glacial lake 45 outburst floods. {9.5.1.1, 12.4.4.4} 46 47 Coastal and oceanic: Around Central and South America, relative sea level has increased at a higher rate than 48 GMSL in the South Atlantic and the subtropical North Atlantic, and at a rate lower than GMSL in the East 49 Pacific over the last 3 decades. The present day 1-in-100-years ETWL is highest in Southern and Southwestern 50 South America subregions, where it can be as large as 5 to 6 m. Satellite observations for 1984–2015 show 51 shoreline retreat rates along the sandy coasts of Southern Central America, Southeastern South America and 52 Southern South America, while shoreline progradation rates have been observed in Northwestern South 53 America and Northern South America. Over the period 1982–2016, the coastlines experienced at least one 54 MHW per year, and more along the Pacific coast of North Central America and the Atlantic coast of Do Not Cite, Quote or Distribute TS-94 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Southeastern South America. {12.4.4.5} 2 3 4 TS.4.3.2.5 Europe 5 6 Additional regional changes in Europe, besides those features described in TS.4.3.1, include observed 7 increases in pluvial flooding in northern Europe and hydrological and agricultural/ecological droughts in the 8 Mediterranean (high confidence), which have been attributed to human contribution with high and medium 9 confidence, respectively. Increased mean precipitation amounts at high latitudes in boreal winter and reduced 10 summer precipitation in southern Europe are projected starting from a 2°C GWL (high confidence). Aridity, 11 agricultural and hydrological droughts and fire weather conditions will increase in the Mediterranean region 12 starting from 2°C GWL (high confidence). Pluvial flooding will increase everywhere with high confidence 13 except for medium confidence in the Mediterranean; in Western and Central Europe this also applies to river 14 flooding starting from a 2°C GWL (high confidence). Most periglacial processes in Northern Europe are 15 projected to disappear by the end of the 21st century, even for a low warming scenario (medium confidence). 16 {8.3, 11.3, 11.9, 12.4.5, 12.5.2, Atlas.8.2, Atlas.8.4} 17 18 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 19 20 Heat and cold: Observed and projected increases in mean temperature and a shift toward heat extreme 21 characteristics are broadly similar to the generic pattern described in TS.4.3.1. {11.3, 11.9, 12.4.5.1, 12.5.2, 22 Atlas.8.2, Atlas.8.4} 23 24 Wet and dry: There is medium confidence that annual mean precipitation in Northern Europe, West and 25 Central Europe and Eastern Europe has increased since the early 20th century and high confidence for extreme 26 precipitation. In the European Mediterranean, the magnitude and sign of observed land precipitation trends 27 depend on time period and exact study region (medium confidence). There is medium confidence that river 28 floods will decrease in Northern, Eastern and Southern Europe for high warming levels. {8.3.1.3, 11.3, 11.9, 29 12.4.5.2, Atlas.8.2, Atlas.8.4} 30 31 Wind: Mean wind speed over land has decreased (medium confidence), but the role of human-induced climate 32 change has not been established. There is high confidence that mean wind speeds will decrease in 33 Mediterranean areas and medium confidence of such decreases in Northern Europe for GWLs exceeding 2°C. 34 The frequency of Medicanes (tropical-like cyclones in the Mediterranean) is projected to decrease (medium 35 confidence). {11.9, 12.4.5.3} 36 37 Snow and ice: In the Alps, snow cover will decrease below elevations of 1500–2000 m throughout the 21st 38 century (high confidence). A reduction of glacier ice volume is projected in the European Alps and Scandinavia 39 with high confidence and with medium confidence for the timing and mass change rates. {12.4.5.4, 9.5.2} 40 41 Coastal and oceanic: Over the last three decades, relative sea level has increased at a lower rate than GMSL 42 in the sub-polar North Atlantic coasts of Europe. The present day 1-in-100-years ETWL is between 0.5–1.5 m 43 in the Mediterranean basin and 2.5–5.0 m in the western Atlantic European coasts, around the United Kingdom 44 and along the North Sea coast, and lower at 1.5–2.5 m along the Baltic Sea coast. Satellite-derived shoreline 45 change estimates over 1984–2015 indicate shoreline retreat rates of around 0.5 m yr–1 along the sandy coasts 46 of Central Europe and the Mediterranean and more or less stable shorelines in Northern Europe. Over the 47 period 1982–2016, the coastlines of Europe experienced on average more than 2.0 MHW per year, with the 48 eastern Mediterranean and Scandinavia experiencing 2.5–3 MHWs per year. {12.4.5.5} 49 50 51 TS.4.3.2.6 North America 52 53 Additional regional changes in North America, besides those features described in TS.4.3.1, include changes 54 in North American wet and dry CIDs, which are largely organized by the northeast (more wet) to southwest Do Not Cite, Quote or Distribute TS-95 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 (more dry) pattern of mean precipitation change, although heavy precipitation increases are widespread (high 2 confidence). Increasing evaporative demand will expand agricultural and ecological drought and fire weather 3 (particularly in summertime) in Central North America, Western North America and North Central America 4 (from medium to high confidence). Severe wind storms, tropical cyclones, and dust storms in North America 5 are shifting toward more extreme characteristics (medium confidence), and both observations and projections 6 point to strong changes in the seasonal and geographic range of snow and ice conditions in the coming decades 7 (very high confidence). General findings for relative sea level, coastal flooding and erosion will not apply for 8 areas with substantial land uplift around the Hudson Bay and Southern Alaska. {8.4, 11.4, 11.5, 11.7, 11.9, 9 12.4, Atlas.9.4} 10 11 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 12 13 Heat and cold: Observed and projected increases in mean temperature and a shift toward heat extreme 14 characteristics are broadly similar to the generic pattern described in TS.4.3.1. {11.3, 11.9, 12.4.6.1, Atlas.9.2, 15 Atlas.9.4} 16 17 Wet and dry: Annual precipitation increased over parts of Eastern and Central North America during 1960– 18 2015 (high confidence) and has decreased in parts of southwestern United States and northwestern Mexico 19 (medium confidence). River floods are projected to increase for all North American regions other than north 20 Central America (medium confidence). {8.4.2.4, 11.4, 11.5, 11.9, 12.4.6.2, Atlas.9.2, Atlas.9.4} 21 22 Agricultural and ecological drought increases have been observed in Western North America (medium 23 confidence), and aridity is projected to increase in the southwestern United States and Northern Central 24 America, with lower summer soil moisture across much of the continental interior (medium confidence). 25 {8.4.1, 11.6.2, 12.4.6.2} 26 27 Wind: Projections indicate a greater number of the most intense TCs, with slower translation speeds and higher 28 rainfall potential for Mexico’s Pacific Coast, the Gulf Coast and the United States East Coast (medium 29 confidence). Mean wind speed and wind power potential are projected to decrease in Western North America 30 (high confidence), with differences between global and regional models lending low confidence elsewhere. 31 {11.4, 11.7, 12.4.6.3} 32 33 Snow and ice: It is very likely that some high-latitude regions will experience an increase in winter snow water 34 equivalent, due to the snowfall increase prevailing over the warming trend. At sustained GWLs between 3℃ 35 and 5℃, nearly all glacial mass in Western Canada and Western North America will disappear (medium 36 confidence). {9.5.1, 9.5.3, 12.4.6.4, Atlas.9.4} 37 38 Coastal and oceanic: Around North America, relative sea level has increased over the last 3 decades at a rate 39 lower than GMSL in the subpolar North Atlantic and in the East Pacific, while it has increased at a rate higher 40 than GMSL in the subtropical North Atlantic. Observations indicate that episodic coastal flooding is increasing 41 along many coastlines in North America. Shoreline retreat rates of around 1 m yr –1 have been observed during 42 1984–2015 along the sandy coasts of Northwestern North America and Northern Central America, while 43 portions of the United States Gulf Coast have seen a retreat rate approaching 2.5 m yr –1. Sandy shorelines 44 along Eastern North America and Western North America have remained more or less stable during 1984– 45 2014, but a shoreline progradation rate of around 0.5 m yr–1 has been observed in Northeastern North America. 46 {12.4.6.5} 47 48 49 TS.4.3.2.7 Small Islands 50 51 Additional regional changes in Small Islands, besides those features described in TS.4.3.1, include a likely 52 decrease in rainfall during boreal summer in the Caribbean and in some parts of the Pacific islands poleward 53 of 20° latitude in both the Northern and Southern Hemispheres. These drying trends will likely continue in 54 coming decades. Fewer but more intense tropical cyclones are projected starting from a 2°C GWL (medium Do Not Cite, Quote or Distribute TS-96 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 confidence). {9.6, 11.3, 11.4, 11.7, 11.9, 12.4.7, Atlas.10.2, Atlas.10.4, Cross-Chapter Box Atlas.2} 2 3 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 4 5 Heat and cold: It is very likely that most Small Islands have warmed over the period of instrumental records, 6 and continued temperature increases in the 21st century will further increase heat stress in these regions. 7 {11.3.2, 11.9, 12.4.7.1, Atlas.10.2, Atlas.10.4, Cross-Chapter Box Atlas.2} 8 9 Wet and dry: Observed and projected rainfall trends vary spatially across the Small Islands. Higher 10 evapotranspiration under a warming climate can partially offset future increases or amplify future reductions 11 in rainfall resulting in increased aridity as well as more severe agricultural and ecological drought in the Small 12 Islands (medium confidence). {11.4.2, 11.9, 12.4.7.2, Atlas.10.2, Atlas.10.4, Cross-Chapter Box Atlas.1} 13 14 Wind: Global changes indicate that Small Islands will face fewer but more intense TCs, with spatial 15 inconsistency in projections given poleward shifts in TC tracks (medium confidence). {11.7.1.2, 11.7.1.5, 16 12.4.7.3} 17 18 Coastal and oceanic: Continued relative sea level rise is very likely in the ocean around Small Islands and, 19 along with storm surges and waves, will exacerbate coastal inundation with the potential to increase saltwater 20 intrusion into aquifers in small islands. Shoreline retreat is projected along sandy coasts of most small islands 21 (high confidence). {9.6.3.3, 12.4.7.4, Cross-Chapter Box Atlas.1} 22 23 24 TS.4.3.2.8 Polar 25 26 It is virtually certain that surface warming in the Arctic will continue to be more pronounced than the global 27 average warming over the 21st century. An intensification of the polar water cycle will increase mean 28 precipitation, with precipitation intensity becoming stronger and more likely to be rainfall rather than snowfall 29 (high confidence). Permafrost warming, loss of seasonal snow cover, and glacier melt will be widespread (high 30 confidence). There is high confidence that both the Greenland and Antarctic Ice Sheets have lost mass since 31 1992 and will continue to lose mass throughout this century under all emissions scenarios. Relative sea level 32 and coastal flooding are projected to increase in areas other than regions with substantial land uplift (medium 33 confidence). {2.3, 3.4, 4.3, 4.5, 7.4, 8.2, 8.4, Box 8.2, 9.5, 12.4.9, Atlas.11.1, Atlas.11.2} 34 35 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 36 37 Heat and cold: Changes in Antarctica showed larger spatial variability, with very likely warming in the 38 Antarctic Peninsula since the 1950s and no overall trend in East Antarctica. Less warming and weaker polar 39 amplification are projected as very likely over the Antarctic than in the Arctic, with a weak polar amplification 40 projected as very likely by the end of the 21st century. {4.3.1, 4.5.1, 7.4.4, 12.4.9.1, Atlas.11.1, Atlas.11.2} 41 Wet and dry: Recent decades have seen a general decrease in Arctic aridity (high confidence), with increased 42 moisture transport leading to higher precipitation, humidity and streamflow and a corresponding decrease in 43 dry days. Antarctic precipitation showed a positive trend during the 20th century. The water cycle is projected 44 to intensify in both polar regions, leading to higher precipitation totals (and a shift to more heavy precipitation) 45 and higher fraction of precipitation falling as rain. In the Arctic, this will result in higher river flood potential 46 and earlier meltwater flooding, altering seasonal characteristics of flooding (high confidence). A lengthening 47 of the fire season (medium confidence) and encroachment of fire regimes into tundra regions (high confidence) 48 are projected. {8.2.3, 8.4.1, Box 8.2, 9.4.1, 9.4.2, 12.4.9.2, Atlas.11.1, Atlas.11.2} 49 50 Wind: There is medium confidence in mean wind decrease over the Russian Arctic and Arctic Northeast North 51 America, but low confidence of changes in other Arctic regions and Antarctica. {12.4.9.3} 52 53 Snow and ice: Reductions in spring snow cover extent have occurred across the Northern Hemisphere since 54 at least 1978 (very high confidence). Permafrost warming and thawing have been widespread in the Arctic Do Not Cite, Quote or Distribute TS-97 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 since the 1980s (high confidence), causing strong heterogeneity in surface conditions. There is high confidence 2 in future glacier and ice sheet loss, permafrost warming, decreasing permafrost extent and decreasing seasonal 3 duration and extent of snow cover in the Arctic. Decline in seasonal sea ice coverage along the majority of the 4 Arctic coastline in recent decades is projected to continue, contributing to an increase in coastal hazards 5 (including open water storm surge, coastal erosion and flooding). {2.3.2, 3.4.2, 3.4.3, 9.4.1, 9.4.2, 9.5, 12.4.6, 6 12.4.9, Atlas.11.2} 7 8 Coastal and oceanic: Higher sea levels contribute to high confidence for projected increases of Arctic coastal 9 flooding and higher coastal erosion (aided by sea ice loss) (medium confidence), with lower confidence for 10 those regions with substantial land uplift (Arctic Northeast North America and Greenland). {12.4.9.5} 11 12 13 TS.4.3.2.9 Ocean 14 15 The Indian Ocean, western equatorial Pacific Ocean and western boundary currents have warmed faster than 16 the global average (very high confidence), with the largest changes in the frequency of marine heatwaves 17 (MHWs) projected in the western tropical Pacific and the Arctic Ocean (medium confidence). The Pacific and 18 Southern Ocean are projected to freshen and the Atlantic to become more saline (medium confidence). 19 Anthropogenic warming is very likely to further decrease ocean oxygen concentrations, and this deoxygenation 20 is expected to persist for thousands of years (medium confidence). Arctic sea ice losses are projected to 21 continue, leading to a practically ice-free Arctic in September by the end of the 21st century under high CO2 22 emissions scenarios (high confidence). {2.3, 5.3, 9.2, 9.3, Box 9.2, 12.4.8} 23 24 In addition to the main changes summarized above and in TS.4.3.1, further details are given below. 25 26 Ocean surface temperature: The Southern Ocean, the eastern equatorial Pacific, and the North Atlantic 27 Ocean have warmed more slowly than the global average or slightly cooled. Global warming of 2°C above 28 1850–1900 levels would result in the exceedance of numerous hazard thresholds for pathogens, seagrasses, 29 mangroves, kelp forests, rocky shores, coral reefs and other marine ecosystems (medium confidence). {9.2.13, 30 12.4.8} 31 32 Marine heatwaves: Moderate increases in MHW frequency are projected for mid-latitudes, and only small 33 increases are projected for the Southern Ocean (medium confidence). Under the SSP5-8.5 scenario, permanent 34 MHWs (more than 360 days per year) are projected to occur in the 21st century in parts of the tropical ocean, 35 the Arctic Ocean and around 45°S; however, the occurrence of such permanent MHWs can be largely avoided 36 under SSP1-2.6 scenario. {Box 9.2, 12.4.8} 37 38 Ocean acidity: With the rising CO2 concentration, the ocean surface pH has declined globally over the past 39 four decades (virtually certain). {2.3.3.5, 5.3.3.2, 12.4.8} 40 41 Ocean salinity: At the basin scale, it is very likely that the Pacific and the Southern Ocean have freshened 42 while the Atlantic has become more saline. {2.3.3.2, 9.2.2.2, 12.4.8} 43 44 Dissolved oxygen: In recent decades, low oxygen zones in ocean ecosystems have expanded. {2.3.4.2, 5.3.3.2, 45 12.4.8} 46 47 Sea ice: Arctic perennial sea ice is being replaced by thin, seasonal ice, with earlier spring melt and delayed 48 fall freeze up. There is no clear trend in the Antarctic sea ice area over the past few decades and low confidence 49 in its future change. {2.3.2.1.1, 9.3.1.1, 12.4.8, 12.4.9} 50 51 52 53 54 Do Not Cite, Quote or Distribute TS-98 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 TS.4.3.2.10 Other Typological Domains 2 3 Some types of regions found in different continents face common climate challenges regardless of their 4 location. These include biodiversity hot spots that will very likely see even more extreme heat and droughts, 5 mountain areas where a projected raising in the freezing level height will alter snow and ice conditions (high 6 confidence), and tropical forests that are increasingly prone to fire weather (medium confidence). {Box 8.2, 7 8.4, 9.5, 12.3, 12.4} 8 9 Biodiversity hotspots located around the world will each face unique challenges in CID changes. Heat, drought 10 and length of dry season, wildfire weather, sea surface temperature and deoxygenation are relevant drivers to 11 terrestrial and freshwater ecosystems and have marked increasing trends. {12.3, 12.4.10.1} 12 13 Desert and semi-arid areas are strongly affected by CIDs such as extreme heat, drought and dust storms, with 14 large-scale aridity trends contributing to expanding drylands in some regions (high confidence). {12.3, 15 12.4.10.3} 16 17 Average warming in mountain areas varies with elevation, but the pattern is not globally uniform (medium 18 confidence). Extreme precipitation is projected to increase in major mountainous regions (medium to high 19 confidence depending on location), with potential cascading consequences of floods, landslides and lake 20 outbursts in all scenarios (medium confidence). {Box 8.2, 8.4.1.5, 9.5.1.3, 9.5.3.3, 9.5.2.3, Cross-Chapter Box 21 10.4, 11.5.5, 12.3, 12.4.1–12.4.6, 12.4.10.4} 22 23 Most tropical forests are challenged by a mix of emerging warming trends that are particularly large in 24 comparison to historical variability (medium confidence). Water cycle changes bring prolonged drought, 25 longer dry seasons, and increased fire weather to many tropical forests (medium confidence). {10.5, 12.3, 12.4} 26 27 28 [START BOX TS.14 HERE] 29 30 Box TS.14: Urban Areas 31 32 With global warming, urban areas and cities will be affected by more frequent occurrence of extreme climate 33 events, such as heatwaves, with more hot days and warm nights as well as sea level rise and increases in 34 tropical cyclone storm surge and rainfall intensity that will increase the probability of coastal city flooding 35 (high confidence). {Box 10.3, 11.3, 11.5, 12.3, 12.4} 36 37 Urban areas have special interactions with the climate system, for instance in terms of heat islands and altering 38 the water cycle, and thereby will be more affected by extreme climate events such as extreme heat (high 39 confidence). With global warming, increasing relative sea level compounded by increasing tropical cyclone 40 storm surge and rainfall intensity will increase the probability of coastal city flooding (high confidence). Arctic 41 coastal settlements are particularly exposed to climate change due to sea ice retreat (high confidence). 42 Improvements in urban climate modelling and climate monitoring networks have contributed to understanding 43 the mutual interaction between regional and urban climate (high confidence). {Box 10.3, 11.3, 11.5, 12.3, 44 12.4} 45 46 Despite having a negligible effect on global surface temperature (high confidence), urbanization has 47 exacerbated the effects of global warming through its contribution to the observed warming trend in and near 48 cities, particularly in annual mean minimum temperature (very high confidence) and increases in mean and 49 extreme precipitation over and downwind of the city, especially in the afternoon and early evening (medium 50 confidence). {2.3, Box 10.3, 11.3, 11.4, 12.3, 12.4} 51 52 Combining climate change projections with urban growth scenarios, future urbanization will amplify (very 53 high confidence) the projected local air temperature increase, particularly by strong influence on minimum 54 temperatures, which is approximately comparable in magnitude to global warming (high confidence). Do Not Cite, Quote or Distribute TS-99 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Compared to present day, large implications are expected from the combination of future urban development 2 and more frequent occurrence of extreme climate events, such as heatwaves, with more hot days and warm 3 nights adding to heat stress in cities (very high confidence). {Box 10.2, 11.3, 12.4} 4 5 Both sea levels and air temperatures are projected to rise in most coastal settlements (high confidence). There 6 is high confidence in an increase in pluvial flood potential in urban areas where extreme precipitation is 7 projected to increase, especially at high global warming levels. {11.4, 11.5, 12.4} 8 9 [END BOX TS.14 HERE] 10 Do Not Cite, Quote or Distribute TS-100 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Figures 2 3 4 5 Figure TS.1: Changes in atmospheric CO2 and global surface temperature (relative to 1850-1900) from the 6 deep past to the next 300 years. The intent is to show that CO2 and temperature covary, both in the 7 past and into the future, and that projected CO2 and temperatures are similar to those only from many 8 millions of years ago. CO2 concentrations from millions of years ago are reconstructed from multiple 9 proxy records (grey dots are data from 2.2.3.1, Figure 2.3 shown with cubic-spline fit). CO2 levels for 10 the last 800,000 years through the mid-20th century are from air trapped in polar ice; recent values are 11 from direct air measurements (Figure TS.9). {1.2.1.2, 2.2.3, Figures 1.5, 2.4, 2.5} Global surface 12 temperature prior to 1850 is estimated from marine oxygen isotopes, one of multiple sources of 13 evidence used to assess paleo temperatures in this report. {2.3.1.1.1, Cross-Chapter Box 2.1, Figure 1} 14 Temperature of the past 170 years is the AR6 assessed mean (Cross-Section Box TS.1). {2.3.1.1} CO2 15 levels and global surface temperature change for the future are shown for three SSP scenarios (TS.1.3) 16 through 2300 CE, using Earth System Model emulators calibrated to the assessed global surface 17 temperatures. {4.7.1, Cross-Chapter Box 7.1} Their smooth trajectories do not account for inter-annual 18 to inter-decadal variability, including transient response to potential volcanic eruptions. {Cross-Chapter 19 Box 4.1} Global maps for two paleo reference periods are based on CMIP6 and pre-CMIP6 multi- 20 model means, with site-level proxy data for comparison (squares and circles are marine and terrestrial, 21 respectively) (Box TS.2). {Cross-Chapter Box 2.1, Figure 7.13} The map for 2020 is an estimate of the 22 total observed warming since 1850-1900. {Figure 1.14} Global maps at right show two SSP scenarios 23 at 2100 (2081-2100) {4.5.1} and at 2300 (2281-2300; map from CMIP6 models; temperature assessed 24 in 4.7.1). A brief account of the major climate forcings associated with past global temperature changes 25 is in Cross-Chapter Box 2.1. 26 27 Do Not Cite, Quote or Distribute TS-101 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Box TS.2, Figure 1: Paleoclimate and recent reference periods, with selected key indicators. The intent of this 4 figure is to list the paleoclimate reference periods used in the WGI report, to summarize three key 5 global climate indicators, and compare CO2 with global temperature over multiple periods. (a) 6 Three large-scale climate indicators (atmospheric CO2, global surface temperature relative to 1850- 7 1900, and global mean sea level relative to 1900), based on assessments in Chapter 2, with 8 confidence levels ranging from low to very high. (b) Comparison between global surface 9 temperature (relative to 1850-1900) and atmospheric CO2 concentration for multiple reference 10 periods (mid-points with 5–95% ranges). {2.2.3, 2.3.1.1, 2.3.3.3, Figure 2.34} 11 Do Not Cite, Quote or Distribute TS-102 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.2, Figure 2: Global surface temperature as estimated from proxy records (reconstructed) and climate 5 models (simulated). The intent of this figure is to show the agreement between observations and 6 models of global temperatures during paleo reference periods. (a) For individual paleoclimate 7 reference periods. (a) For individual paleoclimate reference periods. (b) For the last millennium, 8 with instrumental temperature (AR6 assessed mean,10-year smoothed). Model uncertainties in (a) 9 and (b) are 5-95% ranges of multi-model ensemble means; reconstructed uncertainties are 5-95% 10 ranges (medium confidence) of (a) midpoints and (b) multi-method ensemble median. {2.3.1.1, 11 Figure 2.34, Figure 3.2c, Figure 3.44} 12 13 14 15 16 Do Not Cite, Quote or Distribute TS-103 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.2: Progress in climate models. The intent is to show present improvements in climate models in 5 resolution, complexity and representation of key variables. (a) Evolution of model horizontal 6 resolution and vertical levels (based on Figure 1.19); (b) Evolution of inclusion of processes and 7 resolution from CMIP Phase 3 (CMIP3) to CMIP6 (Annex II). (c) Centred pattern correlations 8 between models and observations for the annual mean climatology over the period 1980–1999. 9 Results are shown for individual CMIP3 (cyan), CMIP5 (blue) and CMIP6 (red) models (one 10 ensemble member is used) as short lines, along with the corresponding ensemble averages (long 11 lines). The correlations are shown between the models and the primary reference observational data 12 set (from left to right: ERA5, GPCP-SG, CERES-EBAF, CERES-EBAF, CERES-EBAF, CERES- 13 EBAF, JR-55, ERA5, ERA5, ERA5, ERA5, ERA5, ERA5, AIRS, ERA5, ESACCI-Soilmoisture, 14 LAI3g, MTE). In addition, the correlation between the primary reference and additional observational 15 data sets (from left to right: NCEP, GHCN, -, -, -, -, ERA5, HadISST, NCEP, NCEP, NCEP, NCEP, 16 NCEP, NCEP, ERA5, NCEP, -, -, FLUXCOM) are shown (solid grey circles) if available. To ensure a 17 fair comparison across a range of model resolutions, the pattern correlations are computed after 18 regridding all datasets to a resolution of 4º in longitude and 5º in latitude. (Expanded from Figure 19 3.43; produced with ESMValTool version 2). 20 21 22 23 Do Not Cite, Quote or Distribute TS-104 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.3: Emergence of changes in temperature over the historical period. The intent of this figure is to 5 show how observed changes in temperature have emerged and that the emergence pattern agrees with 6 model simulations. The observed change in temperature at a global warming level of 1°C (top map), 7 and the signal-to-noise ratio (the change in temperature at a global warming level of 1°C, divided by 8 the size of year-to-year variations, bottom map) using data from Berkeley Earth. The right panels 9 show the zonal means of the maps and include data from different observational datasets (red) and the 10 CMIP6 simulations (black, including the 5-95% range) processed in the same way as the observations. 11 {1.4.2, 10.4.3} 12 13 14 Do Not Cite, Quote or Distribute TS-105 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 Do Not Cite, Quote or Distribute TS-106 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.4: The climate change cause-effect chain: from anthropogenic emissions, to changes in atmospheric 4 concentration, to changes in the Earth’s energy balance (‘forcing’), to changes to changes in global 5 climate and ultimately regional climate and climatic impact-drivers. Shown is the core set of five SSP 6 scenarios as well as emission and concentration ranges for the previous RCP scenarios in year 2100; 7 CO2 emissions (GtCO2 yr–1), panel top left; CH4 emissions (middle) and SO2, NOx emissions (all in 8 Mt yr–1), top right; concentrations of atmospheric CO2 (ppm) and CH4 (ppb), second row left and 9 right; effective radiative forcing for both anthropogenic and natural forcings (W m –2), third row; 10 changes in global surface air temperature (°C) relative to 1850–1900, fourth row; maps of projected 11 temperature change (°C) (left) and changes in annual-mean precipitation (%) (right) at GWL 2°C 12 relative to 1850–1900 (see also Figure TS.5), bottom row. Carbon cycle and non-CO2 biogeochemical 13 feedbacks will also influence the ultimate response to anthropogenic emissions (arrows on the left). 14 {1.6.1, Cross-Chapter Box 1.4, 4.2.2, 4.3.1, 4.6.1, 4.6.2} 15 16 17 18 Do Not Cite, Quote or Distribute TS-107 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.5: How scenarios are linked to global warming levels (GWLs), and examples of the evolution of 5 patterns of change with global warming levels. Left: Illustrative example of GWLs defined as 6 global surface temperature response to anthropogenic emissions in unconstrained CMIP6 simulations, 7 for two illustrative scenarios (SSP1-2.6 and SSP3-7.0). The time when a given simulation reaches a 8 GWL, e.g., +2°C, relative to 1850-1900 is taken as the time when the central year of a 20-year 9 running mean first reaches that level of warming. See the dots for +2°C, and how not all simulations 10 reach all levels of warming. The assessment of the timing when a GWL is reached takes into account 11 additional lines of evidence and is discussed in Cross-Section Box TS.1. Right: Multi-model, multi- 12 simulation average response patterns of change in near-surface air temperature, precipitation 13 (expressed as percentage change) and soil moisture (expressed in standard deviations of interannual 14 variability), for three GWLs. The number to the top right of the panels shows the number of model 15 simulations averaged across including all models that reach the corresponding GWL in any of the 5 16 SSPs. See TS.2 for discussion. (See also Cross-Chapter Box 11.1) 17 18 Do Not Cite, Quote or Distribute TS-108 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.6: A graphical abstract for key aspects of the Technical Summary related to observed and 5 projected changes in global surface temperature and associated regional changes in climatic 6 impact-drivers relevant for impact and risk assessment. Top left: a schematic representation of the 7 likelihood for equilibrium climate sensitivity (ECS), consistent with the AR6 assessment (see Chapter 8 7; TS.3). ECS values above 5°C and below 2°C are termed low-likelihood high warming (LLHW) and 9 low-likelihood low warming, respectively. Top right: Observed (see Cross-Section Box TS.1) and 10 projected global surface temperature changes, shown as global warming levels (GWL) relative to 11 1850-1900, using the assessed 95% (top), 50% (middle) and 5% (bottom) likelihood time series (see 12 Chapter 4; TS.2). Bottom panels show maps of CMIP6 median projections of two climatic impact- 13 drivers (CIDs) at three different GWLs (columns for 1.5, 2 and 4°C) for the AR6 land regions (see 14 Chapters 1, 10, Atlas; TS.4). The heat warning index is the number of days per year averaged across 15 each region at which a heat warning for human health at level ‘danger’ would be issued according to 16 the U.S. National Oceanic and Atmospheric Administration (NOAA) (NOAA HI41, see Chapter 12 Do Not Cite, Quote or Distribute TS-109 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 and Technical Annex VI). The maps of extreme rainfall changes show the percentage change in the 2 amount of rain falling on the wettest day of a year (Rx1day, relative to 1995-2014, see Chapter 11) 3 averaged across each region when the respective GWL is reached. Additional CIDs are discussed in 4 TS.4. 5 6 Do Not Cite, Quote or Distribute TS-110 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 5 6 Cross-Section Box TS.1, Figure 1: Earth’s surface temperature history and future with key findings annotated 7 within each panel. The intent of this figure is to show global surface temperature 8 observed changes from the Holocene to now, and projected changes. (a) Global Do Not Cite, Quote or Distribute TS-111 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 surface temperature over the Holocene divided into three time scales. (i) 12,000 to 2 1000 years ago (10,000 BCE to 1000 CE) in 100-year time steps, (ii) 1000 to 1900 3 CE, 10-year smooth, and (iii) 1900 to 2020 CE (mean of four datasets in panel (c)). 4 Median of the multi-method reconstruction (bold lines), with 5% and 95% percentiles 5 of the ensemble members (thin lines). Vertical bars are 5th to 95th percentile ranges of 6 estimated global surface temperature for the Last Interglacial and mid Holocene 7 (medium confidence) (Section 2.3.1.1). All temperatures relative to 1850–1900. (b) 8 Spatially resolved trends (°C per decade) for (upper map) HadCRUTv5 over 1981– 9 2020, and (lower map, total change) multi-model mean projected changes from 1995– 10 2014 to 2081–2010 in the SST3-7.0 scenario. Observed trends have been calculated 11 where data are present in both the first and last decade and for at least 70% of all years 12 within the period using OLS. Significance is assessed with AR(1) correction and 13 denoted by stippling. Hatched areas in the lower map show areas of conflicting model 14 evidence on significance of changes. (c) Temperature from instrumental data for 15 1850–2020, including annually resolved averages for the four global surface 16 temperature datasets assessed in Section 2.3.1.1.3 (see text for references). The grey 17 shading shows the uncertainty associated with the HadCRUTv5 estimate. All 18 temperatures relative to the 1850–1900 reference period. (d) Recent past and 2015– 19 2050 evolution of annual mean global surface temperature change relative to 1850– 20 1900, from HadCRUTv5 (black), CMIP6 historical simulations (up to 2014, in grey, 21 ensemble mean solid, 5% and 95% percentiles dashed, individual models thin), and 22 CMIP6 projections under scenario SSP2-4.5, from four models that have an 23 equilibrium climate sensitivity near the assessed central value (thick yellow). Solid 24 thin coloured lines show the assessed central estimate of 20-year change in global 25 surface temperature for 2015–2050 under three scenarios, and dashed thin coloured 26 lines the corresponding 5% and 95% quantiles. (e) Assessed projected change in 20- 27 year running mean global surface temperature for five scenarios (central estimate 28 solid, very likely range shaded for SSP1-2.6 and SSP3-7.0), relative to 1995–2014 29 (left y-axis) and 1850–1900 (right y-axis). The y-axis on the right-hand side is shifted 30 upward by 0.85°C, the central estimate of the observed warming for 1995–2014, 31 relative to 1850–1900. The right y-axis in (e) is the same as the y-axis in (d). 32 Do Not Cite, Quote or Distribute TS-112 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 Figure TS.7: Simulated and observed changes compared to the 1995–2014 average in key large-scale 3 indicators of climate change across the climate system, for continents, ocean basins and globally 4 up to 2014. Black lines show observations, orange lines and shading show the multi-model mean and 5 5–95th percentile ranges for CMIP6 historical simulations including anthropogenic and natural 6 forcing, and green lines and shading show corresponding ensemble means and 5-95th percentile ranges 7 for CMIP6 natural-only simulations. Observations after 2014 (including, for example, a strong Do Not Cite, Quote or Distribute TS-113 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 subsequent decrease of Antarctic sea-ice area that leads to no significant overall trend since 1979) are 2 not shown because the CMIP6 historical simulations end in 2014. A 3-year running mean smoothing 3 has been applied to all observational time series. {3.8, Figure 3.41} 4 5 Do Not Cite, Quote or Distribute TS-114 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.8: Observed, simulated and projected changes compared to the 1995–2014 average in 4 key 4 indicators of the climate system through to 2100 differentiated by SSP scenario pathway. Past 5 simulations are based on the CMIP6 multi-model ensemble. Future projections are based on the 6 assessed ranges based upon multiple lines of evidence for (a) global surface temperature (Cross- 7 Section Box TS.1) and (b) global ocean heat content and the associated thermosteric sea level 8 contribution to Global Mean Sea Level (GMSL) change (right-hand axis) using a climate model 9 emulator (Cross-Chapter Box 7.1), and CMIP6 simulations for (c) Arctic September sea ice and (d) 10 Global land precipitation. SSP1-1.9 and SSP1-2.6 projections show that reduced GHG emissions lead 11 to a stabilization of global surface temperature, Arctic sea ice area and global land precipitation over 12 the 21st century. SSP1-2.6 shows that emissions reductions have the potential to substantially reduce 13 the increase in ocean heat content and thermosteric sea level rise over the 21st century but that some 14 increase is unavoidable. {4.3, 9.3, 9.6, Figure 4.2, Figure 9.6} 15 16 Do Not Cite, Quote or Distribute TS-115 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.9: Changes in well-mixed greenhouse gas (WMGHG) concentrations and Effective Radiative 4 Forcing. a) Changes in CO2 from proxy records over the past 3.5 million years; b) Changes in all 5 three WMGHGs from ice core records over the Common Era; c) directly observed WMGHG changes 6 since the mid-20th century; d) Evolution of ERF and components since 1750. Further details on data 7 sources and processing are available in the associated FAIR data table. {2.2, Figures 2.3, 2.4 and 8 2.10} Do Not Cite, Quote or Distribute TS-116 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.10: Observed and projected upper air temperature and circulation changes. Upper panels: Left: 4 Zonal cross-section of temperature trends for 2002-2019 in the upper troposphere region for the ROM 5 SAF radio-occultation dataset. Middle: Change in the annual and zonal mean atmospheric temperature 6 (°C) in 2081-2100 in SSP1-2.6 relative to 1995-2014 for 36 CMIP6 models. Right: the same in SSP3- 7 7.0 for 32 models. Lower panels: Left: Long-term mean (thin black colour) and linear trend (colour) 8 of zonal mean DJF zonal winds for ERA5. Middle: multi-model mean change in annual and zonal 9 mean wind (m s-1) in 2081-2100 in SSP1-2.6 relative to 1995-2014 based on 34 CMIP6 models. The 10 1995-2014 climatology is shown in contours with spacing 10 m s -1. Right: the same for SSP3-7.0 for 11 31 models. {2.3.1, 4.5.1, Figures 2.12, 2.18, and 4.26} 12 13 Do Not Cite, Quote or Distribute TS-117 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 5 Box TS.3, Figure 1: High-warming storylines. (a) CMIP6 multi-model mean linearly scaled to the assessed best 6 global surface temperature estimate for SSP1-2.6 in 2081–2100 relative to 1995–2014, (b) mean 7 across five high-warming models with global surface temperature changes warming nearest to the 8 upper bound of the assessed very likely range, (c) mean across five very high-warming models 9 with global surface temperature changes warming higher than the assessed very likely. (d-f) Same 10 as (a-c) but for SSP5-8.5. Note the different colour bars in (a-c) and (d-f). {4.7, Figure 4.41} 11 12 Do Not Cite, Quote or Distribute TS-118 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.11: Past and future ocean and ice sheet changes. Observed and simulated historical changes and 5 projected future changes under varying greenhouse gas emissions scenarios. Simulated and projected 6 ocean changes are shown as CMIP6 ensemble mean, and 5-95% range (shading) is provided for 7 scenario SSP1-2.6 and SSP3-7.0 (except in panel a where range provided for scenario SSP1-2.6 and 8 SSP5-8.5). Mean and 5-95% range in 2100 are shown as vertical bars on the right-hand side of each 9 panel. (a) Change in multiplication factor in surface ocean marine heatwave days relative to 1995- 10 2014 (defined as days exceeding the 99th percentile in SST from 1995-2014 distribution). Assessed 11 observational change span 1982-2019 from AVHRR satellite SST. (b) AMOC transport relative to 12 1995-2014 (defined as maximum transport at 26°N). Assessed observational change spans 2004-2018 13 from the RAPID array smoothed with a 12-month running mean (shading around the mean shows the 14 12-month running standard deviation around the mean). (c) Global mean percent change in ocean 15 oxygen (100–600 m depth), relative to 1995-2014. Assessed observational trends and very likely range 16 are from the SROCC assessment, and spans 1970–2010 centered on 2005. (d) Global mean surface 17 pH. Assessed observational change span 1985-2019, from the CMEMS SOCAT-based reconstruction 18 (shading around the global mean shows the 90% confidence interval). (e), (f): Ice sheet mass changes. 19 Projected ice sheet changes are shown as median, 5-95% range (light shading), and 17-83% range 20 (dark shading) of cumulative mass loss and sea level equivalent from ISMIP6 emulation under SSP1- 21 26 and SSP5-85 (shading and bold line), with individual emulated projections as thin lines. Median 22 (dot), 17-83% range (thick vertical bar), and 5-95% range (thin vertical bar) in 2100 are shown as 23 vertical bars on the right-hand side of each panel, from ISMIP6, ISMIP6 emulation, and LARMIP-2. 24 Observation-based estimates: For Greenland (e), for 1972-2018 (Mouginot), for 1992-2016 (Bamber), 25 for 1992-2020 (IMBIE) and total estimated mass loss range for 1840-1972 (Box). For Antarctica (f), Do Not Cite, Quote or Distribute TS-119 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 estimates based on satellite data combined with simulated surface mass balance and glacial isostatic 2 adjustment for 1992-2020 (IMBIE), 1992-2016 (Bamber), and 1979-2017 (Rignot). Left inset maps: 3 mean Greenland elevation changes 2010-2017 derived from CryoSat-2 radar altimetry (e) and mean 4 Antarctica elevation changes 1978-2017 derived from restored analog radar records (f). Right inset 5 maps: ISMIP6 model mean (2093- 2100) projected changes under the MIROC5 climate model for the 6 RCP8.5 scenario. {Box 9.2, 2.3.3, 2.3.4, 3.5.4, 4.3.2, 5.3.2, 5.3.3, 5.6.3, 9.2.3, 9.4.1, 9.4.2, Box 9.2 7 Figure 1, Figure 9.10, Figure 9.17, Figure 9.18} 8 9 Do Not Cite, Quote or Distribute TS-120 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.4, Figure 1: Global mean sea level change on different time scales and under different scenarios. (a) 5 GMSL change from 1900 to 2150, observed (1900–2018) and projected under the SSP scenarios 6 (2000–2150), relative to a 1995–2014 baseline. Solid lines show median projections. Shaded 7 regions show likely ranges for SSP1-2.6 and SSP3-7.0. Dotted and dashed lines show respectively 8 the 83rd and 95th percentile low-confidence projections for SSP5-8.5. Bars at right 9 show likely ranges for SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5 in 2150. Lightly 10 shaded thick/thin bars show 17th–83rd/5th–95th percentile low-confidence ranges in 2150 for 11 SSP1-2.6 and SSP5-8.5, based upon projection methods incorporating structured expert judgement 12 and Marine Ice Cliff Instability. Low-confidence range for SSP5-8.5 in 2150 extends to 4.8/5.4 m 13 at the 83rd/95th percentile. (b) GMSL change on 100- (blue), 2,000- (green) and 10,000-year 14 (magenta) time scales as a function of global surface temperature, relative to 1850–1900. For 100- 15 year projections, GMSL is projected for the year 2100, relative to a 1995-2014 baseline, and 16 temperature anomalies are average values over 2081–2100. For longer-term commitments, 17 warming is indexed by peak warming above 1850–1900 reached after cessation of emissions. 18 Shaded regions show paleo-constraints on global surface temperature and GMSL for the Last 19 Interglacial and mid-Pliocene Warm Period. Lightly shaded thick/thin blue bars show 17th– 20 83rd/5th–95th percentile low-confidence ranges for SSP1-2.6 and SSP5-8.5 in 2100, plotted at 2°C 21 and 5°C. (c) Timing of exceedance of GMSL thresholds of 0.5, 1.0, 1.5 and 2.0 m, under different 22 SSPs. Lightly shaded thick/thin bars show 1th7–83rd/5th–95th percentile low-confidence ranges 23 for SSP1-2.6 and SSP5-8.5. 24 Do Not Cite, Quote or Distribute TS-121 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.5, Figure 1: Carbon cycle processes and projections. Carbon cycle response to forcings. The figure shows 5 changes in carbon storage in response to elevated CO2 (a, b) and the response to climate warming 6 (c, d). Maps show spatial patterns of changes in carbon uptake during simulations with 1% per 7 year increase in CO2 {section 5.4.5.5}, and zonal mean plots show distribution of carbon changes 8 is dominated by the land (green lines) in the tropics and northern hemisphere and ocean (blue 9 lines) in the southern hemisphere. Hatching indicates regions where fewer than 80% of models 10 agree on the sign of response. (e) Future CO 2 projections: projected CO2 concentrations in the SSP 11 scenarios in response to anthropogenic emissions, results from coupled ESMs for SSP5-8.5 and 12 from the MAGICC7 emulator for other scenarios {section 4.3.1}. (f) Future carbon fluxes: Do Not Cite, Quote or Distribute TS-122 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 projected combined land and ocean fluxes (positive downward) up to 2100 for the SSP scenarios, 2 and extended to 2300 for available scenarios, 5-95% uncertainty plumes shown for SSP1-2.6 and 3 SSP3-7.0 {section 5.4.5.4, 5.4.10}. The numbers near the top show the number of model 4 simulations used. (g) Sink fraction: the fraction of cumulative emissions of CO 2 removed by land 5 and ocean sinks. The sink fraction is smaller under conditions of higher emissions. {5.4.5, 5.5.1; 6 Figure 5.27; Figure 4.31; Figure 5.25; Figure 5.30; Figure 5.31} 7 8 Do Not Cite, Quote or Distribute TS-123 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.12: Land-related changes relative to the 1850-1900 as a function of global warming levels. a) 4 Changes in the frequency (left scale) and intensity (in °C, right scale) of daily hot extremes occurring 5 every 10- and 50-years; b) as a), but for daily heavy precipitation extremes, with intensity change in 6 %; c) Changes in 10-year droughts aggregated over drought-prone regions (CNA, NCA, SCA, NSA, 7 SAM, SWS, SSA, MED, WSAF, ESAF, MDG, SAU, and EAU; for definitions of these regions, see 8 Atlas.2), with drought intensity (right scale) represented by the change of annual mean soil moisture, 9 normalized with respect to interannual variability; d) Changes in Northern Hemisphere spring (March- 10 April-May) snow cover extent relative to 1850-1900; e,f) Relative change (%) in annual mean of total 11 precipitable water (grey line), precipitation (red solid lines), runoff (blue solid lines) and in standard 12 deviation (i.e. variability) of precipitation (red dashed lines) and runoff (blue dashed lines) averaged 13 over (e) tropical and (f) extratropical land as function of global warming levels. CMIP6 models that Do Not Cite, Quote or Distribute TS-124 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 reached a 5°C warming level in the 21st century in SSP5-8.5 above the 1850-1900 average have been 2 used. Precipitation and runoff variability are estimated by respective standard deviation after 3 removing linear trends. Error bars show the 17-83% confidence interval for the warmest +5°C global 4 warming level. {Figures 11.6, 11.7, 11.12, 11.15, 11.18, 9.24, 8.16, Atlas.2} 5 6 Do Not Cite, Quote or Distribute TS-125 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Box TS.6, Figure 1: Projected water cycle changes. Long-term (2081-2100) projected annual mean changes (%) 4 relative to present-day (1995-2014) in the SSP2-4.5 emission scenario for (a) precipitation, (b) 5 surface evapotranspiration, (c) total runoff and (d) surface soil moisture. Top-right panel numbers 6 indicate the number of CMIP6 models used for estimating the ensemble mean. For other scenarios, 7 please refer to relevant figures in Chapter 8. Uncertainty is represented using the simple approach: 8 No overlay indicates regions with high model agreement, where ≥80% of models agree on sign of 9 change; diagonal lines indicate regions with low model agreement, where <80% of models agree 10 on sign of change. For more information on the simple approach, please refer to the Cross-Chapter 11 Box Atlas.1. {8.4.1, Figures 8.14, 8.17, 8.18, 8.19} 12 13 Do Not Cite, Quote or Distribute TS-126 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.13: Estimates of the net cumulative energy change (ZJ = 10 21 Joules) for the period 1971–2018 5 associated with: (a) observations of changes in the Global Energy Inventory (b) Integrated 6 Radiative Forcing; (c) Integrated Radiative Response. The intent is to show assessed changes in 7 energy budget and ERFs. Black dotted lines indicate the central estimate with likely and very likely 8 ranges as indicated in the legend. The grey dotted lines indicate the energy change associated with an 9 estimated pre-industrial Earth energy imbalance of 0.2 W m -2 (panel a) and an illustration of an 10 assumed pattern effect of –0.5 W m–2 °C–1 (panel c). Background grey lines indicate equivalent 11 heating rates in W m–2 per unit area of Earth’s surface. Panels (d) and (e) show the breakdown of 12 components, as indicated in the legend, for the Global Energy Inventory and Integrated Radiative 13 Forcing, respectively. Panel (f) shows the Global Energy Budget assessed for the period 1971–2018, 14 that is, the consistency between the change in the Global Energy Inventory relative to pre-industrial 15 and the implied energy change from Integrated Radiative Forcing plus Integrated Radiative Response 16 under a number of different assumptions, as indicated in the figure legend, including assumptions of 17 correlated and uncorrelated uncertainties in Forcing plus Response. Shading represents the very likely 18 range for observed energy change relative to pre-industrial and likely range for all other quantities. 19 Forcing and Response timeseries are expressed relative to a baseline period of 1850–1900. 20 21 22 23 Do Not Cite, Quote or Distribute TS-127 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.14: Schematic representation of changes in the top-of atmosphere (TOA) radiation budget following 5 a perturbation. The intent of the figure is to illustrate the concept of adjustments in the climate 6 system following a perturbation in the radiation budget. The baseline TOA energy budget (a) 7 responds instantaneously to perturbations (b), leading to adjustments in the atmospheric meteorology 8 and composition, and land surface that are independent of changes in surface temperature (c). Surface 9 temperature changes (here using an increase as an example) lead to physical, biogeophysical and 10 biogeochemical feedback processes (d). Long term feedback processes, such as those involving ice 11 sheets, are not shown here. {adapted from Chapter 7 Figure 7.2, FAQ 7.2 Figure 1, and Figure 8.3} 12 Do Not Cite, Quote or Distribute TS-128 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.15: Contribution to ERF and b) global surface temperature change from component emissions 5 between 1750 to 2019 based on CMIP6 models and c) net aerosol effective radiative forcing 6 (ERF) from different lines of evidence. The intent of the figure is to show advances since AR5 in the 7 understanding of a) aerosol ERF from different lines of evidence as assessed in Chapter 7, b) 8 emissions-based ERF and c) global surface temperature response for SLCFs as estimated in Chapter 9 6. In panel a), ERFs for well-mixed greenhouse gases (WMGHGs) are from the analytical formulae. 10 ERFs for other components are multi-model means based on ESM simulations that quantify the effect 11 of individual components. The derived emission-based ERFs are rescaled to match the concentration- 12 based ERFs in Figure 7.6. Error bars are 5-95% and for the ERF account for uncertainty in radiative 13 efficiencies and multi-model error in the means. In panel b), the global mean temperature response is 14 calculated from the ERF time series using an impulse response function. In panel c), the AR6 15 assessment is based on energy balance constraints, observational evidence from satellite retrievals, 16 and climate model-based evidence. For each line of evidence the assessed best-estimate contributions 17 from ERF due to ERFari and ERFaci are shown with darker and paler shading, respectively. Estimates 18 from individual CMIP5 and CMIP6 models are depicted by blue and red crosses, respectively. The 19 observational assessment for ERFari is taken from the instantaneous forcing due to aerosol-radiation 20 interactions (IRFari). Uncertainty ranges are given in black bars for the total aerosol ERF and depict 21 very likely ranges. {Sections 7.3.3, 6.4.2, Cross-Chapter Box 7.1, Figures 6.12, 7.5 ; Table 7.8} 22 Do Not Cite, Quote or Distribute TS-129 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 The intent is to show a) the progression in ECS including uncertainty and the lines of evidence used for assessment, b) 4 and c) show the lines of assessment used to assess ECS and TCR in AR6. 5 6 Figure TS.16: a) Evolution of equilibrium climate sensitivity (ECS) assessments from the Charney Report 7 through a succession of IPCC Assessment Reports to AR6, and lines of evidence and combined 8 assessment for (b) ECS and (c) transient climate response (TCR) in AR6. In panel (a), the lines of 9 evidence considered are listed below each assessment. Best estimates are marked by horizontal bars, 10 likely ranges by vertical bars, and very likely ranges by dotted vertical bars. In panel (b) and (c), 11 assessed ranges are taken from Tables 7.13 and 7.14 for ECS and TCR respectively. Note that for the 12 ECS assessment based on both the instrumental record and paleoclimates, limits (i.e., one-sided 13 distributions) are given, which have twice the probability of being outside the maximum/minimum 14 value at a given end, compared to ranges (i.e., two tailed distributions) which are given for the other 15 lines of evidence. For example, the extremely likely limit of greater than 95% probability corresponds 16 to one side of the very likely (5% to 95%) range. Best estimates are given as either a single number or 17 by a range represented by grey box. CMIP6 ESM values are not directly used as a line of evidence but 18 are presented on the Figure for comparison. {Sections 1.5, 7.5; Tables 7.13, 7.14; Figures 7.18} 19 20 21 22 Do Not Cite, Quote or Distribute TS-130 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.17: An overview of physical and biogeochemical feedbacks in the climate system. The intent is to 5 summarize assessed estimates of physical, biogeophysical and biogeochemical feedbacks on global 6 temperature based on Chapters 5, 6 and 7. (a) Synthesis of physical, biogeophysical and non-CO2 7 biogeochemical feedbacks that are included in the definition of ECS assessed in this Technical 8 Summary. These feedbacks have been assessed using multiple lines of evidence including 9 observations, models and theory. The net feedback is the sum of the Planck response, water vapour 10 and lapse rate, surface albedo, cloud, and biogeophysical and non-CO2 biogeochemical feedbacks. 11 Bars denote the mean feedback values and uncertainties represent very likely ranges; (b) Estimated 12 values of individual biogeophysical and non-CO2 biogeochemical feedbacks. The atmospheric 13 methane lifetime and other non-CO2 biogeochemical feedbacks have been calculated using global 14 Earth System Model simulations from AerChemMIP, while the CH 4 and N2O source responses to 15 climate have been assessed for the year 2100 using a range of modelling approaches using simplified 16 radiative forcing equations. The estimates represent the mean and 5-95% range. The level of 17 confidence in these estimates is low owing to the large model spread. (c) carbon-cycle feedbacks as 18 simulated by models participating in the C4MIP of CMIP6. An independent estimate of the additional 19 positive carbon-cycle climate feedbacks from permafrost thaw, which is not considered in most 20 C4MIP models, is added. The estimates represent the mean and 5-95% range. Note that these Do Not Cite, Quote or Distribute TS-131 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 feedbacks act through modifying the atmospheric concentration of CO 2 and thus are not included in 2 the definition of ECS, which assumes a doubling of CO 2, but are included in the definition and 3 assessed range of TCRE. {Sections Box 5.1, 5.4.7, 5.4.8, 6.4.5, 7.4.2, Figure 5.29, Tables 6.9, 7.10} Do Not Cite, Quote or Distribute TS-132 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI (b) (a) 1 2 3 The intent is to show the proportionality between cumulative CO2 emissions and global surface air temperature in 4 observations and models (left) as well as the assessed range of TCRE and the right-hand panel shows how information 5 is combined to derive remaining carbon budgets consistent with limiting warming to a specific level. 6 7 Figure TS.18: Illustration of (a) relationship between cumulative emissions of CO 2 and global mean surface air 8 temperature increase and (b) the assessment of the remaining carbon budget from its 9 constituting components based on multiple lines of evidence. Carbon budgets consistent with 10 various levels of additional warming are provided in Table 5.8 and should not be read from the 11 illustrations in either panel. In panel (a) thin black line shows historical CO 2 emissions together with 12 the assessed global surface temperature increase from 1850-1900 as assessed in Chapter 2 (Box 2.3). 13 The orange-brown range with its central line shows the estimated human-induced share of historical 14 warming. The vertical orange-brown line shows the assessed range of historical human-induced 15 warming for the 2010–2019 period relative to 1850-1900 (Chapter 3). The grey cone shows the 16 assessed likely range for the transient climate response to cumulative emissions of carbon dioxide 17 (TCRE) (Section 5.5.1.4), starting from 2015. Thin coloured lines show CMIP6 simulations for the 18 five scenarios of the WG1 core set (SSP1-1.9, green; SSP1-2.6, blue; SSP2-4.5, yellow; SSP3-7.0, 19 red; SSP5-8.5, maroon), starting from 2015 and until 2100. Diagnosed carbon emissions are 20 complemented with estimated land-use change emissions for each respective scenario. Coloured areas 21 show the Chapter 4 assessed very likely range of global surface temperature projections and thick 22 coloured central lines the median estimate, for each respective scenario, relative to the original 23 scenario emissions. For panel (b), the remaining allowable warming is estimated by combining the 24 global warming limit of interest with the assessed historical human induced warming (Section 25 5.5.2.2.2), the assessed future potential non-CO2 warming contribution (Section 5.5.2.2.3) and the 26 ZEC (Section 5.5.2.2.4). The remaining allowable warming (vertical blue bar) is subsequently 27 combined with the assessed TCRE (Section 5.5.1.4 and 5.5.2.2.1) and contribution of unrepresented 28 Earth system feedbacks (Section 5.5.2.2.5) to provide an assessed estimate of the remaining carbon 29 budget (horizontal blue bar, Table 5.8). Note that contributions in panel (b) are illustrative and are not 30 to scale. For example, the central ZEC estimate was assessed to be zero. {Box 2.3; Sections 5.2.1, 31 5.2.2; Figure 5.31} 32 Do Not Cite, Quote or Distribute TS-133 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.19: Carbon sink response in a scenario with net CO2 removal from the atmosphere. . The intent of 4 this figure is to show how atmospheric CO2 evolves under negative emissions and its dependence on 5 the negative emissions technologies. It also shows the evolution of the ocean and land sinks. Shown 6 are CO2 flux components from concentration-driven Earth system model simulations during different 7 emission stages of SSP1–2.6 and its long-term extension. (a) Large net positive CO2 emissions, (b) 8 small net positive CO2 emissions, (c) – (d) net negative CO2 emissions, (e) net zero CO2 emissions. 9 Positive flux components act to raise the atmospheric CO 2 concentration, whereas negative 10 components act to lower the CO2 concentration. Net CO2 emissions, land and ocean CO2 fluxes 11 represent the multi-model mean and standard deviation (error bar) of four ESMs (CanESM5, 12 UKESM1, CESM2-WACCM, IPSL-CM6a-LR) and one EMIC (UVic ESCM). Net CO2 emissions are 13 calculated from concentration-driven Earth system model simulations as the residual from the rate of 14 increase in atmospheric CO2 and land and ocean CO2 fluxes. Fluxes are accumulated over each 50- 15 year period and converted to concentration units (ppm). {5.6.2.1, Figure 5.33} 16 17 18 Do Not Cite, Quote or Distribute TS-134 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Figure TS.20: Global surface temperature change 10 and 100 years after a one year pulse of present-day 5 emissions. This figure shows the sectoral contribution to present-day climate change by specific 6 climate forcers including CO2 as well as SLCFs. The temperature response is broken down by 7 individual species and shown for total anthropogenic emissions (top), and sectoral emissions on 10- 8 year (left) and 100-year time scales (right). Sectors are sorted by (high-to-low) net temperature effect 9 on the 10-year time scale. Error bars in the top panel show the 5-95% range in net temperature effect 10 due to uncertainty in radiative forcing only (calculated using a Monte Carlo approach and best 11 estimate uncertainties from the literature). Emissions for 2014 are from the CMIP6 emissions dataset, 12 except for HFCs and aviation H2O which rely on other datasets (see Section 6.6.2 for more details). 13 CO2 emissions are excluded from open biomass burning and residential biofuel use. {6.6.2, Figure 14 6.16} 15 16 17 Do Not Cite, Quote or Distribute TS-135 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.7, Figure 1: Effects of short-lived climate forcers (SLCFs) on global surface temperature and air 5 pollution across the WG1 core set of Shared Socio-Economic Pathways (SSPs). The intent of 6 this figure is to show the climate and air quality (surface ozone and PM 2.5) response to SLCFs in 7 the SSP scenarios for near and long-term. Effects of net aerosols, tropospheric ozone, 8 hydrofluorocarbons (HFCs) (with lifetimes less than 50 years), and methane are compared with 9 those of total anthropogenic forcing for 2040 and 2100 relative to year 2019. The global surface 10 temperature changes are based on historical and future evolution of Effective Radiative Forcing 11 (ERF) as assessed in chapter 7 of this report. The temperature responses to the ERFs are calculated 12 with a common impulse response function (RT) for the climate response, consistent with the metric 13 calculations in Chapter 7 (Box 7.1). The RT has an equilibrium climate sensitivity of 3.0°C for a 14 doubling of atmospheric CO2 concentration (feedback parameter of -1.31 W m-2 °C-1). The 15 scenario total (grey bar) includes all anthropogenic forcings (long- and short-lived climate forcers, 16 and land use changes). Uncertainties are 5-95% ranges. The global changes in air pollutant 17 concentrations (ozone and PM2.5) are based on multimodel CMIP6 simulations and represent 18 changes in 5-year mean surface continental concentrations for 2040 and 2098 relative to 2019. 19 Uncertainty bars represent inter-model ±1 standard deviation. {6.7.2, 6.7.3, Figure 6.24} 20 Do Not Cite, Quote or Distribute TS-136 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.10, Figure 1: Synthesis of assessed observed changes and human influence of hot extremes (panel a), 5 heavy precipitation (panel b) and agricultural and ecological drought (panel c) for the 6 IPCC AR6 regions (displayed as hexagons). The colours in each panel represent the four 7 outcomes of the assessment on the observed changes: In Panel a): red – at least medium 8 confidence in an observed increase in hot extremes; blue – at least medium confidence in an 9 observed decrease in hot extremes; white – no significant change in hot extremes is observed for Do Not Cite, Quote or Distribute TS-137 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 the region as a whole; grey – the evidence in this region is insufficient (because of a lack of data 2 and/or literature) to make an assessment for the region as a whole. In panel b): green – at least 3 medium confidence in an observed increase in heavy rainfall; yellow - at least medium 4 confidence in an observed decrease in heavy rainfall; white – no significant change in heavy 5 rainfall is observed for the region as a whole; grey – the evidence in this region is insufficient 6 (because of a lack of data and/or literature) to make an assessment for the region as a whole. In 7 panel c): yellow – at least medium confidence in an observed increase in agricultural and 8 ecological drought; green - at least medium confidence in an observed decrease in agricultural 9 and ecological drought; white – no significant change in agricultural and ecological drought is 10 observed for the region as a whole; grey – the evidence in this region is insufficient (because of a 11 lack of data and/or literature) to make an assessment for the region as a whole. Each panel 12 represents in addition the synthesis of assessment of the human influence on the observed 13 changes based on available trend detection and attribution and event attribution scientific 14 publications. The level of confidence is indicated by a number of dots: high confidence (three 15 dots), medium confidence (two dots), low confidence (one dot), and when no assessment is 16 possible, due to insufficient evidence for the specific region (horizontal bar). For hot extremes, 17 the evidence is mostly drawn from changes in metrics based on daily maximum temperatures, 18 regional studies using other metrics (heatwave duration, frequency and intensity) are used in 19 addition {11.9.2}. For heavy precipitation, the evidence is mostly drawn from changes in metrics 20 based on one-day or five-day precipitation amounts using global and regional studies {11.9.3}. 21 Agricultural and ecological droughts are assessed based on observed and projected changes in 22 total column soil moisture, complemented by evidence on changes in surface soil moisture, 23 water-balance (precipitation minus evapotranspiration) and metrics driven by precipitation and 24 atmospheric evaporative demand. {11.9.3} All assessments are made for each AR6 region as a 25 whole and for the timeframe from 1950 to present thus, more local or assessment made on 26 shorter time scales might differ from what is shown in the figure. {11.9, Table TS.5}. 27 28 Do Not Cite, Quote or Distribute TS-138 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 4 Box TS.12, Figure 1: Example of generating regional climate information from multiple lines of evidence for the 5 case of Mediterranean summer warming, with indication of the information available from 6 the Interactive Atlas. (a) Mechanisms and feedbacks involved in enhanced Mediterranean 7 summer warming. (b) Locations of observing stations from different datasets. (c) Distribution of 8 1960‒2014 summer temperature trends (°C per decade) for observations (black crosses), CMIP5 9 (blue circles), CMIP6 (red circles), HighResMIP (orange circles), CORDEX EUR-44 (light blue Do Not Cite, Quote or Distribute TS-139 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 circles), CORDEX EUR-11 (green circles), and selected single Model Initial-condition Large 2 Ensembles - SMILEs (grey boxplots, MIROC6, CSIRO-Mk3-6-0, MPI-ESM and d4PDF). (d) 3 Time series of area averaged (25°N‒50°N, 10°W‒40°E) land point summer temperature 4 anomalies (°C, baseline period is 1995–2014): the boxplot shows long term (2081–2100) 5 temperature changes of different CMIP6 scenarios in respect to the baseline period. (e) Projected 6 Mediterranean summer warming in comparison to global annual mean warming of CMIP5 7 (RCP2.6, RCP4.5, RCP6.0 and RCP8.5) and CMIP6 (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5- 8 8.5) ensemble means (lines) and spread (shading). {Figure 10.20, Figure 10.21, Figure Atlas.8} 9 10 Do Not Cite, Quote or Distribute TS-140 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI Pathway to understanding past and assessing future climate changes at regional scale The southeastern South America (SES) case study a. Identification of and for interpreting b. Models simulations/evaluation of SES DJF precipitation over SES observed precipitation trend and variability in austral summer (DJF) the historical period and 21st century based on 7 large ensembles Aerosol GHG c. Quantification of the respective weight of uncertainties as a function + O3 depletion of future assessed periods Waves Internal variability uncertainty Model uncertainty trend (GPCC/1950-2014) Scenario uncertainty (mm/month/decade) Near Term Mid Term Long Term 1 2 Figure TS.21: Example of the interplay between drivers of climate variability and change at regional scale to 3 understand past and projected changes. The figure intent is to show an illustrative pathway for 4 understanding past, and anticipating future, climate change at regional scale in the presence of 5 uncertainties. (a) Identification of the climate drivers and their influences on climate phenomena 6 contributing through teleconnection to Southeastern South America summer (DJF) precipitation 7 variability and trends observed over 1950–2014. Drivers (red squares) include MoVs as well as 8 external forcing. Observed precipitation linear trend from GPCC is shown on continents (green-brown 9 colour bar in mm month per decade) and the SES AR6 WGI reference region is outlined with the –1 10 thick black contour. Climate phenomena leading to local impact on SES are schematically presented 11 (blue ovals). (b) Time series of decadal precipitation anomalies for DJF SES simulated from seven 12 large ensembles of historical + RCP8.5 simulations over 1950–2100. Shading corresponds to the 5th– 13 95th range of climate outcomes given from each large ensemble for precipitation (in mm/month) and 14 thick coloured lines stand for their respective ensemble mean. The thick timeseries in white 15 corresponds to the multi-model multi-member ensemble mean with model contribution being 16 weighted according to their ensemble size. GPCC observation is shown in the light black line with 17 squares over 1950–2014 and the 1995–2014 baseline period has been retained for calculation of 18 anomalies in all datasets. (c) Quantification of the respective weight (in percent) between the 19 individual sources of uncertainties (internal in gray, model in magenta and scenario in green) at near- 20 term, mid-term and long-term temporal windows defined in AR6 and highlighted in (b) for SES DJF 21 precipitation. All computations are done with respect to 1995–2014, taken as the reference period and 22 the scenario uncertainty is estimated from CMIP5 using the same set of models as for the large 23 ensembles that have run different RCP scenarios. {Figure 10.12a} 24 25 Do Not Cite, Quote or Distribute TS-141 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 Box TS.13, Figure 1: Global and regional monsoons: past trends and projected changes: The intent of this figure 3 is to show changes in precipitation over regional monsoon domains in terms of observed past 4 trends and related attribution, and in terms of future projections in one intermediate emission 5 scenario in the near, medium and long terms. (a) Global (black contour) and regional monsoons 6 (color shaded) domains. The global monsoon (GM) is defined as the area with local summer- 7 minus-winter precipitation rate exceeding 2.5 mm day –1 (see Annex V). The regional monsoon 8 domains are defined based on published literature and expert judgement (see Annex V), and also 9 accounting for the fact that the climatological summer monsoon rainy season varies across the 10 individual regions. Assessed regional monsoons are South and Southeast Asia (SAsiaM, Jun-Jul- 11 Aug-Sep), East Asia (EAsiaM, Jun-Jul-Aug), West Africa (WAfriM, Jun-Jul-Aug-Sep), North 12 America (NAmerM, Jul-Aug-Sep), South America (SAmerM, Dec-Jan-Feb), Australia and 13 Maritime Continent Monsoon (AusMCM, Dec-Jan-Feb). Equatorial South America (EqSAmer) 14 and South Africa (SAfri) regions are also shown, as they receive unimodal summer seasonal 15 rainfall although their qualification as monsoons is subject to discussion. (b) Global and regional 16 monsoons precipitation trends based on DAMIP CMIP6 simulations with both natural and 17 anthropogenic (ALL), GHG only (GHG), aerosols only (AER) and natural only (NAT) radiative 18 forcing. Weighted ensemble means are based on nine CMIP6 models contributing to the MIP 19 (with at least 3 members). Observed trends computed from CRU GPCP, and APHRO (only for 20 SAsiaM and EAsiaM) datasets are shown as well. (c) Percentage change in projected seasonal 21 mean precipitation over global and regional monsoons domain in the near-term (2021–2040), Do Not Cite, Quote or Distribute TS-142 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 mid-term (2041–2060), and long-term (2081–2100) under SSP2-4.5 based on 24 CMIP6 models. 2 {Figure 8.11, Figure 8.22} 3 4 Do Not Cite, Quote or Distribute TS-143 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 a) 2 3 4 5 6 7 8 9 Do Not Cite, Quote or Distribute TS-144 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 b) 2 3 4 5 Figure TS.22: (Panel a): shows the geographical location of regions belonging to one of five groups characterized by 6 a specific combination of changing climatic impact-drivers (CIDs). The five groups are represented by 7 the five different colours, and the CID combinations associated with each group are represented in the 8 corresponding ‘fingerprint’ and text below the map. Each fingerprint comprises a set of CIDs 9 projected to change with high confidence in every region in the group, and a second set of CIDs, one 10 or more of which are projected to change in each region with high or medium confidence. The CID 11 combinations follow a progression from those becoming hotter and drier (group 1) to those becoming 12 hotter and wetter (group 5). In between (groups 2–-4), the CIDs that change include some becoming 13 drier and some wetter and always include a set of CIDs which are getting hotter. Tropical cyclones 14 and severe wind CID changes are represented on the map with black dots in the regions affected. 15 Regions affected by coastal CID changes are described by text on the map. The five groups are chosen 16 to provide a reasonable level of detail for each region-specific detail whilst not overwhelming the map 17 with a full summary all aspects of the assessment, which is available in Table TS.5. [Placeholder: This 18 summary is also represented visually in the Interactive Atlas.] The CID changes summarized in the 19 figure represent high and medium confidence changes projected if a level of 2°C of global warming is 20 attained around 2050. The bar chart in panel b) shows the numbers of regions where each CID is 21 increasing or decreasing with medium or high confidence for all land regions reported in the map of 22 panel a) and for the ocean regions. The regions coloured in the map comprise the WG I AR6 reference 23 regions, which include inhabited land areas and an additional non-continuous Pacific Islands region 24 labelled PAC. Definitions of the acronyms of the other regions are provided in Atlas.1 and the 25 Interactive Atlas. {Table TS.5, Figure TS.24} Do Not Cite, Quote or Distribute TS-145 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI Year of significant emergence of changes in temperature over land regions (S/N>2) Dataset: Berkeley Earth. Temperature changes relative to 1850-1900. 1 Before 1981 1981-1988 1989-1996 1997-2004 2005-2012 2013-2020 2 3 4 5 Year of significant emergence of changes in temperature over land regions (S/N>2) Dataset: CRUTEM5. Temperature changes relative to 1850-1900. Grey: not enough data. 6 Before 1981 1981-1988 1989-1996 1997-2004 2005-2012 2013-2020 7 8 9 Figure TS.23: Time period during which the signals of temperature change in observed data aggregated over 10 the reference regions emerged from the noise of annual variability in the respective aggregated 11 data, using a signal-to-noise ratio of 2 as the threshold for emergence. Emergence time is Do Not Cite, Quote or Distribute TS-146 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 calculated for two global datasets (a) Berkeley Earth observations and (b) HadCRUT5 observations. 2 Regions in the HadCRUT5 map are shaded grey when data are available over less than 50% of the 3 area of the region. {Figure Atlas.11, TS.1.2.4} 4 5 Do Not Cite, Quote or Distribute TS-147 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 2 3 Figure TS.24: Projected change in the mean number of days per year with maximum temperature exceeding 4 35°C for CMIP5 (first column), CMIP6 (second column) and CORDEX (thirth column). The 5 map shows the median change in the number of days per year between the mid-century (2041–2060) 6 or end-century (2081–2100) and historical (1995–2014) periods for the CMIP5 and 7 CORDEX RCP8.5 and RCP2.6 and CMIP6 SSP5-8.5 and SSP1-2.6 scenarios ensembles. Stippling 8 indicates areas where less than 80% of the models agree on the sign of change. {Interactive Atlas} 9 Do Not Cite, Quote or Distribute TS-148 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 a) 2 3 4 5 6 7 8 Do Not Cite, Quote or Distribute TS-149 Total pages: 150 Final Government Distribution Technical Summary IPCC AR6 WGI 1 b) 2 3 4 5 Figure TS.25: Distribution of projected changes in selected climatic impact-driver indices for selected 6 regions for CMIP6, CMIP5 and CORDEX model ensembles. Different indices are shown for 7 different region: for east Europe and north Asia the mean number of days per year with maximum 8 temperature exceeding 35°C, for Central America and the Caribbean, and the Arabian peninsula, 9 western, southern and easten Asia the mean number of days per year with the NOAA Heat Index 10 exceeding 41°C, for Australasia, East Asia and Russia far East the average shoreline position 11 change, for South America, Europe and Africa the mean change in 1-in-100-years river discharge 12 per unit catchment area (m3 s–1 km–2), and for North America the median change in the number 13 of days with snow water equivalent (SWE) over 100 mm. For each box plot the changes or the 14 climatological values are reported respect to, or compared to, the recent past (1995–2014) period 15 for 1.5 oC, 2oC and 4oC global warming levels and for mid-century (2041–2060) or end-century 16 (2081–2100) periods for the CMIP5 and CORDEX RCP8.5 and RCP2.6 and CMIP6 SSP5-8.5 17 and SSP1-2.6 scenarios ensembles {Figure 12.5, Figure 12.6, Figure 12.9, Figure 12.SM.1, Figure 18 12.SM.2, Figure 12.SM.6}. Do Not Cite, Quote or Distribute TS-150 Total pages: 150