The manuscript “summertime tropospheric ozone source apportionment study in Madrid (Spain) by de la Paz et al. presents a ozone source apportionment study using ISAM in the CMAQ model over the Madrid region for July 2016.

While the topic of the paper is very interesting and fits into the scope of ACP, the manuscript lacks many important information and the results are discussed in an insufficient way. Therefore, I recommend to reject the manuscript. Please find below a more detailed review given some major and minor comments the authors should consider before the plan to resubmit the publication.

Reading the paper I had several major concerns which needs to be clarified.

1) The authors apply a new version of ISAM in CMAQ 5.3.2 which (to my knowledge) has not been used in similar publications before. According to the authors this new version attributes ozone to all involved precursors and NOT to the limiting precursor (e.g. NOx or VOC). Sadly, the manuscript is lacking any details on the new method. I guess the method is somehow similar to the method presented by Grewe et al., 2017? The results presented in the manuscript heavily depend on this method. Therefore, the method either needs to be presented before in a scientific publication (e.g. an update of Kwok et al., 2015 and not a youtube video) which is cited by the authors or the manuscripts needs to include a detailed presentation/discussion of the revised method.

2) Given the new method, I am missing a detailed discussion of the method and the results in comparison to previous publications ( see for example Butler et al., 2018 for a detailed discussion of many ozone tagging methods). Moreover, I am missing a critical discussion of the model results. As example, Fig.4 shows a contributions of more than 14 % of SNAP6 (solvents) to ozone over the mountain range north of Madrid. Given my understanding of the method and results of similar methods I wonder about this high contribution. How can this be explained? Moreover, I wonder about the small contribution of biogenic emissions, even though they account for a large fraction of the VOC emissions. If the method attributed ozone to all precursors, they should account for a larger fraction (?) Please clarify. In addition, the manuscript is lacking an overview of the definition of the different source attribution sectors (e.g. as table etc). From Fig 11 it seems that 12 different source sectors are considered.

3) I am missing any new results. The large importance of boundary conditions to ozone levels over the Iberian Peninsula have been reported by e.g. Pay et al., 2019. Also the larger importance of regional emissions to high ozone values have been presented in previous publications (maybe not focusing on Madrid). I like the detailed investigation of source attribution results for specific weather patterns, however, for a scientific publication in ACP more detailed analyses are needed in my opinion and the author need to highlight new findings in more detail.

4) The authors mix the physical quantities “concentration” and “mixing ratio”. They use the term concentration and use the unit ppb which suggest a (volume) mixing ratio. Please clarify the used physical quantity. Similarly, Fig. 2 does not give any physical quantities for the emissions. In addition, please clarify what emissions of NOx and VOC are. Are they given in amount of N, NO, NO2 and C or NMHC?

5) Parts of the manuscript are confusing and missing a proper proof-reading. As example, on P5l197 the authors write that Fig. 4 shows “average contributions”. The description of Fig.4, however, indicates that contribution to the 90^th percentile of ozone are given. Moreover, I find it very confusing, that the authors only show contributions to ozone attributed to anthropogenic origin. I suggest to always show contributions with respect to total ozone. Otherwise results are very hard to compare to other studies and readers might be confused.

6) In the last subsections the authors present a comparison with measurements. This comparison shows an underestimation of ozone simulated by the model under accumulation conditions during 13 -19 UTC, however the authors do not discuss this model bias. How does it affect the source apportionment results? It seems that the model underestimated local ozone production under this stagnant conditions. To my opinion, the manuscript should start with a model evaluation and discuss the source attribution results critically with respect to the model performance.

7) Given the importance of emissions from the previous day for ozone formation I wonder why the authors attribute them to “IC”. Wouldn’t it be better to account them also sectorwise?

8) Some of the reference seems to be not adequate. As an example, P3l118 cites Borgee et al., 2022 (https://doi.org/10.1016/j.scitotenv.2013.07.093), but I can’t find “tagging” nor “ISAM” in the whole paper. Maybe I misunderstood something, but the authors should check the manuscript carefully.

Some minor comments:

- Introduction: I am missing a discussion of similar source attribution studies (globally, for Europe) and a discussion of comparable source attribution methods.

- Fig 2: COVs instead of VOC

-p5l183 I wonder why the contribution of biogenic emissions is so small (see also major comments above).

- P3l109 What is the temporal resolution of the boundary conditions?

- p2l45 You mean STE is projected to increase? Please clarify.

- P2l47ff I am missing a discussion of the role of the non linearity of the ozone chemistry which lead to an increase of the ozone production efficiency when emissions are reduced. The authors should consider to add this point including a discussion of the relevant literature.

- p2l58 Please fix, should be Paoletti et. al, 2014

- p5l160 How is soil-NOx handled?

- p5l185 But Pay et al, 2019 applies the “old” ISAM tagging, right? So I would expect a difference with the new approach? Please discuss.

References:

Butler, T., Lupascu, A., Coates, J., and Zhu, S.: TOAST 1.0: Tropospheric Ozone Attribution of Sources with Tagging for CESM 1.2.2, Geosci. Model Dev., 11, 2825–2840, https://doi.org/10.5194/gmd-11-2825-2018, 2018

Grewe, V., Tsati, E., Mertens, M., Frömming, C., and Jöckel, P.: Contribution of emissions to concentrations: the TAGGING 1.0 submodel based on the Modular Earth Submodel System (MESSy 2.52), Geosci. Model Dev., 10, 2615–2633, https://doi.org/10.5194/gmd-10-2615-2017, 2017.

Please find the review comments attached.

Thanks,