Review of Okuljar et al., 2023

Okuljar et al. present measurements of HOMs and other condensable vapors from measurements at two sites in 1 km distance from another in Helsinki. The manuscript is well written and uses state-of-the-art scientific methods.

Given that there is still a lot that the atmospheric chemistry community does not know about the formation of HOMs/SOA, these observations in a suburban area provide an important addition to the literature. What I find the most interesting and striking point of this study is the strong difference between the composition and diurnal behavior of observed compounds during the same time, although the two sites are spatially so close to each other. This points to an important lesson for the atmospheric chemistry community regarding the (non-)representativeness of a single observation site for a whole metropolitan area, and generally to a very localized inhomogeneity of emissions, reaction processes and products. Therefore, I wish this point would be brought across more strongly, perhaps graphically. Thus see my comments below.

I recommend this article for publication in ACP after the following comments have been addressed:

General comments:

1. Did you see any evidence for cooking emissions contributing to any of the observed PMF factors? E.g. Zotter et al. (2014, https://doi.org/10.1002/2013JD021114, 2014) mention that at least 25% of the non-fossil SOA is from cooking.
2. You report that only few PMF factors correlated with each other between the two sites. I wonder if some of the reason may be that the air transported from one place to the other by prevailing winds takes some time, so that the correlation would be shifted in time, and would require a method like time-warping to find correlations? Also, was the PMF conducted separately for the time where there is overlapping data between the two sites? I wonder if there could be a bias if the latter is not the case, since both sites were measured for a longer time separately than simultaneously.
3. How deep is the “street canyon”, i.e., how high are the surrounding buildings? If they are high: Do you see any evidence that the street canyon site has more stagnant air, i.e., that emissions spend more time there so that local emissions reach higher oxidation states, vs. a more well-ventilated suburban site, where observed HOMs may stem from longer range transport? I think the discussion lacks a little bit of consideration that local emissions and reactants may not be the only reason for the differences observed between the two sites. Please discuss meteorology/transport.
4. I suggest adding a graphical depiction of the amount of overlap between the observed signals between the two sites since this is such an important message. Maybe a pie chart showing the fraction of total ions (sum of both sites) only observed at SC, fraction only observed at UB, fraction observed at both sites? E.g. similar to Fig. 2 of https://doi.org/10.1021/acs.est.2c07260 .

Specific comments:

l.44: apart from the listed sources, I suggest to mention cooking since it is also a strong source of anthropogenic VOCs/ condensable vapors.

Fig. 1: It would be helpful to add a wind rose to see which of the two stations is up-/downwind of each other.

Sect. 2.5 regarding transmission: I assume the signals are corrected for transmission/normalized to transmission - I think this should be mentioned for clarity that the signal strength is not impacted by the transmission issue between the two instruments.

Fig. 2: I think it would be helpful to add two panels showing the same data just for the time when there were simultaneous measurements at both sites. Otherwise, it is impossible to see how comparable the conditions (temperature etc.) are between both sites during this relevant time.

l.313: Since you mention monoterpene monomer factors with different diel variabilities – was there any diel variability in the monoterpene composition measured with sorbent tubes that could help to find out which monomer stems from which monoterpene or which part of the monoterpenes is anthropogenic/biogenic?

l.316-369: I think it would be interesting to mention the most important compounds/chemical formulas observed in each factor as you do for some, but not all that factors you describe here.

l.495: “This is clearly different from Helsinki” – I am not sure you can make that statement in the way you are making it, since you have only a few weeks of observations in one season and for China you are mentioning variability between seasons. It would be more appropriate to compare concrete cases with concrete numbers (which percentage of condensable vapors in Helsinki is biogenic, vs. which percentage is biogenic in the same season in a Chinese city?).

Technical comments:

l.145: “only we can narrow it usage” – this phrase is unclear to me.

l.195: missing “the” before “identified”

Fig. 3 caption: It was not initially clear to me that “SC-1” stands for street canyon, factor 1. Please make the meaning of the numbering clearer in the caption.

l.412-413: This sentence is not clear to me, it seems as if both grammatically and content-wise something is mixed up in there?

l.482: lower resemblance than what?

l.490: “influenced” should be “influence”.

l.507: “particular” should be “particulate”.

General comment:

The authors present an interesting data set, focusing on the factors governing HOM/SOA formation in biogenically influenced urban areas.  In general, the manuscript is well structured, and the authors clearly communicate their approach and goals, as well as the limits of their analysis. Therefore, I would recommend this manuscript for publication with the minor re-works and additions outlined below.

Specific comments:

-As the transmission is one differing factor between the two employed CIMS instruments, I suggest to include details about the instrument transmission curve determination as well as the transmission curves themselves in the supplement.

-One aspect that intrigued me and that I would like to have seen discussed further is the possibility of NO3 oxidation with subsequent termination by NO (see lines 347-352). Was this observed before or are there any model calculations estimating the likelihood?

-In Section 3.3 the implications for air quality are discussed, and though the section gives a good quick overview over the general present understanding of SOA toxicity if it is intended as it’s own section I would like to see the actual results from this study included more. I.e. how could the major contribution of ONC affect toxicity or how does the change in SOA yields due to NOx termination impact the exposition to SOA?

-Line 151: What reagent ions were used for normalization? (NO3- / HNO3*NO3- / (HNO3)2*NO3-? Inclusion of water clusters of NO3-?)

-I would cut back on the description of what will be shown in Section 3.2. Lines 278-285 give information that is partially repeated later and I think the section would profit from shortening this description.

-For Fig.3 and 4 I recommend shifting the full timeseries to the supplement. In my opinion the diurnal trend plots give a much better impression of the important influences in one PMF factor. Furthermore, the full timeseries are not extensively discussed in the manuscript itself.

-On a minor note, I would also be interested what the gaps in the timeseries of the factors mean, since sometimes it seems that days with missing data (for example Fig. 4 May 14) just linearly interpolated over the missing period. Furthermore, I would mention the scaling of the temperature axis as the non-linear scaling can be a bit misleading.

-For Table 1: I would be interested to understand why the importance of species was judged by different methods (factor time series, or factor mass spectrum or both) and which criteria were used to determine the chosen method.

Technical corrections:

-Lines 65-67: “Additionally, NO2 can terminate oxidation chain in reaction leading in most cases, which decompose back to substrates (Atkinson and Arey, 2003).“ This sentence needs to be reworked.

-Line 72: “attach” instead of “attached”

-Line 229: remove “of”

-Line 484: include to in “compared to the”

-Line 490: “influence” instead of “influenced”