Greater fuel efficiency is potentially preferable to reducing NOx emissions for aviation’s climate impacts

Aviation emissions of nitrogen oxides (NOx) alter the composition of the atmosphere, perturbing the greenhouse gases ozone and methane, resulting in positive and negative radiative forcing effects, respectively. In 1981, the International Civil Aviation Organization adopted a first certification standard for the regulation of aircraft engine NOx emissions with subsequent increases in stringency in 1992, 1998, 2004 and 2010 to offset the growth of the environmental impact of air transport, the main motivation being to improve local air quality with the assumed co-benefit of reducing NOx emissions at altitude and therefore their climate impacts. Increased stringency is an ongoing topic of discussion and more stringent standards are usually associated with their beneficial environmental impact. Here we show that this is not necessarily the right direction with respect to reducing the climate impacts of aviation (as opposed to local air quality impacts) because of the tradeoff effects between reducing NOx emissions and increased fuel usage, along with a revised understanding of the radiative forcing effects of methane. Moreover, the predicted lower surface air pollution levels in the future will be beneficial for reducing the climate impact of aviation NOx emissions. Thus, further efforts leading to greater fuel efficiency, and therefore lower CO2 emissions, may be preferable to reducing NOx emissions in terms of aviation’s climate impacts.


Greater fuel efficiency is potentially preferable to reducing NO x emissions for aviation's climate impacts
Agnieszka Skowron, David S. Lee, Rubén Rodríguez De León, Ling L. Lim, Bethan Owen

Supplementary Tables
Supplementary Table 1: The list of performed simulations. The bolded entries present experiments that have been exploited in Figure 3.

Supplementary Figures
Supplementary Figure 1: The results of the global annual net NOx RFs from aircraft studies that have been published since the IPCC (1999) Special Report 'Aviation and the Global Atmosphere'. The analysis covers a wide range of global atmospheric chemistry/climate models (black -CTMs, blue -CCMs in CTM mode, green -CCMs) and present-day aviation emission inventories (different shapes of data points, see key).

Supplementary Note 1: Comparison of modelled, by 3D CTM MOZART-3, atmospheric constituents with measurement data
The ability of MOZART-3 to represent atmospheric processes and constituents was extensively evaluated by Kinnison et al. 2 and was shown in a number of publications 3,4 , with special attention paid to the upper troposphere and lower stratosphere region (UTLS) [5][6][7][8] . Through these publications, the capability of MOZART-3 in reproducing atmospheric composition, both globally and seasonally, with relatively good accuracy was shown. However, while the chemical tropopause exchanges are qualitatively well represented in MOZART-3, quantitatively, the trace gas profiles show some discrepancies. The main factor that determines the model's accuracy of chemical distribution in the UTLS region is the meteorological data: MOZART-3 driven by the ECMWF reanalysis winds has shown the best agreement with observational data 2,9 .

Modelling data
The 3D CTM, MOZART-3, set-up is described in detail in the Methods section. The monthly averages, starting in January and finishing in December, representing the year 2006 are exploited in the CTM comparison with observational data.

Measurement data
The summary of the geographical distributions of the selected observational stations and regions applied for this analysis is presented in Supplementary Figure 4. TOPSE measurements give a unique view of the spatial and temporal distribution of ozone and ozone precursors. The gridded climatologies and the regional profiles are provided; the latter were utilized in this study. The 3 regions used were defined in TOPSE campaign as follows: Boulder ( Table 2).
The annual average stratospheric (100-1 hPa) O3 column change in the 2 nd year is positive (with July showing the most negative peak -0.0002 DU), whilst the 6 th year shows negative O3 change through all year (with greatest July depletion -0.0137 DU). Despite these significant differences in the stratospheric response to aviation NOx emissions between the 2 nd and 6 th year, the total O3 change is not as much affected, as most of the mass of aircraft O3 is concentrated in the UTLS region. The difference in O3 column change between the 2 nd and 6 th year is 5.1%, the differences in the resultant O3 RF is -0.6%. Thus, the O3 changes presented in this work can be treated as reliable.
The impact of the interannual meteorological variability on aircraft O3 response has been also explored (Supplementary Figure 13). with other meteorological conditions. The uncertainty in aircraft O3 response arising from interannual meteorological variability is estimated here to be a rather of a low significance.
Supplementary Table 2: The global and annual mean O3 column change (in DU) and RF response due to aircraft O3 for consecutive years of MOZART-3 simulations.
Supplementary Figure 12: The results of the global and annual net NOx RFs from aircraft studies as presented in Supplementary Figure 1. The analysis covers a wide range of global atmospheric chemistry/climate models: grey -CTMs, blue -CCMs offline, run in CTM mode (green -CCM online has been excluded due to too few estimates). The upper panel presents a descriptive statistic that summarizes the net NOx RF derived by CTMs (grey) and offline CCMs (blue). The bottom panel shows the characteristic of the distribution of aircraft net NOx RF derived by CTMs (grey) and offline CCMs (blue).