The paper describes observations of supercooled liquid water clouds over Dome C, Antarctica with multiple instruments: balloon-borne Vaisala PTU and Anasphere SLW sondes, lidar, and HAMSTRAD microwave radiometer during field campaigns carried out in the period of 2021-2022. Generally supercooled liquid water clouds were observed near temperature inversion zone where atmospheric motion is capped. Both consistency and discrepancy among the different measurements were observed.

There are a few questions I would like the authors to address:

1. I totally get it that it is difficult to perform in-situ measurements over Antarctic and it is important that someone are doing it. On the other hand, as a scientific paper, can you emphasize the novelty and impact of this paper, i.e., what is the new findings of this paper that readers cannot get somewhere else?

2. Under nominal conditions, the LWP values obtained by integrating SLWC sonde profiles are consistent with the HAMSTRAD measurements. Under non-ideal conditions, they are way off. The authors suggest the HAMSTRAD values are more trustworthy. Is this always the case? If yes, why don't we just perform HAMSTRAD observations all the time? What is the added value of doing SLWC sonde profiling? 

3. Figure 1 is not informative at all, which could be replaced by a cartoon drawing showing how things work.

4. What is the functionality of unwound string or unwinder? It is not clear.

In situ observations of supercooled liquid water clouds over Dome C, Antarctica by balloon-borne sondes

Author(s): Philippe Ricaud et al.

MS No.: amt-2024-8

The manuscript describes the measurement of cloud supercooled liquid using in situ measurements at the Concordia station in Antarctica. Cloud supercooled liquid water is an important parameter that needs to be better measured, especially in Antarctica, because it is not well described in numerical models. In addition, measurements in Antarctica are always very challenging. Therefore, the manuscript has an important contribution to the readers of AMT. However, the manuscript has some major problems, which I comment on below. I proposed to accept the manuscript after a major revision.

Major Comments

The manuscript spends several pages discussing the potential temperature, the flight description, and the radiosonde trajectories and misses the most important part, the measurement of cloud liquid water. I suggest the authors focus the manuscript on the main points. I miss a statistical calculation of the errors between the different instruments. I would like to see how all measurements and the selected ones compare with remote sensing observations. I would like to see the lidar measurement profiles combined with the sonde calculated SLWC in Figures 6, 10, 11 and 12. One point that is not clear to me is the multilayer of supercooled liquid water clouds in different layers. In my interpretation, I would expect only one layer, as shown by the lidar in Figure 2. Is the lidar attenuated for the other layers? I suspect not, these are very thin clouds. My main suggestion is to avoid all this parallel discussion in the manuscript and focus on the main points - how these measurements compare with remote sensing. Do the measurements make sense? What is the bias, the mean square error? Are these measurements relevant for calculating an accurate radiative effect of these clouds compared to just ice clouds? How many W.m-2 is the error, is it larger than the estimated 40 W.m-2, what is the impact on radiative transfer if this multilayer is considered at the wrong height?  What are the limitations of this method? How to avoid such out-of-scale values as in Flight 14? Why is the lidar liquid water cloud so different from the measured one? What is the impact on the radiation simulation?

Minors Comments

Line 58 – Please add references

Line 61 – Please add references

Line 68 – LWP and SLWC are the same, so why call them by different names?

Line 88 – “At Concordia station, several studies from remote-sensed observations already took place to evaluate....”.  “4) the differences between observations and model simulations of SLW clouds.” This is not done at the Concordia station, and remote sensing is not used. Please write this paragraph without using numbers but a coherent description of the studies using Concordia data.

Line 131—Please clarify. When the SLW reaches the wire, I would expect the liquid to be instantly converted to ice.

Line 136 - its derivative with respect to (wrt) time - the abbreviation wrt is standard jargon in mathematical papers - please avoid it.

Line 142 - the vertical velocity is variable and could be updraft or downdraft. Please clarify.

Line 144 - What error is associated with w and droplet collection efficiency (depending on size)?

Line 152 - Why are the values integrated if the data is collected every 100 m?

Line 180 - Is there a control for wet radon and fog events? How does the radiometer compare with lidar and radiosondes?

Line 248 - "the profile of the SLWC" - SLWC is an integrated value - how do you define the profile?

Table I - Why does Lidar sometimes have the SLW layer and other times not? Please add a new table for the clear air flight, it has different variables.

Line 304 - How do you calculate supersaturation with Vaisala radiosonde? What is the accuracy?

Line 307 - Are the clouds 70 km from the radiometer similar to those observed on Concordia?

Line 355 - The authors mention that L03 and L04 are consistent with lidar - where are the lidar data? Are they referring to Table I? The figure in the supplement (S2) shows only one layer.

Figure 8—It was not cloudless because there was LWP all day. In fact, it was a day with thin clouds below the defined threshold.

Figure 10 - Many levels indicate liquid water (frequency peaks) during the flight - How was the LWP calculated on this day? How low relative humidity is associated with the line frequency?

Line 496 - The explanation for the possible error related to the fog that could stick to the wire is still not clear to me. What happens when the