This box provides short descriptions of key concepts that are relevant to the AR6 WGI assessment, with a focus on their use in the Technical Summary and the Summary for Policymakers. The Glossary (Annex VII) includes more information on these concepts along with definitions of many other important terms and concepts used in this Report.

Characteristics of Climate Change Assessment

Global warming: Global warming refers to the change of global surface temperature relative to a baseline depending upon the application. Specific global warming levels, such as 1.5°C, 2°C, 3°C or 4°C, are defined as changes in global surface temperature relative to the years 1850–1900 as the baseline (the earliest period of reliable observations with sufficient geographic coverage). They are used to assess and communicate information about global and regional changes, linking to scenarios and used as a common basis for Working Group II (WGII) and Working Group III (WGIII) assessments. (Section TS.1.3, Cross-Section Box TS.1) Links to chapters1.4.1, 1.6.2, 4.6.1, Cross-Chapter Boxes 1.5, 2.3, 11.1, and 12.1, Atlas Sections 3–11, Glossary

Emergence: Emergence refers to the experience or appearance of novel conditions of a particular climate variable in a given region. This concept is often expressed as the ratio of the change in a climate variable relative to the amplitude of natural variations of that variable (often termed a ‘signal-to-noise’ ratio, with emergence occurring at a defined threshold of this ratio). Emergence can be expressed in terms of a time or a global warming level at which the novel conditions appear and can be estimated using observations or model simulations. (Sections TS.1.2.3 and TS.4.2) Links to chapters1.4.2, FAQ 1.2, 7.5.5, 10.3, 10.4, 12.5.2, Cross-Chapter Box Atlas.1, Glossary

Cumulative carbon dioxide (CO2) emissions: The total net amount of CO₂ emitted into the atmosphere as a result of human activities. Given the nearly linear relationship between cumulative CO₂ emissions and increases in global surface temperature, cumulative CO₂ emissions are relevant for understanding how past and future CO₂ emissions affect global surface temperature. A related term – remaining carbon budget – is used to describe the total net amount of CO₂ that could be released in the future by human activities while keeping global warming to a specific global warming level, such as 1.5°C, taking into account the warming contribution from non-CO₂ forcers as well. The remaining carbon budget is expressed from a recent specified date, while the total carbon budget is expressed starting from the pre-industrial period. (Sections TS.1.3 and TS.3.3) Links to chapters1.6.3, 5.5, Glossary

Net zero CO2 emissions: A condition that occurs when the amount of CO₂ emitted into the atmosphere by human activities equals the amount of CO₂ removed from the atmosphere by human activities over a specified period of time. Net negative CO₂ emissions occur when anthropogenic removals exceed anthropogenic emissions. (Section TS.3.3) Links to chaptersBox 1.4, Glossary

Human Influence on the Climate System

Earth’s energy imbalance: In a stable climate, the amount of energy that Earth receives from the Sun is approximately in balance with the amount of energy that is lost to space in the form of reflected sunlight and thermal radiation. ‘Climate drivers’, such as an increase in greenhouse gases or aerosols, interfere with this balance, causing the system to either gain or lose energy. The strength of a climate driver is quantified by its effective radiative forcing (ERF), measured in W m^–2. Positive ERF leads to warming, and negative ERF leads to cooling. That warming or cooling in turn can change the energy imbalance through many positive (amplifying) or negative (dampening) climate feedbacks. (Sections TS.2.2, TS.3.1 and TS.3.2) Links to chapters2.2.8, 7.2, 7.3, 7.4, Box 7.1, Box 7.2, Glossary

Attribution: Attribution is the process of evaluating the relative contributions of multiple causal factors to an observed change in climate variables (e.g., global surface temperature, global mean sea level), or to the occurrence of extreme weather or climate-related events. Attributed causal factors include human activities (such as increases in greenhouse gas concentration and aerosols, or land-use change) or natural external drivers (solar and volcanic influences), and in some cases internal variability. (Sections TS.1.2.4 and TS.2, Box TS.10) Links to chaptersCross-Working Group Box: Attribution in Chapter 1; 3.5; 3.8; 10.4; 11.2.4; Glossary

Committed change, long-term commitment: Changes in the climate system, resulting from past, present and future human activities, which will continue long into the future (centuries to millennia) even with strong reductions in greenhouse gas emissions. Some aspects of the climate system, including the terrestrial biosphere, the deep ocean and the cryosphere, respond much more slowly than surface temperatures to changes in greenhouse gas concentrations. As a result, there are already substantial committed changes associated with past greenhouse gas emissions. For example, global mean sea level will continue to rise for thousands of years, even if future CO₂ emissions are reduced to net zero and global warming halted, as excess energy due to past emissions continues to propagate into the deep ocean and as glaciers and ice sheets continue to melt. (Section TS.2.1, Box TS.4, Box TS.9) Links to chapters1.2.1, 1.3, Box 1.2, Cross-Chapter Box 5.3

Climate Information for Regional Climate Change and Risk Assessment

Distillation: The process of synthesizing information about climate change from multiple lines of evidence obtained from a variety of sources, taking into account user context and values. It leads to an increase in the usability, usefulness and relevance of climate information, enhances stakeholder trust, and expands the foundation of evidence used in climate services. It is particularly relevant in the context of co-producing regional-scale climate information to support decision-making. (Section TS.4.1, Box TS.11) Links to chapters10.1, 10.5, 12.6

(Climate change) risk: The concept of risk is a key aspect of how the IPCC assesses and communicates to decision-makers about the potential for adverse consequences for human or ecological systems, recognizing the diversity of values and objectives associated with such systems. In the context of climate change, risks can arise from potential impacts of climate change as well as human responses to climate change. WGI contributes to the common IPCC risk framing through the assessment of relevant climate information, including climatic impact-drivers and low-likelihood, high-impact outcomes. (Sections TS.1.4 and TS.4.1, Box TS.4) Links to chaptersCross-Chapter Boxes 1.3 and 12.1, Glossary

Climatic impact-drivers: Physical climate system conditions (e.g., means, events, extremes) that can be directly connected with having impacts on human or ecological systems are described as ‘climatic impact-drivers’ (CIDs) without anticipating whether their impacts are detrimental (i.e., as for hazards in the context of climate change risks) or provide potential opportunities. A range of indices may capture the sector- or application-relevant characteristics of a climatic impact-driver and can reflect exceedances of identified tolerance thresholds. (Sections TS.1.4 and TS.4.3) Links to chapters12.1–12.3, FAQ 12.1, Glossary

Storylines: The term storyline is used both in connection to scenarios (related to a future trajectory of emissions or socio-economic developments) or to describe plausible trajectories of weather and climate conditions or events, especially those related to high levels of risk. Physical climate storylines are introduced in AR6 to explore uncertainties in climate change and natural climate variability, to develop and communicate integrated and context-relevant regional climate information, and to address issues with deep uncertainty⁷, including low-likelihood, high-impact outcomes . (Section TS.1.4, Box TS.3, Infographic TS.1) Links to chapters1.4.4, Box 10.2, Glossary

Low-likelihood, high impact outcomes: Outcomes/events whose probability of occurrence is low or not well known (as in the context of deep uncertainty) but whose potential impacts on society and ecosystems could be high. To better inform risk assessment and decision-making, such low-likelihood outcomes are considered if they are associated with very large consequences and may therefore constitute material risks, even though those consequences do not necessarily represent the most likely outcome. (Section TS.1.4, Box TS.3, Figure TS.6) Links to chapters1.4.4, 4.8, Cross Chapter Box 1.3, Glossary

As part of the AR6 cycle, the IPCC produced three Special Reports in 2018 and 2019: the Special Report on Global Warming of 1.5°C (SR1.5), the Special Report on the Ocean and Cryosphere in a Changing Climate (SROCC), and the Special Report on Climate Change and Land (SRCCL).

The AR6 WGI Report provides a full and comprehensive assessment of the physical science basis of climate change that builds on the previous assessments and these Special Reports and considers new information and knowledge from the recent scientific literature⁸, including longer observational datasets and new scenarios and model results.

The structure of the AR6 WGI Report is designed to enhance the visibility of knowledge developments and to facilitate the integration of multiple lines of evidence, thereby improving confidence in findings. The Report has been peer-reviewed by the scientific community and governments (Annex X provides the Expert Reviewer list). The substantive introduction provided by (Chapter 1 is followed by a first set of chapters dedicated to large-scale climate knowledge (Chapters 2–4), which encompasses observations and paleoclimate evidence, causes of observed changes, and projections; these are complemented by (Chapter 11 for large-scale changes in extremes. The second set of chapters (Chapters 5–9) is orientated around the understanding of key climate system components and processes, including the global cycles of carbon, energy and water; short-lived climate forcers and their link to air quality; and the ocean, cryosphere and sea level change. The last set of chapters (Chapters 10–12 and the Atlas) is dedicated to the assessment and distillation of regional climate information from multiple lines of evidence at sub-continental to local scales (including urban climate), with a focus on recent and projected regional changes in mean climate, extremes, and climatic impact-drivers. The new online Interactive Atlas allows users to interact in a flexible manner through maps, time series and summary statistics with climate information for a set of updated WGI reference regions. The Report also includes 34 Frequently Asked Questions and answers for the general public (https://www.ipcc.ch/report/ar6/wg1/faqs).

Together, this Technical Summary and the underlying chapters aim at providing a comprehensive picture of knowledge progress since the WGI contribution to AR5. Multiple lines of scientific evidence confirm that the climate is changing due to human influence. Important advances in the ability to understand past, present and possible future changes should result in better-informed decision-making.

Some of the new results and main updates to key findings in this Report compared to AR5, SR1.5, SRCCL, and SROCC are summarized below. Relevant Technical Summary sections with further details are shown in parentheses after each bullet point.

Paleoclimate evidence is integrated within multiple lines of evidence across the WGI Report to more fully understand the climate system. Paleo evidence extends instrument-based observations of climate variables and climate drivers back in time, providing the long-term context needed to gauge the extent to which recent and potential future changes are unusual (Section TS.2, Figure TS.1). Pre-industrial climate states complement evidence from climate model projections by providing real-world examples of climate characteristics for past global warming levels, with empirical evidence for how the slow-responding components of the climate system operate over centuries to millennia – the time scale for committed climate change (Core Concepts Box, Box TS.4, Box TS.9). Information about the state of the climate system during well-described paleoclimate reference periods helps narrow the uncertainty range in the overall assessment of Earth’s sensitivity to climate forcing (Section TS.3.2.1). Links to chaptersCross-Chapter Box 2.1, FAQ 1.3, FAQ 2.1

Paleoclimate reference periods. Over the long evolution of Earth’s climate, several periods have received extensive research attention as examples of distinct climate states and rapid climate transitions (Box TS.2, Figure 1). These paleoclimate reference periods represent the present geological era (Cenozoic; past 65 million years) and are used across chapters to help structure the assessment of climate changes prior to industrialization. Cross-Chapter Box 2.1 describes the reference periods, along with a brief account of their climate forcings, and lists where each is discussed in other chapters. Cross-Chapter Box 2.4 summarizes information on one of the reference periods, the mid-Pliocene Warm Period. The Interactive Atlas includes model output from the World Climate Research Programme Coupled Model Intercomparison Project Phase 6 (CMIP6) for four of the paleoclimate reference periods.

Box TS.2, Figure 1 | Paleoclimate and recent reference periods, with selected key indicators. The intent of this figure is to list the paleoclimate reference periods used in this Report, to summarize three key global climate indicators, and compare CO2 with global temperature over multiple periods. (a) Three large-scale climate indicators (atmospheric CO2, global surface temperature relative to 1850–1900, and global mean sea level relative to 1900), based on assessments in Chapter 2, with confidence levels ranging fromlow tovery high. (b) Comparison between global surface temperature (relative to 1850–1900) and atmospheric CO2 concentration (shown on a log scale) for multiple reference periods (mid-points with 5–95% ranges). Links to chapters2.2.3, 2.3.1.1, 2.3.3.3, Figure 2.34

Paleoclimate models and reconstructions. Climate models that target paleoclimate reference periods have been featured by the IPCC since the First Assessment Report. Under the framework of CMIP6-PMIP4 (Paleoclimate Modelling Intercomparison Project), new protocols for model intercomparisons have been developed for multiple paleoclimate reference periods. These modelling efforts have led to improved understanding of the climate response to different external forcings, including changes in Earth’s orbital and plate movements, solar irradiance, volcanism, ice-sheet size and atmospheric greenhouse gases. Likewise, quantitative reconstructions of climate variables from proxy records that are compared with paleoclimate simulations have improved as the number of study sites and variety of proxy types have expanded, and as records have been compiled into new regional and global datasets. Links to chapters1.3.2, 1.5.1, Cross-Chapter Boxes 2.1 and 2.4

Global surface temperature. Since AR5, updated climate forcings, improved models, new understanding of the strengths and weaknesses of a growing array of proxy records, better chronologies and more robust proxy data products have led to better agreement between models and reconstructions. For global surface temperature, the mid-point of the AR6-assessed range and the median of the model-simulated temperatures differ by an average of 0.5°C across five reference periods; they overlap within their 90% ranges in four of five cases, which together span from about 6 [5 to 7]°C colder during the Last Glacial Maximum to about 14 [10 to 18] °C warmer during the Early Eocene, relative to 1850–1900 (Box TS.2, Figure 2a). Changes in temperature by latitude in response to multiple forcings show that polar amplification (stronger warming at high latitudes than the global average) is a prominent feature of the climate system across multiple climate states, and the ability of models to simulate this polar amplification in past warm climates has improved since AR5 (high confidence). Over the past millennium, and especially since about 1300 CE, simulated global surface temperature anomalies are well within the uncertainty of reconstructions (medium confidence), except for some short periods immediately following large volcanic eruptions, for which different forcing datasets disagree (Box TS.2, Figure 2b). Links to chapters2.3.1.1, 3.3.3.1, 3.8.2.1, 7.4.4.1.2

Box TS.2, Figure 2 | Global surface temperature as estimated from proxy records (reconstructed) and climate models (simulated). The intent of this figure is to show the agreement between observations and models of global temperatures during paleo reference periods. (a) For individual paleoclimate reference periods. (b) For the last millennium, with instrumental temperature (AR6 assessed mean, 10-year smoothed). Model uncertainties in (a) and (b) are 5–95% ranges of multi-model ensemble means; reconstructed uncertainties are 5–95% ranges (medium confidence) of (a) midpoints and (b) multi-method ensemble median. Links to chapters2.3.1.1, Figure 2.34, Figure 3.2c, Figure 3.44

Equilibrium climate sensitivity. Paleoclimate data provide evidence to estimate equilibrium climate sensitivity (ECS¹⁵) (Section TS.3.2.1). In AR6, refinements in paleo data for paleoclimate reference periods indicate that ECS is very likely greater than 1.5°C and likely less than 4.5°C, which is largely consistent with other lines of evidence and helps narrow the uncertainty range of the overall assessment of ECS. Some of the CMIP6 climate models that have either high (>5°C) or low (<2°C) ECS also simulate past global surface temperature changes outside the range of proxy-based reconstructions for the coldest and warmest reference periods. Since AR5, independent lines of evidence, including proxy records from past warm periods and glacial–interglacial cycles, indicate that sensitivity to forcing increases as temperature increases (Section TS.3.2.2). Links to chapters7.4.3.2, 7.5.3, 7.5.6, Table 7.11

Water cycle. New hydroclimate reconstructions and model-data comparisons have improved the understanding of the causes and effects of long-term changes in atmospheric and ocean circulation, including monsoon variability and modes of variability (Box TS.13, Section TS.4.2). Climate models are able to reproduce decadal drought variability on large regional scales, including the severity, persistence and spatial extent of past megadroughts known from proxy records (medium confidence). Some long-standing discrepancies remain, however, such as the magnitude of African monsoon precipitation during the early Holocene (the past 11,700 years), suggesting continuing knowledge gaps. Paleoclimate evidence shows that, in relatively high CO₂ climates such as the Pliocene, Walker circulation over the equatorial Pacific Ocean weakens, supporting the high confidence model projections of weakened Walker cells by the end of the 21st century. Links to chapters3.3.2, 8.3.1.6, 8.4.1.6, 8.5.2.1, 9.2

Sea level and ice sheets. Although past and future global warming differ in their forcings, evidence from paleoclimate records and modelling show that ice-sheet mass and global mean sea level (GMSL) responded dynamically over multiple millennia (high confidence). This evidence helps to constrain estimates of the committed GMSL response to global warming (Box TS.4). For example, under a past global warming levels of around [2.5°C to 4°C] relative to 1850–1900, like during the mid-Pliocene Warm Period, sea level was [5 to 25 m] higher than 1900 (medium confidence); under past global warming levels of [10°C to 18°C], like during the Early Eocene, the planet was essentially ice free (high confidence). Constraints from these past warm periods, combined with physical understanding, glaciology and modelling, indicate a committed long-term GMSL rise over 10,000 years, reaching about 8 to 13 m for sustained peak global warming of 2°C and up to 28 to 37 m for 5°C, which exceeds the AR5 estimate. Links to chapters2.3.3.3, 9.4.1.4, 9.4.2.6, 9.6.2, 9.6.3.5

Ocean. Since AR5, better integration of paleo-oceanographic data with modelling along with higher-resolution analyses of transient changes have improved understanding of long-term ocean processes. Low-latitude sea surface temperatures at the Last Glacial Maximum cooled more than previously inferred, resolving some inconsistencies noted in AR5. This paleo context supports the assessment that ongoing increase in ocean heat content (OHC) represents a long-term commitment (see Core Concepts Box), essentially irreversible on human time scales (high confidence). Estimates of past global OHC variations generally track those of sea surface temperatures around Antarctica, underscoring the importance of Southern Ocean processes in regulating deep-ocean temperatures. Paleoclimate data, along with other evidence of glacial–interglacial changes, show that Antarctic Circumpolar flow strengthened and that ventilation of Antarctic Bottom Water accelerated during warming intervals, facilitating release of CO₂ stored in the deep ocean to the atmosphere. Paleo evidence suggests significant reduction of deep-ocean ventilation associated with meltwater input during times of peak warmth. Links to chapters2.3.1.1, 2.3.3.1, 9.2.2, 9.2.3.2

Carbon cycle. Past climate states were associated with substantial differences in the inventories of the various carbon reservoirs, including the atmosphere (Section TS.2.2). Since AR5, the quantification of carbon stocks has improved due to the development of novel sedimentary proxies and stable-isotope analyses of air trapped in polar ice. Terrestrial carbon storage decreased markedly during the Last Glacial Maximum by 300–600 PgC, possibly by 850 PgC when accounting for interactions with the lithosphere and ocean sediments, a larger reduction than previously estimated, owing to a colder and drier climate. At the same time, the storage of remineralized carbon in the ocean interior increased by as much as 750–950 PgC, sufficient to balance the removal of carbon from the atmosphere (200 PgC) and terrestrial biosphere reservoirs combined (high confidence). Links to chapters5.1.2.2

This box synthesizes the outcomes of the assessment of past, current and future global surface temperature. Global mean surface temperature (GMST) and global surface air temperature (GSAT) are the two primary metrics of global surface temperature used to estimate global warming in IPCC reports. GMST merges sea surface temperature (SST) over the ocean and 2 m air temperature over land and sea ice areas and is used in most paleo, historical and present-day observational estimates. The GSAT metric is 2 m air temperature over all surfaces and is the diagnostic generally used from climate models. Changes in GMST and GSAT over time differ by at most 10% in either direction (high confidence), but conflicting lines of evidence from models and direct observations, combined with limitations in theoretical understanding, lead to low confidence in the sign of any difference in long-term trend. Therefore, long-term changes in GMST/GSAT are presently assessed to be identical, with expanded uncertainty in GSAT estimates. Hence the term global surface temperature is used in reference to both quantities in the text of the TS and SPM. Links to chaptersCross-Chapter Box 2.3

Global surface temperature has increased by 0.99 [0.84 to 1.10] °C from 1850–1900 to the first two decades of the 21st century (2001–2020) and by 1.09 [0.95 to 1.20] °C from 1850–1900 to 2011–2020. Temperatures as high as during the most recent decade (2011–2020) exceed the warmest centennial-scale range reconstructed for the present interglacial, around 6500 years ago [0.2°C to 1°C] (medium confidence). The next most recent warm period was about 125,000 years ago during the last interglacial when the multi-centennial temperature range [0.5°C to 1.5°C] encompasses the 2011–2020 values (medium confidence). The likely range of human-induced change in global surface temperature in 2010–2019 relative to 1850–1900 is 0.8°C to 1.3°C, with a central estimate of 1.07°C, encompassing the best estimate of observed warming for that period, which is 1.06°C with avery likely range of [0.88°C to 1.21°C], while The likely range of the change attributable to natural forcing is only –0.1°C to +0.1°C.

Compared to 1850–1900, average global surface temperature over the period 2081–2100 is very likely to be higher by [1.0°C to 1.8°C] in the low CO2emissions scenario SSP1-1.9 and by [3.3°C to 5.7°C] in the high CO2emissions scenario SSP5-8.5. In all scenarios assessed here except SSP5-8.5, the central estimate of 20-year averaged global surface warming crossing the 1.5°C level lies in the early 2030s, which is in the early part of The likely range (2030–2052) assessed in SR1.5. It is more likely than not that under SSP1-1.9, global surface temperature relative to 1850–1900 will remain below 1.6°C throughout the 21st century, implying a potential temporary overshoot of 1.5°C global warming of no more than 0.1°C. Global surface temperature in any individual year could exceed 1.5°C relative to 1850–1900 by 2030 with a likelihood between 40% and 60% across the scenarios considered here (medium confidence). A 2°C increase in global surface temperature relative to 1850–1900 will be crossed under SSP5-8.5 but is extremely unlikely to be crossed under SSP1-1.9. Periods of reduced and increased global surface temperature trends at decadal time scales will continue to occur in the 21st century (very high confidence). The effect of strong mitigation on 20-year global surface temperature trends would be likely to emerge during the near term (2021–2040), assuming no major volcanic eruptions occur. (Figure TS.8, Cross-Section Box TS.1, Figure 1) Links to chapters2.3, 3.3, 4.3, 4.4, 4.5, 4.6, 7.3

Surface Temperature History

Dataset innovations, particularly more comprehensive representation of polar regions, and the availability of new datasets have led to an assessment of increased global surface temperature change relative to the directly equivalent estimates reported in AR5. The contribution of changes in observational understanding alone between AR5 and AR6 in assessing temperature changes from 1850–1900 to 1986–2005 is estimated at 0.08 [–0.01 to 0.12] °C. Global surface temperature increased from 1850–1900 to 1995–2014 by 0.85 [0.69 to 0.95] °C, between 1850–1900 and the first two decades of the 21st century (2001–2020) by 0.99 [0.84 to 1.20] °C, and to the most recent decade (2011–2020) by 1.09 [0.95 to 1.20] °C. Each of the last four decades has in turn been warmer than any decade that preceded it since 1850. Temperatures have increased faster over land than over the ocean since 1850–1900, with warming to 2011–2020 of 1.59 [1.34 to 1.83] °C over land and 0.88 [0.68 to 1.01] °C over the ocean. Links to chapters2.3.1, Cross-Chapter Box 2.3

Global surface temperature during the period 1850–1900 is used as an approximation for pre-industrial conditions for consistency with AR5 and AR6 Special Reports, whilst recognizing that radiative forcings have a baseline of 1750 for the start of anthropogenic influences. It is likely that there was a net anthropogenic forcing of 0.0–0.3 Wm^–2 in 1850–1900 relative to 1750 (medium confidence), and from the period around 1750 to 1850–1900, there was a change in global surface temperature of around 0.1°C (likely range –0.1 to +0.3°C, medium confidence), with an anthropogenic component of 0.0°C to 0.2°C (likely range , medium confidence). Links to chaptersCross-Chapter Box 1.2, 7.3.5

Global surface temperature has evolved over geological time (Figure TS.1, Box TS.2). Beginning approximately 6500 years ago, global surface temperature generally decreased, culminating in the coldest multi-century interval of the post-glacial period (since roughly 7000 years ago), which occurred between around 1450 and 1850 (high confidence). Over the last 50 years, global surface temperature has increased at an observed rate unprecedented in at least the last two thousand years (high confidence). Temperatures as high as during the most recent decade (2011–2020) exceed the warmest centennial-scale range reconstructed for the present interglacial, around 6500 years ago [0.2°C to 1°C] (medium confidence). The next most recent warm period was about 125,000 years ago during the Last Interglacial when the multi-centennial temperature range [0.5°C to 1.5°C] encompasses the 2011–2020 values (medium confidence) (Cross-Section Box TS.1, Figure 1). During the mid-Pliocene Warm Period, around 3.3–3.0 million years ago, global surface temperature was 2.5°C to 4°C warmer (medium confidence). Links to chapters2.3.1, Cross-Chapter Box 2.1 and 2.4

Current Warming

There is very high confidence that the CMIP6 model ensemble reproduces observed global surface temperature trends and variability since 1850 with errors small enough to allow for detection and attribution of human-induced warming. The CMIP6 multi-model mean global surface warming between 1850–1900 and 2010–2019 is close to the best estimate of observed warming, though some CMIP6 models simulate a warming that is outside the assessed very likely observed range. Links to chapters3.3.1

The likely range of human-induced change in global surface temperature in 2010–2019 relative to 1850–1900 is 0.8°C to 1.3°C, with a central estimate of 1.07°C (Figure Cross-Section Box TS.1, Figure 1), encompassing the best estimate of observed warming for that period, which is 1.06°C with avery likely range of [0.88°C to 1.21°C], while the likely range of the change attributable to natural forcing is only –0.1°C to +0.1°C. This assessment is consistent with an estimate of the human-induced global surface temperature rise based on assessed ranges of perturbations to the top of the atmosphere (effective radiative forcing) and with metrics of feedbacks of the climate response (equilibrium climate sensitivity and the transient climate response). Over the same period, well-mixed greenhouse gas forcing likely warmed global surface temperature by 1.0°C to 2.0°C, while aerosols and other anthropogenic forcings likely cooled global surface temperature by 0.0°C to 0.8°C. Links to chapters2.3.1, 3.3.1, 7.3.5, Cross-Chapter Box 7.1

The observed slower increase in global surface temperature (relative to preceding and following periods) in the 1998–2012 period, sometimes referred to as ‘the hiatus’, was temporary (very high confidence). The increase in global surface temperature during the 1998–2012 period is also greater in the data sets used in the AR6 assessment than in those available at the time of AR5. Using these updated observational data sets and a like-for-like consistent comparison of simulated and observed global surface temperature, all observed estimates of the 1998–2012 trend lie within the very likely range of CMIP6 trends. Furthermore, the heating of the climate system continued during this period, as reflected in the continued warming of the global ocean (very high confidence) and in the continued rise of hot extremes over land (medium confidence). Since 2012, global surface temperature has risen strongly, with the past five years (2016–2020) being the hottest five-year period between 1850 and 2020 (high confidence). Links to chapters2.3.1, 3.3.1, 3.5.1, Cross-Chapter Box 3.1

Future Changes in Global Surface Temperature

The AR6 assessment of future change in global surface temperature is, for the first time in an IPCC report, explicitly constructed by combining new projections for the SSP scenarios with observational constraints based on past simulated warming as well as the AR6-updated assessment of equilibrium climate sensitivity and transient climate response. In addition, climate forecasts initialized from the observed climate state have been used for the period 2019–2028. The inclusion of additional lines of evidence has reduced the assessed uncertainty ranges for each scenario (Cross-Section Box TS.1, Figure 1). Links to chapters4.3.1, 4.3.4, Box 4.1, 7.5

Cross-Section Box TS.1, Figure 1 | Earth’s surface temperature history and future with key findings annotated within each panel. The intent of this figure is to show global surface temperature observed changes from the Holocene to now, and projected changes. (a) Global surface temperature over the Holocene divided into three time scales: (i) 12,000 to 1000 years ago (10,000 BCE to 1000 CE) in 100-year time steps, (ii) 1000 to 1900 CE, 10-year smooth, and (iii) 1900 to 2020 CE (mean of four datasets in panel c). Bold lines show the median of the multi-method reconstruction, with 5% and 95% percentiles of the ensemble members (thin lines). Vertical bars are 5–95th percentile ranges of estimated global surface temperature for the Last Interglacial and mid-Holocene (medium confidence) (Section 2.3.1.1). All temperatures are relative to 1850–1900. (b) Spatially resolved trends (°C per decade) for (upper map) HadCRUTv5 over 1981–2020, and (lower map, total change) multi-model mean projected changes from 1995–2014 to 2081–2100 in the SST3-7.0 scenario. Observed trends have been calculated where data are present in both the first and last decade and for at least 70% of all years within the period using ordinary least squares. Significance is assessed with autoregressive AR(1) model correction and denoted by stippling. Hatched areas in the lower map show areas of conflicting model evidence on significance of changes. (c) Temperature from instrumental data for 1850–2020, including annually resolved averages for the four global surface temperature datasets assessed in Section 2.3.1.1.3 (see text for references). The grey shading shows the uncertainty associated with the HadCRUTv5 estimate. All temperatures are relative to the 1850–1900 reference period. (d) Recent past and 2015–2050 evolution of annual mean global surface temperature change relative to 1850–1900, from HadCRUTv5 (black), Coupled Model Intercomparison Project Phase 6 (CMIP6) historical simulations (up to 2014, in grey, ensemble mean solid, 5% and 95% percentiles dashed, individual models thin), and CMIP6 projections under scenario SSP2-4.5, from four models that have an equilibrium climate sensitivity near the assessed central value (thick yellow). Solid thin coloured lines show the assessed central estimate of 20-year change in global surface temperature for 2015–2050 under three scenarios, and dashed thin coloured lines the corresponding 5% and 95% quantiles. (e) Assessed projected change in 20-year running mean global surface temperature for five scenarios (central estimate solid, very likely range shaded for SSP1-2.6 and SSP3-7.0), relative to 1995–2014 (left y-axis) and 1850–1900 (right y-axis). The y-axis on the right-hand side is shifted upward by 0.85°C, the central estimate of the observed warming for 1995–2014, relative to 1850–1900. The right y-axis in (e) is the same as the y-axis in (d). Links to chapters2.3, 4.3, 4.4

During the near term (2021–2040), a 1.5°C increase in global surface temperature, relative to 1850–1900, is very likely to occur in scenario SSP5-8.5, likely to occur in scenarios SSP2-4.5 and SSP3-7.0, and more likely than not to occur in scenarios SSP1-1.9 and SSP1-2.6. The time of crossing a warming level is defined here as the midpoint of the first 20-year period during which the average global surface temperature exceeds the level. In all scenarios assessed here except SSP5-8.5, the central estimate of crossing the 1.5°C level lies in the early 2030s. This is in the early part of the likely range (2030–2052) assessed in SR1.5, which assumed continuation of the then-current warming rate; this rate has been confirmed in the AR6. Roughly half of this difference arises from a larger historical warming diagnosed in AR6. The other half arises because for central estimates of climate sensitivity, most scenarios show stronger warming over the near term than was estimated as ‘current’ in SR1.5 (medium confidence). When considering scenarios similar to SSP1-1.9 instead of linear extrapolation, the SR1.5 estimate of when 1.5°C global warming is crossed is close to the central estimate reported here. (Cross-Section Box TS.1, Table 1) Links to chapters2.3.1, Cross-Chapter Box 2.3, 3.3.1, 4.3.4, Box 4.1

It is more likely than not that under SSP1-1.9, global surface temperature relative to 1850–1900 will remain below 1.6°C throughout the 21st century, implying a potential temporary overshoot of 1.5°C global warming of no more than 0.1°C. If climate sensitivity lies near the lower end of the assessed very likely range, crossing the 1.5°C warming level is avoided in scenarios SSP1-1.9 and SSP1-2.6 (medium confidence). Global surface temperature in any individual year, in contrast to the 20-year average, could by 2030 exceed 1.5°C relative to 1850–1900 with a likelihood between 40% and 60%, across the scenarios considered here (medium confidence). (Cross-Section Box TS.1, Table 1) Links to chapters4.3.4, 4.4.1, Box 4.1, 7.5

During the 21st century, a 2°C increase in global surface temperature relative to 1850–1900 will be crossed under SSP5-8.5 and SSP3-7.0, is extremely likely to be crossed under SSP2-4.5, but is unlikely to be crossed under SSP1-2.6 and extremely unlikely to be crossed under SSP1-1.9. For the mid-term period 2041–2060, this 2°C global warming level is very likely to be crossed under SSP5-8.5, likely to be crossed under SSP3-7.0, and more likely than not to be crossed under SSP2-4.5. (Cross-Section Box TS.1, Table 1) Links to chapters4.3.4

Events of reduced and increased global surface temperature trends at decadal time scales will continue to occur in the 21st century but will not affect the centennial-scale warming (very high confidence). If strong mitigation is applied from 2020 onward as reflected in SSP1-1.9, its effect on 20-year trends in global surface temperature would likely emerge during the near term (2021–2040), measured against an assumed non-mitigation scenario such as SSP3-7.0 or SSP5-8.5. All statements about crossing the 1.5°C level assume that no major volcanic eruption occurs during the near term (Cross-Section Box TS.1, Table 1). Links to chapters2.3.1, Cross-Chapter Box 2.3, 4.3.4, 4.4.1, 4.6.3, Box 4.1

Compared to 1850–1900, average global surface temperature over the period 2081–2100 is very likely to be higher by [1.0°C to 1.8°C] in the low CO₂ emissions scenario SSP1-1.9 and by [3.3°C to 5.7°C] in the high CO₂ emissions scenario SSP5-8.5. For the scenarios SSP1-2.6, SSP2-4.5, and SSP3-7.0, the correspondingvery likely ranges are [1.3°C to 2.4°C], [2.1°C to 3.5°C], and [2.8°C to 4.6°C], respectively. The uncertainty ranges for the period 2081–2100 continue to be dominated by the uncertainty in equilibrium climate sensitivity and transient climate response (very high confidence) (Cross-Section Box TS.1, Table 1). Links to chapters4.3.1, 4.3.4, 4.4.1, 7.5

The CMIP6 models project a wider range of global surface temperature change than the assessed range (high confidence); furthermore, the CMIP6 global surface temperature increase tends to be larger than that in CMIP5 (very high confidence). Links to chapters4.3.1, 4.3.4, 4.6.2, 7.5.6

Cross-Section Box TS.1, Table 1 | Assessment results for 20-year averaged change in global surface temperature based on multiple lines of evidence. The change is displayed in °C relative to the 1850–1900 reference period for selected time periods (first three rows), and as the first 20-year period during which the average global surface temperature change exceeds the specified level relative to the period 1850–1900 (last four rows). The entries give both the central estimate and, in parentheses, the very likely (5–95%) range. An entry n.c. means that the global warming level is not crossed during the period 2021–2100.

Future global warming exceeding the assessed very likely range cannot be ruled out and is potentially associated with the highest risks for society and ecosystems. Such low-likelihood, high-warming storylines tend to exhibit substantially greater changes in the intensity of regional drying and wetting than the multi-model mean. Even at levels of warming within the very likely range, global and regional low-likelihood outcomes might occur, such as large precipitation changes, additional sea level rise associated with collapsing ice sheets (see Box TS.4), or abrupt ocean circulation changes. While there is medium confidence that the Atlantic Meridional Overturning Circulation (AMOC) will not experience an abrupt collapse before 2100, if it were to occur, it would very likely cause abrupt shifts in regional weather patterns and water cycle. The probability of these low-likelihood outcomes increases with higher global warming levels. If the real-world climate sensitivity lies at the high end of the assessed range, then global and regional changes substantially outside the very likely range projections occur for a given emissions scenario. With increasing global warming, some very rare extremes and some compound events (multivariate or concurrent extremes) with low likelihood in past and current climate will become more frequent, and there is a higher chance that events unprecedented in the observational record occur (high confidence). Finally, low-likelihood, high-impact outcomes may also arise from a series of very large volcanic eruptions that could substantially alter the 21st century climate trajectory compared to SSP-based Earth system model (ESM) projections. Links to chaptersCross-Chapter Box 4.1, 4.3, 4.4, 4.8, 7.3, 7.4, 7.5, 8.6, 9.2, 9.6, Box 9.4, Box 11.2, Cross-Chapter Box 12.1

Previous IPCC reports largely focused their assessment on the projected very likely range of future surface warming and associated climate change. However, a comprehensive risk assessment also requires considering the potentially larger changes in the physical climate system that are unlikely orvery unlikely but possible and potentially associated with the highest risks for society and ecosystems (Figure TS.6). Since AR5, the development of physical climate storylines of high warming has emerged as a useful approach for exploring the future risk space that lies outside of the IPCCvery likely range projections. Links to chapters4.8

Uncertainty in the true values of equilibrium climate sensitivity (ECS) and transient climate response (TCR) dominate uncertainty in projections of future warming under moderate to strong emissions scenarios (Section TS.3.2). A real-world ECS higher than the assessed very likely range (2°C–5°C) would require a strong historical aerosol cooling and/or a trend towards stronger warming from positive feedbacks linked to changes in SST patterns (pattern effects), combined with a strong positive cloud feedback and substantial biases in paleoclimate reconstructions – each of which is assessed as eitherunlikely orvery unlikely , but not ruled out. Since CMIP6 contains several ESMs that exceed the upper bound of the assessed very likely range in future surface warming, these models can be used to develop low-likelihood, high warming storylines to explore risks and vulnerabilities, even in the absence of a quantitative assessment of likelihood. Links to chapters4.3.4, 4.8, 7.3.2, 7.4.4, 7.5.2, 7.5.5, 7.5.7

CMIP6 models with surface warming outside, or close to, the upper bound of the very likely range exhibit patterns of large widespread temperature and precipitation changes that differ substantially from the multi-model mean in all scenarios. For SSP5-8.5, the high-warming models exhibit widespread warming of more than 6°C over most extratropical land regions and parts of the Amazon. In the Arctic, annual mean temperatures increase by more than 10°C relative to present-day, corresponding to about 30% more than the best estimate of warming. Even for SSP1-2.6, high-warming models show on average 2°C–3°C warming relative to present-day conditions over much of Eurasia and North America (about 40% more than the best estimate of warming) and more than 4°C warming relative to the present over the Arctic in 2081–2100 (Box TS.3, Figure 1). Such a high global warming storyline would imply that the remaining carbon budget consistent with a 2°C warming is smaller than the assessed very likely range. Put another way, even if a carbon budget that likely limits warming to 2°C is met, a low-likelihood, high-warming storyline would result in warming of 2.5°C or more. Links to chapters4.8

CMIP6 models with global warming close to the upper bound of the assessed very likely warming range tend to exhibit greater changes in the intensity of regional drying and wetting than the multi-model mean. Furthermore, these model projections show a larger area of drying and tend to show a larger fraction of strong precipitation increases than the multi-model mean. However, regional precipitation changes arise from both thermodynamic and dynamic processes so that the most pronounced global warming levels are not necessarily associated with the strongest precipitation response. Abrupt human-caused changes to the water cycle cannot be ruled out. Positive land surface feedbacks, involving vegetation and dust, can contribute to abrupt changes in aridity, but there is onlylow confidence that such changes will occur during the 21st century. Continued Amazon deforestation, combined with a warming climate, raises the probability that this ecosystem will cross a tipping point into a dry state during the 21st century (low confidence). (See also Box TS.9). Links to chapters4.8, 8.6.2

While there is medium confidence that the projected decline in the AMOC (Section TS.2.4) will not involve an abrupt collapse before 2100, such a collapse might be triggered by an unexpected meltwater influx from the Greenland Ice Sheet. If an AMOC collapse were to occur, it would very likely cause abrupt shifts in the regional weather patterns and water cycle, such as a southward shift in the tropical rain belt, and could result in weakening of the African and Asian monsoons, strengthening of Southern Hemisphere monsoons, and drying in Europe. (See also Boxes TS.9 and TS.13). Links to chapters4.7.2, 8.6.1, 9.2.3

Very rare extremes and compound or concurrent events, such as the 2018 concurrent heatwaves across the Northern Hemisphere, are often associated with large impacts. The changing climate state is already altering the likelihood of extreme events, such as decadal droughts and extreme sea levels, and will continue to do so under future warming. Compound events and concurrent extremes contribute to increasing probability of low-likelihood, high-impact outcomes and will become more frequent with increasing global warming (high confidence). Higher warming levels increase the likelihood of events unprecedented in the observational record. Links to chapters9.6.4, Box 11.2

Finally, low likelihood storylines need not necessarily relate solely to the human-induced changes in climate. A low-likelihood, high-impact outcome, consistent with historical precedent in the past 2500 years, would be to see several large volcanic eruptions that could greatly alter the 21st century climate trajectory compared to SSP-based Earth system model projections. Links to chaptersCross-Chapter Box 4.1

Box TS.3, Figure 1 | High-warming storylines. The intent of this figure is to illustrate high warming storylines compared to the CMIP6 multi-model-mean. (a) Coupled Model Intercomparison Project Phase 6 (CMIP6) multi-model mean linearly scaled to the assessed best global surface temperature estimate for SSP1-2.6 in 2081–2100 relative to 1995–2014, (b) mean across five high-warming models with global surface temperature changes nearest to the upper bound of the assessed very likely range, and (c) mean across five very high-warming models with global surface temperature changes higher than the assessed very likely . (d–f) Same as (a–c) but for SSP5-8.5. Note the different colour bars in (a–c) and (d–f). Links to chapters4.7, Figure 4.41

Open figure

Global mean sea level (GMSL) increased by 0.20 [0.15 to 0.25] m over the period 1901 to 2018, with a rate 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 confidence). Human activities were very likely the main driver of observed GMSL rise since 1971, and new observational evidence leads to an assessed sea level rise over the period 1901 to 2018 that is consistent with the sum of individual components contributing to sea level rise, including expansion due to ocean warming and melting of glaciers and ice sheets (high confidence). It is virtually certain that GMSL will continue to rise over the 21st century in response to continued warming of the climate system (Box TS.4, Figure 1). Sea level responds to greenhouse gas (GHG) emissions more slowly than global surface temperature, leading to weaker scenario dependence over the 21st century than for global surface temperature (high confidence). This slow response also leads to long-term committed sea level rise, associated with ongoing ocean heat uptake and the slow adjustment of the ice sheets, that will continue over the centuries and millennia following cessation of emissions (high confidence) (Box TS.9). By 2100, GMSL is projected to rise by 0.28–0.55 m (likely range) under SSP1-1.9 and 0.63–1.01 m (likely range) under SSP5-8.5 relative to the 1995–2014 average (medium confidence). Under the higher CO₂ emissions scenarios, there is deep uncertainty in sea level projections for 2100 and beyond associated with the ice-sheet responses to warming. In a low-likelihood, high-impact storyline and a high CO₂ emissions scenario, ice-sheet processes characterized by deep uncertainty could drive GMSL rise up to about 5 m by 2150. Given the long-term commitment, uncertainty in the timing of reaching different GMSL rise levels is an important consideration for adaptation planning. Links to chapters2.3, 3.4, 3.5, 9.6, Box 9.4, Cross-Chapter Box 9.1, Table 9.5

GMSL change is driven by warming or cooling of the ocean (and the associated expansion/contraction) and changes in the amount of ice and water stored on land. Paleo-evidence shows that GMSL has been about 70 m higher and 130 m lower than present within the past 55 million years and was likely 5 to 10 m higher during the Last Interglacial (Box TS.2, Figure 1). Sea level observations show that GMSL rose by 0.20 [0.15 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-evidence since AR5 show this rate is very likely faster than during any century over at least the last three millennia (high confidence). Since AR5, there is strengthened evidence for an increase in the rate of GMSL rise since the mid-20th century, with an average rate of 2.3 [1.6–3.1] mm yr^–1 over the period 1971–2018 increasing to 3.7 [3.2–4.2] mm yr^–1 for the period 2006–2018 (high confidence). Links to chapters2.3.3, 9.6.1, 9.6.2

Box TS.4, Figure 1 | Global mean sea level (GMSL) change on different time scales and under different scenarios. The intent of this figure is to (i) show the century-scale GMSL projections in the context of the 20th century observations, (ii) illustrate ‘deep uncertainty’ in projections by considering the timing of GMSL rise milestones, and (iii) show the long-term commitment associated with different warming levels, including the paleo evidence to support this. (a) GMSL change from 1900 to 2150, observed (1900–2018) and projected under the SSP scenarios (2000–2150), relative to a 1995–2014 baseline. Solid lines show median projections. Shaded regions showlikely ranges for SSP1-2.6 and SSP3-7.0. Dotted and dashed lines show respectively the 83rd and 95th percentilelow confidence projections for SSP5-8.5. Bars at right showlikely ranges for SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5 in 2150. Lightly shaded thick/thin bars show 17th–83rd/5th–95th percentile low-confidence ranges in 2150 for SSP1-2.6 and SSP5-8.5, based upon projection methods incorporating structured expert judgement and marine ice cliff instability. Low confidence range for SSP5-8.5 in 2150 extends to 4.8/5.4 m at the 83rd/95th percentile. (b) GMSL change on 100- (blue), 2000- (green) and 10,000-year (magenta) time scales as a function of global surface temperature, relative to 1850–1900. For 100-year projections, GMSL is projected for the year 2100, relative to a 1995–2014 baseline, and temperature anomalies are average values over 2081–2100. For longer-term commitments, warming is indexed by peak warming above 1850–1900 reached after cessation of emissions. Shaded regions show paleo-constraints on global surface temperature and GMSL for the Last Interglacial and mid-Pliocene Warm Period. Lightly shaded thick/thin blue bars show 17th–83rd/5th–95th percentile low confidence ranges for SSP1-2.6 and SSP5-8.5 in 2100, plotted at 2°C and 5°C. (c) Timing of exceedance of GMSL thresholds of 0.5, 1.0, 1.5 and 2.0 m, under different SSPs. Lightly shaded thick/thin bars show 17th–83rd/5th–95th percentile low-confidence ranges for SSP1-2.6 and SSP5-8.5. Links to chapters4.3.2, 9.6.1, 9.6.2, 9.6.3, Box 9.4

GMSL will continue to rise throughout the 21st century (Box TS.4, Figure 1a). Considering only those processes in whose projections we have at least medium confidence, relative to the period 1995–2014, 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, likely range; SSP5-8.5) by 2050. By 2100, the projected rise is between 0.38 m (0.28–0.55 m, likely range; SSP1-1.9) and 0.77 m (0.63–1.01 m, likely range; SSP5-8.5) Links to chaptersTable 9.9 . The methods, models and scenarios used for sea level projections in the AR6 are updated from those employed by SROCC, with contributions informed by the latest model projections described in the ocean and cryosphere Sections (Sections TS.2.4 and TS.2.5). Despite these differences, the sea level projections are broadly consistent with those of SROCC. Links to chapters4.3.2, 9.6.3

Importantly, likely range projections do not include those ice-sheet-related processes whose quantification is highly uncertain or that are characterized by deep uncertainty. Higher amounts of GMSL rise before 2100 could be caused by earlier-than-projected disintegration of marine ice shelves, the abrupt, widespread onset of marine ice sheet instability (MISI) and marine ice cliff instability (MICI) around Antarctica, and faster-than-projected changes in the surface mass balance and dynamical ice loss from Greenland (Box TS.4, Figure 1). In a low-likelihood, high-impact storyline and a high CO₂ emissions scenario, such processes could in combination contribute more than one additional meter of sea level rise by 2100 (Box TS.3). Links to chapters4.3.2, 9.6.3, Box 9.4

Beyond 2100, GMSL will continue to rise for centuries to millennia due to continuing deep ocean heat uptake and mass loss from ice sheets, and will remain elevated for thousands of years (high confidence). By 2150, considering only those processes in whose projections we have at least medium confidence and assuming no acceleration in ice-mass flux after 2100, GMSL is projected to rise between 0.6 m (0.4–0.9 m, likely range, SSP1-1.9) and 1.3 m (1.0–1.9 m, likely range) (SSP5-8.5), relative to the period 1995–2014 based on the SSP scenario extensions. Under high CO₂ emissions, processes in which there is low confidence, such as MICI, could drive GMSL rise up to about 5 m by 2150 (Box TS.4, Figure 1a). By 2300, GMSL will rise 0.3–3.1 m under low CO₂ emissions (SSP1-2.6) (low confidence). Under high CO₂ emissions (SSP5-8.5), projected GMSL rise is between 1.7 and 6.8 m by 2300 in the absence of MICI and by up to 16 m considering MICI (low confidence). Over 2000 years, there is medium agreemen t and limited evidence that committed GMSL rise is projected to be about 2–3 m with 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 4°C of peak warming, and 19–22 m with 5°C of peak warming. Links to chapters9.6.3

Looking at uncertainty in time provides an alternative perspective on uncertainty in future sea level rise (Box TS.4, Figure 1c). For example, considering only medium confidence processes, GMSL rise is likely to exceed 0.5 m between about 2080 and 2170 under SSP1-2.6 and between about 2070 and 2090 under SSP5-8.5. Given the long-term commitment, uncertainty in the timing of reaching different levels of GMSL rise is an important consideration for adaptation planning. Links to chapters9.6.3

At regional scales, additional processes come into play that modify the local sea level change relative to GMSL, including vertical land motion, ocean circulation and density changes, and gravitational, rotational, and deformational effects arising from the redistribution of water and ice mass between land and the ocean. These processes give rise to a spatial pattern that tends to increase sea level rise at the low latitudes and reduce sea level rise at high latitudes. However, over the 21st century, the majority of coastal locations have a median projected regional sea level rise within ±20% of the projected GMSL change (medium confidence). Further details on regional sea level change and extremes are provided in Section TS.4. Links to chapters9.6.3

The continued growthof atmospheric CO₂ concentrations over the industrial era is unequivocally due to emissions from human activities. Ocean and land carbon sinks slow the rise of CO₂ in the atmosphere. Projections show that while land and ocean sinks absorb more CO₂ under high emissions scenarios than low emissions scenarios, the fraction of emissions removed from the atmosphere by natural sinks decreases with higher concentrations (high confidence). Projected ocean and land sinks show similar responses for a given scenario, but the land sink has a much higher interannual variability and wider model spread. The slowed growth rates of the carbon sinks projected for the second half of this century are linked to strengthening carbon–climate feedbacks and stabilization of atmospheric CO₂ under medium-to-no-mitigation and high-mitigation scenarios, respectively (see FAQ 5.1). Links to chapters5.2, 5.4

Carbon sinks for anthropogenic CO₂ are associated with mainly physical ocean and biospheric land processes that drive the exchange of carbon between multiple land, ocean and atmospheric reservoirs. These exchanges are driven by increasing atmospheric CO₂, but are modulated by changes in climate (Box TS.5, Figure 1c,d). The Northern and Southern Hemispheres dominate the land and ocean sinks, respectively (Box TS.5, Figure 1). Ocean circulation and thermodynamic processes also play a critical role in coupling the global carbon and energy (heat) cycles. There is high confidence that this ocean carbon–heat nexus is an important basis for one of the most important carbon–climate metrics, the transient climate response to cumulative CO₂ emissions (TCRE; Section TS.3.2.1) used to determine the remaining carbon budget. Links to chapters5.1, 5.2, 5.5, 9.2, Cross-Chapter Box 5.3

Based on multiple lines of evidence using interhemispheric gradients of CO₂ concentrations, isotopes, and inventory data, it is unequivocal that the growth in CO₂ in the atmosphere since 1750 (see Section TS.2.2) is due to the direct emissions from human activities. The combustion of fossil fuels and land-use change for the period 1750–2019 resulted in the release of 700 ± 75 PgC (likely range, 1 PgC = 10¹⁵g of carbon) to the atmosphere, of which about 41% ± 11% remains in the atmosphere today (high confidence). Of the total anthropogenic CO₂ emissions, the combustion of fossil fuels was responsible for about 64% ± 15%, growing to an 86% ± 14% contribution over the past 10 years. The remainder resulted from land-use change. During the last decade (2010–2019), average annual anthropogenic CO₂ emissions reached the highest levels in human history at 10.9 ± 0.9 PgC yr^–1 (high confidence). Of these emissions, 46% accumulated in the 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 PgC yr^–1) was removed by terrestrial ecosystems (high confidence). Links to chapters5.2.1, 5.2.2, 5.2.3

The ocean (high confidence) and land (medium confidence) sinks of CO₂ have increased with anthropogenic emissions over the past six decades (Box TS.5, Figure 1). This coherence between emissions and the growth in ocean and land sinks has resulted in the airborne fraction of anthropogenic CO₂ remaining at 44 ± 10% over the past 60 years (high confidence). Interannual and decadal variability of the ocean and land sinks indicate that they are sensitive to changes in the growth rate of emissions as well as climate variability and are therefore also sensitive to climate change (high confidence). Links to chapters5.2.1

The land CO₂ sink is driven by carbon uptake by vegetation, with large interannual variability, for example, linked to the El Niño–Southern Oscillation (ENSO). Since the 1980s, carbon fertilization from rising atmospheric CO₂ has increased the strength of the net land CO₂ sink (medium confidence). During the historical period, the growth of the ocean sink has been primarily determined by the growth rate of atmospheric CO₂. However, there is medium confidence that changes to physical and chemical processes in the ocean and in the land biosphere, which govern carbon feedbacks, are already modifying the characteristics of variability, particularly the seasonal cycle of CO₂, in both the ocean and land. However, changes to the multi-decadal trends in the sinks have not yet been observed. Links to chapters2.3.4, 3.6.1, 5.2.1

In AR6, ESM projections are assessedwith CO₂ concentrations by 2100 from about 400 ppm (SSP1-1.9) to above 1100 ppm (SSP5-8.5). Most simulations are performed with prescribed atmospheric CO₂ concentrations, which already account for a central estimate of climate–carbon feedback effects. Carbon dioxide emissions-driven simulations account for uncertainty in these feedbacks, but do not significantly change the projected global surface temperature changes (high confidence). Although land and ocean sinks absorb more CO₂ under high emissions than low emissions scenarios, the fraction of emissions removed from the atmosphere decreases (high confidence). This means that the more CO₂ that is emitted, the less efficient the ocean and land sinks become (high confidence), an effect which compensates for the logarithmic relationship between CO₂ and its radiative forcing, which means that for each unit increase in additional atmospheric CO₂ the effect on global temperature decreases. (Box TS.5, Figure 1f,g). Links to chapters4.3.1, 5.4.5, 5.5.1.2

Ocean and land sinks show similar responses for a given scenario, but the land sink has a much higher interannual variability and wider model spread. Under SSP3-7.0 and SSP5-8.5, the initial growth of both sinks in response to increasing atmospheric concentrations of CO₂ is subsequently limited by emerging carbon–climate feedbacks (high confidence) (Box TS.5, Figure 1f). Projections show that the ocean and land sinks will stop growing from the second part of the 21st century under all emissions scenarios, but with different drivers for different emissions scenarios. Under SSP3-7.0 and SSP5-8.5, the weakening growth rate of the ocean CO₂ sink in the second half of the century is primarily linked to the strengthening positive feedback from reduced carbonate buffering capacity, ocean warming and altered ocean circulation (e.g., AMOC changes). In contrast, for SSP1-1.9, SSP1-2.6 and SSP2-4.5, the weakening growth rate of the ocean carbon sink is a response to the stabilizing or declining atmospheric CO₂ concentrations. Under SSP1-1.9, models project that combined land and ocean sinks will turn into a weak source by 2100 (medium confidence). Under high CO₂ emissions scenarios, it is very likely that the land carbon sink will grow more slowly due to warming and drying from the mid-21st century, but it is very unlikely that it will switch from being a sink to a source before 2100.

Box TS.5, Figure 1 | Carbon cycle processes and projections. The intent of this figure is to show the response of the carbon cycle to carbon dioxide (CO2) emissions and climate and its role in determining future CO2levels through projected changes to sinks and sink fractions. The figure shows changes in carbon storage in response to elevated CO₂ (a, b) and the response to climate warming (c, d). Maps show spatial patterns of changes in carbon uptake during simulations with 1% per year increase in CO₂ (Section 5.4.5.5), and zonal mean plots show distribution of carbon changes is dominated by the land (green lines) in the tropics and Northern Hemisphere and ocean (blue lines) in the Southern Hemisphere. Hatching indicates regions where fewer than 80% of models agree on the sign of response. (e) Future CO₂ projections: projected CO₂ concentrations in the Shared Socio-economic Pathway (SSP) scenarios in response to anthropogenic emissions, results from coupled Earth system models for SSP5-8.5 and from the MAGICC7 emulator for other scenarios (Section 4.3.1). (f) Future carbon fluxes: projected combined land and ocean fluxes (positive downward) up to 2100 for the SSP scenarios, and extended to 2300 for available scenarios, 5–95% uncertainty plumes shown for SSP1-2.6 and SSP3-7.0 (Sections 4.3.2.4, 5.4.5.4 and 5.4.10). The numbers near the top show the number of model simulations used. (g) Sink fraction: the fraction of cumulative emissions of CO₂ removed by land and ocean sinks. The sink fraction is smaller under conditions of higher emissions. Links to chaptersFigure 4.3; 5.4.5; Figures 5.25, 5.27 and 5.30

Climate change alone is expected to increase land carbon accumulation in the high latitudes (not including permafrost, which is assessed in Sections TS.2.5 and TS.3.2.2), but also to lead to a counteracting loss of land carbon in the tropics (medium confidence). Earth system model projections show that the overall uncertainty of atmospheric CO₂ by 2100 is still dominated by the emissions pathway, but carbon–climate feedbacks (see Section TS.3.3.2) are important, with increasing uncertainties in high emissions pathways (Box TS.5, Figure 1e). Links to chapters4.3.2, 5.4.1, 5.4.2, 5.4.4, 5.4.5, 11.6, 11.9, Cross-Chapter Box 5.1, Cross-Chapter Box 5.3

Under three SSP scenarios with long-term extensions until 2300 (SSP5-8.5, SSP5-3.4-OS, SSP1-2.6), ESMs project a change of the land from a sink to a source (medium confidence). The scenarios make simplified assumptions about emissions reductions, with SSP1-2.6 and SSP5-3.4-OS reaching about 400 ppm by 2300, while SSP5-8.5 exceeds 2000 ppm. Under high emissions, the transition is warming-driven, whereas it is linked to the decline in atmospheric CO₂ under net negative CO₂ emissions. The ocean remains a sink throughout the period to 2300 except under very large net negative emissions. The response of the natural aspects of the carbon cycle to carbon dioxide removal is further developed in Section TS.3.3.2. Links to chapters5.4.9

Human-caused climate change has driven detectable changes in the global water cycle since the mid-20th century (high confidence), and it is projected to cause substantial further changes at both global and regional scales (high confidence).

Global land precipitation has likely increased since 1950, with a faster increase since the 1980s (medium confidence). Atmospheric water vapour has increased throughout the troposphere since at least the 1980s (likely) . Annual global land precipitation will increase over the 21st century as global surface temperature increases (high confidence). Human influence has been detected in amplified surface salinity and precipitation minus evaporation (P–E) patterns over the ocean (high confidence).

The severity of very wet and very dry events increase in a warming climate (high confidence), but changes in atmospheric circulation patterns affect where and how often these extremes occur. Water cycle variability and related extremes are projected to increase faster than mean changes in most regions of the world and under all emissions scenarios (high confidence).

Over the 21st century, the total land area subject to drought will increase and droughts will become more frequent and severe (high confidence). Near-term projected changes in precipitation are uncertain mainly because of internal variability, model uncertainty and uncertainty in forcings from natural and anthropogenic aerosols (medium confidence).

Over the 21st century and beyond, abrupt human-caused changes to the water cycle cannot be excluded (medium confidence). Links to chapters2.3, 3.3, 4.3, 4.4, 4.5, 4.6, 8.2, 8.3, 8.4, 8.5, 8.6, 11.4, 11.6, 11.9

There is high confidence that the global water cycle has intensified since at least 1980 expressed by, for example, increased atmospheric moisture fluxes and amplified precipitation minus evaporation patterns. Global land precipitation has likely increased since 1950, with a faster increase since the 1980s (medium confidence), and a likely human contribution to patterns of change, particularly for increases in high-latitude precipitation over the Northern Hemisphere. Increases in global mean precipitation are determined by a robust response to global surface temperature (very likely 2–3% per °C) that is partly offset by fast atmospheric adjustments to atmospheric heating by greenhouse gases (GHGs) and aerosols (Section TS.3.2.2). The overall effect of anthropogenic aerosols is to reduce global precipitation through surface radiative cooling effects (high confidence). Over much of the 20th century, opposing effects of GHGs and aerosols on precipitation have been observed for some regional monsoons (high confidence) (Box TS.13). Global annual precipitation over 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% 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 relative to 1995–2014 (Box TS.6, Figure 1). Inter-model differences and internal variability contribute to a substantial range in projections of large-scale and regional water cycle changes (high confidence). The occurrence of volcanic eruptions can alter the water cycle for several years (high confidence). Projected patterns of precipitation change exhibit substantial regional differences and seasonal contrast as global surface temperature increases over the 21st century (Box TS.6, Figure 1). Links to chapters2.3.1, 3.3.2, 3.3.3, 3.5.2, 4.3.1, 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

Global total column water vapour content has very likely increased since the 1980s, and it is likely that human influence has contributed to tropical upper tropospheric moistening. Near-surface specific humidity has increased over the ocean (likely) and land (very likely) since at least the 1970s, with a detectable human influence (medium confidence). Human influence has been detected in amplified surface salinity and precipitation minus evaporation (P–E) patterns over the ocean (high confidence). It is virtually certain that evaporation will increase over the ocean and very likely that evapotranspiration will increase over land, with regional variations under future surface warming (Box TS.6, Figure 1). There is high confidence that projected increases in precipitation amount and intensity will be associated with increased runoff in northern high latitudes (Box TS.6, Figure 1). In response to cryosphere changes (Section TS.2.5), there have been changes in streamflow seasonality, including an earlier occurrence of peak streamflow in high-latitude and mountain catchments (high confidence). Projected runoff (Box TS.6, Figure 1c) is typically decreased by contributions from small glaciers because of glacier mass loss, while runoff from larger glaciers will generally increase with increasing global warming levels until their mass becomes depleted (high confidence). Links to chapters2.3.1, 3.3.2, 3.3.3, 3.5.2, 8.2.3, 8.4.1, 11.5

Warming over land drives an increase in atmospheric evaporative demand and in the severity of drought events (high confidence). Greater warming over land than over the ocean alters atmospheric circulation patterns and reduces continental near-surface relative humidity, which contributes to regional drying (high confidence). Avery likely decrease in relative humidity has occurred over much of the global land area since 2000. Projected increases in evapotranspiration due to growing atmospheric water demand will decrease soil moisture over the Mediterranean region, south-western North America, South Africa, South-Western South America and south-western Australia (high confidence) (Box TS.6, Figure 1). Some tropical regions are also projected to experience enhanced aridity, including the Amazon basin and Central America (high confidence). The total land area subject to increasing drought frequency and severity will expand (high confidence), and in the Mediterranean, South-Western South America, and Western North America, future aridification will far exceed the magnitude of change seen in the last millennium (high confidence). Links to chapters4.5.1, 8.2.2, 8.2.3, 8.4.1, Box 8.2, 11.6, 11.9

Land-use change and water extraction for irrigation have influenced local and regional responses in the water cycle (high confidence). Large-scale deforestation likely decreases evapotranspiration and precipitation and increases runoff over the deforested regions relative to the regional effects of climate change (medium confidence). Urbanization increases local precipitation (medium confidence) and runoff intensity (high confidence) (Box TS.14). Increased precipitation intensities have enhanced groundwater recharge, most notably in tropical regions (medium confidence). There is high confidence that groundwater depletion has occurred since at least the start of the 21st century, as a consequence of groundwater withdrawals for irrigation in agricultural areas in drylands. Links to chapters8.2.3, 8.3.1, 11.1.6, 11.4, 11.6, FAQ 8.1

Water cycle variability and related extremes are projected to increase faster than mean changes in most regions of the world and under all emissions scenarios (high confidence). A warmer climate increases moisture transport into weather systems, which intensifies wet seasons and events (high confidence). The magnitudes of projected precipitation increases and related extreme events depend on model resolution and the representation of convective processes (high confidence). Increases in near-surface atmospheric moisture capacity of about 7% per 1ºC of warming lead to a similar response in the intensification of heavy precipitation from sub-daily up to seasonal time scales, increasing the severity of flood hazards (high confidence). The average and maximum rain-rates associated with tropical and extratropical cyclones, atmospheric rivers and severe convective storms will therefore also increase with future warming (high confidence). For some regions, there is medium confidence that peak tropical cyclone rain-rates will increase by more than 7% per 1°C of warming due to increased low-level moisture convergence caused by increases in wind intensity. In the tropics year-round and in the summer season elsewhere, interannual variability of precipitation and runoff over land is projected to increase at a faster rate than changes in seasonal mean precipitation (Figure TS.12e,f) (medium confidence). Sub-seasonal precipitation variability is also projected to increase, with fewer rainy days but increased daily mean precipitation intensity over many land regions (high confidence). Links to chapters4.5.3, 8.2.3, 8.4.1, 8.4.2, 8.5.1, 8.5.2, 11.4, 11.5, 11.7, 11.9

Box TS.6, Figure 1 | Projected water cycle changes. The intent of this figure is to give a geographical overview of changes in multiple components of the global water cycle using an intermediate emissions scenario. Important key message: without drastic reductions in greenhouse gas emissions, human-induced global warming will be associated with widespread changes in all components of the water cycle. Long-term (2081–2100) projected annual mean changes (%) relative to present-day (1995–2014) in the SSP2-4.5 emissions scenario for (a) precipitation, (b) surface evapotranspiration, (c) total runoff and (d) surface soil moisture. Numbers in top right of each panel indicate indicate the number of Coupled Model Intercomparison Project Phase 6 (CMIP6) models used for estimating the ensemble mean. For other scenarios, please refer to relevant figures in Chapter 8. Uncertainty is represented using the simple approach: No overlay indicates regions with high model agreement, where ≥80% of models agree on sign of change; diagonal lines indicate regions with low model agreement, where <80% of models agree on sign of change. For more information on the simple approach, please refer to the Cross-Chapter Box Atlas.1. Links to chapters8.4.1; Figures 8.14, 8.17, 8.18, and 8.19

(top left) Annual emissions of CO₂ for the five core Shared Socio-economic Pathway (SSP) scenarios (very low: SSP1-1.9, low: SSP1-2.6, intermediate: SSP2-4.5, high: SSP3-7.0, very high: SSP5-8.5). (bottom left) Projected warming for each of these emissions scenarios. (top right) Response of some selected climate variables to four levels of global warming (°C). Changes in the ‘Today’ column are based on a global warming level of 1°C. (bottom right) The long-term effect of each global warming level on sea level. See Section TS.1.3.1 for more detail on the SSP climate change scenarios. This infographic builds from several figures in the Technical Summary: Figure TS.4 (for top left panel), Figure TS.6 (bottom left), Figure TS.12 (top right) and Box TS.4, Figure 1b (bottom right).

Future changes in emissions of short-lived climate forcers (SLCFs) are expected to cause an additional global mean warming, with a large diversity in the end-of-century response across the WGI core set of Shared Socio-economic Pathways (SSPs), depending upon the level of climate change and air pollution mitigation (Box TS.7, Figure 1). This additional warming is either due to reductions in cooling aerosols for air pollution regulation or due to increases in methane (CH₄), ozone and hydrofluorocarbons (HFCs). This additional warming is stable after 2040 in SSPs associated with lower global air pollution as long as CH₄ emissions are also mitigated, but the overall warming induced by SLCF changes is higher in scenarios in which air quality continues to deteriorate (induced by growing fossil fuel use and limited air pollution control) (high confidence).

Sustained CH₄ mitigation reduces global surface ozone, contributing to air quality improvements, and also reduces surface temperature in the longer term, but only sustained CO₂ emissions reductions allow long-term climate stabilization (high confidence). Future changes in air quality (near-surface ozone and particulate matter, or PM) at global and local scales are predominantly driven by changes in ozone and aerosol precursor emissions rather than climate (high confidence). Air quality improvements driven by rapid decarbonization strategies, as in SSP1-1.9 and SSP1-2.6, are not sufficient in the near term to achieve air quality guidelines set by the World Health Organization in some highly polluted regions (high confidence). Additional policies (e.g., access to clean energy, waste management) envisaged to attain United Nations Sustainable Development Goals bring complementary SLCF reduction. Links to chapters4.4.4, 6.6.3, 6.7.3, Box 6.2

The net effect of SLCF emissions changes on temperature will depend on how emissions of warming and cooling SLCFs will evolve in the future. The magnitude of the cooling effect of aerosols remains the largest uncertainty in the effect of SLCFs in future climate projections. Since the SLCFs have undergone large changes over the past two decades, the temperature and air pollution responses are estimated relative to the year 2019 instead of 1995–2014.

Temperature Response

In the next two decades, it is very likely that SLCF emissions changes will cause a warming relative to 2019, across the WGI core set of SSPs (see Section TS.1.3.1), in addition to the warmingfrom long-lived GHGs. The net effect of SLCF and HFC changes in global surface temperature across the SSPs is a likely warming of 0.06°C–0.35°C in 2040 relative to 2019. This near-term global mean warming linked to SLCFs is quite similar in magnitude across the SSPs due to competing effects of warming (CH₄, ozone) and cooling (aerosols) forcers (Box TS.7, Figure 1). There is greater diversity in the end-of-century response among the scenarios. SLCF changes in scenarios with no climate change mitigation (SSP3-7.0 and SSP5-8.5) will cause a warming in the likely range of 0.4°C–0.9°C in 2100 relative to 2019 due to increases in CH₄, tropospheric ozone and HFC levels. For the stringent climate change and pollution mitigation scenarios (SSP1-1.9 and SSP1-2.6), the cooling from reductions in CH₄, ozone and HFCs partially balances the warming from reduced aerosols, primarily sulphate, and the overall SLCF effect is a likely increase in global surface temperature of 0.0°C–0.3°C in 2100, relative to 2019. With intermediate climate change and air pollution mitigations, SLCFs in SSP2-4.5 add a likely warming of 0.2°C–0.5°C to global surface temperature change in 2100, with the largest warming resulting from reductions in aerosols. Links to chapters4.4.4, 6.7.3

Assuming implementation and efficient enforcement of both the Kigali Amendment to the Montreal Protocol on Substances that Deplete the Ozone Layer and current national plans result in limiting emissions (as in SSP1-2.6), the effects of HFCs on global surface temperature, relative to 2019, would remain below +0. 02°C from 2050 onwards versus about +0.04°C–0.08°C in 2050 and +0.1°C–0.3°C in 2100 considering only national HFC regulations decided prior to the Kigali Amendment (as in SSP5-8.5) (medium confidence). Links to chapters6.6.3, 6.7.3

Air Quality Responses

Air pollution projections range from strong reductions in global surface ozone and PM (e.g., SSP1-2.6, with stringent mitigation of both air pollution and climate change) to no improvement and even degradation (e.g., SSP3-7.0 without climate change mitigation and with only weak air pollution control) (high confidence). Under the SSP3-7.0 scenario, PM levels are projected to increase until 2050 over large parts of Asia, and surface ozone pollution is projected to worsen over all continental areas through 2100 (high confidence). In SSP5-8.5, a scenario without climate change mitigation but with stringent air pollution control, PM levels decline through 2100, but high CH₄ levels hamper the decline in global surface ozone at least until 2080 (high confidence). Links to chapters6.7.1

Box TS.7, Figure 1 | Effects of short-lived climate forcers (SLCFs) on global surface temperature and air pollution across the WGI core set of Shared Socio-economic Pathways (SSPs). The intent of this figure is to show the climate and air quality (surface ozone and particulate matter smaller than 2.5 microns in diameter, or PM2.5) response to SLCFs in the SSP scenarios for the near and long-term. Effects of net aerosols, tropospheric ozone, hydrofluorocarbons (HFCs; with lifetimes less than 50 years), and methane (CH₄) are compared with those of total anthropogenic forcing for 2040 and 2100 relative to year 2019. The global surface temperature changes are based on historical and future evolution of effective radiative forcing (ERF) as assessed in Chapter 7 of this Report. The temperature responses to the ERFs are calculated with a common impulse response function (RT) for the climate response, consistent with the metric calculations in Chapter 7 (Box 7.1). The RT has an equilibrium climate sensitivity of 3.0°C for a doubling of atmospheric CO₂ concentration (feedback parameter of –1.31 W m^–2°C^–1). The scenario total (grey bar) includes all anthropogenic forcings (long- and short-lived climate forcers, and land-use changes). Uncertainties are 5–95% ranges. The global changes in air pollutant concentrations (ozone and PM2.5) are based on multimodel Coupled Model Intercomparison Project Phase 6 (CMIP6) simulations and represent changes in five-year mean surface continental concentrations for 2040 and 2098 relative to 2019. Uncertainty bars represent inter-model ±1 standard deviation. Links to chapters6.7.2, 6.7.3, Figure 6.24

Since AR5, further modelling work has been conducted on aerosol-based solar radiation modification (SRM) options such as stratospheric aerosol injection, marine cloud brightening, and cirrus cloud thinning²¹ and their climate and biogeochemical effects. These investigations have consistently shown that SRM could offset some of the effects of increasing greenhouse gases on global and regional climate, including the carbon and water cycles (high confidence). However, there would be substantial residual or overcompensating climate change at the regional scales and seasonal time scales (high confidence), and large uncertainties associated with aerosol–cloud–radiation interactions persist. The cooling caused by SRM would increase the global land and ocean CO₂ sinks (medium confidence), but this would not stop CO₂ from increasing in the atmosphere or affect the resulting ocean acidification under continued anthropogenic emissions (high confidence). It is likely that abrupt water cycle changes will occur if SRM techniques are implemented rapidly. A sudden and sustained termination of SRM in a high CO₂ emissions scenario would cause rapid climate change (high confidence). However, a gradual phase-out of SRM combined with emissions reduction and carbon dioxide removal (CDR) would avoid these termination effects (medium confidence). Links to chapters4.6.3, 5.6.3. 6.4.6, 8.6.3 .

Solar radiation modification (SRM) refers to deliberate, large-scale climate intervention options that are studied as potential supplements to deep mitigation, for example, in scenarios that overshoot climate stabilization goals. SRM options aim to offset some of the warming effects of GHG emissions by modification of Earth’s shortwave radiation budget. Following SR1.5, the SRM assessed in this Report also includes some options, such as cirrus cloud thinning, that alter the longwave radiation budget.

SRM contrasts with climate change mitigation activities, such as emissions reductions and CDR, as it introduces a ‘mask’ to the climate change problem by altering Earth’s radiation budget, rather than attempting to address the root cause of the problem, which is the increase in GHGs in the atmosphere. By masking only the climate effects of GHG emissions, SRM does not address other issues related to atmospheric CO₂ increase, such as ocean acidification. This Report assesses physical understanding of the Earth system response to proposed SRM, and the assessment is based primarily on idealized climate model simulations. There are other important considerations, such as risk to human and natural systems, perceptions, ethics, cost, governance, and trans-boundary issues and their relationship to the United Nations Sustainable Development Goals – issues that the WGII (Chapter 16) and WGIII (Chapter 14) Reports address. Links to chapters4.6.3

SRM options include those that increase surface albedo, brighten marine clouds by increasing the amount of cloud condensation nuclei, or reduce the optical depth of cirrus clouds by seeding them with ice nucleating particles. However, the most commonly studied approaches attempt to mimic the cooling effects of major volcanic eruptions by injecting reflective aerosols (e.g., sulphate aerosols) or their precursors (e.g., sulphur dioxide) into the stratosphere. Links to chapters4.6.3, 5.6.3, 6.4.6

SRM could offset some effects of greenhouse gas-induced warming on global and regional climate, but there would be substantial residual and overcompensating climate change at the regional scale and seasonal time scales (high confidence). Since AR5, more modelling work has been conducted with more sophisticated treatment of aerosol-based SRM approaches, but the uncertainties in cloud–aerosol–radiation interactions are still large (high confidence). Modelling studies suggest that it is possible to stabilize multiple large-scale temperature indicators simultaneously by tailoring the deployment strategy of SRM options (medium confidence) but with large residual or overcompensating regional and seasonal climate changes. Links to chapters4.6.3

SRM approaches targeting shortwave radiation are likely to reduce global mean precipitation, relative to future CO₂ emissions scenarios, if all global mean warming is offset. In contrast, cirrus cloud thinning, targeting longwave radiation, is expected to cause an increase in global mean precipitation (medium confidence). If shortwave approaches are used to offset global mean warming, the magnitude of reduction in regional precipitation minus evapotranspiration (P–E) (Box TS.5), which is more relevant to freshwater availability, is smaller than precipitation decrease because of simultaneous reductions in both precipitation and evapotranspiration (medium confidence). Links to chapters4.6.3, 8.2.1, 8.6.3 .

If SRM is used to cool the planet, it would cause a reduction in plant and soil respiration and slow the reduction of ocean carbon uptake due to warming (medium confidence). The result would be an enhancement of the global land and ocean CO₂ sinks (medium confidence) and a slight reduction in atmospheric CO₂ concentration relative to unmitigated climate change. However, SRM would not stop CO₂ from increasing in the atmosphere or affect the resulting ocean acidification under continued anthropogenic emissions (high confidence). Links to chapters5.6.3

The effect of stratospheric aerosol injection on global temperature and precipitation is projected by models to be detectable after one to two decades, which is similar to the time scale for the emergence of the benefits of emissions reductions. A sudden and sustained termination of SRM in a high GHG emissions scenario would cause rapid climate change and a reversal of the SRM effects on the carbon sinks (high confidence). It is also likely that a termination of strong SRM would drive abrupt changes in the water cycle globally and regionally, especially in the tropical regions by shifting the Inter-tropical Convergence Zone and Hadley cells. At the regional scale, non-linear responses cannot be excluded, due to changes in evapotranspiration. However, a gradual phase-out of SRM combined with emissions reductions and CDR would avoid larger rates of changes (medium confidence). Links to chapters4.6.3, 5.6.3, 8.6.3 .

The present rates of response of many aspects of the climate system are proportionate to the rate of recent temperature change, but some aspects may respond disproportionately. Some climate system components are slow to respond, such as the deep ocean overturning circulation and the ice sheets (Box TS.4). It is virtually certain that irreversible, committed change is already underway for the slow-to-respond processes as they come into adjustment for past and present emissions.

The paleoclimate record indicates that tipping elements exist in the climate system where processes undergo sudden shifts toward a different sensitivity to forcing, such as during a major deglaciation, where 1°C degree of temperature change might correspond to a large or small ice-sheet mass loss during different stages (Box TS.2). For global climate indicators, evidence for abrupt change is limited, but deep ocean warming, acidification and sea level rise are committed to ongoing change for millennia after global surface temperatures initially stabilize and are irreversible on human time scales (very high confidence). At the regional scale, abrupt responses, tipping points and even reversals in the direction of change cannot be excluded (high confidence). Some regional abrupt changes and tipping points could have severe local impacts, such as unprecedented weather, extreme temperatures and increased frequency of droughts and forest fires.

Models that exhibit such tipping points are characterized by abrupt changes once the threshold is crossed, and even a return to pre-threshold surface temperatures or to atmospheric carbon dioxide concentrations does not guarantee that the tipping elements return to their pre-threshold state. Monitoring and early warning systems are being put into place to observe tipping elements in the climate system. Links to chapters1.3, 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, 9.4.1, 9.4.2, 9.6.3, Cross-chapter Box 12.1

Understanding of multi-decadal reversibility (i.e., the system returns to the previous climate state within multiple decades after the radiative forcing is removed) has improved since AR5 for many atmospheric, land surface and sea ice climate metrics following sea surface temperature recovery. Some processes suspected of having tipping points, such as the Atlantic Meridional Overturning Circulation (AMOC), have been found to often undergo recovery after temperature stabilization with a time delay (low confidence). However, substantial irreversibility is further substantiated for some cryosphere changes, ocean warming, sea level rise, and ocean acidification. Links to chapters4.7.2, 5.3.3, 5.4.9, 9.2.2, 9.2.4, 9.4.1, 9.4.2, 9.6.3

Some climate system components are slow to respond, such as the deep ocean overturning circulation and 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 1850–1900, the AMOC will continue to weaken for several decades by about 15%, 20% and 30% of its strength and then recover to pre-decline values over several centuries (medium confidence). At sustained warming levels between 2°C and 3°C, there is limited evidence that the Greenland and West Antarctic ice sheets will be lost almost completely and irreversibly over multiple millennia; both the probability of their complete loss and the rate of mass loss increases with higher surface temperatures (high confidence). At sustained warming levels between 3°C and 5°C, near-complete loss of the Greenland Ice Sheet and complete loss of the West Antarctic Ice Sheet is projected to occur irreversibly over multiple millennia (medium confidence); with substantial parts or all of Wilkes Subglacial Basin in East Antarctica lost over multiple millennia (low confidence). Early-warning signals of accelerated sea level rise from Antarctica could possibly be observed within the next few decades. For other hazards (e.g., ice-sheet behaviour, glacier mass loss and global mean sea level change, coastal floods, coastal erosion, air pollution, and ocean acidification) the time and/or scenario dimensions remain critical, and a simple and robust relationship with global warming level cannot be established (high confidence). Links to chapters4.3.2, 4.7.2, 5.4.3, 5.4.5, 5.4.8, 8.6, 9.2, 9.4, Box 9.3, Cross-Chapter Box 12.1

For global climate indicators, evidence for abrupt change is limited. For global warming up to 2°C above 1850–1900 levels, paleoclimate records do not indicate abrupt changes in the carbon cycle (low confidence). Despite the wide range of model responses, uncertainty in atmospheric CO₂ by 2100 is dominated by future anthropogenic emissions rather than uncertainties related to carbon–climate feedbacks (high confidence). There is no evidence of abrupt change in climate projections of global temperature for the next century: there is a near-linear relationship between cumulative CO₂ emissions and maximum global mean surface air temperature increase caused by CO₂ over the course of this century for global warming levels up to at least 2°C relative to 1850–1900. The increase in global ocean heat content (Section TS.2.4) will likely continue until at least 2300 even for low emissions scenarios, and global mean sea level will continue to rise for centuries to millennia following cessation of emissions (Box TS.4) due to continuing deep ocean heat uptake and mass loss of the Greenland and Antarctic ice sheets (high confidence). Links to chapters2.2.3; Cross-Chapter Box 2.1; 5.1.1; 5.4; Cross-Chapter Box 5.1; Figures 5.3, 5.4, 5.25, and 5.26; 9.2.2; 9.2.4

The response of biogeochemical cycles to anthropogenic perturbations can be abrupt at regional scales and irreversible on decadal to century time scales (high confidence). The probability of crossing uncertain regional thresholds increases with climate change (highBox TS.9

confidence). It is very unlikely that gas clathrates (mostly methane) in deeper terrestrial permafrost and subsea clathrates will lead to a detectable departure from the emissions trajectory during this century. Possible abrupt changes and tipping points in biogeochemical cycles lead to additional uncertainty in 21st century atmospheric GHG concentrations, but future anthropogenic emissions remain the dominant uncertainty (high confidence). There is potential for abrupt water cycle changes in some high emissions scenarios, but there is no overall consistency regarding the magnitude and timing of such changes. Positive land surface feedbacks, including vegetation, dust, and snow, can contribute to abrupt changes in aridity, but there is onlylow confidence that such changes will occur during the 21st century. Continued Amazon deforestation, combined with a warming climate, raises the probability that this ecosystem will cross a tipping point into a dry state during the 21st century (low confidence). (Section TS.3.2.2) Links to chapters5.4.3, 5.4.5, 5.4.8, 5.4.9, 8.6.2, 8.6.3, Cross-Chapter Box 12.1

The attribution of observed changes in extremes to human influence (including greenhouse gas and aerosol emissions and land-use changes) has substantially advanced since AR5, in particular for extreme precipitation, droughts, tropical cyclones, and compound extremes (high confidence). There is limited evidence for windstorms and convective storms. Some recent hot extreme events would have been extremely unlikely to occur without human influence on the climate system. (Section TS.1) Links to chaptersCross-Working Group Box: Attribution in Chapter 1, 11.2, 11.3, 11.4, 11.6, 11.7, 11.8

Since AR5, the attribution of extreme weather events has emerged as a growing field of climate research with an increasing body of literature. It provides evidence that greenhouse gases and other external forcings have affected individual extreme weather events by disentangling anthropogenic drivers from natural variability. Event attribution is now an important line of evidence for assessing changes in extremes on regional scales. (Section TS.1) Links to chaptersCross-Working Group Box: Attribution, 11.1.4

The regional extremes and events that have been studied are geographically uneven (Section TS.4.1). A few events, for example, extreme rainfall events in the United Kingdom, heatwaves in Australia, or Hurricane Harvey that hit Texas in 2017, have been heavily studied. Many highly impactful extreme weather events have not been studied in the event attribution framework, particularly in the developing world where studies are generally lacking. This is due to various reasons, including lack of observational data, lack of reliable climate models, and lack of scientific capacity. While the events that have been studied are not representative of all extreme events that have occurred, and results from these studies may also be subject to selection bias, the large number of event attribution studies provide evidence that changes in the properties of these local and individual events are in line with expected consequences of human influence on the climate and can be attributed to external drivers. Links to chaptersCross-Working Group Box: Attribution, 11.1.4, 11.2.2

It is very likely that human influence is the main contributor to the observed increase in the intensity and frequency of hot extremes and the observed decrease in the intensity and frequency of cold extremes on continental scales. Some specific recent hot extreme events would have been extremely unlikely to occur without human influence on the climate system. Changes in aerosol concentrations have likely slowed the increase in hot extremes in some regions, in particular from 1950–1980. No-till farming, irrigation and crop expansion have similarly attenuated increases in summer hot extremes in some regions, such as central North America (medium confidence). Links to chapters11.3.4

Human influence has contributed to the intensification of heavy precipitation in three continents where observational data are most abundant: North America, Europe and Asia (high confidence). On regional scales, evidence of human influence on extreme precipitation is limited, but new evidence from attributing individual heavy precipitation events found that human influence was a significant driver of the events. Links to chapters11.4.4

There is low confidence that human influence has affected trends in meteorological droughts in most regions, butmedium confidence that they have contributed to the severity of some specific events. There is medium confidence that human-induced climate change has contributed to increasing trends in the probability or intensity of recent agricultural and ecological droughts, leading to an increase of the affected land area. Links to chapters11.6.4

Event attribution studies of specific strong tropical cyclones providelimited evidence for anthropogenic effects on tropical cyclone intensifications so far, but high confidence for increases in precipitation. There is high confidence that anthropogenic climate change contributed to extreme rainfall amounts during Hurricane Harvey (in 2017) and other intense tropical cyclones. Links to chapters11.7.3

The number of evident attribution studies on compound events is limited. There is medium confidence that weather conditions that promote wildfires have become more probable in southern Europe, northern Eurasia, the USA, and Australia over the last century. In Australia a number of event attribution studies show that there is medium confidence of increase in fire weather conditions due to human influence. Links to chapters11.8.3, 12.4.3.2

Box TS.10, Figure 1 | Synthesis of assessed observed and attributable regional changes. The IPCC AR6 WGI inhabited regions are displayed as hexagons of identical sizes in their approximate geographical location (see legend for regional acronyms). All assessments are made for each region as a whole and for the 1950s to the present. Assessments made on different time scales or more local spatial scales might differ from what is shown in the figure. The colours in each panel represent the four outcomes of the assessment on observed changes. Striped hexagons (white and light-grey) are used where there is low agreement in the type of change for the region as a whole, and grey hexagons are used when there is limited data and/or literature that prevents an assessment of the region as a whole. Other colours indicate at least medium confidence in the observed change. The confidence level for the human influence on these observed changes is based on assessing trend detection and attribution and event attribution literature, and it is indicated by the number of dots: three dots for high confidence, two dots formedium confidence and one dot for low confidence (single, filled dot: limited agreement; single, empty dot: limited evidence).

Panel (a) For hot extremes, the evidence is mostly drawn from changes in metrics based on daily maximum temperatures; regional studies using other indices (heatwave duration, frequency and intensity) are used in addition. Red hexagons indicate regions where there is at least medium confidence in an observed increase in hot extremes.

Panel (b) For heavy precipitation, the evidence is mostly drawn from changes in indices based on one-day or five-day precipitation amounts using global and regional studies. Green hexagons indicate regions where there is at least medium confidence in an observed increase in heavy precipitation.

Panel (c) Agricultural and ecological droughts are assessed based on observed and simulated changes in total column soil moisture, complemented by evidence on changes in surface soil moisture, water balance (precipitation minus evapotranspiration) and indices driven by precipitation and atmospheric evaporative demand. Yellow hexagons indicate regions where there is at least medium confidence in an observed increase in this type of drought and green hexagons indicate regions where there is at least medium confidence in an observed decrease in agricultural and ecological drought.

For all regions, Table TS.5 shows a broader range of observed changes besides the ones shown in this figure. Note that Southern South America (SSA) is the only region that does not display observed changes in the metrics shown in this figure, but is affected by observed increases in mean temperature, decreases in frost and increases in marine heatwaves. (Table TS.5) Links to chapters11.9, Atlas, 1.3.3, Figure Atlas.2

Climate services involve providing climate information to assist decision-making, for example, about how extreme rainfall will change to inform improvements in urban drainage. Since AR5, there has been a significant increase in the range and diversity of climate service activities (very high confidence). The level of user-engagement, co-design and co-production are factors determining the utility of climate services, while resource limitations for these activities constrain their full potential. Links to chapters12.6, Cross-Chapter Box 12.2

Climate services include engagement from users and providers and an effective access mechanism; they are responsive to user needs and based on integrating scientifically credible information and relevant expertise. Climate services are being developed across regions, sectors, time scales and user-groups and include a range of knowledge brokerage and integration activities. These involve identifying knowledge needs; compiling, translating and disseminating knowledge; coordinating networks and building capacity through informed decision-making; analysis, evaluation and development of policy; and personal consultation.

Since AR5, climate change information produced in climate service contexts has increased significantly due to scientific and technological advancements and growing user awareness, requirements and demand (very high confidence). Climate services are growing rapidly and are highly diverse in their practices and products. The decision-making context, level of user engagement and co-production between scientists, practitioners and intended users are important determinants of the type of climate service developed and their utility for supporting adaptation, mitigation and risk management decisions. They require different types of user–producer engagement depending on what the service aims to deliver (high confidence), and these fall into three broad categories: website-based services, interactive group activities and focused relationships .

Realization of the full potential of climate services is often hindered by limited resources for the co-design and co-production process, including sustained engagement between scientists, service providers and users (high confidence). Further challenges relate to the development and provision of climate services, generation of climate service products, communication with users, and evaluation of their quality and socio-economic benefit. (Section TS.4.1) Links to chapters1.2.3, 10.5.4, 12.6, Cross-Chapter Box 12.2, Glossary

A key novel element in the AR6 is the Working Group I Atlas, which includes the Interactive Atlas (https://interactive-atlas.ipcc.ch/). The Interactive Atlas provides the ability to explore much of the observational and climate model data used as lines of evidence in this assessment to generate regional climate information. Links to chaptersAtlas.2

A significant innovation in the AR6 WGI Report is the Atlas. Part of its remit is to provide region-by-region assessment on changes in mean climate and to link with other WGI chapters to generate climate change information for the regions. An important component is the new online interactive tool, the Interactive Atlas, with flexible spatial and temporal analyses of much of the observed, simulated past and projected future climate change data underpinning the WGI assessment. This includes the ability to generate global maps and a number of regionally aggregated products (time series, scatter plots, tables, etc.) for a range of observations and ensemble climate change projections of variables (such as changes in the climatic impact-drivers summarized in Table TS.5) from the Coupled Model Intercomparison Project Phases 5 and 6 (CMIP5, CMIP6) and the Coordinated Regional Climate Downscaling Experiment (CORDEX). The data can be displayed and summarized under a range of SSP-RCP scenarios and future time slices and also for different global warming levels, relative to several different baseline periods. The maps and various statistics can be generated for annual mean trends and changes or for any user-specified season. A new set of WGI reference regions is used for the regional summary statistics and applied widely throughout the report (with the regions, along with aggregated datasets and the code to generate these, available at the ATLAS GitHub: https://github.com/IPCC-WG1/Atlas).

Box TS.12, Figure 1 shows how the Interactive Atlas products, together with other lines of evidence, can be used to generate climate information for an illustrative example of the Mediterranean summer warming. The lines of evidence include the understanding of relevant mechanisms, dynamic and thermodynamic processes and the effect of aerosols in this case (Box TS.12, Figure 1a); trends in observational datasets (which can have different spatial and temporal coverage; Box TS.12, Figure 1b, c); and attribution of these trends and temperature projections from global and regional climate models at different resolutions, including single-model initial-condition large ensembles (SMILEs; Box TS.12, Figure 1d, e). Taken together, this evidence shows there is high confidence that the projected Mediterranean summer temperature increase will be larger than the global mean, with consistent results from CMIP5 and CMIP6 (Box TS.12, Figure 1e). However, CMIP6 results project both more pronounced warming than CMIP5 for a given emissions scenario and time period and a greater range of changes (Box TS.12, Figure 1d). Links to chapters10.6.4, Atlas.2, Atlas.8.4

Box TS.12, Figure 1 | Example of generating regional climate information from multiple lines of evidence for the case of Mediterranean summer warming. Box TS.12 The intent of this figure is to provide an example of using different lines of evidence to assess the confidence in or likelihood of a projected change in regional climate and which of these lines of evidence are available to view and explore in the Interactive Atlas. (a) Mechanisms and feedbacks involved in enhanced Mediterranean summer warming. (b) Locations of observing stations from different datasets. (c) Distribution of 1960–2014 summer temperature trends (°C per decade) for observations (black crosses), CMIP5 (blue circles), CMIP6 (red circles), HighResMIP (orange circles), CORDEX EUR-44 (light blue circles), CORDEX EUR-11 (green circles), and selected single model initial-condition large ensembles (SMILEs; grey boxplots, MIROC6, CSIRO-Mk3-6-0, MPI-ESM and d4PDF). (d) Time series of area averaged (25°N–50°N, 10°W–40°E) land point summer temperature anomalies (°C, baseline period is 1995–2014): the boxplot shows long term (2081–2100) temperature changes of different CMIP6 scenarios in respect to the baseline period. (e) Projected Mediterranean summer warming in comparison to global annual mean warming of CMIP5 (RCP2.6, RCP4.5, RCP6.0 and RCP8.5) and CMIP6 (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5) ensemble means (lines) and spread (shading). Links to chaptersFigure 10.20, Figure 10.21, Figure Atlas.8

Global land monsoon precipitation decreased from the 1950s to the 1980s, partly due to anthropogenic aerosols, but has increased since then in response to GHG forcing and large-scale multi-decadal variability (medium confidence). Northern Hemispheric anthropogenic aerosols weakened the regional monsoon circulations in South Asia, East Asia and West Africa during the second half of the 20th century, thereby offsetting the expected strengthening of monsoon precipitation in response to GHG-induced warming (high confidence).

During the 21st century, global land monsoon precipitation is projected to increase in response to GHG warming in all time horizons and scenarios (high confidence). Over South and South East Asia, East Asia and the central Sahel, monsoon precipitation is projected to increase, whereas over North America and the far western Sahel it is projected to decrease (medium confidence). There is low confidence in projected precipitation changes in the South American and Australian-Maritime Continent monsoons. At global and regional scales, near-term monsoon changes will be dominated by the effects of internal variability (medium confidence). Links to chapters2.3, 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

Global Monsoon

Paleoclimate records indicate that during warm climates, like the mid-Pliocene Warm Period, monsoon systems were stronger (medium confidence). In the instrumental records, global summer monsoon precipitation intensity has likely increased since the 1980s, dominated by Northern Hemisphere summer trends and large multi-decadal variability. Contrary to the expected increase of precipitation under global warming, the Northern Hemisphere monsoon regions experienced declining precipitation from the 1950s to 1980s, which is partly attributable to the influence of anthropogenic aerosols (medium confidence) (Box TS.13, Figure 1). Links to chapters2.3.1, Cross-Chapter Box 2.4, 3.3.2, 3.3.3

Box TS.13, Figure 1 | Global and regional monsoons: past trends and projected changes. The intent of this figure is to show changes in precipitation over regional monsoon domains in terms of observed past trends, how greenhouse gases and aerosols relate to these changes, and in terms of future projections in one intermediate emissions scenario in the near, medium and long term. (a) Global (black contour) and regional monsoons (colour shaded) domains. The global monsoon (GM) is defined as the area with local summer-minus-winter precipitation rate exceeding 2.5 mm day^–1 (see Annex V). The regional monsoon domains are defined based on published literature and expert judgement (see Annex V) and accounting for the fact that the climatological summer monsoon rainy season varies across the individual regions. Assessed regional monsoons are South and South East Asia (SAsiaM, Jun–July–August–September), East Asia (EAsiaM, June–July–August), West Africa (WAfriM, June–July–August–September), North America (NAmerM, July–August–-September), South America (SAmerM, December–January–February), Australia and Maritime Continent Monsoon (AusMCM, December–January–February). Equatorial South America (EqSAmer) and South Africa (SAfri)regions are also shown, as they receive unimodal summer seasonal rainfall although their qualification as monsoons is subject to discussion. (b) Global and regional monsoons precipitation trends based on DAMIP CMIP6 simulations with both natural and anthropogenic (ALL), greenhouse gas only (GHG), aerosols only (AER) and natural only (NAT) radiative forcing. Weighted ensemble means are based on nine Coupled model Intercomparison Project Phase 6 (CMIP6) models contributing to the MIP (with at least three members). Observed trends computed from CRU, GPCP and APHRO (only forSAsiaM and EAsiaM) datasets are shown as well. (c) Percentage change in projected seasonal mean precipitation over global and regional monsoons domain in the near term (2021–2040), mid-term (2041–2060), and long term (2081–2100) under SSP2-4.5 based on 24 CMIP6 models. Links to chaptersFigures 8.11 and 8.22

Open figure

With continued global warming, it is likely that global land monsoon precipitation will increase during this century (Box TS.13, Figure 1), particularly in the Northern Hemisphere, although the monsoon circulation is projected to weaken. A slowdown of the tropical circulation with global warming can partly offset the warming-induced strengthening of precipitation in monsoon regions (high confidence). In the near term, global monsoon changes are likely to be dominated by the effects of internal variability and model uncertainties (medium confidence). In the long term, global monsoon rainfall change will feature a robust north–south asymmetry characterized by a greater increase in the Northern Hemisphere than in the Southern Hemisphere and an east–west asymmetry characterized by enhanced Asian–African monsoons and a weakened North American monsoon (medium confidence). Links to chapters4.4.1, 4.5.1, 8.4.1

Regional Monsoons

Paleoclimate reconstructions indicate stronger monsoons in the Northern Hemisphere but weaker ones in the Southern Hemisphere during warm periods, particularly for the South and South East Asian, East Asian, and North and South American monsoons, with the opposite occurring during cold periods (medium confidence). It is very likely that Northern Hemispheric anthropogenic aerosols weakened the regional monsoon circulations in South Asia, East Asia and West Africa during the second half of the 20th century, thereby offsetting the expected strengthening of monsoon precipitation in response to GHG-induced warming (Box TS.13, Figure 1). Multiple lines of evidence explain this contrast over South Asia, with the observed trends dominated by the effects of aerosols, while future projections are mostly driven by GHG increases. The recent partial recovery and enhanced intensity of monsoon precipitation over West Africa is related to the growing influence of GHGs with an additional contribution due to the reduced cooling effect of anthropogenic aerosols, emitted largely from North America and Europe (medium confidence). For other regional monsoons, that is, North and South America and Australia, there is low confidence in the attribution of recent changes in precipitation (Box TS.13, Figure 1) and winds. Links to chapters2.3.1, 8.3.1, 8.3.2, Box 8.1, 10.6.3

Projections of regional monsoons during the 21st century indicate contrasting (region-dependent) and uncertain precipitation and circulation changes. The annual contrast between the wettest and driest month of the year is likely to increase by 3–5% per degree Celsius in most monsoon regions in terms of precipitation, precipitation minus evaporation, and runoff (medium confidence). For the North American monsoon, projections indicate a decrease in precipitation, whereas increased monsoon rainfall is projected over South and South East Asia and over East Asia (medium confidence) (Box TS.13, Figure 1). West African monsoon precipitation is projected to increase over the central Sahel and decrease over the far western Sahel (medium confidence). There is low confidence in projected precipitation changes in the South American and Australian-Maritime Continent regional monsoons (for both magnitude and sign) (Box TS.13, Figure 1). There is medium confidence that the monsoon season will be delayed in the Sahel and high confidence that it will be delayed in North and South America. Links to chapters8.2.2, 8.4.2.4, Box 8.2

Building the Assessment from Multiple Lines of Evidence

Large natural variability of monsoon precipitation across different time scales, found in both paleoclimate reconstructions and instrumental measurements, poses an inherent challenge for robust quantification of future changes in precipitation at regional and smaller spatial scales. At both global and regional scales, there is medium confidence that internal variability contributes the largest uncertainty related to projected changes, at least in the near term (2021–2040). A collapse of the Atlantic Meridional Overturning Circulation could weaken the African and Asian monsoons but strengthen the Southern Hemisphere monsoons (high confidence). Links to chapters4.4.4, 4.5.1, Cross-Chapter Box 4.1, 8.5.2, 8.6.1, 9.2.3, 10.6.3

Overall, long-term (2081–2100) future changes in regional monsoons like the South and South East Asian monsoon are generally consistent across global (including high-resolution) and regional climate models and are supported by theoretical arguments. Uncertainties in simulating the observed characteristics of regional monsoon precipitation are related to varying complexities of regional monsoon processes and their responses to external forcing, internal variability, and deficiencies in representing monsoon warm rain processes, organized tropical convection, heavy orographic rainfall and cloud–aerosol interactions. Links to chapters8.3.2, 8.5.1, 10.3.3, 10.6.3

With global warming, urban areas and cities will be affected by more frequent occurrences of extreme climate events, such as heatwaves, with more hot days and warm nights as well as sea level rise and increases in tropical cyclone storm surge and rainfall intensity that will increase the probability of coastal city flooding (high confidence). Links to chaptersBox 10.3, 11.3, 11.5, 12.3, 12.4

Urban areas have special interactions with the climate system, for instance in terms of heat islands and altering the water cycle, and thereby will be more affected by extreme climate events such as extreme heat (high confidence). With global warming, increasing relative sea level compounded by increasing tropical cyclone storm surge and rainfall intensity will increase the probability of coastal city flooding (high confidence). Arctic coastal settlements are particularly exposed to climate change due to sea ice retreat (high confidence). Improvements in urban climate modelling and climate monitoring networks have contributed to understanding the mutual interaction between regional and urban climate (high confidence). Links to chaptersBox 10.3, 11.3, 11.5, 12.3, 12.4

Despite having a negligible effect on global surface temperature (high confidence), urbanization has exacerbated the effects of global warming through its contribution to the observed warming trend in and near cities, particularly in annual mean minimum temperature (very high confidence) and increases in mean and extreme precipitation over and downwind of the city, especially in the afternoon and early evening (medium confidence). Links to chapters2.3, Box 10.3, 11.3, 11.4, 12.3, 12.4

Combining climate change projections with urban growth scenarios, future urbanization will amplify (very high confidence) the projected local air temperature increase, particularly by strong influence on minimum temperatures, which is approximately comparable in magnitude to global warming (high confidence). Compared to present day, large implications are expected from the combination of future urban development and more frequent occurrence of extreme climate events, such as heatwaves, with more hot days and warm nights adding to heat stress in cities (very high confidence). Links to chaptersBox 10.2, 11.3, 12.4

Both sea levels and air temperatures are projected to rise in most coastal settlements (high confidence). There is high confidence in an increase in pluvial flood potential in urban areas where extreme precipitation is projected to increase, especially at high global warming levels. Links to chapters11.4, 11.5, 12.4