In this paper, Kolinska discusses the differences in VHF emission immediately following return strokes.  Overall, the paper is fairly well written, easy to understand, and quite interesting.  It is good to see careful analysis of the LMA signals recorded at individual stations, rather than just analysis of the located maps.  Eventually I think this paper will make a good contribution to the state of the science.  Right now, there are some citations that are missing which require more than just a cursory mention, and I would like to see the discussion expanded a bit.  But, I believe these will amount to fairly minor revisions.  

Specific Comments:

Citations: This study omits discussion of the results in Lapierre et al 2017 (https://doi.org/10.1002/2016JD026189), and which deserves more than just a passing mention.

You have a citation from Lapierre’s earlier paper on -CG leader growth, but skipped the second one looking at +CGs.  As before, the paper is mostly interested in the behavior during continuing current, but for +CG flashes, this usually happens on the first (and only) stroke.  This means the study has direct and detailed observations of exactly the behavior described in this paper.  There is also a fairly detailed map of a -CG flash, extrapolated from LMA and INTF data.  The INTF map lacks the initial return stroke, but the subsequent strokes all show the behavior discussed in this study.  

Expanded Discussion:  The author’s description of what they believe is happening in the cloud falls a little bit short of what I believe to be happening in the cloud. 

The situation for +CGs is quite well documented in Lapierre 2017.  Here the return stroke causes an immediate and significant growth or ‘bloom’ of negative leaders in the cloud.  During the bloom, both the number of negative leaders and their speed increases.  The growth period appears linked to the duration of the return stroke current, but even return strokes without continuing current will produce a bloom in the negative leader channels in the cloud.  

For -CGs, there is no single reference to point to, but we can piece together the behavior from several studies.  Immediately following a -CG return stroke, negative leader growth usually ceases.  The potential wave from the return stroke itself traverses the entire lightning leader channel structure, all the way to the tips of the positive leaders.  We know this become sometimes (perhaps even often) there will be a burst of positive breakdown at the positive leader tips immediately following the return stroke (Shao 1995 - although those observations are ambiguous and might be K-leaders, and Stock 2017 is not so ambiguous).  In particular, Stock 2017 notes that the positive bursts following a negative return stroke do not appear to produce conducting channels.  

From the Lapierre studies (Lapierre 2014, 2017 both), we know that after the return stroke the positive leaders continue to propagate with speed unaltered.  Emission from the positive leader tips is usually not detectable (unless it is, Pu 2021, Stock 2023).  Still, the LMA will detect emission on positive leaders in the form of needles, as they produce impulsive but weak RF (Hare 2019,2021, Pu 2019).  Needle emission is detectable and mappable by the LMA, but most studies on this predate the term ‘needles’ (e.g. Edens 2012, Van der Valde 2013).  In the case of SAETTA, the needle emission is weak enough that it’s unlikely to be located by the network, but the nearest sensor should still be able to detect it in the logRF trace.  Somewhat counterintuitively, renewing the channel conductivity of the positive leader channel with a stroke is usually associated with a strong reduction in needle activity.  I don’t think this has been directly shown following CGs (other than in this study), but it has been described after K-leader strokes which are similar (Jensen 2023).  It then takes a while for needle activity to resume.  

It is interesting that the negative and positive leaders behave so differently when subjected to very similar stimulus.  Where the positive leader produces bright bursts of radiation, new conducting channels aren’t usually formed.  In contrast, the same potential wave arriving at the tip of a negative leader produces an immediate and dramatic effect which produces significant new channel structure.  

Also Interesting, there is a similar effect to what the authors describe here seen in IC flashes after the negative leader stops propagating, even though there is no conductivity renewing stroke.  

• Shao 1995: https://doi.org/10.1029/94JD01943
• Stock 2017: https://doi.org/10.1002/2016JD025909
• Pu 2021: https://doi.org/10.1029/2021GL093145
• Stock 2023: https://doi.org/10.3390/rs15143657
• Hare 2019: https://doi.org/10.1038/s41586-019-1086-6
• Hare 2021: https://doi.org/10.1029/2020JD034252
• Pu 2019: https://doi.org/10.1029/2019GL085635
• Edens 2012: https://doi.org/10.1029/2012GL053666
• Van der Valde 2013: https://doi.org/10.1002/2013JD020257
• Jensen 2023: https://doi.org/10.1029/2023JD039104
  
  
  
  

Technical Comments:

Line 97: Coquillat is a good citation for the SAETTA deployment, but the context indicates this to be a citation for how the LMA locates lightning (usually this is Rison 1999 Thomas 2004).  You may want to revise this to be a little more clear.  

Line 160: “Figure 1.a” should be “Figure 1a” to be consistent with other references

Line 171: You discuss continuous emission in this paper in a few different contexts, some where the emission is continuous but impulse such that the LMA will well detect it, and in other places where the emission is smoothly continuous and the LMA may detect but not locate it.  It may be good to be a little more careful distinguishing these modes.  

Line 179: I think ‘units’ should be another word

Line 323: As noted above, I don’t think you need much speculation about the presence of negative leaders.  

Line 328: The positive leader does produce impulsive radiation as it moves through the negative charge region in the form of needles (citations above).  The tips also radiate, but usually aren’t detectable (citations also above), this is a smoothly continuous emission and shouldn’t be locatable by the LMA.  

Other comments:

The authors have done very well with the dataset they have available to them.  That said, I also find myself wishing that the dataset in this study were better than it is.  SAETTA appears to be doing a poorer than expected job of mapping out the in-cloud lightning channels, making the results somewhat hard to interpret.  The authors have largely overcome this by analyzing the LMA data at the individual station level, but I have seen much more detailed observations of the behavior the authors are describing.  While well outside the scope of this study, if the authors are interested I believe I can provide them with more highly detailed mapping observations of CGs including fast electric field change records to expand upon the observations in this study.

General Comments:

This is an interesting study that analyzes VHF emissions immediately before and after the RS, and the results are important to the field. The introduction is comprehensive and clearly explains previous work and the primary objective of the study. The instrumentation section is concise and the data analysis is well-organized based on the discharge types and their differences (-CG, +CG). The data analysis and discussion sections could be strengthened further by adding a table of relevant statistics for VHF source counts, power, and durations.



Specific Comments:

It would be fascinating to see the analysis of this study with the addition of electric field data if it is available for the observed lightning discharges. It would also be interesting to see the mean or median VHF peak frequency of the strongest VHF source detected by the selected LMA station.



Technical Comments:

- Lines 135 - 140: Figure 1 resolution appears lower than other figures.

- Line 179: "Units" should probably be changed to a different word.

- Line 241: This should probably start as, "As an example," similarly to line 210.

- Lines 308 - 311: The beginning portion of this sentence could be worded to be more clear/concise.