Summertime nitrate aerosol in the upper troposphere  and lower stratosphere over the Tibetan Plateau  and the South Asian summer monsoon region

Gu, Yixuan; Liao, Hong; Bian, Jianchun

doi:https://doi.org/10.5194/acp-16-6641-2016

Articles | Volume 16, issue 11

https://doi.org/10.5194/acp-16-6641-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

Special issue:

Study of ozone, aerosols and radiation over the Tibetan Plateau...

https://doi.org/10.5194/acp-16-6641-2016

© Author(s) 2016. This work is distributed under
the Creative Commons Attribution 3.0 License.

Articles | Volume 16, issue 11

Research article

|

02 Jun 2016

Research article |

| 02 Jun 2016

Summertime nitrate aerosol in the upper troposphere and lower stratosphere over the Tibetan Plateau and the South Asian summer monsoon region

Yixuan Gu, Hong Liao, and Jianchun Bian

Download

Final revised paper (published on 02 Jun 2016)
Supplement to the final revised paper
Preprint (discussion started on 13 Nov 2015)
Supplement to the preprint

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC C10193: 'Referee comments', Anonymous Referee #2, 05 Dec 2015
- AC C12084: 'Responses to Reviewer #2', Hong Liao, 29 Jan 2016
RC C10245: 'Review of acp-2015-793', Anonymous Referee #1, 07 Dec 2015
- AC C12083: 'Responses to Reviewer #1', Hong Liao, 29 Jan 2016

Peer-review completion

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

AR by Hong Liao on behalf of the Authors (29 Jan 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (01 Feb 2016) by Xiaobin Xu

RR by Anonymous Referee #2 (16 Feb 2016)

Suggestions for revision or reasons for rejection

The authors have made an effort to address the questions raised by the reviewers. The model sensitivity tests as performed by the authors should be helpful for further understanding the sources and formation mechanisms of nitrate aerosols in the UTLS over the TP/SASM region. The authors’ responses are generally fine, but additional issues arising need to be addressed further. Below are my specific comments and suggestions.

It is interesting to see that compared to surface NO3-, nitrate aerosols in the UTLS over the TP/SASM region are not so sensitive to a 50% reduction of anthropogenic NOx emissions in Asia. What could one learn from this result with regards to the sources and formation mechanism of nitrate aerosols (and their gas precursors) in the UTLS over the TP/SASM region? Specifically, does it mean that anthropogenic NOx emissions in Asia (including South and Southeast Asia) do not have large impacts on nitrate aerosols in the UTLS over the TP /SASM region? The authors may want to give a short discussion on the implications of this sensitivity result, instead of merely arguing that nitrate is more important than sulfate for the UTLS aerosol layer even under a situation of 50% NOx emission reduction. The authors are suggested to use a different title for Sect. 7, .e.g. “Sensitivities of simulated nitrate in the UTLS to anthropogenic NOx, NH3 and SO2 emissions in Asia”.

PM2.5 is an important parameter usually used for air quality study, and it might not have to be used to describe the aerosols in the UTLS. Was PM2.5 simulated as a tracer in GEOS-Chem or it was calculated just by summing simulated SO4=, NO3-, NH4+, BC and OC? Mineral dust and sea salt aerosols, which were simulated and used for the calculation of extinction coefficient in this study, should also be taken into account for PM2.5. Otherwise, more clear definition/descriptions should be given in the text. In the abstract (Line37-38), for example, it might be written as “defined as the sum of sulfate, nitrate, ammonium, black carbon, and organic carbon aerosols in this study”. In Fig. 7, the sub-panels for PM2.5 can be omitted, and instead the distributions of mineral dust and sea salt aerosols might be displayed.

The physical, chemical and dynamical processes related to the formation of the UTLS aerosol layer are complex and need further investigations in future studies. Therefore, I would suggest that a more detailed description of the aerosol module used in this study be given for the convenience of comparisons of future work with this study. How many size bins or modes of the aerosols were adopted in the GEOS-Chem used in this study? What ionic aerosols were considered in the ISORROPIA II used in this study, NH4+/SO4=/NO3-, or NH4+/Na+/K+/Ca2+/Ma2+/SO4=/NO3-/Cl-/CO3=, or others? Were the aerosols treated as internal, or external mixed, or both (the same question as in my first round review)? One can see from Table 3 that OC and BC concentrations are constant for different simulations, indicating that they are external mixed.

With respect to the formation mechanisms for SO4= and NO3- in the UTLS (Line 547-553), simple comparison of the changes in the gas-phase oxidation of SO2 and the gas-particle equilibria of NO3- with temperature (altitude) might not be sufficient to explain the difference in their vertical distributions. Note that the reaction of NO2 (not NO as written in Line 546) with OH also changes with temperature, the same as for the reaction of SO2 with OH. Was H2SO4 simulated in the GEOS-Chem? If so, comparisons of both the formation and nucleation rates of H2SO4 and HNO3 might be helpful. It seems that the gas-phase oxidation of SO2 was not addressed in the work of X.Y. Zhang et al. (2012), not mentioned to the change of its reaction rate with temperature. Therefore, this work should not be cited here (Ling 550).

Hide

RR by Anonymous Referee #3 (17 Feb 2016)

Suggestions for revision or reasons for rejection

The paper "Summertime nitrate aerosol in the upper troposphere and lower stratosphere over the Tibetan Plateau and the South Asian summer monsoon region" by Gu et al. describes results from global model simulations of aerosol chemical composition over the Tibetean plateau.
However, the title is misleading as not only UTLS but also surface concentrations are analysed and compared with observations, which are a substantial contribution to the total manuscript.

Generally, the paper is reasonably well written and provides a basic evaluation of the aerosol chemical composition with a focus on aerosol nitrate over the Tibetean plateau. However, the applied horizontal
resolution of the global model, i.e. 2° by 2.5° is quite coarse for a focus on a specific region, such that no detailed information on the regional distribution over the TP can be addressed properly.

Overall, the paper can be published after addressing the major (and minor) points below:

There are three major points in the current manuscript, which inhibit publication in the current form. These are:

1) The effect of nitrate aerosols is difficult to address from a simulation with and without nitrate as the chemical regime of the aerosol thermodynamic partitioning depend strongly on the availability
of nitrate. Consequently, a completely different aerosol composition is derived, and from this a modified aerosol microphysical processing results as well as properties of the size distribution with implications for aerosol extinction. Consequently, this whole part should better be addressed by conclusions drawn from the experiments with a modified emission strength (+- 20%), as this will not completely alter the chemical regime of the thermodynamics.
The comparison with SAGE data for extinction suffers from the same problem. However, total extinction is reasonably well matched. A key question is whether the model is able to represent the stratospheric
aerosol layer (originating mostly from COS oxidation and to a minor extent also anthropogenic SO2), as this will contribute to a major extent to the aerosol layer in the UTLS.

2) The other key problem is the comparison with satellite data with a coarse vertical resolution. As HNO3 (as well as ozone) has an increasing gradient from the troposphere to the stratosphere, the detection from the satellite could easily be influenced by the upper part of the sampling layers which might be dominated by the stratospheric air masses.
This is e.g. obvious from the Figures 4 and 5, which are clearly influenced by the differences in tropospheric and stratospheric air masses at the given altitude. However, also in Fig. 7 there is only a very reduced impact of surface emissions or lifting patterns along the TP or the SASM visible, but the distribution looks basically as influenced by the tropopause distinguishing tropospheric and stratospheric air masses (comparable to a plot of potential vorticity).

3) The authors claim that vertical transport of nitrate (as it is the second most abundant aerosol compound in near surface air) is responsible for enhanced values in the UT. This cannot be true, as
the vertical transport cannot change the ratios of sulphate to nitrate by transport alone; the transport algorithms usually transport all compounds in the same way.
Precipitation scavenging of aerosol species is controlled by their size and the hygroscopic properties of nitrate aerosols are so close to those of sulphate particles such that also cloud activation cannot play a role in altering the ratios of nitrate to sulphate either.
Both gaseous HNO3 and H2SO4 are so soluble that in case of precipitation events, the concentrations
would be equally reduced to almost zero. SO2 which is slightly less soluble might have a larger chance of reaching the UT without a complete scavenging, but as seen from previous studies UTLS sulphate
mostly originates from COS and other long lived species.
Consequently, convective transport of nitrate has to be ruled out almost completely to lead to the enhanced nitrate concentrations, even though the authors attribute a substantial fraction to this process
(see e.g. Fig.13).
STE might offer an alternative pathway of HNO3 entering the UT from above, but most likely the HNO3 is formed in situ by NOx chemistry or by the release of NOy from long lived, hardly soluble compounds
such as PAN. However, chemical in situ production is missing completely in Fig.13 and the corresponding discussion.

Other points:
a)
The comparison with ozone soundings does not really provide a good agreement (except for the general shape). Tropospheric O3 is substantially deviating (factor of 2) from the observations which on the
given scale. Therefore, it is misleading that the agreement is good, at least not below 12 km altitude.

b)
The authors do not discuss whether the S(VI) is in from of sulphate or bisulphate (both close the surface and in the UT) and how the ammonium is distributed between S(VI) and nitrate.
Are other cations available (K+, Mg+, Ca2+)? What are their concentrations in the model simulations?

c) Is there a reasonable correlation between enhanced nitrate and low temperature together with high RH? The patterns of the two meteorological quantities are very smoothed out, such that this
correlation (even though generally valid from a thermodynamically viewpoint) is not necessarily driving enhanced nitrate concentrations.

Hide

ED: Reconsider after major revisions (25 Feb 2016) by Xiaobin Xu

AR by Hong Liao on behalf of the Authors (09 Mar 2016) Author's response Manuscript

ED: Referee Nomination & Report Request started (11 Mar 2016) by Xiaobin Xu

RR by Anonymous Referee #3 (26 Mar 2016)

Suggestions for revision or reasons for rejection

Before a publication some major points must be addressed (and modified in the manuscript accordingly):

The authors have partially responded to many concerns in their authors replies, but have not substantially modified the manuscript. Therefore, the paper still not very much improved to readers not analysing all the discussion concerns and responses.

The most prominent example is the role of chemical production to the nitrate formation. In the current manuscript version, this is still not well discussed and gas phase HNO3 production and gas aerosol partitioning are not treated separately, such that it cannot be concluded properly whether the meteorological influences on the chemical reaction system or the altered thermodynamical stability of NH4NO3 in the UT dominate the importance of nitrate over the TP. This could be elucidated by e.g. analysing the ratio of particulate nitrate to total oxidised nitrogen (N(V), i.e. gaseous HNO3, N2O5, NO3, and particulate nitrate).
Furthermore, upward transport is still one of the dominant processes according to the authors. Even though they mention deep convection, they use ECMWF vertical winds for displaying ascent of air masses, which does not represent convective activity. And as deep convection is mostly associated with precipitation, HNO3 and aerosol particles would be efficiently removed. This might also contribute to the underestimated OC and BC concentrations in the simulations for the UT, i.e. too strong wet removal. Therefore, such a strong contribution to UTLS nitrate from the lower tropopause appears unlikely to the reviewer. It would be much more easy to explain the enhanced simulated nitrate by the vertical transport acting on low solubility compounds such as NOx, which are then converted in the UTLS by chemical reactions into N(V).

The treatment of aerosol species as an external mixture for the extinction calculation is not very reasonable as most aerosol particles containing nitrate will be in the form of solution doplets – therefore the contribution from nitrate cannot such easily be estimated as stated by the authors and their result can only be considered as a rough approximation.

The authors mention that MLS data is assumed to have uncertainties which are of the same magnitude as the actual values – therefore a comparison of HNO3 values of the magnitude of the uncertainty limits is of reduced value. The MLS data is much more trustful in the stratosphere, where higher mixing ratios are found. The questioned aspect of vertical resolution of the data (i.e. 3 to 4 km) is not discussed at all, such that especially at 100hPa stratospheric air masses can influence the observations.
Did the authors use the same averaging kernels for the model data as the MLS retrieval? If not, this alone makes it dificult to compare model results with satellite data and the authors did not mention how the model-observation comparison is done in detail.
The geographical distribution of the MLS data with simulated HNO3 does not provide a good pattern correlation, except for the main features, which are characterised by the differentation of tropospheric and stratospheric air masses, e.g., there is a simulated minimum over the Gulf of Bengal which does not show up in the data. For the 200 hPa data over the TP there is hardly any data.

The surface data points for the various aerosol compounds (Fig. 8) contains a lot of data from Southern India which are of minor importance to the Tibetean plateau but more to the SASM region. As these are two different regions a separation of the data points would be better (even though this reduces the number of data points for a potential regression). Anyhow, this comparison does not show a very good model representation of the measurements, which the authors attribute to different years and other factors.

Concluding, I am not fully convinced that the manuscript is worth publication, as there is hardly any innovative, conclusive content in the paper. I would merely see this as (one of many) modelling studies covering Southeast Asia including a comparison with measurements, which does not show a very good model performance.

Minor (language and formulation points):
196: irons → ions
699-700: in from of → in form of
222: citation should not contain first names
244-245: NH3 emissions follow the global inventory described in Fisher et al.

Hide

RR by Anonymous Referee #2 (28 Mar 2016)

ED: Reconsider after major revisions (28 Mar 2016) by Xiaobin Xu

AR by Hong Liao on behalf of the Authors (19 Apr 2016) Author's response Manuscript

ED: Reconsider after minor revisions (Editor review) (28 Apr 2016) by Xiaobin Xu

AR by Hong Liao on behalf of the Authors (08 May 2016) Author's response Manuscript

ED: Publish as is (15 May 2016) by Xiaobin Xu

Download

Article (4126 KB)
Full-text XML

Short summary

This is the first study to examine nitrate aerosol in the upper troposphere and lower stratosphere (UTLS) over the Tibetan Plateau (TP) and the South Asian summer monsoon (SASM) region in summer. Nitrate aerosol is simulated to be the most dominant aerosol species in the UTLS over the studied region. The mechanisms for the accumulation of nitrate in the UTLS over the TP/SASM region include vertical transport and the gas-to-aerosol conversion of nitric acid to form nitrate.