Comment to “Physical Interpretation and Implications of Convective Impulses in Thunderstorms Based on Lightning and Polarimetric Radar Observations”

Using polarimetric radar and lightning observations, along with the model simulations, this study investigates the potential physical mechanisms that result in the convective impulses in thunderstorms – the breakup of large raindrops and the released energy from frozen small size raindrops. In principle, the results are interesting and worthy for publication, with following revision comments.

Line 59-61, no need to use “;”

Line 74-78, Regarding the aerosol impacts, a study by Wang et al. (2018, doi: 10.5194/acp-18-12797-2018) could be referred.

Line 92-94, Why unlikely?

Line 125, “Figs.”

Line 152, Why do the authors use linear interpolation?

Line 168-190, Uncertainties associated with these assumptions should be briefly introduced. Also, Is 0.31-0.36 mm bias small enough for this study?

Line 247, Is this local time or Beijing time?

Line 315-316, Why cannot the small secondary ice crystals grow into graupel size particles within a few minutes? (can grow into more than 10 um within 1 minute and then collision-coalescence processes or Bergeron process)

Even though I like the proposed mechanism that has been simply evaluated with model simulations, I still wonder the reliability or observational support for this hypothesis, solution of which could highly improve the value of this study. Of course, even without further study, this work is still worthy for publication.

The authors investigate the physical mechanism driving convective impulses (CI) in thunderstorms using polarimetric radar and lightning observations. They found that higher supercooled liquid water and graupel content occur prior to the CI, and raindrop breakup is associated with an increase in supercooled raindrops. These smaller droplets freeze into graupel-like particles, releasing latent heat, which may enhance the CI.

However, before the publication of this manuscript, the concerns below need to be addressed.

Line 61-63 : Although information about the occurrence of CI is given but it is not clearly defined what is CI

Line 159: Is there any reason using the ERA Interim data instead of newer ERA-5, also is this data spatially interpolated to 0.175⁰×0.175⁰? Because as per Dee et al., 2011, the ERA Interim has T159 horizontal resolution which corresponds to a Gaussian lon/lat 0.75° × 0.75°. Also is there any validation conducted for this data? And why 6 hourly and not 1 /2 hourly data used

1. Line 199-200 Does 6 hourly average wind shear and humidity conditions represent the actual pre-convective environment? As the wind shear may have been different near the initiation hour. It would be beneficial to compare the wind-shear for pre-convective hour (nearest 1 or 2 hour prior to first echo) in both A and B cases.
2. Line 199: When it is called wind shear, what type of wind shear exactly is or between what levels(or heights) such as low/high level shear ,

Line 165-166: Although many studies are cited about the “fingerprint” of microphysical process, more details need to be added specifically about how it is done in the context of this study.

Line 178-179: Isn’t the standard error of 1dB between ZDP AND ZH relationship related to the specific study region (Tiwi island) and not the current study?

Line 185-187: Although it is noted that the radar retrieval techniques were valid in stratiform region and Z<30Dbz, How the ice and graupel above melting level estimated

Line 201: Actually both environments are favorable for thunderstorms. Maybe using wording as “more favorable than A” will be more appropriate.

Line: 239 It is mentioned that the thunderstorm is rapidly collapsed in A and slowly dissipated in B. But looking at figure S1. Case A is purely isolated and well separated(no merging with other cells) whereas in S2, It is clear that the track of cell is not properly taken into account for example at 1242 the composite Z is reduced and at the next step of 1248 the Z again started to increase, again similar pattern at 1318 which suggest that this cell is regenerating (it is a newer updraft ), and also signatures of two cells merging together can be observed. So how can we say this cell as an isolated? This answers why Cell B is long lived, Best option would be to create a similar time series of RHI scans to separate the target cell.

Line 315-316, What is the proof for the statement “we do not agree that these secondary ice crystals grow into graupel sized particles within a few minutes”. Even a very small time (a minute) is very significant in growth of hydrometeors in convection.

The details about model configuration such as initiation mechanism, grid spacing, domain size, input sounding, aerosol conditions, etc. need to be included.

Line 340-341: Is there any study supporting this statement “the weaker environmental wind shear and CAPE promote raindrop breakup”?