Measurements and Modelling of Total Ozone Columns near St. Petersburg, Russia

Round 1

Reviewer 1 Report

This manuscript describes the characteristics of several kinds of datasets for total column ozone in St. Petersburg. It is useful to use the plural datasets to research the temporal variability of total ozone. However, some improvements are required.

1) Abstract is too long. I think that the first sentence in the abstract is not essential. In addition, ozone layer depletion is not solely caused by anthropogenic activities.

2) Lines 57-60: Please add more references. 

3) In Section 2, all of the sub-sections are not numbered. Please correct it. 

4) Please check the typo-error. Especially, 'micro meter' and 'No' shows the typo-error in several points. 

5) Lines 194-197: Could you check the spatial resolution? 13x24 and 13x48 km2 are only valid for the nadir viewing.

6) Line 199: 0.2-0.5 nm is not 'spectral resolution', but 'spectral sampling'. Spectral resolution is related to the instrumental response function for the radiance sensitivity.

7) In Section 2, the author listed several tables to describe the characteristics of the dataset. However, the manuscript is essential to describe the detailed data selection and colocation methods for the intercomparison and validation of plural datasets. Please add it. 

8) Table 3-4: I don't exactly know how to estimate the differences of TOC observations. The dataset for the comparison is used as the 'daily average'? 'representative data with some priority order'? or 'temporal time difference span from overpass time'? We can't find the detailed rules of the data selection for the intercomparison.

9) In Table4, what is the 'natural TOC variability'? Please clarify the definition of this word.

10) For the comparison, why the scattering based observation (such as the Zenith-Sky for Dobson) is used? The scattering based observation has large uncertainty and it is not suitable for the exact validation.

11) Line 318-345: As the author mentioned that the tropospheric ozone is 10% of total ozone, and 20-30% of total ozone variability is due to the difference in tropospheric ozone. I don't understand that the tropospheric ozone variability is larger than total tropospheric ozone amount. Is that correct?

Also, The tropospheric ozone variability is not only caused by the anthropogenic effect, but also driven by the natural effects. Please add and revised it.

12) Lines 375-382: Please add the reference and some analysis to explain the SDD characteristics. 

13) From Table 5-7, I can't decide to analyze the statistical characteristics. For checking these results, the information of colocation between observation and modeled datasets are very important. Also, level of the satellite dataset information (Level 2 pixel-based? Level-2 gridded? Level-3 CMG?) is important to analyze in Figures 6 and 7.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This is a useful paper and warrants publication pending minor revisions.

Some comments for authors  to consider:

1.            Maintain consistent manuscript font as per RS guidelines

2.            Page 2, 4th paragraph, Lines 3-4, “In the last decades …” remove ‘s’, should be decade

3.            Same correction as mentioned in 2. 1st paragraph, page 3 line 3 decades should be decade

4.            CTM is improperly abbreviated in page 3 paragraph2 chemistry -transport mechanism? Or models?

5.            Table 2 can be improved. It is congested, suggest breaking up the characteristics into more columns if necessary.

6.            Page 10: “This can be due to larger spatial and temporal observation mismatches which we used (100 km and 24 h vs 70 km and 1 h).”  What would happen if you had reduced your observation mismatches? Did the MD and SDD come down?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This paper devotes to comparison of total ozone measurements from various sources. According my opinion it is a very nice paper with interesting results especially their correction which substantially improves results But I have one question In paper you give the confidence intervals of MD, but these intervals includes zero many times and you stated the differences are significant or important. Could you explain it? 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The manuscript presents and compares total ozone data from three ground-based instruments, three space-borne instruments, two re-analyses, and two model simulations. All total ozone data are near St. Petersburg, Russia. Data, methodology and presentation are generally sound and clear. The manuscript is quite comprehensive, but also quite long. 

All data sets typically agree within better than 5 %, and all report more or less the same ozone variations. The manuscript gives a good description of total ozone data and general bahaviour near St. Petersburg. It think it is acceptable for Remote Sensing, after a few minor revisions.

English copy editing would be helpful.

In Fig. 1, a typical footprint of an OMI measurement could be shown as well. Maybe also a TROPOMI footprint.

Page 4, 2nd paragraph: might be good to also state here at which URL the M124 data are available (analogous to the 1st paragraph on page 5 giving the URL for the FTIR data).

Also: it would be good to make the Dobson data available (at WOUDC and/or NDACC), and give the URL.

Page 7, 1st paragraph, 2nd line, "either": the authors probably mean "as well" 

Page 12, last paragraph:  The authors might want to cite Zhang et al. 2021 here, which use basically the same approach for merging total ozone data at Halley Bay.

Zhang, L. N., Solomon, S., Stone, K. A., Shanklin, J. D., Eveson, J. D., Colwell, S., Burrows, J. P., Weber, M., Levelt, P. F., Kramarova, N. A., and Haffner, D. P.: On the use of satellite observations to fill gaps in the Halley station total ozone record, Atmos. Chem. Phys., 21, 9829–9838, https://doi.org/10.5194/acp-21-9829-2021, 2021.

Page 17, Section 3.4.1. Please clarify whether the original data sets are used in the average, or the data sets corrected to match the Dobson data.  Currently this is not clear / confusing.

Page 19, Eq. 5: This seems wrong to me. Would you not obtain the deseasonalized ozone simply as:

TOC_M,Deseason = TOC_M - sum_n (e_n*sin(2pi*t/I_n) + f_n*cos(2pi*t/I_n) )

please check and correct as appropriate.

Also: since the annual cycle terms (in the Sum) and the trend term (d *t) should be almost orthogonal, you should get virtually the same trend from TOC_M,Deseason and TOC_M    (Eqs. 4 and 5).

Page 19, 3rd paragraph: TOC increase by +0.4+-0.09 DU/year. Is that uncertainty 1sigma or 2sigma? The end of the paragraph seems to indicate 2sigma. Also: Does this uncertainty account for auto-correlation AC in the residuals? If not, you have to increase the uncertainty, e.g. by a factor of sqrt((1+AC)/(1-AC)), or account for auto-correlation in another way, e.g. Cochrane-Orcutt method. If the trend uncertainty changes here, it needs to be corrected also in the abstract and in the conclusions.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I don't have additional comments.