General Comments:

This study evaluated the temporal and spatial distributions of yield and economic losses due to long-term ozone exposure for major crops in China between 2010 and 2017. The results of this study are interesting, however, the novelty is not clearly stated, and the writing is very poor which requires almost a total revision. Please carefully address the comments below.

Major comments:

1. About the Introduction

There are only two paragraphs in the introduction, with the first paragraph does not mention the impact of ozone on crops at all. Moreover, the scientific questions did not raise clearly in the second paragraph. Overall, I feel the introduction is not well structed, and the science question targeted in this study needs to be clearly stated.

2. About the section 2.2

Line 85-90: The authors mentioned that there are many different crop-ozone matrixes available, and they adopted AOT40 in their study. The authors should at least lay out a few crop-ozone matrixes, and talk about the possible advantage and disadvantage among different matrixes. At the end, give a reason why AOT40 is selected.

Line 94-95: The authors should specifically define or point out the crops corresponding to what growing season, as multiple growing seasons were mentioned in the manuscript.

Line 96-97: Are there any differences in the definition of the growing season between NWCACP and FAO? The authors only mentioned NWCACP and FAO with no details.

3. About the section 3.3

There are redundant descriptions. For instance, the authors made some comparison to some literature, i.e., Lines 180, 181, 199, 205 and 206. However, they repeat similar discussions later (i.e., Line 270-273). In addition, the comparison does not specifically mention what year or period, making the comparison invalid.

Line 191-193

The authors highlight the differences in the calculations of the growing season lead to different years for the lowest values. In my opinion, this is nothing new worth highlighting.

Line 212-213:

It seems to be contradictory that the authors stated the lowest CPL in northeast China, and then emphasize Heilongjiang is the highest. Later on, I realize the statement of highest yield in Heilongjiang is probably in another year. The authors need to carefully check out the entire manuscript to make the statement clear and readable.

4. The Discussions section needs a total revision.

There are many sentences duplicated in several places. For instance, Line 81-82, “In general, the model simulated AOT40 values were lower than the observation data, with normalized mean bias ranging from -5% in 2015 to -28% in 2017”. Line 244-246, and 278-279

Lines 238-246: very wordy, should be trimmed substantially and make clear what major message the authors want to convey.

The second paragraph of Discussions only lay out many results without any depth.

The authors mentioned many times of the year 2014 (i.e., Lines 185, 192, and 219), but what special with the year has never been mentioned.

The authors mentioned spatial heterogeneity across regions and provinces, but with no details.

The last few sentences talked about the ozone pollution control over different regions. However, the major role of ozone on CPL over these regions have not been well discussed at all.

Minor comments:

1. Line 59-60: This message of the sentence is not clearly stated. The sentence writes that previous studies have focused on crop production loss from ozone at the global scale. Have any of the studies focused on China?
2. Line 85: Matrixes: should be metrics?
3. Line 129: The price for each crop during 2010-2017 is given based on the min/max, however, the readers do not know the specific price corresponding to each year. A table might be useful to lay out the prices.
4. Line 162: RYLs for specific crops should be clearly written.
5. Line 178: The citation of the Statistical Yearbook of China should be added.
6. Line 183: “CPL” should be replaced by “wheat CPL”
7. Line 223: studies changed to study.
8. Table S13: The production loss of China in 2017 was miscalculated and “374” should be replaced by “74”. Please carefully check all the calculations in the tables.
9. Line 235: References should be added.
10. Line 237: The authors said the previous studies only focused on small regions. However, in the introduction, the authors mentioned there are studies with a focus of the globe. This seems to be contradictory.
11. Line 255: It should be written clearly whether the CPL and EL for a particular crop or the total CPL and EL for all four crops.

General comments:

This is a manuscript delivering important messages towards China’s air quality policymaking. They found that crop yield damages due to ozone air pollution have increased in recent years and are especially large for wheat and rice.  Accumulatively, the economic losses are substantial, i.e. around ~20 billion USD for major crops during the past 8 years. Findings of this study indicate that improving China’s ozone air quality can benefit food security, in addition to human health which has been the dominant driver of previous clean air policies. This reviewer works broadly in the arena of atmospheric chemistry and policy-relevant science instead of being an expert on vegetation impacts of ozone, thus will only judge based on best expertise. This reviewer recommends the acceptance of this manuscript if the following comments can be sufficiently addressed.

Specific comments:

Introduction:

1. In the first paragraph, it is worth adding the mechanisms of observed increasing ozone concentrations in China. The reasons include not only increasing anthropogenic VOC emissions but also decreased ozone titration due to decreased NOx emissions especially in megacities where ozone production is usually NOx-saturated. It is worth reviewing relevant literature.
2. Line 48-54. Literature seems to find very large yield decrease effects for soybean compared to other crops. I wonder why the authors found relatively small impact as indicated by Line 22, which is one order of magnitude smaller than previous research.
3. L57-60: statement of the key innovation of this study does seem as persuasive, since Line 54-57 indicates that a recent study evaluates effects of ozone on yields of 3 crops for 4 years. The authors do 4 more years of analyses with 1 additional crop (i.e. soybean). Are there new data used or improved model simulation or emission inventories adopted in this research? This novelty statement seems a bit weak. In addition, did previous research not at all examine spatial variations of ozone damages to crop yields? If there are any, they need to be included as literature review here.
4. It is probably also useful to mention the uncertain impacts of climate change on crop yields and increasing future food demand associated with increased population and increased meat demand thus animal feed crops, in the introduction or somewhere in discussion. This will make the evaluation of ozone yield effects and potential mitigation appear to be more urgently relevant to air quality and food security.

Methods:

1. Line 83: Model’s underestimation of AOT40 seems a bit severe. Is there a way to constrain model results with observations? Does the under-estimation indicate underestimate of ozone concentrations? If this is a modeling issue pointed out before, relevant literature needs to be described? Possible mechanisms need to be addressed in Discussion.

Results:

1. 1 title ‘ozone concentration change’ is not precise – it is metric (AOT) value change. Consider revising the title.
2. Line 141-144 seems to address my previous comment on Introduction but this review of literature has been put in a weird place.
3. Line 145-146: To explain the peak of AOT40 in one specific year, one needs to figure out whether the seasonality of ozone concentrations have changed over time since the growing season likely remain the same across years, correct?
4. Section 3.2 and 3.3 list many detailed results. I wonder if at the beginning of each paragraph the authors can summarize the findings in one topical sentence. What are the findings that should be noted without getting into all the details? The readers may get very lost with all the details.
5. Line 217-218: results of this research is much much smaller than this previous research.

Discussion:

1. It appears to me that Line 250-279 are still about results, although some comparisons with earlier research has been added.
2. Line 280-end appears to be actually like a real ‘Discussion’ that really expands the findings of the research. There are not very clear messages to policymaking regarding ozone control in which provinces should be prioritized. Consider improving the Discussion. More details could be provided regarding how to address ozone pollution in prioritized regions (i.e. high losses).

Grammar issues need to be fixed, to name a few, line 20 ‘in 2017’; Line 73 ‘outside of China’ instead of ‘outside China’

Tables and Figures:

Table 1 seems to be methods and from previous research, instead of actual research design or results.

Figure 3 consider putting the names of corresponding crops next to the (a) (b) (c) (d)…

Figure 4 For some crops, the losses peak at 2014 while for others the losses peak at 2015.

Figure 5 the caption needs to describe panels a) and b). Do you simply group the provinces based on the magnitude of values?

Li et al (“Surface ozone impacts on major crop production in China from 2010 to 2017”) quantifies the crop production and economic loss from surface ozone (O3) in China over several years. Overall, the method used is sound and has been used by many studies previously. However, significant improvements must be made to the description of results, discussion and implications for this to be a meaningful scientific paper worth of publication in ACP.

Sections 3.2-3.4 should be simplified and reorganized (together or separately) to better highlight the main results, rather than list many values that can be found in tables and figures. Increase comparing/contrasting of different crops and regions and tie these to an improved discussion section.

The current discussion section is largely a restating of the intro, methods and results. Instead, expand the final paragraph to speak more about the implications of the work. Include discussion of the seasonal cycle of O3 that is carried through to the cropping season differences. Add more about the chemistry and policies throughout China that causes the results. For example, why O3 increases when PM regulations were successful. This section should also include discussion of the uncertainties in the model O3 concentration, AOT40 metric and economic valuation.

More specific comments/suggestions are listed below:

Line 74: Add at least the direction of adjustment. Increased due to vertical gradient near surface?

Line 77: Why compare model AOT40 and not model concentrations? AOT40 has also not yet been introduced.

Line 80: A figure showing the observed-model concentration comparison would be helpful, especially the expected seasonal cycle, despite the bias. Do the patterns match?

Line 85: Should “matrixes” be “metrics”?

Line 87: Why not use other metrics such as M12/M7 or others instead of or in addition to AOT40?

Line 125: Can purchase price be assumed to be the same as economic value? How does the value from the initial sale propagate fully into the economy?

Line 126: Is this the global price from FAOSTAT?

Lines 134-135: This section uses the annual values to show the general trends and distribution, not because of the varying growing seasons.

Line 185: “later” than?

Line 192: This is actually due to the seasonal cycling / varying O3 between the growing seasons, not the difference in calculation of the growing season itself

Figure 5: What is the a) and b) each referring to? Missing from the caption.

Figures 5-6: There are too many bars, with the variation between crops in many provinces roughly the same. Consider simplifying to highlight main points.