This study presents results from a suite of thunderstorm simulations in which CCN concentrations, INP concentrations have been varied and in which SIP mechanisms have been activated or deactivated. The authors focus on the response  of the lightning activity to these choices and attempt to understand the response through the analysis of the storms’ microphysical properties and process rates. Three thunderstorm cases are analyzed. It is a very detailed study. The results regarding the response to CCN and INP concentrations appear to be consistent with previous studies although the present study is perhaps somewhat more robust in that it examines multiple cases. The testing of SIP mechanisms appears to be a more novel aspect of the study and here their results are not entirely consistent with the few other studies that exist. Overall I think the study has potential to be a useful contribution to the community but I do have some major questions about the results.

1. The second part of the study regarding SIP mechanisms seems to make the first part of the study (and previous studies regarding CCN/INP concentrations) potentially irrelevant. The authors show that the inclusion of additional SIP mechanisms increasing the lightning flash count by 15-50x – a substantially larger increase than was obtained by varying CCN and INP concentrations. Assuming that the inclusion of SIP mechanisms leads to a more realistic simulation, then how meaningful are the results of the CCN/INP tests with only HM? I would guess that with all SIP mechanisms included, the sensitivity to INP would vanish and perhaps the sensitivity to CCN would also be diminished?
2. I am very surprised by the near total lack of sensitivity of the CWC profiles to SIP mechanisms (aside from NOSIP). It’s not just that the CWC profiles are similar, they are virtually identical. Assuming that this is not an outright error, could it potentially be due to the parameterization of ice crystal collisions or properties in LIMA? For example, perhaps LIMA has a minimum ice crystal size that is being met and so all simulations have the same crystal size despite differences in concentration. Or there is some hard-coded limiter in the collision rate with cloud droplets? Or is there no longer a mixed-phase region? It is just very hard to explain why orders of magnitude differences in the ice crystal number concentration should have absolutely no impact on the cloud water content. It also seems potentially inconsistent with previous discussion of how INP concentrations impacts CWC (for a fixed CCN concentration). Why should HM-only with INP variation impact CWC while HM+other SIP should not?
3. I know that the simulations aren’t meant to be compared to observations, but can the authors comment at least qualitatively on the magnitude of their results? Do previous observational studies support a nearly 10x increase in lightning flashes due to CCN?

Minor Comments:

1. There are several places where citations are needed, including Line 21-22, 48, and 69-72.
2. Line 86 – sentence is unfinished
3. Lines 145-150 – how was MID-WARM triggered
4. Lines 155-161 – what size particles? Can more information be provided about the INP populations?
5. Just in general, the model setup information was minimal and could be described in greater detail, especially since model initialization files are not provided as part of the code/data/software availability.
6. Line 193 – Not a complete sentence.
7. Line 238 – what does it mean that the simulations were treated together? That they were averaged together? Or that a representative simulation is shown?
8. Line 248 – is the charging rate meant to have a unit of kC/s rather than just kC?
9. Line 272 – by my eye CWC drops below 0.01 g/m^3 nearer to -20C than -10C.
10. Line 281 – effect of varying NCCN on what is mainly the same?
11. The Bergeron process is mentioned a few times. Typically I take this to mean the growth of ice and evaporation of droplets. But in a strong updraft, I assume that supersaturation is produced rapidly enough that supersaturation can be maintained with respect to both liquid and ice such that there is no Bergeron effect.
12. Figure 10 – what are SIP tendencies exactly? The ice number production rate?
13. It would be helpful to label the temperature lines in many of the figures.
14. Overall, I found the manuscript to be overly detailed and a little tedious to read. I think that the main points could be conveyed with more concise text. But I leave this to the authors to decide.

Please find my comments in the attached PDF.