The authors use airborne and ground-based air samples to attribute methane emissions in the region to sources, with a focus on distinguishing coal/fossil sources from biogenic/waste sources. The ground-based samples are used to identify source signatures and then, the airborne samples are used to determine the contribution of the two source types. Overall, the study is nicely written and presented but can benefit from some editing to clarify the analysis and results. 

An interesting finding from this paper is the relatively large contribution of biogenic sources in the Upper Silesian Coal Basin. Given this finding, there is a need for some introduction of these biogenic sources in the Introduction and annotations in Figure 2. Although there is some text on this towards the end, additional text earlier on would be helpful.

It appears that a lot of the data being used in this paper have been published in previous papers. However, it’s unclear if there’s new data presented in this paper. This should be clarified (see detailed comments below).

Below are detailed line-by-line comments:

L20: Replace "growth" with "concentration accumulation"

L20: Replace "on global" with "at global"

L24: are the "other anthropogenic sources" related to the coal mine or something else? Landfills and wastewater should be listed here in the abstract given the findings of the paper.

L35: "allowed for the determination of the source signatures..."

L30: what are all the sources in the region? This speaks to an earlier comment about clarifying the sources being studied.

L40: I think the main point is that the d2H is important for source attribution and that the d2H of the ventilation shafts differ from the regional d2H values. Therefore, it would be worth rephrasing this sentence to highlight this point. The fact that the d2H of ventilation shafts match previous studies is relevant but should be the secondary point.

L41-42, 34: Because wetlands and ruminants were mentioned in L34, I expected these to be significant. Suggest revising L34.

L45: It would be good to provide some quantitative comparison of underestimation. Either state that the common inventories estimate 6% biogenic or state the difference as a factor.

L50: “at limiting” to “to limit”

L52: the Global Methane Pledge is specifically for methane emission reductions, not all greenhouse gas emission reductions. Replace “greenhouse gas” with “methane”

L53: Replace "localize" with "locate"

L116: define what is meant by "regional perspective". What is the scale of "regional"? It sounds as though the authors are re-analyzing existing data. Either way, this needs to be clarified. A clearer description of what new analysis is being performed here would be helpful.

L136: Pniowek is not shown in Figure 2. It should be identified.

L141: Define FUB at first mention.

L153-154: Were the data published already? This needs to be clarified

L171: "...concentrations and isotopic composition of CH4 were..."

L218: how can the standard errors be small if the variance is large?

L226: the location of Silesia needs to be shown in Figure 2.

L261: the locations of the cow farm, landfills, manholes and wastewater facilities need to be shown in Figure 2.

L266: what is meant by vicinity? In the conclusions, 1-2km is mentioned. It's surprising that the samples taken directly within the shafts are similar to those taken 1-2 km away.

Figure 5: Specify where in the mines the "mining" samples were taken.

Figure 5: Is the data from the MEMO2 dataset? Also specify that this is from ground-based studies.

Figure 6: Specify whether the data is from ground-based measurements in the caption.

Figure 7: specify that EMID fossil fuel data is ground-based.

Overall, there are a lot of acronyms that are used interchangeably. For the ground-based data, it's referred to as MEMO2, EMID, and "ground-based". I suggest simply calling it "ground-based", if possible, and not using multiple names for the same datasets.

L324: Given the importance of the waste sector in this region, there is a need to describe these sources more. How big are the landfills and wastewater treatment plants both spatially and in terms of waste volumes? Also, is there an inventory of all methane sources in the USCB? 

L333: the better discrimination of d2H signatures than d13C is interesting. Therefore, it would be helpful to see what the error would be if only d13C was used in source attribution.

L359: can the locations of these landfills and wastewater treatment plants be shown in Figure 2?

Figure 8: The authors assume that waste is the dominant biogenic source in the area. However, there may be natural (and agricultural) sources that may be contributing more than assumed. The authors are probably right but it may be worth pointing out that there still are uncertainties in the biogenic methane source.  

L398: how much do the signatures vary over time? How does this affect uncertainties and the comparisons/analysis presented in this paper?