Characterizing sources of high surface ozone events in the southwestern US with intensive field measurements and two global models

Zhang, Li; Lin, Meiyun; Langford, Andrew O.; Horowitz, Larry W.; Senff, Christoph J.; Klovenski, Elizabeth; Wang, Yuxuan; Alvarez II, Raul J.; Petropavlovskikh, Irina; Cullis, Patrick; Sterling, Chance W.; Peischl, Jeff; Ryerson, Thomas B.; Brown, Steven S.; Decker, Zachary C. J.; Kirgis, Guillaume; Conley, Stephen

doi:https://doi.org/10.5194/acp-20-10379-2020

Articles | Volume 20, issue 17

https://doi.org/10.5194/acp-20-10379-2020

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

https://doi.org/10.5194/acp-20-10379-2020

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

Articles | Volume 20, issue 17

Research article

|

08 Sep 2020

Research article |

| 08 Sep 2020

Characterizing sources of high surface ozone events in the southwestern US with intensive field measurements and two global models

Li Zhang, Meiyun Lin, Andrew O. Langford, Larry W. Horowitz, Christoph J. Senff, Elizabeth Klovenski, Yuxuan Wang, Raul J. Alvarez II, Irina Petropavlovskikh, Patrick Cullis, Chance W. Sterling, Jeff Peischl, Thomas B. Ryerson, Steven S. Brown, Zachary C. J. Decker, Guillaume Kirgis, and Stephen Conley

Download

Final revised paper (published on 08 Sep 2020)
Supplement to the final revised paper
Preprint (discussion started on 03 Dec 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

RC1: 'Reviewer Comments Zhang et al', Anonymous Referee #1, 06 Jan 2020
- AC1: 'Author Responses to Reviewer #1', Meiyun Lin, 04 May 2020
RC2: 'Review of Characterizing sources of high ozone events', Anonymous Referee #2, 07 Feb 2020
- AC2: 'Author Responses to Reviewer #2', Meiyun Lin, 04 May 2020

Peer-review completion

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

AR by Meiyun Lin on behalf of the Authors (07 May 2020) Author's response Manuscript

ED: Referee Nomination & Report Request started (23 May 2020) by Jason West

RR by Anonymous Referee #2 (08 Jun 2020)

Suggestions for revision or reasons for rejection

The revised version of this manuscript has resolved most of my questions with regard to this paper. I would recommend publication after the authors clarify the details below (all line numbers refer to the marked version at the end of the comments to reviewers). Most of these are rather minor although I am concerned that AM4 might be somewhat overzealous in its STT. This is addressed in comments (1) and (9). It would be good if the authors addressed this point, perhaps in the conclusions.

1. l349 of revised document. It is unclear how the STT events in Figure 4 are defined. On approximately May 7 for example the AM4 shows a peak in stratospheric ozone at SM Youth Camp with elevated ozone (not seen in the observations) which appears to be a STT event in the model. However, this event is not marked. There are other instances when the model seems to show high stratospheric ozone values that are not marked as STT events. It appears that the AM4 may show a number of false positives? Please explain and clarify. Also please comment with regards to these so called “false positives”. (One of the major points of the paper is the importance of the ability of AM4 to simulate STT events. However, it looks to me like it may actually simulate too many of these events.)

2. Figure S4, Figure 9 and analogous figures. Please give the spatial correlation between the simulated and measured ozone for both models? And perhaps the mean bias. It looks like AM4 is does not capture the geographic anomaly pattern. In fact in a number of locations in seems like the USB in AM4 is higher than that measured.

3. L537 in track change manuscript. But in fact, despite the similarity in structure on an isentropic surface the two patterns in ozone, and its relations to isentropes still look remarkably different. Where the ozone is, the shape of the 322 isentrope with respect to the fold, the shape of the tropopause etc.

4. Figure 7 and discussion. The pattern of isentropic surfaces between the two simulations still look significantly different. For example, looking at the 322 K isentropic surface: at the southern edge of the slice this surface is significantly lower than in AM4 than in GEOS-chem; in AM4 it goes right through the fold whereas in GEOS-chem it is above the fold; north of the fold it is located higher in AM4 than in GEOS-chem and then descends near the north boundary in AM4 whereas in GEOS-chem it remains level. This difference really suggests a remarkable difference in the two meteorological analysis over a data rich region. It seems this should be pointed out. It is unclear to the extent to which the model differences are caused by difference in the meteorology. The difference in the 322 K contour may explain, in part, some of the differences in ozone between the two simulations. It is certainly worthwhile pointing out the large differences between the simulations south of the ozonesonde profile also. These differences are large and seem significant.

5. l675. “likely due to excessive O3”. I understand that in general lightning NOx is too high in the GEOS-chem simulations, but what is the evidence that on this particular day (June 22) it is likely the excessive ozone is due to lightning NOx. Were there thunderstorms in the vicinity, or …. ? “likely” seems a strong word without some in depth analysis.

6. Figure 9 and similar plots. I may have missed this, but why are some of the observed measurements shown in large dots, other in small dots?

7. I think the paper did not comment on Figure 16b. All figures show should certainly be commented on in the text. In fact it is quite remarkable that USB has a different North-South gradient in the two models. I don’t believe this can be solely attributed to lightning NOx. Maybe I missed it, but is USB calculated the same way in both simulations. This figure is certainly worth a comment.

8. Figure S12 was never comment on (again unless I missed it) in the main manuscript. It should be.

9. Figure S14. While it is true that the highest observed points show high values of stratospheric ozone it should be emphasized that for the most part those points in which AM4 are biased the stratospheric ozone is high. This seems like an important point to bring up. At least to me it seems quite likely that the AM4 model has too much STT (point 1). I think the authors should comment on this possibility.

Hide

RR by Anonymous Referee #1 (01 Jul 2020)

Suggestions for revision or reasons for rejection

The authors have been responsive to reviewer comments and the revised manuscript addresses many of my concerns. I have a few remaining suggestions for improvement to the article before publication.

Abstract: The authors added a lot of nuanced language and discussion about uncertainties to the main text of the paper, but this language did not make it into the abstract which still talks about “pinpointing” sources and appears to suggest definitive attribution of O3 events.

Section 4: I still think a bit more information about model performance would be useful here. I suggest that the authors add a table which includes the model MB for AM4 and GC at the LV surface sites on each of the specific episode days examined (i.e. have a separate row for each of the dates analyzed in section 4).

Minor comments:
Note that the term “exceptional events” has specific regulatory meaning and specifically results in exclusion of data from the regulatory process.

Figure S5 (Apr 23) looks like O3strat_anomoloyg is much lower (5-10 ppb) but this doesn't seem to match purple lines on Fig 4

Page 13, second paragraph: Note that AM4 does not definitively show the strat intrusion reaching ground level.

Page 21, end of second paragraph: This is a good place to note that these quantitative model attributions are only as good as the precision/capability of the model which is why model eval is so important. This will also transition well into the next paragraph.

Page 23: increasing surface O3 in China may not indicate increased long-range transport of ozone to the US if the higher O3 is due to less titration from decreased NOx emissions. Atmospheric chemistry theory predicts that even when O3 is locally suppressed from NOx, high O3 forms downwind. So, if O3 is increasing due to less suppression that does not mean that more total ozone will be created/transported downwind.

Figure 6: r2 value appears to be missing from the panel showing June 22.

Hide

ED: Publish subject to minor revisions (review by editor) (03 Jul 2020) by Jason West

AR by Meiyun Lin on behalf of the Authors (13 Jul 2020) Manuscript

ED: Publish as is (25 Jul 2020) by Jason West

AR by Li Zhang on behalf of the Authors (28 Jul 2020)

Short summary

Measuring and quantifying the sources of elevated springtime ozone in the southwestern US is challenging but relevant to the implications for control policy. Here we use intensive field measurements and two global models to study ozone sources in the region. We find that ozone from the stratosphere, wildfires, and Asia is an important source of high-ozone events in the region. Our analysis also helps understand the uncertainties in ozone simulations with individual models.