Review of the manuscript with the title "Total ozone variability and trends over the South Pole during the

wintertime" by Fioletov et al.

The manuscript studies total ozone measurements at the South Pole during winter time. Dobson, Brewer, ozone soundings and MERRA-2 data are used. The study includes quality assurance of the data as well as uncertainty estimates. This is the first time the analysis is done for such a long period in the winter time, as previous studies are more focused on springtime Antarctic ozone loss. The study brings new information on total ozone trends in the Southern polar area and adds valuable information to global ozone trends at those latitudes. The manuscript is well-written and includes appropriate references. The authors should address the general comments and the more specific comments before publication of the manuscript.   

General Comments:

My main concern is related to the correction of MERRA-2 data which is based on only one correction factor over the whole winter when analysing the MERRA-2 SBUV period. Why not use a correction factor per month, as suggested by the Figure 3? Please see also the specific comments. The second concern is the use of the EESC fit. For a reader not familiar with the methodology the text is rather confusing. This needs to be clarified.

Specific Comments:

Page 2, Line 5: Polar vortex breakdown time: November-December?

Page 2, Line 22: total ozone declining trend: for which time period? Yearly mean?

Page 3, Line 28: Please specify how the temperature and ozone profiles are used in the retrieval algorithm.

Page 4, Line 8 and page 5, line 4: Until which period you can still measure good DS measurements? In the winter, with no Sun, I suppose it is impossible.

Page 4, Line 9: How do you make the choice of the representative daily value?

Page 4, line 30. I don't think Kerr et al. 2010 is a "recent" paper ....13 years old. Maybe better to delete "recent". Please add some sentences on the calibration and harmonization of the Brewer including used corrections to the data, like the use of sl-lamp measurements.

Page 6, line 4: For MERRA-2 during the polar night: is the total ozone for the South Pole actually the total ozone at latitude 82 degrees South? Or is it extrapolated somehow from that latitude to the South Pole? Please specify.  

Page 6, line 14: "...existence of some systematic seasonal biases..." Does this include any information about the winter (the season you studied)? If yes, please add it to the paragraph.

Page 7, line 1: "This could be related to some horizontal inhomogeneity of the ozone distribution over the pole that led to variations in measured ozone due to changing lunar azimuth angle." I can't follow the sentence. Please clarify. Are the changes real or an artifact caused by changing lunar azimuth angle?

Page 7, line 31: "There was also a 3% difference..."   Was it the mean difference? Please specify.

Page 7 + Figure 3: Correction of MERRA-2 data for 1980-2004. Why didn't you use monthly mean differences for the correction in April-August? There seems to be a clear month-dependent difference. You can still see a month-dependent difference in the corrected data ranging from -5% to 5%.

Page 9, line 25: "The estimated ozone variability is relatively low, about 15 DU for daily averages and 10 DU for monthly

values or about 6% and 4%, respectively." -> Do you mean over the winter period April-August? Please specify.

Page 11, Section 4.2. For a reader not familiar with EESC fit it is very difficult to understand what is fitted to what. You should describe the methodology in more detail, and how did you end up with the fit in the right columns of Fig. 7. 

page 14, line 10: "A decline in ozone due to gas-phase ozone destruction from ODSs is probably the largest since the time for an

air parcel to travel from the tropics to high latitudes is the longest." -> Please open this statement. Is this related to the Brewer-Dobson circulation transporting ozone from the tropics to the pole? Does ODS deplete it the whole way? During which months?

This paper reports on ozone changes observed since the 1960s at the

Amundsen-Scott South Pole Station (SPO). A particular focus lies in

the winter months just before the period of rapid ozone depletion in

spring ("ozone hole period"). This time of the year is covered by full

moon (FM) measurements by Brewer and Dobson spectrophotometers

complemented by regular ozone sondes and MERRA2 reanalysis data.

Regular direct sun (DS) measurements by Brewer and Dobsons are also

used to contrast the ozone changes in the winter months to other

seasons and latitudes. Wintertime ozone declined by about 12% from the

pre-1980s until the late 1990s, a larger change than observed at lower

latitudes and in other sesaons, except for the ozone hole period.

The paper is well written and results are well presented. After

adressing some rather minor issues as outlined below, the paper will

be well suited for publication in ACP.

p. 6, l. 15ff: A great part of the paper deals with the adjustments of

the various datasets. As a reference for adjusting all data types the

long-term Dobson data is used (p. 7, l. 14). For justification only the

study by Bernhard et al. (2005) is cited. I think a few more sentences

are here needed to explain why the Dobson record is most suitable as a

refence dataset here.

p. 7, l. 32: "... remove that bias for some of the plots". So the

corrections are only applied in the plots but not for the data. I find

this a bit awkward, why not say simply the data has been corrected,

which would be important if the adjusted data im made publicly

available (see my later comments on Data Availability)

p. 8, l. 5: I think that the bottom panel is not showing what is

described in the main text and figure caption. MERRA2 data are the

same as in the top panel, but all other data have been adjusted (not

the other way around).

p. 11, l. 5:  "four seasons" --> "three seasons"

p. 11., l. 25: not clear what is meant with "analysis of the

residuals". Please specify. A fit of only the EESC curve to the data

will result in large residuals as the short-term variability is not

fitted. Maybe it would be good to show some plots of residuals to make

the point here (could be put in the appendix).

Fig. 8 (and other plots): Light blue color lines are dificult to

distinguish, in particular with different line styles. The light blue

color is not a particular good color for color blind people. I

strongly suggest to use a different color. This applies also for the

other plots using the same color.

p. 14, l. 7: "Rapid ozone destruction on polar stratospheric clouds in

the springtime Antarctic vortex affects ozone levels in subsequent

months everywhere in the southern hemisphere, but its impact on the

polar ozone should be at least as strong as anywhere else." I find

this sentence a bit awkward. I think what was meant to be said is that

the polar ozone loss (ozone ozone hole) is not affecting the

wintertime ozone, so that gas-phase chemistry is only relevant in

winter. Apart from ODS gas-phase chmistry some dynamic contributions can

not be ruled out as suggested in the main text. I think this should be

mentiond here as well.

p. 14, l. 17: The data at SPO, in particular the adjusted data

should be made available publicly for tracability.

p. 34, l. 4: "in increasing" --> "is increasing"