[eprints@whiterose.ac.uk](mailto:eprints@whiterose.ac.uk) [https://eprints.whiterose.ac.uk](https://eprints.whiterose.ac.uk/) **Universities of Leeds, Sheffield and York** ## Deposited via The University of Leeds. White Rose Research Online URL for this paper: [https://eprints.whiterose.ac.uk/id/eprint/165692/](https://eprints.whiterose.ac.uk/id/eprint/165692/) ## Version: Accepted Version **Article:** Lee, DS, Fahey, DW, Skowron, A et al. (2021) The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. Atmospheric Environment, 244. 117834. ISSN: 1352-2310 [https://doi.org/10.1016/j.atmosenv.2020.117834](https://doi.org/10.1016/j.atmosenv.2020.117834) © 2020, Elsevier. This manuscript version is made available under the CC-BY-NC-ND 4.0 license [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/). **Reuse** This article is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. This licence only allows you to download this work and share it with others as long as you credit the authors, but you can’t change the article in any way or use it commercially. More information and the full terms of the licence here: [https://creativecommons.org/licenses/](https://creativecommons.org/licenses/) **Takedown** If you consider content in White Rose Research Online to be in breach of UK law, please notify us by emailing [eprints@whiterose.ac.uk](mailto:eprints@whiterose.ac.uk) including the URL of the record and the reason for the withdrawal request. * * * 1 2 3 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018 a, 1 b a c,n d e f D. S. Lee, D. W. Fahey, A. Skowron, M. R. Allen, U. Burkhardt, Q. Chen, S. J. Doherty, S. 4 a g h i a a h Freeman, P.M. Forster, J. Fuglestvedt, A. Gettelman, R. R. De León, L. L. Lim, M. T. Lund, R. J. 5 c,o a j l k d m Millar, B. Owen, J. E. Penner, G. Pitari, M. J. Prather, R. Sausen, L. J. Wilcox 6 a Faculty of Science and Engineering, Manchester Metropolitan University, John Dalton Building, Chester Street, Manchester M1 5GD, United Kingdom; b b NOAA Chemical Sciences Laboratory (CSL), Boulder, CO USA; c c School of Geography and the Environment, University of Oxford, Oxford, UK; d d Deutsches Zentrum für Luft- und Raumfahrt (DLR), Institut für Physik der Atmosphäre, Oberpfaffenhofen, Germany; e e State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking University, Beijing 100871, China; f f Cooperative Institute for Research in Environmental Sciences (CIRES), University of Colorado, Boulder, CO, USA; g h CICERO—Center for International Climate Research—Oslo, PO Box 1129, Blindern, 0318 Oslo, Norway; i g School of Earth and Environment, University of Leeds, Leeds LS2 9JT, United Kingdom; h CICERO—Center for International Climate Research—Oslo, PO Box 1129, Blindern, 0318 Oslo, Norway; i National Center for Atmospheric Research, Boulder, CO, USA; j j Department of Climate and Space Sciences and Engineering, University of Michigan, 2455 Hayward St., Ann Arbor, MI 48109-2143, USA; k Department of Earth System Science, University of California, Irvine, 3329 Croul Hall, CA 92697-3100, USA; l Department of Physical and Chemical Sciences, Università dell'Aquila, Via Vetoio, 67100 L'Aquila, Italy; l Department of Physical and Chemical Sciences, Università dell'Aquila, Via Vetoio, 67100 L'Aquila, Italy; m m National Centre for Atmospheric Science, Department of Meteorology, University of Reading, Earley Gate, Reading RG6 6BB, UK; n n also at the Department of Physics, University of Oxford, Oxford, UK; o o also at the Committee on Climate Change, 151 Buckingham Palace Road, London, SW1W 9SZ, UK. 1 To whom correspondence should be addressed. Email: [d.s.lee@mmu.ac.uk](mailto:d.s.lee@mmu.ac.uk) Tel: +44 161 247 3663 * * * 24 August 2020 Revised ## 32 Highlights - Global aviation warms Earth's surface through both CO2 and net non-CO2 contributions. 33 - Global aviation contributes a few percent to anthropogenic radiative forcing. 34 - Non-CO2 impacts comprise about 2/3 of the net radiative forcing. 35 - Comprehensive and quantitative calculations of aviation effects are presented. 36 - Data are made available to analyze past, present and future aviation climate forcing. 37 38 ## 39 Abstract Global aviation operations contribute to anthropogenic climate change via a complex set of processes that 40 lead to a net surface warming. Of importance are aviation emissions of carbon dioxide (CO2), nitrogen 41 oxides (NOx), water vapor, soot and sulfate aerosols, and increased cloudiness due to contrail formation. 42 Aviation grew strongly over the past decades (1960–2018) in terms of activity, with revenue passenger 43 kilometers increasing from 109 to 8269 billion km yr -1 , and in terms of climate change impacts, with CO2 44 emissions increasing by a factor of 6.8 to 1034 Tg CO2 yr -1 . Over the period 2013–2018, the growth rates 45 in both terms show a marked increase. Here, we present a new comprehensive and quantitative approach 46 for evaluating aviation climate forcing terms. Both radiative forcing (RF) and effective radiative forcing 47 (ERF) terms and their sums are calculated for the years 2000–2018. Contrail cirrus, consisting of linear 48 contrails and the cirrus cloudiness arising from them, yields the largest positive net (warming) ERF term 49 followed by CO2 and NOx emissions. The formation and emission of sulfate aerosol yields a negative 50 (cooling) term. The mean contrail cirrus ERF/RF ratio of 0.42 indicates that contrail cirrus is less 51 effective in surface warming than other terms. For 2018 the net aviation ERF is +100.9 milliwatts (mW) 52 m -2 (5–95% likelihood range of (55, 145)) with major contributions from contrail cirrus (57.4 mW m -2 ), 53 CO2 (34.3 mW m -2 ), and NOx (17.5 mW m -2 ). Non-CO2 terms sum to yield a net positive (warming) ERF 54 that accounts for more than half (66%) of the aviation net ERF in 2018. Using normalization to aviation 55 fuel use, the contribution of global aviation in 2011 was calculated to be 3.5 (4.0, 3.4) % of the net 56 anthropogenic ERF of 2290 (1130, 3330) mW m -2 . Uncertainty distributions (5%, 95%) show that non-57 CO2 forcing terms contribute about 8 times more than CO2 to the uncertainty in the aviation net ERF in 58 2018. The best estimates of the ERFs from aviation aerosol-cloud interactions for soot and sulfate remain 59 undetermined. CO2-warming-equivalent emissions based on global warming potentials (GWP\* method) 60 indicate that aviation emissions are currently warming the climate at approximately three times the rate of 61 that associated with aviation CO2 emissions alone. CO2 and NOx aviation emissions and cloud effects 62 remain a continued focus of anthropogenic climate change research and policy discussions. 63 **Key words:** \| aviation \| contrail cirrus \| climate \| radiative forcing \| CO2 \| NOx \| 64 **Dedication: This paper is dedicated to the memory of Professor Ivar S. A. Isaksen of the University of 65** 66 Oslo, whose scientific excellence, friendship, and mentorship is sorely missed. ## 68 1. Introduction Aviation is one of the most important global economic activities in the modern world. Aviation emissions 69 of CO2 and non-CO2 aviation effects result in changes to the climate system (Figure 1). Both aviation 70 CO2 and the sum of quantified non-CO2 contributions lead to surface warming. The largest contribution to 71 anthropogenic climate change across all economic sectors comes from the increase in CO2 concentration, 72 which is the primary cause of observed global warming in recent decades (IPCC, 2013; 2018). Aviation 73 contributions involve a range of atmospheric physical processes, including plume dynamics, chemical 74 transformations, microphysics, radiation, and transport. Aggregating these processes to calculate changes 75 in a greenhouse gas component or a cloud radiative effect is a complex challenge for contemporary 76 * * * 24 August 2020 Revised atmospheric modeling systems. Given the dependence of aviation on burning fossil fuel, its significant 77 78 CO2 and non-CO2 effects, and the projected fleet growth, it is vital to understand the scale of aviation’s impact on present-day climate forcing. 79 Historically, estimating aviation non-CO2 effects has been particularly challenging. The primary 80 (quantified) non-CO2 effects result from the emissions of NOx, along with water vapor and soot that can 81 result in contrail formation. Aviation aerosols are small particles composed of soot (black and organic 82 carbon (BC/OC)) and sulfur (S) and nitrogen (N) compounds. The largest positive (warming) climate 83 forcings adding to that of CO2 are those from contrail cirrus and from NOx-driven changes in the chemical 84 composition of the atmosphere (Lee et al., 2009 (L09)). L09 estimated that in 2005, aviation CO2 85 radiative forcing (RF (Wm -2 )) was 1.59% of total anthropogenic CO2 RF and that the sum of aviation CO2 86 and non-CO2 effects contributed about 5% of the overall net anthropogenic forcing. 87 Understanding of aviation’s impacts on the climate system has improved over the decade since the last 88 comprehensive evaluation (L09), but remains incomplete. Published studies of aviation contributions to 89 climate change generally focus on one or a few ERF terms. For example, about 20 studies are cited here 90 that quantify the contribution from global NOx emissions. In contrast, only a few studies have addressed 91 the net RF from global aviation (IPCC, 1999; Sausen et al., 2005; L09). A more recent study updated 92 some aviation terms without providing a net RF (Brasseur et al., 2016). Here, a comprehensive analysis of 93 individual aviation ERFs is undertaken in order to provide an overall ERF for global aviation, along with 94 the associated uncertainties, which is an analysis unavailable elsewhere. This step updates and improves 95 the analysis of L09. Best estimates of individual aviation ERF terms are derived here for the first time and 96 97 combined to provide a net ERF for global aviation. Quantifying the terms required new analyses of CO2 and NOx ERFs and recalibration of other individual ERFs accounting for factors not previously applied in 98 a common framework. 99 In L09, the net RF was calculated with and without the full contrail cirrus term but including an estimate 100 for linear contrails. The exclusion was based on the lack of a best estimate derived from existing studies. 101 At that time radiative forcing estimates were limited to linear or line-shaped contrails since the modelling 102 approaches required scaling contrail formation frequency to observed coverage and only satellite 103 observations of linear contrails existed (Burkhardt et al., 2010). The contrail cirrus term requires the 104 simulation of the whole contrail cirrus life cycle, starting from persistent linear contrails which spread and 105 often become later indistinguishable from natural cirrus. Persistent contrail formation requires ice-106 107 supersaturated conditions along a flight track, which are variable in space and time in the troposphere and tropopause region (Irvine et al., 2013). Estimating the RF from contrail cirrus requires knowledge of 108 complex microphysical processes, radiative transfer, and the interaction with background cloudiness 109 (Burkhardt et al., 2010). Contrail cirrus forcing dominates that of persistent linear contrails with the latter 110 on the order of 10% of the combined forcing (Burkhardt and Kärcher, 2011). In the present study, we 111 present a best estimate and uncertainty based on the results from global climate models employing 112 process-based contrail cirrus parameterizations. 113 Emissions of NOx from aviation lead to photochemical changes that increase global ozone (O3) formation 114 while decreasing the lifetime and abundance of methane (CH4). The changes result in positive and 115 116 negative (cooling) RF contributions, respectively. Since L09, improved understanding and modeling capabilities have emerged, as well as additional RF terms in response to NOx emissions, namely a longer-117 term decrease in background O3 and a reduction in H2O in the stratosphere in response to decreased CH4. 118 Here, model results are used to calculate the additional RF terms, and to incorporate the updated CH4 119 forcing as assessed by Etminan et al. (2016) and the equilibrium-to-transient corrections for the CH4 term 120 (see Appendix D. Finally, aviation-specific efficacies (Appendix C) of the individual NOx components 121 are used to estimate a net NOx ERF for the first time. 122 * * * 24 August 2020 Revised L09 includes best estimates for the RFs resulting from the aerosol-radiation interactions (previously 123 called direct effects) of soot and sulfate aerosols from aviation. However, no best estimates of RFs from 124 aerosol-cloud interactions (previously called indirect effects) were available in 2009. Subsequent studies 125 discussed here have yet to provide a basis for best estimates of ERFs from aviation aerosol-cloud 126 interactions that may be significant. 127 The primary motivations for the present study are to provide an updated, comprehensive evaluation of 128 aviation climate forcings in terms of RF and ERF based on new calculations and the normalization of 129 values from published modeling studies, and to combine the resulting best estimates via a Monte-Carlo 130 analysis to yield a best estimate for the net ERF for global aviation for the years 2000 to 2018. The three 131 years 2018, 2011, and 2005 are notable because the year 2018 is the latest year for which air traffic and 132 fuel use datasets are available, 2011 is the most recent year evaluated for net anthropogenic climate 133 forcing by the IPCC (IPCC, 2013), and 2005 is the year evaluated in the latest comprehensive aviation 134 and climate evaluation (L09). By normalizing the calculations across these years, more specific and self-135 consistent comparisons can be made of the changes in aviation contributions over time. The normalization 136 step requires addressing in each study, for example, the choice of air traffic inventory, the integration of 137 emissions along flight tracks, and the assumed jet-engine emission indices. The new best estimates of 138 aviation ERF, for example, show that the 2018 value is about 48% larger than the updated 2005 value. 139 In general, previous global aviation climate assessments have made different assumptions concerning 140 emissions, cloudiness effects, and aviation operations (e.g., IPCC, 1999). Here, our self-consistent set of 141 component and net aviation ERFs for 2000 to 2018 allows historical and scenario projections of aviation 142 climate impacts to be assessed in context with other sectors, such as maritime shipping, ground 143 transportation and energy generation. This updated understanding is especially important given the 144 potential role of international aviation in meeting the goals of the Paris Agreement (Section 2) on limiting 145 future temperature increases. 146 The remaining sections address global aviation growth statistics (Section 2); a brief summary of methods 147 used in the analysis (Section 3); results for the ERF estimates of CO2, NOx, water vapor, contrail cirrus, 148 and aerosol-radiation and aerosol-cloud interactions with soot and sulfate (Section 4); results for the net 149 ERF of global aviation (Section 5); emission metrics (Section 6); and aviation CO2 vs non-CO2 forcings 150 (Section 7). The appendices contain additional detailed information on trends in aviation emissions (App. 151 A); aviation CO2 radiative forcing calculations (App. B); radiative forcing, efficacy and ERF definitions 152 (App. C); aviation NOx RF calculations (App. D); contrail cirrus RF scaling factors and uncertainty (App. 153 E); and emission equivalency metric calculations (App. F). A Supplemental Data (SD) file is provided 154 containing the interactive spreadsheet used to calculate RFs and ERFs for each aviation term. 155 ## 156 2. Global aviation growth Global aviation fuel use and CO2 emissions have increased in the last four decades with large growth 157 occurring in Asia and other developing regions due to the rapid expansion of civil aviation (Figure 2 and 158 Appendix A). Looking forward, this pattern of growth is expected to be maintained—for example, of the 159 1229 orders of Airbus and 1031 orders of Boeing in 2017, 20.3% and 37.5%, respectively, are for airlines 160 in the Asia region (Airbus, 2017; Boeing, 2018). Airbus projects 41% of orders over the next two decades 161 to be from the Asia-Pacific region (Airbus, 2017). The uncertainty in this expectation has increased due to 162 the slowdown in aviation operations in the early months of 2020 due to the COVID-19 pandemic (Le 163 Quéré et al., 2020). Annual aviation emissions in 2020 are now expected to be below recent projections 164 that are based on historical growth. 165 A striking feature of Figure 2a is the sustained multi-decade growth in CO2 emissions; the average rate 166 for the period 1960–2018 is 15 Tg CO2 yr -1 . The growth rate for 2013 through 2018 is much larger (44 Tg 167 CO2 yr -1 ). The annually averaged growth rate over the period 1970 to 2012 is 2.2% yr -1 and for 2013 to 168 2018 is 5% yr -1 (increase of 27%). In 2018, global aviation CO2 emissions exceeded 1000 million tonnes 169 * * * 24 August 2020 Revised per year for the first time (see methodology for scaling 2016 IEA data in Appendix A). The cumulative 170 emissions of global aviation (1940 to 2018) are 32.6 billion (10 9 ) tonnes of CO2, of which approximately 171 50% were emitted in the last 20 years. Current (2018) CO2 emissions from aviation represent 172 approximately 2.4% of anthropogenic emissions of CO2 (including land use change) (Figure 2c). 173 Aviation has grown strongly over time (Figure 2b) in terms of available seat kilometers (ASK, a measure 174 175 of capacity) and revenue passenger kilometers (RPK, a measure of transport work). Fuel usage and hence CO2 emissions have grown at a lesser rate than RPK, reflecting increases in aircraft efficiency derived 176 from changes in technology, larger average aircraft sizes and increased passenger load factor. Aviation 177 transport efficiency has improved by approximately eightfold since 1960, to 125 gCO2 (RPK) -1 . 178 At present and for some considerable time into the future, aviation growth is likely to be largely 179 dependent upon the combustion of kerosene fossil fuel (Jet A-1/A) (OECD, 2012), resulting in emission 180 of CO2. Renewable biofuels partially offset fossil fuel emissions but these have yet to be produced in 181 sufficient quantities to offset growth of fossil fuel use. Furthermore, considerable uncertainties remain 182 regarding the life-cycle emissions of biofuels, which determine the reductions in net CO2 emissions (e.g., 183 Hari et al., 2015). There are current regulations regarding aviation emissions of CO2, NOx, and soot mass 184 and number based on decisions by the International Civil Aviation Organization (ICAO). Under the 2016 185 Paris climate agreement, nations are committing to limiting future increases in global temperatures with 186 Nationally Determined Contributions (NDCs) (UNFCCC). Whereas domestic aviation CO2 emissions are 187 included in the NDCs, CO2 emissions from international aviation are not mentioned in the agreement. It 188 remains open as to whether emissions from international aviation or global emissions beyond greenhouse 189 gases (e.g., short-lived (non-CO2) climate forcers) will be included in future international agreements. 190 ## 191 3. Methods The methodologies used to calculate ERF and RF for individual aviation terms are described in this 192 section, and results of these calculations are given in Section 4. Common to the methodologies is a 193 comprehensive multi-page spreadsheet (see SD) that begins with a user's guide. The spreadsheet pages 194 include those for contrail cirrus, CO2, NOx, H2O, and sulfate and soot aerosol, along with CO2-equivalent 195 metrics, ERF probability distributions, ERF time series, and estimates of forcings from aerosol-cloud 196 effects. The spreadsheet displays the results of aviation forcings provided by individual published studies. 197 ERF and RF values were calculated for 2018 and other years based on the normalized values of ERF or 198 RF per unit emission or distance, choice of appropriate emission indices, and times series data on fuel use 199 and distance travelled. In the case of the contrail cirrus forcing, the flight-track distance was chosen as the 200 proxy over fuel usage. Annual global emissions are derived from fuel burn by multiplying by the average 201 emission indices (Table 1). The combined and normalized results are used to create sets of RF and ERF 202 aviation terms for the years 2000 to 2018. In addition to facilitating the present study, the spreadsheet also 203 provides a quantitative framework for follow-on analyses. 204 Calculations of radiative forcing are expanded here beyond the approach in L09 to include ERF values in 205 addition to the traditional RF values (Tables 2 and 3 and Figure 3). The distinction between ERF and 206 RF is presented in Appendix C. ERF is the preferred metric for comparing the expected impacts of 207 climate forcing terms (Myhre et al., 2013). Its use derives from the stronger correlation between ERF and 208 the change in the equilibrium global-mean surface temperature for some forcing agents than for the 209 210 corresponding RF. ERF is calculated as the change in net top-of-the-atmosphere (TOA) downward 211 radiative flux after allowing for rapid adjustments in atmospheric temperatures, water vapor and clouds 212 with globally-averaged sea surface and/or land surface temperatures unchanged. ERF is preferred over RF 213 estimates because the imposed forcing and rapid responses to the forcing cannot always be separately evaluated, especially for aerosols. In general, the largest differences between ERF and RF are expected 214 for aerosol-cloud interactions and contrail cirrus (Myhre et al., 2013; Boucher et al., 2013). In calculating 215 ERF values for 2000-2018, the ERF/RF ratio is assumed to be constant with time. 216 * * * 24 August 2020 Revised Most of the results for the non-CO2 terms have associated statistics from which the median was chosen as 217 the best estimate, including the net aviation ERF and RF, and the net non-CO2 ERF and RF. For CO2 and 218 contrail cirrus, for which the sample sizes are small (3, in both cases), the mean was used as the best 219 estimate. The best estimates of the non-CO2 terms except contrail cirrus have associated uncertainties 220 expressed as 5% and 95% confidence intervals calculated from 5, 95% percentile statistics. The 221 uncertainty distributions for all forcing terms other than CO2 and contrail cirrus are lognormal and that for 222 net NOx has a discrete probability distribution function (PDF). The uncertainties for the ERF and RF of 223 CO2 were taken from IPCC (2013) and fitted with a Monte Carlo analysis with a normal distribution (see 224 Section 5). The uncertainties for contrail cirrus were estimated partly from expert judgement of the 225 underlying processes, as described in Appendix E, again fitted with a Monte Carlo analysis with a normal 226 distribution. 227 228 **4\. Calculations of ERFs for aviation terms** ## 229 4.1. CO2. The time series of aviation CO2 emissions is shown in Figure 2 as derived from combined kerosene and 230 avgas usage (UKDS, 2016). Calculating CO2 concentrations from emissions requires use of a global 231 carbon-cycle model, which has a range of complexity from a comprehensive Earth system model (ESM) 232 233 to a simple climate model (SCM), with the latter being based on a box model or impulse response function (IRF) model. Three SCMs were used here: LinClim, an IRF model based on Sausen and 234 Schumann (2000) (Appendix B); the Finite-amplitude Impulse Response (FaIR) model (Millar et al., 235 2017); and the CICERO-SCM (Fuglestvedt and Berntsen, 1999; Skeie et al., 2017). The performance of 236 LinClim and CICERO-SCM with respect to aviation emissions is documented in the multi-model 237 comparison of Khodayari et al. (2013). The CO2 concentrations attributable to aviation in 2018 based on 238 LinClim, CICERO-SCM and FaIR are 2.9, 2.4 and 2.4 ppm, respectively, with concentrations nearly 239 doubling in the last 20 years (see SD spreadsheet). The ERF/RF ratio for CO2 is assumed to be unity. The 240 resulting CO2 ERFs, as derived from global concentrations using standard IPCC expressions (IPCC, 241 2001), are 38.6, 32.0 and 32.4 mW m -2 , respectively. With only three model estimates, the average of 34.3 242 mW m -2 (5 and 95% percentiles of 29 and 40 mW m -2 ), is chosen be the CO2 RF best estimate. 243 ## 244 4.2. NOx The photochemical effects of aviation NOx emissions on the atmospheric abundances of O3, CH4, carbon 245 monoxide (CO) and reactive hydrogen (HOx) are well established (Fuglestvedt et al., 1999). Earlier 246 studies assessed the short-term increase of O3 and the longer-term reduction in CH4 lifetime and 247 abundance, which yield positive and negative RFs, respectively (IPCC, 1999; Sausen et al., 2005). L09 248 introduced the concept of the ‘net NOx’ effect by combining the two components, extending and updating 249 the study of Sausen et al. (2005). Later studies expanded the analysis of NOx effects to include the long-250 term decreases in both O3 and stratospheric water vapor (SWV) resulting from the CH4 reduction. Both 251 252 effects yield negative RFs (Holmes et al., 2011; Myhre et al., 2011). In the present study, an ensemble of 20 NOx studies is assessed to provide NOx forcing best estimates based on a wide range of global 253 atmospheric chemistry/climate models and a broad range of present-day aviation emission inventories 254 (details in Appendix D and SD spreadsheet). Results from 6 of the studies were adopted from Holmes et 255 al. (2011). 256 The study ensemble represents various model methodologies in calculating and treating both the short-257 term and the long-term NOx components. In order to avoid gaps and additional uncertainties, standardized 258 ERFs were developed that estimated disparate elements (e.g., CH4 mediated decreases in SWV and long-259 term O3). Moreover, most of the studies were based upon a parameterization of the CH4 response that 260 assumed a full equilibrium response. In order to calculate the transient response for a specific year more 261 accurately, a correction factor is needed (Myhre et al., 2011). Here, the CH4 responses for individual 262 years were calculated (see Appendix D) using the difference between two simulations with differing 263 * * * 24 August 2020 Revised aviation NOx emissions. A number of transient and equilibrium simulations were conducted with a 2D 264 chemical-transport model to find that the requirement for a correction factor is well supported and that the 265 266 2018 value is 0.79 (see Transient vs. equilibrium in Appendix D and Appendix Table D.2). In addition, a scaling factor (1.23) is applied to derived CH4 ERF numbers to account for the effect of shortwave CH4 267 forcing, following Etminan et al. (2016) (see Appendix D). The existence and nature of correlations 268 between the NOx RF components were also explored (see Correlations in Appendix D and Appendix 269 Figure D.1) since the degree of correlation between short-term O3 and CH4 terms was a source of 270 uncertainty in the calculation of the net-NOx forcing in L09. The work of Holmes et al. (2011) supports 271 the prior assumption of correlation, which is greatly expanded here. Regardless of inter-model 272 differences, significant correlations are observed; for example, a significant negative correlation (p = -0.7) 273 exists between the short-term and the long-term NOx RF components. 274 The normalized sensitivity results for net NOx in units of mW m -2 (Tg (N) yr -1 ) -1 for the individual 275 modeling studies are shown in Figure 4 along with statistical parameters (see Ensemble values in 276 Appendix D). Given the diversity of studies conducted over nearly two decades, the standard deviations 277 of the distributions are reasonably small. In contrast, the sign of the net-NOx RF obtained from summing 278 over the 4 component values varies from positive to negative. The spread in NOx RF values is caused by 279 280 various factors (e.g., emissions inventories, experimental design or inter-model differences) and is particularly sensitive to the NOx distribution in the model background troposphere (Holmes et al., 2011). 281 The NOx efficacies are 1.37 for the short-term ozone increases and 1.18 for methane decreases (Ponater et 282 al., 2006). The efficacies do not equal the ERF/RF ratios, in general (Ponater et al., 2020; Appendix C); 283 nonetheless, in the present study, we assume the efficacies and the ERF/RF ratios are equal, in the 284 absence of better information. The factor of 1.18 was similarly adopted for the CH4-mediated decreases in 285 long-term ozone and SWV. It is noted that these ratios are from one study and that, in general, the ratio of 286 ERF to RF for CH4 and tropospheric O3 are currently the subject of some debate (Smith et al., 2018; Xie 287 et al., 2016; Richardson et al., 2019). Given the strength of the net effect of the ERF adjustment on the net 288 NOx forcing (more than doubling over its stratosphere-adjusted RF), these ratios warrant further study. 289 The net-NOx ERF sensitivity of 5.5 ± 8.1 mW m -2 (Tg (N) yr -1 ) -1 yields a 2018 best estimate of 17.5 (0.6, 290 28.5) mW m -2 . This best estimate includes the correction factor for non-steady state conditions as well as 291 the revised formulation of CH4 RF (Appendix D). 292 Other potential short-term effects from NOx emissions involve the direct formation of nitrate aerosol and 293 indirect enhancement of sulfate aerosol. These effects, addressed in a few modelling studies, are 294 associated with large uncertainties (Righi et al., 2013; Pitari et al., 2017; Unger, 2011). The effects of 295 NOx on aerosol abundances are not further considered here owing to the limited number of studies and the 296 large associated uncertainties. 297 ## 298 4.3. Water vapor emissions. A large fraction of annual aircraft emissions from the global fleet occurs in the stratosphere, primarily in 299 the northern hemisphere (Forster et al., 2003). The accumulation of water vapor emissions perturbs the 300 low background humidity in the lower stratosphere and changes the water vapor radiative balance. 301 Calculating the water vapor RF is complicated by the sensitivity to the vertical and horizontal distribution 302 of emissions, seasonal changes in tropopause heights, and short stratospheric residence times. Some 303 earlier studies do not include the water vapor effect. 304 The water vapor effects were explored in detail (see SD) using results from nine studies: IPCC (1999), 305 Marquart et al. 2001, Gauss et al. (2003), Ponater et al. (2006), Frömming et al. (2012), Wilcox et al. 306 (2012), Lim et al. (2015), Pitari et al. (2015) and Brasseur et al. (2016). The reported RFs from these 307 studies vary from 0.4 mW m -2 (Wilcox et al., 2012) through 1.5 mW m -2 (Frömming et al. 2012, Lim et 308 al., 2015) to 3.0 mW m -2 (IPCC, 1999). The differences are attributed to the different transport models 309 used, with some contribution from the different meteorologies in different studies. Normalizing to the 310 * * * 24 August 2020 Revised same emissions and averaging these reported estimates yields a water vapor sensitivity of 0.0052 ± 311 0.0026 mW m -2 (Tg (H2O) yr -1 ) -1 . Scaling this value linearly to emissions of 382 Tg H2O yields an ERF 312 best estimate of 2.0 (0.8, 3.2) mW m -2 for 2018, which is well within the uncertainty range of the 2005 313 L09 value of 2.8 (0.39, 20.3) mW m -2 . The ERF/RF ratio for stratospheric water increases is assumed to 314 be unity. We have greater confidence in the new estimate and its smaller uncertainty since it is based on 315 detailed physical studies, rather than a scaling of the earlier IPCC (1999) estimate. The new best estimate 316 is also in good agreement with the earlier results of Gauss et al. (2003) and Ponater et al. (2006), after 317 scaling their results to account for emissions differences. 318 ## 319 4.4. Contrail cirrus. The aviation fleet increases global cloudiness through the formation of persistent contrails when the 320 ambient atmosphere is supersaturated with respect to ice (IPCC, 1999). Contrail cirrus, consisting of 321 linear contrails and the cirrus cloudiness arising from them, have cooling (short-wave) and warming 322 (long-wave) effects, with the effect at night being exclusively warming. In past assessments (e.g., IPCC, 323 1999; L09), a best estimate was only available for the RF of linear persistent contrails, in part because of 324 the difficulty of quantifying the cloudiness contribution of aging and spreading contrails (Minnis et al., 325 2013). The ERF of contrail cirrus was estimated for 2011 as 50 (20, 150) mW m -2 by Boucher et al. 326 (2013). Results of a recent assessment of contrail cirrus and other aviation effects are included here, 327 although the study did not propose new best estimates (Brasseur et al., 2016). 328 A persistent contrail requires ice-supersaturated conditions along the flight track. Contrail cirrus life 329 cycles are dependent on the temporal and spatial scales of the ice supersaturated areas, which are highly 330 variable in the troposphere and tropopause region (e.g., Lamquin et al., 2012; Irvine et al., 2013; Bier et 331 332 al., 2017). Estimating the impact of contrail cirrus on upper tropospheric cloudiness requires the simulation of complex microphysical processes, contrail spreading, overlap with natural clouds, radiative 333 transfer, and the interaction with background cloudiness (Burkhardt et al., 2010). We present new best 334 estimates based on the results of global climate models employing process-based contrail cirrus 335 parameterizations (Appendix E). Due to the small number of independent estimates the uncertainty must 336 be estimated from the sensitivities of the respective processes and the uncertainty in the underlying 337 338 parameters and fields. Here, we consider RF and ERF estimates from global climate models (Burkhardt and Kärcher, 2011; 339 Bock and Burkhardt, 2016; Chen and Gettelman, 2013; Schumann et al., 2015; Bickel et al., 2019) to 340 ultimately produce an ERF best estimate. For the present study, the Chen and Gettelman study was 341 repeated with lower prescribed initial ice-crystal diameters, thereby bringing assumptions in line with 342 measurements (e.g., Schumann et al., 2017). Since the RF estimates differ regarding the air traffic 343 inventory, the measure of air traffic distance (i.e., taking only surface-projected or overall flight distances 344 into account) and the temporal resolution of the air traffic data, the estimates were homogenized using 345 known sensitivities (Bock and Burkhardt, 2016) (see Appendix E). Furthermore, the estimates were 346 corrected to account for the underestimation of the contrail cirrus RF, as calculated by climate models that 347 use frequency bands, relative to more detailed line-by-line radiative transfer calculations (Myhre et al., 348 2009). The Chen and Gettelman (2013) study is closer to a calculation of an ERF, since it accounts for 349 fast feedbacks on natural clouds, which Bickel et al. (2019) show in their model explains most of the 350 differences between an ERF and an RF calculation. Bickel et al. (2019) presents an explicit calculation of 351 the contrail cirrus ERF and uses the same basic model formulation of Bock and Burkhardt, so the ERF 352 calculation was not used here directly but rather the estimation of the ERF/RF ratio was used. 353 The RF best estimate for 2011 was calculated here for comparison to the most recent IPCC estimate 354 (Boucher et al., 2013). With each study weighted equally, the resulting 2011 RF best estimate for contrail 355 cirrus (excluding any adjustments) is approximately 86 (25, 146) mW m -2 (see Table 3). The IPCC best 356 estimate of 50 (20, 150) mW m -2 (including the natural cloud feedback) was derived from scaling and 357 * * * 24 August 2020 Revised averaging two studies. IPCC assigned a large uncertainty and low confidence to reflect important aspects 358 with incomplete knowledge (e.g., spreading rate, optical depth, and radiative transfer). The RF best 359 estimate derived here for 2018 is 111 (33, 189) mW m -2 . The uncertainties in the present study are 360 reduced due to the development of process-based approaches simulating contrail cirrus in recent years. 361 The uncertainty in the new RF estimate, excluding the uncertainty in the ERF/RF scaling of individual RF 362 363 values, is ±70%, a value substantially lower than the factor of three stated in IPCC. The ±70% uncertainty was derived differently than for the NOx forcing due to the smaller number of 364 available studies. Instead, the uncertainty was derived from the combined uncertainties associated with 365 the processes involved (see Appendix E). The processes fall into two groups: those connected with the 366 upper tropospheric water budget and the contrail cirrus scheme itself, and those associated with the 367 change in radiative transfer due to the presence of contrail cirrus. We considered uncertainty in upper 368 tropospheric ice-supersaturation frequencies and their simulation in global models and the uncertainty of 369 ice-crystal numbers due to uncertainty in soot-number emissions, ice nucleation within the plume, and 370 loss processes in the contrail’s vortex phase. Finally, an important uncertainty comes from the adjustment 371 of natural clouds (Burkhardt and Kärcher, 2011). There is also a small uncertainty associated with the 372 contrail cirrus life cycle, which affects the difference in nighttime and daytime contrail cirrus cover 373 (Stuber et al., 2006) based on work analyzing the diurnal cycle (Chen and Gettelman, 2013; Newinger 374 and Burkhardt, 2012). 375 Uncertainty connected with the radiative response to contrail cirrus is largely due to the differences in the 376 radiation schemes across climate models and the approximations made therein (Myhre et al., 2009; 377 Gounou and Hogan, 2007); the background cloud field and its vertical overlap with contrail cirrus; and 378 assumptions about the homogeneity of the contrail cirrus field. Furthermore, the presence of very small 379 380 ice crystals (<5µm) (Bock and Burkhardt, 2016) and unknown ice-crystal habits (Markowicz and Witek, 2011. add to the uncertainty. 381 Our best estimate of the contrail cirrus uncertainty does not include the impact of contrails forming within 382 natural clouds, which was recently shown to be observable from space (Tesche et al., 2016), or the change 383 in radiative transfer due to soot cores in contrail cirrus ice crystals (Liou et al., 2013), which decreases the 384 albedo at solar wavelengths and increases the top of atmosphere net RF. Both effects are very likely to 385 lead on average to an increase in contrail cirrus RF, causing our best estimate to be conservative. The 386 estimated uncertainty relates to the average contrail cirrus RF. In specific synoptic situations, 387 uncertainties may be much larger and correlated with each other. 388 In contrast to other aviation forcing terms, the average ERF/RF ratio for contrail cirrus is estimated to be 389 0.42, much less than unity. The associated uncertainty is thought to be very large and dependent on 390 391 prevailing aviation traffic and its geographic distribution. The low ERF/RF value is largely due to the reduction in natural cloudiness caused by increased contrail cirrus similar to the reduction in natural cirrus 392 cloudiness as reported by Burkhardt and Kärcher (2011). The ERF/RF value is the average of three global 393 climate model studies: two that estimated climate efficacies of 31% and 59% (Ponater et al., 2005; Rap et 394 al., 2010) and a third that gave a direct estimate of the ERF of contrail cirrus that is 35% of the 395 corresponding RF (Bickel et al., 2019). These studies conclude that efficacies equal to that of CO2 396 overstate the role of cirrus changes due to aviation on global mean surface temperatures. The average 397 ERF/RF ratio was applied to the homogenized estimates of RF, while the RF of Chen and Gettelman 398 (2013) was interpreted as an ERF (see above). Weighting each study equally, the resulting ERF for 399 contrail cirrus is 57 (17, 98) mW m -2 for 2018. It is important to note that the uncertainty does not include 400 any contribution coming from the ERF/RF estimate. Despite the large ERF/RF adjustment, this ERF term 401 is the largest for global aviation in 2018 and is comparable in magnitude to the CO2 term in the 402 normalized results for 2000 to 2018 (Figure 6). While comparable in magnitude, these ERFs have 403 different implications for future climate change (Section 6). 404 * * * 405 406 427 441 446 4.5. Aerosol-radiation interaction. 2− Aircraft engines directly emit soot, defined as mixture of BC and OC, and precursors for sulfate (SO4) − and nitrate (NO3) aerosol along flight tracks. Soot aerosol is formed from the condensation of unburnt 407 aromatic compounds in the combustor (e.g. Ebbinghaus and Wiesen, 2001) and sulfate aerosol from the 408 oxidation of sulfur in the fuel (Dstan 91-91, 2015). Most of the sulfur is emitted as SO2, whilst a small 409 fraction (~3%) is emitted as oxidized H2SO4 (Petzold et al., 2005). Most of the sulfate aerosol is produced 410 after emission from sulfur precursor compounds by oxidation in the ambient atmosphere. Both aerosol 411 types create RFs from aerosol-radiation interactions: soot absorbs short-wave radiation leading to net 412 warming and sulfate aerosol scatters incoming short-wave radiation leading to net cooling (IPCC, 1999). 413 As figures of merit, year 2000 global aviation emissions increase aerosol mass for both soot and sulfate 414 by a few percent and aerosol number by 10–30% near air traffic flight corridors in the northern 415 extratropics (Righi et al., 2013). 416 \\mathrm{(S O\_{4}^{2-})} {\\mathrm{S O}}\_{2}. \ \\mathrm\ O\_{3}^{-}) \\mathrm{H\_{2}S O\_{4}} Past calculations of aerosol-radiation RF values using a variety of global aerosol models have yielded 417 -2 values of a few mW m and with large uncertainties (e.g., Righi et al., 2013; Gettelman and Chen, 2013; 418 L09). In the present study, 10 estimates across 8 models were used to evaluate soot and sulfate aerosol 419 normalized RFs (IPCC, 1999; Sausen et al., 2005; Fuglestvedt et al., 2008; Balkanski et al., 2010; 420 Gettelmann and Chen, 2013; Unger et al., 2013; Pitari et al., 2015; Brasseur et al., 2016) (see SD 421 spreadsheet). Averaging the normalized values yields a 2018 best estimate of the soot aerosol-radiation 422 -2 RF of 0.9 (0.1, 4.0) mW m for 0.0093 Tg soot emitted. The corresponding best estimate for sulfate 423 -2 aerosol is -7.4 (-19, -3) mW m for 0.37 Tg SO2 emitted. The uncertainties are derived from the standard 424 deviation of the model values. The ERF/RF ratios for soot and sulfate are assumed to be unity in the 425 absence of any estimates of this ratio. 426 \\mathrm{m}^{-2} \\mathrm{m}^{-2} \\mathrm{S O}\_{2} -7.4,(-19,\ ),\\mathrm{m W},\\mathfrak{m}^{\ 2} 4.6 Aerosol-cloud interaction. Aerosol-cloud interactions are those processes by which aerosols influence cloud formation. For example, 428 cloud droplets and ice crystals nucleate on aerosol particles. Thus, aerosol-cloud interactions involving 429 aviation aerosol potentially result in an ERF. Aviation soot and sulfate particles are the predominant 430 primary and secondary aerosol from aircraft. The uncertainties in evaluating the aerosol-cloud 431 interactions of aviation soot and sulfate preclude best estimates of ERF contributions. Given the potential 432 importance of these ERF terms, placeholders are included in Figure 3. Furthermore, to promote progress 433 towards future best estimates, the results of relevant modeling studies were compiled and normalized to 434 global aviation fuel usages in 2005, 2011, 2018, to a soot emission index, and to a fuel S content of 600 435 pm (except in the cases of low fuel-S content tests) (see Figure 5 and spreadsheet). As noted in the 436 caption of Figure 5, some earlier wide-ranging values for the soot aerosol-cloud interaction have been 437 superseded by a more recent study (Penner et al., 2018). 438 Aviation sulfate aerosol primarily affects liquid clouds in the background atmosphere. Sulfate aerosol is 440 very efficient as a cloud condensation nuclei (CCN) for liquid clouds, and for promoting homogeneous freezing of solution particles at cold temperatures, thus nucleating ice clouds. Two integrated model 442 simulations (Kapadia et al., 2016; Gettelman and Chen, 2013) found large impacts on liquid clouds from 443 aviation sulfate aerosol that is transported to liquid clouds at lower altitudes over oceans, which have low 444 -2 albedo. The reported RF values in these studies, when scaled appropriately, are -37 to -76 mW m in 445 2018, excluding a low fuel-sulfur case. Note that the study of Righi et al. (2013) that yields an RF of -213 mW m2 in 2018 includes sulfate aerosol-cloud interactions but cannot be directly compared with Kapadia 447 et al. (2016) and Gettelman and Chen (2013), since the former treats the combined effects of sulfate, 448 nitrate and particulate organic matter (POM) rather than isolating the effects of sulfate as done in the 449 latter studies. While these RF estimates do not support a best estimate at present, they do suggest that the 450 sign of the sulfate aerosol-cloud effect on low-level clouds is likely to be negative (i.e., a cooling), similar 451 -76,\\mathrm{m W\ }m^{2} 4.6.1 Sulfate aerosol. 439 \\mathrm{m W},\\mathrm{m}^{2} * * * 24 August 2020 Revised to the ERF for the aerosol-cloud interactions of other anthropogenic sources of sulfate aerosol (IPCC, 452 2013). 453 Sulfate aerosol-cloud interaction forcing estimates are highly dependent on the sensitivity (or 454 susceptibility) of the cloud radiative field to aerosol perturbations, which is dependent on uncertain model 455 processes and the model background aerosol state. Clouds that form with small CCN number 456 concentrations in the background atmosphere are more sensitive to CCN perturbations. Forcing by these 457 cloud effects are largely concentrated near flight corridors over oceans because the high albedo contrast 458 between the ocean surface and clouds increases forcing sensitivity to CCN perturbations. 459 A large uncertainty was also reported for the magnitude of the aerosol-cloud ERF from all anthropogenic 460 activities, estimated for 2011 to be -450 (-1200, 0.0) mW m -2 (Myhre et al., 2013). A more recent estimate 461 of the aerosol-cloud RF from all anthropogenic activities has a 68% confidence interval of -650 to -1600 462 mW m -2 (Bellouin et al., 2019). In general, aerosol-cloud interactions contribute the largest uncertainty in 463 calculations of anthropogenic ERF (IPCC, 2013). 464 _4.6.2 Soot. 465_ The magnitude and the sign of the global RF from aviation soot effects on background cloudiness remain 466 highly uncertain. The uncertainties center on the difficulties in accurately simulating homogeneous and 467 heterogeneous ice nucleation in the background atmosphere, variations in the treatment of updraft 468 velocities during cirrus formation, and the lack of knowledge of the ice nucleating (IN) ability of aviation 469 soot particles during their atmospheric lifetime (Zhou and Penner, 2014; Penner et al., 2018). 470 471 Two studies find moderate effects of soot aerosol on ice clouds, depending on the ice nucleating efficiency and the size distribution. RF values of about 11–13 mW m -2 (normalized to 2018 emissions) 472 are calculated in some studies for moderate ice-nucleating efficiencies (Pitari et al., 2015, Gettelman and 473 Chen, 2013). 474 In sensitivity tests, if soot processed within contrails is assumed to be an efficient IN particle, then the RF 475 may be negative by up to -330 mW m -2 due to reductions in ice crystal number in regions dominated by 476 homogeneous freezing (Penner et al., 2018; see Figure 5). The RF could be significantly smaller (less 477 negative) if additional ice-forming particles, such as secondary organic aerosol (SOA), are already present 478 479 in the background atmosphere (Penner et al., 2018; Gettelman and Chen, 2013). In addition, increases in ice crystal numbers occur when the background atmosphere has much lower sulfate or haze-forming 480 aerosol number concentrations and is dominated by heterogeneous freezing, causing forcings near zero or 481 even positive (Zhou and Penner, 2014). Other studies predict decreases in cirrus number for smaller 482 numbers of larger soot particles (Hendricks et al., 2011), resulting in a slight warming (Gettelman and 483 Chen, 2013). 484 A dominant uncertainty for the aerosol-cloud effect from soot is the IN properties of aviation soot aerosol. 485 Some laboratory studies indicate soot particles are not efficient ice nuclei (DeMott et al., 1999), while 486 other studies indicate higher efficiencies (Möhler et al., 2005; Hoose and Möhler, 2012). The possibility 487 488 that contrail-processed soot particles would show enhanced IN activity after sublimation in the 489 background atmosphere was addressed in the laboratory (Mahrt et al., 2020). The effect was limited to large soot particles, suggesting that the impact of aviation soot on cloudiness may be overestimated in 490 previous studies that assume soot processed through contrails and not covered by a sulfate coating is an 491 efficient IN (Penner et al., 2018). 492 Another source of uncertainty is soot number concentrations. For individual engines, the soot number can 493 vary by two orders of magnitude (Agarwal et al., 2019). Soot number concentrations from aviation vary 494 with the assumed size of the particles emitted as well as the mass emissions. Soot emissions from aircraft 495 are set as a regulatory parameter for the landing/take-off (LTO) cycle by ICAO and are measured in terms 496 of mass. Robust conversion factors from mass to number have recently been developed for the ICAO-497 * * * 24 August 2020 Revised LTO cycle (Agarwal et al., 2019) but have not yet been made for cruise, although other methodologies 498 exist (Teoh et al., 2019). 499 500 **5\. Calculated net aviation ERF and RF values** ERF and RF values for the terms associated with global aviation emissions and cloudiness are given in 501 **Tables 2 and 3, respectively, for the years 2018, 2011, and 2005, along with uncertainties, sensitivities to 502** emissions and the ERF/RF ratio for selected terms. ERF values are shown for all years in Figure 6. All 503 ERF and RF values are available in the analysis spreadsheet (SD). Through normalization and scaling, all 504 2000 to 2018 values are self-consistent. The sensitivity of each term to emission magnitudes or flight 505 track distances is derived in the normalization process. ERF best estimates and uncertainties (95% 506 confidence limits) are highlighted for year 2018 in Figure 3 along with their assessed confidence levels. 507 No best estimates are included for sulfate and soot aerosol-cloud interactions because of the substantial 508 uncertainties noted above. However, placeholder spaces are included in both the Tables 2 and 3 and 509 510 **Figure 3 to indicate the potential importance of these terms and to flag the associated knowledge gaps for** consideration in future research and assessment activities. The confidence levels and their justifications 511 shown in Figure 3 are obtained by employing the methodology of Mastrandrea et al. (2011), which is 512 based on evidence and agreement in accordance with IPCC guidance (Table 4). 513 In Figure 3, contrail cirrus formation yields the largest positive (warming) ERF term, followed by CO2 514 and NOx emissions. For the 1940 to 2018 period, the net aviation ERF is +100.9 mW m -2 (5–95% 515 likelihood range of (55, 145)) with major contributions from contrail cirrus (57.4 mW m -2 ), CO2 (34.3 516 mW m -2 ), and NOx (17.5 mW m -2 ). The aerosol and water vapor terms represent minor contributions. The 517 formation and emission of sulfate aerosol yields the only significant negative (cooling) term. Non-CO2 518 terms sum to yield a positive (warming) ERF that accounts for 66% of the aviation net ERF in 2018 (66.6 519 (21, 111) mW m -2 ). The application of ERF/RF ratios more than halves the RF value of contrail cirrus 520 while approximately doubling the NOx value. ERF/RF ratios were not included in the L09 analysis. 521 Uncertainty distributions (5%, 95%) show that non-CO2 forcing terms contribute about 8 times more than 522 CO2 to the uncertainty in the aviation net ERF in 2018. The best estimates of the ERFs from aviation 523 aerosol-cloud interactions remain undetermined. 524 The time series of ERF values for individual terms is shown in Figure 6 for the 2000–2018 period. 525 Through normalization and scaling the terms are self-consistent over this period. The increase in all of the 526 terms with time is consistent with the growth of aviation fuel burn and CO2 emissions over the same 527 period (Figure 2). Note that net ERF values shown for each year are not linear sums over the component 528 terms due to the separate probability distributions associated with each component term in the sum, and 529 instead are calculated with a Monte Carlo sampling method described below. 530 A comparison of updated RF estimates with L09 values for 2005 is given in Table 3. The large increase 531 in the contrail cirrus RF between 2005 and 2018 results in part because the 2005 value only includes 532 linear contrails. In L09, only an estimate of 2005 contrail cirrus was provided rather than a best estimate. 533 The present study now includes a process-based model estimate of the contrail cirrus term (Section 4.4). 534 The NOx treatment in L09 did not include the negative forcing contributions of the long-term O3 decrease 535 or the SWV decrease, the updated treatment of CH4 of Etminan et al. (2016), nor an equilibrium-to-536 transient correction. As a result, the updated RF values for NOx are approximately a factor of 2 smaller. 537 Incorporating all the updated information in the RF calculations of the NOx and contrail cirrus terms 538 yields an approximately 30% increase in the net aviation RF for 2005, from 78.0 to 95.2 mW m -2 . In the 539 ERF evaluation for 2005 the net aviation forcing is reduced from 95.2 to 66.9 mW m -2 because the 540 ERF/RF ratios for NOx and contrail cirrus are different than unity. 541 In seeking comparison of net aviation ERF with net anthropogenic ERF, we note that IPCC (Myhre et al., 542 2013. provides a value for 1750–2011 of 2290 (1130, 3330) mW m -2 . The percentage contributions of 543 aviation to the net ERF in 2011 are 3.5% (4.0, 3.4%) and 1.59% (1.65, 1.56%) for the sum of all terms 544 * * * 24 August 2020 Revised and the CO2 term alone, respectively. The 2005 and 2018 percentages are likely the same because the 545 fraction of aviation CO2 emissions of total anthropogenic CO2 emissions has averaged 2.1% (±0.15) for 546 the last two decades (see Figure 2). Normalized relative probabilities of CO2 and non-CO2 ERFs for 2018 547 as derived from the Monte Carlo simulations show that non-CO2 uncertainties are the predominant 548 contribution to the uncertainty in the aviation net ERF (Figure 7). IPCC also separately estimated the 549 contrail cirrus term for 2011 as 50 (20, 150) mW m -2 as discussed above, which compares well with the 550 updated value of 44.1 (13, 75) mW m -2 . 551 The determination of net aviation ERFs and their uncertainties shown in Figure 3 and accompanying 552 tables required a Monte Carlo approach to summing over terms with discrete probability distributions. A 553 similar method was employed in L09. PDFs of each term were constructed from the respective individual 554 studies as normal, lognormal or discrete distributions (see SD spreadsheet). Monte Carlo samplings (one 555 million random points) of the individual forcing PDFs were then used to combine terms to yield net ERFs 556 557 and the uncertainties (95% likelihood range) for the sum of all terms and for only non-CO2 terms (Figure **7). The forcing terms are generally assumed to be independent (uncorrelated) with the notable exception 558** 559 of the NOx component terms which have strong paired correlations as shown in Appendix Figure D.1. Only the short-term O3 and CH4 terms were included in L09 and a 100% correlation was assumed, in part, 560 because the assumption of uncorrelated effects was deemed less acceptable. A subsequent study showed 561 that these terms are indeed strongly correlated (R2 = 0.79) (Holmes et al., 2011), similar to the present 562 results in Appendix Figure D.1. The Holmes et al. (2011) study further concluded that the assumption of 563 100% correlation in this case would lead to an underestimate of uncertainty in the NOx RF. Another 564 565 correlation of forcing terms not considered here may be the dependence of the soot direct effect and contrail properties on the soot number index since ice nucleation at the time of contrail formation depends 566 on the soot number index (e.g., Kärcher, 2018). 567 ## 568 6. Emission equivalency metrics Using the best estimate ERFs, we calculate updated aviation-specific Global Warming Potential (GWP) 569 and Global Temperature change Potential (GTP) values, presented for 20-, 50-, and 100-year time 570 horizons in Table 5. These metrics assign so-called ‘CO2-emission equivalences’ for non-CO2 emissions 571 via ratios of time-integrated ERF and changes in future temperatures, respectively. The choice of metric 572 depends upon the particular underlying application (Fuglestvedt et al., 2010) such that there is no 573 574 uniquely ‘correct’ metric or time horizon, and alternative metrics are available. GWP and GTP are the most commonly applied metrics and the values calculated here allow a comparison with previous 575 estimations (e.g., Lee et al., 2010; Lund et al. 2017). In calculating the GWPs and GTPs, the CO2 IRF 576 from Joos et al. (2013) is used and the climate response IRF from Boucher and Reddy (2008) for the 577 GTPs (see Appendix F for futher details about the metrics calculations). 578 GWPs and GTPs for contrail cirrus and for water vapor reported here are similar to, albeit slightly smaller 579 than, corresponding results previously reported, while soot and sulfate numbers are larger in magnitude 580 (positive and negative) than previous estimates (Fuglestvedt et al. 2010; Lund et al. 2017). The 581 582 Fuglestvedt et al. (2010) estimates for soot are based on RF due to soot emissions from all sources, not just aviation, which yields a lower radiative efficiency (i.e., forcing per unit emission) than in the present 583 study. Also given in Table 5 are CO2-equivalent aviation emissions, along with ratios of total CO2-584 585 equivalent emissions to CO2 emissions. Such ratios are sometimes used as ‘multipliers’ to illustrate the additional climate impact from aviation non-CO2 terms over those from CO2 emissions alone. Here, 586 estimated multipliers for 2018 range from 1.0 to 4.0 depending on the choice of time horizon and 587 emission metric. This is broadly consistent with what has been reported and used previously (Lee et al., 588 589 2010). The broad range emphasizes the challenges associated with developing comparisons of emission equivalences for short- and long-lived climate forcers within a common framework and how such 590 considerations strongly depend on the chosen perspective. 591 * * * 604 One of the significant uncertainties in calculating GWPs and GTPs is the treatment of climate-carbon (C-592 cycle) feedbacks in the modeling framework. The efficiency of carbon sinks reduces with increasing 593 warming (Ciais et al., 2013) and this climate feedback is implicitly included in the Absolute GWP of CO2 594 through the IRF used (Joos et al., 2013). However, Myhre et al. (2013) highlighted that this introduces an 595 inconsistency since the numerators for the GWP and GTP do not include such a climate carbon feedback. 596 One of the studies that have proposed ways of addressing this inconsistency is Gasser et al. (2017). They 597 show that when the C-cycle feedback is consistently accounted for, the non-CO2 emission metrics 598 increase, but less so than initially suggested by Myhre et al. (2013). They also find that removing the C-599 cycle feedback from both numerator and denominator give similar metric values as including it in both 600 places. Using the CO2 IRF without the C-cycle feedback provided by Gasser et al. (2017), we calculate a 601 second set of aviation emission metrics (Table F.1), showing that the changes to the GWP100 and 602 GTP100 values from those given in Table 5 are rather small. 603 \\mathrm{C O\_{2}} \ mathrm-\\mathrm{{C C}}\_{2} \\mathrm{C O\_{2}} In response to the challenges related to comparing short-lived and long-lived forcing components, a number of new ‘flow-based’ methods have been introduced representing both short-lived and long-lived 605 climate forcers explicitly as ‘warming-equivalent’ emissions that have approximately the same impact on 606 the global average surface temperature over multi-decade to century timescales (Lauder et al., 2012; Allen 607 et al., 2016; 2018; Cain et al., 2019; Collins et al., 2019). A simple version of these methods, known as 608 _CO2e_ _GWP_, defines the average annual rate of CO2-warming-equivalent emissions (E) over a period of ∆𝑡 609 years arising from a particular component of RF or ERF by (Cain et al., 2019): 610 ∗ {E}{}\_{C O2e}^{\*} \\mathrm{C O\_{2}} \\mathrm{G W P\*} \\Delta t E\_{C O2e}^{\*}=\ \[\[\\mathbf{1}-\\alpha)H/\\mathbf{A G W}\\mathbf{P} _{H}\],\\Delta F,/\\Delta\ +\\,\[\\alpha/\\mathbf{A G W P_\_{H H}}\]F,\ \ (1) 611\ \ where ∆𝐹 is the ERF change and 𝐹̅ the average ERF arising from that component over that period, 612\ -2-1\ AGWP𝐻 is the Absolute GWP of CO2 (Wm kg year) over time-horizon 𝐻 and 𝛼 is a small coefficient 613\ depending on the previous history of that RF component. This equation gives the rate of CO2 emission 614\ that would, alone, create the same rate of global temperature increase as the combined effect of aviation 615\ climate forcings. For historically small and/or rapidly changing RF components, 𝛼 may be neglected, and 616\ ∗\ hence to a good approximation, total CO2-warming-equivalent emissions over this period (∆𝑡𝐸𝐶𝑂2𝑒) are 617\ approximated by an increase in forcing, ∆𝐹, multiplied by 𝐻⁄AGWP𝐻(see Appendix F), which is about 618\ 1000 GtCO2 per W/m2 for 𝐻 in the range 20 to 100 years (Myhre et al, 2013; IPCC, 2018, Figure SPM.1, 619\ caption). This result follows from the definition of AGWP: since all GWP calculations assume a 620\ linearization, the AGWP𝐻 is equivalent to the forcing change resulting from the emission of 𝐻 tonnes of 621\ CO2 spread over 𝐻 years (Shine et al, 2005), so AGWP𝐻⁄𝐻 is the forcing change per tonne of CO2. Under 622\ the historical profile of increasing global annual aviation-related emissions and associated ERFs, CO2-623\ warming-equivalent emissions based on GWP\* indicate that aviation emissions are currently warming the 624\ climate around three times faster than that associated with aviation CO2 emissions alone (Table 5). 625\ \ \\bar{F}\ \ \\Delta F\ \ \\mathrm{C O\_{2},(W m^{2},k g^{-1},y e a r}\ \ \\mathrm{C O\_{2}}\ \ \\mathsf{A G W P}\_{H}\ \ \\left(\\Delta t E\_{C02e}^{\*}\\right)\ \ \ {/\ }\\mathtt{A G W P}\_{H},(\ \ \\Delta F\ \ 1000:\\mathrm{G t C O\_{2}}\ \ \\mathsf{A G W P}\_{H}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ It is important to note that, unlike the conventional GWP and GTP metrics given in Table 5, the ratio 626\ between total CO2-warming-equivalent emissions from all forcing agents and those from CO2 alone will 627\ change substantially if future aviation emissions deviate from their current growth trajectory (calculated 628\ here over the period 2000–2018). If annual global aviation emissions were to stabilize, this ratio declines 629\ towards unity, as ∆𝐹⁄∆𝑡 would decline to zero. This does not indicate, however, that the non-CO2 effects 630\ do not have a warming affect. This human-induced warming still represents a mitigation potential. 631\ Warming-equivalent emissions capture the fact that constant emission of short-lived climate forcers 632\ maintain an approximately constant level of warming, whilst constant emissions of long-lived climate 633\ forcers, such as CO2, continue to accumulate in the atmosphere resulting in a constantly increasing level 634\ of associated warming. Hence warming-equivalent emissions show that the widely-used assumption of a 635\ constant ‘multiplier’, assuming that net warming due to aviation is a constant ratio of warming due to 636\ aviation CO2 emissions alone, only applies in a situation in which aviation emissions are rising 637\ exponentially such that the rate of change of non-CO2 RF is approximately proportional to the rate of CO2 638\ emissions (assuming non-CO2 RF is proportional to CO2 emissions, and noting that the rate of change any 639\ \ \\mathrm{G W P\*}\ \ {\\mathrm O\_{2}}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{n o n-C O\_{2}\ R F}\ \ \\mathrm{\\it O}}}\ \ \\mathrm{O}}\_{2}\ \ * * *\ \ 24 August 2020 Revised\ \ quantity is proportional to that quantity only when both are growing exponentially). In contrast, under a 640 future hypothetical trajectory of decreasing aviation emissions, this GWP\* based multiplier could fall 641 below unity, as a steadily falling rate of emission of (positive) short-lived climate forcers has the same 642 643 effect on global temperature as active removal of CO2 from the atmosphere. The GWP\* based ‘multiplier’ calculated here (which depends on the ratio of the increase in net aviation warming to the increase in 644 warming due to aviation CO2 emissions alone over the recent past), should not be applied to future 645 scenarios that deviate substantially from the current trend of increasing aviation-related emissions. The 646 broad range of values for a ‘multiplier’ presented here is an illustration of the limitations of using a 647 constant multiplier in the assessment of climate impacts of aviation, and a reminder that the choice of 648 metric for such a multipler involves subjective choices. 649\ \ 650 **7\. Aviation CO2 vs non-CO2 forcings**\ \ Since IPCC (1999), the comparison of aviation CO2 RF with the non-CO2 RFs has been a major scientific 651 topic, as well as a discussion point amongst policy makers and civil society (ICAO, 2019). Aviation as a 652 sector is not unique in having significant non-CO2 forcings; the same is true of agriculture with significant 653 CH4 and N2O emissions, or maritime shipping with net-negative current-day RF despite CO2 emissions of 654 a similar magnitude to those from aviation (Fuglesvedt et al., 2009). However, unlike direct emissions of 655 the greenhouse gases N2O and CH4 from the agricultural sector, aviation non-CO2 forcings are not 656 covered by the former Kyoto Protocol. It is unclear whether future developments of the Paris Agreement 657 or ICAO negotiations to mitigate climate change, in general, will include short-lived indirect greenhouse 658 gases like NOx and CO, aerosol-cloud effects, or other aviation non-CO2 effects. Aviation is not 659 mentioned explicitly in the text of the Paris Agreement, but according to its Article 4, total global 660 greenhouse-gas emissions need to be reduced rapidly to achieve a balance between anthropogenic 661 662 emissions by sources and removals by sinks of greenhouse gases in the second half of this century.\ \ 663 The IPCC concludes: “Reaching and sustaining net-zero global anthropogenic CO2 _emissions and_ 664 _declining net non-CO2 radiative forcing would halt anthropogenic global warming on multi-decadal time_ 665 _scales.” (IPCC, 2018, bullet A2.2, SPM). Crucially, both conditions would need to be met to halt global_ warming. Hence, to halt aviation’s contribution to global warming, the aviation sector would need to 666 achieve net-zero CO2 emissions and declining non-CO2 radiative forcing (unless balanced by net negative 667 emissions from another sector): neither condition is sufficient alone. Some combination of reductions in 668 CO2 emissions and non-CO2 forcings might halt further warming temporarily, but only for a few years: it 669 would not be possible to offset continued warming from CO2 by varying non-CO2 radiative forcing, or 670 _vice versa, over multi-decade timescales. 671_\ \ That aviation’s non-CO2 forcings are not included in global climate policy has resulted in studies as to 672 whether they could be incorporated into existing policies, such as the European Emissions Trading 673 Scheme, using an appropriate overall emissions ‘multiplier’; however, scientific uncertainty has so far 674 675 precluded this (Faber et al., 2008). In addition, as noted above, the multiplier is highly dependent on the future emissions scenario (Section 6). Alternatively, proposals have been made to reduce aviation’s non-676 CO2 forcings by, for example, avoiding contrail formation by re-routing aircraft (Matthes et al., 2017), or 677 optimizing flight times to avoid the more positive (warming) fractional forcings (e.g., by avoiding night 678 flights, Stuber et al., 2006). There is a developing body of literature on this topic (e.g., Newinger and 679 Burkhardt, 2012; Yin et al., 2018). Similarly, studies have assessed whether changes in cruise altitudes 680 could mitigate NOx impacts (e.g. Frömming et al., 2012). The potential impacts of changes in technology 681 have also been examined to reduce the non-CO2 forcings such as lowering the emission index for NOx 682 (Freeman et al., 2018) or soot particle number emissions (Moore et al., 2017) to reduce net NOx and 683 contrail cirrus forcings, respectively (Burkhardt et al., 2018). 684\ \ Avoidance of contrail formation through re-routing can incur a fuel penalty and therefore additional CO2 emissions during a flight, and changes in combustor technology to minimize NOx generally increases 686\ \ * * *\ \ 24 August 2020 Revised\ \ marginal fuel burn and CO2 emission. Both methods invoke the usage of climate metrics such as those 687 calculated and presented in Section 6 to evalulate whether there is a net climate benefit or disbenefit over 688 a defined period. In examining such mitigation scenarios involving tradeoffs (e.g. Teoh et al., 2020), the 689 perceived success or otherwise of the outcome will be a function of the user’s choice of metric and time 690 691 horizon. A limitation noted for the GWP is that it has an ‘artificial memory’ over longer time horizons, 692 since the integrated-RF nature of the metric accumulates ‘signal’ over time that the climate system has 693 ‘forgotten’ (Fuglestvedt et al., 2010). The GTP, being an ‘end point’ metric that captures the temperature response, overcomes this limitation of the GWP but is not yet in usage within current climate policy. 694\ \ Changes to aviation operations or technology that result in a reduction of a non-CO2 forcing with the 695 added consequence of increased CO2 emissions can result in net reductions of forcing on short timescales 696 while increasing the net forcing on longer timescales (e.g., Freeman et al., 2018). In a case study of 697 contrail avoidance through routing changes, Teoh et al. (2019) found that the resultant small increase in 698 699 CO2 emissions still reduces the net forcing over a timescale of 100 years. In such ‘tradeoff cases’ the balance between non-CO2 and CO2 forcings have to be weighted carefully, since CO2 accumulates in the 700 atmosphere and a fraction has millennial timescales (Archer and Brovkin, 2008; IPCC, 2007). Prior to the 701 COVID-19 pandemic, global aviation traffic and emissions were projected to grow to 2050 (Fleming and 702 703 Lepinay, 2019). As the COVID-19 pandemic diminishes, aviation traffic is likely to recover to meet projected rates on varying timescales (IATA, 2020), with continued growth further increasing CO2 704 emissions. Thus, reducing CO2 aviation emissions will remain a continued focus in reducing future 705 anthropogenic climate change, along with aviation non-CO2 forcings. The latter increase the current-day 706 impact on global average temperatures by a factor of around 3 (using GWP\*) above that due to CO2 707 alone. 708\ \ ## 709 Author Contributions\ \ 710 **D. S. Lee, D. W. Fahey**\ \ Role: Investigation, Methodology, Writing–review & editing, Data curation; Formal analysis, Project 711 administration, Supervision 712\ \ ## 713 A. Skowron\ \ Role: Investigation, Methodology, Writing–review & editing, Data curation, Formal analysis; 714 Software 715\ \ 716 **M. R. Allen, U. Burkhardt, Q. Chen, S. J. Doherty, S. Freeman, P.M. Forster, J. Fuglestvedt, A.** 717 **Gettelman, R. R. De León, L. L. Lim, M. T. Lund, R. J. Millar, B. Owen, J. E. Penner, G. Pitari,** 718 **M. J. Prather, R. Sausen, L. J. Wilcox**\ \ Role: Writing–review & editing, Investigation, Methodology, Writing–original draft, Data curation; 719 Formal analysis; 720\ \ ## 721 Declaration of competing interest\ \ The authors declare that they have no known competing financial interests or personal relationships that 722 could have appeared to influence the work reported in this paper. 723\ \ ## 724 Acknowledgements\ \ We gratefully acknowledge discussions with many colleagues during the preparation of this paper, in 725 particular Andreas Bier and Bernd Kärcher. We acknowledge help with graphical displays from Beth 726 Tully (Figure 1) and Chelsea R. Thompson (Figures 5, 6 and 7). 727\ \ **Funding**\ \ * * *\ \ 24 August 2020 Revised\ \ DSL, AS, RRdL, LL, BO acknowledge support from the UK Department for Transport. PMF 729 acknowledges support of the European Union’s Horizon 2020 Research and Innovation Programme under 730 grant agreement number 820829 (CONSTRAIN) by the UK National Environment Research Council 731 (NERC) SMURPHS project (NE/N006038/1). MRA acknowledges support from the EU H2020 grant 732 agreement number 821205 (FORCeS) and the Oxford Martin Programme on Climate Pollutants. MTL 733 and JSF acknowledges support from the Norwegian Research Council (RCN) grant number 300718 734 (AVIATE), for which DSL and RS have a collaboration agreement. JEP acknowledges support from the 735 National Science Foundation (NSF 1540954). 736\ \ ## 737 Data Availability\ \ Supplementary data to this article is a spreadsheet that can be found online at: [https://doi](https://doi/). org/xxxxx. 738\ \ ## 740 References\ \ Airbus, Global Market Forecast 2017–2036 (Airbus, France 2017). 741\ \ Allen, M. R. J. S. Fuglestvedt, K. P. Shine, A. Reisinger, R. T. Pierrehumbert, and P. M. Forster, New use 742 743 of global warming potentials to compare cumulative and short-lived climate pollutants. Nature Climate _Change 6 (8), 773–776, [https://doi.org/10.1038/nclimate2998](https://doi.org/10.1038/nclimate2998) (2016). 744_\ \ Allen, M. R. K. P. Shine, J. S. Fuglestvedt, R. J. Millar, M. Cain, D. J. Frame, A. H. Macey, A solution to 745 the misrepresentations of CO2-equivalent emissions of short-lived climate pollutants under ambitious 746 mitigation. npj Climate and Atmospheric Science 1:16; [https://doi.org/10.1038/s41612-018-0026-8](https://doi.org/10.1038/s41612-018-0026-8) 747 (2018). 748\ \ Agarwal, A., R. L. Speth, T. M. Fritz, S. D. Jacob, T. Rindlisbacher, R. Iovinelli, B. Owen, R. C. Miake-749 Lye, J. S. Sabnis, S. R. H. Barrett, SCOPE11 method for estimating aircraft black carbon mass and 750 particle number emissions. Environmental Science and Technology 53, 1364–1373, 751 [https://doi.org/10.1021/acs.est.8b04060](https://doi.org/10.1021/acs.est.8b04060), (2019). 752\ \ Alfsen, K. H. and T. Berntsen, T., An Efficient and Accurate Carbon Cycle Model for Use in Simple 753 Climate Models. CICERO, Oslo, Norway, [https://core.ac.uk/reader/52082516](https://core.ac.uk/reader/52082516) 754\ \ Archer, D. and V. Brovkin, The millennial atmospheric lifetime of anthropogenic CO2. Climatic Change 755 90, 283–297, [https://doi.org/10.1007/s10584-008-9413-1](https://doi.org/10.1007/s10584-008-9413-1) (2008). 756\ \ Balkanski, Y., G. Myhre, M. Gauss, G. Rädel, E. J. Highwood, K. P. Shine, Direct radiative effect of 757 aerosols emitted by transport: from road, shipping and aviation. Atmospheric Chemistry and Physics 758 10(10), 4477-4489, [https://doi.org/10.5194/acp-10-4477-2010](https://doi.org/10.5194/acp-10-4477-2010) (2010). 759\ \ Barrett, S., M. Prather, J. Penner, H. Selkirk, S. Balasubramanian, A. Dopelheuer, G. Fleming, M. Gupta, 760\ \ R. Halthore, J. Hileman, M. Jacobson, S. Kuhn, S. Lukachko, R. Miake-Lye, A. Petzold, C. Roof, M. 761 Schaefer, U. Schumann, I. Waitz, R. Wayson R., Guidance on the use of AEDT gridded aircraft emissions 762 in atmospheric models. Massachusetts Institute for Technology, Laboratory for Aviation and the 763 Environment, LAE-2010-008-N. (2010) 764 [http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.719.2090&rep=rep1&type=pdf](http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.719.2090&rep=rep1&type=pdf) 765 Bellouin, N., J. Quaas, E. Gryspeerdt, S. Kinne, P. Stier, D. Watson‐Parris, O. Boucher, K.S. Carslaw, M. 766 Christensen, A.‐L. Daniau, J.‐L. Dufresne, G. Feingold, Bounding global aerosol radiative forcing of 767 climate change. Reviews of Geophysics 58, e2019RG000660, [https://doi.org/10.1029/2019RG000660](https://doi.org/10.1029/2019RG000660) 768 (2019). 769 Bickel, M., M. Ponater, L. Bock, U. Burkhardt, S Reineke, Estimating the effective radiative forcing of 770 contrail cirrus, Journal of Climate, 33, 1991-2005, [https://doi.org/10.1175/JCLI-D-19-0467.1](https://doi.org/10.1175/JCLI-D-19-0467.1) (2020). 771\ \ * * *\ \ 24 August 2020 Revised\ \ Bier, A., U. Burkhardt, L. Bock, Synoptic control of contrail cirrus life cycles and their modification due 772 to reduced soot number emissions. Journal of Geophysical Research Atmospheres 122 (21), 11,584-773 11,603 [https://doi.org/10.1002/2017JD027011](https://doi.org/10.1002/2017JD027011) (2017). 774\ \ Bier, A. and A. U. Burkhardt, Variability in contrail ice nucleation and its dependence on soot number 775 emissions. Journal of Geophysical Research Atmospheres 124, 3384–3400, 776 [https://doi.org/10.1029/2018JD029155](https://doi.org/10.1029/2018JD029155) (2019). 777\ \ Bock, L. and U. Burkhardt, Reassessing properties and radiative forcing of contrail cirrus using a climate 778 model. Journal of Geophysical Research Atmospheres 121, 9717–9736, 779 [https://doi.org/10.1002/2016JD025112](https://doi.org/10.1002/2016JD025112) (2016). 780\ \ Boeing, Orders and Deliveries for January 2018, [http://www.boeing.com/commercial/#/orders-deliveries](http://www.boeing.com/commercial/#/orders-deliveries) 781 (2018). 782\ \ Boucher, O. and M. S. Reddy, Climate trade-off between black carbon and carbon dioxide emissions. 783 _Energy Policy 36, 193–200, [https://doi.org/10.1016/j.enpol.2007.08.039](https://doi.org/10.1016/j.enpol.2007.08.039) (2008). 784_\ \ Boucher, O., D. Randall, P. Artaxo, C. Bretherton, G. Feingold, P. Forster, V.-M. Kerminen, Y. Kondo, 785\ \ H. Liao, U. Lohmann, P. Rasch, S.K. Satheesh, S. Sherwood, B. Stevens, and X.Y. Zhang, 2013: Clouds 786\ 787 and aerosols. 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. T.F. Stocker, D. Qin, G.-788_\ \ K. Plattner, M. Tignor, S.K. Allen, J. Doschung, A. Nauels, Y. Xia, V. Bex, and P.M. Midgley, Eds. 789 Cambridge University Press, pp. 571-657, doi:10.1017/CBO9781107415324.016. 790 Brasseur, G. P., M. Gupta, B. E. Anderson, S. Balasubramanian, S. Barrett, D. Duda, G. Fleming, P. M. 791\ 792 Forster, J. Fuglestvedt, et al., Impact of Aviation on Climate: FAA’s Aviation Climate Change Research Initiative (ACCRI) Phase II. Bulletin of the American Meteorological Society 97, 561–583, 793 [https://doi.org/10.1175/BAMS-D-13-00089.1](https://doi.org/10.1175/BAMS-D-13-00089.1) (2016). 794\ \ 795 Burkhardt, U., B. Kärcher, U. Schumann, Global Modelling of the contrail and contrail cirrus climate impact. Bulletin of the American Meteorological Society 91, 479-484, 796 [https://doi.org/10.1175/2009BAMS2656.1](https://doi.org/10.1175/2009BAMS2656.1) (2010). 797\ \ Burkhardt, U. and B. Kärcher, Global radiative forcing from contrail cirrus. Nature Climate Change 1, 798 54–58, [https://doi.org/10.1038/nclimate1068](https://doi.org/10.1038/nclimate1068) (2011). 799\ \ Burkhardt, U., L. Bock, A. Bier, Mitigating the contrail cirrus climate impact by reducing aircraft soot 800 number emissions. npj Climate and Atmospheric Science 1:37, [https://doi.org/10.1038/s41612-018-0046-801](https://doi.org/10.1038/s41612-018-0046-801) 4 (2018). 802\ \ Cain, M., J. Lynch, M. R. Allen, J. S. Fuglestvedt, D. J. Frame, A. H. Macey, Improved calculation of 803 warming-equivalent emissions for short-lived climate pollutants. npj Climate and Atmospheric Science, 804 2:29, [https://doi.org/10.1038/s41612-019-0086-4](https://doi.org/10.1038/s41612-019-0086-4) (2019). 805\ \ Carlin, B., Q. Fu, U. Lohmann, G. Mace, K. Sassen, J. Comstock, High-cloud horizontal inhomogeneity 806 and solar albedo bias. Journal of Climate 15, 2321–2339, [https://doi.org/10.1175/1520-807](https://doi.org/10.1175/1520-807) 0442(2002)015<2321:HCHIAS>2.0.CO;2 (2002). 808\ \ Chen, C.-C., A. Gettelman, C. Craig, P. Minnis, D. P. Duda, Global contrail coverage simulated by 809 810 CAM5 with the inventory of 2006 global aircraft emissions. Journal of Advances in Modeling Earth _Systems 4, 04003. [https://doi.org/10.1029/2011MS000105](https://doi.org/10.1029/2011MS000105) (2012). 811_\ \ 812 Chen, C.-C. and A. Gettelman, Simulated radiative forcing from contrails and contrail cirrus. Atmospheric _Chemistry and Physics, 13, 12525–12536, [https://doi.org/10.5194/acp-13-12525-2013](https://doi.org/10.5194/acp-13-12525-2013) (2013). 813_\ \ * * *\ \ 24 August 2020 Revised\ \ Ciais, P., Sabine, C., Bala, G., Bopp, L., Brovkin, V., Canadell, J., Chhabra, A., DeFries, R., Galloway, J., 814 Heimann, M., Jones, C., Le Quéré, C., Myneni, R. B., Piao, S., and Thornton, P. Carbon and Other 815 Biogeochemical Cycles, in: Climate Change 2013: The Physical Science Basis. Contribution of Working 816 Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change, edited by: 817 Stocker, T. F., Qin, D., Plattner, G.-K., Tignor, M., Allen, S. K., Boschung, J., Nauels, A., Xia, Y., Bex, 818\ \ V., and Midgley, P. M., Cambridge University Press, Cambridge, UK and New York, NY, USA (2013) 819\ 820 Clarke, L., J. Edmonds, H. Jacoby, H. Pitcher, J. Reilly, R. Richels, “Scenarios of Greenhouse Gas Emissions and Atmospheric Concentrations”. Sub-report 2.1A of Synthesis and Assessment Product 2.1 821 by the U.S. Climate Change Science Program and the Subcommittee on Global Change Research 822 823 (Department of Energy, Office of Biological & Environmental Research, Washington, 7 DC. 2007) pp. 54, [https://globalchange.mit.edu/sites/default/files/CCSP\_SAP2-1a-FullReport.pdf](https://globalchange.mit.edu/sites/default/files/CCSP_SAP2-1a-FullReport.pdf). 824\ \ Collins, W. J., D. J. Frame, J. S. Fuglestvedt, K. P. Shine, Stable climate metrics for emissions of short 825 and long-lived species – combining steps and pulses. Environmental Research Letters 15(2), 024018, 826 [https://doi.org/10.1088/1748-9326/ab6039](https://doi.org/10.1088/1748-9326/ab6039) (2019). 827\ \ Dalsøren, S. B., C. L. Myhre, G. Myhre, A. J. Gomez-Pelaez, O. A. Søvde, I. S. A. Isaksen, R. F. Weiss, 828\ \ C. M. Harth, Atmospheric methane evolution the last 40 years. Atmospheric Chemistry and Physics 16, 829 3099–3126, [https://doi.org/10.5194/acp-16-3099-2016](https://doi.org/10.5194/acp-16-3099-2016) (2016). 830 DeMott, P. J., Y. Chen, S. M. Kreidenweis, D. C. Rogers, D. E. Sherman, Ice formation by black carbon 831 particles. Geophysical Research Letters 26, 2429–2432, [https://doi.org/10.1029/1999GL900580](https://doi.org/10.1029/1999GL900580) (1999). 832 Derwent, R. G., W. J. Collins, C. E. Johnson, D. S. Stevenson, Transient behaviour of tropospheric ozone 833 precursors in a global 3-D CTM and their indirect greenhouse effects. Climatic Change 49, 463–487, 834 [https://doi.org/10.1023/A:1010648913655](https://doi.org/10.1023/A:1010648913655) (2001). 835\ 836 Dstan 91-91 “Turbine fuel, kerosene type, Jat A-1. Ministry of Defence, Defence Standard 91-91”, Issue 7, Amendment 3. Defence Equipment and Support (UK Defence Standardization, Glasgow, UK, 2015). 837\ \ 838 Ebbinghaus, A. and P. Wiesen, Aircraft fuels and their effects upon engine emissions. Air and Space _Europe 3, 101-103, [https://doi.org/10.1016/S1290-0958(01](https://doi.org/10.1016/S1290-0958(01)) 90026-7 (2001). 839_\ \ Etminan, M., G. Myhre, E. J. Highwood, K. P. Shine, Radiative forcing of carbon dioxide, methane, and 840 nitrous oxide: A significant revision of the methane radiative forcing. Geophysical Research Letters 43, 841 12,614–12,623 [https://doi.org/10.1002/2016GL071930](https://doi.org/10.1002/2016GL071930) (2016). 842\ \ Faber, J., D. Greenwood, D. S. Lee, M. Mann, P. M. de Leon, D. Nelissen, B. Owen, M. Ralph, J. Tilston, 843\ \ A. van Velzen, G. van de Vreede, “Lower NOx at higher altitudes: policies to reduce the climate impact of 844 aviation NOx emissions”. (CE-Delft, 08.7536.32, Delft, The Netherlands, 2008). 845\ 846 Fleming, G. and U. Ziegler, Environmental trends in aviation to 2050. In ‘ICAO Environmental Report, 2016’, International Civil Aviation Organization, Montreal. (2016) [https://www.icao.int/environmental-847](https://www.icao.int/environmental-847) protection/Documents/EnvironmentalReports/2019/ENVReport2019\_pg17-23.pdf 848\ \ 849 Fleming, G. and I. de Lepinay, “Environmental trends in aviation to 2050”, in ICAO Environmental Report, 2019 Destination Green the Next Chapter, (ICAO Montreal, 2019), 850 [https://www.icao.int/environmental-851](https://www.icao.int/environmental-851) protection/Documents/EnvironmentalReports/2019/ENVReport2019\_pg17-23.pdf (2019) 852\ \ Forster, P.M.d.F. and K. P. Shine, Radiative forcing and temperature trends from stratospheric ozone 853 changes. Journal of Geophysical Research 102, 10841–10855, [https://doi.org/10.1029/96JD03510](https://doi.org/10.1029/96JD03510) 854 (1997). 855\ \ * * *\ \ 24 August 2020 Revised\ \ Forster, C., A. Stohl, P. James, V. Thouret, The residence times of aircraft emissions in the stratosphere 856 857 using a mean emission inventory and emissions along actual flight tracks. Journal of Geophysical _Research Atmospheres 108, 8524, [https://doi.org/10.1029/2002JD002515](https://doi.org/10.1029/2002JD002515) (2003). 858_\ \ Freeman, S., D. S. Lee, L. L. Lim, A. Skowron, R. R. De León, Trading off aircraft fuel burn and NOx 859 emissions for optimal climate policy. Environmental Science and Technology 52, 2498–2505, 860 [https://doi.org/10.1021/acs.est.7b05719](https://doi.org/10.1021/acs.est.7b05719) (2018). 861\ \ Friedlingstein, P., P. Cox, R. Betts, L. Bopp, W. von Bloh, V. Brovkin, P. Cadule, S. Doney, M. Eby, I. 862 Fung, G. Bala, J. John, C. Jones, F. Joos, T. Kato, M. Kawamiya, W. Knorr, K. Lindsay, H. D. Matthews, 863\ \ T. Raddatz, P. Rayner, C. Reick, E. Roeckner, K.-G. Schnitzler, R. Schnur, K. Strassmann, A. J. Weaver, 864\ C. Yoshikawa and N. Zeng, Climate-carbon cycle feedback analysis: Results from the C4MIP model 865 intercomparison, Journal of Climate 19, 3337–3353, [https://doi.org/10.1175/JCLI3800.1](https://doi.org/10.1175/JCLI3800.1) (2006) 866 Frömming, C., M. Ponater, K. Dahlmann, V. Grewe, D. S. Lee, R. Sausen, Aviation-induced radiative 867\ 868 forcing and surface temperature change in dependency of the emission altitude. Journal of Geophysical _Research Atmospheres 117, 9717–9736, [https://doi.org/10.1029/2012JD018204](https://doi.org/10.1029/2012JD018204) (2012). 869_\ \ 870 Fuglestvedt, J. S. and T. Berntsen, “A simple model for scenario studies of changes in climate, Version\ \ 1.0”, (CICERO, Oslo, Norway, 1999) pp. 59, [https://cicero.oslo.no/no/publications/internal/326](https://cicero.oslo.no/no/publications/internal/326). 871 Fuglestvedt, J. S., T. K. Berntsen, I. S. A. Isaksen, H. T. Mao, X. Z. Liang, W. C. Wang, Climatic forcing 872 of nitrogen oxides through changes in tropospheric ozone and methane; global 3D model studies. 873 _Atmospheric Environment 33, 961–977 (1999). 874_ Fuglestvedt, J., T. Berntsen, G. Myhre, K. Rypdal, R. B. Skeie Climate forcing from the transport sectors. 875\ 876 _Proceedings of the National Academy of Sciences U.S.A. 105(2), 454-458,_ [https://doi.org/10.1073/pnas.0702958104](https://doi.org/10.1073/pnas.0702958104) (2008). 877\ \ Fuglestvedt, J. S., T. Berntsen, V. Eyring, I. Isaksen, D. S. Lee, R. Sausen, Shipping emissions: from 878 cooling to warming of climate–and reducing impacts on health. Environmental Science and Technology 879 43, 9057–9062, [https://doi.org/10.1021/es901944r](https://doi.org/10.1021/es901944r) (2009). 880\ \ Fuglestvedt, J. S., K. P. Shine, T. Berntsen, J. Cook, D. S. Lee, A. Stenke, R. B. Skeie, G. J. M. Velders, 881 882 I. A. Waitz, Transport impacts on atmosphere and climate: Metrics. Atmospheric Environment 44, 4648– 4677, [https://doi.org/10.1016/j.atmosenv.2009.04.044](https://doi.org/10.1016/j.atmosenv.2009.04.044) (2010). 883\ \ Gauss, M., I. S. A. Isaksen, S. Wong, W. C. Wang, Impact of H2O emissions from cryoplanes and 884 kerosene aircraft on the atmosphere, Journal of Geophysical Research Atmospheres 108 (D10), 4304, 885 [https://doi.org/10.1029/2002JD002623](https://doi.org/10.1029/2002JD002623) (2003). 886\ \ Gasser, T., G. P. Peters, J. S. Fuglestvedt, W. J. Collins, D. T. Shindell and P. Ciais, Accounting for the 887 888 climate-carbon feedback in emission metrics. Earth Syst. Dynam. 8, 235–253, [https://doi.org/10.5194/esd-](https://doi.org/10.5194/esd-) 889 8-235-2017 (2017).\ \ Gettelman, A. and C. Chen, The climate impact of aviation aerosols. Geophysical Research Letters 40, 890 2785–2789, [https://doi.org/10.1002/grl.50520](https://doi.org/10.1002/grl.50520) (2013). 891\ \ Gounou, A. and R. J. Hogan, A sensitivity study of the effect of horizontal photon transport on the 892 893 radiative forcing of contrails. Journal of the Atmospheric Sciences 64, 1706–1716, [https://doi.org/10.1175/JAS3915.1](https://doi.org/10.1175/JAS3915.1) (2007). 894\ \ Gottschaldt, K., C. Voigt, P. Jöckel, M. Righi, R. Deckert, S. Dietmüller, Global sensitivity of aviation 895 NOx effects to the HNO3-forming channel of the HO2 + NO reaction. Atmospheric Chemistry and Physics 896 13, 3003–3025, [https://doi.org/10.5194/acp-13-3003-2013](https://doi.org/10.5194/acp-13-3003-2013) (2013). 897\ \ * * *\ \ 24 August 2020 Revised\ \ Grewe, V., and A. Stenke, AirClim: an efficient tool for climate evaluation of aircraft technology. 898 _Atmospheric Chemistry and Physics 8, 4621–4639, [https://doi.org/10.5194/acp-8-4621-2008](https://doi.org/10.5194/acp-8-4621-2008) (2008). 899_\ \ 900 Hansen, J., M. Sato, R. Ruedy, Radiative forcing and climate response. Journal of Geophysical Research _Atmospheres 102 (D6), 6831–6864, [https://doi.org/10.1029/96JD03436](https://doi.org/10.1029/96JD03436) (1997). 901_\ \ 902 Hansen, J., and L. Nazarenko Soot climate forcing via snow and ice albedos. Proceedings of the National _Academy of Sciences U.S.A. 101, 423–428, [https://doi.org/10.1073/pnas.2237157100](https://doi.org/10.1073/pnas.2237157100) (2004). 903_\ \ Hansen, J., M. Sato, R. Ruedy, L. Nazarenko, A. Lacis, G. A. Schmidt, G. Russell, I. Aleinov, M. Bauer, 904\ \ S. Bauer, N. Bell, B. Cairns, V. Canuto, M. Chandler, Y. Cheng, A. Del Genio, G. Faluvegi, E. Fleming, 905\ A. Friend, T. Hall, C. Jackman, M. Kelley, N. Kiang, D. Koch, J. Lean, J. Lerner, K. Lo, S. Menon, R. 906 Miller, P. Minnis, T. Novakov, V. Oinas, Ja. Perlwitz, Ju. Perlwitz, D. Rind, A. Romanou, D. Shindell, P. 907 Stone, S. Sun, N. Tausnev, D. Thresher, B. Wielicki, T. Wong, M. Yao, S. Zhang, Efficacy of climate 908 forcings. Journal of Geophysical Research Atmospheres 110, D18104. 909 [https://doi.org/10.1029/2005JD005776](https://doi.org/10.1029/2005JD005776) (2005). 910 Hari, T. K., Z. Yaakob, N. Binitha, Aviation biofuel from renewable resources: routes, opportunities and 911 challenges. Renewable and Sustainable Energy Reviews 42, 1234–1244 912 [https://doi.org/10.1016/j.rser.2014.10.095](https://doi.org/10.1016/j.rser.2014.10.095) (2015). 913 Hasselmann K., S. Hasselmann, R. Giering, V. Ocana, H. von Storch, Sensitivity study of optimal CO2 914 emission paths using a Simplified Structural Integrated Assessment Model (SIAM). Climatic Change 37, 915 345–386, [https://doi.org/10.1023/A:1005339625015](https://doi.org/10.1023/A:1005339625015) (1997). 916 Hendricks, J., B. Kärcher, U. Lohmann, Effects of ice nuclei on cirrus clouds in a global climate model. 917 _Journal of Geophysical Research Atmospheres 116, 2156–2202, [https://doi.org/10.1029/2010JD015302](https://doi.org/10.1029/2010JD015302) 918_ (2011). 919 Hodnebrog, Ø., T. K. Berntsen, O. Dessens, M. Gauss, V. Grewe, I. S. A. Isaksen, B. Koffi, G. Myhre, D. 920 Olivié, M. J. Prather, J. A. Pyle, F. Stordal, S. Szopa, Q. Tang P. van Velthoven, J. E. Williams, K. 921 Ødemark, Future impact of non-land based traffic emissions on atmospheric ozone and OH – an 922\ 923 optimistic scenario and a possible mitigation strategy. Atmospheric Chemistry and Physics 11, 11,293– 11,317, [https://doi.org/10.5194/acp-11-11293-2011](https://doi.org/10.5194/acp-11-11293-2011) (2011). 924\ \ Hodnebrog, Ø., T. K. Berntsen, O. Dessens, M. Gauss, V. Grewe, I. S. A. Isaksen, B. Koffi, G. Myhre, D. 925 Olivié, M. J. Prather, F. Stordal, S. Szopa, Q. Tang, P. van Velthoven, J. E. Williams, Future impact of 926 927 traffic emissions on atmospheric ozone and OH based on two scenarios. Atmospheric Chemistry and _Physics 12, 12,211–12,225, [https://doi.org/10.5194/acp-12-12211-2012](https://doi.org/10.5194/acp-12-12211-2012) (2012). 928_\ \ Holmes, C. D., Q. Tang, M. J. Prather, Uncertainties in climate assessment for the case of aviation 929 NO. Proceedings of the National Academy of Science U.S.A. 108(27), 10997–11002, 930 [https://doi.org/10.1073/pnas.1101458108](https://doi.org/10.1073/pnas.1101458108) (2011). 931\ \ Holmes, C. D., M. J. Prather, O. A. Søvde, G. Myhre, Future methane, hydroxyl, and their uncertainties: 932 933 key climate and emission parameters for future predictions. Atmospheric Chemistry and Physics 13, 285– 302, [https://doi.org/10.5194/acp-13-285-2013](https://doi.org/10.5194/acp-13-285-2013) (2013). 934\ \ Hoor, P., J. Borken-Kleefeld, D. Caro, O. Dessens, O. Endresen, M. Gauss, V. Grewe, D. Hauglustaine, I. 935\ \ S. A. Isaksen, P. Jöckel, J. Lelieveld, G. Myhre, E. Meijer, D. Olivié, M. Prather, C. Schnadt-Poberaj, K. 936\ P. Shine, J. Staehelin, Q. Tang, J. van Aardenne, P. van Velthoven, R. Sausen, The impact of traffic 937 emissions on atmospheric ozone and OH: results from QUANTIFY. Atmospheric Chemistry and Physics 938 9, 3113–3136, [https://doi.org/10.5194/acp-9-3113-2009](https://doi.org/10.5194/acp-9-3113-2009) (2009). 939\ \ * * *\ \ 24 August 2020 Revised\ \ 940 Hoose, C. and O. Möhler, Heterogeneous ice nucleation on atmospheric aerosols: a review of results from laboratory experiments. Atmospheric Chemistry and Physics 12, 9817–9854, [https://doi.org/10.5194/acp-941](https://doi.org/10.5194/acp-941) 12-9817-2012 (2012). 942\ \ Hough, A. M., The development of a two-dimensional global tropospheric model – 1. The model 943 transport. Atmospheric Environment 23, 1235–1261, [https://doi.org/10.1016/0004-6981(89](https://doi.org/10.1016/0004-6981(89)) 90150-9 944 (1989). 945\ \ 946 Hough, A. M., Development of a two-dimensional global tropospheric model: model chemistry. Journal _of Geophysical Research Atmospheres 96, 7325–7362, [https://doi.org/10.1029/90JD01327](https://doi.org/10.1029/90JD01327) (1991). 947_\ \ Irvine, E. A., B. J. Hoskins, K. P. Shine, A Lagrangian analysis of ice-supersaturated air over the North 948 949 Atlantic. Journal of Geophysical Research Atmospheres 119, 90–100, [https://doi.org/10.1002/2013JD020251](https://doi.org/10.1002/2013JD020251) (2013). 950\ \ IATA, Economic Performance of the Airline Industry. 951 [https://www.iata.org/contentassets/f88f0ceb28b64b7e9b46de44b917b98f/iata-economic-performance-of-952](https://www.iata.org/contentassets/f88f0ceb28b64b7e9b46de44b917b98f/iata-economic-performance-of-952) the-industry-end-year-2018-report.pdf (2019). 953\ \ IATA, Outlook for air travel in the next 5 years, [https://www.iata.org/en/iata-954](https://www.iata.org/en/iata-954) repository/publications/economic-reports/covid-19-outlook-for-air-travel-in-the-next-5-years/ (2020) 955\ \ ICAO (2018) ICAO Carbon Emissions Calculator Methodology, version 11, June 2018, 956 ( [https://www.icao.int/environmental-957](https://www.icao.int/environmental-957) 958 protection/CarbonOffset/Documents/Methodology%20ICAO%20Carbon%20Calculator\_v11-2018.pdf) accessed 19-05-2020. 959\ \ ICAO, ‘Destination Green the Next Chapter’, ICAO Environmental Report, Montreal, 960 [https://www.icao.int/environmental-protection/Documents/ICAO-ENV-Report2019-F1-WEB%20(1](https://www.icao.int/environmental-protection/Documents/ICAO-ENV-Report2019-F1-WEB%20(1)).pdf 961 (2019). 962\ \ IEA, International Energy Agency. International Energy Agency Oil Information, 1960-2017. \[data 963 collection\]. 12th Edition. UK Data Service. SN: 5187, [http://doi.org/10.5257/iea/oil/2019-1](http://doi.org/10.5257/iea/oil/2019-1) (2019). 964\ \ IPCC (1999), “Aviation and the Global Atmosphere”, Intergovernmental Panel on Climate Change 965 Special Report, J. E. Penner, D. H. Lister, D. J. Griggs, D. J. Dokken, M. McFarland, Eds. (Cambridge 966 University Press, Cambridge, UK, 1999) [https://www.ipcc.ch/report/aviation-and-the-global-atmosphere-967](https://www.ipcc.ch/report/aviation-and-the-global-atmosphere-967) 2/. 968\ \ IPCC (2001) “Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third 969 Assessment Report of the Intergovernmental Panel on Climate Change”. J.T. Houghton, Y. Ding, D.J. 970 Griggs, M. Noguer, P.J. van der Linden, X. Dai, K. Maskell and C.A. Johnson (eds). Cambridge 971 University Press, UK. [https://www.ipcc.ch/site/assets/uploads/2018/07/WG1\_TAR\_FM.pdf](https://www.ipcc.ch/site/assets/uploads/2018/07/WG1_TAR_FM.pdf) 972\ \ 973 IPCC (2007), “Climate change 2007. “Mitigation of climate change”, in: Contribution of Working Group 974 III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change”, B. Metz, O. R. Davidson, P. R. Bosch, R. Dave, L. A. Meyer, eds (Cambridge University Press, UK) 975 [https://www.ipcc.ch/report/ar4/wg3/](https://www.ipcc.ch/report/ar4/wg3/) 976\ \ IPCC (2013) “Climate Change 2013: The Physical Science Basis, Contribution of Working Group I to the 977 Fifth Assessment Report of the Intergovernmental Panel on Climate Change”, T. F. Stocker, D. Qin, G. -978\ \ K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, P. M. Midgley, Eds. 979 (Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2013). 980 [https://www.ipcc.ch/report/ar5/wg1/](https://www.ipcc.ch/report/ar5/wg1/) 981\ \ * * *\ \ IPCC (2018) “Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 982\ 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of 983\ strengthening the global response to the threat of climate change, sustainable development, and efforts to 984\ eradicate poverty”, Masson-Delmotte, V., P. Zhai, H.-O. Pörtner, D. Roberts, J. Skea, P.R. Shukla, A. 985\ Pirani, W. Moufouma-Okia, C. Péan, R. Pidcock, S. Connors, J.B.R. Matthews, Y. Chen, X. Zhou, M.I. 986\ Gomis, E. Lonnoy, T. Maycock, M. Tignor, and T. Waterfield (eds), (2018). 987\ [https://www.ipcc.ch/sr15/download/](https://www.ipcc.ch/sr15/download/) 988\ Joos, F., M. Bruno, R. Fink, T. F. Stocker, U. Siegenthaler, C. LeQuéré, J. L. Sarmiento, J.L., An efficient 989\ and accurate representation of complex oceanic and biospheric models for anthropogenic carbon uptake. 990\ Tellus 48B, 397e417, [https://doi.org/10.1034/j.1600-0889.1996.t01-2-00006.x](https://doi.org/10.1034/j.1600-0889.1996.t01-2-00006.x) (1996) 991\ Joos, F., R. Roth, J. S. Fuglestvedt, G. P. Peters, I. G. Enting, W. von Bloh, V. Brovkin, E. J. Burke, M. 992\ Eby, N. R. Edwards, T. Friedrich, T. L. Frolicher, P. R. Halloran, P. B. Holden, C. Jones, T. Kleinen,F. T. 993\ Mackenzie, K. Matsumoto, M. Meinshausen, G.-K. Plattner, A. Reisinger, J. Segschneider, G. Shaffer, 994\ M. Steinacher, K. Strassmann, K. Tanaka, A. Timmermann, A. J. Weaver, Carbon dioxide and climate 995\ impulse response functions for the computation of greenhouse gas metrics: a multi-model analysis. 996\ Atmospheric Chemistry and Physics 13, 2793–2825, [https://doi.org/10.5194/acp-13-2793-2013](https://doi.org/10.5194/acp-13-2793-2013) (2013). 997\ Kapadia, Z. Z., D. V. Spracklen, S. R. Arnold, D. J. Borman, G. W. Mann, K. J. Pringle, S. A. Monks, C. 998\ L. Reddington, F. Benduhn, A. Rap, C. E. Scott, E. W. Butt, M. Yoshioka, Impacts of aviation fuel sulfur 999\ content on climate and human health. Atmospheric Chemistry and Physics 16, 10521–10541, 1000\ [https://doi.org/10.5194/acp-16-10521-2016](https://doi.org/10.5194/acp-16-10521-2016) (2016). 1001\ 1002 Kärcher, B., U. Burkhardt, A. Bier, L. Bock, I. J. Ford, The microphysical pathway to contrail formation.\ Journal of Geophysical Research Atmospheres 120, 7893–7927, [https://doi.org/10.1002/2015JD023491](https://doi.org/10.1002/2015JD023491) 1003\ (2015). 1004\ Kärcher, B. Formation and radiative forcing of contrail cirrus. Nature Communications 9:1824, 1005\ [https://doi.org/10.1038/s41467-018-04068-0](https://doi.org/10.1038/s41467-018-04068-0) (2018). 1006\ Khodayari, A., D. J. Wuebbles, S. Olsen, J. S. Fuglestvedt, T. Berntsen, M. T. Lund, I. Waitz, P. Wolfe, 1007\ P. M. Forster, M. Meinshausen, D. S. Lee, L. L. Lim, Intercomparison of the capabilities of simplified 1008\ climate models to project the effects of aviation CO2 on climate. Atmospheric Environment 75, 321–328, 1009\ [https://doi.org/10.1016/j.atmosenv.2013.03.055](https://doi.org/10.1016/j.atmosenv.2013.03.055) (2013). 1010\ Khodayari, A., S. C. Olsen, D. J. Wuebbles, Evaluation of aviation NOx-induced radiative forcings for 1011\ 2005 and 2050. Atmospheric Environment 91, 95–103, [https://doi.org/10.1016/j.atmosenv.2014.03.044](https://doi.org/10.1016/j.atmosenv.2014.03.044) 1012\ (2014a). 1013\ Khodayari, A., S. Tilmes, S. C. Olsen, D. B. Phoenix, D. J. Wuebbles, J.-F. Lamarque, C.-C. Chen, 1014\ Aviation 2006 NOx-induced effects on atmospheric ozone and HOx in Community Earth System Model 1015\ (CESM). Atmospheric Chemistry and Physics 14, 9925–9939, [https://doi.org/10.5194/acp-14-9925-2014](https://doi.org/10.5194/acp-14-9925-2014) 1016\ (2014b). 1017\ Köhler, M. O., G. Rädel, O. Dessens, K. P. Shine, H. L. Rogers, O. Wild, J. A. Pyle, Impact of 1018\ 1019 perturbation of nitrogen oxide emissions from global aviation Journal of Geophysical Research\ Atmospheres 113, D11305, [https://doi.org/10.1029/2007JD009140](https://doi.org/10.1029/2007JD009140) (2008). 1020\ Köhler, M. O., G. Rädel, K. P. Shine, H. L. Rogers, J. A. Pyle, Latitudinal variation of the effect of 1021\ 1022 aviation NOx emissions on atmospheric ozone and methane and related climate metrics. Atmospheric\ Environment 64, 1–9, [https://doi.org/10.1016/j.atmosenv.2012.09.013](https://doi.org/10.1016/j.atmosenv.2012.09.013) (2013). 1023\ Lamarque, J.-F., T. C. Bond, V. Eyring, C. Granier, A. Heil, Z. Klimont, D. Lee, C. Liousse, A. Mieville, 1024\ B. Owen, M. G. Schultz, D. Shindell, S. J. Smith, E. Stehfest, J. van Aardenne, O. R. Cooper, M. 1025\ \ 1.5^{\\circ}\\mathrm{C}.\ \ 1.5^{\\circ}\\mathrm{C}\ \ \\mathrm{C O\_{2}}\ \ .mathrm O,\ \\ \ * * *\ \ 1035\ 1041\ \ Kainuma, N. Mahowald, J. R. McConnell, V. Naik, K. Riahi, D. P. van Vuuren, Historical (1850–2000) 1026\ gridded anthropogenic and biomass burning emissions of reactive gases and aerosols: methodology and 1027\ application. Atmospheric Chemistry and Physics 10, 7017–7039, [https://doi.org/10.5194/acp-10-7017-1028](https://doi.org/10.5194/acp-10-7017-1028)\ 2010 (2010). 1029\ Lamquin, N., C. J. Stubenrauch, K. Gierens, U. Burkhardt, H. Smit, A global climatology of upper 1030\ tropospheric ice supersaturation occurrence inferred from the Atmospheric Infrared Sounder calibrated by 1031\ MOZAIC. Atmospheric Chemistry and Physics 12, 381–405, [https://doi.org/10.5194/acp-12-381-2012](https://doi.org/10.5194/acp-12-381-2012) 1032\ (2012). 1033\ Lauder, A. R., I. G. Enting, J. O. Carter, N. Clisby, A. L. Cowie, B. K. Henry, M. R. Raupach, Offsetting 1034\ 1035 methane emissions — An alternative to emission equivalence metrics. International Journal of\ Greenhouse Gas Control 12, 419–429, [https://doi.org/10.1016/j.ijggc.2012.11.028](https://doi.org/10.1016/j.ijggc.2012.11.028) (2012). 1036\ Lee, D. S., D. Fahey, P. M. Forster, P. J. Newton, R. C. N. Wit, L. L. Lim, B. Owen, R. Sausen, Aviation 1037\ and global climate change in the 21st century. Atmospheric Environment 43, 3520–3537 1038\ [https://doi.org/10.1016/j.atmosenv.2009.04.024](https://doi.org/10.1016/j.atmosenv.2009.04.024) (2009). 1039\ Lee, D. S., G. Pitari, V. Grewe, K. Gierens, J. E. Penner, A. Petzold, M. Prather, U. Schumann, A. Bais, 1040\ 1041 T. Berntsen, D. Iachetti, L. L. Lim, R. Sausen, Transport impacts on atmosphere and climate: Aviation.\ Atmospheric Environment 44, 4678–4734, [https://doi.org/10.1016/j.atmosenv.2009.06.005](https://doi.org/10.1016/j.atmosenv.2009.06.005) (2010). 1042\ Le Quéré, C. and 76 others, Global carbon budget 2018. Earth System Science Data 10, 2141–2194, 1043\ [https://doi.org/10.5194/essd-10-2141-2018](https://doi.org/10.5194/essd-10-2141-2018) (2018). 1044\ Le Quéré, C., R. B. Jackson, M. W. Jones, A. J. P. Smith, S. Abernethy, R. M. Andrew, A. J. De-Gol1, D. 1045\ R. Willis, Y. Shan, J. G. Canadell, P. Friedlingstein, F. Creutzig and G. P. Peters, Temporary reduction in 1046\ daily global CO2 emissions during the COVID-19 forced confinement, Nature Climate Change, 1047\ [https://doi.org/10.1038/s41558-020-0797-x](https://doi.org/10.1038/s41558-020-0797-x) (2020). 1048\ Lim, L. L., D. S. Lee, B. Owen, A. Skowron, S. Matthes, U. Burkhardt, S. Dietmuller, G. Pitari, G. Di 1049\ Genova, D. Iachetti, I. Isaksen, O. A. Søvde, REACT4C: Simplified mitigation study. TAC-4 1050\ Proceedings, June 22nd to 25th, 2015, Bad Kohlgrub, 181-185, 1051\ [https://www.pa.op.dlr.de/tac/2015/Proceedings\_of\_TAC4\_conference\_final.pdf](https://www.pa.op.dlr.de/tac/2015/Proceedings_of_TAC4_conference_final.pdf) (2015). 1052\ Liou, K. N., Y. Takano, Q. Yue, P. Yang, On the radiative forcing of contrail cirrus contaminated by 1053\ black carbon. Geophysical Research Letters 40, 778–784, [https://doi.org/10.1002/GRL.50110](https://doi.org/10.1002/GRL.50110) (2013). 1054\ Lund, M. T., B. Aamaas, T. Berntsen, L. Bock, U. Burkhardt, J. S. Fuglestvedt, K. P. Shine, Emission 1055\ metrics for quantifying regional climate impacts of aviation. Earth System Dynamics 8, 547–563, 1056\ [https://doi.org/10.5194/esd-8-547-2017](https://doi.org/10.5194/esd-8-547-2017) (2017). 1057\ Mahrt, F., K. Kilchhofer, C. Marcolli, P. Grönquist, R. O. David, M. Rösch, U. Lohmann, Z. A. Kanji, 1058\ The impact of cloud processing on the ice nucleation abilities of soot particles at cirrus temperatures. 1059\ Journal of Geophysical Research 125, e2019JD030922, [https://doi.org/10.1029/2019JD030922](https://doi.org/10.1029/2019JD030922) (2020). 1060\ Maier-Reimer, E. and K. Hasselmann, Transport and storage of CO2 in the ocean—An inorganic ocean-1061\ circulation carbon cycle model. Climate Dynamics 2, 63–90, [https://doi.org/10.1007/BF01054491](https://doi.org/10.1007/BF01054491) (1987). 1062\ Matthes, M., V. Grewe, K. Dahlmann, C. Frömming, E. Irvine, L. Lim, F. Linke, B. Lührs, B. Owen, K. 1063\ Shine, S. Stromatas, H. Yamashita, F. Yin, A concept for multi-criteria environmental assessment of 1064\ aircraft trajectories. Aerospace 4 42, [https://doi.org/10.3390/aerospace4030042](https://doi.org/10.3390/aerospace4030042) (2017). 1065\ Markowicz, K. M. and M. L. Witek, Simulations of contrail optical properties and radiative forcing for 1066\ various crystal shapes. Journal of Applied Meteorology and Climatology 50, 1740–1755, 1067\ [https://doi.org/10.1175/2011JAMC2618.1](https://doi.org/10.1175/2011JAMC2618.1) (2011). 1068\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ * * *\ \ 24 August 2020 Revised\ \ Marquart, S., R. Sausen, M. Ponater, V. Grewe, Estimate of the climate impact of the cryoplanes, 1069 _Aerospace Science and Technology, 5, 73-84, [https://doi.org/10.1016/S1270-9638(00](https://doi.org/10.1016/S1270-9638(00)) 01084-1 (2001). 1070_\ \ Mastrandrea, M. D., K. J. Mach, G. K. Plattner, O. Edenhofer, T. F. Stocker, C. B. Field, K. L. Ebi, P. R. 1071 Matschoss, The IPCC AR5 guidance note on consistent treatment of uncertainties: a common approach 1072 across the working groups. Climatic Change 108, 675–691, [https://doi.org/10.1007/s10584-011-0178-6](https://doi.org/10.1007/s10584-011-0178-6) 1073 (2011). 1074\ \ Meinshausen, M., S. J. Smith, K. Calvin, J. S. Daniel, M. L. T. Kainuma, J. -F. Lamarque, K. Matsumoto, 1075\ \ S. A. Montzka, S. C. B. Raper, K. Riahi, A. Thomson, G. J. M. Velders, D. P. P. van Vuuren, The RCP 1076 greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change 109, 213–241, 1077 [https://doi.org/10.1007/s10584-011-0156-z](https://doi.org/10.1007/s10584-011-0156-z) (2011). 1078 Millar, R. J., Z. R. Nicholls, P. Friedlingstein, M. R. Allen, A modified impulse-response representation 1079 of the global near-surface air temperature and atmospheric concentration response to carbon dioxide 1080 emissions. Atmospheric Chemistry and Physics. 17, 7213–7228, [https://doi.org/10.5194/acp-17-7213-1081](https://doi.org/10.5194/acp-17-7213-1081) 2017 (2017). 1082 Miller, M., P. Brook and C. Eyers, Reduction of Sulphur limits in aviation fuel standards (SULPHUR). 1083 EASA research project EASA.2008/C11, European Aviation Safety Agency. (2010) 1084 [https://www.easa.europa.eu/sites/default/files/dfu/2009-SULPHUR-1085](https://www.easa.europa.eu/sites/default/files/dfu/2009-SULPHUR-1085) Reduction%20of%20sulphur%20limits%20in%20aviation%20fuel%20standards-Final%20Report.pdf 1086 Minnis, P., S. T. Bedka, D. P. Duda, K. M. Bedka, T. Chee, J. K. Ayers, R. Palikonda, D. A. 1087 Spangenberg, K. V. Khlopenkov, R. Boeke, Linear contrail and contrail cirrus properties determined from 1088 satellite data. Geophysical Research Letters, 40, 3220–3226, [https://doi.org/10.1002/grl.50569](https://doi.org/10.1002/grl.50569) (2013). 1089 Möhler, O., S. Büttner, C. Linke, M. Schnaiter, H. Saathof, O. Stetzer, R. Wagner, M. Krämer, A. 1090\ 1091 Mangold, V. Ebert, U. Schurath, Effect of sulfuric acid coating on heterogenous ice nucleation by soot aerosol particles. Journal of Geophysical Research 110 (D11), [https://doi.org/10.1029/2004JD005169](https://doi.org/10.1029/2004JD005169) 1092 (2005). 1093\ \ Montzka, S. A., C. M. Spivakovsky, J. H. Butler, J. W. Elkins, L. T. Lock, D. J. Mondeel, New 1094 1095 observational constraints for atmospheric hydroxyl on global and hemispherical scales. Science 288, 500– 503, [https://doi.org/10.1126/science.288.5465.500](https://doi.org/10.1126/science.288.5465.500) (2000). 1096\ \ Moore, R. H., K. L. Thornhill, B. Weinzierl, D. Sauer, E. D'Ascoli, J. Kim, et al., Biofuel blending 1097 reduces particle emissions from aircraft engines at cruise conditions. Nature, 543, 411–415, 1098 [https://doi.org/10.1038/nature21420](https://doi.org/10.1038/nature21420) (2017) 1099\ \ Myhre, G., J. S. Nilsen, L. Gulstad, K. P. Shine, B. Rognerud, I. S. A. Isaksen, Radiative forcing due to 1100 stratospheric water vapor from CH4 oxidation. Geophysical Research Letters 34, L01807, 1101 [https://doi.org/10.1029/2006GL027472](https://doi.org/10.1029/2006GL027472) (2007). 1102\ \ Myhre, G., M. Kvalevåg, G. Rädel, J. Cook, K. P. Shine, H. Clark, F. Karcher, K. Markowicz, A. Kardas, 1103\ \ P. Wolkenberg, Y. Balkanski, M. Ponater, P. Forster, A. Rap, R. R. de Leon, Intercomparison of 1104 radiative forcing calculations of stratospheric water vapour and contrails. Meteorologische Zeitschrift 18, 1105 585–596, [https://doi.org/10.1127/0941-2948/2009/0411](https://doi.org/10.1127/0941-2948/2009/0411) (2009). 1106 Myhre, G., K. P. Shine, G. Rädel, M. Gauss, I. S. A. Isaksen, Q. Tang, M. J. Prather, J. E. Williams, P. 1107 van Velthoven, O. Dessens, B. Koffi, S. Szopa, P. Hoor, V. Grewe, J. Borken-Kleefeld, T. K. Berntsen, J. 1108\ S. Fuglestvedt, Radiative forcing due to changes in ozone and methane caused by the transport sector. 1109 _Atmospheric Environment 45, 387–394, [https://doi.org/10.1016/j.atmosenv.2010.10.001](https://doi.org/10.1016/j.atmosenv.2010.10.001) (2011). 1110_ Myhre, G., D. Shindell, F. -M Breon, W. Collins, J. Fuglestvedt, J. Huang, D. Koch, J. -F. Lamarque, D. 1111 Lee., B. Mendoza, T. Nakajima, A. Robock, G. Stephens, T. Takemura, H. Zhang, “Anthropogenic and\ \ * * *\ \ 1113\ 1120\ 1124\ 1149\ 1154\ \ 1113 Natural Radiative Forcing” 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, 1114\ (Cambridge University Press, 2013). [https://www.ipcc.ch/report/ar5/wg1/](https://www.ipcc.ch/report/ar5/wg1/) 1115\ Newinger, C. and U. Burkhardt, Sensitivity of contrail cirrus radiative forcing to air traffic scheduling. 1116\ Journal of Geophysical Research Atmospheres 117, D10205, [https://doi.org/10.1029/2011JD016736](https://doi.org/10.1029/2011JD016736) 1117\ (2012). 1118\ OECD, Green growth and the future of aviation. Paper prepared for the 27th Round Table on Sustainable 1119\ 1120 Development held at OECD Headquarters 23-24 January 2012 (OECD 2012). [https://www.oecd.org/sdroundtable/papersandpublications/49482790.pdf](https://www.oecd.org/sdroundtable/papersandpublications/49482790.pdf) 1121\ Olivié, D. J. L., D. Cariolle, H. Teyssèdre, D. Salas, A. Voldoire, H. Clark, D. Saint-Martin, M. Michou, 1122\ F. Karcher, Y. Balkanski, M. Gauss, O. Dessens, B. Koffi, R. Sausen, Modeling the climate impact of 1123\ 1124 road transport, maritime shipping and aviation over the period 1860–2100 with an AOGCM. Atmospheric\ Chemistry and Physics 12, 1449–1480, [https://doi.org/10.5194/acp-12-1449-2012](https://doi.org/10.5194/acp-12-1449-2012) (2012). 1125\ Olsen, S. C., G. P. Brasseur, D. J. Wuebbles, S. R. H. Barrett, H. Dang, S. D. Eastham, M. Z. Jacobson, 1126\ A. Khodayari, H. Selkirk, A. Sokolov, N. Unger, Comparison of model estimates of the effects of 1127\ aviation emissions on atmospheric ozone and methane. Geophysical Research Letters 40, 6004–6009, 1128\ [https://doi.org/10.1002/2013GL057660](https://doi.org/10.1002/2013GL057660) (2013). 1129\ Penner, J.E., Y. Chen, M. Wang, and X. Liu, Possible influence of anthropogenic aerosols on cirrus 1130\ clouds and anthropogenic forcing. Atmospheric Chemistry and Physics, 9, 879–96, 1131\ [https://doi.org/10.5194/acp-9-879-2009](https://doi.org/10.5194/acp-9-879-2009) (2009). 1132\ Penner, J. E., C. Zhou, A. Garnier, D. L. Mitchell, Anthropogenic aerosol indirect effects in cirrus clouds. 1133\ Journal of Geophysical Research Atmospheres, 123, 11,652–11,677, 1134\ [https://doi.org/10.1029/2018JD029204](https://doi.org/10.1029/2018JD029204) (2018). 1135\ Petzold, A., M. Gysel, X. Vancassel, R. Hitzenberger, H. Puxbaum, S. Vrochticky, E. Weingartner, U. 1136\ Baltensperger, P. Mirabel, On the effects of organic matter and sulphur-containing compounds on the 1137\ CCN activation of combustion particles. Atmospheric Chemistry and Physics 5, 3187–3203, 1138\ [https://doi.org/10.5194/acp-5-3187-2005](https://doi.org/10.5194/acp-5-3187-2005) (2005). 1139\ Pitari, G., D. Iachetti, G. Genova, N. De Luca, O. A. Søvde, Ø. Hodnebrog, D. S. Lee, L. L. Lim, Impact 1140\ of coupled NOx/aerosol aircraft emissions on ozone photochemistry and radiative forcing. Atmosphere 6, 1141\ 751–782, [https://doi.org/10.3390/atmos6060751](https://doi.org/10.3390/atmos6060751) (2015). 1142\ Pitari, G., I. Cionni, G. Di Genova, O. A. Søvde, L. Lim, Radiative forcing from aircraft emissions of 1143\ NOx: model calculations with CH4 surface flux boundary condition. Meteorologische Zeitschrift 26(6), 1144\ 663-687, [https://doi.org/10.1127/metz/2016/0776](https://doi.org/10.1127/metz/2016/0776) (2017). 1145\ Pomroy, H. R. and J. A. Illingworth, Ice cloud inhomogeneity: Quantifying bias in emissivity from radar 1146\ observations. Geophysical Research Letters 27, 2101–2104, [https://doi.org/10.1029/1999GL011149](https://doi.org/10.1029/1999GL011149) 1147\ (2000). 1148\ 1149 Ponater, M., S. Marquart, R. Sausen, U. Schumann, On contrail climate sensitivity. Geophysical Research\ Letters 32, L10706, [https://doi.org/10.1029/2005GL022580](https://doi.org/10.1029/2005GL022580) (2005). 1150\ Ponater, M., S. Pechtl, R. Sausen, U. Schumann, G. Hüttig, Potential of the cryoplane technology to 1151\ reduce aircraft climate impact: a state-of-the-art assessment. Atmospheric Environment 40, 6928-6944, 1152\ [https://doi.org/10.1016/j.atmosenv.2006.06.036](https://doi.org/10.1016/j.atmosenv.2006.06.036) (2006). 1153\ 1154 Ponater, M., M. Bickel, L. Bock, and U. Burkhardt, Towards determining the efficacy of contrail cirrus.\ In Matthes, S. and A. Blum, Making Aviation Environmentally Sustainable, 3rd ECATS Conference, 1155\ \ * * *\ \ 24 August 2020 Revised\ \ Book of Abstracts, Volume 1. ISBN 978-1-910029-58-9. 51-44 (2020). ( [http://www.ecats-1156](http://www.ecats-1156/) network.eu/uploads/2020/06/ECATS\_Main\_BookOfAbstracts\_Vol1\_final.pdf) 1157\ \ Prather, M. J., Lifetimes and eigenstates in atmospheric chemistry. Geophysical Research Letters 21, 1158 801–804, [https://doi.org/10.1029/94GL00840](https://doi.org/10.1029/94GL00840) (1994). 1159\ \ 1160 Prather, M., D. Ehhalt, F. Dentener, R. Derwent, E. Dlugokencky E, “Atmospheric chemistry and 1161 greenhouse gases”, in Climate Change 2001: The Scientific Basis, Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, J. T. Houghton ed. 1162 1163 (Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2001) pp. 239–\ \ 287. [https://www.ipcc.ch/site/assets/uploads/2018/03/TAR-04.pdf](https://www.ipcc.ch/site/assets/uploads/2018/03/TAR-04.pdf) 1164 Prather, M. J., C. D. Holmes, J. Hsu, Reactive greenhouse gas scenarios: Systematic exploration of 1165 uncertainties and the role of atmospheric chemistry, Geophysical Research Letters, 39, L09803, 1166 [https://doi.org/10.1029/2012GL051440](https://doi.org/10.1029/2012GL051440) (2012). 1167\ 1168 Rap, A., P. M. Forster, J. M. Haywood, A. Jones, O. Boucher, Estimating the climate impact of linear contrails using the UK Met Office climate model. Geophysical Research Letters 37, L20703, 1169 [https://doi.org/10.1029/2010GL045161](https://doi.org/10.1029/2010GL045161) (2010). 1170\ \ Revelle, R. and H. E. Suess, Carbon dioxide exchange between atmosphere and ocean and the question of 1171 an increase of atmospheric CO2 during the past decades, Tellus, 9, 18–27, [https://doi.org/10.1111/j.2153-1172](https://doi.org/10.1111/j.2153-1172)\ \ 3490.1957.tb01849.x (1957). 1173 Richardson, T. B., P.M. Forster, C. J. Smith, A. C. Maycock, T. Wood, T. Andrews, O. Boucher, G. 1174 Faluvegi, D. Fläschner, Ø. Hodnegrog, M. Kasoar, A. Kirkevåg, J.-F. Lamarque, J. Mülmenstädt, G. 1175 Myhre, D. Olivié, R. W. Portmann, B. H. Samset, D. Shawki, D. Shindell, P. Stier, T. Takemura, A. 1176\ 1177 Voulgarakis, D. Watson-Parris, Efficacy of climate forcings in PDRMIP models. Journal of Geophysical _Research: Atmospheres 124, [https://doi.org/10.1029/2019JD030581](https://doi.org/10.1029/2019JD030581) (2019). 1178_\ \ 1179 Righi, M., J. Hendricks, R. Sausen, The global impact of the transport sectors on atmospheric aerosol: simulations for year 2000 emissions. Atmospheric Chemistry and Physics 13, 9939–9970, 1180 [https://doi.org/10.5194/acp-13-9939-2013](https://doi.org/10.5194/acp-13-9939-2013) (2013). 1181\ \ Sander, S. P., R. R. Friedl, A. R. Ravishankara, D. M. Golden, C. E. Kolb, M. J. Kurylo, M. J. Molina, G. 1182\ \ K. Moortgat, B. J. Finlayson-Pitts, “Chemical Kinetics and Photochemical Data for Use in Atmospheric 1183 Studies”, (JPL Publ. 06-2, No. 15, 2006). [https://jpldataeval.jpl.nasa.gov/pdf/JPL\_02-25\_rev02.pdf](https://jpldataeval.jpl.nasa.gov/pdf/JPL_02-25_rev02.pdf) 1184 Sausen R. and U. Schumann, Estimates of the climate response to aircraft CO2 and NOx emissions 1185 scenarios. Climatic Change 44, 27–58 (2000). 1186 Sausen, R. I. Isaksen, V. Grewe, D. Hauglustaine, D. S. Lee, G. Myhre, M. O. Köhler, G. Pitari, U. 1187 Schumann, F. Stordal, C. Zerefos, Aviation radiative forcing in 2000: An update on IPCC (1999). 1188 _Meteorologische Zeitschrift 14, 555–561, [https://doi.org/10.1127/0941-2948/2005/0049](https://doi.org/10.1127/0941-2948/2005/0049) (2005). 1189_ Schumann, U., J. E. Penner, Y. Chen, C. Zhou, K. Graf, Dehydration effects from contrails in a coupled 1190 contrail-climate model. Atmospheric Chemistry and Physics 15, 11179–11199, 1191 [https://doi.org/10.5194/acp-15-11179-2015](https://doi.org/10.5194/acp-15-11179-2015) (2015). 1192 Schumann, U., R. Baumann, D. Baumgardner, S. T. Bedka, D. P. Duda, V. Freudenthaler, J.-F. Gayet, A. 1193\ J. Heymsfield, P. Minnis, M. Quante, E. Raschke, H. Schlager, M. Vázquez-Navarro, C. Voigt, Z. Wang, 1194\ 1195 Properties of individual contrails: a compilation of observations and some comparisons. Atmospheric _Chemistry and Physics 17, 403–438, [https://doi.org/10.5194/acp-17-403-2017](https://doi.org/10.5194/acp-17-403-2017) (2017). 1196_\ \ * * *\ \ 24 August 2020 Revised\ \ Shine, K. P., J.S. Fuglestvedt, K. Hailemariam, N. Stuber, Alternatives to the global warming potential 1197 for comparing climate impacts of emissions of greenhouse gases. Climatic Change 68, 281–302, 1198 [https://doi.org/10.1007/s10584-005-1146-9](https://doi.org/10.1007/s10584-005-1146-9) (2005). 1199\ \ Skeie, R. B., J. Fuglestvedt, T. Berntsen, G. P. Peters, R. Andrew, M. Allen, S. Kallbekken, Perspective 1200 1201 has a strong effect on the calculation of historical contributions to global warming. Environmental _Research Letters 12, 024022, [https://doi.org/10.1088/1748-9326/aa5b0a](https://doi.org/10.1088/1748-9326/aa5b0a) (2017). 1202_\ \ 1203 Skowron, A., D. S. Lee, J. Hurley, “Aviation NOx Global Warming Potential”, in 2nd International Conference on Transport, Atmosphere and Climate, 25-28 June 2009, Aachen/Maastricht, 1204 Germany/Netherlands, [https://www.pa.op.dlr.de/tac/2009/proceedings/FB2010-10.pdf](https://www.pa.op.dlr.de/tac/2009/proceedings/FB2010-10.pdf) (2009). 1205\ \ Skowron, A., D. S. Lee, R. R. de León, The assessment of the impact of aviation NOx on ozone and other 1206 1207 radiative forcing responses–The importance of representing cruise altitudes accurately. Atmospheric _Environment 74, 159–168, [https://doi.org/10.1016/j.atmosenv.2013.03.034](https://doi.org/10.1016/j.atmosenv.2013.03.034) (2013). 1208_\ \ Skowron, A., D. S. Lee, R. R. de León, Variation of radiative forcings and global warming potentials 1209 from regional aviation NOx emissions. Atmospheric Environment 104, 69–78, 1210 [https://doi.org/10.1016/j.atmosenv.2014.12.043](https://doi.org/10.1016/j.atmosenv.2014.12.043) (2015). 1211\ \ Søvde, O. A., S. Matthes, A. Skowron, D. Iachetti, L. Lim, B. Owen, Ø. Hodnebrog, G. Di Genova, G. 1212 Pitari, D. S. Lee, G. Myhre, I. S. A. Isaksen, Aircraft emission mitigation by changing route altitude: A 1213 multi-model estimate of aircraft NOx emission impact on O3 photochemistry. Atmospheric Environment 1214 95, 468–479, [https://doi.org/10.1016/j.atmosenv.2014.06.049](https://doi.org/10.1016/j.atmosenv.2014.06.049) (2014). 1215\ \ Smith, C. J., R. J. Kramer, G. Myhre, P. M. Forster, B. J. Soden, T. Andrews, O. Boucher, G. Faluvegi, 1216\ \ D. Fläschner, Ø. Hodnebrog, M. Kasoar, V. Kharin, A. Kirkevåg, J.‐F. Lamarque, J. Mülmenstädt, D. 1217\ 1218 Olivié, T. Richardson, B. H. Samset, D. Shindell, P. Stier, T. Takemura, A. Voulgarakis, D. Watson‐ Parris, Understanding rapid adjustments to diverse forcing agents. Geophysical Research Letters 45, 1219 doi:10.1029/2018GL079826 (2018) 1220\ \ Stevenson, D. S., C. E. Johnson, W. J. Collins, R. G. Derwent, K. P. Shine, J. M. Edwards, Evolution of 1221 tropospheric ozone radiative forcing. Geophysical Research Letters 25, 3819–3822, 1222 [https://doi.org/10.1029/1998GL900037](https://doi.org/10.1029/1998GL900037) (1998). 1223\ \ Stevenson, D. S., R. M. Doherty, M. G. Sanderson, W. J. Collins, C. E. Johnson, R. G. Derwent, 1224 1225 Radiative forcing from aircraft NOx emissions: mechanisms and seasonal dependence Journal of _Geophysical Research Atmospheres 109, D17307, [https://doi.org/10.1029/2004JD004759](https://doi.org/10.1029/2004JD004759) (2004). 1226_\ \ Stordal, F., M. Gauss, G. Myhre, E. Mancini, D. A. Hauglustaine, M. O. Köhler, T. Berntsen, E. J. G. 1227 Stordal, D. Iachetti, G. Pitari, I. S. A. Isaksen, TRADEOFFs in climate effects through aircraft routing: 1228 forcing due to radiatively active gases. Atmospheric Chemistry and Physics Discussions 6, 10733–10771 1229 (2006). 1230\ \ Stuber, N., M. Ponater, R. Sausen, Why radiative forcing might fail as a predictor of climate change. 1231 _Climate Dynamics 24, 497–510 doi:10.1007/s00382-004-0497-7 (2005). 1232_\ \ Stuber, N., P. Forster, G. Rädel, K. Shine, The importance of the diurnal and annual cycle of air traffic for 1233 contrail radiative forcing. Nature 441, 864–867, [https://doi.org/10.1038/nature04877](https://doi.org/10.1038/nature04877) (2006). 1234\ \ Teoh, R., M. E. J. Stettler, A. Majumdar, U. Schumann, B. Graves, A. M. Boies, A methodology to relate 1235 black carbon particle number and mass emissions. Journal of Aerosol Science 132, 44–59, 1236 [https://doi.org/10.1016/j.jaerosci.2019.03.006](https://doi.org/10.1016/j.jaerosci.2019.03.006) (2019). 1237\ \ * * *\ \ Teoh, R., Schumann, U., Majumdat A., Stettler, M. E. J., Mitigating the climate forcing of aircraft 1238\ contrails by small-scale diversions and technology adoption. Environmental Science and Technology, 54 1239\ 2941–2950, doi: 10.1021/acs.est.9b05608 (2020). 1240\ 1241 Tesche, M., P. Achtert, P. Glantz, K. J. Noone, Aviation effects on already-existing cirrus clouds. Nature\ Commumications 7, 12016, [https://doi.org/10.1038/ncomms12016](https://doi.org/10.1038/ncomms12016) (2016). 1242\ UKDS (2016) [http://stats.ukdataservice.ac.uk/index.aspx?r=349678&DataSetCode=IEA\_COAL\_BA](http://stats.ukdataservice.ac.uk/index.aspx?r=349678&DataSetCode=IEA_COAL_BA), 1243\ 2016\. 1244\ UNFCCC, [https://unfccc.int/nationally-determined-contributions-ndcs](https://unfccc.int/nationally-determined-contributions-ndcs) 1245\ Unger, N., T. C. Bond, J. S. Wang, D. M. Koch, S. Menon, D. T. Shindell, S. Bauer, Attribution of 1246\ 1247 climate forcing to economic sectors. Proceedings of the National Academy of Sciences U.S.A. 107, 3382–\ 3387, [https://doi.org/10.1073/pnas.0906548107](https://doi.org/10.1073/pnas.0906548107) (2010). 1248\ 1249 Unger, N., Global climate impact of civil aviation for standard and desulfurized jet fuel. Geophysical\ Research Letters 38, 1–6, [https://doi.org/10.1029/2011GL049289](https://doi.org/10.1029/2011GL049289) (2011). 1250\ Unger, N., Y. Zhao, H. Dang, Mid-21st century chemical forcing of climate by the civil aviation sector. 1251\ Geophysical Research Letters 40, 641–645, [https://doi.org/10.1002/grl.50161](https://doi.org/10.1002/grl.50161) (2013). 1252\ Unterstrasser, S., Large-eddy simulation study of contrail microphysics and geometry during the vortex 1253\ phase and consequences on contrail-to-cirrus transition. Journal of Geophysical Research Atmospheres 1254\ 119, 7537–7555, [https://doi.org/10.1002/2013JD021418](https://doi.org/10.1002/2013JD021418) (2014). 1255\ 1256 Voulgarakis, A., V. Naik, J.-F. Lamarque, D. T. Shindell, P. J. Young, M. J. Prather, O. Wild, R. D. Field,\ D. Bergmann, P. Cameron-Smith, I. Cionni, W. J. Collins, S. B. Dalsøren, R. M. Doherty, V. Eyring, G. 1257\ Faluvegi, G. A. Folberth, L. W. Horowitz, B. Josse, I. A. McKenzie, T. Nagashima, D. A. Plummer, M. 1258\ Righi, S. T. Rumbold, D. S. Stevenson, S. A. Strode, K. Sudo, S. Szopa, G. Zeng, Analysis of present-day 1259\ and future OH and methane lifetime in the ACCMIP simulations. Atmospheric Chemistry and Physics 13, 1260\ 2563–2587, [https://doi.org/10.5194/acp-13-2563-2013](https://doi.org/10.5194/acp-13-2563-2013) (2013). 1261\ Wilcox, L., K. P. Shine, B. J. Hoskins, Radiative forcing due to aviation water vapour emissions. 1262\ Atmospheric Environment 63, 1–13, [https://doi.org/10.1016/j.atmosenv.2012.08.072](https://doi.org/10.1016/j.atmosenv.2012.08.072) (2012). 1263\ Wild, O., M. J. Prather, H. Akimoto, Indirect long-term global radiative cooling from NOx emissions. 1264\ Geophysical Research Letters 28, 1719–1722, [https://doi.org/10.1029/2000GL012573](https://doi.org/10.1029/2000GL012573) (2001). 1265\ Xie B., H. Zhang, Z. Wang, S. Zhao, Q. Fu, A modelling study of effective radiative forcing and climate 1266\ response due to tropospheric ozone. Advances in Atmospheric Sciences 33, 819–828 doi: 10.1007/s00376-1267\ 1268 016-5193-0 (2016).\ Yin, F., V. Grewe, C. Frömming, H. Yamashita, Impact on flight trajectory characteristics when avoiding 1269\ 1270 the formation of persistent contrails for transatlantic routes. Transportation Research Part D: Transport\ and Environment 65, 466–484, [https://doi.org/10.1016/j.trd.2018.09.017](https://doi.org/10.1016/j.trd.2018.09.017) (2018). 1271\ Zhou, C. and J. E. Penner, Aircraft soot indirect effect on large-scale cirrus clouds: Is the indirect forcing 1272\ by aircraft soot positive or negative? Journal of Geophysical Research Atmospheres 119, 11,303-11,320, 1273\ [https://doi.org/10.1002/2014JD021914](https://doi.org/10.1002/2014JD021914) (2014). 1274\ 1275\ \ * * *\ \ 1276\ \ Table 1. Emission indices used in ERF and RF calculations\ \ | Emission | Emission index |\ | --- | --- |\ | CO2 | 3.16kg/kg fuel |\ | NOx | 15.14g/kg fuel |\ | 14.12g/kg fuel | |\ | Water vapor | 1.231kg/kg fuel |\ | Soot | 0.03g/kg fuel |\ | $2\\times10^{14}$ particles/kg fuela | |\ | Sulphur(SO\_{2}$) | 1.2g/kg fuel |\ \ | Reference | Notes |\ | --- | --- |\ | ICAO(2018) | |\ | Fleming and Ziegler(2016) | 2018,2011 |\ | Barrett et al.(2010) | 2005 |\ | Barrett et al.(2010) | |\ | Barrett et al.(2010) | |\ | Miller et al.(2010) | Assumed S content of 600ppm |\ \ 1277\ \ \\mathrm{c O z}\ \ \\mathrm{N O}\_{}\ \ 2{\\times}10^{14}\ \ a\ Assumes mean particle size in the range of 11–79 nm diameter.\ \ * * *\ \ 1280\ 1281\ \ Table 2. Best estimates and high/low limits of the 90% likelihood ranges for aviation ERF components\ derived in this study\ \ | ERF(mWm-2) | 2018a | 2011a |\ | --- | --- | --- |\ | Contrail cirrus | 57.4(17,98) | 44.1(13,75) |\ | CO2 | 34.3(28,40) | 29.0(24,34) |\ | Short-termO3increase | 49.3(32,76) | 37.3(24,58) |\ | Long-termO3decrease | -10.6(-20,-7.4) | -7.9(-15,-5.5) |\ | CH4decrease | -21.2(-40,-15) | -15.8(-30,-11) |\ | Stratospheric watervapor decrease | -3.2(-6.0,-2.2) | -2.4(-4.4,-1.7) |\ | NetNOx | 17.5(0.6,29) | 13.6(0.9,22) |\ | StratosphericH2Oincrease | 2.0(0.8,3.2) | 1.5(0.6,2.4) |\ | Soot(aerosolradiation) | 0.94(0.1,4.0) | 0.71(0.1,3.0) |\ | Sulfate(aerosol-radiation) | -7.4(-19,-2.6) | -5.6(-14,-1.9) |\ | Sulfate and soot(aerosol-cloud) | \-\-\-- | \-\-\-- |\ | NetERF(only non-CO2terms) | 66.6(21,111) | 51.4(16,85) |\ | NetaviationERF | 100.9(55,145) | 80.4(45,114) |\ | Net anthropogenicERF in2011 | \-\-\-- | 2290(1130,3330)b |\ \ | $2005^{a}$ | Sensitivity to emissions | ERF/RF |\ | --- | --- | --- |\ | 34.8(10,59) | 9.36x10^{-10}mW m^{-2}km^{-1}$ | 0.42 |\ | 25.0(21,29) | | 1.0 |\ | 33.0(21,51) | 34.4±9.9mW m^{-2}(Tg(N)yr^{-1})^{-1}$ | 1.37 |\ | -6.7(-13,-4.7) | -9.3±3.4mW m^{-2}(Tg(N)yr^{-1})^{-1}$ | 1.18 |\ | -13.4(-25,-9.4) | -18.7±6.9mW m^{-2}(Tg(N)yr^{-1})^{-1}$ | 1.18 |\ | -2.0(-3.8,-1.4) | -2.8±1.0mW m^{-2}(Tg(N)yr^{-1})^{-1}$ | 1.18 |\ | 12.9(1.9,20) | 5.5±8.1mW m^{-2}(Tg(N)yr^{-1})^{-1}$ | |\ | 1.4(0.6,2.3) | 0.0052±0.0026mW m^{-2}(Tg(H\_{2}O)yr^{-1})^{-1}$ | \-\-\- |\ | 0.67(0.1,2.8) | 100.7±165.5mW m^{-2}(Tg(BC)yr^{-1})^{-1}$ | \-\-\- |\ | -5.3(-13,-1.8) | -19.9±16.0mW m^{-2}(Tg(SO\_{2})yr^{-1})^{-1}$ | \-\-\- |\ | \-\-\-- | \-\-\-- | \-\-\- |\ | 41.9(14,69) | \-\-\-- | \-\-\- |\ | 66.9(38,95) | \-\-\-- | \-\-\- |\ | \-\-\-- | \-\-\-- | \-\-\- |\ \ (m w m^{-2})\ \ \ 018^{a}\ \ 2011^{\\textsf{a}}\ \ 9.36\\times10^{-10}m m m^{-2}k m\ \ \\mathrm{C O}\_{2}\ \ 34.4\\pm9.9m m^{-2}(T g(N)y y^{-1})\\cdot1\ \ 0\_{3}\ \ \\log\\mathsf{t-t e m}O\_{3}\ \ ..9.\ 3.45m m^{2}(T g(N)y r^{-1})\\cdot\ \ {\\mathsf C H}\_{4},d e c r e a s e\ \ -18.7\\pm0.m9m m^{2}(T g(N)y r^{-1})\\cdot1\ \ -.2.\\pm1.0m m m^{2}(T g(N)y r^{-1})\\cdot\ \ \\mathrm{N O}\_{}\ \ 5.5\\pm8.1m M m^{2}(T g(N)y y r^{-1})\\cdot1\ \ 0.0052\\pm0.0026m m^{2}\ \ (\\mathsf{T g},(\\mathsf{H}\_{2}mathsf O,)\\mathsf{y r}^{-1})^{-1}\ \ 100.7\\pm165.m M m^{2}(T g(B C)y r^{-})\ \ 1282\ 1283\ 1284\ 1285\ \ -9.9\\pm16.0m m^{2}(F g(S O\_{2})y r^{-1})\ \ \\mathrm{C O\_{2}}\ \ a\ 1282 The uncertainty distributions for all forcing terms are lognormal except for CO2 and contrail cirrus (normal) and Net\ 1283 NOx (discrete pdf).\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{N o}\_{}\ \ b-2\ Boucher et al., 2013. IPCC also separately estimated the contrail cirrus term for 2011 as 50 (20, 150) mW m.\ \ m W\\mathfrak m{{}}^{{2}}.\ \ * * *\ \ 1286\ 1287\ \ Table 3. Best estimates and low/high limits of the 95% likelihood ranges for aviation RF components\ a\ derived in this study\ \ | RF(mWm-2) | 2018b | 2011b |\ | --- | --- | --- |\ | Contrail cirrus | 111.4(33,189) | 85.6(25,146) |\ | CO2 | 34.3(31,38) | 29.0(26,32) |\ | Short-termO3increase | 36.0(23,56) | 27.3(17,42) |\ | Long-termO3decrease | -9.0(-17,-6.3) | -6.7(-13,-4.7) |\ | CH4decrease | -17.9(-34,-13) | -13.4(-25,-9.3) |\ | Stratospheric water vapor decrease | -2.7(-5.0,-1.9) | -2.0(-3.8,-1.4) |\ | NetNOx | 8.2(-4.8,16) | 6.5(-3.3,12) |\ | StratosphericH2Oincrease | 2.0(0.8,3.2) | 1.5(0.6,2.4) |\ | Soot(aerosol-radiation) | 0.94(0.1,4.0) | 0.71(0.1,3.0) |\ | Sulfate(aerosol-radiation) | -7.4(-19,-2.6) | -5.6(-14,-1.9) |\ | Sulfate and soot(aerosol-cloud) | \-\-\-- | \-\-\-- |\ | NetRF(only non-CO2terms) | 114.8(35,194) | 88.4(27,149) |\ | NetaviationRF | 149.1(70,229) | 117.4(56,179) |\ | aERFvalues are shown inTable2. | | |\ \ | $2005^{b}$ | L09 2005 values | Sensitivity to emissions (this work) |\ | --- | --- | --- |\ | 67.5(20,115) | (11.8 $ ^{\\circ} $) | 1.82x10-9mW m-2km-1 |\ | 25.0(23,27) | 28.0 | |\ | 24.0(15,37) | 26.3 | 25.1±7.3mW m-2(Tg(N)yr-1)$ ^{-1}$ |\ | -5.7(-11,-4.0) | \-\-\-- | -7.9±2.9mW m-2(Tg(N)yr-1)$ ^{-1}$ |\ | -11.4(-21,-7.9) | -12.5 | -15.8±5.9mW m-2(Tg(N)yr-1)$ ^{-1}$ |\ | -1.7(-3.2,-1.2) | \-\-\-- | -2.4±0.9mW m-2(Tg(N)yr-1)$ ^{-1}$ |\ | 6.6(1.9,12) | 13.8 $ ^{\\circ} $ | 1.0±6.6mW m-2(Tg(N)yr-1)$ ^{-1}$ |\ | 1.4(0.6,2.3) | 2.8 | 0.0052±0.0026mW m-2(Tg(H2O)yr-1)$ ^{-1}$ |\ | 0.67(0.1,2.8) | 3.4 | 100.7±165.5mW m-2(Tg(BC)yr-1)$ ^{-1}$ |\ | -5.3(-13,-1.8) | -4.8 | -19.9±16.0mW m-2(Tg(SO2)yr-1)$ ^{-1}$ |\ | \-\-\-- | \-\-\-- | \-\-\-- |\ | 70.3(22,119) | \-\-\-- | \-\-\-- |\ | 95.2(47,144) | 78.0 | \-\-\-- |\ \ 2188^{b}\ \ 211^\ {\\tt b}\ \ \\mathbf{2005^{b}}\ \ 1.82\\times10^{-9}m W m^{-2}k m^{-1}\ \ \\mathrm{C O\_{2}}\ \ \ mathrm O{}\_{3}\ \ 25.1\\pm7.3m\\minmin^{2}(T g(N)y y^{-1})\\cdot1\ \ \ mathrm O{{}}\_33{\ \ -7.9\\pm2.m g M m^{2}(T g(N)y r^{-1})\\cdot\ \ -15.8\\pm5.m9m m^{2}(T g(N)y y^{-1})^cdot{\ \ -2.4\\pm0.9m m^{2}(T g(N)y r^{-1})^{-}\ \ 1.0\\pm6.6m m{m}^{2}(T g(N)y r^{-1})^{\\cdot}\ \ \\mathrm{N o}\_{}\ \ 0.0052\\pm0.0026m m^^{2}\ \ (mathsf{T g},(\\mathsf{H}\_{2}\\mathsf{O}),\\mathsf{y r}^{-1})^{-1}\ \ 1288\ 1289\ 1290\ 1291\ 1292\ 1293\ \ 100.7\\pm165.m M m^{2}(T g(B C)y r^{-1})\ \ -9.9\\pm18.0m m^{2}(T g(S O\_{2})y r^{-1})\ \ \ n\ n C O\_{2}\ \ b\ 1289 The uncertainty distributions for all forcing terms are lognormal except for CO2 and contrail cirrus (normal) and Net\ 1290 NOx (discrete pdf).\ c\ 1291 Linear contrails only; excludes the increase in cirrus cloudiness due to aged spreading contrails.\ \ \\mathrm{c O}}}\ \ \\mathrm{N O}\_{}\ \ d\ Excludes updated CH4 RF evaluation of Etminan et al. (2016) and equilibrium-to-transient correction.\ \ \\mathrm{C H}\_{4}\ \ * * *\ \ 1294 Table 4a. Confidence levels for the ERF estimates in Figure 3\ \ 1295\ 1296\ 1297\ 1298\ 1299\ 1300\ 1301\ \ | Terms | Evidence | Agreement | Conf. level |\ | --- | --- | --- | --- |\ | Contrail cirrus formation in high-humidity regions | Limited | Medium | Low\* |\ | Carbon dioxide (CO2) emissions | Robust | Medium | High\*\* |\ | | | | |\ | Short-term ozone increase | Medium | Medium | Medium\* |\ | Long-term ozone decrease | Limited | Medium | Low\* |\ | Methane decrease | Medium | Medium | Medium\* |\ | Stratospheric water vapour decrease | Limited | Medium | Low\* |\ | Net NOx | Medium | Limited | Low\* |\ | Water vapor emissions in the stratosphere | Medium | Medium | Medium |\ | Aerosol-radiation interactions | | | |\ | From soot emissions | Limited | Medium | Low |\ | From sulfur emissions | Limited | Medium | Low |\ | Aerosol-cloud interactions | | | |\ | From sulfur emissions | Limited | Low | Very |\ \ | Basis for uncertainty estimates | Understanding change since L09 |\ | --- | --- |\ | Robust evidence for the phenomenon. Large remaining uncertainties in magnitude in part due to incomplete representation of key processes | The inclusion of contrail cirrus processes in global climate models. |\ | Trends in aviation CO2 emissions and differences between simplified C-cycle models | Better assessment of uncertainties from multiple models |\ | Observed trends of tropospheric ozone and laboratory studies of chemical kinetics, reliance on a large number of model results for aviation emissions | Elevated owing to many more studies |\ | Reliance on chemical modelling studies | Not provided previously |\ | Observed trends of tropospheric methane and laboratory studies of chemical kinetics, reliance on a large number of model results for aviation emissions | Elevated owing to many more studies |\ | Reliance on chemical modelling studies | Not provided previously |\ | Associated uncertainties with combining above effects | Elevated owing to more studies but lowered in total owing to additional terms and methodological constraints |\ | Limited studies of perturbation of water vapor budget of UT/LS | Elevated owing to more studies |\ | | |\ | Limited studies and uncertain emission index | More studies |\ | Limited studies and uncertain emission index | More studies |\ | | |\ | None available; few studies, probably a negative EFE | Not provided previously |\ \ 1w^^{\\star}\ \ (\\mathbf{c o\_{2}})\ \ {\\mathrm O}\_{2}\ \ \\mathsf{M e d i u m}^{\\star}\ \ L0W^{\\pm}\ \ L0W^{x}\ \ Lcos w^{\\pm}\ \ N\_\_x{x}\ \ - This term has the additional uncertainty of the derivation of an effective radiative forcing from a radiative forcing.\ \ | From sulfur emissions | Limited | Low | Very low | None available; few studies, probably a negative ERF | Not provided previously |\ | --- | --- | --- | --- | --- | --- |\ | From soot emissions | Limited | Low | Very low | None available; few studies, varying in sign and magnitude of ERF constrained by poor understanding of processes | Not provided previously |\ \ \\mathsf{E R F}\ \ 1296 \*\* This term differs from ‘Very High’ level in IPCC (2013) because additional uncertainties are introduced by the\ 1297 assessment of marginal aviation CO2 emissions and their resultant concentrations in the atmosphere from simplified\ 1298 carbon cycle models.\ \ \\mathrm{c O z}\ \ * * *\ \ 1303\ 1304\ 1305\ 1306\ \ a\ 1302 Table 4b. Basis for confidence levels in Table 4a\ \ 4a^{a}\ \ | MediumHigh agreementLimited evidence | HighHigh agreementMedium evidence | Very HighHigh agreementRobust evidence |\ | --- | --- | --- |\ | LowMedium agreementLimited evidence | MediumMedium agreementMedium evidence | HighMedium agreementRobust evidence |\ | Very LowLow agreementLimited evidence | LowLow agreementMedium evidence | MediumLow agreementRobust evidence |\ \ a\ The basis for the confidence level is given as a combination of evidence\ (limited, medium, robust) and agreement (low, medium and high) based\ on guidance given by Mastrandrea et al. (2011).\ \ * * *\ \ 1307\ 1308\ 1309\ \ Table 5. Emission metrics and corresponding CO2-equivalent emissions for the ERF components of 2018\ aviation emissions and cloudiness\ \ \ \\mathrm\ c O\_{2}\ \ 1310\ 1311\ \ \\mathsf{G W P}\_{20}\ \ \\mathsf{G W P}\_{50}\ \ | ERF term | GWP20 | GWP50 |\ | --- | --- | --- |\ | CO2 | 1 | 1 |\ | Contrail cirrus(Tg CO2 basis) | 2.32 | 1.09 |\ | Contrail cirrus(km basis) | 39 | 18 |\ | Net NOx | 619 | 205 |\ | Aerosol-radiation | | |\ | Soot emissions | 4288 | 2018 |\ | SO2 emissions | -832 | -392 |\ | Water vapor emissions | 0.22 | 0.10 |\ \ | GWP100 | GTP20 | GTP50 | GTP100 |\ | --- | --- | --- | --- |\ | 1 | 1 | 1 | 1 |\ | 0.63 | 0.67 | 0.11 | 0.09 |\ | 11 | 11 | 1.8 | 1.5 |\ | 114 | -222 | -69 | 13 |\ | 1166 | 1245 | 195 | 161 |\ | -226 | -241 | -38 | -31 |\ | 0.06 | 0.07 | 0.01 | 0.008 |\ \ \\mathsf{G M P}\_{100}\ \ \\tt{G T P}\_{50}\ \ \\mathrm{{c O\_\_{2}}}\ \ \\mathtt{G T P}\_{100}\ \ (\\mathsf{T g}C\\mathsfmathsf{O}\_{2},\\mathsf{b a s i s})\ \ \\mathrm{S O}\_{2}\ \ -1\ CO2-eq emissions (Tg CO2 yr) for 2018\ \ y r^{-1})\ \ \\mathbf{c{}{bf}}{{\\bf{}}\_{2}}\ \ c0\_{2}=e q\ \ | ERF term | GWP20 | GWP50 |\ | --- | --- | --- |\ | CO2 | 1034 | 1034 |\ | Contrail cirrus(Tg CO2 basis) | 2399 | 1129 |\ | Contrail cirrus(km basis) | 2395 | 1127 |\ | Net NOx | 887 | 293 |\ | Aerosol-radiation | | |\ | Soot emissions | 40 | 19 |\ | SO2 emissions | -310 | -146 |\ | Water vapor emissions | 83 | 39 |\ | Total CO2-eq(usingkm basis) | 4128 | 2366 |\ | Total CO2-eq/CO2 | 4.0 | 2.3 |\ \ | GWP100 | GTP20 | GTP50 | GTP100 | GWP\*100(E$^{+}$CO2e) |\ | --- | --- | --- | --- | --- |\ | 1034 | 1034 | 1034 | 1034 | 1034 |\ | 652 | 695 | 109 | 90 | 1834 |\ | 651 | 694 | 109 | 90 | 1834 |\ | 163 | -318 | -99 | 19 | 339 |\ | 11 | 12 | 2 | 2 | 20 |\ | -84 | -90 | -14 | -12 | -158 |\ | 23 | 27 | 4 | 3 | 42 |\ | 1797 | 1358 | 1035 | 1135 | 3111 |\ | 1.7 | 1.3 | 1.0 | 1.1 | 3.0 |\ \ \\mathsf{G W P}\_{50}\ \ \\tt{G T P}\_{20}\ \ \\mathsf{G W P}^{\*}{}\_{100}\ \ \\mathtt{G T P}\_{100}\ \ (\\mathsf{E}\_{\\mathsf{C o2e2}})\ \ \\mathrm{{c O\_{2}}}\ \ (\\mathsf{T g},{\\mathsf{C O}}\_{2},{\\mathsf{b a s i s}})\ \ \\mathrm{N O}\_{x}\ \ \ {mathrm C o o-e q}\ \ 1312\ \ \\mathsf{C O}e q,/\\mathsf{C O}\_{2}\ \ * * *\ \ 1 July 2020 Revised\ \ 1313 1314\ \ **Figure 1. Schematic overview of the processes by which aviation emissions and increased cirrus 1315**\ \ cloudiness affect the climate system. Net positive RF (warming) contributions arise from CO2, water 1316 vapor, NOx, and soot emissions, and from contrail cirrus (consisting of linear contrails and the cirrus 1317 cloudiness arising from them). Negative RF (cooling) contributions arise from sulfate aerosol production. 1318 Net warming from NOx emissions is a sum over warming (short-term ozone increase) and cooling 1319 (decreases in methane and stratospheric water vapor, and a long-term decrease in ozone) terms. Net 1320 warming from contrail cirrus is a sum over the day/night cycle. These contributions involve a large number 1321 of chemical, microphysical, transport and, radiative processes in the global atmosphere. The quantitative 1322 ERF values associated with these processes are shown in Figure 3 for 2018. 1323\ \ * * *\ \ 1 July 2020 Revised\ \ 1325 1326\ \ **Figure 2. Data related to the growth of aviation traffic and CO2** emissions from 1940 to 2018. Panel (a): 1327\ \ Global aviation CO2 emissions. Underlying fuel usage data for 1940 to 1970 are derived from Sausen and 1328 Schumann (2000) and for 1970–2016 from International Energy Agency (UKDS, 2016) data, which 1329 1330 include international bunker fuels. For 2017/18, the values are scaled from information from the International Air Transport Association (see Appendix A). The average annual increase of global 1331 emissions from 1960 to 2018 is 15 Tg CO2 yr -1 and the corresponding decadal average growth rates are 8.0, 1332\ \ 2.2, 3.0, 2.3 and 1.1% yr\ -1 , yielding an overall average of 3.3% yr -1 . Panel (b): Global aviation traffic in 1333 RPK and ASK from airlines.org ( [http://airlines.org/dataset/world-airlines-traffic-and-capacity/](http://airlines.org/dataset/world-airlines-traffic-and-capacity/)), and the 1334 transport efficiency of global aviation in kg CO2 per RPK. The passenger load factor defined as RPK/ASK 1335 increased from about 60% in 1960 to 82% in 2018. Panel (c): Total anthropogenic CO2 emissions and the 1336 aviation fractions of this total with and without the inclusion of CO2 emissions from land use change 1337 (LUC) from the Global Carbon Budget 2018 (Le Quéré et al., 2018). Panel (d)–(f): Additional aviation 1338 emissions data by region and year. The yearly sums of OECD and non-OECD values in (d) equal the 1339 respective global total values. The regional values in (e) and (f) also sum to equal the yearly global total 1340 values. Note different vertical scales. ( [http://www.oecd.org/about/membersandpartners/](http://www.oecd.org/about/membersandpartners/)) (UKDS, 2016) 1341 (Country listings in SD Spreadsheet). 1342\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure 3. Best-estimates for climate forcing terms from global aviation from 1940 to 2018. The bars and 1347**\ \ whiskers show ERF best estimates and the 5–95% confidence intervals, respectively. Red bars indicate 1348 1349 warming terms and blue bars indicate cooling terms. Numerical ERF and RF values are given in the columns with 5–95% confidence intervals along with ERF/RF ratios and confidence levels. ERF and RF 1350 values are shown for other years in Tables 2 and 3, Figure 6 and the SD spreadsheet. RF values are 1351 multiplied by the respective ERF/RF ratio to yield ERF values. ERF/RF values designated as \[1\] indicate 1352 that no estimate is available yet. The basis for confidence levels is presented in Table 4. 1353\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure 4. Results from an ensemble of 18 models from 20 studies for aviation NO** x impacts: short-term O3 1356\ \ increases; CH4 reductions, CH4-induced long-term reductions of O3, CH4-induced reductions of 1357 stratospheric water vapor (SWV) and Net NOx. Each data point represents a value of RF per unit emission 1358 (mW m -2 (Tg N yr -1 ) -1 ) as normalized from a published study (see SD). CH4-induced O3 and SWV are 1359 calculated using standardized methodology (see text for details). Note that the displayed values do not 1360 include correction factors to account for the non-steady-state CH4 responses to NOx emissions and the new 1361 CH4 RF parameterization. These adjustments are applied in forming the best estimates as discussed in 1362 Appendix D. 1363\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure 5. Summary of RF estimates for aerosol-cloud interactions for aviation aerosol as calculated in the 1366**\ \ SD spreadsheet for a variety of published results normalized to 2018 air traffic and 600 ppm fuel sulfur. 1367 The results are shown for soot; total particulate organic matter (POM), sulfate and ammonia (NH3); and 1368 sulfate aerosol from the indicated studies. The color shading gradient in the symbols indicates increasing 1369 positive or negative magnitudes. No best estimate was derived in the present study for any aerosol-cloud 1370 effect due to the large uncertainties. In previous studies, the estimates for the soot aerosol-cloud effect are 1371 associated with particularly large uncertainty in magnitude and uncertainty in the sign of the effect (Penner 1372 et al., 2009; Zhou and Penner, 2014; Penner et al., 2018). As part of the present study, an author (JEP) re-1373 evaluated these earlier studies and concluded that the Penner et al. (2018) results supersede the earlier 1374 Penner et al. (2009) and Zhou and Penner (2014) results because of assumptions regarding updraft 1375 velocities during cloud formation. In addition, a bounding sensitivity case in which all aviation soot acts as 1376 an IN in Penner et al. (2018) is not included here.\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure 6. Timeseries of calculated ERF values and confidence intervals for annual aviation forcing terms 1379**\ \ from 2000 to 2018. The top panel shows all ERF terms and the bottom panel shows only the NOx terms 1380 and net NOx ERF. All values are available in the SD spreadsheet, in Tables 2 and 3, and in Figure 3 for 1381 1382 2018 values. The net values are not arithmetic sums of the annual values because the net ERF, as shown in\ \ **Figure 3 for 2018, requires a Monte Carlo analysis that properly includes uncertainty distributions and 1383**\ \ correlations (see text). 1384\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure 7. Probability distribution functions (PDFs) for aviation ERFs in 2018 based on the results in 1387**\ \ **Figure 3 and Table 2. PDFs are shown for separately for CO2**, the sum of non-CO2 terms, and the net 1388\ \ aviation ERF. Since the area of each distribution is normalized to the same value, relative probabilities can 1389 be intercompared. Uncertainties are expressed by a distribution about the best-estimate value that is normal 1390 for CO2 and contrail cirrus, and lognormal for all other components. A one-million-point Monte Carlo 1391 simulation run was used to calculate all PDFs. 1392\ \ * * *\ \ 1 July 2020 Revised\ \ ## 1394 Appendices\ \ 1395 **A. Trends in aviation CO2 emissions**\ \ Global aviation CO2 emissions for 1940–1970 were taken from Sausen and Schumann (2000) and for the 1396 years 1971–2016 were calculated from International Energy Agency (IEA) data on usage of JET-A and 1397 aviation gasoline, largely from annual ‘Oil Information’ digests (e.g., [https://webstore.iea.org/oil-1398](https://webstore.iea.org/oil-1398) information-2019). The regional data are from the same source but accessed online from the IEA Oil 1399 Information (1960–2017) held at the UK Data Service (IEA, 2019). Note that these data are proprietary 1400 and must be purchased from IEA. Data were unavailable for 2017 and 2018, so incremental annual 1401 percentage increases in global aviation fuel usage and, therefore CO2 emissions, for those years were taken 1402 from reports of the International Air Transport Association (IATA, 2019). Some uncertainties exist from 1403 the annual fuel estimations and to a much smaller extent, the emission factors. The IEA does not give 1404 uncertainties for annual kerosene fuel sales or usage. Sausen and Schumann (2000), from which the 1940 1405 to 1970 data are based here, estimated that the uncertainty in cumulative fuel consumption from 1940 to 1406 1407 1995 (their dataset) is 20%. There is a known discrepancy of IEA estimates of aviation fuel usage being greater by about 10% than that derived from bottom-up global civil aviation inventories. Actual fuel usage 1408 is likely to be somewhere between the two estimates: aviation emissions inventories are known to be 1409 incomplete, with only scheduled traffic being available from some air traffic regions, and fuel usage 1410 potentially being underestimated from flight routing and cruise altitudes; IEA data on the other hand 1411 includes military aviation fuel (not included in civil aviation inventories) and a small fraction of kerosene 1412 not used in aviation, but sold for that purpose (L09). The CO2 emission factors for aviation fuel on the 1413 other hand are well determined, and the uncertainty is likely within 1%. 1414\ \ ## 1415 B. Aviation CO2 radiative forcings\ \ 1416 **Calculation of CO2 concentrations from emissions—LinClim SCM**\ \ The response of CO2 concentrations, C(t), to a CO2 aviation emissions rate, E(t), is modelled using the 1417 method described in Hasselmann et al., (1997) and is expressed as: 1418 _t_ ò _t_ 0 _C_ −=D _tEttGtC dt')'()'()(_ 1419 (B.1)\ \ 1420 where 5 −t / t _j_ )( =åa _jC_ _etG_ _j =0_ (B.2) 1421\ \ 1422 and t _j_ is the e-folding time of mode j and the equilibrium response of mode j to a unit emissions of a _j_ t _j_.\ \ The mode parameters used in this study are presented in Sausen and Schumann (2000) and approximate 1423 the carbon-cycle model in Meier-Reimer and Hasselmann (1987). The applicability of these parameters in 1424 the context of aviation response was tested in a model intercomparison exercise (Khodayari et al., 2013). 1425 1426 For the time horizon of 50-60 years into the future, these were found to compare well with other more sophisticated carbon-cycle models such as MAGICC 6.0, which is widely used in the IPCC Fourth 1427 Assessment Report (IPCC, 2007). Beyond this horizon, aviation CO2 concentrations begin to have an 1428 impact on the ocean and biosphere uptake of CO2 and the non-linearities of the system must be accounted 1429 for. 1430\ \ 1431 **Calculation of CO2 concentrations from emissions—CICERO-2 SCM**\ \ The CICERO-2 SCM (Fuglestvedt and Berntsen, 1999; Skeie et al., 2017) uses interconnected process-1432 specific IRFs with explicit treatment of air-sea and air-biosphere exchange of CO2 (Joos et al., 1996; 1433 Alfsen and Berntsen, 1999) that forms a nonlinear carbon cycle. The ocean and biosphere IRFs in 1434\ \ * * *\ \ 1441\ 1447\ 1454\ 1468\ \ CICERO-2 express how the CO2 impulse decays within each reservoir. The CO2 partial pressure in each 1435\ reservoir is calculated as a function of the carbon in that reservoir, and the CO2 partial pressure in each 1436\ reservoir is related to the CO2 partial pressure in the atmosphere by explicitly solving for the 1437\ atmosphere/ocean/biosphere CO2 mass transfer. Therefore, the CICERO-2 carbon cycle takes into account 1438\ the nonlinearity in ocean chemistry and biosphere uptake at high CO2 partial pressures since it represents 1439\ the atmospheric change in CO2 as a function of total background. 1440\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O}\_{2}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ Calculation of CO2 concentrations from emissions—FaIR SCM\ \ \\mathbf{c o}\_{2}\ \ The FaIR SCM is described by Millar et al. (2017) and summarized as follows. FaIR is a modified version 1442\ of the IPCC AR5 four time-constant impulse response function (IRF) model, which represents the 1443\ evolution of atmospheric CO2 by partitioning emissions of anthropogenic CO2 between four reservoirs of 1444\ an atmospheric CO2 concentrations change, following a pulse emission (see Myhre et al., 2013 for more 1445\ details). In more comprehensive models, ocean uptake efficiency declines with accumulated CO2 in ocean 1446\ sinks (Revelle and Suess, 1957) and uptake of carbon into both terrestrial and marine sinks are reduced by\ warming (Friedlingstein et al., 2006). FAIR captures some of these dynamics within the simple IRF 1448\ structure, mimicking the behaviour of Earth System Models/Earth System Models of Intermediate 1449\ Complexity in response to finite-amplitude CO2 injections; this is achieved by introducing a state-1450\ dependent carbon uptake with a single scaling factor, α, to all four of the time constants in the carbon cycle 1451\ of the IPCC AR5 impulse response model used for the calculation of CO2-equivalence metrics. This 1452\ approach is described in more detail by Millar et al. (2017). 1453\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ \\mathrm{C O\_{2}}\ \ C. Radiative forcing, efficacy and effective radiative forcing (ERF)\ \ Radiative forcing (RF) has been introduced as a predictor for the expected equilibrium global mean of the 1455\ (near) surface temperature change DTs that results from the introduction of climate forcers, such as 1456\ additional atmospheric CO2 or a change in the solar irradiation (e.g., IPCC, 2007): 1457\ \ \\mathrm{C O\_{2}}\ \ \\Delta\\mathrm{T\_{s}}\ \ \\Delta\\mathrm{T\_{s}}!=!\\lambda,\\mathrm{R F}\ \ -2-1\ where l is the climate sensitivity parameter (K (W m)). Several definitions of RF exist. According to 1459\ the simplest one, the instantaneous RF is the change in the total irradiation (incoming short-wave solar 1460\ radiation minus the outgoing long-wave terrestrial radiation) at the top of the atmosphere over the 1461\ industrial era. However, for most of the climate forcers a better definition (with respect to the linearity of 1462\ Eq. (C.1)) is the stratosphere-adjusted RF at the tropopause. Here, after the introduction of the new climate 1463\ forcer, the temperature of the stratosphere is allowed to reach a new radiative equilibrium, while all other 1464\ atmospheric state variables are kept constant. The stratosphere-adjusted RF at the tropopause was used in 1465\ many of the earlier IPCC reports (IPCC, 1999) and in earlier assessments of aviation climate impacts 1466\ (Sausen et al., 2005; L09). 1467\ \ (\\mathrm{K,(W,m^{-2})^{-1})}\ \ (C.2) 1474\ \ Here lCO2 is the climate sensitivity parameter for a CO2 perturbation. While l in (C.1) is considered a 1475\ universal constant, which can only be determined by climate models and hence is model dependent, li 1476\ depends on the type of forcing, as does ri. (While rCO2 is 1 by definition, rlinear contrails is <1 (Ponater, et al., 1477\ 2005; Rap et al., 2010)). Eq. (C.2) can also be expressed differently: 1478\ \ \\lambda\_{\\Omega2}\ \ \ \_\\mathrm{i,}}\ \ \\lambda\_{\\mathrm{i}}\ \ \ {\\mathrm{I}}\_{\ }}\ {\ \ \\mathrm{r\_{C O2}}\ \ \\mathrm{\_l i n e a r:c o n t r a i l s},\ \\mathrm{\_S}}&{{}<1\ \ T\_{s}= _{C02}R\\!{F_ _{{1}}^{\*}\ \ \ \ !!{i t h},!!!{!R F_{1}}^{ _}}^{_}\ {!{:}!{!}}^{\*}!{\ {!:}}^^{\*\*}\ \ * * *\ \ 1 July 2020 Revised\ \ Here RFi\* is the forcing modified by the efficacy, which yields a better approximation for the surface 1480 temperature change than RF. However, the calculation of the RFi\* is computationally much more 1481 expensive than the calculation of RF, as it requires the determination of the equilibrium temperature 1482 change, DTs, with a comprehensive climate model. 1483\ \ As an alternative, the effective radiative forcing (ERF) has been introduced as a more practical indicator of 1484 the eventual global mean temperature response (IPCC, 2013). While RFi\* assumes equilibrium climate 1485 change, ERF only includes all 'fast' atmospheric responses to a given climate forcer. For example, rapid 1486 adjustments in cloud cover, such as from aerosols, or in properties that respond to changes in water vapor, 1487 can either increase or decrease the initial RF. In contrast, the instantaneous, stratosphere-adjusted, and 1488 1489 effective RFs for well-mixed greenhouse gases are nearly equal. In practice, ERF is determined with a comprehensive climate model, which calculates a new equilibrium radiative imbalance, while the sea 1490 surface temperature and/or the global surface temperature is kept constant. As a consequence, an ERF 1491 value is expected to be somewhere between RF and RFi\* values and closer to RFi\* values. 1492\ \ ## 1493 D. Aviation NOx radiative forcings\ \ 1494 **Impacts of NOx emissions on ozone, methane and stratospheric water vapor**\ \ _**Model studies. In this ensemble analysis of the climate forcing from aviation NO**_ x emissions, the results of 1495 20 studies published since the IPCC (1999) aviation report were considered: IPCC (1999), Sausen et al. 1496 (2005), Stordal et al. (2006), Köhler et al. (2008), Hoor et al. (2009), Myhre et al. (2011), Frömming et al. 1497 (2012), Olivié et al. (2012), Gottschaldt et al. (2013), Köhler et al. (2013), Olsen et al. (2013), Skowron et 1498 al. (2013), Khodayari et al. (2014a), Khodayari et al. (2014b), Søvde et al. (2014), Skowron et al. (2015), 1499 Pitari et al. (2015), Kapadia et al. (2016), Pitari et al. (2017, Lund et al. (2017). Three studies that reported 1500 results from a 100-year integration of a pulse NOx emission (Wild et al. 2001, Derwent et al. 2001, 1501 1502 Stevenson et al. 2004) were not included in this analysis, nor has as Unger et al. (2010) which uses a different methodology to the aforementioned. 1503\ \ This model ensemble represents various methodologies in calculating and treating the long-term effects; in 1504 order to avoid gaps and additional uncertainties, standardized RFs for reductions in CH4-induced O3 and 1505 1506 SWV were adopted, except for one study that calculates the ‘real’ long-term effects from their 50-yr integrations (Pitari et al., 2017): 1507\ \ - All analyzed short-term O3 RFs account for a stratospheric adjustment: Assuming that it reduces the 1508 instantaneous RF by ~20% (Myhre et al., 2013, Stevenson et al., 1998), a factor of 0.8 was applied to 1509 any O3 RF that is an instantaneous RF (e.g., in the cases of Khodayari et al. (2014a,b) and Olsen et al. 1510 (2013)). 1511\ \ - Reductions in CH4-induced O3 and SWV are defined as 50% (Myhre et al., 2013) and 15% (Myhre et 1512 al., 2007) of reported CH4 RFs, respectively. This is applicable for studies that either originally did not 1513 provide CH4-induced O3 and SWV estimates (e.g., IPCC, 1999, Sausen et al., 2005, Olsen et al., 2013) 1514 or derived these RFs using another assumptions (e.g., Stordal et al., 2006, Köhler et al., 2008, Hoor et 1515 al., 2009, Gottschaldt et al., 2013, Köhler et al., 2013, Skowron et al., 2013, Khodayari et al., 2014a). 1516\ Further assumptions regarding data treatment are: 1517\ \ - Frömming et al. (2012), Olivié et al. (2012), Khodayari et al. (2014b) and Kapadia et al. (2016) 1518 provide the short-term O3 RFs only and p-TOMCAT in Stordal et al. (2006) calculates just the long-1519 term effects; thus, these numbers are included in the respective NOx variable analysis but do not 1520 contribute to the net NOx estimate. 1521\ \ - Whenever the same estimate appears repetitively in subsequent studies, it is treated as a single entry: 1522 this is the case for CAM4 short-term O3 RF that appears in Khodayari et al. (2014a; b) and Olsen et al. 1523\ \ \ * * *\ \ 1557\ 1562\ \ (2013), CAM5 short-term O3 RF that can be found in Khodayari et al. (2014a; b) and NASA ModelE2 1524\ short-term O3 and CH4 RFs presented by Unger et al. (2013) and Olsen et al. (2013). 1525\ \ \ mathrm O{}\_{3}\ \ \ \_33\ \ \\mathrm{C H}\_{4}\ \ In addition, the ERF estimates for the CH4 term include shortwave RF (Etminan et al., 2016). The 1526\ inclusion of shortwave forcing in the simplified expression increases CH4 RF from aviation NOx emissions 1527\ by 23% (based on MOZART-3 CTM runs driven for all the aircraft emission inventories represented in the 1528\ model ensemble) (Table D.1). 1529\ \ \\mathrm{C H\_{4}}\ \ \\mathrm{N O\_{x}}\ \ \\mathrm{C H\_{4}}\ \ Ensemble values. This ensemble analysis covers a period of almost two decades; however, none of the RF 1530\ per unit of emitted N estimates show any trends over time of publication and the spread in RF per unit of 1531\ -2-1-1\ emitted N values has not changed. The short-term O3 RF varies from 6.2 to 45.1 mW m (Tg (N) yr), 1532\ where these values come from the NASA ModelE2 (Olsen et al., 2013) and p-TOMCAT (Hoor et al., 1533\ -2-1-1\ 2009) models, respectively. The long-term CH4 RF varies from -27.9 to -8.1 mW m (Tg (N) yr), from 1534\ the p-TOMCAT (Köhler et al., 2008) and MOZART3 (Skowron et al., 2015) models, respectively. The 1535\ spread of other CH4-induced long-term effects follows that of CH4. The net-NOx RF varies from -17.5 to 1536\ -2-1-1\ 11.9 mW m (Tg (N) yr) from ECHAM/MESSy (Gottschaldt et al., 2013) and CAM4 (Khodayari et al., 1537\ 2014a), respectively. The results from the mid-1990s CTMs are within the envelope of RFs generated 1538\ more recently (Figure 3). The numbers from IPCC (1999) and related studies, Sausen et al. (2005) and 1539\ L09, where the non-CO2 effects were originally calibrated to the results from IPCC (1999), do not alter the 1540\ best NOx RF values and their uncertainties (Table D.2). 1541\ \ 45.1,\\mathrm{m W},\\mathrm{m}^{2},((\\mathrm{T g},(\\mathrm{N}),\\mathrm{y},\\mathrm{r}^{1})^{-1}\ \ \\up{{\ }O\_{3}}\ \ \\mathrm{C H\_{4}}\ \ \ 18.1,\\mathfrak{m W}m^{,2}(\\mathrm{T g},(\\mathrm{N}),\\mathfrak{y r}^{,1})^{\\cdot1};\ \ \\mathbf{C H}\_{4}\ \ 11.9,\\mathrm{m{}}\\mathrm{{W},m^{2},(\\mathrm{{T g}\ N,r r^{1})}^{-1}}\ \ \\mathrm{-N O\_{x}}\ \ \\mathrm{n o n o O\_{2}}\ \ \\mathrm{N O\_{X}}\ \ Correlations. The correlations between the NOx RF components are shown in Figure D.1. In addition to 1542\ the significant negative correlations between the short-term and the long-term aviation RF components, 1543\ correlations between the net-NOx effect and its components are also apparent, especially for the short-term 1544\ O3 and net-NOx components; however, their strength is around half. The high correlations (p=1, R2=1) 1545\ across the long-term effects is expected since CH4-induced O3 and SWV are all derived based on CH4 RFs. 1546\ -2-1-1\ In units of mW m (Tg(N yr), 49% of this ensemble short-term O3 RF is concentrated between 20 and 1547\ 35, 43% of CH4 RFs is found between -14 and -10, 41% of CH4-induced O3 RFs is between -7 and -5 and 1548\ 45% of SWV RFs vary from -2.5 to -1.5. Of the normalized net-NOx RFs resulting from this ensemble, 1549\ -2-1-1\ 44% are observed between 5 and 10 mW m (Tg(N) yr). 1550\ \ \\mathrm{N O\_{x}}\ \ \\mathrm{n e t N0}\ \ ({\\tt p}{=}1,{\\tt R}^{2}{=}1)\ \ \ mathrm O{{}}\_333\ \ \\mathrm{C H\_{4}}\ \ \ \_33\ \ \\mathrm{C H}\_{4}\ \ \\mathbf{O}\_{3}\ \ \\mathfrak{m W},\\mathfrak{m}^{2},(\\mathrm{T g}(mathrm N,,\\mathrm{y r}^{1})^{-1}\ \ \\mathrm{C H\_{4}i n d d c}\ \ \ \\mathbf{O}\_{3}\ \ \\mathrm{-N O\_{x}}\ \ 10,\\mathrm{{m W},\\mathfrak{m}^{22},(\\mathrm{{T g}(N),y r^{-})^{-1}}}\ \ ^5\ \ Transient vs. equilibrium. In calculating the CH4 RF response to aviation NOx emissions, the lack of steady-1551\ state conditions is an important consideration. Since methane (CH4) has a lifetime of the order 8–12 years 1552\ (largely model-dependent) any NOx perturbation takes on the order ~40 years to come within 2% of the 1553\ steady state solution. Moreover, the timescale of removal of CH4 from the atmosphere is made longer 1554\ through a positive chemical feedback (Prather, 1994). In order to overcome the necessity to run a global 1555\ chemical transport model (CTM) with full chemistry for such long integrations, a parameterization to 1556\ account for this perturbation was originally developed by Fuglestvedt et al. (1999) and has been widely\ adopted since then. However, with the significant annual increases in aviation NOx emissions over the last 1558\ several decades (Figure D.2a) the CH4 response does not reach its steady-state value in any given year of 1559\ emissions, so the steady-state solution generally overstates the CH4 response in a particular year from 1560\ historical time-evolving emissions. Similar considerations apply to other sectors with substantial NOx 1561\ emissions such as shipping (Myhre et al., 2011). If steady-state conditions are utilized, there is a\ conceptual and quantitative mismatch when comparing the NOx RF from aviation with other RF terms, 1563\ since RF represents a particular condition at a point in time, not the steady-state conditions. To remedy this 1564\ mismatch, Myhre et al. (2011) suggested that a factor accounting for the non-steady-state condition of CH4 1565\ be introduced, thereby modifying the CH4 impact for a given year of interest, and further suggested that for 1566\ the aviation RF in the year 2000 the CH4 term be reduced by approximately 35% for aircraft emissions 1567\ using a simplified estimation derived from Grewe and Stenke (2008). 1568\ \ \\mathrm{N O\_{x}}\ \ \\mathrm{C H}\_{4},\\mathrm{R F}\ \ \ \\mathrm(\\mathrm H\_{4})\ \ \\mathrm{C H\_{4}}\ \ \\mathrm{N O\_{x}}\ \ \\mathbf{C H}\_{4}\ \ \\mathrm{H}\_{4}\ \ \\mathrm{C H\_{4}}\ \ * * *\ \ 1 July 2020 Revised\ \ TROPOS 2D CTM. The results of the steady-state runs using constant emissions for a given year are 1571 compared with those of transient runs using background historical surface emissions from anthropogenic 1572 activities and the corresponding aviation NOx emissions. The latter requires full implementation of time-1573 varying CH4 emissions into the model simulation, a requirement that is not a standard set-up for many of 1574 the CTM/GCMs currently in use where CH4 conditions are defined from observations as fixed 1575 concentrations with relaxation terms introduced to accommodate perturbations to these concentrations. The 1576 use of CTM runs explicitly accounts for changing background atmospheric conditions over the integration 1577 period as well as the change in emission rate dependence of the O3 and CH4 responses. 1578\ \ _Method. In order to compare these two methods, two types of experiments were performed: 1579_\ \ - Transient experiment: a long-term simulation with anthropogenic (surface and aviation) emissions 1580 evolving over time covering the period 1950–2050, using historical data up to 2000 and the RCP-4.5 1581 scenario after 2000 (Figure D.2a), 1582\ 1583 • Steady-state experiment: a 100-year simulation with constant anthropogenic (surface and aviation) emissions representing the year 2000, 2018 or 2050 (Figure D.2a); the steady-state CH4 response starts 1584 to be observed 60–70 years into the run. 1585\ \ Each of these experiments was run twice, with and without aviation emissions, and the difference between 1586 these two results defined as the aircraft response (e.g., Figure D.2d-f). The initial concentrations of CH4 1587 were set using the observations from NOAA surface stations (Montzka et al., 2000) for 1950 and 2000; for 1588 the year 2050 the CH4 concentrations are taken from projections of the MAGICC model (Meinshausen et 1589 al., 2011). The background anthropogenic emissions of CO, CH4, NOx, N2O, and non-methane volatile 1590 organic carbon (NMVOC) compounds, as well as aircraft NOx emissions, evolve during the period 1950-1591 2050 (Lamarque et al., 2010; Clarke et al., 2007) (Figure D.2a). The natural emissions from soils and 1592 oceans were kept constant and represent the year 2000 (Prather et al., 2001). 1593\ \ The TROPOS CTM is a latitudinally-averaged, two-dimensional Eulerian global tropospheric chemistry 1594 model extensively evaluated by Hough (1989; 1991). The model’s domain extends from pole-to-pole (24 1595 latitudinal grid cells) and from the surface to an altitude of 24 km (12 vertical layers). TROPOS is driven 1596 by chemistry, emissions, transport, removal processes and upper boundary conditions. There are 56 1597 chemical species in the chemical mechanism of the model, which consists of 91 thermal reactions, 27 1598 photolytic reactions and 7 more reactions, which include night-time NO3 chemistry. The reaction rates and 1599 cross sections were updated to the evaluation of Sander et al. (2006) (see Skowron et. al, 2009). There are 1600 no fixed concentrations within the model domain other than the upper boundary conditions, which are 1601 1602 specified for long-lived species and for gases that have stratospheric sources. This 2D CTM has the disadvantage of zonal symmetry but has the advantage of an adequate chemical scheme and computational 1603 efficiency, such that long-term integrations can be reasonably performed. Owing to the aforementioned 1604 reasons, the O3 response in TROPOS is overestimated by a factor of ~2 by comparison with a range of up-1605 to-date 3D models. As a consequence, the CH4 results in Figures. D.2d-f were reduced accordingly. This 1606 modification of the original TROPOS responses does not affect the core result of this study, which is the 1607 _relative difference of CH_ 4 responses between transient and equilibrium methods. 1608\ \ _Results. Figure D.2b shows the evolution of the global CH4_ burden over the period 1950–2050 in the 1609 transient TROPOS simulation. There is a steady growth in the atmospheric CH4 burden, with a small 1610 1611 decline over the period 1997–2007 in response to the decrease in CH4 emissions over the period 1990–\ \ 2000. The steady-state simulations for the year 2000 and 2050 agree well (within 1%) with transient CH4 1612 responses for the respective years. A similar agreement is observed for modelled transient and steady-state 1613 CH4 lifetimes in Figure D.2c. Most of the CH4 loss in the atmosphere is driven by OH and the oxidative 1614 capacity of the atmosphere changes over time (thus CH4 lifetime as well), influenced by emissions of CO, 1615 NOx, NMVOC or CH4. 1616\ \ * * *\ \ 1 July 2020 Revised\ \ **Figure D.2c shows the evolution of global CH4** lifetime (LT) over the period 1950–2050: there is a 1617 decrease in the CH4 lifetime between 1950 and 2000 (until around 2007), whilst under the RCP-4.5 1618 scenario the opposite is observed, with the CH4 lifetime increasing by 3.5% by the end of 2050 compared 1619 with 2000. The TROPOS CH4 lifetimes agree relatively well with other studies (e.g., Holmes et al., 2013; 1620 1621 Voulgarakis et al.; 2013, Dalsøren et al., 2016) not only in terms of absolute numbers but also the rate of changes; a detailed comparison is presented in Table D.3. The perturbation lifetime of CH4 in TROPOS is 1622 37% longer than its global lifetime and the sensitivity coefficient s = ∂ln(LT) / ∂ln(CH4) is 0.27, placing 1623 these estimates in the middle of model ranges (e.g., Prather 2001, Holmes et al. 2011). These terms were 1624 calculated using a 5% increase of CH4 global levels for the year 2000. There is no need to apply the 1625 feedback factor (1.37) to the TROPOS CH4 estimates as it is already included in the observed responses; 1626 TROPOS does not have a fixed boundary conditions, so CH4 and OH can freely interact. 1627\ \ Aircraft NOx emissions, via the chemical coupling to OH and HO2, enhance OH, which reduces the global 1628 CH4 lifetime. Figure D.2d shows the evolution of the CH4 lifetime reduction in the transient 1950–2050 1629 1630 simulation and in steady-state runs for conditions representing the years 2000 and 2050. In the transient run, there is a steady decrease of global CH4 lifetime as a consequence of a constant increase of aviation 1631 NOx emissions during the period 1950–2050. The agreement in 2000 and 2050 between the transient and 1632 steady-state CH4 lifetime reductions is within 6% (on a global scale) (see Table D.3). These relatively 1633 small differences in CH4 lifetime lead to much more pronounced differences in the associated global CH4 1634 burdens as shown in Figure D.2e. In contrast to the lifetime results, the CH4 burden response in the 1635 transient run lags behind the steady-state CH4 response with differences of 27% in the year 2000 and 20% 1636 in the year 2050. Similarly, the calculations for 2018 emissions yield a multiplicative correction factor of 1637\ \ 0.79 (Figure D.2f), which has been incorporated into the ERF values of CH4, long-term O3 and SWV 1638 shown in Figure 5. 1639 The CH4 results contrast with O3 changes from aircraft NOx emissions, which agree within 3% between 1640 transient and steady-state experiments with aircraft O3 burdens of 10.3 and 10.6 Tg (O3), respectively, in 1641 the year 2000. These TROPOS O3 magnitudes are at the upper limit of model ranges, as present-day 1642 aircraft O3 perturbations found in the literature vary from 3 to 11 Tg (O3) (e.g., Hoor et al., 2009; Holmes 1643 et al., 2011; Khodayari et al., 2014a). The aircraft O3 burden increases by 41% in 2050, reaching 17.2 and 1644\ 18.0 Tg(O3) for transient and steady-state experiments, respectively. This agrees with other studies (e.g., 1645 Olsen et al., 2013) that report a multi-model average increase of 44% in O3 burden from future aircraft 1646 NOx emissions under the RCP-4.5 scenario. 1647 The present approach is in general agreement with that presented by Grewe and Stenke (2008), which 1648 accounts for CH4 concentrations not being in steady-state with OH changes in the year of simulation. The 1649 present CTM results further demonstrate the importance of explicitly calculating CH4 changes in response 1650 to time-dependent aviation NOx emissions rather than assuming constant emissions. The difference 1651 between transient and steady-state CH4 for the year 2000 found with TROPOS is smaller than that 1652 resulting from the Grewe and Stenke (2008) approach (Myhre et al., 2011) (27% and 35%, respectively). 1653 **Table D.4 presents a further comparison of CH4** correction factors derived in this study. The systematic 1654 differences are likely due to the Grewe and Stenke (2008) values being based on a simplified 1655 chemistry/climate model (AirClim) and the present TROPOS simulations having a different experimental 1656 setup (all our emissions (surface + aircraft) are time-varying) and a full chemical reaction scheme with 1657 explicit calculations performed on time-varying emissions. Indeed, if TROPOS is run with constant 1658 background emissions representing the year 2000 in a similar manner using Grewe and Stenke (2008) 1659 methodology, the difference between transient and steady-state CH4 for the year 2000 increases from 27% 1660 to 31%. This change shows that background emissions modify the CH4 correction factor and further 1661 emphasizes the need to have surface and aircraft emissions that simultaneously follow historical pathways. 1662 In other studies using the Grewe and Stenke (2008) methodology, CH4 correction factors vary from 0.74 to 1663\ 1.15 depending on the investigated year (2025 or 2050) and aircraft emission scenario (SRES A1B, B1 and 1664\ \ * * *\ \ 1674\ \ B1 ACARE) (the factor can be larger than 1 if the aircraft emissions are assumed to decrease in the 1665\ preceding years) (Hodnebrog et al., 2011; 2012). 1666\ \ Uncertainties in the CH4 correction factor are associated mainly with inter-model differences and the 1667\ applied emission scenarios; the correction factor is sensitive, within ~10%, to inter-model differences 1668\ (based on two models, TROPOS and AirClim) and it can vary by another ± 10% depending on emission 1669\ scenario (based on a range of RCP projections up to 2050). Given that the uncertainties of the CH4 1670\ correction factor on the net-NOx RF are rather small, especially when compared with overall uncertainties, 1671\ we do not include in the estimated uncertainty of the net-NOx RF value a separate uncertainty due to the 1672\ correction factor. 1673\ \ \\mathrm{C H\_{4}}\ \ \\mathrm{C H\_{4}}\ \ \\mathtt{n e t N O}\_{x}\ \ \\mathrm{-N O\_{x}}\ \ E. Contrail cirrus\ \ The global contrail cirrus RF is calculated by homogenizing existing estimates through the use of specific 1675\ scaling factors. The factors relate to the choice of air traffic inventory and its basis year; the use of the full 1676\ 3D flight distance; the use of hourly air traffic data; the feedback of natural clouds; and correcting for 1677\ weaknesses in the radiative transfer calculations. The corrections and scaling actions are: 1678\ \ • The estimate of Chen and Gettelman (2013) was corrected by redoing the CAM simulation using a 1679\ lower ice crystal radius of 7 µm and a larger contrail cross-sectional area of 0.09 km2 for the 1680\ initialization of contrails at an age of about 15–20 minutes, in agreement with observations (Schumann 1681\ et al., 2017. The resulting change in cirrus cloudiness including the adjustment in cloudiness due to the 1682\ -2\ presence of contrail cirrus leads to a radiative forcing of 57 mW m. 1683\ \ \\mathrm{m}^{-2}\ \ • A scaling S1 of 1.4 is applied for estimates based on the AERO2k inventory for the year 2002 instead 1684\ of the AEDT inventory for the year 2006 (Bock and Burkhardt, 2016); 1685\ \ \\mathrm{S}\_{1}\ \ • A scaling S2 of 1.14 is applied to estimates that are based on track distance instead of slant distance 1686\ (Bock and Burkhardt, 2016). The ‘slant’ air traffic distance is the full flight distance and not the ground 1687\ projected ‘track’ distance. 1688\ \ • A scaling S3 of 0.87 is applied to estimates that used monthly instead of hourly resolved air traffic 1689\ data. This scaling is based on an estimate for the impact of the temporal resolution of the air traffic data 1690\ of -25% to -30% within CAM (Chen et al., 2012) and one of no significant change in ECHAM4-1691\ CCMod. 1692\ \ \\mathrm{S{3}}\ \ • A scaling S4 of 1.15 is applied to account for the underestimation of RF in radiative transfer 1693\ calculations that use frequency bands instead of line by line calculations (Myhre et al. 2009). 1694\ \ \\mathrm{S{}4}\ \ The statistical uncertainty of global contrail cirrus RF cannot be estimated from the small number of 1702\ available studies. Uncertainties affecting our contrail cirrus estimates are, on the one hand, due to (A) 1703\ uncertainties in the radiative response to the presence of contrail cirrus and, on the other hand, (B) 1704\ uncertainties in the upper tropospheric water budget and the contrail cirrus scheme. In most cases, we can 1705\ only infer very rough estimates for the uncertainties related to specific processes. 1706\ \ \\mathrm{\_I1-\_22}\ \ (A) Uncertainties associated with the radiative response to contrail cirrus are: 1707\ \ \\mathrm{S\_{3}S-\_\_{4}}\ \ \ 76,\\mathrm{m W},\\mathrm{m}^{\ -}\ \ * * *\ \ 1 July 2020 Revised\ \ A1. Uncertainty related to the model’s radiative transfer scheme of approximately 35% (Myhre et al., 1708\ \ 2009). 1709 A2. Uncertainty in the inhomogeneity of ice clouds within a grid box of a climate model (Carlin et al., 1710 2002; Pomroy and Illingworth, 2000), the vertical cloud overlap, and the use of plane parallel geometry 1711 as compared to full 3D radiative transfer (Gounou and Hogan, 2007), which together amount to 1712 approximately 35%. 1713 A3. Uncertainty estimating radiative transfer in a global climate model in the presence of very small ice 1714 crystals within young contrails, which may amount to about 10% (Bock and Burkhardt, 2016). The 1715 uncertainty is dependent on the contrail cirrus ice water content. 1716 A4. Uncertainty due to the ice crystal habit is approximately 20% according to Markowicz and Witek 1717 (2011). 1718 A5. Uncertainty in the radiative transfer due to soot cores within the contrail cirrus ice crystals is 1719 thought to be large, as the change in the shortwave (SW) albedo is large (Liou et al., 2013). The soot 1720 impact on contrail cirrus RF has not yet been quantified. 1721\ Overall, uncertainty in the radiative response to contrail cirrus (excluding A3) is estimated to be about 1722 55%, assuming independence of different uncertainties and excluding the impact of ice crystal soot cores. 1723 The uncertainty A3 is included in the uncertainty estimate under (B) because A3 and B2 are dependent 1724 uncertainties. 1725\ \ (B) Uncertainty in contrail cirrus RF associated with the upper-tropospheric water budget and the contrail 1726 cirrus scheme are: 1727 B1. Uncertainty in contrail cirrus RF associated with the uncertainty in upper-tropospheric ice 1728 supersaturation. This results from a lack of knowledge in ambient conditions due to the low vertical 1729 resolution of satellite instruments (Lamquin et al., 2012) and to the ability of models to reproduce the 1730 observed statistics of ice supersaturation. This contributes about 20% to uncertainty. 1731 B2. There is uncertainty related to ice crystal number densities within young contrails. Ice nucleation 1732 within the plume can vary drastically depending on the water supersaturation reached within the plume 1733 and on the soot emissions (Kärcher et al., 2015; 2018). This dependency on the atmospheric state leads 1734 to a reduction in the number of nucleated ice crystals in particular in the tropics and at lower flight 1735 levels (Bier and Burkhardt, 2019) leading to a large uncertainty in the impact of tropical and subtropical 1736 air traffic. Depending on the atmospheric state and ice crystal numbers, a varying fraction of ice crystals 1737 can be lost in the contrail vortex phase (Unterstrasser, 2014). We assume an uncertainty in average 1738 contrail ice crystal numbers after the vortex phase of about 50% leading to an uncertainty in contrail 1739 cirrus RF of about 20%. This estimate of the sensitivity of contrail cirrus RF to ice crystal numbers in 1740 newly formed contrails is based on simulations with ECHAM5-CCMod (Burkhardt et al., 2018). 1741 B3. The uncertainty in the lifetime of contrail cirrus, affecting the day-/night-time contrail cover, has 1742 only a small impact on the estimated contrail cirrus RF (Chen and Gettelman, 2013; Newinger and 1743 Burkhardt, 2012). We estimate the associated uncertainty to be 5–10%. 1744 B4. From the sensitivity of the contrail cirrus RF to the temporal resolution in the air traffic dataset in 1745 ECHAM5 and CAM, we deduce an uncertainty of about 10%. 1746 B5. The estimate of the feedback of natural clouds, due to contrail cirrus changing the water and heat 1747 budget of the upper troposphere, is very uncertain and has not been properly quantified yet (Burkhardt 1748 and Kärcher, 2011; Schumann et al., 2015). We assume here the uncertainty related to this estimate to 1749 be only slightly smaller than the estimate itself, or about 15%. 1750\ \ * * *\ \ 1 July 2020 Revised\ \ B6. Uncertainty in the RF estimate of Chen and Gettelman (2013) to assumptions in the initial ice-1751 crystal radii and contrail cross-sectional areas is about 33%. 1752\ \ 1753 We assume independence of the uncertainties except for the dependence of A3 and B3 on the uncertainty in B2. The overall uncertainty due to the water budget and the contrail cirrus scheme (including 1754 uncertainty A3) is about 40% and more than 50% in the case of the Chen and Gettelman (2013). From the 1755 two different sources of uncertainty (list A, radiative, and list B, contrail cirrus properties, above) we 1756 calculate an overall contrail cirrus RF uncertainty of about 70%, assuming independence of the overall 1757 uncertainties described in A and B. 1758\ \ Note that we do not attempt to infer an estimate for the uncertainty of the factor ERF/RF. When 1759 calculating the contrail cirrus ERF, the error range given refers to the error range of contrail cirrus RF and 1760 not ERF. 1761\ \ ## 1762 F. Emission metrics calculations\ \ We calculate the AGWP and AGTP, and corresponding GWPs and GTPs, for aviation CO2, NOx (which 1763 encompasses the ERF of short-term O3, CH4, CH4-induced O3 and SWV), soot, SO2, and contrail cirrus. 1764 The methodology and analytical expressions for the emissions metrics are described in detail in previous 1765 literature (e.g., Fuglestvedt et al. 2010; Myhre et al. 2013). The impulse response function (IRF) that 1766 describes the atmospheric decay of CO2 upon emission is taken from Joos et al. (2013). For the other 1767 species, the atmospheric decay is given by a constant e-folding time taken as the ‘perturbation lifetime’. 1768 The lifetimes used here are broadly consistent with Fuglestvedt et al. (2010). The radiative efficiency (RE) 1769 for CO2 is calculated using year 2018 background concentrations of 407 ppm (annual mean, from monthly 1770 mean observed concentrations from NOAA GMD - 1771 [ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2\_mm\_gl.txt](ftp://aftp.cmdl.noaa.gov/products/trends/co2/co2_mm_gl.txt)). This yields a RE of 1.68 x 10 -15 W m -2 kg -1\ \ ), 1772\ 4% lower than used in the IPCC Fifth Assessment report (AR5) (Myhre et al., 2013). The climate response 1773 IRF is taken from Boucher and Reddy (2008). The latter has an inherent equilibrium climate sensitivity 1774 (ECS) of 1.06K (W m -2 ) -1 , equivalent to a 3.9K equilibrium response to a doubling of CO2. 1775\ \ For the calculation of the average rate of CO2-warming-equivalent emissions for aviation non-CO2 forcings 1776 (ECO2e\*) under the GWP\* metric in Table 5, we use the relationship between recent changes in effective 1777 RF and CO2-equivalent emissions from Allen et al. (2018) (or Equation (1) with 𝛼 = 0), 1778\ \ ECO2e\* = \[DF / Dt\] ´ \[H / AGWPH(CO2)\] (F.1) 1779\ \ where DF is the change in ERF over the recent period, Dt, and AGWPH(CO2) is the absolute global warming 1780 potential of CO2 at time horizon H. We use updated AGWPH(CO2) values incorporating the updated 1781 radiative efficiency of CO2 as described in the previous paragraph. Allen et al. (2018) used a backward-1782 looking period of 20 years as Dt, whereas here we use a backward-looking 18-yr period as our time series 1783 of ERF components only extends back to 2000. 1784\ \ ## G. List of Acronyms and abbreviations used in tables and figures of the Appendices 1785\ \ ACARE—Advisory Council for Aeronautical Research in Europe 1786 ACCMIP—Atmospheric Chemistry and Climate Model Intercomparison Project 1787 AEDT—Aviation Environmental Design Tool 1788 AEM—Advanced Emission Model 1789 AERO2K—Global aircraft emissions data project for climate impacts evaluation 1790 AGAGE—Advanced Global Atmospheric Gases Experiment 1791 CAM—Community Atmosphere Model 1792 CCMod—Contrail Cirrus Module 1793 CH3CCl3—Methyl chloroform 1794 COCIP—Contrail Cirrus Prediction Tool 1795\ \ * * *\ \ 1 July 2020 Revised\ \ CTM—Chemical Transport Model 1796 ECHAM—European Centre/Hamburg Model 1797 IPCC—Intergovernmental Panel on Climate Change 1798 MAGICC—Model for the Assessment of Greenhouse Gas Induced Climate Change 1799 MOZART—Model for OZone And Related chemical Tracers 1800 NOAA—National Oceanic and Atmospheric Administration 1801 QUANTIFY—Quantifying the Climate Impact of Global and European Transport System 1802 REACT4C—Reducing Emissions from Aviation by Changing Trajectories for the benefit of Climate 1803 RCP—Representative Concentration Pathway 1804 SRES—Special Report on Emission Scenarios 1805 TAR—Third Assessment Report 1806 1807 TRADEOFF—Aircraft emissions: contribution of different climate components to changes in radiative forcing–tradeoff to reduce atmospheric impact 1808 TROPOS—2D global TROPOSpheric model 1809 WDCGG—World Data Centre for Greenhouse Gases 1810\ \ * * *\ \ 1813\ 1814\ 1815\ 1816\ 1817\ 1818\ 1819\ 1820\ \ 1811\ \ Table D.1. The CH4 RFs derived for all the aircraft emission\ a\ inventories that are present in the model ensemble. 1812\ \ | Inventories |\ | --- |\ | AEDT |\ | AEM |\ | AERO2K |\ | REACT4C |\ | QUANTIFY |\ | TRADEOFF |\ \ | CH4RF，mWm-2 | |\ | --- | --- |\ | Old | New |\ | -6.67 | -8.22 |\ | -6.82 | -8.41 |\ | -7.09 | -8.74 |\ | -6.97 | -8.59 |\ | -6.96 | -8.58 |\ | -7.11 | -8.76 |\ \ 1821\ 1822\ \ \\mathrm{C H}\_{4}\ \ a\ Values are those represented in the model ensemble based on MOZART-3\ CTM simulations (Old) and recalculated values using a revised simplified\ expression for the CH4 RF (New) as presented by Etminan et al. (2016). The\ NOx emissions of each inventory are normalized so that all RFs are scaled to\ -1\ the same global total emissions (0.71 Tg(N) yr) as in the REACT4C model.\ \ \\mathrm{C H\_{4}}\ \ \\mathrm{N O}\_{x}\ \ y r^{-1})\ \ Table D.2. The best NOx RFs per unit emission derived for datasets that include and exclude late 1990s\ numbers and related estimates, see text for details.\ \ \\mathrm{N O\_{x}}\ \ | Components | Value | Uncertainty\* | Value | Uncertainty\* |\ | --- | --- | --- | --- | --- |\ | (mW m-2(Tg(N)yr-1)-1) | | | | |\ | with IPCC(1999) | | without IPCC(1999) | | |\ | Short-termO3 | 25.6 | ±7.3 | 25.1 | ±7.2 |\ | CH4 | -13.8 | ±4.7 | -13.4 | ±4.5 |\ | CH4-inducedO3 | -6.9 | ±2.3 | -6.7 | ±2.3 |\ | SWV | -2.1 | ±0.7 | -2.0 | ±0.7 |\ | NetNOx | 3.9 | ±5.7 | 4.0 | ±5.8 |\ \ (m W^{-}2T g\\left(N\\right)y r^{-1})^{-1}\ \ \ mathrm O{{}}\_33{\ \ \\pm7.2\ \ \\pm7.3\ \ \ mathbf O{{}\_\_{{}3}}\ \ \\pm4.7\ \ \\pm4.5\ \ \\pm2.3\ \ \\pm2.3\ \ \\pm0.7\ \ \\mathrm{N O}\_{x}\ \ \\pm5.7\ \ \\pm5.8\ \ \*Stated uncertainties are one standard deviation (68% confidence interval).\ \ * * *\ \ | Variable | Year | 2D CTM,TROPOS | |\ | --- | --- | --- | --- |\ | Transient | Steady statea | | |\ | CH4 burden,Tg | 2000 | 4770.8 | 4785.1 |\ | 2050 | 5051.6 | 5081.4 | |\ | CH4 abundance,ppb | 2000 | 1784.2 | 1787.5 |\ | 2050 | 1886.2 | 1897.6 | |\ | CH4 lifetime(TCH4+OH)b,yr | 2000 | 10.6 | 10.5 |\ | 2050 | 11.0 | 11.0 | |\ | aircraftCH4 lifetime(TCH4+OH),yr | 2000 | -0.137 | -0.145 |\ | 2050 | -0.293 | -0.311 | |\ \ | Literature | | | |\ | --- | --- | --- | --- |\ | Study | Ref | Model/Years | Variable estimate/change |\ | IPCC TAR | | 1998 | 4850 Tg |\ | Voulgarakis et al 2013 | | ACCMIP | 4750dTg |\ | Dalsøren et al 2016 | | Oslo CTM3 | 4560dTg |\ | Dalsøren et al 2016 | | 1970-2012 | +15% |\ | This studyc | | | +13% |\ | Voulgarakis et al 2013 | | ACCMIP | 5000dTg |\ | Voulgarakis et al 2013 | | | +5.3d% |\ | This studyc | | 2000-2050 | +5.9% |\ | Observations | | NOAA | 1773 ppb |\ | | AGAGE | 1774 ppb | |\ | | WDCGG | 1783 ppb | |\ | Meinshausen et al 2011 | | MAGICC | 1833 ppb |\ | Prather et al 2012 | | CH3CCl3-based | 11.2±1.3yr |\ | | | 9.8±1.6yr | |\ | Voulgarakis et al 2013 | | ACCMIP | -2.2±1.8% |\ | Holmes et al 2013 | | 1980/85-2000/05 | -2.06% |\ | This studyc | | | -4% |\ | Voulgarakis et al 2013 | | 1980-2000 | -2% |\ | This studyc | | 2000-2050 | +1.0d%+3.5% |\ | Hoor et al 2009 | | AERO2K | -1.55%Tg(N)-1 |\ | | QUANTIFY | -1.46%Tg(N)-1 | |\ | Myhre et al 2011 | | Model ensemble | -1.77%Tg(N)-1 |\ | Holmes et al 2011 | | REACT4C | -1.36%Tg(N)-1 |\ | Søvde et al 2014 | | dE NOx=QUANTIFY | -1.48%Tg(N)-1 |\ | This studyc | | | |\ | | SRES B1 | -1.61%Tg(N)-1 | |\ | Hodnebrog et al 2011 | | B1 ACARE | -1.48%Tg(N)-1 |\ | | SRES A1B | -1.22%Tg(N)-1 | |\ | Khodayari et al 2014a | | AEDT Scenario1 | -1.88%Tg(N)-1 |\ | | AEDT Baseline | -1.59%Tg(N)-1 | |\ | This studyc | | RCP45 | -1.36%Tg(N)-1 |\ \ 4750^{\\mathsf{d}},\\mathsf{T g}\ \ \\top\\mathfrak{g}\ \ 11.2\\pm1.3;\\mathsf{y r}\ \ 9.8\\pm1.6;\\mathsf{y r}\ \ -2.2\\pm1.8,%\ \ -1.55%\ {mathsf\\mathsf T{}\ \ {mathsf T g g({sf N})}^{{-1}}}\ \ -1.46%,\\mathsf{T g}(\\mathsf{N})^{-1}\ \ -1.61\\not\\sim\\mathsf{T g}(\\mathsf{N})^{-1}\ \ d E\_{N O x}=O U A N T I Y\ \ * * *\ \ 1826\ 1827\ 1828\ 1829\ \ 1830\ \ 1831\ 1832\ \ 1833\ 1834\ 1835\ 1836\ 1837\ 1838\ 1839\ 1840\ \ b) lifetime is around 7% greater than the total CH4 lifetime, as modelled by TROPOS\ the chemcial (τCH4+OH\ c\ numbers are based on transient simulation\ \ (\\sf{I C}H4+O H)\ \ c\ numbers are based on transient simulation\ d\ numbers might not be very accurate as they are read directly from the graphs found in the respecitve papers\ \ numbers are based on transient simulation\ d\ numbers might not be very accurate as they are read directly from the graphs found in the respecitve papers\ \ Table D.4. Calculated CH4 correction factors\ \ | Aviation emissions year | CH4 correction factors | |\ | --- | --- | --- |\ | This study | Grewe and Stenke(2008) methodology | |\ | 2000 | 0.73 | 0.65 |\ | 2005 | 0.75 | 0.73 |\ | 2011 | 0.78 | 0.81 |\ | 2018 | 0.79 | 0.86 |\ \ \\mathrm{C H}\_{4}\ \ a\ Table E.1. Scaling of contrail cirrus RF and ERF results\ \ | Model | Inventory | Representation of flight distance | RF(mW/m2) | Scalings | Scaled RF(mW/m2)b | Reference |\ | --- | --- | --- | --- | --- | --- | --- |\ | ECHAM4-CCMod | AERO2K2002 | track | 38 | S1,S2,S4 | 70 | Burkhardt and Kärcher(2011) |\ | ECHAM5-CCMod | AEDT2006 | slant | 56 | S3,S4 | 56 | Bock and Burkhardt(2016) |\ | COCIP | AEDT2006 | flight vectors | 63 | S4 | 72 | Schumann et al.(2015) |\ | CAM5 | AEDT2006 | slant | 13\[57\]c | S3,S4 | 57 | Chen and Gettelman(2013) |\ | Best estimate | | | | | 66d | |\ \ (m W/m^{2})\ \ (m N/m^{2})^{b}\ \ \ {\\sf S} _{1},{\\sf S}_{2},\ \ \\mathfrak{S4}\ \ \\mathbf{S} _{3},\\mathbf{S}_{4}\ \ \ {bf S S}\_\ \ 13,\[57\]^{\\mathrm{ ~~c~~}}\ \ \ {bf S},{\\bf S}\_{4}\ \ 66^{\\mathrm{d}}\ \ a\ Adapted from Table 1 of Bock and Burkhardt (2016).\ b\ RF that would be expected in 2006 when using slant distance from the AEDT inventory with hourly resolution.\ \ b\ RF that would be expected in 2006 when using slant distance from the AEDT inventory with hourly resolution.\ c-2\ An updated simulation (see text) yielded 57 mW m.\ \ c-2\ An updated simulation (see text) yielded 57 mW m.\ d\ The best estimate is of RFs, and excludes the Chen and Gettelman (2013) results since this is closer to an ERF\ \ d\ The best estimate is of RFs, and excludes the Chen and Gettelman (2013) results since this is closer to an ERF\ (see main text).\ \ * * *\ \ 1841\ 1842\ 1843\ 1844\ \ Table F.1a. Emission metrics and corresponding CO2-equivalent emissions for the ERF components of\ 2018 aviation emissions and cloudiness using CO2 IRF without C-cycle feedbacks from Gasser et al.\ (2017), and climate IRF from Boucher and Reddy (2008).\ \ {\\mathrm{c O}}\_{2}.\ \ 1846\ 1847\ 1848\ 1849\ \ \ mathrm\\mathrm{c O O}\ \ \\mathsf{G W P}\_{50}\ \ Metrics\ \ | ERF term | GWP20 | GWP50 |\ | --- | --- | --- |\ | CO2 | 1 | 1 |\ | Contrail cirrus(Tg CO2 basis) | 2.39 | 1.15 |\ | Contrail cirrus(km basis) | 40 | 19 |\ | Net NOx | 637 | 216 |\ | Aerosol-radiation | | |\ | Soot emissions | 4409 | 2125 |\ | SO2 emissions | -856 | -412 |\ | Water vapor emissions | 0.22 | 0.11 |\ \ | GWP100 | GTP20 | GTP50 | GTP100 |\ | --- | --- | --- | --- |\ | 1 | 1 | 1 | 1 |\ | 0.68 | 0.70 | 0.11 | 0.10 |\ | 11 | 12 | 1.9 | 1.6 |\ | 122 | -231 | -75 | 14 |\ | 1252 | 1295 | 210 | 177 |\ | -243 | -251 | -41 | -34 |\ | 0.06 | 0.07 | 0.01 | 0.009 |\ \ \\mathsf{G M P}\_{100}\ \ \\tt{G T P}\_{20}\ \ \\tt{G T P}\_{50}\ \ \\mathrm{c{O\_\_2{2}}}\ \ (\\mathsf{T g},\\mathsf{C O}\_{2},\\mathsf{b a s i s})\ \ \\mathrm{N o}\_{}\ \ \\mathrm{S O}\_{2}\ \ Table F.1b. Emission metrics and corresponding CO2-equivalent emissions for the ERF components of\ 2018 aviation emissions and cloudiness using CO2 IRF without C-cycle feedbacks, and climate IRF from\ Gasser et al. (2017).\ \ {\\mathrm{c O}}\_{2}.\ \ \\mathrm{C O\_{2}}\ \ Metrics\ \ | ERF term | GWP20 | GWP50 |\ | --- | --- | --- |\ | CO2 | 1 | 1 |\ | Contrail cirrus(Tg CO2 basis) | 2.39 | 1.15 |\ | Contrail cirrus(km basis) | 40 | 19 |\ | Net NOx | 637 | 216 |\ | Aerosol-radiation | | |\ | Soot emissions | 4409 | 2125 |\ | SO2 emissions | -856 | -412 |\ | Water vapor emissions | 0.22 | 0.11 |\ \ | GWP100 | GTP20 | GTP50 | GTP100 |\ | --- | --- | --- | --- |\ | 1 | 1 | 1 | 1 |\ | 0.68 | 0.3 | 0.19 | 0.15 |\ | 11 | 4 | 3.3 | 2.6 |\ | 122 | -420 | -18 | 22 |\ | 1252 | 466 | 360 | 284 |\ | -243 | -90 | -70 | -55 |\ | 0.06 | 0.03 | 0.018 | 0.014 |\ \ \\mathsf{G W P}\_{50}\ \ \\mathsf{G W P}\_{100}\ \ \\overline{{\\mathrm{{}o}\_{2}}}\ \ \\mathtt{G T P}\_{100}\ \ \\mathrm{C O}\_{i}\ \ .\\mathrm{N O}\_{\\times}\ \ 1850\ \ \\mathrm{S O}\_{2}\ \ * * *\ \ 1852\ 1853\ 1854\ 1855\ 1856\ \ Figure D.1. Matrix of pair-wise scatter plots of RF values from NOx terms: short-term O3, CH4, CH4-\ induced O3, SWV and net NOx (i.e., the sum of all 4 components), all represented as normalized RFs\ -2-1-1\ (mW m (Tg(N)yr)) from the ensemble studies (see details in text). The red line is the linear fit, the\ ellipse shows the 95% confidence level and histograms present frequencies.\ \ \ up\_{3}\ \ \ {mathrm N O}\_{\\mathrm{x}}\ \ \ mathsf O{}\_{3}\ \ \\mathrm{N O\_{x}}\ \ (m W^{2}(\ g(N)y\ ^{-1})\ \ * * *\ \ 1859\ 1860\ 1861\ 1862\ 1863\ 1864\ \ \[Image: R329\]\ \ \[Image: R329\]\ \ \[Image: R329\]\ \ (b) Evolution of the global CH4 burden in TROPOS for transient aircraft NOx emissions combining\ historical emissions (1950–2000) and RCP-4.5 emissions (2000–2050); and constant emissions for the\ years 2000 and 2050.\ \ 2000^{\\circ}\ \ * * *\ \ 1865\ 1866\ 1867\ 1868\ 1869\ 1870\ 1871\ 1872\ 1873\ 1874\ 1875\ 1876\ 1877\ 1878\ 1879\ 1880\ \ (c) Global CH4 lifetime due to aircraft NOx emissions in TROPOS for transient emissions combining\ historical emissions (1950–2000) and RCP-4.5 emissions (2000–2050); and constant emissions for the\ years 2000 and 2050.\ \ \\mathrm{N O\_{x}}\ \ (d) Global CH4 lifetime reduction due to aircraft NOx emissions in TROPOS for transient emissions\ combining historical emissions (1950–2000) and RCP-4.5 emissions (2000–2050); and constant\ emissions for the years 2000 and 2050. The dashed lines represent 2000 and 2050 equilibrium values\ (light and dark blue) and 2000 and 2050 transient values (red).\ \ \\mathrm{N O\_{x}}\ \ \\mathrm{C H}\_{4}\ \ (e) Global CH4 burden reduction due to aircraft NOx emissions in TROPOS for transient emissions\ combining historical emissions (1950–2000) and RCP-4.5 emissions (2000–2050); and constant\ emissions for the years 2000 and 2050. The dashed lines represent 2000 and 2050 equilibrium values\ (light and dark blue) and 2000 and 2050 transient values (red).\ \ \\mathrm{N O\_{x}}\ \ \ \\mathrm{C H}\_{4}\ \ (f) Global CH4 burden reduction due to aircraft NOx emissions in TROPOS for transient emissions\ combining historical emissions (1950–2000) and RCP-4.5 emissions (2000–2050); and constant\ emissions for the year 2018. The dashed lines represent 2018 equilibrium (green) and transient values\ (red).\ \ \\mathrm{N O\_{x}}\ \ \\mathrm{C H}\_{4}