Global inorganic nitrate production mechanisms: comparison of a global model with nitrate isotope observations

Alexander, Becky; Sherwen, Tomás; Holmes, Christopher D.; Fisher, Jenny A.; Chen, Qianjie; Evans, Mat J.; Kasibhatla, Prasad

doi:https://doi.org/10.5194/acp-20-3859-2020

Articles | Volume 20, issue 6

https://doi.org/10.5194/acp-20-3859-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/acp-20-3859-2020

© Author(s) 2020. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 20, issue 6

Research article

|

31 Mar 2020

Research article |

| 31 Mar 2020

Global inorganic nitrate production mechanisms: comparison of a global model with nitrate isotope observations

Becky Alexander, Tomás Sherwen, Christopher D. Holmes, Jenny A. Fisher, Qianjie Chen, Mat J. Evans, and Prasad Kasibhatla

Download

Final revised paper (published on 31 Mar 2020)
Supplement to the final revised paper
Preprint (discussion started on 08 May 2019)
Supplement to the preprint

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

SC1: 'A specific comment on the manuscript by Alexander et al.', Xin Yang, 20 May 2019
- AC1: 'response to Xin Yang', Becky Alexander, 20 Sep 2019
RC1: 'Nitrate Production Mechanisms', Anonymous Referee #1, 02 Jun 2019
- AC2: 'response to anonymous referee #1', Becky Alexander, 20 Sep 2019
RC2: 'Review of Alexander et al.', Anonymous Referee #2, 08 Jun 2019
- AC3: 'response to anonymous referee #2', Becky Alexander, 20 Sep 2019
RC3: 'revisions', Greg Michalski, 17 Jul 2019
- AC4: 'response to Greg Michalski', Becky Alexander, 20 Sep 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Becky Alexander on behalf of the Authors (20 Sep 2019) Author's response Manuscript

ED: Referee Nomination & Report Request started (23 Sep 2019) by Jan Kaiser

RR by Anonymous Referee #2 (11 Oct 2019)

Suggestions for revision or reasons for rejection

The authors have certainly improved the manuscript in response to the reviewer’s comments. Submission of the revised manuscript and continuing onto publication in ACP is warranted. There are few areas that the authors should revisit and consider further revision based upon the original reviewers’ comments:

(1) The authors added a qualitative explanation for the lack of agreement with observations in Mt. Lulin as lack of heterogeneous chemistry “due to minimal aerosol surface area.” However, this statement contrasts with the conclusions drawn in the Guha et al. observational study, so the response by the authors needs to be refined to better explain this interpretation (do they mean that the model predicted aerosol surface area is too lacking to have heterogeneous chemistry?).

(2) The authors make the excuse in considering a comment about the Wang et al., GCA, 2014 paper and the Fibiger et al., 2016 paper that the data is “not available”. It has long been the practice to contact corresponding authors for data if it is not available in the manuscript. And I found that the Fibiger et al., 2016 reference actually states the following: Data from this paper are available at ACADIS. Data sets https://www.aoncadis. org/project/collaborative_research_the_ impact_of_bromine_chemistry_on_the_ isotopic_composition_of_nitrate_at_ summit_greenland.html.

Looking at this website it appears to include the isotope data from both Fibiger et al 2016 and Fibiger et al 2013 (the 2013 one reports the D17O data). The D17O data from the Fibiger et al., 2013 (Geophysical Research Letters, VOL. 40, 3484–3489, doi:10.1002/grl.50659, 2013) should be considered in the current study and does include values that look to be close to 39 per mil (or at least definitely >>30 per mil!). The authors should revisit this and consider the implications for their response in the manuscript. Also consider contacting F. Wang or G. Michalski for the data from Wang et al. so this can be included as well.

(3) The statement added by the authors that “Although lack of transport of the isotope tracers hinders direct comparison of the model with observations at any particular location” contrasts with the fact that they make direct comparison with a range of time series in Figure 5. So maybe restate this that the lack of transport adds uncertainty to direct comparisons – but you do make direct comparisons in space and in time!

(4) The phrasing of “the influence of clouds on nitrate formation” does not really make sense. This should be rephrased to account for the fact that precipitation will represent a column average of nitrate (i.e. long-range transported nitrate, nitrate formed in clouds, and nitrate formed near the surface). The point that the meteorology tends to have clouds near 1 km means that the model sampling is robust for comparison on this point, but the impact of clouds on nitrate formation does not seem to be the point here.

(5) It is not clear whether the authors added a clear reasoning for the cloud chemistry simulation versus the standard simulation to the manuscript. (Response to comment marked “Page 16, Section 4.2”). The manuscript needs to be clear about how, when and where the results from different simulations are being used and why.

(6) Regarding the comments on understanding D17O(NO3-) more regionally (e.g., showing how D17O(NO3-) changes regionally based upon the sensitivity studies). Perhaps another way to consider this is to add a figure to the SI that shows the results of the different simulations for the times series comparison with observations (ie Figure 6). This would give much more quantitative information for researchers conducting observations and give much more information about how sensitive the D17O is in different regions where these processes are more/most important in different seasons. This would only add 1 figure to the SI (i.e. Figure 6 with different color lines representing a few different sensitivity simulations?).

(7) The dashed lines in Figure 5 appear to represent +/- 50%. These should be defined in the figure caption and the authors should consider whether it would make more sense to include dashed lines at +/- 25%.

Hide

RR by Greg Michalski (16 Jan 2020)

Suggestions for revision or reasons for rejection

The authors have substantially improved their manuscript. However I believe they need more throughly and directly address two issues raised by several of the reviewers.

1) The troublesome of value of the O3 D17O value as some fixed value. Using Vicars et al. data does not address the T and P effect demonstrated by numerous lab experiments. The argument that stratospheric O3 "resets" avoids the issue. Any NO oxidation or NO3- production above the mixed layer will likely have a different D17O because the O3 D17O in those layers will be a function of T and P and not fixed at 25 permil. The authors seem to argue that using 25 best "fits the data". This seems a circular argument. One could also argue that the experimental O3 D17O are correct and the pathways are actually wrong. There should be a measure of NO3- production in each model layer...How important is NO3- production at say 5 km and what might the O3 D17O be t this T and P? It would be difficult to hash all this out in the current paper but my fear is that there is a mantra of "its 25 permil always and everywhere" is being repeated by a host of recent papers at the expense of numerous other studies that say otherwise. This makes it increasing difficult to challenge. There should be a least one paragraph that there is somethings we don't understand about O3 D17O and a critical assessment of these conflicting estimates.

2) The role of NO emissions at night is still not satisfactory addressed. Morin et al.s model did not include emissions, thus their conclusions about 5% are not valid. In most of the domain of a global model the nighttime emissions are comparable to daytime. Only urban areas with vehicles is there a significant difference between daytime and night emissions. Thus NO emitted at night retains its source O until sunrise scrambling. How much of this oxidized at night to NO2 to exchange or form NO3-? Clearly this would have a major impact in high latitudes in the winter. Are we to be convinced the NO emitted in Alaska in Jan. is photochemically equilibrated with O3 within 5 %? Seem implausible. I do not expect the authors to redo their model, but there should be another full paragraph is the discussion of the limits of the equilibration assumption.

Hide

ED: Reconsider after major revisions (16 Jan 2020) by Jan Kaiser

Dear Dr Alexander

Thank you for submitting your revised manuscript to Atmospheric Chemistry and Physics.

Both reviewers have suggested minor revisions, but actually raised quite substantial points requiring modifications of several sections and figures, as well as inclusion of previously published data for model validation.

In particular, the uncertainties associated with both lab measurements of the 17O excess of O3 (Morton et al. 1990) and potential methodological biases in the NO2- oxidation method of Vicars et al. may have been underestimated.

Please address the points raised by the reviewers, as well as my complementary comments below, when you submit your revised manuscript and reply to reviewers' comments.

Sincerely
Jan Kaiser
Editor ACP

Introduction (p. 5/l. 9) and conclusions (19/12):
Please add a caveat that previous modelling efforts have made different assumptions about the preferential transfer of central and terminal O atoms to NO2 and NO3, and the 17O enrichment of different ozone isotopomers. This is still not clear enough.

5/2: Remove tilde sign and adjust interval to encompass full range of observations (6 to 54 ‰ based on Krankowsky et al. 1995; 19 to 41 ‰ based on Johnston & Thiemens 1997).

5/4: Likewise, the range shown here is too narrow. It's 30 to 46 ‰ for Morton et al. (1990). Please also add "et al." to the reference.

17/26 & 20/1: Replace tilde sign by actual range value with uncertainty. All measurement results should be rounded according to their uncertainty and stated with an estimate of their measurement uncertainty. Approximation symbols should therefore not be used (unless you are approximating a mathematically exact number, e.g. π ≈ 3.14). In any case, the correct approximation symbol has two wavy lines (≈). It is not the tilde sign (~), a symbol which has perhaps made it into the literature due to limitations of early typewriters.

Figure S1: More than half of the plot appears with the color corresponding to the colorbar maximum. Please include a variant of the plot with an increased maximum value so that variations in τ >= 2 d can be distinguished, or perhaps add contour lines for values higher than 2 days.

Figure S3: Please explain the meaning of the dashed lines in the figure caption.

Figure S6: The caption should refer to Fig. S3.

Hide

AR by Becky Alexander on behalf of the Authors (15 Feb 2020) Author's response Manuscript

ED: Publish subject to technical corrections (17 Feb 2020) by Jan Kaiser

AR by Becky Alexander on behalf of the Authors (17 Feb 2020) Author's response Manuscript

Download

Article (8897 KB)
Full-text XML

Short summary

Nitrogen oxides are important for the formation of tropospheric oxidants and are removed from the atmosphere mainly through the formation of nitrate. We compare observations of the oxygen isotopes of nitrate with a global model to test our understanding of the chemistry nitrate formation. We use the model to quantify nitrate formation pathways in the atmosphere and identify key uncertainties and their relevance for the oxidation capacity of the atmosphere.