The manuscript "Long-term analysis of atmospheric propane over Southern Europe based on observations conducted at the WMO-GAW station of Monte Cimone" presents a 13-year time series of propane, measured at Monte Cimone, Italy. Notably, all the data are available and can be immediately accessed, and I applaud the authors for keeping the monitoring station active and making the data openly accessible.

I really enjoyed reading this manuscript and consider it a nice piece of work. While some readers might find the detailed analysis tedious, I believe this work is essential for detailed investigations of our atmosphere.

Nevertheless, I have a few minor comments for the authors to consider.

Major comment:

This preprint is under review for Atmospheric Chemistry and Physics (ACP). According to the journal's scope, "Articles should have important and clearly argued implications for our understanding of the state and behavior of the atmosphere [...]". I consider this work valuable and important, as it confirms much of the previous work on C3H8. However, not all readers may share this view. Please, consider presenting this manuscript as a "Measurement Report".

Minor comments:

Line 6): JFJ is used but has not been defined. Please check the consistency with the following sentence.

Line 132): I suggest removing "slight" as the trend is anyhow statistically significant.

Line 290): It would be beneficial to briefly compare other emissions databases (e.g., CAMS or CEDS) for the three countries investigated. Similarly, it would be helpful to again cite a few articles mentioned in the introduction that show the underestimation of emissions has been known since 2018 (Dalsoren et al., 2018).

Line 295): There is a missing period (.) before "Similar".

Figure 10): I recommend using "CH" for Switzerland (instead of "CHE").

General Comments:

The manuscript presents a comprehensive long-term analysis of atmospheric propane at the Monte Cimone WMO-GAW station, based on an impressive 2011–2023 dataset that is also publicly available. The authors deserve congratulations for their sustained monitoring efforts and for making the data accessible to the scientific community. However, I agree with the second reviewer that the manuscript sometimes reads like a "measurements report" rather than a cohesive scientific analysis. The links between sections are often missing, which gives the impression of a series of independent analyses rather than a unified study. Below, I suggest ways to strengthen the narrative and improve the scientific rigor of the paper.

Major Comments:

1. Linking sections 3.1 and 3.2: seasonal trends and long-term trends

The authors mention an increase in the amplitude of the seasonal cycle (Section 3.2). To deepen the analysis, it would be valuable to disaggregate the long-term trend (Section 3.1) by season. Is the increase in amplitude due to higher winter concentrations or lower summer concentrations?

Suggestion: Perform a seasonal trend analysis and discuss how it relates to the findings in Section 3.2. This would provide insight into whether the observed changes are driven by emission variations, meteorological factors, or chemical processes (e.g., OH variability).

2. Contextualizing trends (2018–2020 and 2021–2023) with extreme values (Section 3.5)

The trends discussed in Section 3.1 (2018–2020 and 2021–2023) should be analyzed in the context of the numerous extreme values reported in 2018 and 2023 (Section 3.5). Could these extreme values be skewing the trends? A discussion on the robustness of the trends in light of these extremes would strengthen the analysis.

3. COVID-19 impact analysis: accounting for meteorological variability

The analysis of COVID-19 impacts lacks a discussion of meteorological influences (e.g., Petit et al., 2021). Changes in air mass origin or meteorological conditions (e.g., PBL height, wind patterns) could have influenced propane concentrations independently of emission changes.

Suggestion: Include a comparative analysis of meteorological conditions across years (e.g., using ERA5 reanalysis data or backward trajectories for multiple years). The 2022 trajectory analysis is interesting but insufficient for drawing conclusions about interannual variability.

4. Potential OH Bias in the Inversion (Section 3.6)

The inversion results (Section 3.6) may be sensitive to the OH concentrations used. Could a potential overestimation of OH lead to an underestimation of propane emissions?

Suggestion: Discuss the uncertainties in OH concentrations and whether there is a seasonal bias in the OH fields used. If possible, perform a sensitivity analysis to assess how variations in OH impact the inversion results.

5. Methods section: clarifications needed

- Quality Control (QC) Procedures

The QC procedure for propane measurements is not detailed. To ensure confidence in the trend analysis, the authors should provide:

• A table summarizing blank and standard measurements over time.
• Relative standard deviations for standards.
• Any observed instrumental drift and how it was corrected.

Additional request: Provide details on the QC procedures for CH₄ and CO measurements, as these are used in the analysis.

- JFJ Dataset: The Jungfraujoch (JFJ) dataset is used in the inversion but is not described. Clarify the QC procedures, measurement frequency, and any averaging methods applied to this dataset.

- Averaging 2-Hourly C₃H₈ Data to 3-Hourly Bins

The authors mention that 2-hourly C₃H₈ measurements were averaged to 3-hourly resolution for the inversion. However, the methodology is unclear:

• How were measurements assigned to 3-hourly bins (e.g., 00:00–03:00, 03:00–06:00)?
• How were bins with only one measurement (e.g., 04:00 in the 03:00–06:00 window) handled?

Suggestion: Clarify the averaging method and justify its suitability for the inversion.

Minor Comments:

Abstract: Define "JFJ" (Jungfraujoch) at first mention.

Line 143: Define "BB" (biomass burning).

Figure 2:

• Clarify the names and locations of the sites (ALT, BRW, KUM, etc.) in the caption.
• Explain the two y-axes (the red axis for LEF is confusing as LEF is shown in a different color).
• Clarify what is meant by "CMN trend curve (2 years) or (1 year)".

Figure 5:

• Define the seasonal periods (e.g., winter = December–February) in the text.
• Specify whether UTC or local time is used (in other figures as well).

Figure 9: Remove the connecting lines during data gaps to avoid misleading interpretations.

Conclusion:

This study provides a valuable long-term dataset and important insights into atmospheric propane trends. By strengthening the links between sections, addressing potential biases, and providing more methodological details, the authors can significantly enhance the impact and clarity of their work. I look forward to seeing a revised version that incorporates these suggestions.

References:

Petit, J.-E., Dupont, J.-C., Favez, O., Gros, V., Zhang, Y., Sciare, J., Simon, L., Truong, F., Bonnaire, N., Amodeo, T., Vautard, R., and Haeffelin, M.: Response of atmospheric composition to COVID-19 lockdown measures during spring in the Paris region (France), Atmospheric Chem. Phys., 21, 17167–17183, https://doi.org/10.5194/acp-21-17167-2021, 2021.

This manuscript presents an analysis of long-term atmospheric propane observations over Southern Europe, based on measurements conducted at the Monte Cimone GAW station (Italy). Long-term, high-quality VOC datasets are relatively scarce, and the availability of such measurements is valuable and worthy of publication. The manuscript focuses exclusively on propane measurements from 2011 to 2023.

That said, most of the analyses presented here are not particularly novel. Similar datasets, methodologies, and interpretations have been reported in numerous previous studies. While the data themselves are of interest, the scientific advancement beyond the existing literature is limited.

I am surprised that the authors did not incorporate measurements of other VOCs from the same instrument at Monte Cimone into the analysis. Including additional compounds would have enabled a more comprehensive and nuanced investigation of propane atmospheric behavior, sources, and chemical processing.

In my view, the comparison between bottom-up emission inventories and top-down estimates (Figure 11) represents the most valuable contribution of this study. I strongly encourage the authors to expand this part of the manuscript, providing a clearer explanation of the methodology and how the results were derived. This should also be accompanied by a more critical discussion of the uncertainties associated with the inversion results. While I am not a specialist in statistical methods, it seems that additional uncertainty analysis—potentially including a Monte Carlo evaluation—would substantially strengthen this section. Furthermore, the apparent underestimation of propane emissions in southern Italy by the inventory warrants deeper investigation and discussion of potential sources.  And again, incorporating other VOCs as emission tracers might provide insight into emission categories.

Specific Comments

• Line 4: The term “vary” is likely inappropriate in this context. It would be more accurate to state that atmospheric propane mole fractions did not exhibit a statistically significant trend over the observational period.
• Line 5: I recommend reporting changes in seasonal amplitude in absolute units (ppb) rather than percent, as the reference value for the percentage change is unclear.
• Line 6: Jungfraujoch (JFJ) is mentioned without prior introduction. Please introduce the site and its relevance before referencing it.
• Lines 15–16: The meaning of the reported 99.3% value is unclear. Please clarify what this percentage represents.
• Line 57: Please specify the calibration frequency in hours or days; the term “regularly” is too vague. In addition, further details on propane quantification are needed. Was propane quantified using selected ion integration? Was the mass spectrometer operated in scan or SIM mode? How were blanks determined and treated? Was a single-point calibration used, or were dilution curves and system linearity assessed? How was instrumental drift corrected? Given that multiple VOCs were likely quantified during each run, please explain why the analysis focuses exclusively on propane. Were the propane measurements audited by the World Calibration Center for VOCs?
• Line 105: If JFJ data are included in the analysis, the measurement techniques and quality assurance procedures for that site should be described with the same level of detail as those for Monte Cimone.
• Line 133: The text first states that no clear trend is present but then claims a statistically significant decrease at the 95% confidence level. Please reconcile this apparent contradiction and use appropriate statistical terminology. It is also unclear how the reported slope values were derived. The curve shown in Figure 1 does not appear to represent a linear trend. Please explain how these slope values were calculated and how they represent changes over the full data period.
• Line 134: The unit should be ppb yr⁻¹, not ppb.
• Lines 134–146: Differences observed over relatively short time windows (e.g., 2–3 years) may be strongly influenced by interannual meteorological variability. Such variations should not be readily interpreted as emission changes without longer observational periods (at least five years) or appropriate meteorological normalization.
• Lines 148–166: The seasonal behavior of propane has been well documented and explained in the existing literature, including studies cited here. This section offers limited new interpretation and could be substantially shortened.
• Line 167: Meteorological influences should be considered more explicitly in this context.
• Line 171: Please provide additional detail on how seasonal minima and maxima were determined, as this procedure is more complex than implied. Is the reported change in seasonal amplitude statistically significant? An annual change of 0.16% over 12 years amounts to less than 2% in total, which is relatively small. It is unclear whether this signal exceeds measurement noise or reflects meaningful changes in chemistry or emissions.
• Line 181: According to the recent literature, NOₓ emissions in the Northern Hemisphere have generally been decreasing over the past decade rather than increasing. Please revisit this statement.
• Line 203: Please clearly define how the seasons are classified.
• Line 230: Several studies report increased ozone levels during the COVID-19 pandemic. Please discuss your findings in the context of this broader literature.
• Lines 230–243: This section appears highly speculative and inconclusive. I recommend removing it or substantially shortening it.
• Line 248: Please be more precise in this description. If a 96% cutoff is applied, it is unclear how 11–15% high-occurrence values can simultaneously be reported. These metrics appear to use different reference definitions, which need to be clearly explained.
• Figure 6: I recommend adding bold numerical labels for the median values, as these are used for the statistical comparisons.
• Line 295: A period is missing after the word “located”.
• Line 310: Deriving a linear trend over a period that includes a pronounced anomalous minimum during the COVID-19 pandemic is questionable. This limitation should be explicitly acknowledged and discussed.