Please find comments on manuscript in attached file

Review of “Historical and future changes and present-day uncertainties of ozone in China from CMIP6 models”

The manuscript analyzes historical and future changes, as well as present-day uncertainties, in surface ozone concentrations over China using multiple CMIP6 models and the TAP dataset. It evaluated the performance of nine CMIP6 models and discussed the historical changes of surface ozone over China since 1850. Through comparison of CMIP6 models under a single future scenario, the potential sources of inter-model discrepancies were discussed. Overall, this paper presents a meaningful assessment of present-day surface ozone over China and highlights potential future variations under multiple scenarios. I consider the manuscript suitable for publication after minor revisions addressing the specific comments provided.

Specific Comments

1. The abstract may require some revision. The current version reads more like a conclusion section, and restructuring it to improve clarity.
2. It would be helpful to provide additional information on the source and basic characteristics of the cloud cover data used in Section 3.3.
3. Lines 147–151: The manuscript states that “due to the limited availability of model data for the Tier 2 CMIP6 scenarios (SSP1-1.9, SSP4-3.4, SSP4-6.0, and SSP5-3.4-over), the analysis focuses on SSP3-7.0 and the Tier 1 scenarios.” However, it is unclear which scenarios are included in the MME used in the analysis. Additionally, it is not evident where the Tier 2 scenarios were applied in the study. I recommend clarifying these points in Section 2, specifying exactly which SSP scenarios are used in the MME and explaining whether/where the Tier 2 scenarios contribute to the analysis.
4. Lines 165–179: The necessity of analyzing TAP data in relation to PM₅ and its chemical components is unclear. Please clarify. If this analysis does not directly support the study’s objectives or contribute to the main conclusions, I suggest removing this part to maintain focus and coherence.
5. In Section 1 (Lines 218–224), the manuscript would benefit from a clearer explanation of why surface ozone concentrations remain high across multiple seasons in Northwest China, specifically whether the elevated levels are driven primarily by local photochemical production or by regional transport? In addition, it would be helpful to elaborate on the reasons for the comparatively lower ozone levels observed in Northeast China.
6. In Section 3.1, why UKESM1-0-LL exhibits the largest underestimation? A discussion of potential reasons would aid in interpreting the model results.
7. In Section 3.4, could the manuscript provide a more detailed discussion of the primary mechanisms through which aerosols affect surface ozone?
8. Section 5. Splitting it into several sub-sections could help improve the clarity of the manuscript.
9. Section 6 (Summary) could be made more concise.

Comments on figures and tables

1. Figure 1, the PM₅ components are difficult to discern. Since the focus of this study is on ozone, it is recommended to either remove this panel or adjust it to improve clarity.
2. Figure 2, the Bias color scheme may be misleading, as the current gradient mixes small negative and positive values. A more effective scheme would assign white to zero, blue gradients to negative values, and red gradients to positive values, enhancing clarity.
3. Figures 3 and 10. Adjusting the color or line style would improve differentiation from individual CMIP6 models and enhance figure clarity.
4. It would be helpful to include a summary table in the supplementary materials presenting key statistical metrics for each model. This would allow readers to more clearly compare model performance and improve the clarity of the results.
5. It is recommended to include basic information on the Tier 2 CMIP6 scenarios in Table 1, or alternatively provide an additional table in the Supplementary Material summarizing the Tier 2 scenarios (e.g., data availability and participating models). This would help readers clearly distinguish between Tier 1 and Tier 2 scenarios and better understand the data selection in this study.