The paper introduces the first SO₂ column retrievals from the new FY-3F/OMS instrument using the DOAS method, leveraging the TSIS HSRS solar reference and incorporating essential calibration and background correction steps. While the initial validation against TROPOMI provides a strong foundation for the OMS SO₂ product, the manuscript would benefit from clearer explanations of uncertainties and the implications of its current simplifications. Below are specific comments and questions for further consideration.

1. The authors refer to this as the “first version” of the OMS SO₂ It would be useful to clearly state this in the title or abstract to manage reader expectations about its development stage and limitations.

2. The paper (L95-96)recommends filtering OMS data for SZA < 70°, near-nadir, and cloud-free pixels. Were these recommended filters (including cloud-free) applied to the OMS data before comparison with TROPOMI? Please clarify the exact filtering used, especially given the statement about no cloud products.

3. The paper mentions the use of the TSIS HSRS hybrid solar reference spectrum instead of OMS L1 solar irradiance due to degradation. Could this lead to systematic biases in the long term?What plans are in place to account for this degradation?

4. The results in Figure 2 from using the 325–335 nm and 360–390 nm windows alone appear unconvincing. While these windows might be noisy in isolation, their value (as seen in TROPOMI's algorithm) lies in their combined use with stronger windows to avoid saturation at high SO₂ Since OMS shows saturation and underestimation for high SO₂, have the authors explored a multi-window fitting strategy to improve this?

5. Section 3.3.2 states a single fixed Ring spectrum was used. However, Figure 19 clearly shows the Ring spectrum varies significantly with atmospheric and viewing conditions. Please clarify this apparent contradiction: was a single fixed Ring spectrum or a variable one used? If a single fixed spectrum was used, the significant error introduced by ignoring these variations should be quantitatively discussed andthe actual Ring spectrum settings should be reflected in Table 3.

6. Whatimprovements are considered at orbit edges in a future update?

7. For the Persian Gulf, the 0.5-0.6 correlations are not particularly high. What does this imply regarding the agreement between the OMS and TROPOMI products, and what factors contribute to these differences?

8. How much does the difference of the overpass time between OMS and TROPOMI contribute to the observed differences in SO₂retrievals, particularly in dynamic regions like volcanic or strong anthropogenic plumes?

9. The paper highlights the simplified AMF approach as introducing significant errors and large uncertainties. Could the authors elaborate further on the estimated magnitude of biases introduced by these simplifications across a range of atmospheric and surface conditions? Furthermore, to enhance the accuracy and robustness of the product, particularly in complex regions like urban areas or near industrial sources, what are the plans for incorporating a more physically-based AMF calculation?

10. The current retrieval does not account for cloud and aerosol effects, which can introduce significant biases, particularly for boundary layer SO₂. What are the plans to integrate cloud and aerosol effects into future updates of the retrieval algorithm?