Characteristics of 1 year of observational data of VOCs, NOx and O3 at a suburban site in Guangzhou, China

Zou, Y.; Deng, X. J.; Zhu, D.; Gong, D. C.; Wang, H.; Li, F.; Tan, H. B.; Deng, T.; Mai, B. R.; Liu, X. T.; Wang, B. G.

doi:https://doi.org/10.5194/acp-15-6625-2015

Articles | Volume 15, issue 12

https://doi.org/10.5194/acp-15-6625-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

https://doi.org/10.5194/acp-15-6625-2015

© Author(s) 2015. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 15, issue 12

Research article

|

16 Jun 2015

Research article |

| 16 Jun 2015

Characteristics of 1 year of observational data of VOCs, NO_x and O₃ at a suburban site in Guangzhou, China

Y. Zou, X. J. Deng, D. Zhu, D. C. Gong, H. Wang, F. Li, H. B. Tan, T. Deng, B. R. Mai, X. T. Liu, and B. G. Wang

Download

Final revised paper (published on 16 Jun 2015)
Preprint (discussion started on 17 Jul 2014)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C5330: 'Reviewer's comments', Anonymous Referee #1, 27 Jul 2014
- AC C7204: 'Interactive comment on“An analysis of the impacts of VOCs and NOx on the ozone formation in Guangzhou”by Y. Zou, et al', xuejiao deng, 18 Sep 2014
RC C7505: 'Comments-ACPD-14-18849-2014', Anonymous Referee #2, 29 Sep 2014

Peer-review completion

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

AR by Xuejiao Deng on behalf of the Authors (27 Oct 2014) Author's response Manuscript

ED: Referee Nomination & Report Request started (30 Oct 2014) by Steven Brown

RR by Anonymous Referee #1 (04 Nov 2014)

Suggestions for revision or reasons for rejection

1. The VOC/NOx ratio approach
The reviewer appreciated the effort of the authors on the revisions of the manuscript according to the reviewers’ comments. However, I am still not convinced by the revised manuscript. As stated in my previous comments, VOC/NOx can only give rough idea about VOC or NOx limited. Other methods/models must be used to confirm the traditional method. The limitations of the VOC to NOx ratio approach have already been well addressed by the U.S. scientists in the report (NESCAUM (1995): Preview of the 1994 ozone precursor concentrations in the northeastern U.S. 5/1/94 draft report; prepared by the Ambient Monitoring and Assessment Committee of the Northeast States for Coordinated Air Use Management, Boston, MA). In brief, 1) the measured absolute concentrations of precursors at the site are determined by local emissions and transport from upwind locations. It will ultimately be critical to distinguish between ozone contributions from local precursor emissions, from transported ozone formed in upwind locations, and from in-situ ozone production from transported upwind precursors. 2) It is questionable about which species should be included in a calculation of “total VOCs” and “total NOx”. The VOC/NOx ratio can be very different if you chose different VOC species to make total VOCs, leading to VOC/NOx variable. Traditional NOx measurements include NO + variable combination of NO2 and some other reactive nitrogen species (NOy). 3) The simple ratio of the two disregards the composition and reactivity of individual hydrocarbon and nitrogen compounds. During a multi-day episode, concentrations of TNMHCs may build to high concentrations under a stagnating high pressure system. However, without fresh injections of VOCs, the most reactive compounds will become depleted, leaving high VOC levels composed of less reactive species with little potential for ozone production. Sillman has already established equations by considering the reactivity of VOCs. 4) The VOC/NOx ratio approach focuses exclusively on the 6 to 9 AM time frame, while this manuscript calculated the ratios at all hours of a day. Also, reactive nitrogen compounds like PAN, HONO and nitric acid exhibit diurnal patterns which differ from those of NO and NO2.
Although the authors emphasized that the measured ozone isopleths were used and confirmed by the measured VOC/NOx ratios, both are basically the same and a kind of self-confirmation. The conclusion is dangerous only based on the measured VOC/NOx ratios which have a number of limitations described above and are not confirmed by other methods and models. Therefore, if NOx limited result is true, you can say this is one of the highlights in this paper. If the result is not reliable, this paper will significantly mislead the readers.
In addition, though other studies may use the VOC/NOx ratio approach to roughly estimate the sensitivity of VOCs or NOx, all these studies also used other methods and models to verify the traditional method. But in this manuscript, the authors exclusively relied on this traditional method.

2. Sampling site selection
The statement of “The site has been demonstrated by many experts….” does not make sense at all. Who are the experts? Please give references recognized by the national/international experts.
3. QA/AC of VOC measurements
My question was not fully answered. QA/QC includes internal and external calibration. Since the authors claimed that this is the first time to measure VOCs online in the region, and the online measurement method is not the same as other online instruments, validation of the results in the study is important. Have you conducted any inter-comparison with other recognized groups nationally/internationally? In other words, how could readers trust your measurement data? As a matter of fact, I am not 100% confident with the VOC online data, based on the authors' reply to the specific comment 4) below.
My another question on the separation efficiency of low carbon number VOC species at 13C was not answered. Although at 13C the instrument can separate C2 and C3 VOCs, it can only be achieved by series connection of two columns i.e. DB-1 and plot-Q. If only one column i.e. DB-1 was used, the C2 and C3 VOCs could not be separated at such temperature. Our experience is that it is difficult to well separate C2 such as ethene from ethyne for such kind of two-analyzer instrument.
Even if the pre-concentration temperature was negative, we still need two columns in series. Otherwise the separation efficiency was poor.

4. Statistical analysis was still not conducted

Though the authors replied that they have added the statistical analysis in the manuscript, they actually did not. I doubt they even did not understand my question about statistical analysis.

References about VOC- and NOx-limited
Sillman S, OBSERVATION-BASED METHODS (OBMS) FOR ANALYZING URBAN/REGIONAL OZONE PRODUCTION AND OZONE-NOx-VOC SENSITIVITY. http://www-personal.engin.umich.edu/~sillman

Sillman S, OVERVIEW: TROPOSPHERIC OZONE, SMOG AND OZONE-NOx-VOC SENSITIVITY. http://www-personal.engin.umich.edu/~sillman
USEPA, EPA-454/R-96-006 Chapter 4: OBSERVATIONAL BASED METHODS FOR DETERMINING VOC/NOEFFECTIVENESS.

Hide

RR by Anonymous Referee #3 (22 Nov 2014)

Suggestions for revision or reasons for rejection

General Comments:

This is a valuable contribution showing the seasonal and diurnal variation of ozone, NOx, VOCs and VOC reactivity at a polluted site downwind of a major Chinese urban and industrial area. The interpretation of some aspects of the paper, especially with regard to VOC and NOx limitations and how they are derived, and the role of transport in determining ozone concentrations, needs to be further improved prior to final publication. Please follow specific recommendations outlined below.

Specific Comments:

Title: The new title is a step in the right direction. The original reviewer’s comments suggest that the main value in this work is the reporting of the VOC, NOx and O3 data itself, while the analysis of photochemistry is somewhat more difficult to interpret. Thus, a title more along the lines of “One year of observational data of VOCs, NOx and O3 at a suburban site in Guagnzhou, China” would be more appropriate.

Page 3, line 19: “ozone isopleth diagrams” (no s on the end of isopleth). Search on “isopleths” throughout the manuscript and change to “isopleth”.

Page 3, line 22: The Dodge 1977 reference is a conference proceeding and not easy to find. It is cited as the key support for the 8:1 ratio of VOC to NOx below which O3 production is VOC limited. The authors should make two changes here.
1) Use the following citation, which is much more easily accessed than Dodge 1977.

Seinfeld, J. H. (1989), Urban Air Pollution: State of the Science, Science, 243, 745-752.

The ozone isopleth diagram in this paper is roughly consistent with the stated 8:1 VOC to NOx ratio, so long as the VOC is quoted in parts per million (or parts per billion) of carbon rather than as a direct mixing ratio.

2) Include a statement that the actual VOC to NOx ratio at which O3 production is VOC limited will depend on specific conditions within a given area, so that the use of a single ratio (such as 8:1) is only approximate, to be used if detailed photochemical modeling is not available.

Page 3, line 38: Replace “characterizations” with “characteristics.”

Page 3. Line 39-41: The sentence on observational ozone isopleth diagrams may need to be removed. See comments below.

Page 4, line 1: replace “regime” with “regimes”

Page 4, line 41-43: How is NO2 measured in the NOx system? Is it by photolytic conversion to NO, or by using a molybdenum converter? If molybdenum, the actual NOx may be too large. See further comments below. At this point in the paper, please add specification for the type of NO2 converter on the NOx instrument.

Page 5, line 45: The quoted NOx concentrations in Table 3 and shown in Figure 4 are very large. If the converter used on the NOx instrument is heated molybdenum, then a statement should be included at this point that the NOx may include some oxidized reactive nitrogen that is converted by the molybdenum, and thus that the stated mixing ratios are upper limits to the actual NOx.

Page 6, line 5: the word photochemical is used twice. Eliminate the first one.

Page 6, line 8: Suggest replacing “variation trend” with “derivative with respect to time”. On line 11, replace “growth of variation” with “derivative.”

Page 6, lines 10-22: There is no discussion of the role of transport in the interpretation of Figure 5. The morning rise in ozone, for example, is as likely to arise from the breakup of the nocturnal boundary layer during morning hours as from chemical production of ozone. The authors should consider if transport of ozone rich or ozone poor air to this location could influence the derivatives. At a minimum, they need to state that the assumption is that ozone is uniform vertically and horizontally if they are to interpret the diurnal variation in the ozone derivative as being due to chemistry alone. Much better, however, would be to include some discussion of the possible role of transport, especially vertical transport, since breakup of the nocturnal boundary layer leads to growth in surface level ozone even without significant chemical production in many locations.

Also, the authors should have sufficient data to analyze this on a seasonal basis, rather than an average for the entire year, since the preceding figure shows that the diurnal variation of ozone depends on season. They should either add additional figures for different seasons to figure 5, or add a sentence to the text explaining the choice of a single plot.

Page 6, Equation 3: The description is not clear, especially the quantity u¬_ozone. This is the “molecular mass of species J in the ozone”, where the index J refers to a VOC. Also it appears from the description that the MIR factors are taken from a table or otherwise looked up from the Carter (1994) reference. If so, please make this description clearer. Also note that MIR determined in one location may be different in another, so that the application of MIRs modeled in a different city would be only approximate correct in Guangzhou.

Page 6, lines 28-29: Remove sentence fragment “on ozone formation”

Page 7, line 15: Replace “most important categories of the atmosphere” with “highest mixing ratio”

Page 7, line 17: “Alkane content is high”

Page 7, line 40: Dong Guan should be indicated on the map in figure 1 if it is to be cited here.

Page 7, lines 44-45: Add “, respectively.” After “OH activity and MIR”. The following sentence does not make sense and should be revised. This reads like a policy recommendation for the types of vegetation that should be grown around the site, but is this something that could really be controlled? Better would be a simple statement of the facts, such as what type of vegetation surrounds the site (e.g., forest vs grassland) and if there is any significant variation in the vegetation around the site that would lead to variations in isoprene concentration with wind direction.

Page 8, section 3.3 and figure 7: This new figure showing an ozone isopleth based on observational data is unconventional and likely out of place in this analysis. The addition of this figure seems to take the paper in a new direction, different from the original submission. It is also not common to present observationally based ozone isopleths. Normally these are generated only by models, since they are intended to represent chemical processes only and not transport. The figures do not lead the reader to any very obvious conclusion either, since they do not show the common transition between VOC and NOx limited regimes, showing instead several local maxima for ozone formation. Inclusion of this figure is not appropriate, especially as an addition to the paper after the first stage of review. Also, while figure 8 shows NOx and VOCs to be correlated in different seasons, it is not clear why this correlation supports the notion that ozone production is NOx limited, unless it is being used to derive the slope of VOC:NOx. In that case, the authors should include fits on this figure, show the slopes explicity, and including in the discussion.

Page 9, line 10: See comments above regarding the Dodge reference and the citation of a single number for the VOC:NOx ratio. Text here also seems to be a repeat of the introduction, and is thus redundant.

Page 9, line 19 and discussion of VOC:NOx ratio. A dashed line should appear at the stated ratio of 8 in Figure 9 to aid the reader in visualizing how the diurnal variation in this ratio changes with season. Also, the text here should include the caveat that the NOx is an upper limit to actual NOx if a molybdenum converter was used for NOx measurement. If that is it the case, the VOC:NOx ratio would possibly be larger than shown, and the authors should include appropriate caveats to this discussion if that is the case.

Page 9, line 43. New paragraph beginning with “In order to”

Page 10, line 3: State how many days there were with ozone in excess of 93 ppbv. Also, add a line at VOC:NOx = 8 to Figure 10.

Page 10, lines 16-24: Include transport in this discussion, see comments above.

Page 10, lines 32-36: Modify as needed after removing figure 7.

Hide

ED: Reconsider after major revisions (22 Nov 2014) by Steven Brown

AR by Xuejiao Deng on behalf of the Authors (03 Jan 2015) Author's response Manuscript

ED: Reconsider after major revisions (19 Jan 2015) by Steven Brown

ED: Referee Nomination & Report Request started (22 Jan 2015) by Steven Brown

RR by Anonymous Referee #3 (31 Jan 2015)

Suggestions for revision or reasons for rejection

The paper has improved since the last round of revisions, but still requires further revision before it is acceptable for publication in ACP. The first part of the manuscript (excluding the abstract) that describes the data is better than the latter part that infers the dependence of ozone on NOx and VOCs. Since the authors do not present a photochemical model, the paper needs to be very clear that it cannot make recommendations about NOx and VOC control that are specific to Guangzhou, although it can make comparisons to other locations. Statements about the specific NOx and VOC regime in the latter part of the paper need to be revised with this comparison in mind, but without stating that it has determined a control strategy for Guangzhou that is appropriate.

Page 3, line 6: Particulate matter rather than particulate matters (no “s”), and pollutant rather than pollutants (again, no “s”).

Page 3, line 17: “smog” rather than “somg”

Page 3, line 21-25: The reference here is now improved, but the statement about the 8:1 ratio is not as general as the sentence indicates. Suggest the following revision. “Although specific actual situations differ, numerical simulations in other polluted regions, such as Los Angeles in the 1980s, have shown ozone formation to be NOx limited when the VOC/NOx ratio is greater than 8:1 for VOC expressed as the mixing ratio of total carbon (Seinfeld 1989).”

Page 3, line 25-27: There is a great deal of literature examining observational relationships between VOC and NOx in O3 formation. See for example, Trainer, M., et al: Correlation Of Ozone With NOy In Photochemically Aged Air, Journal Of Geophysical Research-Atmospheres, 98, 2917-2925, 1993, which has been widely cited. Suggest simply omitting the sentence beginning with “However, little attention …”, since it is not needed to justify the work in this manuscript.

Page 4, line 7: If this is the first use of the term “MIR”, give the definition here. Also, since the authors do not actually use a photochemical model to calculate MIR but look it up from work on other places, replace “MIR” with “estimated MIR” here. Also see further comments on the use of MIR below.

Page 5, line 1: Suggest that “single measurement cell” is a better description than “single glass measuring tank”

Page 5, line 5: Replace “respectively” with “alternately”

Page 5, line 9: Add an appropriate reference to the conversion of higher oxides of nitrogen in molybdenum converters, such as: Dunlea, E. J., et al..: Evaluation of nitrogen dioxide chemiluminescence monitors in a polluted urban environment, Atmos. Chem. Phys., 7, 2691-2704, 10.5194/acp-7-2691-2007, 2007.

Page 5, line 36: Remove the phrase “you can see that”. Sentence should read simply “The correlation coefficient ranged …”

Page 5, line 40: A sentence is needed here to state that the instrument has not been intercompared with other VOC measurements, and that such an intercomparision is an important priority for future work.

Page 6, line 2: Remove the word “characteristic”, replace “relatively” with “readily” and remove “which displays the seasonal pattern” to read “The seasonal variation of ozone is readily apparent, being lower in the spring ….”

Page 6, line 11: Replace “showed” with “shown”

Page 6, line 19: Remove “in this”

Page 6, line 20: Replace “of that” with “the”, and change “may include” to “probably includes”

Page 6, line 26: Remove the word “variation”. Also, begin this sentence with an appropriate caveat about transport effects. “In the absence of transport effects, a negative derivative in ozone …”

Page 6, line 26: Low wind speed does not mean that transport is necessarily unimportant. The statement is not quantitative in any case, since “very small” is not defined, and the authors probably don’t have a means to make a quantitative statement here. Rather, the statement should read something to this effect.

“Although wind speeds are low at this site (1.4 m s 1 in different seasons, see Table 1), the effect of horizontal transport on daily variation in ozone is difficult to estimate and cannot be ruled out. Here, we interpret all variation as due to chemical production and loss, understanding that such interpretation is necessarily an upper limit to the effect of chemistry due to the likelihood of transport also influencing daily ozone variation.”

Page 6, line 32: remove “of variation”

Page 6, line 33: There is no measurement of OH radicals as the statement implies. The authors probably mean that sunlight begins to decrease at this time. Sentence should be clearer. “At 15:00, sunlight and associated photochemistry begin to decrease ….”

Page 6, line 37: remove “variation” at beginning of line. Replace “high OH radicals” with “increased photochemistry”

Page 6, bottom: First two sentences in section 3.2 are redundant. Omit the first one.

Page 6, line 15: Is it accurate to say “developed by Carter in model scenarios of Los Angeles in the 1980s” ? The MIR calculations must have some reference point that should be specified. If they are developed for a different city or time period, then give that reference point, but the reference must be stated.

Page 7, line 33: Remove the phrase “of the atmosphere at the sampling point.” End of this sentence should read “… alkanes and aromatics are shown to have the highest mixing ratio.”

Page 7, line 39. The point about alkenes and aromatics is important, but is consistent with findings from other locations. There should be a sentence and a reference to this effect. A good reference would be: Russell, A., Milford, J., Bergin, M. S., McBride, S., McNair, L., Yang, Y., Stockwell, W. R., and Croes, B.: Urban Ozone Control and Atmospheric Reactivity of Organic Gases, Science, 269, 491-495, 1995.

Page 8, line 11: The sentence is not consistent with itself. If the MIR varies with location, why should this demonstrate that it is accurate for Guangzhou? A better statement would be that the MIR values were not calculated for Guangzhou, so they may not be accurate.

Page 8, line 14: Too many significant figures. It is not possible that the ozone formation potential has been determined to four significant digits. Quote as either “approximately 30%” or “approximately 32%”.

Page 8, line 28: replace “were” with “are”

Page 8, line 28-32: the definition of the ozone isopleth is not clear. Please revise this sentence to state this more clearly. It is simply a two dimensional plot of the variation of peak ozone with VOC and NOx. Also, the Dodge 1977 reference is a difficult one to find and should be omitted.

Section 3.3, Figure 7, and associated discussion: The figure is difficult to interpret. There is no description of how the underlying VOC and NOx data are used to arrive at the contour lines on the figure, and as the authors state, there are other variables (mainly transport) beyond the VOC and NOx concentrations. There are 2 options.

1) Simplest: Simply omit this analysis and move on to the discussion of the remaining figures.

2) Average all the data in Areas 1, 2 and 3 to produce peak ozone vs. NOx for different ranges of VOCs and three single lines rather than as a contour plot. Any such plot will need to have the caveats in the text that NOx is an upper limit to actual NOx due to measurement artifacts on the molybdenum converter. If the authors choose this option, the paper will necessarily require an additional stage of review, further delaying publication. If the authors choose option (1), review may be shorter.

Page 9, line 20: New paragraph corresponding to the description of figure 8.

Page 9, line 33: “Fair” rather than “fare”. The sentence should reference the source of the number used for VOC to NOx ratio. “Simulations in Los Angeles from the 1980’s indicated a transition from VOC limited to NOx limited at a ratio of approximately 8:1 (Seinfeld, 1989). Although we have not performed specific modeling for Guangzhou, we use this approximate relationship to divide VOC and NOx limited regimes for Guangzhou.”

Page 9-10, discussion following the sentence modified in the last comment: The discussion on here must avoid explicitly stating that NOx control would be more effective than VOC control, since Guangzhou is not Los Angeles, and there is no photochemical modeling analysis in this paper that is specific to Guangzhou. Rephrase all conclusions to state more simply that Guangzhou falls in the NOx limited regime according to the simple comparison to Los Angeles, but that no definite conclusion can be reached about NOx control compared to VOC control in the absence of a photochemical model together with more accurate measurements of NOx.

Page 10, line 25: Replace “You could be seen in Figure 11” with “Figure 11 shows”

Hide

ED: Reconsider after major revisions (31 Jan 2015) by Steven Brown

AR by Xuejiao Deng on behalf of the Authors (27 Apr 2015) Author's response Manuscript

ED: Publish subject to technical corrections (18 May 2015) by Steven Brown

The manuscript has substantially improved. Most of the reviewer’s concerns have been addressed satisfactorily. Listed below are two minor comments the authors should consider in addition to the list of technical corrections.

Minor comments:

Page 8, lines 5-8: A redundant sentence that has no meaning. Sentence states that the MIR method is not reliable, with the reason that it has “possible uncertainty.” A better justification of the lack of reliability is required.

Page 8, line 27: The authors could consider replacing “no definitive conclusions can be drawn, in terms of limiting NOx, or VOCs to control ozone formation, in the absence of a photochemical model, together with more accurate measurements of NOx in the Guangzhou region.” With “the analysis is only approximate in terms of the assessment of VOC or NOx limitations for ozone control. A more accurate analysis would require a photochemical model and more accurate NOx measurements of NOx in the Guangzhou region.”

Technical Corrections:

Abstract, line 9: “the seasonal variation”

Page 3, line 2: “include but are not limited”

Page 3, line 19: “expressed as the concentration of carbon atoms”

Page 4, line 29: replace “homogenized from various” with “homogenous with respect to various”

Page 5, line 34: Define the meaning of outliers. How large or small were the values that were eliminated?

Page 5, line 36: remove “that”, “and such an intercomparison”

Page 6, line 17: remove “is”

Page 6, line 23 and later: replace “change rate” with “rate of change” or “derivative”

Page 6, line 24: remove “the fact”, “indicates that the”

Page 6, line 33: replace “there were no OH radical present” with “there was no sunlight present to generate OH radicals”

Page 8, line 10: “such that the application of MIR may be only approximate in Guangzhou”

Hide

AR by Xuejiao Deng on behalf of the Authors (22 May 2015) Author's response Manuscript