The paper demonstrates two applications of using an existing emission inventory comparison method (Thunis et al. 2022) with ensemble concept introduced to illustrate inconsistencies in selected emission inventories. One application is for EU-wide emission inventory comparison and the other application is for local inventory emission inventory comparison. The paper is well-written and well-organized with detailed results and thorough analysis. The results are important and meaningful in terms of shedding lights on ‘problematic’ inventory with specific pollutant and sector combinations which would require attentions and check-ups from emission inventory developers.

However, the method used in the analysis lacks novelty even with ensemble concept introduced. The method has been described in detail in Thunis et al. 2022 and directly used in this paper with very limited modifications or improvements while the use of ensemble concept relates to input data not the method itself. Therefore, I would not recommend this work to be published on GMD. But I do believe the findings about emission inventory inconsistencies are important and provide insights on next generation emission inventory development. I would suggest the authors to re-organize the paper and submit to other journals which focuses more on applications and findings.

Comments on ensemble approach:

Though mentioned in discussion section, the limited number of ensemble members (3) might be problematic while typical ensemble approaches used in earth sciences in general requires more ensemble members, for example, WetCHARTS (Bloom et al. 2017b), a global wetland methane emissions dataset generated using bottom-up approach, has 18 ensemble members. Also, the approach of creating ensemble (taking median) with limited number of ensemble members, may result in selecting the same ensemble member for many [p, s] tuples so that the comparison is essentially inventory-to-inventory instead of desired inventory-to-ensemble.

Minor comments:

• The diamond diagram in the results section may cause confusion for readers, especially to those who is not familiar with Thunis et al. 2022.
• Line 345: Is this line a separate figure caption?
• Figure 1 and following figures: what are the numbers in parentheses next to each legend items?

Thunis et al. present in their manuscript an ensemble (median) inventory to assess the quality of emission inventories. This new inventory is used to screen differences in terms of country totals, country sectorial share and share of the country emissions to the urban areas for emissions of several pollutants. For the screening, they use the method they already presented in Thunis et al. (2022).

There are clear differences between the current study and the former publication of Thunis et al. (2022). However, here no new technique is provided. What the authors actually provide is a new dataset that can be used to assess other emission inventories. Thus, I think that this manuscript does not fulfill the requirements for a publication in Geoscientific Model Development, but rather to a journal with focuses on data provision and/or application/evaluation of data sets.

Further, the paper is not well written. Without reading the previous paper (Thunis et al., 2022) it is impossible to understand what has been done here; especially, which parameters are used for the assessment of the emission inventories and how to read the diamond plots. The paper, especially, the results section, reads more like an scientific report than a scientific publication. Therefore, I would suggest to reject the current version and encourage the authors to revise and resubmit their manuscript to another journal.

General comments:

• The whole manuscript needs to be significantly revised. The abstract has not the structure an abstract for a publication should have and the result section reads more like a scientific report. The introduction does not give a clear overview over the topic. Further, all abbreviations need to be introduced and a list of abbreviations should be provided in the appendix since quite many are used here. I will provide more detailed comments under specific comments.
• I think the usage of the term “ensemble” in your context is not correct. Here, it is rather confusing. Usually the term “ensembles” describes the performance of a model simulation several times with different initial conditions. However, what you are doing here, to my understanding, is averaging a certain number of emission inventories and providing a median value of these inventories that is screened (beforehand or afterwards?) to keep a certain quality standard. Thus, I would pick a different term than ”ensemble”.

Specific comments:

P1, L21ff: The method part is too complicated  and should be shortened. State more clearly what has been done in your study and what can your data be used for. I would suggest the following structure for your abstract: Start with why are emission inventories important, what are they used for and then describe what has been done in this study and what are the major results.

P2, L53: The first sentence is hard to understand. Please rephrase. As stated in my general comments, what you are doing here is far away from what is done when ensemble simulations are used. Thus, this sentence rather confuses than helping to find an introduction to your study or the topic of emission inventories.

P2, L53ff: Start the introduction with how are emission inventories created, which ones are existing (give a short overview over the most important/known ones including references). What are the uncertainties of these inventories. What can be done or has be done to reduce the uncertainties and then describe your previous work and what has been done in this study.

P3, L113ff: Here, you describe how you compare the inventories, but shouldn’t you first describe how you create your inventory?

P5, L200: How is the ensemble calculated. What actually do you understand under “ensemble”?

P6, L221: Not clear, if you mean here the whole inventory or only a specific value.

P6, L232: Chose a more precise section title than just “Input data”.

P6, 264-266: Sentence not clear. I think there is a doubling of what the difference between the two versions is.

P8, L308: Repeat here once again the GFNR sectors.

P8, L310: Provide here an example or give more explanations what is meant with polygons.

P8, L325: The abbreviation ECI has not been introduced and it should be more clearly stated what this parameter is used for or what it is describing.

P9, Figure 1: More explanation/guidance is needed to read the diamond plots. Further, in my opinion there is too much information in one figure.

P9, L352-353: What is given by the numbers in parenthesis? The ECI? Also here more guidance is needed what the respective values mean.

P11, L412: It would be much more helpful if you could give the inconsistencies in percent. I still don’t know how to judge these numbers. How many inconsistencies could be there?

P13, L463: Since you do the application for Poland I would suggest to add Poland to the title.

P13, L464: I would suggest to chose a more precise title than just “Input Data”. Name the data sets used here, thus “The Central Emission Database”.

P13, L465-466: Why do you compare Poland with Europe? Wouldn’t this be like comparing apples with pears? Are you selecting the Poland area from the European inventory to make it a correct comparison? Also here more explanations are needed.

P14, L492: I already have forgotten what is denoted by A, B, D, E and J emissions. Which areas these letters are referring to could be repeated.

P14, L503: Which cities and sectors have been chosen? Are these all located in Poland?

P14, L516: A better section header than just “Results” would be “Comparison of the CED inventory to the Ensemble”.

P17, L540-541: What do you mean here? Is that a repetition again? Check sentence and correct.

P22, L753ff: Since you already have more than one page discussion, I do not understand why you need additionally more than one page of conclusions. The main message and key points of your study are not coming through. These two sections need to be significantly improved.

P23, L767: Simple manner? You have not described at all how a user could download and use your data set. What requirements are needed, e.g disk space, software etc.?

P23, L792:” …..whereas are available on country level”. This sentence is not correct. Please check and rephrase.

P23, L796: “The latter point is key” -> Please rephrase.

Technical corrections:

P6, L269: till nowadays -> up to date

P7, L288: Abbreviation GFNR already on P6 introduced.

P8, L295: Abbreviation “EEA” introduced?

P8, L296: Same holds for “TNO”, has this abbreviation introduced?

P8, L324: emissions ratios -> emission ratios

P9, L347: No underlining of text is used in Copernicus journals.

P10, L370: Has the abbreviation “LPT” been introduced?

P10, L376: Same for IEA.

P10, L378: of the -> for the

P11, L398-399: Have the abbreviations “FAS” and “PMco” been introduced?

P11, L405: appear -> appears

P11, L408: Remove underlining of text.

P11, L408: a factor 15 -> a factor of 15

P12, L438: Remove underlining of text.

P13, L450: what reported -> what is reported

P13, L473: localization -> location?

P14, L503: add “the” before “ensemble”

P14, L503: I would rather use here “for” than “is” -> for 14 cities

P17, L534: Sentence not correct. Please rephrase.

P19, L621: Add section number.

P20; L655 and 663: Remove underlining of text.

P21, L679 and 690: Same here.

P23, L789: be also -> also be

P23, L794: Full stop is missing.