In this paper, the authors used two methods to quantify the effect of soil NO emissions on surface ozone concentrations during the simulation period. The first is the traditional approach (BFM), which involves comparing simulated ozone concentrations in the base case with scenarios without soil NO emissions. The difference between the two scenarios is thought to represent the contribution of soil NO emissions to ozone. The second method uses the widely used ozone source allocation technique (OSAT) implemented by CAMx to label soil NO emissions as separate emission groups. The two methods basically reached the same conclusion: soil NO emission has a great effect on ozone concentration. This study emphasizes the importance of considering soil NO emissions in future ozone mitigation strategies in China. The article is integrative and innovative, and can be published. But it also has the following problems that need to be addressed by the authors.

Main question:

1. The Manuscript needs to add a discussion section in which the inclusion of comparisons with other models and the provision of comparative results with other relevant studies or models can provide a broader context for the application of the findings. More detailed information on the evaluation results and potential uncertainties in the model simulations would improve the robustness of the analysis.
2. For the simplicity of the article, some methods can be described briefly, but the main methods involved in the article need to be described carefully to enhance the readability of the article. The paper lacks the detailed introduction of WRF model configuration and model parameterization scheme. In addition, did the author consider the influence of regional differences in precipitation between north and south China on the model when selecting the parameterization scheme?
3. Please explain why the NO value simulated by OSAT model is higher than BFM
4. The conclusion of this paper is that the reduction of soil NO emission leads to the overall decrease of monthly MDA8 ozone concentration, and the reduction of ozone becomes more obvious with the increase of the reduction amount. NO can be titrated with O3. Does the author consider the depletion of ozone by NO?
5. NOx and VOCs are important ozone precursors. Is it reasonable to consider only NO without considering the effect of VOCs on ozone concentration?
6. In lines 195-196, the author pointed out that the spatial distribution of NO emissions in soil was very close to that of fertilization. However, I found that in the YRD area, there is a significant difference between the two, please explain the reason. And, Figure 4 of the article shows the simulation results of ozone concentration in China. In addition to the accidents in several hot cities introduced in the article, I noticed that the ozone concentration in the northwest of the Qinghai-Tibet Plateau was higher, please explain the reason. Refer to https://doi.org/10.1016/j.scitotenv.2022.160928.

Minor questions:

1. The caption of Figure 4 is not very clear. How are the results of observation and simulation represented in the figure? Does the author mean that scatter points indicate observations and base colors indicate simulation results?
2. (p11 Fig 5. Fig 6.) The map of soil NO ozone contribution from the Brute force method, and the model post-processing map can be done with map mask whitening.
3. Some references have the problem of being too old, and newer research results can be added to the references.
4. The resolution and clarity of all figures in the supplementary materials need to be improved. And, the results of ozone simulation in many areas of China exceed the boundary of China. Please confirm again whether it is reasonable.

General Comments:

This study adopted and applied the Berkeley-Dalhousie Soil NOx Parameterization (BDSNP) algorithm to develop soil NOx emission inventory, and evaluated their contribution to surface ozone concentration in China. Both the Brute-Force Method and Ozone Source Apportionment Technology were used in CAMx to assess spatial and temporal (diurnal and monthly) variations in five major regions of China.  It is concluded that soil NOx emissions substantially contributed to maximum daily 8-hr (MDA8) ozone concentrations by 8 to 12.5 ug/m3 on average for June 2018, and led to an increase of exceedance days by 10 to 43.5% in selected regions. This study highlights the importance of soil NOx in ozone formation in selected regions of China and suggests that control strategies aimed at soil NOx emission reductions, along with other sectors, should be considered. This manuscript is well written, and the methodology used in, and the results derived from this study are scientifically sound but some improvement or clarification should be considered as follows:

Specific Comments:

• Lines 37-39 (Abstract): It states that even with complete reductions in soil NO emissions about 450 million people are still exposed to unhealthy ozone levels, necessitating additional control policies such as transportation, power plant, etc. for selected regions. It is not clear whether or not the study suggests that soil NOx emission reductions (e.g., use of nitrification inhibitors) are considered a priority over other NOx emission sources.
• Lines # 43-46 (Introduction): It states that with the substantial decrease in ambient fine PM concentrations, ozone emerges as a simultaneously targeted air pollutant. Please note that nitrogen oxides serve as important precursors to both tropospheric ozone and fine PM with consequent adverse effects so NOx control strategies would lead to reductions of both ozone and fine PM.
• Lines # 66-69 (Introduction): a number of references related to soil NOx with upper values were summarized here but other studies (e.g., https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2020jd033304) should be acknowledged as well for completeness. For Sha et al, 2021 study, please note that in this study, the cropland regions included both high rates of fertilizer application and regular irrigation with the largest soil NOx of about 9 times higher than that of default in July.
• Lines # 109-143 (2.1. Estimation of soil NO emissions in China):
  • BDSNP estimates soil NOx emissions based on available soil nitrogen, soil temperature, and soil moisture. This parameterization classifies each grid cell as either arid or non-arid and applies one of two static relationships between soil NOx and soil moisture depending on that classification. Previous studies have shown that this relationship can be more dynamic, with emissions exhibiting different regional relationships between biome-specific NO emission factors (such as soil moisture) and soil NOx emissions. Hence, the uncertainty of A’biome values can be significant so some general discussion is warranted in terms of the limitation of BDSNP capability in producing more self-consistent emissions estimates regardless of the choice of data input.
  • Lines 116-122: What are the A’biome values used in the study, especially for cropland? How were these values determined? Suggest providing units for all terms in Eq. 1 here to help readers understand the physical meaning of the equation and the relationship of the terms included.
  • Line # 138-139: Please explain how this range of ‑147% to 69% was derived and its potential impacts on emission estimates.
• Lines # 172-183 (2.3. Brute-force and OSAT): It is stated in the Introduction section that brute-force method might be inappropriate given the strong nonlinearity of the ozone chemistry but nevertheless it was used to simulate ozone concentration between the base case and a scenario case without soil NO emissions. Some discussion is warranted.
• Lines # 200-201 (Fig. 2): I assume the authors used the same values of A’biome from Hudman et al. (2012). In this case, it would be 0.62% of available N in soil or 1.5% of fertilizer N applied for cropland. However, Fig. 2 indicated approximately 10% of fertilizer N was emitted as soil NOx (judged on scales of the color schemes). Please explain the discrepancy between the emission rate (1.5% of the fertilizer N applied) of the original paper (Hudman et al., 2012) and that of this study (~10%). Additionally, NOx is only one of many gaseous N species (NH3, N2O, NOx, and N2) emitted from fertilizer N from soils. Significant leaching/runoff losses (about 30% per IPCC defaults) exist. What are the expected N balances given the fact there are other more prominent N gas emissions with higher emission rates (NH3, N2O, and N2), leaching/runoff losses, and plant uptakes?

Minor comments:

Pls use “Gg N/a” or “Gg N a^-1“ throughout the manuscript for consistency