Exploring hail and lighting diagnostics over the Alpine-Adriatic region in a km-scale climate model

The study presents simulations of severe thunderstorms in the Alpine and Adriatic areas. The eight case studies are run with ERA-5 boundary conditions and initiated at 12 UTC of the previous day to allow for spin-up. The model simulations include, for the first time Hailcast and LPI diagnostics to identify and characterize hail and lightning over the study areas in Europe. The simulations are compared in detail to observations.

The paper is well-written, it is very well structured, and the figures are very clear and well-readable. The results are highly relevant, and the detailed process discussions are instructive and illustrate the diversity of thunderstorm environments in the complex topography. The paper is ready for publication after some minor revisions.

Kind regards

Olivia

Major point

L312ff I think that there is an issue with this comparison. If you compare two relative frequency distributions that both sum up to 1 for a variable that is expected to have an exponential distribution and one distribution starts at 0mm and one at 20mm, then by construction you will always find relatively fewer large values in the distribution that starts at 0mm. To check if this is true, you can artificially cut the model data at 20mm. Since the issue is not present in the comparison with the crowd-sourced reports, I suspect that your comparison improves a lot if you use only hailcast data of >=20mm.

Minor points:

L 20 e.g. Punge and Kunz …

L30 averaged --> average

L31 is located between

L33 A transition sentence is needed before the sentence “It remains ….”

L35 local-scale

L39 With this --> Radar provides

L45 due to multiple

L49 The statement “has been used … “ is very vague, maybe be a bit more specific.

L59 troughs

L119 Please specify in more detail the variable that hailcast stores at hourly intervals. Is it the mean, max or instantaneous hailstone size?

L131 is epsilon a function of height?

L150 Maybe add a sentence on the ability of RhighresD to capture convective precipitation? Local maxima could be missed due to too coarse station density.

L153 in-situ observations and remote radar-based observations

L167 crowd-sourced reports --> insurance loss reports

L169 delete “on the other side” since there is no matching on the one side

L169 in units of cm starting at 20mm :-)

L172 2009 -->2002

L186 we use data from

L200ff Please explain the calculation of tau in a bit more detail. Is dCAPE/dt a daily mean or daily max value?

L223 heavy --> severe

L255 near the mountainous region of central Europe --> a bit more specific?

L265 Please explain in a bit more detail how you compare across the different spatial resolutions and how sensitive the method is to the interpolation in space. I suggest adding 1 – 2 sentences of discussion.

L298 Note that the black forest is already quite far away from the radar, and the data quality may become an issue because the radar gates become very big and the lowest beam height is quite high. If you compare only within a 140km area around the radars, the radar quality should be very good.

L303 Please mention that both radar-based hail products only see hail in the air and not on the ground

L371 Please mention the time of the hailfall at the beginning of the case study description.

L380ff I do not yet fully get my head around the timing of the individual steps and the position of the cold front. You mention the advection of cold air; this would happen behind the cold front. But then you mention the increased dew point ahead of the cold front. It would be great to know where the front is, at which altitude, and at which time to get the full (complex) 4-D picture of the event.

L409 becomes warmer --> I do not see this; to me, the temperature near the surface looks colder at 00 UTC than at 12 UTC.

L418 Given the southeasterly flow --> this implies causality; please explain in more detail

L432 Please explain in a bit more detail why the stable was eroded by the upper-level westerly flow

L439 add not shown

L461 It is unclear to what “this” refers and what is meant by these kind of events (I think there is some grammatical confusion as to whether you mean the synoptically driven cases or not)

L471 three cases that were selected

L485 In this context … this statement is not very clear to me yet

Figure 1, caption four --> three

Figure 11 text says the sounding is at Payerne, but the figure title suggests the sounding is at Stuttgart.

Figure 11f Please add the borders of Switzerland.