2006) carbon cycle models, assuming a PDF for climate sensitivity that corresponds to the assessment of IPCC AR4 (Meehl et al., 2007b, Box.
120 KB – 108 Pages
PAGE – 1 ============
102912This chapter should be cited as:Collins, M., R. Knutti, J. Arblaster, J.-L. Dufresne, T. Fichefet, P. Friedlingstein, X. Gao, W.J. Gutowski, T. Johns, G. Krinner, M. Shongwe, C. Tebaldi, A.J. Weaver and M. Wehner, 2013: Long-term Climate Change: Projections, Com -mitments and Irreversibility. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Coordinating Lead Authors: Matthew Collins (UK), Reto Knutti (Switzerland) Lead Authors: Julie Arblaster (Australia), Jean-Louis Dufresne (France), Thierry Fichefet (Belgium), Pierre Friedlingstein (UK/Belgium), Xuejie Gao (China), William J. Gutowski Jr. (USA), Tim Johns (UK), Gerhard Krinner (France/Germany), Mxolisi Shongwe (South Africa), Claudia Tebaldi (USA), Andrew J. Weaver (Canada), Michael Wehner (USA) Contributing Authors: Myles R. Allen (UK), Tim Andrews (UK), Urs Beyerle (Switzerland), Cecilia M. Bitz (USA), Sandrine Bony (France), Ben B.B. Booth (UK), Harold E. Brooks (USA), Victor Brovkin (Germany), Oliver Browne (UK), Claire Brutel-Vuilmet (France), Mark Cane (USA), Robin Chadwick (UK), Ed Cook (USA), Kerry H. Cook (USA), Michael Eby (Canada), John Fasullo (USA), Erich M. Fischer (Switzerland), Chris E. Forest (USA), Piers Forster (UK), Peter Good (UK), Hugues Goosse (Belgium), Jonathan M. Gregory (UK), Gabriele C. Hegerl (UK/Germany), Paul J. Hezel (Belgium/ USA), Kevin I. Hodges (UK), Marika M. Holland (USA), Markus Huber (Switzerland), Philippe Huybrechts (Belgium), Manoj Joshi (UK), Viatcheslav Kharin (Canada), Yochanan Kushnir (USA), David M. Lawrence (USA), Robert W. Lee (UK), Spencer Liddicoat (UK), Christopher Lucas (Australia), Wolfgang Lucht (Germany), Jochem Marotzke (Germany), François Massonnet (Belgium), H. Damon Matthews (Canada), Malte Meinshausen (Germany), Colin Morice (UK), Alexander Otto (UK/Germany), Christina M. Patricola (USA), Gwenaëlle Philippon- Berthier (France), Prabhat (USA), Stefan Rahmstorf (Germany), William J. Riley (USA), Joeri Rogelj (Switzerland/Belgium), Oleg Saenko (Canada), Richard Seager (USA), Jan Sedlá˜ek (Switzerland), Len C. Shaffrey (UK), Drew Shindell (USA), Jana Sillmann (Canada), Andrew Slater (USA/Australia), Bjorn Stevens (Germany/USA), Peter A. Stott (UK), Robert Webb (USA), Giuseppe Zappa (UK/Italy), Kirsten Zickfeld (Canada/Germany) Review Editors: Sylvie Joussaume (France), Abdalah Mokssit (Morocco), Karl Taylor (USA), Simon Tett (UK) Long-term Climate Change:Projections, Commitments and Irreversibility
PAGE – 2 ============
1030Table of Contents Executive Summary .103112.1 Introduction ..103412.2 Climate Model Ensembles and Sources of Uncertainty from Emissions to Projections ..1035 12.2.1 The Coupled Model Intercomparison Project Phase 5 and Other Tools 1035 12.2.2 General Concepts: Sources of Uncertainties 1035 12.2.3 From Ensembles to Uncertainty Quanti˚cation .1040 Box 12.1: Methods to Quantify Model Agreement in Maps ..1041 12.2.4 Joint Projections of Multiple Variables ..104412.3 Projected Changes in Forcing Agents, Including Emissions and Concentrations .1044 12.3.1 Description of Scenarios 1045 12.3.2 Implementation of Forcings in Coupled Model Intercomparison Project Phase 5 Experiments .1047 12.3.3 Synthesis of Projected Global Mean Radiative Forcing for the 21st Century 105212.4 Projected Climate Change over the 21st Century .1054 12.4.1 Time-Evolving Global Quantities ..1054 12.4.2 Pattern Scaling ..1058 12.4.3 Changes in Temperature and Energy Budget ..1062 12.4.4 Changes in Atmospheric Circulation ..1071 12.4.5 Changes in the Water Cycle 1074 12.4.6 Changes in Cryosphere .1087 12.4.7 Changes in the Ocean 1093 12.4.8 Changes Associated with Carbon Cycle Feedbacks and Vegetation Cover .1096 12.4.9 Consistency and Main Differences Between Coupled Model Intercomparison Project Phase 3/Coupled Model Intercomparison Project Phase 5 and Special Report on Emission Scenarios/Representative Concentration Pathways .1099 12.5 Climate Change Beyond 2100, Commitment, Stabilization and Irreversibility ..1102 12.5.1 Representative Concentration Pathway Extensions 1102 12.5.2 Climate Change Commitment 1102 12.5.3 Forcing and Response, Time Scales of Feedbacks .1105 12.5.4 Climate Stabilization and Long-term Climate Targets .1107 Box 12.2: Equilibrium Climate Sensitivity and Transient Climate Response 1110 12.5.5 Potentially Abrupt or Irreversible Changes ..1114References ..1120Frequently Asked Questions FAQ 12.1 Why Are So Many Models and Scenarios Used to Project Climate Change ? ..1036 FAQ 12.2 How Will the Earth™s Water Cycle Change? .1084 FAQ 12.3 What Would Happen to Future Climate if We Stopped Emissions Today? .1106
PAGE – 3 ============
10311 In this Report, the following terms have been used to indicate the assessed likelihood of an outcome or a result: Virtually certain 99Œ100% probability, Very likely 90Œ100%, Likely 66Œ100%, About as likely as not 33Œ66%, Unlikely 0Œ33%, Very unlikely 0Œ10%, Exceptionally unlikely 0Œ1%. Additional terms (Extremely likely: 95Œ100%, More likely than not >50Œ100%, and Extremely unlikely 0Œ5%) may also be used when appropriate. Assessed likelihood is typeset in italics, e.g., very likely (see Section 1.4 and Box TS.1 for more details).2 In this Report, the following summary terms are used to describe the available evidence: limited, medium, or robust; and for the degree of agreement: low, medium, or high. A level of con˜dence is expressed using ˜ve quali˜ers: very low, low, medium, high, and very high, and typeset in italics, e.g., medium con˜dence. For a given evidence and agreement statement, different con˜dence levels can be assigned, but increasing levels of evidence and degrees of agreement are correlated with increasing con˜dence (see Section 1.4 and Box TS.1 for more details). Executive SummaryThis chapter assesses long-term projections of climate change for the end of the 21st century and beyond, where the forced signal depends on the scenario and is typically larger than the internal variability of the climate system. Changes are expressed with respect to a baseline period of 1986Œ2005, unless otherwise stated. Scenarios, Ensembles and Uncertainties The Coupled Model Intercomparison Project Phase 5 (CMIP5) presents an unprecedented level of information on which to base projections including new Earth System Models with a more complete representation of forcings, new Representative Concentration Pathways (RCP) scenarios and more output avail -able for analysis. The four RCP scenarios used in CMIP5 lead to a total radiative forcing (RF) at 2100 that spans a wider range than that estimated for the three Special Report on Emission Scenarios (SRES) scenarios (B1, A1B, A2) used in the Fourth Assessment Report (AR4), RCP2.6 being almost 2 W m Œ2 lower than SRES B1 by 2100. The mag -nitude of future aerosol forcing decreases more rapidly in RCP sce -narios, reaching lower values than in SRES scenarios through the 21st century. Carbon dioxide (CO 2) represents about 80 to 90% of the total anthropogenic forcing in all RCP scenarios through the 21st century. The ensemble mean total effective RFs at 2100 for CMIP5 concen -tration-driven projections are 2.2, 3.8, 4.8 and 7.6 W m Œ2 for RCP2.6, RCP4.5, RCP6.0 and RCP8.5 respectively, relative to about 1850, and are close to corresponding Integrated Assessment Model (IAM)-based estimates (2.4, 4.0, 5.2 and 8.0 W m Œ2). {12.2.1, 12.3, Table 12.1, Fig -ures 12.1, 12.2, 12.3, 12.4} New experiments and studies have continued to work towards a more complete and rigorous characterization of the uncertain -ties in long-term projections, but the magnitude of the uncer -tainties has not changed signi˜cantly since AR4. There is overall consistency between the projections based on CMIP3 and CMIP5, for both large-scale patterns and magnitudes of change. Differences in global temperature projections are largely attributable to a change in scenarios. Model agreement and con˚dence in projections depend on the variable and spatial and temporal averaging. The well-established stability of large-scale geographical patterns of change during a tran-sient experiment remains valid in the CMIP5 models, thus justifying pattern scaling to approximate changes across time and scenarios under such experiments. Limitations remain when pattern scaling is applied to strong mitigation scenarios, to scenarios where localized forcing (e.g., aerosols) are signi˚cant and vary in time and for varia -bles other than average temperature and precipitation. {12.2.2, 12.2.3, 12.4.2, 12.4.9, Figures 12.10, 12.39, 12.40, 12.41} Projections of Temperature Change Global mean temperatures will continue to rise over the 21st century if greenhouse gas (GHG) emissions continue unabat -ed. Under the assumptions of the concentration-driven RCPs, global mean surface temperatures for 2081Œ2100, relative to 1986Œ2005 will likely 1 be in the 5 to 95% range of the CMIP5 models; 0.3°C to 1.7°C (RCP2.6), 1.1°C to 2.6°C (RCP4.5), 1.4°C to 3.1°C (RCP6.0), 2.6°C to 4.8°C (RCP8.5). Global temperatures averaged over the period 2081Œ 2100 are projected to likely exceed 1.5°C above 1850-1900 for RCP4.5, RCP6.0 and RCP8.5 (high con˜dence), are likely to exceed 2°C above 1850-1900 for RCP6.0 and RCP8.5 (high con˜dence) and are more likely than not to exceed 2°C for RCP4.5 (medium con˜dence). Temper -ature change above 2°C under RCP2.6 is unlikely (medium con˜dence). Warming above 4°C by 2081Œ2100 is unlikely in all RCPs (high con˜-dence) except for RCP8.5, where it is about as likely as not (medium con˜dence). {12.4.1, Tables 12.2, 12.3, Figures 12.5, 12.8} Temperature change will not be regionally uniform. There is very high con˜dence2 that globally averaged changes over land will exceed changes over the ocean at the end of the 21st century by a factor that is likely in the range 1.4 to 1.7. In the absence of a strong reduction in the Atlantic Meridional Overturning, the Arctic region is project -ed to warm most ( very high con˜dence). This polar ampli˚cation is not found in Antarctic regions due to deep ocean mixing, ocean heat uptake and the persistence of the Antarctic ice sheet. Projected region-al surface air temperature increase has minima in the North Atlantic and Southern Oceans in all scenarios. One model exhibits marked cool -ing in 2081Œ2100 over large parts of the Northern Hemisphere (NH), and a few models indicate slight cooling locally in the North Atlantic. Atmospheric zonal mean temperatures show warming throughout the troposphere, especially in the upper troposphere and northern high latitudes, and cooling in the stratosphere. {12.4.2, 12.4.3, Table 12.2, Figures 12.9, 12.10, 12.11, 12.12} It is virtually certain that, in most places, there will be more hot and fewer cold temperature extremes as global mean temper -atures increase. These changes are expected for events de˚ned as extremes on both daily and seasonal time scales. Increases in the fre -quency, duration and magnitude of hot extremes along with heat stress are expected; however, occasional cold winter extremes will continue to occur. Twenty-year return values of low temperature events are project -ed to increase at a rate greater than winter mean temperatures in most regions, with the largest changes in the return values of low tempera -tures at high latitudes. Twenty-year return values for high temperature events are projected to increase at a rate similar to or greater than the rate of increase of summer mean temperatures in most regions. Under RCP8.5 it is likely that, in most land regions, a current 20-year high temperature event will occur more frequently by the end of the 21st
PAGE – 4 ============
1032century (at least doubling its frequency, but in many regions becoming an annual or 2-year event) and a current 20-year low temperature event will become exceedingly rare. {12.4.3, Figures 12.13, 12.14} Changes in Atmospheric Circulation Mean sea level pressure is projected to decrease in high lati -tudes and increase in the mid-latitudes as global temperatures rise. In the tropics, the Hadley and Walker Circulations are likely to slow down. Poleward shifts in the mid-latitude jets of about 1 to 2 degrees latitude are likely at the end of the 21st century under RCP8.5 in both hemispheres (medium con˜dence), with weaker shifts in the NH. In austral summer, the additional in˛uence of stratospheric ozone recovery in the Southern Hemisphere opposes changes due to GHGs there, though the net response varies strongly across models and scenarios. Substantial uncertainty and thus low con˜dence remains in projecting changes in NH storm tracks, especially for the North Atlantic basin. The Hadley Cell is likely to widen, which translates to broad -er tropical regions and a poleward encroachment of subtropical dry zones. In the stratosphere, the BrewerŒDobson circulation is likely to strengthen. {12.4.4, Figures 12.18, 12.19, 12.20} Changes in the Water Cycle It is virtually certain that, in the long term, global precipitation will increase with increased global mean surface temperature. Global mean precipitation will increase at a rate per degree Celsius smaller than that of atmospheric water vapour. It will likely increase by 1 to 3% °CŒ1 for scenarios other than RCP2.6. For RCP2.6 the range of sensitivities in the CMIP5 models is 0.5 to 4% °CŒ1 at the end of the 21st century. {12.4.1, Figures 12.6, 12.7} Changes in average precipitation in a warmer world will exhibit substantial spatial variation. Some regions will experience increases, other regions will experience decreases and yet others will not experience signi˜cant changes at all. There is high con˜dence that the contrast of annual mean precipitation between dry and wet regions and that the contrast between wet and dry seasons will increase over most of the globe as temperatures increase. The general pattern of change indicates that high latitude land masses are likely to experience greater amounts of precipitation due to the increased speci˚c humidity of the warmer troposphere as well as increased transport of water vapour from the tropics by the end of this century under the RCP8.5 scenario. Many mid-latitude and subtropical arid and semi-arid regions will likely experience less precipitation and many moist mid-latitude regions will likely experience more precipitation by the end of this century under the RCP8.5 scenario. Globally, for short-duration precipitation events, a shift to more intense individual storms and fewer weak storms is likely as temperatures increase. Over most of the mid-latitude land-masses and over wet tropical regions, extreme precipitation events will very likely be more intense and more frequent in a warmer world. The global average sensitivity of the 20-year return value of the annual maximum daily precipitation increases ranges from 4% °CŒ1 of local temperature increase (average of CMIP3 models) to 5.3% oCŒ1˝of local tempera -ture increase (average of CMIP5 models) but regionally there are wide variations. {12.4.5, Figures 12.10, 12.22, 12.26, 12.27} Annual surface evaporation is projected to increase as global temperatures rise over most of the ocean and is projected to change over land following a similar pattern as precipitation. Decreases in annual runoff are likely in parts of southern Europe, the Middle East, and southern Africa by the end of the 21st century under the RCP8.5 scenario. Increases in annual runoff are likely in the high northern latitudes corresponding to large increases in winter and spring precipitation by the end of the 21st century under the RCP8.5 scenario. Regional to global-scale projected decreases in soil moisture and increased risk of agricultural drought are likely in presently dry regions and are projected with medium con˜dence by the end of the 21st century under the RCP8.5 scenario. Prominent areas of projected decreases in evaporation include southern Africa and north western Africa along the Mediterranean. Soil moisture drying in the Mediterra -nean, southwest USA and southern African regions is consistent with projected changes in Hadley Circulation and increased surface tem-peratures, so surface drying in these regions as global temperatures increase is likely with high con˜dence by the end of this century under the RCP8.5 scenario. In regions where surface moistening is projected, changes are generally smaller than natural variability on the 20-year time scale. {12.4.5, Figures 12.23, 12.24, 12.25} Changes in Cryosphere It is very likely that the Arctic sea ice cover will continue shrink -ing and thinning year-round in the course of the 21st century as global mean surface temperature rises. At the same time, in the Antarctic, a decrease in sea ice extent and volume is expected, but with low con˜dence . Based on the CMIP5 multi-model ensem-ble, projections of average reductions in Arctic sea ice extent for 2081Œ 2100 compared to 1986Œ2005 range from 8% for RCP2.6 to 34% for RCP8.5 in February and from 43% for RCP2.6 to 94% for RCP8.5 in September (medium con˜dence). A nearly ice-free Arctic Ocean (sea ice extent less than 1 × 106 km2for at least 5 consecutive years) in Septem-ber before mid-century is likely under RCP8.5 (medium con˜dence ), based on an assessment of a subset of models that most closely repro-duce the climatological mean state and 1979Œ2012 trend of the Arctic sea ice cover. Some climate projections exhibit 5- to 10-year periods of sharp summer Arctic sea ice declineŠeven steeper than observed over the last decadeŠand it is likely that such instances of rapid ice loss will occur in the future. There is little evidence in global climate models of a tipping point (or critical threshold) in the transition from a peren-nially ice-covered to a seasonally ice-free Arctic Ocean beyond which further sea ice loss is unstoppable and irreversible. In the Antarctic, the CMIP5 multi-model mean projects a decrease in sea ice extent that ranges from 16% for RCP2.6 to 67% for RCP8.5 in February and from 8% for RCP2.6 to 30% for RCP8.5 in September for 2081Œ2100 com-pared to 1986Œ2005. There is, however, low con˜dence in those values as projections because of the wide inter-model spread and the inability of almost all of the available models to reproduce the mean annual cycle, interannual variability and overall increase of the Antarctic sea ice areal coverage observed during the satellite era. {12.4.6, 12.5.5, Figures 12.28, 12.29, 12.30, 12.31} It is very likely that NH snow cover will reduce as global tem -peratures rise over the coming century. A retreat of permafrost extent with rising global temperatures is virtually certain. Snow
PAGE – 5 ============
1033cover changes result from precipitation and ablation changes, which are sometimes opposite. Projections of the NH spring snow covered area by the end of the 21st century vary between a decrease of 7% (RCP2.6) and a decrease of 25% (RCP8.5), with a pattern that is fairly consistent between models. The projected changes in permafrost are a response not only to warming but also to changes in snow cover, which exerts a control on the underlying soil. By the end of the 21st cen -tury, diagnosed near-surface permafrost area is projected to decrease by between 37% (RCP2.6) and 81% (RCP8.5) (medium con˜dence). {12.4.6, Figures 12.32, 12.33} Changes in the OceanThe global ocean will warm in all RCP scenarios. The strongest ocean warming is projected for the surface in subtropical and tropi -cal regions. At greater˝depth the warming is projected to˝be most pronounced in the Southern Ocean. Best estimates of˝ocean warm -ing in the top one˝hundred meters are about 0.6°C (RCP2.6) to 2.0°C (RCP8.5), and about 0.3°C (RCP2.6) to 0.6°C˝(RCP8.5) at a depth of about 1 km by the end of the 21st century. For RCP4.5 by the end of the 21st century, half of the energy taken up by the ocean is in the upper -most 700 m and 85% is in the uppermost 2000 m. Due to the long time scales of this heat transfer from the surface to depth, ocean warming will continue for centuries, even if GHG emissions are decreased or concentrations kept constant. {12.4.7, 12.5.2Œ12.5.4, Figure 12.12} It is very likely that the Atlantic Meridional Overturning Circu -lation (AMOC) will weaken over the 21st century but it is very unlikely that the AMOC will undergo an abrupt transition or col -lapse in the 21st century. Best estimates and ranges for the reduc-tion from CMIP5 are 11% (1 to 24%) in RCP2.6 and 34% (12 to 54%) in RCP8.5. There is low con˜dence in assessing the evolution of the AMOC beyond the 21st century. {12.4.7, Figure 12.35} Carbon CycleWhen forced with RCP8.5 CO 2 emissions, as opposed to the RCP8.5 CO 2 concentrations, 11 CMIP5 Earth System Models with interactive carbon cycle simulate, on average, a 50 ppm (min to max range Œ14 0 to +210 ppm) larger atmospheric CO 2 concen-tration and 0.2°C (min to max range Œ0.4 to +0.9°C) larger global surface temperature increase by 2100. {12.4.8, Figures 12.36, 12.37} Long-term Climate Change, Commitment and Irreversibility Global temperature equilibrium would be reached only after centuries to millennia if RF were stabilized. Continuing GHG emis-sions beyond 2100, as in the RCP8.5 extension, induces a total RF above 12 W m Œ2 by 2300. Sustained negative emissions beyond 2100, as in RCP2.6, induce a total RF below 2 W m Œ2 by 2300. The projected warm -ing for 2281Œ2300, relative to 1986Œ2005, is 0.0°C to 1.2°C for RCP2.6 and 3.0°C to 12.6°C for RCP8.5 (medium con˜dence). In much the same way as the warming to a rapid increase of forcing is delayed, the cooling after a decrease of RF is also delayed. {12.5.1, Figures 12.43, 12.44} A large fraction of climate change is largely irreversible on human time scales, unless net anthropogenic CO 2 emissions were strongly negative over a sustained period. For scenarios driven by CO2 alone, global average temperature is projected to remain approximately constant for many centuries following a com-plete cessation of emissions. The positive commitment from CO 2 may be enhanced by the effect of an abrupt cessation of aerosol emissions, which will cause warming. By contrast, cessation of emission of short- lived GHGs will contribute a cooling. {12.5.3, 12.5.4, Figures 12.44, 12.45, 12.46, FAQ 12.3} Equilibrium Climate Sensitivity and Transient Climate ResponseEstimates of the equilibrium climate sensitivity (ECS) based on observed climate change, climate models and feedback analy -sis, as well as paleoclimate evidence indicate that ECS is likely in the range 1.5°C to 4.5°C with high con˜dence, extreme -ly unlikely less than 1°C (high con˜dence) and very unlikely greater than 6°C ( medium con˜dence). The transient climate response (TCR) is likely in the range 1°C to 2.5ºC and extremely unlikely greater than 3°C, based on observed climate change and climate models. {Box 12.2, Figures 1, 2} Climate StabilizationThe principal driver of long-term warming is total emissions of CO2 and the two quantities are approximately linearly related. The global mean warming per 1000 PgC (transient cli -mate response to cumulative carbon emissions (TCRE)) is likely between 0.8°C to 2.5°C per 1000 PgC, for cumulative emissions less than about 2000 PgC until the time at which temperatures peak. To limit the warming caused by anthropogenic CO 2 emissions alone to be likely less than 2°C relative to the period 1861-1880, total CO2 emissions from all anthropogenic sources would need to be limit-ed to a cumulative budget of about 1000 PgC since that period. About half [445 to 585 PgC] of this budget was already emitted by 2011. Accounting for projected warming effect of non-CO 2 forcing, a possible release of GHGs from permafrost or methane hydrates, or requiring a higher likelihood of temperatures remaining below 2°C, all imply a lower budget. {12.5.4, Figures 12.45, 12.46, Box 12.2} Some aspects of climate will continue to change even if temper-atures are stabilized. Processes related to vegetation change, chang -es in the ice sheets, deep ocean warming and associated sea level rise and potential feedbacks linking for example ocean and the ice sheets have their own intrinsic long time scales and may result in signi˚cant changes hundreds to thousands of years after global temperature is stabilized. {12.5.2 to 12.5.4} Abrupt Change Several components or phenomena in the climate system could potentially exhibit abrupt or nonlinear changes, and some are known to have done so in the past. Examples include the AMOC, Arctic sea ice, the Greenland ice sheet, the Amazon forest and mon -soonal circulations. For some events, there is information on potential consequences, but in general there is low con˜dence and little con-sensus on the likelihood of such events over the 21st century. {12.5.5, Table 12.4}
PAGE – 6 ============
103412.1 Introduction Projections of future climate change are not like weather forecasts. It is not possible to make deterministic, de˚nitive predictions of how climate will evolve over the next century and beyond as it is with short-term weather forecasts. It is not even possible to make projections of the frequency of occurrence of all possible outcomes in the way that it might be possible with a calibrated probabilistic medium-range weath-er forecast. Projections of climate change are uncertain, ˚rst because they are dependent primarily on scenarios of future anthropogenic and natural forcings that are uncertain, second because of incomplete understanding and imprecise models of the climate system and ˚nally because of the existence of internal climate variability. The term cli -mate projection tacitly implies these uncertainties and dependencies. Nevertheless, as greenhouse gas (GHG) concentrations continue to rise, we expect to see future changes to the climate system that are greater than those already observed and attributed to human activi -ties. It is possible to understand future climate change using models and to use models to characterize outcomes and uncertainties under speci˚c assumptions about future forcing scenarios. This chapter assesses climate projections on time scales beyond those covered in Chapter 11, that is, beyond the mid-21st century. Informa -tion from a range of different modelling tools is used here; from simple energy balance models, through Earth System Models of Intermediate Complexity (EMICs) to complex dynamical climate and Earth System Models (ESMs). These tools are evaluated in Chapter 9 and, where pos -sible, the evaluation is used in assessing the validity of the projections. This chapter also summarizes some of the information on leading-order measures of the sensitivity of the climate system from other chapters and discusses the relevance of these measures for climate projections, commitments and irreversibility.Since the AR4 (Meehl et al., 2007b) there have been a number of advances: New scenarios of future forcings have been developed to replace the Special Report on Emissions Scenarios (SRES). The Represen -tative Concentration Pathways (RCPs, see Section 12.3) (Moss et al., 2010), have been designed to cover a wide range of possible magnitudes of climate change in models rather than being derived sequentially from storylines of socioeconomic futures. The aim is to provide a range of climate responses while individual socioeco-nomic scenarios may be derived, scaled and interpolated (some including explicit climate policy). Nevertheless, many studies that have been performed since AR4 have used SRES and, where appro -priate, these are assessed. Simpli˚ed scenarios of future change, developed strictly for understanding the response of the climate system rather than to represent realistic future outcomes, are also synthesized and the understanding of leading-order measures of climate response such as the equilibrium climate sensitivity (ECS) and the transient climate response (TCR) are assessed. New models have been developed with higher spatial resolution, with better representation of processes and with the inclusion of more processes, in particular processes that are important in simu -lating the carbon cycle of the Earth. In these models, emissions of GHGs may be speci˚ed and these gases may be chemically active in the atmosphere or be exchanged with pools in terrestrial and oceanic systems before ending up as an airborne concentration (see Figure 10.1 of AR4). New types of model experiments have been performed, many coordinated by the Coupled Model Intercomparison Project Phase 5 (CMIP5) (Taylor et al., 2012), which exploit the addition of these new processes. Models may be driven by emissions of GHGs, or by their concentrations with different Earth System feedback loops cut. This allows the separate assessment of different feedbacks in the system and of projections of physical climate variables and future emissions. Techniques to assess and quantify uncertainties in projections have been further developed but a full probabilistic quanti˚ca -tion remains dif˚cult to propose for most quantities, the exception being global, temperature-related measures of the system sensitiv -ity to forcings, such as ECS and TCR. In those few cases, projections are presented in the form of probability density functions (PDFs). We make the distinction between the spread of a multi-model ensemble, an ad hoc measure of the possible range of projections and the quanti˚cation of uncertainty that combines information from models and observations using statistical algorithms. Just like climate models, different techniques for quantifying uncertainty exist and produce different outcomes. Where possible, different estimates of uncertainty are compared.Although not an advance, as time has moved on, the baseline period from which climate change is expressed has also moved on (a common baseline period of 1986Œ2005 is used throughout, consistent with the 2006 start-point for the RCP scenarios). Hence climate change is expressed as a change with respect to a recent period of history, rather than a time before signi˚cant anthropogenic in˛uence. It should be borne in mind that some anthropogenically forced climate change had already occurred by the 1986Œ2005 period (see Chapter 10).The focus of this chapter is on global and continental/ocean basin-scale features of climate. For many aspects of future climate change, it is possible to discuss generic features of projections and the processes that underpin them for such large scales. Where interesting or unique changes have been investigated at smaller scales, and there is a level of agreement between different studies of those smaller-scale changes, these may also be assessed in this chapter, although where changes are linked to climate phenomena such as El Niño, readers are referred to Chapter 14. Projections of atmospheric composition, chemistry and air quality for the 21st century are assessed in Chapter 11, except for CO 2 which is assessed in this chapter. An innovation for AR5 is Annex I: Atlas of Global and Regional Climate Projections, a collection of global and regional maps of projected climate changes derived from model output. A detailed commentary on each of the maps presented in Annex I is not provided here, but some discussion of generic features is provided. Projections from regional models driven by boundary conditions from global models are not extensively assessed but may be mentioned in this chapter. More detailed regional information may be found in Chapter 14 and is also now assessed in the Working Group II report, where it can more easily be linked to impacts.
PAGE – 8 ============
1036Frequently Asked Questions FAQ 12.1 | Why Are So Many Models and Scenarios Used to Project Climate Change? Future climate is partly determined by the magnitude of future emissions of greenhouse gases, aerosols and other natural and man-made forcings. These forcings are external to the climate system, but modify how it behaves. Future climate is shaped by the Earth™s response to those forcings, along with internal variability inherent in the climate system. A range of assumptions about the magnitude and pace of future emissions helps scientists develop different emission scenarios, upon which climate model projections are based. Different climate models, mean -while, provide alternative representations of the Earth™s response to those forcings, and of natural climate variabil -ity. Together, ensembles of models, simulating the response to a range of different scenarios, map out a range of possible futures, and help us understand their uncertainties.Predicting socioeconomic development is arguably even more dif˜cult than predicting the evolution of a physical system. It entails predicting human behaviour, policy choices, technological advances, international competition and cooperation. The common approach is to use scenarios of plausible future socioeconomic development, from which future emissions of greenhouse gases and other forcing agents are derived. It has not, in general, been pos-sible to assign likelihoods to individual forcing scenarios. Rather, a set of alternatives is used to span a range of possibilities. The outcomes from different forcing scenarios provide policymakers with alternatives and a range of possible futures to consider. Internal ˚uctuations in climate are spontaneously generated by interactions between components such as the atmosphere and the ocean. In the case of near-term climate change, they may eclipse the effect of external per -turbations, like greenhouse gas increases (see Chapter 11). Over the longer term, however, the effect of external forcings is expected to dominate instead. Climate model simulations project that, after a few decades, different scenarios of future anthropogenic greenhouse gases and other forcing agentsŠand the climate system™s response to themŠwill differently affect the change in mean global temperature (FAQ 12.1, Figure 1, left panel). Therefore, evaluating the consequences of those various scenarios and responses is of paramount importance, especially when policy decisions are considered.Climate models are built on the basis of the physical principles governing our climate system, and empirical under-standing, and represent the complex, interacting processes needed to simulate climate and climate change, both past and future. Analogues from past observations, or extrapolations from recent trends, are inadequate strategies for producing projections, because the future will not necessarily be a simple continuation of what we have seen thus far. Although it is possible to write down the equations of ˚uid motion that determine the behaviour of the atmo -sphere and ocean, it is impossible to solve them without using numerical algorithms through computer model simulation, similarly to how aircraft engineering relies on numerical simulations of similar types of equations. Also, many small-scale physical, biological and chemical processes, such as cloud processes, cannot be described by those equations, either because we lack the computational ability to describe the system at a ˜ne enough resolution to directly simulate these processes or because we still have a partial scienti˜c understanding of the mechanisms driving these processes. Those need instead to be approximated by so-called parameterizations within the climate models, through which a mathematical relation between directly simulated and approximated quantities is estab-lished, often on the basis of observed behaviour. There are various alternative and equally plausible numerical representations, solutions and approximations for modelling the climate system, given the limitations in computing and observations. This diversity is considered a healthy aspect of the climate modelling community, and results in a range of plausible climate change projections at global and regional scales. This range provides a basis for quantifying uncertainty in the projections, but because the number of models is relatively small, and the contribution of model output to public archives is voluntary, the sampling of possible futures is neither systematic nor comprehensive. Also, some inadequacies persist that are common to all models; different models have different strength and weaknesses; it is not yet clear which aspects of the quality of the simulations that can be evaluated through observations should guide our evaluation of future model simulations. (continued on next page)
120 KB – 108 Pages