Summary: In the manuscript titled “Impact of urbanization on fine particulate matter concentrations over central Europe”, the authors examine urban contribution to particulate matter and its constituents over central Europe using a modeling approach. The study is overall written well and mostly well designed. I do have several questions about the modeling approach and uncertainties and how they might impact the results.

Major comments:

1.    In general, I found the modeling approach difficult to follow. There seem to be multiple models driving each other and other models being used to derive or modify the emission inventories (for instance, MEGAN for BVOC emissions). It would really help to have a schematic of the overall modeling setup and how each component feeds into each other regarding input and output. 

2.    As far as I can tell, the RegCM model uses the urban representation of CLM4.5, which is a single layer canopy model with no urban vegetation and no anthropogenic heat flux from vehicles. How does this affect the BVOC estimates. Does the MEGAN algorithm actually resolved urban vegetation and biogenic emissions from them (or deposition on them?). Additionally, where did the leaf area index and plant functional type data for urban areas come from? Urban areas can have very different vegetation characteristics from their surrounding rural areas (Paschalis et al. 2021).

3.    I have other questions about the emission inventory. How are vehicular emissions considered in these estimates? What about impacts from asphalt-related emissions (Khare et al. 2020)?

4.    Finally, as I understand it, these are offline simulations. Does that mean that there is no feedback from the land to the atmosphere? In the real world, urban heat and moisture islands (Chakraborty et al. 2022) may impact near-surface chemistry and convection; and thus overall pollutant concentrations within and sourced from urban areas. There needs to be a broader discussion about these uncertainties.

Minor comments:

1.    Figs 1 and 2: Please add the y axis labels with the units. 

2.    Page 1: ‘We evaluated the RUT impact on PM2.5 over an ensemble of 19 central European cities’: That is not the typical usage of ensemble. 19 is just the sample size here?

References:

1.    Paschalis, A., Chakraborty, T. C., Fatichi, S., Meili, N., & Manoli, G. (2021). Urban forests as main regulator of the evaporative cooling effect in cities. AGU Advances, 2(2), e2020AV000303.

2.    Khare, P., Machesky, J., Soto, R., He, M., Presto, A. A., & Gentner, D. R. (2020). Asphalt-related emissions are a major missing nontraditional source of secondary organic aerosol precursors. Science advances, 6(36), eabb9785.

3.    Chakraborty, T., Venter, Z. S., Qian, Y., & Lee, X. (2022). Lower urban humidity moderates outdoor heat stress. Agu Advances, 3(5), e2022AV000729.

This manuscript discusses the effect of urbanization on PM 2.5 levels in central Europe. Four factors were considered, i.e., the urban emissions, the urban canopy meteorological forcing, the impact of modified dry deposition velocities and the impact of modified biogenic emissions. A cascade of simulations with RegCM and CAMx were performed to separate the effect of these factors. The authors found that urban emission contributes the most to increasing the PM 2.5 concentration, while urban canopy has a counter effect due to stronger vertical eddy-diffusions. The transformation of land-use also tends to decrease PM 2.5 levels by increasing the particle dry deposition velocity. Overall, the conclusions drawn from the simulations seem sound, but I believe the presentation of the results can be improved and the manuscript can become more concise by revision.

Major comments:

• Regarding the design of the simulation cascade, more explanation of the simulation sequence of is needed. If the simulations follow a different sequence, will the results change significantly?
• The authors have somewhat strictly separated results from discussions. I find this way of writing very difficult to follow. One the one hand, in the results section some paragraphs are only recounting figure contents and barely provide any explanations (e.g., sections 3.3.1-3.3.4). This makes the reading process really dry (actually, a table with numbers may actually outperform the texts in 3.3.1-3.3.4). One other hand, when I arrive at the discussion part (far away from the results), I have a hard time correlating the discussions to the corresponding results; as a result, I have difficulty in assessing the validity of some statements in the discussion section. Therefore, I recommend restructuring of the paper to some extent by combining some of the discussions with the results.

Minor comments:

• Line 175: Eqn (3) assumes that the processes are additive. This sentence seems to suggest the opposite causal relation.
• Figure 5: Please state clearly whether figure 5 shows the measured or simulated concentrations.

Technical:

There are quite a number of tiny mistakes in the manuscript. The authors need to carefully go through the manuscript for a more complete check.

For the majority of the figures: Please use larger font size for the axis tick labels. Figures 1, 2, 11, 12 also need proper labels for the axes. In Figures 11 and 12, the phrase ‘Diurnal cycle of urbanization impact’ does not need to appear above every panel.

Line 32: the abbreviation of DV has not been defined previously.

Line 65: messed up citation of Yang et al.

Line 83: this sentence needs to be revised.

Line 163: This sentence is very confusing.

Line 237: although -> despite

Line 282: the impacts of urbanization on POA.

Line 285: … and their components to observations.

Line 291: and the underestimation is stronger in winter.

Line 310: missing units

ug/m3 are sometimes italic and sometimes normal.

Line 541: we further discuss…

Line 559: know -> known

Line 561: responded -> respond

Line 623: the citation of Ortega et al. is in the wrong format

......