Please find below our response to the referee comments on our Manuscript “The Co-benefits of a Low-Carbon Future on Air Quality in Europe”  

To guide the guide the review process, the comments from both reviewers are in italics and our responses are in bold. We have numbered the reviewer comments to make it easier to refer to them in the response. Text added to the manuscript in response to the comments is in quotation marks below. We have produced a tracked changes manuscript that can be submitted when required. Where a figure has been updated or added for the purpose of addressing these comments, it has been attached to an additional document submitted alongside this response. 

We would like to thank both referees for the time taken to review our manuscript and for providing helpful and constructive comments. We have addressed all comments and have modified our manuscript accordingly. Our manuscript has been strongly improved through the review process and we hope our revised manuscript is now suitable for publication. 

Reviewer 1 

  This work describes the impact of three different SSP scenarios on future PM2.5 and O3 concentrations in Europe. The WRF-Chem model is applied over Europe on 30 x 30 km for the year 2050 and the SSP scenarios are compared to the Base Case year 2014. The WRF-Chem model is first evaluated by comparing calculated gas and aerosol species with observations followed by the analysis of the SSP scenarios on calculated PM2.5 and O3 concentrations. 

The overall structure of the paper is good and the topic is meaningful, but it lacks a profound explanation for the results. For example, the authors make no attempt to physically and chemically relate the changes in emissions (SSP scenarios) to the observed changes in PM2.5 and O3 concentrations. The manuscript amounts to little more than a statement that where emissions of aerosol precursors have decreased in 2050, i.e. a decrease/increase in calculated PM2.5 and O3 are found. 

Therefore, the article needs substantial revisions before it can be accepted for publication. 

We thank the commenter for the comments, they have been very helpful particularly in providing sources to extrapolate some of the trends. We have improved our analysis significantly to add information on the underlying chemistry behind our results.  

Major comments. 

1. The Introduction is quite long. Please try to shorten it. 

We agree with this comment and have taken steps to reduce unneeded background in the introduction to provide a more concise and focused paper. We have cut the introduction from 1970 to 1424 words. 

2. Recent studies (e.g. Thunis et al. (2021, 2022) and Clappier et al. (2021)) showed that the relationship between changing the emissions (e.g. 10%, 25% and 50%) and resulting concentration changes on PM2.5 can be non-linear. Furthermore, they also showed contrasting chemical regimes over a distance of a few hundreds of kilometers. PM2.5 chemical regimes are mainly determined by the relative importance of NOx versus NH3 responses to emission reductions and show large variations seasonally and spatially. For example, in the Po Valley, responses of PM2.5 concentrations to NOx emission reductions beyond 25% become non-linear mainly during wintertime. The seasonality of emission reductions is not well addressed in the paper.  

We have reworked our discussion section using most of the suggested references to discuss non-linear chemical responses to emissions changes, with a focus on PM2.5 and O3 responses to NH3, NOx and SO2 emissions changes and how these may explain some of the trends we see. We also plot these emissions reductions spatially. Please see our responses to the comment number 37 for more details. 

Also, the size of the receptor (city definition) is important on the sensitivity of the emission reductions (especially in city centres) and 30 x 30 km could be too coarse to address this issue properly. 

Therefore, I would recommend to add to this study 6 simulations (3 SSP scenarios for PM2.5 and 3 for O3) on a higher resolution over a smaller domain (e.g. Po Valley area, Poland, Benelux or Ruhr area) to investigate the impact of the SSP scenarios on PM25 and O3, and describe the underlying chemical mechanisms on the calculated PM2.5 and O3 concentrations and the differences between the three scenarios. 

While performing local scale simulations is a very interesting direction to take future research, it is out of scope for this paper. We specifically designed our methodology with country/regional policy relevance in mind – simulating at scales that could improve understanding of air quality co-benefits at the national level that policymakers in Europe often consider. We are aware the current resolution is not optimal for understanding the chemical mechanisms at city level and as such, we have not focused on this level. This paper was intended to be similar in focus as other ACP publications such as Turnock et al. (2020) and Silva et al. (2016) which don’t focus in depth on the underlying chemistry behind the changes in concentrations of particular air pollutants . Our intended niche was using a mid-point between global and local scale models, hence the comparison we provide with Turnock et al’s results.  

Additionally, to set up local scale simulations would require effectively a whole new model setup. The emissions provided for CMIP6 are designed for global models and are not at a resolution appropriate for local-scale simulations. The computational expense of the model (each of our simulations took upwards of a month and produced several terabytes of output) also means that we could not effectively integrate this idea into the current paper with the attention it deserves. It would instead be a potential direction for future work in the vein of Sa et al. (2016) or Coelho et al. (2020). 

Coelho, S. R., S; Fernades, A.P; Lopes, M; Carvalho, D.: How the New Climate Scenarios Will Affect Air Quality Trends: An Exploratory Research, Urban Climate, 49, 2023. 

Sa, E. M., H; Ferreira, J; Marta-Almeida, M; Rocha, A; Carvalho, A; Freitas, S;  Borrego, C;: Climate change and pollutant emissions impacts on air quality in 2050 over Portugal, Atmospheric Environment, 131, 2016. 

Turnock, S. T., Allen, R. J., Andrews, M., Bauer, S. E., Deushi, M., Emmons, L., Good, P., Horowitz, L., John, J. G., Michou, M., Nabat, P., Naik, V., Neubauer, D., O'Connor, F. M., Olivié, D., Oshima, N., Schulz, M., Sellar, A., Shim, S., Takemura, T., Tilmes, S., Tsigaridis, K., Wu, T., and Zhang, J.: Historical and future changes in air pollutants from CMIP6 models, Atmospheric Chemistry and Physics, 20, 14547-14579, 10.5194/acp-20-14547-2020, 2020. 

Silva, R. A., West, J. J., Lamarque, J.-F., Shindell, D. T., Collins, W. J., Dalsoren, S., Faluvegi, G., Folberth, G., Horowitz, L. W., Nagashima, T., Naik, V., Rumbold, S. T., Sudo, K., Takemura, T., Bergmann, D., Cameron-Smith, P., Cionni, I., Doherty, R. M., Eyring, V., Josse, B., Mackenzie, I. A., Plummer, D., Righi, M., Stevenson, D. S., Strode, S., Szopa, S., and Zengast, G.: The effect of future ambient air pollution on human premature mortality to 2100 using output from the ACCMIP model ensemble, Atmospheric Chemistry and Physics, 16, 9847-9862, 10.5194/acp-16-9847-2016, 2016. 

Minor comments: 

3) The paper describes the impact of the different SSP scenarios on PM2.5 and O3 in Europe. Therefore, I suggest to change the title that captures these two pollutants. 

We agree that this makes the topic of the paper more clear and have amended the title to The Co-benefits of a Low-Carbon Future on PM2.5 and O3 Air Pollution in Europe 

4) Line 25: Be consistent with the naming of the species Ammonia, Sulfur, Nitrogen (non-capitals) throughout the whole text. 

Amended 

5) Line 30: remove the text between brackets. 

Amended 

6) Line 40: Remove the dot after m3. 

Amended 

7) Line 43: For reducing O3 levels in cities, this is not so ‘feasible’. Please rephrase. 

We have rephrased to the following to account for O3 increases: 

“Improving air quality in Europe is feasible: primary air pollution responds quickly to air pollutant emissions reductions, potentially resulting in lower population exposure. Notably, some secondary air pollutants such as O3 can worsen depending on emissions reductions in some circumstances, but reducing emissions largely leads to an overall air quality benefit.” 

8) Line 47-49: Note that for some locations the PM10 and O3 concentrations increased, see Putaud et al., (2021, 2023). Please rephrase. 

See response to comment 9 

9) Line 52: Also background CH4 concentrations have increased over the years. 

In response to these comments, the paper now specifies that PM2.5 air quality in Europe has improved over recent years and ozone has worsened. We do not focus on CH4 or PM10, so have omitted this in order to keep the length of the introduction down. 

10) Line 55: Is mentioned earlier. 

Amended to remove repetition 

11) Line 70: Add references here. 

We have added a reference to Von Schneidemesser et al. (2015), which describes in detail the chemical processes relevant to climate and air quality policies. 

Von Schneidemesser, E., Monks, P. S., Allan, J. D., Bruhwiler, L., Forster, P., Fowler, D., Lauer, A., Morgan, W. T., Paasonen, P., Righi, M., Sindelarova, K., and Sutton, M. A.: Chemistry and the Linkages between Air Quality and Climate Change, Chemical Reviews, 115, 3856-3897, 10.1021/acs.chemrev.5b00089, 2015. 

12) Line 74-77: This is a repetition of line 30. 

Removed with reference to previous discussion 

13) Line 79: “Modelling” 

Amended 

14) Line 90: Can the authors address if future land-use/cover changes are included in the SSP scenarios? If not included, what would be the impact of future land cover changes on the outcome of the scenarios? 

We can confirm that the SSP scenarios do have land use projections, and these are used in the gridded emission projections. Notably, we only change anthropogenic emissions in our simulations. 

15) Line 114-119: This part fits better in a section Discussion, not in the Introduction. 

We included this paragraph as a way to segue into our methodology. Without only considering emissions, rather than emissions + meteorology, we could not have used WRF-Chem as we would not be able to perform long enough simulations to capture meteorological variability. I have moved the first two sentences to the methodology where I introduce the emissions-only simulations and removed the second two sentences. 

Section 2.1 

16) Are the simulations performed with feedback switched on in WRF-Chem? If so, could you please describe if the meteorology changes between the scenarios and how much this affects the air pollutants? This might be important for SO2 oxidation in clouds (SO4= formation) and O3 photochemistry. 

The aerosol-radiation feedback is switched on, however as the simulations are frequently nudged to meteorology, there is no meaningful meteorological difference between the scenarios. We have added this information to our methodology. 

17) Line 166: Remove “The emissions… of futures” and replace by something like that you performed three different emission reduction scenarios. 

Amended 

18) Line 175: Can you name the countries where NH3 has different trajectories between SSP2-4.5 and SSP3-7.0? 

There are not notably different trajectories in NH3 emissions between SSP2-4.5 and SSP3-7.0, in fact the same countries seem to show the same trends in these scenarios (increases in France, Spain, Italy, decreases in Western Germany). This line is meant to say that these scenarios show different trajectories compared to SSP1-2.6. I have amended to reflect this and provide examples. 

19) Page 7: Can you please elaborate more what the differences are between the numbers in Figure 1 and Table 2. The numbers in Table 2 do not correspond with the number is Figure 1. You can also put the relative reductions between brackets in each column. 

Thank you very much for pointing this out – upon looking into this, I had erroneously put the global emissions in the table when I intended to put the European emissions. This has been amended and Figure 1 is now representative of the changes in table 1. 

20) Page 7: Table 2 should be “2014 emissions”. 

Amended 

21) It would be nice to have this information for the different countries in the model domain. Or only for the bigger countries, e.g. UK, FR, ES, DE, I. Not every country reduces in the same manner their emissions. This would help to understand the emission ratios and the corresponding chemical regimes (e.g. NOx vs. NH3 for PM2.5 formation & VOC/NOX ratios for O3) for the different scenarios. 

Also, it would be interesting to see the difference in spatial distribution in the emissions between the SSP scenarios for NOx, NH3 and PM2.5. 

We haven’t specifically extracted the emissions changes in individual countries in order to avoid adding a lot of extra text and tables, but what we have done is plot the emissions changes spatially and added them to the supplementary material (Supplementary figs A3, A4, A5). These are shown in a document attached with our response, labelled as Figures 4,5 and 6. This means that we can refer to emissions changes in particular countries/regions (and have done so in describing the PM2.5 and O3 formation).  

21) Line 197: what do you mean by “For some analysis..” ? 

For some figures and tables, we weighted PM2.5 and O3 by population, but not all. The ones where population weighting is used mention it in the captions. This is just to provide the formula where relevant.  

22) Line 203: Add “other” between “and aerosol”. 

Amended 

23) Line 206: Please provide reference. 

Have amended to “as measurement and modelling of air quality in complex terrain is challenging and frequently less accurate (Giovannini et al. 2020)”  

Giovannini, L. F., E; Karl, T; Rotach, M.W; Staquet, C; Trini Castelli, S; Zardi, D: Atmospheric Pollutant Dispersion over Complex Terrain: Challenges and Needs for Improving Air Quality Measurements and Modeling, Atmosphere, 11, 2020. 

24) Line 207: Add dot at the end of the sentence. 

Amended 

25) Line 209: Use the correct syntax for Figure(s) -> Fig. 

Amended 

26) Line 210: Indicate which panels (a) and (b). 

Amended 

27) Figure 2: Please provide Titles of the plots, plus add to the plot for example the Bias. 

We have updated figure 2 based on these suggestions, and some suggestions from reviewer 2 to improve the readability and have added some more validation sites (K-Puszta, Hungary; Rucava, Latvia) to cover more sites in Eastern Europe. It is shown as figure 1 in the document submitted with our respnse. 

28) Table 3: 5th Column can be named “Bias”. 

Amended 

29) Table 3: Why is O3 missing in the evaluation? 

It was excluded as the purpose of the table was to focus on PM2.5 components and diagnose the overall PM2.5 bias.  

30) Line 218: I don’t see the added value of Figure 3. 

Figure 3 was included to demonstrate whether it was a systemic bias or seasonal. 

31) Also I don’t see the relevance for Figure 4a. Knowing that the emissions are different, source sector allocation, temporal and spatial profiles. 

 Understandable, we added this simply as another comparison than the EBAS observations to provide additional validation. We have removed it. 

32) Line 249: PM2.5 is overestimated by 8 ug/m3 on average, but I wonder if that’s true for the Eastern part of Europe. Can the author say something about the inclusion of condesables in the emissions? I guess they aren’t. The condensables have a large fraction to the PM2.5 emissions for the residential heating sector, especially in Eastern Europe. How is the model performing in that area with respect to PM2.5 concentrations? 

To our knowledge, the emissions do not factor in condensables. The papers discussing the production of these emissions make no reference to them (Hoesly et al. 2018; Feng et al. 2020). While fewer suitable observations were available in Eastern Europe, those that are included show lower bias than the mean. We have added this detail to our validation section.  

Hoesly, R. M., Smith, S. J., Feng, L., Klimont, Z., Janssens-Maenhout, G., Pitkanen, T., Seibert, J. J., Vu, L., Andres, R. J., Bolt, R. M., Bond, T. C., Dawidowski, L., Kholod, N., Kurokawa, J.-I., Li, M., Liu, L., Lu, Z., Moura, M. C. P., O'Rourke, P. R., and Zhang, Q.: Historical (1750–2014) anthropogenic emissions of reactive gases and aerosols from the Community Emissions Data System (CEDS), Geoscientific Model Development, 11, 369-408, 10.5194/gmd-11-369-2018, 2018. 

Feng, L., Smith, S. J., Braun, C., Crippa, M., Gidden, M. J., Hoesly, R., Klimont, Z., Van Marle, M., Van Den Berg, M., and Van Der Werf, G. R.: The generation of gridded emissions data for CMIP6, Geoscientific Model Development, 13, 461-482, 10.5194/gmd-13-461-2020, 2020. 

33) Line 250: It’s a classical problem that models overestimate ammonium nitrate aerosol when compared to quartz filters, due to the evaporation of ammonium nitrate aerosol when temperatures are higher than 20 degrees Celsius (Schaap et al., 2003, De Meij et al., 2006). Did the authors investigate if for these measurement sites quartz filters are used? This could explain the bias. 

Thank you for this suggestion – it was something we had not considered. The measurement sites we have accessed tend to use beta-attenuation measuring devices with continuous glass fibre filters. 

34) Figure 5 caption, remove “of” form “of the annual”. 

Amended 

35) Also, I don’t see the added value of Fig.5a. For Fig5. b-d, I propose to show the differences between 2014 and the scenarios in absolute terms. You can put the relative difference to the Appendix. Changes in low concentrations might lead to large relative differences, while larger differences in absolute values in lower relative differences. Changes in absolute values would make more sense. 

We provide 5a and 8a simply to show the concentrations we simulate for the present day as useful context for our results. We have focused on relative improvements as this stops the systematic PM2.5 overestimation from watering down our conclusions. We do agree that providing the absolute changes is valuable contextual information, so we have included similar figures to Fig 5 and Fig 8 showing the absolute to the appendix (Supplementary Figures A2 and A6). Attached to extra document as Figures 2 and 3. 

36) The same applies to Figure 8. 

As above 

37) Line 265-276: You describe the changes in PM2.5 concentrations for the different scenarios and provide possible explanations for that. Please quantify the emission reductions for PM, NOx and SO2 and describe the chemical mechanisms for the areas that you mention. 

We have already quantified the changes on the whole domain in Table 2 and Figure 1. We now present spatial analysis showing the emissions changes in supplementary figures A3, A4 and A5.  

We have retooled our discussion to describe the relevant chemical mechanisms This can be seen in the paragraphs between lines 265 and 300 in the updated paper. For convenience, we have pasted these below. 

“The PM2.5 reductions in SSP1-2.6 are larger across most of the domain compared to the other scenarios. This is most notable across Central/Eastern Europe (e.g. Germany, Poland, the Czech Republic, Austria). This is potentially because these regions have a larger proportion of anthropogenic PM2.5 sources than natural sources. Smaller improvements are projected in countries such as Portugal and Ireland where natural sources of PM2.5 dominate. In the regions where the reductions in SSP1-2.6 compared to SSP2-4.5 270 and SSP3-7.0 are the largest (such as Central/Eastern Germany), the size of the reduction is likely the result of the combined reduction in NH3 and NOx emissions in SSP1-2.6 (domain-wide, -25% and -70%, respectively). 

Across the domain, SSP2-4.5 and SSP3-7.0 have increases in NH3 emissions (both approximately 20%) and reductions in NOx (both approx -30%) (Figure 1). Where only NOx is reduced, in some regions (usually NOx-abundant regions) the increased oxidising capacity of the atmosphere can result in increased formation of secondary organic aerosol, thus limiting the efficacy of emissions reductions. This impact can be mitigated by joined reductions of both NOx and NH3 (Clappier et al. 2021). This is supported by the near-universal decreases of over 60% of NO3 PM2.5 following SSP1-2.6, which is the result of a domain-wide reduction of 18% of agricultural NOx emissions and 7% of agricultural NH3 emissions. While following SSP2- 4.5 and SSP3-7.0, NO3 PM2.5 does decrease universally, however, the reductions are much larger in urban regions than rural. Notably, both of these scenarios have approximately 20% increases in agricultural NH3 emissions and increases in agricultural NOx emissions (41% following SSP2-4.5, 10% following SSP3-7.0). This suggests that the difference in agricultural emissions will be a large driver of the extra reductions in PM2.5 following SSP1-2.6 and mitigation of emissions in this sector will be key to achieving improved air quality in Europe. It should however be noted, that some rural regions (e.g Western France. Scotland, Wales and most of Spain) do still have overall NH3 emissions increases following SSP1-2.6. This may explain the smaller reductions in PM2.5 in these regions (Supplementary Figures A3 & A4). 

Reductions in SO2 emissions also contribute to the reduced PM2.5. The countries that tend to see the largest decreases in PM2.5 concentrations following the future scenarios are Central/European countries. These countries, particularly in urban areas usually have a higher contribution from SOx aerosols (including SO4, which will cover much of these aerosols in the chemistry scheme) to PM2.5 (Zauli-Sajani et al. 2024). SOx aerosol is formed by reactions between SO2, NOx and NH3. For example, atmospheric sulfuric acid is formed when SO2 reacts with OH radicals. The sulfuric acid can then react with NH3 to form SO4 particulates (Clappier et al. 2021). Unlike NH3 and NOx, there is limited non-linearity of SO4 PM2.5 reductions resulting from mitigating SO2 emissions (Clappier et al. 2021). This means that they retain efficacy in reducing PM2.5 despite trends in NH3 and NOx reductions that may work against each other when only one is present. SO2 emissions reduce in all scenarios by approximately 30% for both SSP2-4.5 and SSP3-7.0 and by approximately 75% following SSP1-2.6 (Figure 1). This may explain why PM2.5 reductions are most consistently seen in Central/Eastern Europe across all the future scenarios here, reductions in SO2 emissions (see Supplementary Figure A5) lower PM2.5 in the sulfate fraction to a greater degree than in Western Europe. In Western Europe, following SSP2-4.5 and SSP3-7.0, the reductions in NOx in the absence of NH3 hamper PM2.5 reductions. This impact is reduced in Eastern Europe, where NH3 and NOx provide a lesser contribution to PM2.5 compared 300 to SO2, meaning Eastern Europe has larger PM2.5 reductions following SSP2-4.5 and SSP3-7.0.” 

Zauli-Sajani, S., Thunis, P., Pisoni, E., Bessagnet, B., Monforti-Ferrario, F., De Meij, A., Pekar, F., and Vignati, E.: Reducing biomass burning is key to decrease PM2.5 exposure in European cities, Scientific Reports, 14, 10.1038/s41598-024-60946-2, 2024. 

Clappier, P. Thunis, M. Beekmann, J.P. Putaud, A. de Meij, Impact of SOx, NOx and NH3 emission reductions on PM2.5 concentrations across Europe: Hints for future measure development, Environment International, Volume 156, 2021, 106699, ISSN 0160-4120, https://doi.org/10.1016/j.envint.2021.106699. 

38) Line 285: Can you please clarify the effect of NH3 emission reductions in NH3-sensitive regions, i.e. NOx abundant and NOx/NH3 ratios? 

As the computational expense of our model setup does not allow sensitivity simulations in which only singular emissions species are reduced, we cannot definitively state the impact of NH3 emissions reductions in the absence of other emissions changes. Our output also does not provide a variable demonstrating exactly which areas are NH3 limited and which are NOx limited. We have however enhanced our discussion of the impacts of NH3 and NOx reductions, providing a supplementary figure showing NO2 (extra document Figure 7) in the present and future (Supplementary figure A7) along with the new emissions changes figures we mention previously. We also use many of the suggested references to explain how the trends we see in NH3 and NOx emissions reductions Please see the response above for this enhanced discussion section. 

39) Line 290. Can you indicate for which source sector the largest reductions are found for Slovenia? 

Amended – residential emissions show the largest reductions in Slovenia. 

40) Line 297: I understand that sea salt is important for coastal sites, but does sea salt travel so much inland that it has a significant contribution to PM2.5? If so, please quantify it  

I have now referenced literature (Manders et al. 2010) demonstrating that seas salt PM2.5 can have a notable impact on European PM2.5 (up to 5 ug/m3) at up to 300km inland.  

Manders, A. M. M. S., M; Querol, X; Albert, M.F.M.A; Vercauteren, J; Kuhlbusch, T.A.J; Hoogerbrugge, R;: Sea salt concentrations across the European continent, Atmosperic Environment, 44, 2010. 

41) Line 310: Figure 7. I wonder if this is relevant, knowing that in Europe, the spatial and sectoral distribution of the emissions varies a lot (e.g. Po Valley vs central France or Poland vs Ireland). I would focus over smaller areas such as the southern part of Poland, Po Valley, Benelux or Ruhr area to highlight the impact of the three SSP scenarios on the yearly distribution of the PM2.5 concentrations and its components. 

We agree that it varies across the continent, however, this paper is not written to focus on local areas and is meant to show trends on a regional/continental scale. This would be outside the scope of our work. 

42) Line 350: Link this work to Air Quality legislation, it would be interesting to see the impact of these scenarios on the new WHO AQ limit values. The authors can select a few cities to show the impact of the SSP scenarios on WHO AQ limit values. 

We already cover this in Figure 5 (Figure 6 in previous version of the paper), where we have compared population-weighted mean PM2.5 in European countries to AQ limit values. We have not extracted data for cities due to the resolution of the model. 

43) Line 354, “such as” 

Amended 

44) Figure 8 caption: Move the Figs. 8c and d to a new Figure. 

We think keeping them as a single figure helps the reader understand at a glance the differences between our model setup and Turnock et al’s. While we understand the idea that separating them out keeps PM2.5 figures with the PM2.5 section and O3 with O3, we think the benefit of having them as one figure is more useful to summarise the benefits of our work. It also helps keep the length of the paper down. 

45) Line 361. It’s true that changes in O3 concentrations caused by changes in NOx emissions over urban areas are small. Reducing NOx increases O3 in city centres. However, the model resolution is 30 x 30km, which is quite coarse. Could the authors perform (as indicated before) additional simulations over a smaller area and see if the statement holds? 

We appreciate that the reviewer would like for this to be investigated further, but as discussed above we do not have the capacity to perform these model simulations, and this recommendation is outside the scope of our research. We think this would be best addressed as further work in a dedicated paper for this purpose. 

References: 

Clappier, P. Thunis, M. Beekmann, J.P. Putaud, A. de Meij, Impact of SOx, NOx and NH3 emission reductions on PM2.5 concentrations across Europe: Hints for future measure development, Environment International, Volume 156, 2021, 106699, ISSN 0160-4120, https://doi.org/10.1016/j.envint.2021.106699. 

De Meij, A., Krol, M., Dentener, F., Vignati, E., Cuvelier, C., and Thunis, P.: The sensitivity of aerosol in Europe to two different emission inventories and temporal distribution of emissions, Atmos. Chem. Phys., 6, 4287–4309, https://doi.org/10.5194/acp-6-4287-2006, 2006. 

Putaud, J.-P., Pozzoli, L., Pisoni, E., Martins Dos Santos, S., Lagler, F., Lanzani, G., Dal Santo, U., and Colette, A.: Impacts of the COVID-19 lockdown on air pollution at regional and urban background sites in northern Italy, Atmos. Chem. Phys., 21, 7597–7609, https://doi.org/10.5194/acp-21-7597-2021, 2021. 

Putaud, J.-P., Pisoni, E., Mangold, A., Hueglin, C., Sciare, J., Pikridas, M., Savvides, C., Ondracek, J., Mbengue, S., Wiedensohler, A., Weinhold, K., Merkel, M., Poulain, L., van Pinxteren, D., Herrmann, H., Massling, A., Nordstroem, C., Alastuey, A., Reche, C., Pérez, N., Castillo, S., Sorribas, M., Adame, J. A., Petaja, T., Lehtipalo, K., Niemi, J., Riffault, V., de Brito, J. F., Colette, A., Favez, O., Petit, J.-E., Gros, V., Gini, M. I., Vratolis, S., Eleftheriadis, K., Diapouli, E., Denier van der Gon, H., Yttri, K. E., and Aas, W.: Impact of 2020 COVID-19 lockdowns on particulate air pollution across Europe, Atmos. Chem. Phys., 23, 10145–10161, https://doi.org/10.5194/acp-23-10145-2023, 2023. 

Schaap, M., Van Loon, M., Ten Brink, H. M., Dentener, F., and Builtjes, P. J. H.: The nitrate aerosol field over Europe: simulations with an atmospheric chemistry-transport model of intermediate complexity, Atmos. Chem. Phys., 3, 5919–5976, 2003, http://www.atmos-chem-phys.net/3/5919/2003/. 

Thunis, P., Clappier, A., de Meij, A., Pisoni, E., Bessagnet, B., and Tarrason, L.: Why is the city's responsibility for its air pollution often underestimated? A focus on PM2.5, Atmos. Chem. Phys., 21, 18195–18212, https://doi.org/10.5194/acp-21-18195-2021, 2021. 

Thunis, P., Clappier, A., Beekmann, M., Putaud, J. P., Cuvelier, C., Madrazo, J., and de Meij, A.: Non-linear response of PM2.5 to changes in NOx and NH3 emissions in the Po basin (Italy): consequences for air quality plans, Atmos. Chem. Phys., 21, 9309–9327, https://doi.org/10.5194/acp-21-9309-2021, 2021. 

Reviewer 2 

Overall this paper presents a nice evaluation of future scenarios and the implications for air quality using a regional model with higher resolution. That said, I think the manuscript could be improved if some of these minor comments could be addressed. 

We thank reviewer 2 for their comments. The paper has been updated to reflect the suggested changes and we have added a limitations section. Some comments we decided not to follow the specific request, but we hope that our explanations for each below are sufficient.  

50) L25/L195-197: I don’t know why the chemical species are capitalized, they should not be. 

Amended 

51) L27: I would rather say ‘has consequences’ rather than may have. I think there is enough literature to show that. 

Amended 

52) L37-38: For clarity it would be good to specify ‘133 per 100,000’ deaths? People?… 

Amended to specify 133 deaths per 100,000 people 

53) L50-52: Isn’t it in urban areas also that we tend to target NOx emissions and so there is less titration of O3 happening? There are papers out there that discuss this and it would be worth mentioning because just saying intercontinental transport seems a bit of an oversimplification. 

Amended to include this - Conversely, O3 concentrations in Europe have increased in the latter half of the 20th century and early 21st century. See new lines: 

“The speed of this response to changes in emissions indicates that considerable improvements in air quality can be achieved when air pollutant emissions are reduced. Conversely, O$_3$ concentrations in Europe have increased in the latter half of the 20th century and early 21st century (Turnock et al. 2020) despite considerable reductions in local, anthropogenic O3 precursor emissions. This is potentially due to increased intercontinental transport of O3 precursors (Guerreiro et al. 2014) . Despite improving trends in PM2.5, O3 concentrations may increase due to reduced NOx emissions causing reduced titration of O3 (Miyazaki et al.2021). A different approach may therefore be required to reduce exposure to O3.” 

Turnock, S. T., Allen, R. J., Andrews, M., Bauer, S. E., Deushi, M., Emmons, L., Good, P., Horowitz, L., John, J. G., Michou, M., Nabat, P., Naik, V., Neubauer, D., O'Connor, F. M., Olivié, D., Oshima, N., Schulz, M., Sellar, A., Shim, S., Takemura, T., Tilmes, S., Tsigaridis, K., Wu, T., and Zhang, J.: Historical and future changes in air pollutants from CMIP6 models, Atmospheric Chemistry and Physics, 20, 14547-14579, 10.5194/acp-20-14547-2020, 2020. 

Guerreiro, C. B. B., Foltescu, V., and De Leeuw, F.: Air quality status and trends in Europe, Atmospheric Environment, 98, 376-384, 10.1016/j.atmosenv.2014.09.017, 2014. 

Miyazaki, K., Bowman, K., Sekiya, T., Takigawa, M., Neu, J. L., Sudo, K., Osterman, G., and Eskes, H.: Global tropospheric ozone responses to reduced NOx emissions linked to the COVID-19 worldwide lockdowns, Science Advances, 7, eabf7460, 10.1126/sciadv.abf7460, 2021. 

54) L75-76: Aerosols are not only cooling, some are also warming depending on the components and mixing, and the ‘masking’ is very different depending on the region.  Please make sure to communicate this correctly even if you want to just note it in one sentence. 

The information on aerosol’s impact on climate has been edited out to comply with reviewer 1’s comments on the length of the introduction. We decided that it was superfluous information not needed to clarify that climate can impact on air quality and the complexity requires modelling. 

55) L83-98: It is true that using consistent scenarios with the same underlying assumptions improves comparability. But is also has downsides, where various aspects of these scenarios, such as the evolution of air pollutants (in particular for RCPs) are poorly represented and people use them anyway. It also limits the diversity of scenarios considered and evaluated, which provides some limitations on the scientific evaluation. 

Thank you for bringing this up, we have added a sentence to ensure this caveat is included 

“Using scenarios for research this way does have disadvantages; the RCPs and SSPs are optimised for climate modelling, not air quality modelling and over-reliance on them in the literature may reduce the use of more specific scenarios.” 

56) L127: remove ‘of’ 

Amended – paragraph reworded so no longer in there. 

57) L162: do you mean µm? 

Microns in diameter – amended to reflect this 

58) L168: when describing the middle of the road scenario, you state GHG mitigation and sustainable development does not accelerate or decelerate strongly. I was unaware that we were currently on a sustainable development pathway. Can you explain what this is in reference to? This sounds to me more like BAU not sustainable development. 

This is in reference to sustainable behaviours, which are mostly reflected in the emissions, for example there is no great shift to vegetarianism and corresponding reduction in methane emissions. I have amended the description to hopefully reflect this better. 

59) L183: Any information on regridding? Or a publication? Or the regridded data provided through an open access link?  

The emissions data is freely available on the Input4MIPs website. We used standard xarray bilinear regridding methodology. 

60) Table 2: The column heading states 2015 emissions, but everywhere else in the text it mentions 2014. Please clarify. 

Should be 2014 as in the text, this has been amended. 

61) L199-200: Why use population data for 2020 when you are using emissions data for 2014? Why not be consistent and use 2014 population data? 

The population data did not have 2014 as an option – we chose 2020 to ensure we could have a consistent source for all population data, reducing the impact of differences between population grid methodologies. The paper has been amended to mention this. 

62) Figure 2: The colors are almost not recognizable for some of the symbols. I would use symbols that are not outlined in black, or make them bigger. 

We have removed the black borders, made the symbols bigger and changed the colour scheme in order to make them more readable. The new version of Figure 2 is available as Figure 1 in a document uploaded with our comment showing all new/updated figures. 

63) L266-267: some of the urban areas are self-explanatory. Paris and Madrid stand out. But the industrial regions? Are we considering most of eastern Europe an industrial region? Would be good to be a bit more explicit. In the following sentences where there are e.g., major combustion plants, might be good to put a dot or note those somehow on the map because I think most people don’t know where Deax or Belchatow are. 

More information has been given on which regions are considered industrial. We appreciate the comment about highlighting specific locations, however we decided we did not want to add extra detail to the plots as we think this would make them messier. As such, we have enhanced the text where specific locations such as Drax and Belchatow are named to explain where they are, (e.g Drax, North Yorkshire, UK). 

64) L305: If the (average) model overestimation were accounted for, would this put it under the guideline? Or still not? To understand the range being discussed here. 

Amended to cover this, see below: 

“It suggests that following SSP1-2.6, many countries could see PM2.5 exposure reduce below interim target values (guidelines the WHO suggest as targets to aim for before reaching the guideline value), representing a significant potential benefit for human health. Even the emissions reductions from SSP1-2.6 do not result in annual mean population-weighted PM2.5 concentrations under this guideline, but when factoring in the average overestimation of PM2.5 from our model, it is likely that this is achievable in some locations.” 

65) L325: starting from the comma in the line above, there seems to be a word missing or something because the sentence is super awkward. 

 This sentence has been reworded for clarity 

66) General: It would be nice if you could add some text to address limitations based on the model performance and the implications thereof. In particular related to PM2.5 since there are not insubstantial differences there and that is also where most health effects come from. If possible, adding some sort of uncertainty estimates would be nice, but if not, discussing whether this is over/under-estimated or higher/lower bounds, etc. would be good. In general, including a limitations section would be helpful. Also aspects such as the mismatch in ‘present day’ emissions vs population would likely have implications. Some studies have also found that including information on population demographic change can have a big influence and nothing even close to that is even mentioned here. 

This makes a lot of sense and a limitations section has been added (see below), we have emphasised the potential overestimations where relevant, noting how this may have impacted our conclusions, although it is hard to quantify this uncertainty. The population issue is a very fair point. We try to focus primarily on the percentage changes in concentration between scenarios, providing the population weighting as an additional source of information for those interested. We are currently working on an additional paper where the scenarios will be used for understanding the health impacts of future air quality changes. In that paper, we are comparing the changes when population uses the projections from the scenarios and where it is kept constant. To mitigate this, we have explained that we do not control for population and the implications of this. 

Limitations section: 

“There are limitations with this study to be considered - primary among these is the overestimation of PM2.5 concentrations in the present-day. While these are mostly systemic, they should be taken into account when considering the absolute concentrations reported, especially when compared to air quality guideline values. It is likely that the percentage changes we see would be less drastic had the model not overestimated PM2.5. As the model overestimates NO3, and underestimates SO4 PM2.5 we may overestimate the impacts on changes in NH3 and NOx emissions, particularly in the agriculture sector on future air quality compared to other sectors, such as industry.  

The resolution that we use is designed for the region/country scale and not the urban scale. 30km horizontal resolution is unlikely to faithfully capture atmospheric chemistry at city scale. Although we can represent the locations urban and industrial peaks, be aware that the model may not simulate chemistry at this scale as effectively as a finer-scale model.  

We also do not control for the impacts of population change in the future scenarios for Table 4, Figure 6 or Table 5. “