Fine particles (PM2.5) and reactive N (N_r) and S (S_r) are important air quality mitigation targets due to their negative effects on human health, ecosystem diversity, and climate. This study uses the EMEP MSC-W model coupled with WRF meteorology for 2015 to explore the sensitivity of global and regional N_r, S_r, and PM_2.5 to 20% and 40% reductions in anthropogenic emissions of NH₃, NO_x, and SO_x. They found the multiple co-benefits and small disbenefits of individual emissions on different species are highly geographically variable, which reveals the importance of prioritizing emissions strategies and order of precursor reduction. The subject matter is appropriate for Atmospheric Chemistry and Physics. The manuscript is also generally well-written, but I recommend a major revision before the paper is accepted for publication.

General comments:

The paper systematically describes an evaluation of PM and N/S concentrations and deposition sensitivity to precursor emissions, but overall, I feel a too weak connection with any empirical evidence:

1. Although the paper is already very long, I suggest adding a brief section and 1-2 new figures that describe a “base” scenario that shows the simulated concentrations of related chemical species compared against available observations.
2. Such a 20-40% emission reduction occurred in many parts of the world, for example, the United States/Canada, EMEP, and East Asia regions. There are also long-term air quality observations available in these regions. How are your sensitivity results compared with this observational evidence?

The abstract and conclusion sections need to be significantly shortened, less repetitive, and with a higher-level summary. The results section is generally too long, while the discussion section is too short and weak. The discussion should be expanded significantly with subsections. The research implication of the article is not emphasized enough, giving the readers the impression that it is a purely technical report. Besides, the discussion on the sources of uncertainty needs to be elaborated in more detail.

The equation to calculate the sensitivity and the definition of sensitive areas need to be described in more detail within the method section. Why are only these four regions analyzed? Why not include more regions in the world, such as Middle/South America, Africa, and the Middle East? These regions may have sensitive ecosystems, emerging economic growth, and also large populations. The omission of these regions needs better reasoning.

Specific comments:

1. Line 49-50. There is a grammatical error, consider changing “also its form” to “also by its form”.

2. Line 57-58. This sentence is vague. It is not enough to support your point on emphasizing the global dominance of emissions in East and South Asia.

3. Line 89-95. The description of the single reference cited here is too long, you may consider summarizing it.

4. Line 99-100. Consider changing “global” to “the global”.

5. Line 175-178. These sentences may be moved to the discussion section.

6. Line 367. The method of the specific definition of sensitivity regime through the precursor includes sensitivities of NH₃, NO_x, and SO_x, which could be elaborated in your method section.

7. Line 382. It seems a bit obscure to be a separate paragraph here, are you trying to explain the sensitivity of SO_x in the marine area?

8. Line 514-515. This sentence may be not proper here, please consider moving it to the discussion section.

9. Line 717. Is it possible that the significant seasonality in these few areas is related to the vegetation on the land surface?

10. Line 737-738. Consider elaborating and expanding it by citing some related references.

11. Line 778-779. This sentence seems a bit vague. Please check it.

12 . Line 789-790. There are several grammatical errors here. Please check and rewrite it.

The manuscript presented a suite of model sensitivity simulations, which investigated the global responses of atmospheric reactive nitrogen and sulfur levels and their deposition fluxes to their emission reductions. Global 20% and 40% emission reductions for NH3, NOx, and SO2 (individually and jointly) were simulated to analyze the resulting changes in global nitrogen and sulfur concentrations and deposition, as well as annual mean PM2.5 concentrations. The results show that the individual emissions reduction (NH3, NOx, or SO2) has multiple co-benefits and small disbenefits on different species. Such changes reflect geographically-varying non-linear chemical responses of nitrogen, sulfur, and PM2.5 to emissions reductions 

I think the manuscript, by comprehensively analyzing the response of multiple indicators (air pollutants levels and deposition fluxes), includes important findings to understand the underlying drivers or mechanisms in response to emission reductions regionally and globally. This study is thus well-suitable for the scope of ACP, and I would like to see its publication. Before that, some moderate comments shall be addressed.

Specific comments:

1) Model set-up

Global 20% and 40% emission reductions were applied in this study. Some sentences are needed to justify the selection of 20% and 40%. Why a greater emission reduction (i.e., 60%) was not considered?

2) Section 3.1.2, Line 260

What are the sources of fine and coarse nitrate aerosol? Can you explain why coarse nitrate would increase associated with NH3 emission reduction? Please clarify.

3) Page 18, Figure 8

Figure 8 showed the spatial sensitivity regimes based on 40% emission reductions and annual mean PM2.5 concentrations. Such sensitivity regimes shall have large seasonal variations. Do you have the model datasets to generally the seasonal maps? That shall provide valuable information to understand the SIA formation regimes.

4) For the green/red circles and stars in Figures 2, 4, 6, 9, and 11, in the main text, the symbols that are discussed as the maximum reductions (e.g., Page 11, Line 330-335) were denoted as “Min” in these figures. Please be consistent.

5) Page 25, Line 708-710

It is not clear what “non-linearity” mean here in the text. We can see from Figures 3, 5, 7, and 10, the responses with respect to 20% vs. 40% emission reductions are rather linear. They deviate from the 1:1 line, however, the responses are linear. Please clarify. 

6) The writing of the manuscript is rather intensive. Many results are described in parallel, which makes the manuscript less focused. I understand that many results can be derived from the set of sensitivity simulations, still, the key findings of the study shall be better emphasized in the abstract and conclusions.