Effects of Ozone and Clouds on Temporal Variability of Surface UV Radiation and UV Resources over Northern Eurasia Derived from Measurements and Modeling

Round 1

Reviewer 1 Report

All my comments are addressed.

Author Response

Thank you!

Reviewer 2 Report

Please see the attached file and thanks

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

***General comments

I thank the authors for their efforts to address many of my comments in my review of the original version of the manuscript. However, the paper still requires substantial efforts before it can be considered for publication. In my very detailed second review below, I tried to be as explicit as possible and provide text that can replace those parts of the paper that need to be improved the most, both in terms of contents and language. I also made suggestion for removing material that either contributes little to the overall message of the paper or is inconsistent. For example, the analysis of the different SST/SIC datasets raises more questions than it answers. More work (e.g., additional model runs) would be needed to address this issue. I also feel that the TOMS/OMI dataset is not appropriate to assesses changes in cloudiness at high latitudes, partly because of the uncertainty caused by the interaction of clouds with surface albedo, and therefore suggest to remove this dataset from Figure 13 and 14.. 

In my original review, I pointed out that the definition of Qery (now Eery) is ambiguous. I asked whether Eery is the erythemal irradiance at noontime, at the time of the satellite overpass, or a general measure. I further asked whether trend estimates based on spectral irradiance at noontime, spectral irradiance averaged over 24 hours, or something else. In their rebuttal, the authors asserted that “We analyzed daily sums.” (L92 of rebuttal letter). This is in conflict with the revised paper, which defines Eery as an irradiance (L15 & L48). Also trends in Eery (e.g., Fig. 10) refer to irradiance, not a daily sum (also known as a daily dose). The authors should double check whether their analyses are really based on irradiance, and correct their results if this is not the case. Hopefully the confusion is only caused by an incorrect statement in the rebuttal letter.

 

***Specific comments:

L31: If the abstracts needs to be shorter to meet the journal’s guidelines, the last sentence of the abstract could be deleted because there are virtually no people living in the “Central Arctic” regions (or the Arctic Ocean). Changes in UV resources in this region are irrelevant for human health.

L88: The sentence contradicts line 84 as the WACCM model mention in line 84 should have included this interaction. (But perhaps interaction was switched off in the results presented in publications 5, 35, and 36.). Please clarify.

Eq. (4): It's still not clear to me how X_2000 was calculated. Is it  X_i,j averaged over all months? The way X_2000 was calculate should be explicated stated, perhaps in another equation.

L212: What are Dobson sondes? Do you mean a Dobson spectrophotometer? If so, then these measurements are just part of the "independent ground-based observations" mentioned in the preceding sentence.

L215: If you compare the total ozone content (by the way "column" would be a more appropriate word), how can you discriminate between the stratosphere and troposphere? Did you use Umkehr data from the Dobson measurements?

L235: Regarding: “TOMS and OMI satellite datasets were used for comparisons with the Eery retrievals from the RSHU CCM and the ERA-Interim reanalysis. For characterizing Eery variability, we applied daily erythemal doses (Jm-2) from the combined satellite dataset.”

The first sentence implies the RSHU CCM and the ERA-Interim reanalysis are based on spectral irradiance, Eery. However, the second sentence asserts that the combined satellite dataset (I presume this refers to TOMS and OMI) is based on daily erythemal dose. So it seems that two different quantities are being compared despite my request to harmonize the quantities in my original review.

Eq. (8): It seems that tau_ext and omega are provided by the Macv2 climatology, but this is not clear. I am actually surprised that this climatology apparently provides these parameters at 320 nm considering that aerosol measurements at wavelengths shorter than 340 nm are very sparse.

Remove Eq. (10). It is confusing and does not explain anything. Just say that the threshold is defined by the Minimum Erythemal Dose MED_k, which depends on skin type. It would actually be good to spell out this dose (e.g., from [72]). For example, the Minimum Erythemal Dose for skin type 2 is 210 J/m2 (or whatever number was used) and provide also numbers for other skin types. This would be helpful because MEDs for different skin types are not unambiguously defined. Eq. (10), does not really specify the MED because it is defined by the integration limit t_MED_k. So it is defined in a circular way.

Eq. (11) and the following sentence are also very confusing. Please state your assumptions in plain English and provide a formula that translate your vitamin D threshold (i.e., 1000 IU) to a threshold in terms of J/m2 to be comparable to the threshold defined for erythema. Please also state that this relationship (whatever it is) is a very crude simplification because there is now overwhelming evidence that the amount of Vitamin D produced by sun exposure depends greatly on the Vitamin D status of the person receiving the dose. For example, someone with high Vitamin D levels (e.g., measured as 1,25-dihydroxyvitamin D) will make very little additional Vitamin D when exposed to the sun.

L308: The sentence “For this purpose, we calculated the absolute noon hourly Eery averages with 3-minute resolution for clear sky conditions…” is incomprehensible. Do you mean that you calculated an hourly average of Ery data, centered at noon, from data provided in a time resolution of 3 minutes? Please rephrase!

Table 1: Please round all value to 1 DU. Considering the uncertainties of the various datasets, providing ozone at a precision of 0.01 DU is not appropriate as it may imply that ozone can be predicted or measured with such a high precision. Also, please describe how the values in the Table were calculated. For example, is the mean the average of all pixels shown for a specific month in Fig. 3? The same question applies to median and standard deviation. What does “Case number” mean? Is the number of pixels from which the mean/median/stdev were calculated?

Regarding the mean, what does the range (+/-) and confidence interval refer to? Fig. 3 shows a climatology, not a trend, so I don’t understand what these +/- numbers mean.

L409-412 and Fig. 5: As mentioned in my original review, the percentages in the “all factors together” panel of Fig. 5 seem to be about 2% too low. I am aware that their could be synergistic effects such that the effect of four factors quantified in the other panels don’t add up linearly, but I am still surprised that the  “all factors together” panel is not more dominated by the “anthropogenic effect”. I encourage the authors to double-check their calculations.

Caption Figure 5: The expression “Annual difference in Eery anomalies between the 2000-2015 and 1979-1999” is hard to understand. Is the word “annual” really required considering that data are either averaged for the 1979-1999 and 2000-2015 periods? Also, according to Eq. (1) anomalies were calculated on an annual basis. So it would be more accurate to say: “Difference in the average of Eery anomalies for the period 2000-2015 and the period 1979-1999”.

L413 - L431 and Figure 6: The material presented here is not convincing. Since the same CCM was used for all data shown in Fig. 6, it seems that the greatly different results for three SST/SIC datasets are due to differences in the SST/SIC data. I conclude from this that our knowledge of the actual change in SST/SIC over time is too poor to allow firm conclusions how these changes have affected Eery. I therefore suggest that the authors remove the paragraph and Figure 6 from the manuscript.

Of note, to my knowledge, CCMs do not necessarily produce the same results when driven by the same set of initial conditions. For example, difference in the distribution of cyclone and anticyclones around the planet on day one of the simulations may have an effect on the patterns found years latter at the time when the simulations end. When using CCMs it is therefore standard practice to perform several model runs (or ensembles) and use the ensemble average to describe the effect being studied. For the same start parameters (e.g., the same SST/SIC dataset), does the CCM used here give the same results or does every run produce different results? If the latter were the case, differences between the models shown in Figure 6 may not be due to differences in the pre-described  SST/SIC but may reflect differences in model ensembles. The paragraph should only be kept if the differences in patterns shown in Figure 6 cannot be attributed to fluctuations about a ensemble mean of the CCM results.

Caption Figure 7: similar to my comment above, the expression “Annual Eery decadal trends” makes little sense. Either drop “annual” or say what you actually did: “Decadal trends in annually averaged Eery anomalies”.

L448: As mentioned above, “difference in Eery anomalies between 2000-2015 and 1979-1999” does not describe the figure accurately as anomalies are calculated on a yearly basis according to Eq. (1). I suggest: “Figure 8 shows the difference in the average of Eery anomalies for the period 2000-2015 and the period 1979-1999.”

L453 - 461 and Figure 8: I do not see any resemblance in the seasonal patters of the results from the INM-RSHU CCM model on one hand and the ERA-Interim and satellite datasets on the other. I don’t find that the text describing the Figures in L453 - L458 (“The discrepancy in … over Moscow [28]”) describes the figure appropriately. I suggest to replace this text with the following:

“The discrepancy in Eery changes between the different datasets is much higher for the seasons. While the results for the ERA-Interim and TOMS/OMI satellite datasets are in reasonable agreement for April and July, the agreement for January and October is poor. Results from the INM-RSHU CCM model are not able to match the observation for any month.”

Because of this change, the paragraph in L458 - L461 needs some adjustment, too. I suggest to replace the text in L458 - L461 with the following:

“Figure 9 presents time series of Eery anomalies at 48° N for longitudes between 0° E and 150° E based on the INM-RSHU CCM, ERA-Interim, and TOMS/OMI satellite datasets. Results from the three datasets agree within a few percent and suggest that Eery has increased between 1979 and 1995. In contrast, Eery has not changed perceivably between 2000 and 2015. These result corroborate the conclusion from Figure 8 that Eery was lower during 1979-1999 compared to 2000-2015.”

Change caption Figure 8 to: “Difference in the average of Eery anomalies for the period 2000-2015 and the period 1979-1999 due to total ozone changes calculated from the different datasets. Statistically significant differences at 95% are shown by white hatching. Note that the left-bottom panel is identical to the bottom panel of Figure 5.”

L462 - L469 and Figure 10: Similar to my comment above regarding Figures 6 and 8, the differences in the seasonal patterns (first four rows of Figure 10) are very large. The text in L462 - L469 does not describe the figure well and I suggest to replace it with the following:

“Figure 10 shows the spatial distribution of the trend in Eery caused by changes in total ozone over the period 1979 - 2015, calculated from the INM-RSHU CCM, ERA-Interim, and TOMS/OMI satellite datasets. For annual means (last row in Figure 10), trends computed from the three datasets are generally consistent and range between 0% and 1% per decade. One exception are larger trends of up to 2% for two regions in the ERA-Interim dataset. However, similar to the results shown in Figure 8, the discrepancy is much larger for the seasonal datasets. Results for the ERA-Interim and TOMS/OMI satellite datasets agree again reasonably well, while results from the INM-RSHU CCM model do not capture the spatial patterns of the observations.”

Change caption of Figure 10 to: “Spatial distribution of the trend in Eery caused by changes in total ozone over the period 1979 - 2015, calculated from the INM-RSHU CCM, ERA-Interim, and TOMS/OMI satellite datasets. Statistically significant differences at 95% are shown by white hatching. Note that the left-bottom panel is identical to bottom panel of Figure 7.”

L499: Delete “Antarctica”. Even if clouds were to increase substantially over Antarctica, the effect on UV would be very small because the very high (>96%) albedo over snow covered regions of Antarctica will greatly reduce cloud effects.

L521 - L530 and Figure 13: In general, I see almost no resemblance in the results for the three dataset for any month. The only exception is the good agreement for the annual mean between the CCM model and ERA-Interim datasets. I am particularly concerned that there is so little agreement with the ERA-Interim and satellite datasets considering that both are based on observations and partly draw on the same raw data. I can only conclude from the figure that the understanding of changes in Eery resulting from changes in ozone and clouds is very limited. I don’t have the means to double-check whether the results presented in Fig. 13 were calculate correctly and therefore urge the authors do double-check their calculations, in particularly those for the satellite dataset. If no processing error is found, I urge the authors to remove the satellite dataset from Figure 13 and 14. OMI and TOMS measurements were not designed to track changes in cloudiness at high latitudes and the results presented in these figures do not do justice to these instruments.

The text in L521 - L530 is far to optimistic. Presuming that the final figures do not include the satellite datasets, I suggest to replace it with the following:

“Figure 13 shows seasonal and annual differences in Eery anomalies between the 2000-2015 and 1979-1999 periods due to the combined effects of total ozone and cloudiness. For annual means (last row in Figure 13), results for the CCM model show increases by 0-3% between the two periods while changes for the ERA-Interim dataset range between -2 and 6%, with few exceptions. ERA-Interim results for July exhibit a quasi-wave spatial structure in mid-latitudes with a significant increase in Eery over Europe, South-Eastern Siberia, and over the Russian Far East. A similar structure, albeit less pronounced, is also apparent in the CCM model. The agreement in the spatial patterns of the two datasets is generally poor for other months. This disagreement may partly be caused by the interaction between clouds and high surface albedo from snow cover (e.g., see discussion in [67]). This interaction is particularly a problem over northern areas that are affected by snow cover from October through June.”

Change caption Figure 13 to:” Seasonal and annual differences in Eery anomalies between the 2000-2015 and 1979-1999 periods due to combined effects of changes in total ozone column and cloudiness according to the INM-RSHU CCM model and the ERA-Interim dataset. Statistically significant difference at 95% are shown by hatching.”

L542 - L551 and 14: Similar to the reasons given above, please remove the satellite dataset from Figure 14 and adjust the text accordingly. I suggest the following:

“Figure 14 presents Eery decadal trends due to the combined effect of total ozone and cloudiness according to the INM-RSHU CCM and ERA-Interim simulations. For annual means (last row in Figure 14), trends calculated from the CCM and ERA-Interim datasets generally agree, range between 0 and 3% per decade, and are statistically significant over large areas. Trends of up to 8% per decade are observed in the ERA-Interim dataset for July and April at several areas in Eastern Europe, South Eastern Siberia, and the Russian Far East. Spatial patterns of the CCM and ERA-Interim results agree qualitatively but trends calculated with the CCM model are generally smaller.”

Caption Figure 14: Please adjust, similar to my suggestion for Caption 13.

L565 - L567 (“We also see … from this dataset.”)

Delete this paragraph as a result of my request to remove OMI/TOMS satellite data from Section 3.2.

L572 (“skin type 4 is characterized by minimum sun-sensitivity.)

No. That would be skin type VI, which of course is not very prevalent in Russia. Still, it is incorrect to say that skin type IV is the one with minimum sun sensitivity.

Figure 15: In Section 2, only two thresholds were defined, one between UV deficiency and UV optimum and one between UV optimum and UV excess. Thresholds for “high” and “very high” UV excess were not defined. Either define these thresholds or remove these categories from the figures.

L594-L596 (“The application of the Met Office … Northern Siberia.”)

This sentence should be removed if Figure 6 and the associated text is removed following my comment above.

L640: Delete “However, large variations in the results are observed in seasonal patterns due to the application of different SST/SIC datasets.” if Figure 6 is removed.

L647: “In the central Arctic region, on contrary, negative Eery trend was” > “In the central Arctic region, negative Eery trends were”

L650: Delete “pronounced”. (A few percent is not “pronounced”.)

L653 - L655: “However, in summer in the central Arctic region, on contrary, due to pronounced CMFuv decrease, there is a change from UV excess to UV optimum conditions for the skin types 2 and 3.” > “However, during summer months in the central Arctic region, there is a change from UV excess to UV optimum conditions for the skin types II and III due to a decrease in CMFuv.”

L656: “the on-going” > “an on-going”

L657: “into the account for aerosol and reflectivity temporal changes as” > “into the account temporal changes in aerosol and reflectivity as”

 

***Technical corrections / language

The final manuscript should be read by a copy editor or person with good English skills. While I tried to point out most language issues, many errors remain with respect to grammar, punctuation, and the use of the articles “the” and “a”, which are often omitted when required and vice versa.

The authors frequently refer to the region of eastern Russia as the “Far East”. However, the “Far East” also includes South East Asia, see: https://en.wikipedia.org/wiki/Far_East. The term “Far East” should be changed to “the Russian Far East” throughout (see:

https://en.wikipedia.org/wiki/Russian_Far_East)

 

The authors frequently use “string of nouns” This is bad writing style. I provided suggestions for improvement, but my list is not complete and the authors should scrutinize their text for such phrases. Examples include:

the Eery trends” > trends in Eery (L54)

stratospheric ozone decrease > decrease in stratospheric ozone (L46)

of UV temporal variations > temporal variations in UV radiation (L69)

interpret the UV long-term changes > interpret long-term changes in UV radiation (L103)

the Eery temporal variability > temporal variability in Eery (L112)

pronounced Eery trend > pronounced trend in Eery (L635)

 

L45: within > over

L50: “However, not over all regions there is a direct relationship between total ozone column (X) and erythemal irradiance…” > “However, there is no clear relationship between total ozone column (X) and erythemal irradiance for many regions…”

L70: “According to these publications over many European sites an increase of UV irradiance has been revealed since the middle of 20 century mainly” > “According to these publications, an increase of UV irradiance has been revealed over many European sites since the middle of 20 century, mainly…”

L72: i.e. > e.g. (see: https://www.grammarly.com/blog/know-your-latin-i-e-vs-e-g/)

L72: “the Eery increase to some extent can be also attributed” > “increases in Ery can be attributed to some extent to”

L73: “Arctic area” is too general. Refer to the region this paper is dealing with.

L83: “(CESM1) Whole Atmosphere” > “(CESM1) and the Whole Atmosphere”

L89: “are also applied for estimating the changes” > “have also been used for estimating changes”

L90: “made using the clear-sky data from the first phase of the Chemistry-Climate Model Initiative and TUV” > “made by using the clear-sky data from the first phase of the Chemistry-Climate Model Initiative (CCMI) as input to the TUV”

L92: “analysis has revealed an average Eery increase” > “analysis projects [or predicts] an increase in average Eery” (one can only “reveal” something that has occurred in the past)

L94: “are opposite” > “partly contradict”

L95: “The projected Eery increase in [37] is observed due to the assumption of the large decrease in atmospheric aerosol loading by the end of 21 century, which is debatable.“ > “The projected increase in Eery reported in [37] results from the assumption that the atmospheric aerosol loading will decrease greatly over the course of the 21^st century, which is debatable.“

L97: “has revealed the Eery decrease” > “projects Eery to decrease”

L108: “with additional account for skin types and open body fraction. Similar approach has been” > “and takes into account the skin type and the fraction of the skin surface that is exposed to UV radiation. A similar approach has been”

L113: “cloudiness during the historical period by the application of the” > “cloudiness between 1979 and 2015 based on”

L114: “and chemical –climate model over” > “and a chemistry–climate model over”

L115: delete “during the 1979-2015 period.” (now part of L113)

L118: “Re-analysis” > “reanalysis”

L119: “applied the model runs of Russian Chemistry” > “applied model runs of the Russian Chemistry”

L120: “developed by the joint efforts of Institute” > “ which was jointly developed by the Institute”

L124: “the UV retrievals from” > “UV data”

L127: efficient > appropriate (The word "efficient" does not quite work, "efficient" means "can be done with little or no waste in time or materials)"

L129: “its efficiency and a good agreement with” > “good agreement between results obtained with this method and the observed ...”

L131: i-year > year i

L132: “respectively, Wj(he) – is the weighting” > “respectively. Wj(he) is a weighting”

L139: “(X)) and” > “(X) and”

L144: “the RAF dependence on effective solar elevation for Eery” > “that the RAF for erythemal irradiance depends on solar elevation. We used the following equation to parameterize this relationship:”

L148: Start line with “,where the index 2000 indicates the year 2000.”

L159: “in other conditions we should use the correction.” > a correction should be used in other conditions”

L163: “To accounting for” > “To account for” and “we used the same method applied to” > “we applied the same method to”

L165: “the changes” > “changes”

L168: “we evaluated the year-to-year relative Eery anomalies relative to 2000 with account for only total ozone, for only cloud modification factor,” > “we evaluated anomalies in Eery relative to the year 2000 taking into account either only total ozone or only the cloud modification factor,”

L172: “applied t-test” > “applied a t-test”

L174: “the trend analysis for Eery anomalies using linear “ > “a trend analysis for Eery anomalies using a linear”

L180: “and photochemical module developed at the RSHU (Russian State Hydrometeorological University).” > “and a photochemical module developed at the Russian State Hydrometeorological University (RSHU).” (The names of the other two institutions are mentioned before their acronym. This sequence should also be applied to the RSHU).

L183: I don’t understand “with 39σ levels equal the difference between the layer bottom and model top pressure divided by the difference between the ground surface and model top pressures up to 0.003 hPa.” Please reword. For example, are the pressure layers equally spaced on a logarithmic axis? The sentence could be removed if the definition of the layers is described in a publication (e.g., 43 & 44).

L185: “allows to account for” > “accounts for”

L186: “Chemical block” > “The chemical module”

L188: “The detailed” > “A detailed”

L190: “for accounting the greenhouse gases emissions we applied representative concentration pathways (RCP4.5) emissions scenario” > “the change in greenhouse gas emissions was parameterized according to the RCP4.5 scenario” (RCP was already define before - no need to do it again)

L191: “solar fluxes variability” > “solar flux variability”

L196: Change to “For evaluating CMFuv, a spectral correction was applied to the CMF data according to Eq. (6).”

L204: “For CMF evaluating” > “For calculating CMF”

Eq. (7): Move comma in denominator

L211: “ground-based independent ozone” > “independent ground-based ozone”

L214: “total ozone content” > “total ozone measurements”

L216: “ozone retrievals using”   “total ozone data from”

L219: “lies typically within ±5” > “is typically ±5”

L222: “The validation of the ERA-Interim total solar irradiance at ground has been made in several papers [61, 62, 63] against ground-based” > “ERA-Interim total solar irradiances at the ground have been validated with ground-based ...”

L223: “In [61] according to the comparisons with 674 sites a high correlation was obtained between the Era-Interim and ground-based datasets (R2=0.97) with the smallest deviations from observations, compared with the other re-analysis data.” > “According to a comparison of data from 674 sites, Era-Interim and ground-based measurements were highly correlated (R2=0.97) [61]. Biases relative to observations were smaller compared to other reanalysis data evaluated in this study.”

L227: “RMSE (Root-Mean-Square Error) - about 27.7 Wm-2,” > “the RMSE (Root-Mean-Square Error) was 27.7 Wm-2,” (don’t use “about” if you specify a value at high precision!)

L228: ”was about 4.15 Wm- 2 and RMSE – about 19.6 Wm-2. Similar small bias”  > ”was 4.15 Wm- 2 and the RMSE was 19.6 Wm-2. A similarly small bias”

L238: “Level 3 data were available at” > either “Level 3 data are available at” or “Level 3 data were downloaded from”

L238: “with 1x1.25° grid” > “with a spatial resolution of 1° x 1.25°.”

L239: “Delete “and the” and start new sentence with  “Level 3…”

L239: “Daily product archives were available at” > “Daily product archives are available at”

L242: “a new aerosol” > “the new aerosol” (as this refers to Macv2)

L253: “first we” > “we first”

L255: “both previous” > “both the previous”

L256: “take into account for change aerosol in time.” > “take into account changes in aerosols  over time.”

L261: decreasing > smaller

L262: “up to” > “of up to”

L274: “has been applied [69].” > “described in [69] has been applied.”

L277: “in this study we mark it as TOMS/OMI ozone dataset.” > “we refer to it in this study as the TOMS/OMI ozone dataset”

L282: “ERA-Interim. We” > “ERA-Interim datasets. We”

L294: “of the two” > “of two”

L312: “for total ozone” > “total ozone”

L323: “reproduce well both ozone climatology and its temporal variations.” > “calculate both the ozone climatology and its temporal variation with low uncertainty.”

L332: Western > the Western

L350: “cloud modification factor over Europe is” > “cloud modification factors over Europe are”

L354: “in agreement” > “in reasonably good agreement”

L377 “the ozone depleting” > “ozone depleting”

L387: “2000-2015 (21 century) and 1979-1999 (20 century) due” > “2000-2015 and 1979-1999 due” (Delete 21 century and 20 century. These periods span only 15-20 years, not a century!)

L401: “the strong volcanic eruptions in 20 century” > “strong volcanic eruptions at the end of the 20th century”

L402: “in the active ozone loss initiated by the heterogeneous” > “in ozone loss initiated by heterogeneous”

L416: “with account of the same data on the ODS, stratospheric aerosol and solar activity, but with different SST/SIC (MetOffice, ERA-Interim, and SOCOL) datasets.” > “using the same datasets for ODS, stratospheric aerosol and solar activity as before, but the MetOffice SST/SIC dataset was replaced with ERA-Interim and SOCOL data.”

L423: “There are larger differences in seasonal changes:” > “Differences between the three model runs are larger for seasonal changes:”

L438: “the statistically” > statistically

L451: “to ozone reduction in 2000-2015.” > “due to lower ozone in the 2000-2015 period compared to the 1979-1999 period.”

L490: “over Atlantic and Far East,” > “over the Atlantic Ocean and the Russian Far East,”

L491: “over Northern Arctic region” > “over the Northern Arctic region”

L495: “with the other model” > “with independent model”

L511: “-2-4 %/decade” > “2-4 %/decade” (the text already says that these trends are negative)

L513: “that model does” < “that the model does”

L514: “regions - in” > “regions in”

L559: “chosen according to the following” > “chosen for the following”

L579: “on contrary” > “on the contrary”

L593: “which increase led” > “the increase of which led”

L602: “of about -2-4 %” > “of about 2-4%”

L614 “over the large areas” > “over large areas”

L615: “For the first and the second, most susceptible to sunburn, skin types” > “For skin types I and II, which are most susceptible to sunburn,”

L616: “substituted by the UV excess over large areas.” > “substituted by UV excess conditions over large areas.”

L619: Delete :” on contrary,”

L621: “We should note that in our assessment of UV resources we do not take into account for changes in open body fraction S during” > “In our assessment of UV resources, we did not take into account changes in the surface area of the skin that is exposed to sunlight or possible changes in the behavior of people, which could result from climate change and increasing temperatures. Instead, we focused …”

L637: “have confirmed the largest impact of” > “have confirmed that the largest effect on ozone and Eery stem from”

L638: “substances and some effects of volcanic aerosol” > “substances. Additional factors include volcanic aerosol”

Author Response

Please, find our response in the attached file

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

I just provide a few minor comments/corrections as attached. 

Comments for author File: Comments.pdf

Author Response

The answers to the comments:

Mirror comments/corrections:
1. In the abstract, line 28-
sentence to make it read better. The abstract is very concise now and clearly conveys the
message.

Thank you. This sentence has been changed:

"When defining a “UV optimum” conditions with the best balance in Eery for human health, the observed increases in Eery led to a noticeable reduction of the area with UV optimum for skin types 1 and 2, especially in April. "

2. Line 152: We used the according to [47] the following equation
We used the following equation according to [47]

Done. Thanks!

"We used the following equation according to [47] to parameterize this relationship: "

3.Line 221: What is WOUDC?

We have inserted the following clarification in the text:

"...from the World Ozone and Ultraviolet Radiation Data Centre (WOUDC) archive .."

4. Line 616: the effects in to effects on

Done. Thank you!

"In this paper we evaluated the separate effects of total ozone and cloudiness and their combined effect on temporal variability of UV erythemal daily doses and UV resources according to the Eery retrievals from the ERA-Interim, the INM-RSHU CCM datasets, and TOMS/OMI satellite measurements with the Macv2 aerosol correction over Northern Eurasia for the 1979-2015 period."

5. Line 626: should be fulfilled to should be conducted

Done.

"The application of the Met Office and the SOCOL SST datasets in the INM-RSHU CCM provided the similar annual Eery increase in 2000-2015 over Central part of Northern Siberia, however, further studies should be conducted for understanding the physical mechanism of this phenomenon. "

6. Line 626: In overall to overall; model Eery to modeled Eery

In the first case we used "on the whole", and changed to "modeled Eery" in the second case:

"On the whole, there is a positive change in the modeled Eery of up to 1-2% at the northern regions of Eurasia in 2000-2015 compared with the 1979-1999 period. Maximum positive linear Eery trends (up to 3% per decade) due to ozone loss were observed in April."

7. Line 689: Fulfil to perfom 

Done. 
 
 "This work is a part of an on-going project, within which we plan to perform additional numerical experiments with the INM-RSHU- CCM taking into the account temporal changes in aerosol and surface reflectivity as well as aerosol-cloud interaction, which possibly help in evaluating real CMFuv long-term changes over 1979-2015 period. "
 
 

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Review of Chubarova et al. [2019]

This study examines trends in Qery in Northern Eurasia due to ozone and cloud using reanalysis data, modeling outputs, and satellite retrievals. The authors report a significant trend (3% per decade) in Query data driven by ozone loss. I suggest a major revision for the current manuscript based on my concerns below.

In my view, the manuscript is not well written, and sometimes even awkward and boring to read. The authors may want to improve quality of the manuscript. The paper reads like a redundant report rather than an insightful paper. I may suggest the authors deliver the main points and offer more scientific insights in their revised manuscript.

Is there any interaction terms between ozone and cloudiness? The authors may want to reexamine this part in Section 2.

Satellite retrievals may suffer from high uncertainties at high latitudes due to longer light path. The authors may want to discuss more on this, and probably with more comparison between satellite ozone retrievals with model/ reanalysis data.

Minor comments:

The authors may want to consider breaking a long sentence into parts using a comma. Otherwise, it really takes a long breath to finish reading. For example, L52, you can write it as "Using OMI retrievals, Qery changes..."

About citing a reference, it’s awkward to just use numbers. For example, in L50, it would be great to write it as “In Herman 2015 [16]” instead of “In [16]”.

L33, please remove “does”, change “cause” to “causes”

L34, “does have”, change it to “has”

L35, should be UV “levels”

L36, delete “it was”

L62, “drivers”

L65, remove “both”

L88, remove “how”

L90, what is the “skin type?”

There are too many of this kind of little things. I just pick up some of them from Page 1-2.

Author Response

Please, find the responses and an updated text with the revised parts shown in yellow color in the attached file. 

Author Response File: Author Response.docx

Reviewer 2 Report

Please see the attached comments and thanks. 

Comments for author File: Comments.pdf

Author Response

Please, find attached our response together with the updated text where the revised parts shown in yellow color. 

Author Response File: Author Response.docx

Reviewer 3 Report

General comments

The manuscript by Chubarova et al. discusses trends in erythemally (sunburning) UV irradiance (QEry) over Northern Eurasia and the adjacent Arctic ocean. The subject is appropriate for the journal Atmosphere, however, the paper has substantial shortcomings, which need to be addressed before publication can be considered.

Most importantly, trends in QEry derived from results of the CCM model, the ERA-Interim dataset, and from satellite data shown in Figures 12 and 13 are very different, both in magnitude and pattern. The authors assert that trends derived from the ERA-Interim “are the most reliable”, but provide no evidence that this is indeed the case. In fact, uncertainties are not specified anywhere in the manuscript. So the reader has no indication whether any trends calculated in the manuscript are realistic. (Of note, in line 432, the authors refer to Section 2.2.1 for a discussion on the reliability of ERA-Interim trend estimates, however, there is no Section 2.2.1, and I could not find suitable information elsewhere in the manuscript.)  Without information on the accuracy of the ERA-Interim dataset (e.g., uncertainty estimates, validation of this dataset against other sources such as ground-based measurements), trends derived from this dataset are not  useful and conclusions derived from these trends (e.g., the change in UV resources discussed in Section 3.3) are without basis. The fact that the reader is not given any clues whether trend estimates are realistic calls into question the purpose of the manuscript. If the authors consider to submit a revised version of the manuscript, it is therefore of utmost importance that they discuss the accuracy of the ERA-Interim dataset and the uncertainty of trends derived from it.

Another important shortcoming of the manuscript is that the definition of the central quantity of the manuscript, namely “Qery”, is ambiguous. Is it the erythemal irradiance at noontime, at the time of the satellite overpass, or a general measure (as Eq. 10 suggests), which varies during the course of the day? More specifically, are trend estimates based on spectral irradiance at noontime, spectral irradiance averaged over 24 hours, or something else? Is the trend analysis for the three data products (ERA-INTERIM reanalysis, TOMS/OMI, INM-RSHU CCM) based on the same quantity of erythemal irradiance? For example, the Radiation Amplification Factor RAF defined in Eq. (3) depends on solar elevation and if one data product were to refer to noontime erythemal irradiance and one to a daily average, this would result in inconsistent trend estimates.

There are many other issues that need to be addressed as described in my “specific comments” below. Critical issues are indicated by the word “(critical)” before the line number.

 

Specific comments

 

L14: the symbol “Q” for irradiance is unusual. Usually the symbol “E” is used. The authors should consider to use E (or more specifically E_ery) to be consistent with the vast bulk of the literature.

L14: The authors should explain the term “UV resources”. I have seen this term used by the authors in their previous publications, however, to the best of my knowledge, nobody else uses this term. A definition (e.g., provided in the introduction) is therefore necessary. As an alternative, the term “UV resources” should be replaced with a more descriptive term, e.g., “UV categories for optimum health”. To avoid repeating this term, it could be defined once and then a acronym (e.g., UV_opt) could be used.

(critical) L17-19: The abstract is very confusing and should be rewritten. According to line 17-18, trends in clear-sky Qery calculated from the three datasets (ERA-INTERIM reanalysis, TOMS/OMI satellite measurements, and INM-RSHU CCM) are consistent and range between 2.5 and 3% per decade. However, I do not understand the trend estimates provided for the all-sky case (i.e. data derived during clear-sky and cloudy conditions). For example, the clause “were significantly enhanced (up to +6-9% per decade) over Eastern Europe, Siberia and Far East due to joint ozone and cloud decrease” suggests that trends calculated for the all sky case are much larger. However, the next sentence (“much lower cloud Qery trends were evaluated there according to the INM-20 RSHU CCM”) suggest that these large trends were only calculated from the ERA-INTERIM reanalysis and TOMS/OMI data sets but not for the INM-RSHU CCM dataset. The next sentence  (“Negative Qery changes were observed from measurements and modelling due to cloudiness in summer over Arctic region”) greatly adds to the confusion as large positive trends suddenly turn in to negative trends. After reading the abstract several times, I guessed that the positive trends apply to  Eurasia (specifically Eastern Europe, Siberia and the Far East of Russia) and the negative trends to the Arctic north of Russia.

I suggest to change the abstract to the following, however, my suggested text may need revision because my interpretation of the existing abstract may not be correct and changes resulting from revising the manuscript would also affect the abstract:

“Temporal variability in erythemal irradiance (Eery) over Northern Eurasia (Eastern Europe, Siberia, and the Russian Far East) was assessed using retrievals from ERA-INTERIM reanalysis, TOMS/OMI satellite measurements, and the INM-RSHU chemistry-climate model (CCM). The study is based on the period 1979 – 2015 and analyzes the effects of changes in total ozone column (TOC) and cloudiness on Eery. For clear sky conditions during spring and summer, consistent trends in Eery of up to 3% per decade were calculated from the three datasets and attributed to decreases in TOC. For all-sky conditions (i.e., clear and cloudy days), trends in Eery calculated from the ERA-INTERIM and TOMS/OMI data ranged between 6 and 9% and resulted from a combination of decreases in TOC and decreases in cloudiness. In contrast, all-sky trends in Eery calculated from the CCM results were only x% per decade [specify x!]. While trends for Northern Eurasia were generally positive, negative trends were calculated for the Arctic, north of Eurasia, due to an increase in cloudiness during summer months. Model experiments suggest that anthropogenic emissions of ozone-depleting substances are the largest contributor to trends in Eery. In addition, volcanic aerosols and changes in sea surface temperature (which affect evaporation and cloud formation) [please check whether this is true!], have contributed to Eery trends. Finally, changes in the availability of UV radiation for optimal human health was assessed. When defining a “UV optimum” as the UV intensity that best balances adverse (e.g., sunburn) and beneficial (e.g., vitamin D production) effects, we determine that the observed increases in Eery over Northern Eurasia led to a noticeable reduction of the area with UV optimum for skin types 1 and 2, especially in April. However, in some Arctic regions, decreases in Eery in July resulted in a change from “UV excess” to “UV optimum” conditions for skin types 2 and 3.”

L35: “Ozone and cloudiness…” > “Apart from the solar elevation, ozone and cloudiness…”

L40: As far as I know, the quoted increase in ozone of about 0.3 to 1.2% is not yet statistically significant. Please double-check with Ref. [8]!

L50: “pronounced” is exaggerated and misleading. It would be better to quote a range of percentages.

L53: Please quantify trends discussed in Ref. [17]. Furthermore, change “In [18, 19] it was shown a significant change in UV reflectivity…” to “A significant change in UV reflectivity due to clouds over the 1979-2011 period was derived from different satellite measurements [18,19].”

L60: Please provide a range of years instead of saying “during the last years”. For example, during the last 10 years there was not much of a change, except at some sites where increases in UV radiation due to reduction in air pollution were observed (e.g., Thessaloniki).

L65: I note that ref [32] is now almost 10 years old.

L76: [9, 35, 36] > [9, 11, 35, 36]

L80: It should be noted that increases in Qery projected by ref [35] (which contradicts ref [33]) is mostly due to the fact that ref [35] assumes large decrease in atmospheric aerosol loading, which are based on results of three of the CCM models that participated in the CCMI-1 initiative. Whether or not these large decreases are realistic is debatable. In addition, decreases in aerosols are spatially inhomogeneous, for example, will likely be  much larger in China than in other northern hemisphere regions with similar latitude. For the “fixed aerosol” case discussed in ref [35], Qery is projected to decrease, more or less consistent with the results of ref [33].

L89: As mentioned above, “UV resources” is not  a good term even though it has been used in ref. [14]. I suggest “UV categories for optimum health”, but leave it to the author to find a more appropriate term.

(Critical) Section 2 should be expanded to better explain the math used to calculate trends. Also in section 2, the word “changes” is misleading. V_i, v1_i,j, and v2_i,j express anomalies relative to a mean.

(Critical) Section 2 describes how anomalies V_i are calculated, however, the paper is about trends and the section does not describe how trends (and their uncertainties) are calculated from these anomalies. This is a serious omission and information on trend estimates needs to be added. Were trends calculated via a simple regression or was a more sophisticated statistical model used? Was autocorrelation taken in to account?

L109. The quantities v1 and v2 should be better explained and they should be introduced as v1_ij and v2_ij. Are these quantities anomalies relative to the mean of all values (irrespective of month) or are these anomalies relative to the value in the year 2000 (as perhaps suggested in line 137? Also, the expression “X (v1)” is confusing. It would be better to delete “X” and explain “X” somewhere else. As it is written, it looks as if X is a function depending on the variable v1 while the opposite is the case.

Eq (1): It seems to me (also suggested by Eq. (4)) that “X” and “cloud” are monthly averages, resulting in discrete values depending year and month. This should be clarified and “X” and “cloud” should be written as “X_ij” and “cloud_ij”. I also suggest to introduce a symbol for “cloud” (perhaps “c”) instead of using the word “cloud” in a formula.

(critical) L113. The weighting function W(h) needs to be explained. According to Eq(1), the relative anomaly in UV radiation for a given year i is a weighted sum of the monthly UV anomalies, which in turn depend on ozone and cloud variability. W(h) could be designed to we weighted towards the summer, the winter, or could treat all months equally. The weighting that was actually used to calculate annual trends should be explained and the reason for this particular weighting should be given.

Eq. (3): It should be noted that the RAF is also a function of total ozone and not just solar elevation. (I plotted the function and it seems that the values (and their dependence on solar elevation) roughly correspond to a RAF calculated for a ozone column of about 320 DU.)

Eq. (4): I think it would be better to change “X” to “X_i,j” on the left side of the equation.

Eq. (6): I did not check whether this parameterization is reasonable. Has this parameterization been used in a previous publication that could be cited? Is the right part of the equation, which is written in very small font, an exponent? If so, the font size should be increased for better readability.

L137: The mean value of the Vi calculated with the approach described on Page 3 is unlikely equal to zero for the year 2000. Does “deviations of Qery against 2000” mean that the Vi values were renormalized to so that the anomaly for year 2000 becomes zero? Also, instead of saying “against 2000” it would be better to say “relative to the year 2000”.

L181: Please specify the source of the TOMS and OMI datasets.

L188: In addition to ref [55], the following ref should be cited (Arola et al., 2005):

https://agupubs.onlinelibrary.wiley.com/doi/pdf/10.1029/2009GL041137

Also, I believe that the factor of 3 in Eq. (9) was empirically determined by comparing OMI UV estimates with ground based measurements using the previous climatology of absorbing aerosols. If a new aerosol climatology is being used (i.e., ref [39]), the factor of 3 may no longer be appropriate. I realize that the question of how to correct for absorbing aerosols is beyond the scope of the manuscript, but it should at least be mentioned that there is a large uncertainty in the aerosol corrections, regardless of the aerosol climatology that is being used.

L 202: Please clarify what “higher CFs” means. I presume that CFs are always smaller than one. So “higher CFs” would means that CFs calculated with the MACv2 climatology are closer to one. Please confirm.

Section 2.3.: Please state somewhere in this section that the MACv2 aerosol climatology does not depend on time. (If my assessment is incorrect, and the MACv2 data do in fact depend on time, this would be in conflict with line 147, which states that “we do not take into account changes in aerosol properties).

Section 2.4.: Again, I hope that a better word can be found for “UV resources” and a good definition can be provided!

(critical) Eq. (10): I don’t understand this equation and how the threshold is determined from this. I would expect that the threshold is reached if the integral of Qery over one day is larger than the minimum erythemal dose for a given skin type j. However, that’s not what the equation says. In addition, delete extra equal sign and change “j” (the indicator of the skin type) to a different symbol because “j” was already used to indicate a month.

(critical) Eq. (11) Similar to my problem understanding Eq. (10), it is not clear to me how Eq. (11) defines a threshold. I would think that a threshold for vitamin D production is defined by the desired daily amount of vitamin D (expressed in IUs) that is considered adequate. For example, some studies recommend a daily intake of 1000 IUs (either from supplements or cutaneous synthesized from solar radiation), but there is not a universally accepted value to the best of my knowledge, and recommended daily doses may vary between 400 and 4000 IU, depending on source. The authors should specify the amount of vitamin D (expressed in IUs) assumed for their threshold and modify Eq. (11) to clearly define the threshold.

L252: Please mention the satellites that were used for the data shown in the last column of Figure 4.

L249 - 271: Figure 4 shows the cloud climatology for January, April, July, and October. When referring to the figure in the text, please also use month, and not spring, summer, autumn, etc.

L274 & L281: The word “dramatically” is too strong. Trends in QEry are in the range of a few percent per decade (e.g., Fig. 7). These changes are hardly “dramatic”.

Fig. 5 and L321-323: I am surprised that the “all factors together” panel only shows changes in the -1 to 2% range (which, by the way, are not “dramatic”). The anthropogenic effect is about 3%, the stratospheric aerosol effect is about -1%, solar activity contributes +1% and SST +1%. So the “all factor” effect should be about 3% – 1% + 1%  +1% = 4%, which is considerably higher than shown in the “all factors together” panel. Why?

L304: Please clarify what was actually done. It seems that SST/SIC are provided independently by ERA-Interim, SOCOL and the “Met Office” and these data are then ingested by the INH-RSHU CCM model to calculate changes in ozone, and in turn changes in Qery. However, this is not entirely clear.

L312: during summer and fall > during July and October

L326, L350, L361. I can’t discern hatching in Fig. 7, 8, and 9.

(critical) L332 - L357 and Figures 8 and 9: It seems to me that Figures 8 and 9 (plus their captions) were swapped. The text in lines 332 - 347 refers to difference in Qery between 2000 - 2015 and 1979 - 1999 while Figure 8 shows trends in percent per decade. On the other hand, the text in lines 351-357 discusses seasonal and annual trends in QEry while Figure 9 shows seasonal and annual difference in Qery. (Of note, the caption of Figure 8 seems to be correct as the bottom left panel of Figure 8 looks like the bottom panel of Figure7, which also shows a decadal trend.) 

L340: The statement “there is still negative ozone trend of about 3% over 30-60N area against 1960-1980” does not make much sense because a trend refers to a change over a given period, not a difference between the present and a period past (in this case 1960-1980).

(critical) L342-343: Are you suggesting that the ERA-Interim dataset is more accurate than the actual satellite measurements? What is the evidence? Of note, the latest Scientific Assessment of Ozone Depletion (ref. [8]) does not use ERA-Interim data for trend estimates. Instead, ground based data and various merged satellite data (which were validated against ground data) were used (see Appendix 3A of ref [8]).

L344-347: I would say: “Difference of the three datasets for January are of little consequence because QEry is very low for these months. For April, the three datasets agree reasonably well. However, in July and October, difference between the three datasets are large as the model predicts more rapid ozone recovery (resulting in negative Qery trends) than observed.” Again, this statement may need to be modified as it likely refers to Figure 9, not Figure 8.

L351 - 357. Again, don’t use seasons in the text when you show months in the figures.

L343-347: spring> April, summer> July, etc.

(critical) L365: I have problems seeing a “a significant increase in  CMFuv of more than 5-10% according to the ERA-Interim over the large areas of Northern Eurasia”. Instead, I see a patchwork of increases and decreases, generally ranging between -5% and +5% with a few areas in the 5-10% range. Increases of more than 10% are limited to two rather small areas in July. I am also hard-pressed to see a significant decrease over the northern Arctic. The text should be improved to match Fig. 10.

(critical) L372: I see little resembles between the patterns shown in the left and right column of Figure 10. While the monthly panels are very noisy, the “yearly” panel (last row of Figure 10) shows that CMFuv values have declined according to the INM-RSHU model but have increased according to the ERA-Interim analysis. So my assessment is that the model does not reproduce the changes of the ERA-Interim analysis.

L385: I can’t see crosses, perhaps because the quality of the figure precludes this.

L386- 392 and Figure 11: For me, the take home message is that the INM-RSHU model is not able to reproduce the actual trends in Qery due to changes in cloudiness. This should be emphasized.

(critical) L399: The satellite datasets (right column) shown in Figures 12 and 13 also include cloud effects according to text and figure captions. However, the text does not explain how the effect of clouds was quantified. While Figure 4 shows a climatology of cloud modification factors also for the satellite dataset, Figures 12 and 13 require information how CMF_UV changed over time. Where is this information? According to the caption of Figure 4, the CMF_UV climatology is based on data from the years 1979-2002, so this dataset would not be suitable for quantifying cloud changes over the period of interest (i.e., 1979 - 2015).

(critical) L402: I can’t discern a “a quite perfect agreement between satellite and ERA-Interim datasets for July”. While there is some resembles between the two panels in questions, this is far from “quite perfect”.  More importantly, I can’t discern any resemblance between the three datasets for the other months. Even the yearly data (last row) are quite different, with the satellite data showing significant increases of up to ~15% while the CCM model and ERA-Interim results show increases of less than 5%. Considering the rather poor agreement, is it possible that the wrong panels were copied into Figure 12?

(critical) Figures 12 and 13: Why are the satellite data only shown for latitudes between 40 and 65 degrees for April, July and October? Ozone data (Fig. 3) and cloud data (Fig. 4) are available for latitudes between 40 and 80 degrees, so I don’t understand why the panels showing the effects of both ozone and clouds are restricted in latitude.

(critical) L415-423, Figure 13: Also here the resemblance of the datasets for the model, ERA-Interim and satellite is generally poor, suggesting that no firm conclusions can be drawn.

L425: The figure caption is incorrect. The figure shows decadal trends not “Seasonal and annual differences in Qery between the 2000-2015 and 1979-1999 periods”.

(critical) L432: The results shown in Figures 12 and 13 obtained with the CCM model, ERA-Interim and satellite data are greatly different. I therefore have little confident that the results calculated with ERA-interim are “reliable”. Furthermore, there is no “Section 2.2.1” and I couldn’t find a discussion on the “reliability” of the ERA-interim dataset anywhere else in the manuscript.

Considering the large uncertainty of the QEry trends calculated in the manuscript I think it is not appropriate to evaluate “changes in UV resources”.

(critical) Section 3.3. Based on my last point, I suggest to either remove Section 3.3 or provide evidence that the ERA-Interim trends are indeed realistic. The assessment of changes in “UV resources” hinges on the correctness of UV trends according to the ERA-interim analysis. In addition to those uncertainties, climate change will change people’s behavior, for example, increasing temperatures at a cold location will likely result in more time spent outdoors with no UV protection. So in addition to changes in QEry exposure times will change, making it even harder to assess adverse and beneficial effects.

Figure 14: The panels A-D presumably refer to skin types 1-4, however, this not stated anywhere.

L466: “pronounced and” > “pronounced than those caused by ozone changes and”

L480 - 488: Again, considering the uncertainty of the trend estimates, I feel that discussions on changes in UV resources are not appropriate.

 

General technical comments

All acronyms used in the manuscript should be explained. This also applies to acronyms such as CESM1, WACCM, CMIP, TUV, SOCOL, NAO, AO, AAO, etc.

It should be avoided to use words in mathematical expressions such as “Qery_no correct” (to give an example; there are numerous other awkward constructs). Instead, one-letter symbols should be defined and used in equations.

 

Specific technical comments, language

L33: UV high doses > High UV doses

L38: “the erythemal irradiance (Qery) increase” > “increase in erythemal irradiance (Eery)”

L39: “of 1990s due …” > “of the 1990s” and move “due to the reduction of the ODSs in the stratosphere” to the end of the sentence

L42: aerosol > aerosols

L47: “data different” > “data, different”

L59. Delete “etc.”

L69: “future global” > “future, global”

L78: “They show” > “The study projects”

L87: to interpreting > to interpret

L95: If “this study” refers to the manuscript at hand, it should be “this study is” instead of “this study was”.

L98 “the whole Northern Eurasia” > “the whole of Northern Eurasia” or just “Northern Eurasia”.

L117: where C is the constant > where C is a constant

L141: “2000 year” > “the year 2000”

L147: “do not take into account for the changes in” > “do not take into account changes in”

L156: “5°LON × 4°LAT” > 5° x  4° (longitude x latitude)  What are “sigma” levels?

L237: “reproduce well” to “adequately reproduce the”

L248: ozone content > ozone column

L279: fulfilled > conducted or performed

L280: of main ozone drivers> of the main ozone drivers

L286: joined > combined

L288: to the other model > to other model

L314: at the South > in the South

L332: with accounting > that account

L455: “affecting Qery is the ODS, which increase led to the significant ozone loss.” > “affecting Qery is the concentration of ODS, the increase of which led to the significant ozone loss.”

L439 & L484: Instead of 3d and 4^th skin type, say skin type 3 and 4. (By the way, the abbreviation for “third” is 3^rd, not 3d.)

Author Response

Please, find our response in the attachment . It also contains the an updated text with the revised parts shown in yellow color.

Author Response File: Author Response.docx