Review of “Water vapor in cold and clean atmosphere: a 3-year data set in the boundary layer of Dome C, East Antarctic Plateau”, by Genthon et al (ESSD-2021-414)

General

This paper documents an important new dataset of near-surface humidity profiles from Dome C on the East Antarctic Plateau, obtained using a novel hygrometer previously described by Genthon et al, 2017. Unlike many conventional hygrometers, the new instrument is capable of accurately measuring humidities which are supersaturated with respect to ice. Since ice supersaturation occurs frequently at Dome C this dataset provides for the first time an accurate climatological description of near-surface humidity profiles and their variability on daily to annual time scales at this station.

The paper provides a clear description of the measurements and the means by which they were obtained. Some basic climatological analysis is presented, which gives useful insight into the processes that control humidity at Dome C. I have a few comments on the paper (set out below) that require attention but these are all relatively minor and should all be easily addressed by the authors before final publication.

Specific comments

1. Throughout the paper, the term “supersaturation” is used without qualification, and is generally used to imply supersaturation with respect to ice. There is thus some ambiguity in the use of this term and I would recommend that it should always be qualified by “wrt ice” or “wrt liquid water” as appropriate unless it is absolutely clear from the context which of these is being meant. Examples of places where clarification is definitely needed include the caption to fig.3, line 167 and line 473.
2. Lines 23-24: Make it clear that these temperatures are for Dome C, not averages for the whole plateau.
3. Lines 78-79: Strictly speaking, the humidity gradient isn't the origin of the turbulence (which is generated by wind shear or convection). Maybe say "...because it enables the calculation of vertical moisture transport and exchange with the surface."?
4. Line 163: I'm not sure that this is the "traditional" view. Cloud physicists have known for a long time that ice supersaturation occurs in mixed-phase clouds - it is the basis of the Bergeron-Findeisen process, formulated in the 1930s. However, the occurrence of near-surface supersaturation wrt ice seems to have been largely overlooked until appropriate measurements (King and Anderson, 1999; Genthon et al, 2017) became available).
5. Figure 8: Is “Hour” local time? Give the difference to UTC.
6. Figure 9: There is very little information on this figure. You could make it more informative. Maybe show seasonal mean profiles?
7. Lines 326-327: King et al (2001, DOI: https://doi.org/10.3189/172756501781832548) present estimates of water vapour fluxes for Halley from humidity profiles.
8. Lines 373-374: Not sure what you mean here - that water vapour profiles cannot be used to diagnose convection?
9. Lines 475-476: “Thus, it is no wonder…” Do you have a reference that shows that models with appropriate microphysical parametrisations do produce ice supersaturation near the surface at Dome C?
10. Section 4: Also using a heated hygrometer, King and Anderson (1999) observed frequent ice supersaturation at Halley but no saturation wrt liquid water. They suggested that this indicated that, even in the clean polar air, cloud condensation nuclei, which will initiate droplet formation at very low supersaturations wrt liquid, are relatively abundant, while ice nucleating particles are rare.

Minor points and typographical corrections

1. Line 76: “operational”
2. Line 115: Capital “D” for Dome C
3. Line 144: Delete space in “humidity”
4. Line 183: “divergence” (not convergence)
5. Lines 308-310: I don’t understand this, please clarify.
6. Line 335: “upward” (not downward)
7. Lines 467-470: Insert “increasing” before “upward” and “downward”

John King

Review of "Water vapor in cold and clean atmosphere: a 3-year data set in the boundary layer of Dome C, East Antarctic Plateau" by Genthon et al.

This manuscript describes humidity data generated from a tower over Antarctica. This is an important data set in a unique environment. The authors do a good job of describing the relevance of the data. The manuscript is generally well written. It spends most of its time doing basic analysis of the data, with some interesting results. It is quite curious that there is no supersaturation with respect to liquid seen, even if the authors imply it from qualitative observations of possible droplets. I have a few major concerns, and minor comments. 

My major concerns are:

1. It seems odd to me that the manuscript pays more attention to analysis than to description of the sensors, calibration and errors. That seems appropriate for ESSD, while the scientific analysis perhaps belongs somewhere else. Section 2 should be expanded with more detail on the error characterization. 

2. In particular, I think the paper should use the best available conversions for water vapor saturation pressure (e.g. Murphy and Koop 2005), or show several different ones. Particularly for Relative Humidity over Liquid. 

3. In general the plot quality for line plots is not that great. As noted, some of the plots are redundant (Figure 7, Figure 10). 

4. Some of the plots could be improved. The PDF historgrams at different heights (Figure 5, 13) would be easier to interpret with overlaid transparent filled colors for the bars.

5. There should be more discussion of possible errors in section 2 or in section 4. This  seems strange given the discussion of potential supercooled liquid water, but no evidence of RH liquid > 100%. 

6. Links: not sure why the DOI refers to 3 more DOIs, one at each level. But I guess that is the Authors' choice. 

Specific comments:

Page 5, L106: what is this reference to Genthon et al 2022: it’s not in the references. Is it supposed to be the data itself?

Page 6, L144: Does the Humicap calibration assume a saturation vapor pressure relationship? Seems like it must in the empirical calibration somewhere. Please elaborate and specify what is used or why it is not used.

Page 7, L163: Is the modified sensor in 3b the heated inlet on the right side of figure 1? Please clarify.

Page 8, L167: does "adapted" mean "heated"?

Page 9, L181-8: Most of the discussion of mechanisms here is speculative. Do you have data that can bear on this? Figure 3b supports this figure, but not the mechanisms. Maybe show low winds with low temperatures? Or other measures of subsidence? Subsidence would lead to drying, but also warming, so pushing air out of those temperature ranges. Air aloft is going to have a higher potential temperature than surface Air, so without large radiative cooling it would be warmer.

Page 11, L240: Figure 7 doesn’t really add new information. Couldn’t you add standard deviations as colored shading to figure 6? Then you could do it for RHi and PPW too…

Page 15, L320: Figure 10. What does this figure show that is different from figure 9? I don’t think this figure adds value to the paper and could be removed.

Page 18, L410: might want to note that since RhL < RHi then if RhL > 100% then RHi > 100%. That’s the logic here right? If not, then how do you get supercooled liquid if RhL < 100%?

Page 19, L445: Was the observed haze liquid or ice? That would seem to be an important distinction. Was there ever evidence of supercooled liquid at Dome C?

Page 19, L447: if you used a different saturation vapor pressure like Murphy and Koop would you get a different answer? That should reduce conversion inaccuracies.

Page 19, L449: But earlier you argued that Dome C was pretty homogeneous? How does that match with the heterogeneity you imply here.

Page 22, L508: This is the same repository and doi as Genthon et al 2021? Just checking that is supposed to be the case. What is the reference to Genthon et al 2022 noted above?

Review of "Water vapor in cold and clean atmosphere: a 3-year data set in the boundary layer of Dome C, East Antarctic Plateau" by Genthon et al. submitted to ESSD

The paper describes a 3 year timeseries of data from an Antarctic station tower with improved instrumentation to observe supersaturation with respect to ice. The discussion of the observations highlights the frequent occurrence of supersaturation with respect to ice at this location, occasionally up to water saturation. It is suggested this is a useful dataset for atmospheric model validation, with similar conditions to the upper troposphere.

This is an important and useful dataset for helping to quantify and understand supersaturated conditions and for evaluating models. The data is well described and the paper includes an informative analysis of the data.

I have a number of comments where further clarity is required, particularly relating to gradients and turbulent fluxes, some of the Figures and use of supersaturation with respect to (wrt) ice or water.

Comments

In the "Short summary" for the paper, the last sentence is

"While supersaturation with respect to ice is frequent throughout the column, relative humidity with respect to (supercooled) liquid water reaches close to saturation."

As a one sentence summary of the results, I think this sentence, as it is, could be confusing.

Suggest something like:

"Supersaturation with respect to ice is frequently observed throughout the column, with relative occasionally humidities reaching water saturation."

Line 34

"and even more as freezing nuclei", but you mean even less? Suggest reword to make clearer.

Line 41, Tomkins -> Tompkins

Line 78

"the vertical humidity gradient … is the origin of turbulence". It's not the origin, do you mean that it is a result of the turbulence. Please reword this sentence to make it clearer.

Line 93

Only use of "MO" is here and it is not defined. Suggest just state explicitly "Monin-Obukov"

Line 96-97

"the vertical moisture gradients used to represent the vertical distribution and mixing of moisture in the boundary layer". Again, a confusing statement. Please simplify/reword, e.g. "the vertical moisture gradient as a result of the mixing of moisture in the boundary layer".

Line 115, dome C -> Dome C

Line 117, according to Goff and Gratch

Lines 124-131.

There is a bit of duplication in this paragraph and it could be made simpler/shorter to improve clarity. Line 129 -> saturation vapor pressure.

I don't think Figure 2 is necessary and would suggest removing it. The same curve (although on a linear scale) is in Figure 3.

Figure 3. For context for later discussions, you could also consider putting svp wrt water on the Figure, highlighting the upper bound of RHi?

Line 164, "by the CC curve"

Line 160-180

Because of the previous discussion that the instrument reports vapor pressure with respect water, for clarity it is really important you are clear that Figure 3 and this paragraph discusses vapor pressure with respect to ice, e.g. on line 167, and "ice saturation" on line 178. Also mention this in the caption for Fig 3.

Line 167,  "frequently largely" -> "frequently"

Line 186, "adiabatic and radiative cooling combine"

Radiative cooling is more obvious, but could you say more about the conditions in this mid-temperature range that are leading to the adiabatic cooling (i.e. ascent) as this is perhaps less obvious over this region.

Line 197, the 18m level also does not show in the 105%-115% range. For improved clarity and simpler explanation I suggest showing the bars in a different way in Figure 5 so all can be seen - either offset or a set of three bars for each bin.

Figure 7  - this is exactly the same data as used for the temperature data in Fig 6. Is it needed? Could the information be put on Fig 6 or just described in the text?

Line 268, "mid" summer/winter rather than "full" summer/winter?

Line 269, refers -> refer

Line 270 , remove "at"

Line 312, "vertical profile of moisture content determines the turbulent fluxes"

Again, surely the other way round, the vertical profile is determined by the turbulent fluxes. Or you mean it determines the sign of the transport due to the turbulent fluxes? 

Does Figure 10 add much? You could just say you looked at individual times to confirm it is not a result of the averaging?

Lines 330, 332. Which gradient? Be clear you are talking about the PPW gradient in this case.

Line 335, "water vapor flux is downward, exporting surface sublimated moisture"

Did you mean upward?

Line 372, for dry convection, surely potential temperature is the key quantity for instability, not temperature?

Line 468,"diurnal cycle in the summer, upward.."

Diurnal cycle of the moisture gradient…

Figure 13, same comment as for Fig 5 for improved clarity.