This manuscript investigates 2015−2023 trends of major air pollutants in South Korea using the AirKorea surface network and satellite observations. The results indicate that further NO_x emission decreases in Korea will reap benefits for both O₃ and PM_2.5 pollution. The figures are well prepared, and the analyses are relatively sound based on the results. This research's quality and scope are suitable for publication in ACP. However, the manuscript still requires revision to ensure a high-quality analysis that meets ACP standards, subject to the following concerns.

Major Comments:

1. The manuscript analyses trends in major air pollution in Korea using multi-source data, including satellites, ground-based observations, and emission inventories…, but there are significant differences in trends between the multi-source data that need to be clarified. (a) Authors claim CO trend observed by MOPITT decreased slower than surface concentrations because of the background contribution to the CO VCD (Line155-156), why there is a consistent downward trend in MOPITT and surface concentrations in the period 2015-2018 and a huge difference in their downward trends in 2019-2023, it is clear that there is more than just the effect of background concentrations here. (b) Surface SO₂ concentrations and OMI VCDs have decreased at similar rates but there are differences (Line173), For example, SO₂ observed by OMI rises significantly in 2019-2020, and SO₂ observed by both OMI and GEMS rises in 2022-2023, whereas CAPSS and AirKorea only show a downward trend, and these details should be clarified. (c) I really don't understand why the diurnal variation of NO₂ VCD observed by GEMS in warm season (8-11:00 local time) and cold season (10-13:00 local time) is opposite to that of surface NO₂ (Figure 4e). The authors try to explain this phenomenon by using the variation of the mixed layer height, it is insufficient. Besides, the high NO₂ concentration in the morning and evening is affected by meteorological conditions. Vehicle emissions during the morning and evening rush hours are also an important factor. NO₂ is mainly concentrated near the surface and rapidly photolysis after sunrise, and the satellite and the surface observations should show similar diurnal trends, which can be confirmed by previous observations in some mega-cities (Tian et al., 2018) and background stations (Cheng et al., 2019). What's more, an observation from the GEMS also showed that NO₂ column concentrations began to decline at 10:00 (local time) (Xu et al., 2023). I recommend first comparing the GEMS and surface NO₂ concentrations on an hour-by-hour basis, and then carefully analysing the reasons for the opposite trend.
2. Line288-289 “Based on the criteria from Duncan et al. (2010) the positive trend in RFN implies that Korea is now mostly in the NOx-sensitive regime (RFN > 2).” In order to avoid the misjudgment of O₃ formation sensitivity caused by arbitrary selection of FNR thresholds, I strongly suggest using a third–order polynomial model to investigate the empirical relationship between FNR and surface O₃ concentrations, which has been widely used in other studies (Ren et al., 2022; Jin et al., 2020). The criteria presented in Duncan et al. (2010) may not be applicable to the current diagnosis of O₃ formation sensitivity, the threshold is usually small (1 and 2), which causes the contribution of the NOx limit regime to be overestimated.

Minor Comments:

1. Line 57-58 “Synoptic meteorology and transport from China also contribute to seasonal and long-term variations of pollutants over Korea.” Missing relevant references.
2. Line 204-205 “Both surface and column NO₂ are higher by a factor of two during the cold season, which can be explained by the longer NO_x lifetime (Shah et al., 2020).” Differences in warm- and cold-season emission patterns should have a greater impact.
3. Line 242-243 “but CHOCHO shows hotspots for manufacturing industries while HCHO shows hotspots for petrochemical facilities.” Unclear HCHO shows hotspots for petrochemical facilities, since HCHO observations are also more distributed, HCHO didn't just indicate petrochemical facilities.
4. Line 262-263 “has been previously reported as systematic low biases in satellite observations of CHOCHO and HCHO.” Please specify it.
5. GEMS is observed every hour during the day and the time should be clarified. For example, in Fig. 3d, does GEMS use all the observations during the day or just a certain hour of the mid-day.
6. Figure 5g “OMI CHOCHOÍ20”, Does it mean 20 times magnification? This should be clarified in the legend.

 Suggestion:

Although well known, some instrument name abbreviations should indicate the full name when they first appear, i.e. OMI, TROPOMI, MOPITT…

Reference:

Cheng S, Ma J, Cheng W, et al. Tropospheric NO2 vertical column densities retrieved from ground-based MAX-DOAS measurements at Shangdianzi regional atmospheric background station in China[J]. Journal of Environmental Sciences, 2019, 80: 186-196.

Tian X, Xie P, Xu J, et al. Long-term observations of tropospheric NO2, SO2 and HCHO by MAX-DOAS in Yangtze River Delta area, China[J]. Journal of Environmental Sciences, 2018, 71: 207-221.

Xu T, Zhang C, Xue J, et al. Estimating hourly nitrogen oxide emissions over East Asia from geostationary satellite measurements[J]. Environmental Science & Technology Letters, 2023, 11(2): 122-129.

Ren J, Guo F, Xie S. Diagnosing ozone–NO x–VOC sensitivity and revealing causes of ozone increases in China based on 2013–2021 satellite retrievals[J]. Atmospheric Chemistry and Physics, 2022, 22(22): 15035-15047.

Jin X, Fiore A, Boersma K F, et al. Inferring changes in summertime surface Ozone–NO x–VOC chemistry over US urban areas from two decades of satellite and ground-based observations[J]. Environmental science & technology, 2020, 54(11): 6518-6529.

General comments

Oak et al. present a valuable analysis of surface, airborne and satellite measurements that provides important insights for air pollution management in South Korea. The text is well structured, presenting the trends over 2015-2023 for various air pollutants in turn, including insights on O­₃­ production regime and the sensitivity of PM_­2.5­­ NO_­3­^- formation, before highlighting the implications for air quality policy. The work could be strengthened by including more recognition of the differences between and limitations of the various datasets, e.g., time of day of the measurement, sensitivity to the lower troposphere or uncertainty associated with a given dataset. The explanations of the drivers of change in the pollutants are also, on occasion, incomplete.

Specific comments

1. Line 16: Include a statement on the context/background of the research at the start of the abstract.
2. Line 59: Specify how the summer monsoon impacts O­₃­. The O_­3­ decreases after June due to the monsoon doing what? The current implication of the sentence is that the monsoon causes a peak in May-June O­_3­.
3. ­Lines 74-77: Restructure to first introduce O_­3­ sensitivity to VOCs versus NO_­x­ and then the relevant trace gas ratio. Similarly, for NO_­3­^- You could also expand on why O­₃­ and NO_­3­^- are sensitive to these specific compounds earlier in the paragraph.
4. Line 87-95: This paragraph would be a good place to highlight the purpose and novelty of your work in a brief statement.
5. Table 1: What about the temporal resolution or overpass time of the satellites? The LEO orbit satellites will be measuring at a specific time of day over South Korea, so they could be catching the daily min or max values. How does this compare to the times other measurements are available for? A discussion of how this might affect the differences in the results for different datasets is missing. This could be included in the relevant results sections or section 2.
6. Line 149: Expand on how the topography affects the CO along the east coast
7. Line 158: The 2019 spike seems quite small – is it greater than the uncertainty of the data?
8. Line 165: Specify what the continuing motivation for SO_­2­ emission controls is
9. Lines 174-177: Link the other studies' results back to your findings, e.g., are they consistent, what are the implications of the different sources of SO_­2­
10. Line 218: Clarify why the transportation contribution may be a severe underestimate
11. Line 250: “values in Korea are higher everywhere” is inconsistent with previous statements. Below 0.03 can be greater than 0.02.
12. Line 254: Although there is no significant trend in surface BTEX, can you comment on the higher values over 2019-2021? Is this within the data uncertainty or a significant signal?
13. Line 269: Can you comment on why the satellite data do not show the late afternoon rise?
14. Line 279: This seems to be the only result for 2005-2014 in the paper. Is it relevant to the rest of the work? If not, I would suggest removing it.
15. Line 279: Can you comment on the O_­3­ change between 2019 and 2020?
16. Lines 301-305: Link the US data back to your results, otherwise they just read as additional, slightly random, facts.
17. Figures 2-9: It would be useful to see some measure of uncertainty or error on the line graphs, or a statement on the associated uncertainty in the main text.
18. Additional detail on data analysis could be added to the supplement or processing scripts shared in a code availability section

Technical corrections

1. Line 40: Clarify ‘Subsequent atmospheric chemistry (of these trace gases?) produces’
2. _­Line 68-69: I would change ‘would respond nonlinearly’ to ‘responds nonlinearly’, as this a general statement
3. Line 79: clarify you are listing the relevant LEO instruments
4. Line 116: “We do not use them here.” Be explicit that the O_­3­ measurements are what is not used. This sentence and the previous one could be combined for clarity: “(…) are inconsistent over Korea (Gaudel et al., 2018), therefore we do not use them here.”
5. Line 135: Explain ‘SMA’ acronym in the main text
6. Line 144: ‘plays an important role in driving ozone formation’
7. Line 169-170: I suggest rephrasing to “(…), consistent with OMI SO_­2­ hotspots previously identified for 2011-2016 (Chong et al., 2020).” for easier reading.
8. Line 182: replace ‘accounting’ with ‘which account’ for clarity
9. Line 184: add ‘the potential’ to match ‘motivated by’: “but also the potential to reduce PM2.5”
10. Line 185: replace ‘diesel engines in 2016’ with ‘diesel engines since 2016’
11. Line 198: “CAPSS shows an increase”
12. Line 202: “additional information on the diurnal variation of NO_­2­­”
13. Line 320: “and is at its minimum in summer”
14. Line 327: “PM_2.­5­ observations in Seoul show­” (not shows)
15. Line 386: “(…) component not found to show”
16. Line 393: “­for in terms of decreasing O­_3­”