In fact, I also reviewed another paper by the author (Yang et al. 2024; https://doi.org/10.5194/acp-24-5989-2024), which can be considered a sister paper in some respects. In comparison to these two papers, the first one (Yang et al. 2024) seems more innovative because it appears to have established a new model. This paper only conducted sensitivity experiments with different aerosol concentrations using the model from the previous paper, which provided useful references but seemed to have lessened in terms of innovation.

          First, we acknowledge that different secondary ice production (SIP) processes or different aerosol concentrations play an important role in cloud electrification, which has been previously studied in many related studies (Mansell and Ziegler 2013; Tan et al. 2015; Phillips and Patade 2022; Yang et al. 2024). At this stage, it seems less urgent to continue discussing the impact of different SIP processes on electrification under different aerosol conditions. Because cloud microphysical characteristics change successively after the SIP process and aerosol concentration changes, the part related to electricity will also naturally change. When the charging rate changes, the evolution of the charge structure is only a more superficial feature. We know that there are many experiments or hypotheses related to the SIP processes, but which process of SIP actually plays a role in clouds and whether they all play a role at the same time is still a question worth exploring. Therefore, we are still unsure whether 4SIP will play a role at the same time.

          Regarding the structure of the paper, in the description of the model in section 2.2.2, the content of this section is almost a duplicate of Appendix B in Yang et al. (2024). Is there a need for duplicate descriptions here?

          From the results in Fig. 10, compared to the experiments with N₀=400, the experiments with N₀=4000 show a stronger charging rate, which also corresponds to a stronger electric field intensity in these two groups of experiments (Fig. 16).  However, in Fig. 10, compared with noSIP-4000, the non-inductive charging rate of SIP-4000 is stronger, but the polarity and height of the charging rate change very little, which may not cause a significant change in the polarity of the charge structure. On the contrary, the inductive charging rate has undergone significant changes. This may be the reason for the change in the polarity of the charge structure of 4SIP-4000 in Fig. 13.

          The paper mentions that without SIP, the aerosol does not change its charge structure. However, as shown in Fig. 11a, even without the SIP process, the charging rate within the cloud significantly changes with the change in aerosol. The height-time variation diagram may not reflect the actual charge structure, and a cross-section diagram of the charge structure should be provided.

Review for: Impact of secondary ice production on thunderstorm electrification under different aerosol conditions

General comments:

Huang et al. present a detailed investigation into the role of secondary ice production (SIP) processes and aerosols in cloud electrification within thunderstorms. Utilizing the Weather Research and Forecasting (WRF) model with a spectral bin microphysics (SBM) scheme, this study examines how different SIP processes interact with varying cloud condensation nuclei (CCN) concentrations to influence cloud microphysics and charge separation. This topic is both timely and relevant, exploring key uncertainties in our understanding of cloud microphysics and electrification mechanisms under different aerosol conditions. Nevertheless, the manuscript requires substantial revisions to improve clarity and strengthen the validity of conclusions. The following specific comments and technical corrections should be addressed before the paper is considered for publication:

Specific comments:

•    In the abstract: (i) Please be more precise when referring to terms like "clean" and "polluted" environments, as well as "relatively warm/cold temperatures," both in the abstract and throughout the manuscript. Explicitly define these terms or specify the aerosol conditions and temperature ranges you are referring to each time to avoid ambiguity. (ii) It would be beneficial to mention some implications of your findings within the abstract. For example, you could explain the importance of correctly capturing the charging reversal at different temperatures due to variations in aerosol concentrations or the inclusion of SIP. This would help readers understand the relevance and impact of your results. (iii) Finally, please define what is meant by the term "normal charge structure" (Line 29).

•    Lines 41-52: Here you mention the aerosol indirect effects in liquid clouds (cloud albedo and lifetime effect). Consider also discussing the indirect aerosol effects in mixed-phase clouds (MPCs; e.g., glaciation, riming and thermodynamic indirect effect in convective clouds). This is particularly relevant to your study, and these indirect effects could help explain some of the results you present (e.g., Lines 241-247 or 296-298).

•    Lines 54-63: Before discussing the results from various model simulations, it would be beneficial to provide a brief introduction to SIP. Additionally, introducing Primary Ice Production (PIP) as a prerequisite for SIP would clarify the mechanisms by which an increased presence of cloud droplets promotes ice formation in MPCs.

•    Section 2.1: You should explicitly state that the case study discussed in this manuscript is essentially the same as in Huang et al. (2024). Clearly link this study to your previous work, explaining how it builds upon earlier findings to examine the sensitivity of results to varying CCN concentrations. Notably, the highest CCN concentration tested in this study (i.e., 4000 cm⁻³) is the same as in the Huang et al. (2024) paper. Given that the sublimational break-up mechanism is found to be inefficient (as evidenced by the orders-of-magnitude difference shown in Figure 9), the primary results from this study should be comparable to those from your previous work, particularly for the SBM-3SIP simulation (e.g., compare Fig. 13l of this manuscript to Fig. 13d of the Huang et al. (2024) paper). However, there is little discussion comparing the two papers, and the specific innovation being tested in this new paper is not clearly articulated.

•    Figure 1: Clarify whether the maps presented here are generated from WRF simulations. If so, have you evaluated the meteorological fields against any available observations? If this was done in previous work, please mention it explicitly.

•    Section 2.2.1: This paragraph is quite lengthy; consider breaking it into smaller sections for better readability. I suggest either reordering the content to begin with the Model Setup (Section 2.2.3) or, when referencing the "two domains" of WRF (Line 118), guiding readers to the relevant domains shown in Figure 2. Additionally, please include the following details in this section: (i) specify the number of ice and liquid species included in the scheme; (ii) in Line 123, clearly state the assumed value of "k" in the Twomey slope; (iii) provide a more detailed reference to the Primary Ice Production (PIP) scheme used, as any overestimation of primary ice crystals could diminish the perceived significance of SIP. Have you observed any changes in PIP rates with varying CCN concentrations? (iv) For the SIP implementation, it would be more effective to refer to your previous work (e.g., Yang et al., 2024) rather than the original parameterization papers, and to emphasize key assumptions made in this study (e.g., what was the rimed fraction and ice habit assumed for the involved ice species, above which raindrop diameter did you allow droplet-shattering to occur?).

•    Section 2.2.3: Please justify the chosen range of CCN concentrations in your simulations (Line 198). Are these values based on literature or measurements or random? The title for Table 1 should also be more descriptive. 

•    Figure 3: It appears that the simulated squall line is slightly "shifted" compared to the radar observations. Is the observation panel identical to the one presented in Figure 3a of the Huang et al. (2024) paper? While I understand that it is challenging to achieve perfect alignment in the spatial extent and timing of the system in the model, I wonder if you have considered using a different forcing dataset besides the FNL reanalysis. Is this dataset the most widely used and validated for modeling studies in this particular region?

•    Figure 4: could you please explain why you focused the statistical analysis on the four-hour period between 00:00 and 04:00 on May 30, given that the stated period of interest is between 16:00 UTC on May 29 and 06:00 UTC on May 30, 2022. It might be more insightful to compare median profiles and/or additional quantiles of the simulated reflectivity, as these could provide a better understanding of the distribution of simulated values. Additionally, have you conducted any statistical tests to determine whether the observed differences between the sensitivity simulations are statistically significant?

•    Lines 226-227: Could you clarify why increased CCN concentrations worsen the agreement with observed radar reflectivities? This appears to contradict the statement in Line 216 that the highest CCN concentration (4000 cm⁻³) is the most realistic for this case study.

•    Line 239: Please specify how "The results are averaged over the cloud at different heights" was achieved. Did you average spatially over model grids representing the squall line, and what criteria did you use to identify in-cloud grids? 

•    Line 252: Please consider defining the CCN thresholds distinguishing "clean" versus "polluted" cases in the Methodology section, and use this distinction consistently throughout the text.

•    Line 253-254: If the shattering of freezing drops is the most important SIP process to ice production at relatively warm subzero temperatures, would this not reduce the number or mass concentration of big raindrops within those temperature ranges?

•    Figure 6: Confirm whether the mean vertical profiles in this figure correspond to the same period shown in Figure 5. Additionally, have you tried to use logarithmic scale for the horizontal axes showing the number concentration of particles in this figure?

•    Figure 7: This figure appears overly complex, potentially obscuring key insights from your sensitivity experiments. Please consider reducing the number of sensitivities discussed in the main paper or discussing any plateau or asymptotic behavior as CCN concentrations increase. A logarithmic vertical axis might also be beneficial in this figure. Is there an issue with the y-axis in Figure 7b?

•    Line 285-286: Have you examined how PIP rates vary with changes in CCN concentrations?

•    Line 313 and Figure 9: Please elaborate on why IC is most significant within the -10°C to -20°C temperature range. The enhanced efficiency of IC within the so-called dendritic growth layer is discussed in von Terzi et al. (2022) and Georgakaki et al. (2024). Did you implement both IC formulations from Table 1 of Phillips et al. (2017) for dendritic and spatial planar ice crystal habits? If so, was the selection of ice habit based on temperature? This should be further clarified in Section 2.2.1, as it might help explain the results discussed here. Also, please consider using a logarithmic scale for the colorbar in this figure.

•    Line 314 and Figure 9c-d: Shouldn’t the production rate due to RS zero outside of the -3 °C and -8 °C temperature range?

•    Section 3.3: Is there any way to evaluate the modeling results and conclusions from this section? In your previous work (Yang et al., 2024), you included flashing rates observations. Were such observations available for the squall line discussed in this study?

•    Lines 435-437: Could you clarify the need to alter the "inverted tripole structure" of the total charge density? What is meant by "normal charge structure," and to which subplot in Figure 15 does this refer?

•    Lines 501-502: To support your statement here, you should include a more extensive comparison and model evaluation against observations.

•    Line 516: Please revise this sentence, as the objective should be to align model simulations with observations, not to maximize ice splinter production. Similarly, the claim about the "incredibly important role" of SIP processes mentioned in Line 524 is not substantiated by your modeling results and lacks support from cloud microphysical measurements (e.g., ice crystal number concentrations, ice/liquid water content).

•    Data availability section: The repository linked only provides the default WRF code, not the updated scheme developed and used in this study. Additionally, no sounding or lightning data has been used in this manuscript (Line 529), so please update the text here accordingly.

Technical corrections:

•    Line 16: Consider changing "eastward tendency" to "eastward propagation".

•    Line 37: Change "The uncertainty in modeling the microphysics" to "The uncertainty in modeling cloud microphysics".

•    Line 39: Did you mean "play a crucial role" in improving the accuracy of lightning prediction in numerical simulations?

•    Lines 45-46: It would be helpful to add a few references to support this statement.

•    Line 60: Change "remarkably ‘as’ CCN concentration increases".

•    Line 65: Remove "the" before "SIP processes" and "electrification".

•    Line 72: Change to "And they found ‘that’ graupel-snow collisions".

•    Line 76: Remove "the" in front of "SIP processes" and "ice generation".

•    Line 77: Consider using "inverted tripole structure".

•    Line 79: Change "suggested the rime-splintering" to "suggesting that the rime-splintering".

•    Line 84: I would suggest to use the plural form: "Observations and simulations".

•    Line 93: "Model set-up" might be more accurate than "design".

•    Line 105: Swap the order of Figures 2 and 3 since you first refer to the figure with simulated reflectivities and then to the WRF domains.

•    Line 118: “Fast-SBM” has not yet been defined here.

•    Line 148: Move the definition of SP98 here from Line 164.

•    Line 154: Did you mean "describe detailed discharge channels"? You might also use "sophisticated lightning models" instead of "elaborate".

•    Line 184: “field” instead of “filed”.

•    Line 191: “shortwave and longwave ‘spectra’”?

•    Line 228: Remove "of" and consider providing a reference to support the statement.

•    Line 273: Add the figure number you are referring to here (likely Fig. 6d).

•    Line 308: Replace "revolution" with "evolution".

•    Line 310: Consider using "observation" instead of "phenomenon".

•    Line 333: Replace "revolution" with "evolution" and please mention that the results in this figure come from the sensitivity experiments accounting for all four SIP mechanisms.

•    Line 350: “collisions”

•    Line 369: This should be Fig. 11l if mentioned first.

•    Line 399: Use "level ‘that’ dominates the aerosol impact".

•    Line 400: Include the units for the RAR threshold of 0.1 mentioned here and consider providing a reference.

•    Line 420: “inverted tripole” structure?

•    Line 436: “inverted tripole” structure?

•    Caption of Figure 15: can you please update the text to define the abbreviations in the legend (scg, scsl, sctot)?

•    Line 449: “increases”

•    Line 451: Use “causes”, and did you mean “-20 °C” rather than “-30 °C” here? 

•    Line 461: squall “line”

•    Line 470: higher “cloud” levels

•    Line 520: Clarify what is meant by "different convective intensities".

•    Line 521: aerosol “concentration”

References

- Georgakaki, P., Billault-Roux, A.-C., Foskinis, R., Gao, K., Sotiropoulou, G., Gini, M., Takahama, S., Eleftheriadis, K., Papayannis, A., Berne, A. and Nenes, A.: Unraveling ice multiplication in winter orographic clouds via in-situ observations, remote sensing and modeling, npj Clim. Atmos. Sci., 7(1), doi:10.1038/s41612-024-00671-9, 2024.

- Huang, S., Jing, X., Yang, J., Zhang, Q., Guo, F., Wang, Z. and Chen, B.: Modeling the Impact of Secondary Ice Production on the Charge Structure of a Mesoscale Convective System, J. Geophys. Res. Atmos., 129(9), 1–22, doi:10.1029/2023JD039303, 2024.

- von Terzi, L., Dias Neto, J., Ori, D., Myagkov, A. and Kneifel, S.: Ice microphysical processes in the dendritic growth layer: a statistical analysis combining multi-frequency and polarimetric Doppler cloud radar observations, Atmos. Chem. Phys., 22(17), 11795–11821, doi:10.5194/acp-22-11795-2022, 2022.

- Yang, J., Huang, S., Yang, T., Zhang, Q., Deng, Y. and Liu, Y.: Impact of ice multiplication on the cloud electrification of a cold-season thunderstorm: A numerical case study, Atmos. Chem. Phys., 24(10), 5989–6010, doi:10.5194/acp-24-5989-2024, 2024.