|Re-Review of “A quantitative analysis of the reactions involved in stratospheric polar ozone depletion” by Wohltmann et al.|
The authors have done an excellent job of revising their manuscript in response to referee comments. In my opinion, most of the issues raised by the reviewers have been adequately addressed. In particular, much more information has been provided on how generally applicable the results shown in the main text are for other winters and at other pressure levels. Discussion of the serious shortcoming of the ATLAS model (and other state-of-the-art models) in reproducing the observed behavior of HCl has also been substantially expanded and moved into the body of the paper, which is appropriate. A pervasive insufficiency in referencing existing literature has largely been corrected. Numerous other changes have been made throughout the text and figures in response to specific comments. I must admit that at first I was dismayed to discover that the revised manuscript included a 184-page supplement containing 172 additional figures, but as I read through the paper I came to agree that such a lengthy supplement is very valuable. I have only a few remaining comments (most prompted by new material / discussion added during revision) that I would like the authors to consider before the paper is published.
* Supplement: As noted above, I believe that the supplement is a very valuable resource and all of the references to it in the main text are useful. However, I feel strongly that simply stating “see the supplement” with no pointers to specific figures therein is not adequate. In my opinion it is the authors, not readers, who should shoulder the burden of tracking down which figures are relevant for a particular point (especially since often as many as four figures that are widely dispersed throughout the supplement – for two different years in both hemispheres – are needed to fully make a point). Thus I recommend that specific figure information be added in almost all places throughout the body of the paper where the supplement is referred to.
* Interhemispheric differences in peak ClOx values: The simulations in this manuscript show that ClO (ClOx) reaches maximum abundances of about 2.0 (2.5) ppb in the NH but only about 1.3 (2.0) ppb in the SH at 54 hPa; comparisons between the hemispheres are similar at other altitudes. This is a very interesting result, but one that I have trouble reconciling with MLS measurements. The paper by Santee et al.  that is now referenced in the revised manuscript shows larger peak Aura MLS ClO values (as well as enhanced values over a greater vertical range) in the SH than in the NH. Similarly, Santee et al. [JGR 2003] found, based on UARS MLS measurements, that ClO enhancement within the Arctic vortex is slightly smaller in magnitude and spatial extent than that in the Antarctic at 465 K but considerably smaller (by 0.5-1.0 ppbv) at 520 and 585 K, where the spatial extent of activated chlorine is also significantly greater in the Antarctic. Thus the ATLAS results and MLS (both UARS and Aura) measurements are in conflict about which hemisphere experiences greater ClO enhancement. The comparisons to MLS in the supplement appear to indicate that, although the general morphology of the modeled fields is in good agreement with that observed, ATLAS overestimates ClO abundances under enhanced conditions in the NH but underestimates them in the SH. Further discussion of the quantitative agreement between ATLAS and MLS ClO is warranted, along with its implications for some of the conclusions of the paper.
* HCl model/measurement discrepancy:
-- It is now mentioned (P34, L28-29) that, whereas the “uncorrected” ATLAS runs substantially overestimate HCl in mid and late winter in the NH and midwinter in the SH, the SLIMCAT CTM fairly consistently underestimated MLS HCl in both hemispheres in an earlier study. I note, however, that the runs with “corrected” HCl solubility (Figures 21 and 22) display behavior quite similar to that of SLIMCAT shown by Santee et al. . Those authors attributed the low modeled HCl to premature chlorine activation induced by the simplified SLIMCAT PSC parameterization. Although the fact that a large discrepancy in the corrected ATLAS runs remains in December in the NH is pointed out (P38, L9), its implications are not discussed. Does the underestimate of HCl in this case also suggest too early activation of chlorine (as for SLIMCAT) – and if so, is the modeled ClO consistent with that picture – or merely too much uptake of HCl into STS droplets? I suggest clarifying this so that readers do not get the wrong impression.
-- It is stated that the discrepancy cannot be caused by unrealistic modeled subsidence “as long as errors in subsidence do not compensate for errors in mixing” (P37, L14). Is this a robust assumption? Some justification for this statement would be good. The authors note the marked discrepancy between MLS and ATLAS N2O in the SH in October/November but state that that period is beyond the timeframe of interest in this paper. It seems to me, however, that the modeled and measured N2O curves have started to diverge already by mid-August. Couldn’t the gradual accumulation of errors over the course of the long simulation account for the larger deviation in October/November? I am not completely convinced that deficiencies in modeled (reanalysis) transport play no role even earlier in the season. The NH runs are two months shorter, which could account for the smaller apparent discrepancy (though even there the departure is larger at the end of the run).
-- In their replies to the referees and the Introduction, the authors make the point that the HCl discrepancy is a sufficiently important issue that it deserves a dedicated paper exploring the problems in the different models. I agree, and I wonder whether it might be good to make such a statement also in the Conclusions section of this manuscript (where the discrepancy and the steps taken in this study to deal with it should probably be mentioned in any case).
* Model validation:
-- From a quick glance at the plots, the overestimation of observed O3 outside the vortex appears to be a slightly bigger issue in the NH than in the SH. Do the authors have any explanation for that?
-- The significant overestimation of ClONO2 around 600 K in the SH is attributed to the initialization. But the agreement appears to be reasonably good in June and then worsens as the season progresses. Is that consistent with an initialization issue?
* Figure changes: I noticed that quite a few of the figure panels have been updated without explanation, e.g. ones depicting the partitioning within the various chemical families. Especially in the NH the new figures do look more reasonable. I wonder, though, if all figures in the supplement (e.g., Figures 15, 39) have been similarly updated.
Minor points of clarification, wording suggestions, and grammar / typo corrections:
* P2, L3: I suggest adding here a statement that the additional results in the supplement are summarized throughout the paper.
* P2, L21-22: “which are explored” --> “which is explored”
* P3, L1: “explores in how” – delete “in”
* P5, L14: “homogenous” --> “homogeneous”
* P6, L22: I suggest adding “directly” in front of “contribute”.
* P12, L10: “shift” --> “shifts”
* P12, L15: elsewhere “gas-phase” is hyphenated
* P13, L16: It would be helpful to add a pointer to Figure 5 at the end of this sentence.
* P14, L15-18: This sentence is still awkward. I think it would flow better if it were reorganized, perhaps along the lines of: “Particularly in the SH, CH4 oxidation plays an important role. Production from CH4 oxidation can be initiated by … with X=Cl, O(1D), or OH and then continues …”.
* P17, L5: add “but” before the second half of the sentence (“the relative importance …”)
* P18, L6: I think it would be better to move the end parenthesis from after “deactivated” to after “March 2005”.
* P18, L7: “the relative partitioning of OH”. Strictly speaking, it is not OH that is being partitioned, but HOx. It would be better to say “the relative importance of OH within the HOx family” or something similar.
* P19, L5: I suggest adding “over the winter” after “The increase of Cly”.
* P19, L6: The interhemispheric difference in wintertime diabatic descent rates now mentioned in the Cly discussion is indeed an important point. It might be good to add a reference for this point (e.g., Manney et al., JAS 1994, or Rosenlof et al., JGR 1995, or something similar).
* P19, L12: “inside Cly” --> “within Cly” or “between Cly species”
* P21, L14-16: add a comma after “Cl”, add a comma after “ClO”, delete “there is an”, replace “by” with “is maintained through”, and replace “in all” with “at all”
* P21, L23: maybe “motivated” is a better word here than “caused”?
* P21, L24: add “Section” in front of “2.2.2” for the WMO Report
* P21, L26: use either “In addition” or “also” but not both
* P21, L30: First, it is only necessary to state “2006” once, so delete it after “May”. Second, “September” should be “October” here – according to Figure 12 (and previous literature), chlorine is activated at least through September and even into October.
* P23, L18: add a period at the end of R4.
* P24, L1-4: This paragraph is a little confusing. First it is stated that “initially” more HCl is removed in the NH than in the SH, but then it is stated that the “initial” values of HCl are similar in the two hemispheres (“initial” is used in one form or another three times in these lines). Please clarify whether “initially” in the first sentence means December or February?
* P25, L29: “hpa” --> “hPa”
* P29, L9: R38 should be written here in the same manner as on P32.
* P29, L26: I suggest: “in the NH 2005 (where values range from 1.5 ppm to 2.5 ppm at 54 hPa) compared to the SH 2006, since …”.
* P31, L6: delete “a”
* P32, L6: delete the comma after “effective” and “the” before “Cl2O2”; also delete “the” before “Cl2O2” in L7
* P32, L16: “70 ppb” --> “70 ppb per day”; also “at 42 hPa” rather than “in”
* P33, L2: “increasing” --> “increases”
* P33, L5-7: “at all latitudes”; “at 475 K for 2005” (i.e., delete “and”); “See also Grenfell et al. (2006) for case studies”
* P33, L12: add “or” in front of “Br”
* P33, L25: “rare cases, such as”
* P33, L27: “… winter, respectively, and reflect …”
* P33, L29-30: “December 2009” and “December 2004”; also “heterogeneous”
* P34, L2-3: “or” would be better than “and” here; it might also be good to add at the end of the sentence (after “until April”): “and experience severe ozone loss”. Another reference to Manney et al.  and other papers examining multiple Arctic winters might also be appropriate here.
* P34, L13: “ACE-FTS” should be spelled out the first time it is used.
* P34, L17: “restrict discussion to” (not “on”)
* P34, L22: The appropriate (version 3) Data Quality Document should be cited for the accuracy information used to produce the error bars on MLS data in Figures 21 and 22.
* P35, Figure 21 caption: N2O should be added to the list of species shown. I also suggest adding “at 54 hPa” at the end of the first sentence. Finally, the error bars that have been added to the MLS measurements need to be mentioned in the caption.
* P37, L11: “into” --> “in”
* P37, L32: “above” --> “larger than”
* P38, L2: delete “a” in front of “good agreement”
* P38, L7: The authors might consider reiterating here that another reason not to take the temperature bias approach is that ERA-I temperatures are unlikely to be off to the degree necessary to improve the match between modeled and measured HCl.
* P38, L8-9: I suggest deleting “than before” and adding “than those from the original runs” at the end of the sentence. I also suggest adding “However,” at the beginning of the next sentence.
* P38, L12: I think this should be “ClOx”, not “ClO”.
* P38, L24: It might be good to add “in both hemispheres” after “vortex”.
* P38, L29: “there is only a limited amount of measurements” would be better as “there are few measurements” (or “there are only limited measurements”)
* P39, L12: “inside” --> “within”
* P39, L18-19: Add “cycle” after “ClO-O” and make “cycle” after “ClO-BrO” plural