Summertime OH reactivity from a receptor coastal site in the Mediterranean Basin

Zannoni, Nora; Gros, Valerie; Sarda Esteve, Roland; Kalogridis, Cerise; Michoud, Vincent; Dusanter, Sebastien; Sauvage, Stephane; Locoge, Nadine; Colomb, Aurelie; Bonsang, Bernard

doi:https://doi.org/10.5194/acp-17-12645-2017

Articles | Volume 17, issue 20

https://doi.org/10.5194/acp-17-12645-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

CHemistry and AeRosols Mediterranean EXperiments (ChArMEx)...

https://doi.org/10.5194/acp-17-12645-2017

© Author(s) 2017. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 17, issue 20

Research article

|

25 Oct 2017

Research article |

| 25 Oct 2017

Summertime OH reactivity from a receptor coastal site in the Mediterranean Basin

Nora Zannoni, Valerie Gros, Roland Sarda Esteve, Cerise Kalogridis, Vincent Michoud, Sebastien Dusanter, Stephane Sauvage, Nadine Locoge, Aurelie Colomb, and Bernard Bonsang

Download

Final revised paper (published on 25 Oct 2017)
Supplement to the final revised paper
Preprint (discussion started on 04 Oct 2016)
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: 'Review', Anonymous Referee #2, 20 Dec 2016
- AC1: 'Answer to anonymous referee 2', Nora Zannoni, 22 Feb 2017
RC2: 'Review of Zenoni et al. Summertime OH reactivity from a receptor coastal site in the Mediterranean basin', Anonymous Referee #1, 30 Dec 2016
- AC2: 'Answer to anonymous referee 1', Nora Zannoni, 23 Feb 2017

Peer-review completion

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

AR by Nora Zannoni on behalf of the Authors (24 Feb 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (27 Feb 2017) by William Lahoz (deceased)

RR by Anonymous Referee #3 (24 Mar 2017)

Suggestions for revision or reasons for rejection

This manuscript presents an OH reactivity and trace gas observational datasets during the CARBOSOR-ChArMex campaign in 2013. The field site is on the hill top of an island surrounded by the Mediterranean vegetation. In addition, the location is influenced by different background air masses such as polluted continental outflows and clean marine influences. As the authors argue in the manuscript, this is an ideal environment to examine roles of fresh BVOC emissions in different air masses from surrounding environments. The data analysis is mostly focused on identifying the potential sources of missing OH reactivity. As BVOCs are the most dominant contributor in OH reactivity assessments using the observed trace gas dataset, the discussion is mainly developed to attribute the source either unmeasured primary BVOCs such as reactive monoterpenes or unmeasured BVOC oxidation products. The authors applied the PMF analysis method to discern relative importance of primary BVOCs and secondary BVOCs any given time then conclude the large missing OH reactivity was mostly found when secondary BVOC factor was contributed higher.
The dataset will be an important contribution to the research community as the geographical coverage of OH reactivity observations is not still good enough. However, in my judgement, the main conclusion is appeared to be drawn rather inconclusively. According to Figure 2, most of high missing OH reactivity in terms of % not an absolute number can be observed July 23, 24, 25, 26, 27,28, and 29. However, the oxidation product shows very different pattern before and after July 26th (Figure 6). This observation reflects in PMF analysis results presented in Figure 8, as the sBVOC factor is assessed in the higher level during the time frame between July 26th to July 28th but the lower level before July 26th for a few days. Nonetheless, missing OH reactivity is also observed in higher level before July 26th. In this sense, it does not seem that the main conclusion, attributing the main source of missing OH reactivity to the oxidation products of BVOCs, is well supported by the analysis. Most of previous studies that this manuscript introduces have used a chemical box model to explore potential roles of unobserved BVOC oxidation products towards the observed missing OH reactivity. I would suggest that the authors consider including furthermore direct analysis on potential sources for the observed missing OH reactivity using a box model framework on top of the presented statistical analysis.

Hide

RR by Anonymous Referee #4 (31 Mar 2017)

Suggestions for revision or reasons for rejection

The revised submission is a significant improvement over the previous submission. However, some comments were not satisfactorily addressed (see below). The response text makes impression of rather chaotic - it would have been helpful if the response text was in different color or font style because it was not immediately apparent which text was from a reviewer and which corresponds to the authors response.
I think the story is interesting and would be a valuable contribution to the existing literature on OH reactivity from this Mediterranean region, but I do have remaining concerns which I suggest are dealt with before the publication.

- I completely agree with the other reviewer that the PMF addition to this paper is confusing and cannot be properly evaluated in this paper. I am surprised that the authors persist on keeping it, because in this case it is necessary to provide sufficient information for the PMF data to be interpreted rather than repeating the methods from the companion paper. There are also other questions which make me wonder about the meaningfulness of this addition. The concentrations of several ppm on PMF factors (see Fig. 8) are surprisingly high. Is it because the loadings included trace gases other than VOCs? Why was the sBVOC factor not sensitive to the period on 23-25 July if isoprene and its oxidation products were clearly emitted? Is it because the uncertainties were set too high in the model? If the PMF graph is kept it would be nice to see how each factor is broken down by the profile weights of constituent VOCs which would help in the PMF interpretation.

- I am surprised how the comment on toluene source apportionment was addressed. I do not think the reviewer was suggesting toluene was primarily biogenic at this site, but that it does have different known source categories which were not even mentioned by the authors. This is evident from the companion paper (Michoud et al.) showing toluene split across all the factors, and the figure with benzene and toluene shows distinctively different patterns of toluene and benzene. While benzene enhancements would not be expected biogenic, on many days toluene shows a distinct pattern uncorrelated with benzene and during daytime (e.g. 2-4 Aug). Although toluene biogenic concentrations were low relative to other compounds, the toluene contribution was elevated in sBVOC factor (Michoud et al., Figure 7) so some of it could be advected together with longer lived isoprene oxidation products. Unfortunately, other common biogenic benzenoids such as benzaldehyde were not measured. Benzaldehyde fluxes were recently reported from red oaks (Cappelin et al., 2017). The argument about the mixed classification also concerns MEK which authors also attribute exclusively to anthropogenic category while it has recently been observed emitted from plants taking up MVK and converting it to MEK (Cappelin et al., 2017). The general issue is that a single compound can often have more than just one source category, but this could be denoted with an asterisk next to a compound and briefly discussed.

- The authors nicely mention unmeasured monoterpenes in the discussion, but there could be many other unmeasured compounds some of which would not be easily detected by GC. PTRTOF often sees hundreds of ions from forested systems, so I wonder if it would be possible to estimate calculated OH reactivity from remaining PTR-TOF ions (e.g. with a default OH reaction rate constant or upper and lower limits). Alternatively, it could be discussed whether any unidentified PTRTOF ions or unmeasurable species could significantly explain the missing reactivity.

Technical:
- Figure and Fig. are still used inconsistently.

Hide

ED: Reconsider after major revisions (03 Apr 2017) by William Lahoz (deceased)

AR by Nora Zannoni on behalf of the Authors (02 Jul 2017) Author's response Manuscript

ED: Referee Nomination & Report Request started (08 Jul 2017) by William Lahoz (deceased)

RR by Anonymous Referee #4 (21 Jul 2017)

RR by Anonymous Referee #3 (15 Aug 2017)

Suggestions for revision or reasons for rejection

This is a review for the first revision of the manuscript, submitted by Zannoni et al. Overall, the manuscript has been improved in terms of scientific scopes and data presentation. However, I still believe that it needs to be significantly improved by addressing some inconsistencies to be considered for the final production.

The main weak point of the presented notion of data analysis is that the authors just focused on one period when the maximum OH reactivity was quantified (e.g. July 29th and the three-consecutive day after that). The justification was that this was the time period when local biogenic influences were dominant. Using a correlation analysis between temperature and missing OH reactivity, the authors argued the case (Figure 7). However, there are a couple of obvious observations that the authors need to address to substantiate this reasoning. First, it does not seem to anthropogenic influences did not change too much over the time by observing the temporal variations of AVOCs and others (probably mostly trace gases such as CO and NOX). The BVOC contribution to the total calculated OH reactivity changes significantly and the variance does not correlate with the notated wind direction shown in Figure 2. In this context, I would suggest that the authors need to discuss the obvious inconsistency in the level of assessed missing OH reactivity. For example, in Aug 8th, the BVOC contribution to the calculated OH reactivity is appeared to be the largest during the whole observational period. However, it does not seem to have any missing OH reactivity, which is inconsistent with the main conclusion of the current form of manuscript and requires further justification. Moreover, the duration between July 23rd and July 26th, the fraction or percentage of missing OH reactivity with respect to observed OH reactivity is appeared the highest although the BVOC contribution towards calculated OH reactivity was relatively small. This also requires for further reasoning. In this reason, I am strongly convinced that the data analysis should be redone for holistic discussion rather than just focused on one time period as the current manuscript is framed. One notable difference I can tell the episode (July 26th to July 28th) when the authors focused on from the other period is high temperature and low wind speed, which may indicate a strong stagnation episode. This may allow accumulating uncharacterized/unknown oxidation products, I suspect.

Minor comments

1) Recently in AMTD, the results of an OH reactivity intercomparison study has been published. It would better to describe the results in the perspectives of potential limitations of the analytical method.

2) Page 10 Line 8: Please be consistent by either using hydroxyl radical reactivity or OH reactivity

3) Please define how you evaluate missing OH reactivity

4) Page 14 Line 4 to Line 11: What about the other periods? The explanation for the dataset between 23rd to 26th is not clear enough. The authors need to discuss more specific how they have reach that conclusion.

5) Page 23 Tabel 1 it is “alpha” pinene and “beta” pinene not a and b-piene, repsctively.

Hide

ED: Reconsider after minor revisions (Editor review) (24 Aug 2017) by William Lahoz (deceased)

AR by Nora Zannoni on behalf of the Authors (03 Sep 2017) Author's response Manuscript

ED: Publish subject to technical corrections (05 Sep 2017) by William Lahoz (deceased)

AR by Nora Zannoni on behalf of the Authors (13 Sep 2017) Manuscript

Download

Article (2678 KB)
Full-text XML

Short summary

Our paper presents results of hydroxyl radical (OH) reactivity from a field study conducted during summer 2013 in a western Mediterranean coastal site (Corsica, France). Here, the total OH reactivity, measured with the comparative reactivity method, is compared with the summed OH reactivity from the reactive gases measured with a multitude of different technologies. Our results demonstrate the relatively high observed reactivity and the large impact of biogenic compounds.