General comment

The manuscript by Vogel et al. (2025) present the analysis of 21 years of optical particle counter (OPC) in-situ observations at the remote Monte Cimone site as an extension of the work by Duchi et al. (2016).  The authors analyse the OPC data to provide an estimate of the trend for the annual fractions of dust transport days (DTDs) and PMcoarse, describe the PMcoarse variability compared to background values (enhancement factor) and investigate the correlation between dust events and PMcoarse concentrations.

From a general point of view, although the paper is well written, the analysis could benefit from further refinement and inclusion of additional data sources (i.e. satellite data or reanalysis) and a more accurate and clear presentation of the results (e.g. discussion of measurement uncertainties), which I see as the added value of this work compared to Duchi et al. (2016). I would recommend considering publication once the comments have been adequately addressed.

Specific comments

• This work appears to be an extension of Duchi et al. (2016). How do the results of your analysis compare quantitatively with those presented in that study? This comparison is missing in the paper. Which is the added value compared to that study? Please discuss.
• I would see the added value extending this analysis to include spatially-resolved data such as satellite data and/or reanalysis, especially considering that Monte Cimone is located at 2165 m a.s.l., where aerosol optical depth can provide additional insights into dust episodes. OPC in-situ observations alone may not be sufficient to provide a robust estimate of trends in DTD/DTE and to set a "milestone in DTE identification".
• What happens at your analysis if you neglect the data before the 2008 (i.e. after the inlet in the line has been heated)? I rather see a decreasing trend from 2008 to 2023 (Fig. 9) in PMcoarse and a more increasing trend in DTDs (Fig. 2). Please detail and consider adding supporting information.
• You present a dataset without discussing the uncertainties in the measurements/plots. This aspect must be better clarified and taken into account in the statistical analysis/plots avoid limiting the analysis of dataset variability to percentile-based metrics only. How significant is your trend estimation considering the uncertainties you have in the measurements? 
• L.96 the FLEXTRA back trajectory configuration is missing. A section (i.e. meteorological inputs, configuration, model description) should be incorporated into the methods, with details on the configuration of the back-trajectories, given that it represents a major factor in determining the DTD. Why do you decrease the 10 days from Duchi et al. (2016) back trajectory to 7 days? Is there any reason? Please explain.

Minor Comments

- I would rather suggest adding a section in the introduction on aerosol optical depth as a proxy for dust outbreaks, as it represents a more informative parameter compared to surface PM measurements.

- Please consider adding a few sentences describing the structure of the paper at the end of the introduction.

L.37 Please add a reference

L.38 if dust particles remain in the upper layers there will be no increase in the surface PM concentration.

L.76-77 the GRIMM 1.108 starts at 0.3 µm and not at 0.25 µm with the 780 nm operating wavelength

L.81 Please include the number of bins that you consider for the “coarse” mode.

L.92-103 I suggest introducing a list of items instead of the text to identify the different steps of the algorithm.

L.102 how many months are available in 21 years of data?

L.105 Why are back trajectories missing? Due to missing meteorological data?

L.108-110 “consecutive days”, how many consecutive days do you consider? I see it in Figure 1 but it should be written also here.

L.117 Please add more details, which is the average particle density you obtain?

L.138 “user-defined alpha value”, which is?

L.168 replace “merging” by “grouping” and “years” by “DTD yearly values”.

L.168-170 here it could be very interesting to compare with the aerosol optical depth. Your results are consistent with the seasonal cycle in the aerosol optical depth climatology observed for the Po Valley in (Di Antonio et al, 2023) using satellite data.

https://acp.copernicus.org/articles/23/12455/2023/

L.199-200 which is the uncertainty linked to this value?

L.226 Background concentrations are also expected to show diurnal variability in summer compared to winter.

L.218-239 What is the key message then here? What do we learn with the seasonal cycle? It is not very clear to me.

L.242-243 this should go in the methods

L.279 the median is always below the WMO threshold.

L.286-87 I would avoid to make such a suggestion based on the analysis of a single point data source.

L.297-301 I do not fully agree with this statement. On the one hand, dust transport over the Mediterranean basin is generally favoured during the summer months due to the development of a deeper planetary boundary layer (PBL) over the Sahara, on the other hand, it is also strongly influenced by synoptic-scale weather patterns that facilitate such transport. I would rather say that DTE appears to be more closely associated with the persistence of high-pressure systems over the region, rather than directly linked to PBL development over the Sahara. Given the considerable distance from the source areas, the observations primarily reflect atmospheric transport processes rather than continuous Saharan emissions. It may be helpful to investigate reanalysis data to assess whether the occurrence of longer DTEs is associated with more specific stable atmospheric conditions (looking at geopotential height for example).

L.314 aerosol or dust optical depth?

L.314 what do you mean with “reanalysis data from satellites”?

Technical comments

Methods and results can be presented in a clearer way:

- Could you kindly list the different key processes that the dataset has undergone (i.e. Sec. 2.2, 2.3, 2.4) rather than describing them in a block of text?"

- You can summarize major results (such as the average conditions for background/non-background) in tables.

L.40 replace “mineral aerosol” by “dust particles”

L.40 PM “surface” values

Figures: Adding markers to the line can improve the clarity of the plot.

L.263 add space

The paper presents 21 years (2003-2023) of aerosol optical size distribution in a WMO/GAW, ACTRIS-RI and ICOS-RI sampling place at Monte Cimone peak, which have been used to identify dust transport days on site and to analyze its interannual and seasonal patterns for both frequency of occurrence and observed particle concentration. Air masses pathways were also analyzed to confirm the selected days. The paper is an extension of the analysis shows in Duchi et al. (2016), which similar goals and methodology but with 11 years more of database.

General comments

The topic of the paper is suitable for ACP, the methodology is clear, the results are logically interpreted and the manuscript is well written. But, it is well-known the doubts of the scientific community when using the aerosol optical size distribution to study desert dust episodes. Therefore, it is observed a lack of explaining the uncertainty in the experimental technique, and how this may affect the estimated parameters. While information of the inventory and duration of the events is strong, other parameters as total mass concentration, enhancement or the comparison with the WHO threshold are subject to some uncertainty. It is due to, as authors comment in Sections 2.1 and 2.3, OPS instruments measure the particle number size distribution, which is converted into a mass distribution using a particle density dependent by particle size, but it also depends on aerosol composition. The dependence on particles composition becomes much stronger, due to the variability in the aerosol typing studied in this work: desert dust and background aerosol.

Therefore, before publishing this work, the authors should resolve this issue, explaining the uncertainties in the estimations. Perhaps, the use of complementary information such as off-line mass concentration, multi-wavelength absorption coefficient, and/or aerosol optical depth could be useful, event if it is of shorter periods.

On the other hand, this paper is an extension of Duchi et al (2016), but there is no comparison of the current results with those of the previous article, nor an explanation of the new results. Furthermore, this citation is very important for the paper, and is not accessible via the indicated DOI. The authors are suggested to make an effort on it as well.

Specific comments

• Please, consider to include information about the backward-trajectories source within Section 2.
• Figure 1 – the legend about (a), (b) and (c) is not corresponding to the text in paper. Please, revise.
• Lines 162-163: it is commented that there was no significant temporal trend (slope of 0.063) in the fraction of DTDs obtained from the trend analysis. But dashed line shows an increasing trend. Please, consider to delete the line, because this may lead to misinterpretation.
• Line 235: opposing - > opposite
• Line 247: in the majority of the years was -> in most years was
• Line 249: longer of shorter -> longer or shorter
• Lines 262-263: dust mobilization in summer -> dust injection into the atmosphere during the summer.
• Lines 269-271: care must be taken because the results may have a high uncertainty. It is important to solve this issue.