In elucidating the behavior of nitrate (NO₃⁻) contained aerosol, research using stable isotope ratios has been actively conducted in recent years. In particular, the study of the δ¹⁵N (NO₃⁻) and Δ¹⁷O (NO₃⁻) of PM_2.5 in Seoul from 2018 to 2019 to quantitatively estimate the oxidation pathways of particulate NO₃⁻ and the source apportionment of major NOx emission sources is a very innovative and academically valuable study. However, even though discussing dual isotopes, they were not being discussed effectively. In addition, several issues were raised, such as the lack of clarity in the explanation of many of the discussions. Therefore, it could not be accepted by Atmospheric Chemistry and Physics international journal.

Major points

1. The studies on the environmental kinetics using Δ¹⁷O (NO₃⁻) and the source analysis using δ¹⁵N (NO₃⁻) were not organically linked. For example, how was the source contribution affected by the use of Δ¹⁷O (NO₃⁻) in the source analysis? How did the use of Δ¹⁷O (NO₃⁻) to elucidate the environmental dynamics affect the source contribution? It be made clearer how it was clarified and how the findings differed from past studies?
2. The estimated source contribution should be discussed more quantitatively from the viewpoint other than isotope, because biomass burning is very large. Research on isotopes is still unexplored, and it is thought that it will be put to practical use only after such discussions.
3. Many of the figures were characteristic figures, and many of them were very difficult to understand.

In addition, the other points are shown below.

Introduction part: It would be better to discuss isotope-related research and other research separately. In addition, Line 40 is an introduction to isotope-related research, and these should be discussed separately.

Lines 89-97: Please move to the method part.

Lines 98-113: The introduction of the research on Δ¹⁷O (NO₃⁻) should be described more clearly. Please rewrite.

Line 123: Table 1 should be added in line.

Lines 147-157: For the analysis method of Δ¹⁷O (NO₃⁻), it is stated to refer to Morin, 2009, but more details should be written.

Line 156: Is the standard deviation the mean of the standard deviations, or does it mean within that number? Please describe clarify.

Lines 162-164: It is not clear whether it is two steps or three steps. Please make the description clearer.

Line 267: Shouldn't source data be added here?

Section 3.1: There were few comparisons with previous papers, and more consideration should be given to the reliability and characteristics of the data. In particular, the discussion of isotopes after line 306 should be revised, since it only compares figures and does not refer to references.

Line 320: high relative humidity? Please describe clearly and statistically.

Line 452-Line 461: This part should be included in the section3.1? Isn't it better to summarize the discussion of source analysis in this section?

Line 476-479: Please move the method part.

Line 479-482: I would like you to clearly describe the discussion on the difference between the case where fractionation is considered and the case where it is not.

Figure 2: The figure was very difficult to understand. Rather than this figure, a scatter diagram may provide clearer discussion.

Figure 3: The figure was very difficult to understand. The comparison of sources is made, but is it the data of Rose, 2019? Do you consider isotope fractionation?

In this study by Lim et al., the authors measured the D17O and d15N of atmospheric nitrate particulates in Seoul, Korea during two different seasons and used this isotopic information to understand the oxidation formation pathways of atmospheric nitrate and to source partition NOx emission sources. The basic rationale is to estimate the relative fractions of the three major nitrate oxidation pathways using the conversative D17O tracer and then use these fractions, along with the relevant nitrogen isotope effects (mostly derived from theoretical calculations), to re-construct the original d15N of ambient NOx. This isotope effect-corrected d15N-NOx signal can then be combined with d15N-NOx source signatures to source partition local NOx emissions. While I found this research important and timely, and that the analysis and discussions related to the D17O of nitrate are thorough and convincing, there are multiple flaws and ambiguities in the discussion of the d15N data.

First, nitrogen isotope effects that are important for the conversion of gaseous NOx to particulate nitrate are not clearly discussed and treated during the data analyses and discussion. The first of those is the nitrogen isotope effect for the Leighton NOx cycle. The authors used a lumped isotope effect of 1.018 to represent the combined kinetic and equilibrium isotope effect for the photochemical NO-NO2 cycling. However, this value was determined by simultaneous d15N-NO2 and d15N-NOx in a very different atmospheric environment (in terms of NOx concentration, NO-to-NO2 ratio, and O3 concentration; i.e., Julich Germany) (Freyer et al., 1993). How this value is representative to the study site of this work is unknown. More importantly, this value is an annual average, without considering temperature effects. A following laboratory study confirming this value (Li et al., 2020) was also conducted at room temperature. Because seasonal temperature fluctuations likely have strong effects on nitrogen isotopic fractionations in the atmosphere (Walters and Michalski, 2015), these impacts need to be acknowledged and considered during the nitrogen isotope data correction. This is a particularly important issue for this study as the temperature effects are expected to manifest to different extent in the data collected during the summer and winter seasons.

It is also not clear how the isotope effect for the NO2-OH oxidation was represented during the data analysis. In several places, the authors claimed that this step exerts no isotope effect; in some other places, a small isotope effect of -3 permil was mentioned (e.g., Line 182-184). The empirical and/or theoretical grounds for these deductions were not sufficiently stated.

Furthermore, the isotope effect for the conversion from gaseous HNO3 to particulate nitrate was completed ignored. A previous study found that the isotope effect for this gas-to-solid conversion step dominated the variability of atmospheric nitrate aerosols, was dependent on aerosol pH, and can be potentially as large as 22 permil (Geng et al., 2014).

In general, I found more work is needed to better consider the large uncertainties surrounding the nitrogen isotope effects during the d15N data correction. While some of these uncertainties have been mentioned in this manuscript, failures to incorporate these uncertainties into the NOx source partitioning invalidates the partitioned results. Consequently, these results should be deemed only qualitative, but not quantitative. In this sense, a thorough sensitivity analysis, like the one used to treat NOx source signatures, is required to tackle these uncertainties associated with the nitrogen isotope effects.

Minor comments:

[1] 87-89: The mention of negative fractionation effect here is very confusing. First, you need to define what a positive (or normal) isotope effect is and then to assign either a positive or a negative sign to indicate positive/negative isotope effects.

[2] 172-179: Throughout the manuscript, it is not clear how this A factor was derived using in situ data and used to correct d15N data.

[3] 182-184: this sentence is difficult to follow. Did you assume no fractionation during NO2-OH oxidation or use -3 permil for this step? Need more clarifications.

[4] 264: I understand these source endmembers were presented somewhere else in the paper. But to ease readability, I would suggest including several sentences or a table here to summarize or contrast the different d15N ranges.

Reference:

Freyer, H.D., Kley, D., Volz‐Thomas, A. and Kobel, K., 1993. On the interaction of isotopic exchange processes with photochemical reactions in atmospheric oxides of nitrogen. Journal of Geophysical Research: Atmospheres, 98(D8), pp.14791-14796.

Li, J., Zhang, X., Orlando, J., Tyndall, G. and Michalski, G., 2020. Quantifying the nitrogen isotope effects during photochemical equilibrium between NO and NO 2: implications for δ 15 N in tropospheric reactive nitrogen. Atmospheric Chemistry and Physics, 20(16), pp.9805-9819.

Walters, W.W. and Michalski, G., 2015. Theoretical calculation of nitrogen isotope equilibrium exchange fractionation factors for various NOy molecules. Geochimica et Cosmochimica Acta, 164, pp.284-297.

Geng, L., Alexander, B., Cole-Dai, J., Steig, E.J., Savarino, J., Sofen, E.D. and Schauer, A.J., 2014. Nitrogen isotopes in ice core nitrate linked to anthropogenic atmospheric acidity change. Proceedings of the National Academy of Sciences, 111(16), pp.5808-5812.