This manuscript presents a comprehensive investigation of organic aerosol (OA) composition, sources, and oxidative evolution in Hong Kong, utilizing a combination of high-resolution analytical techniques, including HR-AMS, TAG-TOF-MS, and HPLC-HR-MS. The authors demonstrate that during the COVID-19 lockdown period, low-oxidation OA was primarily influenced by biomass burning and gas-phase secondary organic aerosol (gas-pSOA), largely driven by air mass transport from mainland China. In contrast, high-oxidation OA was dominated by secondary inorganic aerosols (SIA), gas-pSOA, and biogenic SOA, with isoprene-derived SOA identified as the major contributor. HR-MS results indicate that CHO and CHON compounds dominated the OA composition, contributing over 60% of signal intensity and more than 40% of molecular formulas. Constrained-NMF analysis suggests that atmospheric oxidation promoted CHO accumulation, while nitrogen-related reactions increased molecular diversity. Overall, this study offers valuable insights into the molecular complexity and transformation of OA under different atmospheric conditions. I recommend publication after minor revision. However, several specific aspects require clarification or improvement prior to acceptance:

1. Line 20：This is an interesting and innovative approach to source apportionment at the molecular level. The authors applied a constrained NMF model to offline HR-MS data to investigate molecular differences among various OA sources. However, the specific differences identified between these sources are not clearly presented. It is recommended that the authors elaborate on the key distinguishing features—such as dominant compound classes, oxidation levels, or representative molecular formulas—that characterize each OA source.
2. Lines 175-180：The authors suggest that LO-OOA1 is more influenced by anthropogenic photochemical processes involving VOCs and NOx, while MO-OOA is primarily linked to biogenic SOA formation through ozone oxidation. However, the distinction between their "primary nature" and "secondary formation pathways" remains somewhat ambiguous. Could the authors clarify how LO-OOA1 is considered to have a more primary character despite its apparent link to secondary photochemical reactions?
3. Lines 179：The authors state that MO-OOA shows good correlations with O₃ and biogenic SOA tracers. Could the authors provide the specific correlation coefficients (e.g., r values) and p-values to quantitatively support this statement, particularly for the relationship between MO-OOA and O₃?Including these statistical details in the main text or supplementary figures would enhance the transparency and robustness of the analysis. The same places that need to be modified are in Lines 198 and 215.
4. The x-axis in Figure 2a seems to be incorrect, and the symbol/icon used in Figure 3 is unclear. Please revise it to improve clarity and readability.
5. Lines 235-238: The authors state that SIA contributed the most to MO-OOA (38%), yet also suggest that highly oxidized organics in MO-OOA might not be primarily related to SIA processes. This appears somewhat contradictory. Could the authors clarify how they reconcile the high SIA contribution with the conclusion that MO-OOA is more closely linked to biogenic SOA processes?
6. Line 370-375: The authors note that the contribution of gas-pSOA estimated by the constrained-NMF method was 1.4 times higher than that from the tracer-based PMF model, while the SIA contribution was significantly lower. Could the authors clarify how they addressed this discrepancy, and whether any compound misclassification between SIA-related and gas-pSOA-related species may have biased the NMF results?

Comments on egusphere-2025-2264

In this study, Jiang et al. utilized advanced analytical techniques, including high-resolution AMS, TAG-TOF-MS, and HPLC-HR-MS, to investigate the composition, sources, and evolution of OA in Hong Kong. Their findings reveal that during the COVID-19 lockdown, biomass burning and gas-phase secondary organic aerosols (gas-pSOA) were the primary contributors to low-oxidation OA, influenced by air mass transport from mainland China. In contrast, secondary inorganic aerosols (SIA), gas-pSOA, and biogenic SOA accounted for over 90% of high-oxidation OA, with isoprene-derived SOA emerging as the dominant contributor to the high oxidation levels. HR-MS analysis showed that CHO and CHON compounds were key components of OA, accounting for more than 60% of the total intensity and over 40% of molecular formulas, respectively. This indicates that atmospheric oxidation processes led to the accumulation of specific CHO compounds, while nitrogen addition reactions increased the diversity of the molecular composition.

A key innovation of this study lies in addressing the challenges of interpreting factors in PMF analysis due to the complexity of atmospheric chemical processes and the diverse sources of OA. By integrating molecular composition data into a non-negative matrix factorization (NMF) model constrained by PMF factors, the study improved the interpretability of these factors and shed light on the molecular composition differences among sources and formation pathways. Overall, this research employed robust analytical methods, provided comprehensive discussions, and delivered novel insights into the sources, formation mechanisms, and regional air pollution associated with OA. The manuscript aligns well with the journal's scope, and I recommend its acceptance following minor revisions.

Specific comments

Line 43: This sentence is not well organized. Please rephrase them.

Line 159: Capitalize the first letter “yet”.

Line 176-178: This long sentence is suggested to be broken down into multiple short sentences to enhance the logic of the sentence “.

Line 179: “Good correlations between MO-OOA and O₃…”, please list what is their correlation coefficient (R²).

Line 259-260: The subject of the sentence is long and begins with "Although", making the logic seem complicated. Please simplify it.

Line 263: The word "formular" is misspelled; it should be "formula".

Line 316: “might not entirely belong” This phrase is vague.

Line 370: Please specify what constraints were applied in the NMF model.

Line 391-392: The phrase "IsopreneSOA being dominant contributor" is missing the article "a" before "dominant contributor." It should read: "IsopreneSOA being a dominant contributor..." for grammatical accuracy.

Line 400: The term "conducive to forming" is slightly awkward. Consider rephrasing to "is more likely to result in the formation of..." for improved readability.

Please see my comments in the attached file.