Petrescu et al. compiled observation- and inventory-based methane emission estimates for 8 large global emitters. These data from different sources all come with inconsistent formats and sector partitions. The authors harmonized the dataset with a consistent sector partition, so different estimates can be compared properly, which is valuable. However, I have several concerned that (1) the paper contains substantial discussions that are not directly related to the dataset, which distracts the main purpose; and (2) the statement that observation- and inventory-based methane emissions are reconciled is an exaggeration and is misleading, so, I cannot recommend the paper, in its current form, to publish in ESSD. Below explains my main comments.

Main comments

1. The objective of the paper is not clearly stated. As a data paper, one would think that main objective is to describe the dataset. In the case of this paper, since the data are taken from other studies, the key is to describe how diverse data are harmonized and what the harmonized data tell us. However, Section 4 contains substantial discussion that is very general and not directly related to the dataset, which is distractive. The recommendations given in the end are random and not backed up by the findings of the paper.  I would suggest that the authors clearly state their objective and organize the content around that.

Specifically, in Line 134-135, the authors claimed that the paper "aims to inform and attract attention of the use of the results for diverse climate stakeholder needs beyond research", which is a very good statement of the objective. However, it was then not discussed anywhere in the method and results. It is not never explicitly explained what prevented stakeholders from using the existing methane emission data, why this dataset compiled by the authors would be more appealing, and what efforts have been made to make the data easy to use. Moreover, I checked out the data in the repository. The datasheets still look very complicated to me, and I am not sure that non-researchers can easily find the information they needed. Anyway, more explanations are needed if the above statement is indeed the objective of the paper.

2. "Reconciliation of observation- and inventory-based methane emissions" is used as the title and presented as the main finding of the paper. I find it an overstatement. The paper presented "total inversion methane emissions > BU anthropogenic emissions" as a discrepancy, which was reconciled by considering "posterior total flux from inversions (roughly) = BU anthropogenic flux + BU natural flux" at the national scale. However, this level of consistency/reconciliation is not surprising at all. Why would anyone want to directly compare total methane emissions from an inversion with just anthropogenic emissions from an inventory and ignore natural sources? So, I feel that the title exaggerated what was found in the paper. It is very likely that a more in-depth investigation into the data would identify significant discrepancies between the observation- and inventory- based methane emissions.

A more important questions is the reliability of the comparison made here between the inversions and inventories. Inversions are regarded as independent top-down verification of bottom-up inventories. But they are not. All of these inversions rely on prior information and therefore not independent of bottom-up emission inventories. Comparing bottom-up inventories and inversions without characterizing this dependence makes it difficult to judge whether the two are actually consistent. For example, we do not know whether the agreement of the EU emission trend from various inversions was due to strong observation evidence, or due to similar prior information used by the inversions and a weak observation constraint. We also do not know if the disagreement in the USA emission trend was due to a weak observation constraint and different prior information. If this is the case, it makes little sense to talk about the consistency between the inversions and bottom-up inventories in terms of the USA emission trend.

Minor comments:

Line 128-129: The statement indicates that achieving the climate goal will automatically lead to gains in areas of energy, food, etc., which can be misleading. In fact, controlling methane emissions may pose significant challenges to energy and food security. I suggest rephasing the statement to be more balanced.

Line 190: Spell out LULUCF at the first appearance.

Line 191: Missing information. "...which according to the ? are defined as... "

Line 199: Period sign before "Furthermore".

Line 241: What is IPPU? Spell it out and explain if necessary.

Line 291: e.g. -> i.e.

Line 301-302: It may be useful to report the rate of reduction in USA emissions, as a comparison to the EU value reported above.

Line 339: Perhaps be more specific that Russian CH4 emissions remained rather low "relative to its pre-2000 levels"? Compared to other countries, Russian emissions are not low at all. 

Line 383: What is AD? Activity data? Spell out and explain if necessary.

Line 464-473: The paragraph appears to be out of context. All the remainder of the section discusses the BU and TD comparison (in terms of both average emissions and trends), while this paragraph talks generally about sectors driving CH4 growth. 

Line 484-490: This result shows that the emission trends derived from the inversions are strongly dependent on the prior choice, indicating that the atmospheric observations used in these inversions are inadquate to constrain the emission trend. The relatively good agreement of emission trends in other countries (e.g., EU) also does not provide strong evidence, because the agreement can be driven by similar prior information.

Line 501-502: Again, this may be due to different choices of prior emissions. 

Figure 6 and Line 600: Biomass burning is considered anthropogenic in Figure 6 but natural in Line 600. I understand both anthropogenic and natural processes contribute to biomass burning. However, the current description is unclear and confusing. A clear description and terminology should be given to distinguish its anthropogenic and natural components.

Table 3 and 4: The orders of inversions are listed differently in the two tables, making it difficult to compare.

Table 4: For an inversion, there is a difference between "missing" and "unreported" sources. If these natural sources are included in the model simulation but are not reported as results (for example because they are not optimized by the algorithm), it makes sense to use BU estimate in place in order to compare "apple to apple". However, if these sources are not included in the prior simulation, the posterior total flux inferred from observations may still implicitly include their contributions because the observation sees the total flux, although the fluxes from these sources can be mis-attributed to other sources. If this is the case, adding BU estimates to inversion estimates will actually lead to inconsistent comparison. Therefore, it is important to distinguish between "missing" or "unreported" sources, or discuss the complication.

Line 701-707: The discussion on city-level and even event- or facility-level inversion is irrelevent to this study. The entire paper is on national emissions. It is still unclear how information is integrated on these very different scales.

Line 723-724: Worden et al. (2023) provides a framework to properly compare inventory and observation-based inversions.

Worden, J. R., Pandey, S., Zhang, Y., Cusworth, D. H., Qu, Z., Bloom, A. A., et al. (2023). Verifying methane inventories and trends with atmospheric methane data. AGU Advances, 4, e2023AV000871.

Line 724: "Some" attempts

Line 770: I think this is a very good recommendation. However, it is inadequately discussed and justified in the paper. A reader may want to learn the justification of this and other recommendations.

Line 738-740, Line 776: What do you mean by measurement of fluxes? Please be explicit whether you talking about regional fluxes or fluxes of specific sources? In the case of latter, it is actually emission factors rather than fluxes that are directly useful for a better prior estiamte. Moreover, activity data, in many cases, are also bottle-necks for better priors and better uncertainty estimates, in addition to emission factors. 

Line 922-930: Redundant reference information.

Line 832: Inversions are not entirely independent data, as they rely on the prior information. 

This study analyzes methane emissions and their annual variability using multiple bottom-up (BU) inventories and top-down (TD) studies for the European Union (EU) and seven additional countries with substantial emissions. The authors specifically aim to reconcile the BU and TD results by harmonizing the source sectors, and also offer insights to enhance the intercomparison between BU and TD studies. Acknowledging the significance and substantial workload involved in synthesizing a large volume of data, I believe there is considerable room to enhance the discussion, clarity, and readability of the study. Here are some suggestions

1. The purpose for such an “update” needs to be clearer. The authors state that “this study updates earlier syntheses (Petrescu et al., 2020, 2021, 2023) and provides a consolidated synthesis of CH4 emissions”. What's the need for the update? Compared to the previous syntheses, what new focus, data, and methods have been incorporated in this update?

2. The article does not describe the scope of the inversion results included in the discussion. Why were only the inversion results in Table 1 included in the discussion? What were the reasons for selecting these inversion results? Existing literature contains far more TD inversion studies than those discussed in the paper, with detailed discussions of the regions of interest and providing inversion results for interannual variations. These existing studies highlight inconsistency between BU and TD studies for many hotspot regions, such as oil and gas methane emissions in North America, coal emissions in East Asia, and wetland emissions in Europe, North America, and Africa, yet these are not reflected in the paper's discussion. This is a significant weakness. A large portion of these existing inversion studies can be found in Jacob (2022)'s review paper.

3. Figure 5 and Figure 6: why the uncertainty of CTE-GCP is so large? The uncertainty of total emissions can not be estimated as the sum of uncertainty from each sector.

4. Figure 7: It is very difficult to understand the bar of “BU Natural” and “TD Natural”? Why were the same BU anthropogenic emissions subtracted from the total emissions? Please enhance the clarity of this figure for better readability.

5. It would be great to have a figure that highlights the regions and emission sectors where the differences between current top-down and bottom-up results are most significant.

6. There are too many abbreviations in the text. It is recommended to include a table that summarizes the corresponding full forms of these abbreviations, maybe in the Appendix.

7. Line 199：missing a “.”

Reference

Jacob, D.J., D.J. Varon, D.H. Cusworth, P.E. Dennison, C. Frankenberg, R. Gautam, L. Guanter, J. Kelley, J. McKeever, L.E. Ott, B. Poulter, Z. Qu, A.K. Thorpe, J.R. Worden, and R.M. Duren, Quantifying methane emissions from the global scale down to point sources using satellite observations of atmospheric methane, Atmos. Chem. Phys., 22, 9617–9646, https://doi.org/10.5194/acp-22-9617-2022, 2022.