the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 23 Jan 2023
Modeling SOA contributions of VOC, IVOC and SVOC emissions and large uncertainties associated with OA aging
Abstract. Secondary organic aerosols (SOA) are an important component of atmospheric fine particulate matter (PM2.5) in China, and elsewhere, with contributions from anthropogenic and biogenic volatile organic compounds (AVOC and BVOC) and semi- (SVOC) and intermediate volatility organic compounds (IVOC). Policy makers need to know which SOA precursors are important but accurate simulation of SOA magnitude and contributions remains uncertain. We reviewed SOA modelling studies in the past decade that have reported the relative contributions of different precursors to SOA concentration and the findings have many inconsistencies due to differing emission inventory methodologies/assumptions, air quality model (AQM) algorithms, and other aspects of study methodologies. We investigated the role of different AQM SOA algorithms by applying two commonly used models, CAMx and CMAQ, with consistent emission inventories to simulate SOA concentrations and contributions for July and November 2018 in China. Both models have a volatility basis set (VBS) SOA algorithm but with different parameters and treatments of SOA photochemical aging. BSOA (SOA produced from BVOC) is found to be more important over southern China whereas SOA generated from anthropogenic precursors is more prevalent in the North China Plain (NCP), Yangtze River Delta (YRD), Sichuan Basin and Central China. Both models indicate negligible SOA formation from SVOC emissions as compared to other precursors. In July when BVOC emissions are abundant, SOA is predominantly contributed by BSOA (except for NCP), followed by IVOC-SOA (i.e. SOA produced from IVOC) and ASOA (i.e. SOA produced from anthropogenic VOC). In contrast in November, IVOC becomes the leading SOA contributor for all selected regions except PRD, illustrating the important contribution of IVOC emissions to SOA formation. Therefore, future control policies should aim at reducing IVOC emissions as well as traditional VOC emissions.
While both models generally agree in terms of the spatial distributions and seasonal variations of different SOA components, CMAQ tends to predict higher BSOA while CAMx generates higher ASOA concentrations. As a result, CMAQ results suggest that BSOA concentration is always higher than ASOA in November while CAMx emphasizes the importance of ASOA. Utilizing a conceptual model, we found that different treatment of SOA aging between the two models is a major cause of differences in simulated ASOA concentrations. The step-wise SOA aging scheme implemented in CAMx (based on gas-phase reactions with OH radical and similar to other models) exhibits a strong enhancement effect on simulated ASOA concentrations and this effect increases with the ambient OA concentrations. The CMAQ VBS implements a different SOA aging scheme that represents particle-phase oligomerization and has smaller impacts, or no impact, on total OA. A brief literature survey shows that different structure and/or parameters of the SOA aging schemes are being used in current models, which could greatly affect model simulations of OA in ways that are difficult to anticipate. Our results indicate that large uncertainties still exist in the simulation of SOA in current air quality models due to the aging schemes as well as uncertainties of the emission inventory. More sophisticated measurement data and/or chamber experiments are needed to better characterize SOA aging and constrain model parameterizations.
- Preprint
(1910 KB) - Metadata XML
-
Supplement
(1051 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2022-1502', Anonymous Referee #1, 14 Feb 2023
General comments:
The manuscript of egusphere-2022-1502 presented the model comparison for OA. Generally, I am impressed by the low presentation quality in this manuscript (especially the introduction section is not well organized) and cannot fully follow the authors’ intention. I have judged to reject this manuscript and encourage the resubmission after revisions. I hope the following major and specific comments will help to improve this manuscript.
Major comments:
- Introduction: I would like to request a more organized and reader-friendly introduction. Please see my specific comments.
- Model configuration: Despite the discussion of the importance of VOC emissions, the simulation period is 2018 whereas the MEIC emission inventory in China is 2017. I can understand the importance to understand the SOA modeling framework; however, emission status can be much more significant. In addition, S/IVOC emissions considered in this study targeted 2016. This also causes a difference in the simulation period, and also differ in the MEIC inventory. Is it ignoring the difference? I do not think so. Why the different year was applied and how can we understand the result caused by this difference?
Specific comments:
- P1, L17-18: I feel AVOC and BVOC are sources whereas SVOC and IVOC are emission statuses, am I correct? Please clearly state this point. Do they simply summarize as “precursors”? In the case of BVOC, how can we suggest to policy makers? That is why the latter part of the abstract might introduce some confusion.
- P1, L32-33: “ASOA (SOA produced from anthropogenic VOC)” should be defined in L28.
- P1, L36: I cannot follow the wording “traditional VOC emissions” within this abstract.
- P1, L45: “OA” is not defined in the abstract.
- P2, L51-51: “more sophisticated measurement data and/or chamber experiments” for what? SOA aging? This sentence is needed to be rewritten.
- P2, L56-58: Where was indicated for these emission reductions, entire the world or China? It also needs the appropriate references to support them.
- P2, L58-61: However, this reference was older than 10 years. Does this support the increasing importance of OA?
- P2, L63 or L65: No explanation for POA.
- P2, L66: This 2ndparagraph in the introduction section is required to be reorganized. I would like to suggest just state emissions. The latter part of this paragraph contains other topics. Please also see the following specific comments.
- P3, L75: “missing SOA precursors” is ambiguous.
- P3, L88: This sentence makes confusion. Does this CAMx model have more reliability to represent SOA modeling? As the authors introduced, the current models have two frameworks of two-product and VBS (from L93). Without the introduction to the modeling framework, I do not understand why this study was introduced abruptly.
- P3, L89-92: Again, what is the status of this WRF-Chem model? Why this study is introduced without a detailed introduction to the modeling framework? Moreover, this WRF-Chem model is not applied in this study. If the authors recognize this model, why this WRF-Chem model was not applied in this study?
- P3, L107-P4, L110: Is this DDM only applied in Houston, U.S.? If so, is it reliable to apply this method in China? I do not capture what is the intention to introduce this method.
- P4, L113-115: I am further confused because this sentence cited the other model of GEOS-Chem. As the authors explained two modeling approaches, this and the following discussion for the difference of SOA concentration in China are unclear. What is the model used in Miao et al. (2021), An et al. (2022), Chang et al. (2022), and Wu et al. (2021)? This is related to the comment of Section 3.1, and I strongly suggest to re-organize this part for the review of previous studies. It is very hard to follow all of them.
- P4, L127: Nevertheless of the introduction of WRF-Chem model and GEOS-Chem model, why the authors only applied two models of CAMx and CMAQ in this study? The reason is not clear.
- P4, L127-129: Repeatedly, we can easily find the application of WRF-Chem and GEOS-Chem model in China (also as shown in Table 2), so I do not figure out the approach taken in this study. Why only CAMx and CMAQ was applied? Is it enough to answer the authors’ motivation to clarify the SOA modeling?
- P7, L211-212: In the case of the inclusion of S/IVOC emissions, is it excluded a possibility of double-count in conventional VOC emissions?
- P8, Section 3.1 and Table 2: This is not the results and discussions in this study. I would like to suggest shortening this section to highlight the points related to this study, and then moving this section to the introduction or supplemental materials. Moreover, this review contains both modeling approaches and uncertainty in emissions. It is much more readable to possibly divide based on these two viewpoints.
- P11, Section 3.2: I do not agree with the organization of this section. The authors applied the step-by-step scenarios as _base, _IVOC, and _S/IVOC; hence, the model evaluation should also be step-by-step. Why only _S/IVOC results are shown? We can expect the improvement of modeling performance in _S/IVOC rather than _base and _IVOC; however, if there is a degradation in modeling performances, what brings this study approach?
- P11, Figure 2: What indicates a high concentration in the bottom-left corner in (c) and (d)? Is it long-range transport? But southwest China posed a lower level of PM2.5. Is it a high concentration in other countries? Why the map was not shown? This figure itself posed confusion.
- P11, L311-P12, L315: In this context, CAMx and CMAQ mean CAMx_S/IVOC and CMAQ_S/IVOC listed in Table 1, right? It is better to explicitly identify the model name (like L294).
- P12, L324 and hereafter: The wording such as “CMAQ PRD” and “CAMx Sichuan Basin” are not understandable. Please rewrite these wording.
- P12, L334-335: We can expect this BVOC status, but cannot fully understand it as input data in this study. I think the need to explicitly show the emission itself used in this study.
- P12, L331: In this subsection, the relative percentage was well discussed; however, we do not follow these values from Fig. 3. I would like to suggest preparing the supplemental figure for the relative percentage (as spatial distribution).
- P13, L361-365: This discussion might be understandable; however, it could be confirmed by applying the older version of CMAQ. I can partly agree that the scope of this study is on anthropogenic emissions; however, the portion of anthropogenic and biogenic are important points as presented in this study. Does this also imply that the model framework and configuration in CMAQ can cause differences in anthropogenic/biogenic sources?
- P14, Figure 3: Same point for Figure 2. What is a high concentration outside China?
- P15, Figs. 4 and 5: The light blue color seems to be out of alignment. Is it a corrupted figure? Please confirm. The wording “IVOCs” and “SVOCs” did not correspond to the main text. It is better to unify.
- P16, L390: Again, do CAMx and CMAQ means CAMx_S/IVOC and CMAQ_S/IVOC listed in Table 1? It should be explicitly mentioned.
- P16, L397: We do not follow the detail of this RAQMS model from this manuscript.
- P17, L430: What are the explicit definitions for “high-” and “low-” NOx conditions? Are there some values to divide them?
- P17, L432: I do not fully understand the wording “outside the models”. Is it a standalone SOA model from CAMx and CMAQ? How can these be outside?
- P17, L443: “CMAQ_no-aging” is unlabeled “CMAQ” in Fig. 7? It is better to be unified.
- P17, L448: To be consistent with Fig. 6, this should be “CMAQ_OLIG”. Which is correct?
- P17, L462: There is no Table 6. Does this mean “Figure 6”?
- P18, L460: According to this sentence, N in Eq. (1) is 5?
Technical corrections:
- P3, L87: No need to use the parenthesis because this reference was used as a subject.
- P4, L117: The typo of “Miao’s (2021)”?
- P9, L272: I do not find the reference of Miao et al. (2017), Typo in a year?
Citation: https://doi.org/10.5194/egusphere-2022-1502-RC1 -
CC1: 'Reply on RC1 (initial reply)', Ling Huang, 17 Feb 2023
The authors thank the reviewer for reading the manuscript carefully and providing helpful comments. Our study applied two commonly used air quality models to simulate SOA and its components over China and one of our major findings is that different implementations of the VBS schemes can produce substantially different SOA due to treatment of photochemical aging. We read through all the comments and provide an initial response to address the reviewer’s major concerns. We will update the manuscript with improvements suggested by the reviewer’s comments at a later date when other comments can be considered.
After reading all the comments, we feel that the reviewer is most concerned with the following three aspects:
1. Definition of AVOC, BVOC, IVOC and SVOC
Among the specific comments, we found that there are a few comments on AVOC, BVOC, IVOC and SVOC, e.g.:
- “P1, L17-18: I feel AVOC and BVOC are sources whereas SVOC and IVOC are emission statuses, am I correct? Please clearly state this point. Do they simply summarize as “precursors”?”
- “P1, L36: I cannot follow the wording “traditional VOC emissions” within this abstract.”
- “P3, L75: “missing SOA precursors” is ambiguous.”
- “P7, L211-212: In the case of the inclusion of S/IVOC emissions, is it excluded a possibility of double-count in conventional VOC emissions?”
The classification of VOC, IVOC, and SVOC is usually defined according to the effective saturation concentration (e.g. Pye and Seinfeld, 2010; Woody et al. 2015; Lu et al. 2018). Based on the effective saturation concentration C*, organic compounds are usually classified as VOC (C*= 107~1011 µg/m3), IVOC (C*= 103~106 µg/m3), and SVOC (C*= 100~102 µg/m3). VOC and IVOC are predominantly in the gas phase whereas SVOC can partition between gas and particle phases depending on temperature and total organic aerosol loading. Classifying VOC as biogenic (BVOC, mostly emitted by vegetation) or anthropogenic (AVOC, e.g. industrial solvent usage, vehicle exhaust) is useful for air quality management and policy. Traditionally, VOC emission inventories have omitted S/IVOC emissions (Robinson et al. 2007) primarily because the emission factor measurement techniques classified organic emissions either as gasses (VOC) or particles (PM) without considering intermediate cases (S/IVOC). Accordingly, S/IVOC emissions are sometimes referred as “non-traditional SOA” (Woody et al. 2015). There is potential for double counting organic emissions (e.g., between SVOC and PM emissions) which must be considered in emission inventory development (Wu et al. 2021) which occurred outside the current study. Our study considers IVOC and SVOC from anthropogenic sources but we did not label these emissions ASVOC and AIVOC for simplicity and to be consistent with many other published studies.
2. Confusion between air quality models (AQM) and SOA modeling frameworks
There are many comments that are related to different air quality models (e.g. WRF-Chem, RAQMS, etc.), as listed below:
- “P16, L397: We do not follow the detail of this RAQMS model from this manuscript.”
- “P3, L89-92: Again, what is the status of this WRF-Chem model? Why this study is introduced without a detailed introduction to the modeling framework? Moreover, this WRF-Chem model is not applied in this study. If the authors recognize this model, why this WRF-Chem model was not applied in this study?”
- “P4, L113-115: I am further confused because this sentence cited the other model of GEOS-Chem. As the authors explained two modeling approaches, this and the following discussion for the difference of SOA concentration in China are unclear. What is the model used in Miao et al. (2021), An et al. (2022), Chang et al. (2022), and Wu et al. (2021)? This is related to the comment of Section 3.1, and I strongly suggest to re-organize this part for the review of previous studies. It is very hard to follow all of them.”
- “P4, L127: Nevertheless of the introduction of WRF-Chem model and GEOS-Chem model, why the authors only applied two models of CAMx and CMAQ in this study? The reason is not clear.”
- “P4, L127-129: Repeatedly, we can easily find the application of WRF-Chem and GEOS-Chem model in China (also as shown in Table 2), so I do not figure out the approach taken in this study. Why only CAMx and CMAQ was applied? Is it enough to answer the authors’ motivation to clarify the SOA modeling?”
- “P16, L397: We do not follow the detail of this RAQMS model from this manuscript.”
Air quality models (e.g. CAMx, CMAQ, WRF-Chem, GEOS-Chem, RAQMS) contain many component algorithms to represent distinct physical (e.g. transport, diffusion, deposition) and chemical (e.g. photochemistry, aqueous chemistry, and heterogeneous reactions) processes, e.g., several models may choose a VBS scheme to represent SOA chemistry but that does not mean that all VBS schemes are identical. Our study focusses on the SOA schemes rather than the host air quality model. The two-product approach and VBS approach are two most widely recognized SOA modeling framework. By choosing CAMx and CMAQ, which are frequently applied in China due to their source apportionment features (PSAT in CAMx and ISAM in CMAQ), we were able to compare a two-product scheme with VBS and compare two different VBS schemes. This approach met our study objective of focusing on SOA schemes and produced clear findings on the importance of SOA aging assumptions and semi-volatile POA emissions.
3. Different emission inventory and modeling year
The reviewer asks why our simulation year (2018) and emission inventory year (2017) are different and the reason is data availability. The observed OC/EC dataset that we used for model evaluation is for 2018 and earlier data were not available. However, emission inventory data (MEIC and S/IVOC) for 2018 were not available and the most recent year was 2017. We agree with the reviewer that it would be best to have consistent emission inventory for the modeling year. While recognizing that using the same emission inventory and modeling years would be ideal, the uncertainties in the emission inventory itself are almost certainly larger changes in emissions from 2017 to 2018. Therefore, we conclude that this difference does not substantially influence our findings.
References
Lu, Q., Zhao, Y., & Robinson, A. L. (2018). Comprehensive organic emission profiles for gasoline, diesel, and gas-turbine engines including intermediate and semi-volatile<? xmltex\break?> organic compound emissions. Atmospheric Chemistry and Physics, 18(23), 17637-17654.
Pye, H. O., & Seinfeld, J. H. (2010). A global perspective on aerosol from low-volatility organic compounds. Atmospheric Chemistry and Physics, 10(9), 4377-4401.
Woody, M. C., West, J. J., Jathar, S. H., Robinson, A. L., & Arunachalam, S. (2015). Estimates of non-traditional secondary organic aerosols from aircraft SVOC and IVOC emissions using CMAQ. Atmospheric Chemistry and Physics, 15(12), 6929-6942.
Robinson, A. L., Donahue, N. M., Shrivastava, M. K., Weitkamp, E. A., Sage, A. M., Grieshop, A. P., ... & Pandis, S. N. (2007). Rethinking organic aerosols: Semivolatile emissions and photochemical aging. Science, 315(5816), 1259-1262.
Wu, L., Ling, Z., Liu, H., Shao, M., Lu, S., Wu, L., & Wang, X. (2021). A gridded emission inventory of semi-volatile and intermediate volatility organic compounds in China. Science of the total environment, 761, 143295.
Citation: https://doi.org/10.5194/egusphere-2022-1502-CC1
-
RC2: 'Comment on egusphere-2022-1502', Anonymous Referee #2, 24 Feb 2023
This manuscript provides insight into model predictions of SOA over China using two common regional models: CAMx and CMAQ. The ability to find robust messages, such as the important role of IVOCs, across diverse model representations is useful. However, the current manuscript insufficiently describes the base models and should go farther in providing insight into what is well represented vs not.
Major comments:
- Provide a better description of CMAQ SOA model with appropriate citations.
- CMAQ AERO7 uses a VBS treatment for some systems, but not all. Specifically, many aqueous pathways and the “pcSOA” approach are not VBS style. Relabel the CMAQ treatment from “CMAQ VBS” to “CMAQ AERO7” throughout the manuscript to be more complete.
- Several SOA pathways in CMAQ are not mentioned and the current description does not adequately cite CMAQ developments (none of the biogenic SOA articles are cited, for example). Provide a more complete description of SOA in CMAQ vs CAMx and add references as appropriate (See https://www.epa.gov/cmaq/how-cite-cmaq). Consider:
- CMAQ IEPOX SOA approach (not cited)
- CMAQ oligomer approach (not cited)
- CMAQ organic nitrate SOA (not mentioned nor cited)
- CMAQ monoterpene photooxidation SOA (mentioned but not cited)
- CMAQ semivolatile POA approach (could be better cited)
- CMAQ glyoxal/methylglyoxal SOA approach (neither mentioned nor cited)
- One of the conclusions is: “CMAQ tends to estimate higher BSOA concentration, while CAMx generates more ASOA.” You could mention that CMAQ has more pathways to SOA from biogenic precursors, including aqueous pathways that are not present in CAMx.
- Emissions and model choices (e.g., MEGAN) on lines 151-160 could also use literature citations.
- Is biogenic SOA in CAMx subject to VBS aging?
- Can the authors go further in determining whether parametrizations are realistic and what improvements might be needed?
- Figure 6: Is there experimental data to support one parameterization over another? Have you compared the CAMx aging scheme to more recent 2-D VBS parameterizations such as the work of Zhao et al. (2016)? Given 5.75 hours of aging could be captured in a chamber experiment is there data to confirm a 30x increase in yield? Is the aging scheme in CAMx plausible?
- The text on line 401 through 417 (differences in the CMAQ vs CAMx benzene system) should be refocused. The experimental data (Ng et al., 2007) demonstrates why there is only mass in the lowest C* bin for CMAQ—the SOA was observed to be nonvolatile in terms of yield behavior. Providing the total yield of C* 1000 ug/m3 and lower species in CAMx vs CMAQ confuses the story which is better captured on the following page. Table S9 indicates CMAQ would predict an SOA yield from benzene of 0.146 for all atmospherically relevant conditions. CAMx would predict an SOA yield of about 10% (loading of 10ug/m3 assumed) so the CAMx SOA yield for benzene before any multigenerational aging is lower than CMAQ (later shown in Fig 6). This suggests that it isn’t the initial benzene SOA yields (whether wall loss corrected are not) that are driving differences.
- Reword sentences to bring clarity and specificity. For example: “Our results indicate that large uncertainties still exist in the simulation of SOA in current air quality models due to the aging schemes as well as uncertainties of the emission inventory” can be reworded to: “Our results indicate aging schemes are the major driver in CMAQ vs CAMx treatments of ASOA and their resulting predicted mass.” (The role of emission inventories wasn’t specifically addressed and could be removed.)
- Can the regional model bias be used to help inform which representations are plausible? For example, what is the spatial pattern of bias? Since BSOA and ASOA have some spatial separation, how does performance very by model species? Observations could be added to Fig 4.
- The authors map IVOC emissions in CMAQ to the pcSOA precursor (pcVOC). From Murphy et al. (2017): “We further introduce a new surrogate species, potential SOA from combustion emissions (pcSOA) to account for missing mass from IVOC oxidation, multigenerational aging of (anthropogenic) secondary organic vapors (from IVOC and VOC precursors), biases in SOA yields from vapor wall losses, and enhanced organic partitioning to the condensed aqueous phase. In addition to these sources, pcSOA could account for mass from oxidation of as-yet unidentified sources of SOA precursors.” Are IVOCs a good fit for the pcSOA precursor? How much does the emission magnitude of IVOCs differ from what Murphy et al. proposed as the emission rate (which was not IVOC specific)? How does the yield of SOA from pcSOA compare to that expected for IVOCs?
Minor comments:
- Clarify your definition of biogenic (BSOA and BVOC). Are BVOCs strictly from vegetation or defined as specific VOCs such as isoprene and monoterpenes? If BVOCs are defined based on isoprene or monoterpene identity, please highlight that anthropogenic monoterpene emissions can be substantial (Coggon et al., 2021) and anthropogenic NOx modulates monoterpene SOA (Pye et al., 2015) and thus biogenic does not mean the SOA is entirely biogenic.
- Line 101: What is meant by brute-force SOA estimation? Is that a zero out?
- Line 125: I recommend removing “No clear conclusions can be drawn.” Often, the different results reflect different model parameterizations. The reason they are giving different answers can be (at least partially) identified.
- Section 2.2: Include a brief overview of how emissions from previous were developed (were they scaled to POA)?
- Line 193: Are these percents of total S/IVOC or total VOC?
- The 1.6 OM/OC ratio attributed to Feng et al. is actually from Turpin and Lim 2001. The value seems a bit low considering primary wood burning emissions often have OM/OC ratios of 1.7. Consider updating the OM/OC from 1.6 to a more recent value. Alternatively, model output can be converted to OC as the model often has a specific molecular weight and other properties assigned to the species. CMAQ specifies OM/OC ratios in the Species Definition files supplied with the model (https://github.com/USEPA/CMAQ/blob/5.3.2/CCTM/src/MECHS/cb6r3_ae7_aq/SpecDef_cb6r3_ae7_aq.txt).
- Section 3.1: Experimental data to feed parameterizations has increased in concentration range over time which is what allows a greater range of volatility to be fit in the VBS vs older data sets. Similarly, older data tended to be from experiments from very high loading which made extrapolation to ambient atmospheres more difficult and likely drove errors. Consider adding this context.
- Line 364: Cite peer-reviewed original references rather than model release notes.
- Line 417: Remove personal communication citation. The CMAQ benzene yields can be traced back to experimental data which indicates if vapor wall loss was performed.
- Figure 6: Add aging time of 5.75 hours to caption.
- At least one critical reference is missing from Table 1 (Zhao et al., 2016).
- Reword the citation on line 64-65—the reason for the association with SOA and mortality has not been determined.
- Table 1 could be moved to the SI. Also consider relabeling as some figure labels use the Table 1 labels (Fig 2) and others do not (Fig 3).
References
Coggon, M. M., Gkatzelis, G. I., McDonald, B. C., Gilman, J. B., Schwantes, R. H., Abuhassan, N., Aikin, K. C., Arend, M. F., Berkoff, T. A., Brown, S. S., Campos, T. L., Dickerson, R. R., Gronoff, G., Hurley, J. F., Isaacman-VanWertz, G., Koss, A. R., Li, M., McKeen, S. A., Moshary, F., Peischl, J., Pospisilova, V., Ren, X., Wilson, A., Wu, Y., Trainer, M., and Warneke, C.: Volatile chemical product emissions enhance ozone and modulate urban chemistry, P. Natl. Acad. Sci. USA, 118, e2026653118, https://doi.org/10.1073/pnas.2026653118, 2021.
Ng, N. L., Kroll, J. H., Chan, A. W. H., Chhabra, P. S., Flagan, R. C., and Seinfeld, J. H.: Secondary organic aerosol formation from m-xylene, toluene, and benzene, Atmos. Chem. Phys., 7, 3909-3922, https://doi.org/10.5194/acp-7-3909-2007, 2007.
Pye, H. O. T., Luecken, D. J., Xu, L., Boyd, C. M., Ng, N. L., Baker, K. R., Ayres, B. R., Bash, J. O., Baumann, K., Carter, W. P. L., Edgerton, E., Fry, J. L., Hutzell, W. T., Schwede, D. B., and Shepson, P. B.: Modeling the current and future roles of particulate organic nitrates in the southeastern United States, Environ. Sci. Technol., 49, 14195-14203, https://doi.org/10.1021/acs.est.5b03738, 2015.
Zhao, B., Wang, S., Donahue, N. M., Jathar, S. H., Huang, X., Wu, W., Hao, J., and Robinson, A. L.: Quantifying the effect of organic aerosol aging and intermediate-volatility emissions on regional-scale aerosol pollution in China, Sci. Rep., 6, 28815-28815, https://doi.org/10.1038/srep28815, 2016.
Citation: https://doi.org/10.5194/egusphere-2022-1502-RC2 - Provide a better description of CMAQ SOA model with appropriate citations.
-
AC1: 'Comment on egusphere-2022-1502', Li Li, 16 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1502/egusphere-2022-1502-AC1-supplement.pdf
Status: closed
-
RC1: 'Comment on egusphere-2022-1502', Anonymous Referee #1, 14 Feb 2023
General comments:
The manuscript of egusphere-2022-1502 presented the model comparison for OA. Generally, I am impressed by the low presentation quality in this manuscript (especially the introduction section is not well organized) and cannot fully follow the authors’ intention. I have judged to reject this manuscript and encourage the resubmission after revisions. I hope the following major and specific comments will help to improve this manuscript.
Major comments:
- Introduction: I would like to request a more organized and reader-friendly introduction. Please see my specific comments.
- Model configuration: Despite the discussion of the importance of VOC emissions, the simulation period is 2018 whereas the MEIC emission inventory in China is 2017. I can understand the importance to understand the SOA modeling framework; however, emission status can be much more significant. In addition, S/IVOC emissions considered in this study targeted 2016. This also causes a difference in the simulation period, and also differ in the MEIC inventory. Is it ignoring the difference? I do not think so. Why the different year was applied and how can we understand the result caused by this difference?
Specific comments:
- P1, L17-18: I feel AVOC and BVOC are sources whereas SVOC and IVOC are emission statuses, am I correct? Please clearly state this point. Do they simply summarize as “precursors”? In the case of BVOC, how can we suggest to policy makers? That is why the latter part of the abstract might introduce some confusion.
- P1, L32-33: “ASOA (SOA produced from anthropogenic VOC)” should be defined in L28.
- P1, L36: I cannot follow the wording “traditional VOC emissions” within this abstract.
- P1, L45: “OA” is not defined in the abstract.
- P2, L51-51: “more sophisticated measurement data and/or chamber experiments” for what? SOA aging? This sentence is needed to be rewritten.
- P2, L56-58: Where was indicated for these emission reductions, entire the world or China? It also needs the appropriate references to support them.
- P2, L58-61: However, this reference was older than 10 years. Does this support the increasing importance of OA?
- P2, L63 or L65: No explanation for POA.
- P2, L66: This 2ndparagraph in the introduction section is required to be reorganized. I would like to suggest just state emissions. The latter part of this paragraph contains other topics. Please also see the following specific comments.
- P3, L75: “missing SOA precursors” is ambiguous.
- P3, L88: This sentence makes confusion. Does this CAMx model have more reliability to represent SOA modeling? As the authors introduced, the current models have two frameworks of two-product and VBS (from L93). Without the introduction to the modeling framework, I do not understand why this study was introduced abruptly.
- P3, L89-92: Again, what is the status of this WRF-Chem model? Why this study is introduced without a detailed introduction to the modeling framework? Moreover, this WRF-Chem model is not applied in this study. If the authors recognize this model, why this WRF-Chem model was not applied in this study?
- P3, L107-P4, L110: Is this DDM only applied in Houston, U.S.? If so, is it reliable to apply this method in China? I do not capture what is the intention to introduce this method.
- P4, L113-115: I am further confused because this sentence cited the other model of GEOS-Chem. As the authors explained two modeling approaches, this and the following discussion for the difference of SOA concentration in China are unclear. What is the model used in Miao et al. (2021), An et al. (2022), Chang et al. (2022), and Wu et al. (2021)? This is related to the comment of Section 3.1, and I strongly suggest to re-organize this part for the review of previous studies. It is very hard to follow all of them.
- P4, L127: Nevertheless of the introduction of WRF-Chem model and GEOS-Chem model, why the authors only applied two models of CAMx and CMAQ in this study? The reason is not clear.
- P4, L127-129: Repeatedly, we can easily find the application of WRF-Chem and GEOS-Chem model in China (also as shown in Table 2), so I do not figure out the approach taken in this study. Why only CAMx and CMAQ was applied? Is it enough to answer the authors’ motivation to clarify the SOA modeling?
- P7, L211-212: In the case of the inclusion of S/IVOC emissions, is it excluded a possibility of double-count in conventional VOC emissions?
- P8, Section 3.1 and Table 2: This is not the results and discussions in this study. I would like to suggest shortening this section to highlight the points related to this study, and then moving this section to the introduction or supplemental materials. Moreover, this review contains both modeling approaches and uncertainty in emissions. It is much more readable to possibly divide based on these two viewpoints.
- P11, Section 3.2: I do not agree with the organization of this section. The authors applied the step-by-step scenarios as _base, _IVOC, and _S/IVOC; hence, the model evaluation should also be step-by-step. Why only _S/IVOC results are shown? We can expect the improvement of modeling performance in _S/IVOC rather than _base and _IVOC; however, if there is a degradation in modeling performances, what brings this study approach?
- P11, Figure 2: What indicates a high concentration in the bottom-left corner in (c) and (d)? Is it long-range transport? But southwest China posed a lower level of PM2.5. Is it a high concentration in other countries? Why the map was not shown? This figure itself posed confusion.
- P11, L311-P12, L315: In this context, CAMx and CMAQ mean CAMx_S/IVOC and CMAQ_S/IVOC listed in Table 1, right? It is better to explicitly identify the model name (like L294).
- P12, L324 and hereafter: The wording such as “CMAQ PRD” and “CAMx Sichuan Basin” are not understandable. Please rewrite these wording.
- P12, L334-335: We can expect this BVOC status, but cannot fully understand it as input data in this study. I think the need to explicitly show the emission itself used in this study.
- P12, L331: In this subsection, the relative percentage was well discussed; however, we do not follow these values from Fig. 3. I would like to suggest preparing the supplemental figure for the relative percentage (as spatial distribution).
- P13, L361-365: This discussion might be understandable; however, it could be confirmed by applying the older version of CMAQ. I can partly agree that the scope of this study is on anthropogenic emissions; however, the portion of anthropogenic and biogenic are important points as presented in this study. Does this also imply that the model framework and configuration in CMAQ can cause differences in anthropogenic/biogenic sources?
- P14, Figure 3: Same point for Figure 2. What is a high concentration outside China?
- P15, Figs. 4 and 5: The light blue color seems to be out of alignment. Is it a corrupted figure? Please confirm. The wording “IVOCs” and “SVOCs” did not correspond to the main text. It is better to unify.
- P16, L390: Again, do CAMx and CMAQ means CAMx_S/IVOC and CMAQ_S/IVOC listed in Table 1? It should be explicitly mentioned.
- P16, L397: We do not follow the detail of this RAQMS model from this manuscript.
- P17, L430: What are the explicit definitions for “high-” and “low-” NOx conditions? Are there some values to divide them?
- P17, L432: I do not fully understand the wording “outside the models”. Is it a standalone SOA model from CAMx and CMAQ? How can these be outside?
- P17, L443: “CMAQ_no-aging” is unlabeled “CMAQ” in Fig. 7? It is better to be unified.
- P17, L448: To be consistent with Fig. 6, this should be “CMAQ_OLIG”. Which is correct?
- P17, L462: There is no Table 6. Does this mean “Figure 6”?
- P18, L460: According to this sentence, N in Eq. (1) is 5?
Technical corrections:
- P3, L87: No need to use the parenthesis because this reference was used as a subject.
- P4, L117: The typo of “Miao’s (2021)”?
- P9, L272: I do not find the reference of Miao et al. (2017), Typo in a year?
Citation: https://doi.org/10.5194/egusphere-2022-1502-RC1 -
CC1: 'Reply on RC1 (initial reply)', Ling Huang, 17 Feb 2023
The authors thank the reviewer for reading the manuscript carefully and providing helpful comments. Our study applied two commonly used air quality models to simulate SOA and its components over China and one of our major findings is that different implementations of the VBS schemes can produce substantially different SOA due to treatment of photochemical aging. We read through all the comments and provide an initial response to address the reviewer’s major concerns. We will update the manuscript with improvements suggested by the reviewer’s comments at a later date when other comments can be considered.
After reading all the comments, we feel that the reviewer is most concerned with the following three aspects:
1. Definition of AVOC, BVOC, IVOC and SVOC
Among the specific comments, we found that there are a few comments on AVOC, BVOC, IVOC and SVOC, e.g.:
- “P1, L17-18: I feel AVOC and BVOC are sources whereas SVOC and IVOC are emission statuses, am I correct? Please clearly state this point. Do they simply summarize as “precursors”?”
- “P1, L36: I cannot follow the wording “traditional VOC emissions” within this abstract.”
- “P3, L75: “missing SOA precursors” is ambiguous.”
- “P7, L211-212: In the case of the inclusion of S/IVOC emissions, is it excluded a possibility of double-count in conventional VOC emissions?”
The classification of VOC, IVOC, and SVOC is usually defined according to the effective saturation concentration (e.g. Pye and Seinfeld, 2010; Woody et al. 2015; Lu et al. 2018). Based on the effective saturation concentration C*, organic compounds are usually classified as VOC (C*= 107~1011 µg/m3), IVOC (C*= 103~106 µg/m3), and SVOC (C*= 100~102 µg/m3). VOC and IVOC are predominantly in the gas phase whereas SVOC can partition between gas and particle phases depending on temperature and total organic aerosol loading. Classifying VOC as biogenic (BVOC, mostly emitted by vegetation) or anthropogenic (AVOC, e.g. industrial solvent usage, vehicle exhaust) is useful for air quality management and policy. Traditionally, VOC emission inventories have omitted S/IVOC emissions (Robinson et al. 2007) primarily because the emission factor measurement techniques classified organic emissions either as gasses (VOC) or particles (PM) without considering intermediate cases (S/IVOC). Accordingly, S/IVOC emissions are sometimes referred as “non-traditional SOA” (Woody et al. 2015). There is potential for double counting organic emissions (e.g., between SVOC and PM emissions) which must be considered in emission inventory development (Wu et al. 2021) which occurred outside the current study. Our study considers IVOC and SVOC from anthropogenic sources but we did not label these emissions ASVOC and AIVOC for simplicity and to be consistent with many other published studies.
2. Confusion between air quality models (AQM) and SOA modeling frameworks
There are many comments that are related to different air quality models (e.g. WRF-Chem, RAQMS, etc.), as listed below:
- “P16, L397: We do not follow the detail of this RAQMS model from this manuscript.”
- “P3, L89-92: Again, what is the status of this WRF-Chem model? Why this study is introduced without a detailed introduction to the modeling framework? Moreover, this WRF-Chem model is not applied in this study. If the authors recognize this model, why this WRF-Chem model was not applied in this study?”
- “P4, L113-115: I am further confused because this sentence cited the other model of GEOS-Chem. As the authors explained two modeling approaches, this and the following discussion for the difference of SOA concentration in China are unclear. What is the model used in Miao et al. (2021), An et al. (2022), Chang et al. (2022), and Wu et al. (2021)? This is related to the comment of Section 3.1, and I strongly suggest to re-organize this part for the review of previous studies. It is very hard to follow all of them.”
- “P4, L127: Nevertheless of the introduction of WRF-Chem model and GEOS-Chem model, why the authors only applied two models of CAMx and CMAQ in this study? The reason is not clear.”
- “P4, L127-129: Repeatedly, we can easily find the application of WRF-Chem and GEOS-Chem model in China (also as shown in Table 2), so I do not figure out the approach taken in this study. Why only CAMx and CMAQ was applied? Is it enough to answer the authors’ motivation to clarify the SOA modeling?”
- “P16, L397: We do not follow the detail of this RAQMS model from this manuscript.”
Air quality models (e.g. CAMx, CMAQ, WRF-Chem, GEOS-Chem, RAQMS) contain many component algorithms to represent distinct physical (e.g. transport, diffusion, deposition) and chemical (e.g. photochemistry, aqueous chemistry, and heterogeneous reactions) processes, e.g., several models may choose a VBS scheme to represent SOA chemistry but that does not mean that all VBS schemes are identical. Our study focusses on the SOA schemes rather than the host air quality model. The two-product approach and VBS approach are two most widely recognized SOA modeling framework. By choosing CAMx and CMAQ, which are frequently applied in China due to their source apportionment features (PSAT in CAMx and ISAM in CMAQ), we were able to compare a two-product scheme with VBS and compare two different VBS schemes. This approach met our study objective of focusing on SOA schemes and produced clear findings on the importance of SOA aging assumptions and semi-volatile POA emissions.
3. Different emission inventory and modeling year
The reviewer asks why our simulation year (2018) and emission inventory year (2017) are different and the reason is data availability. The observed OC/EC dataset that we used for model evaluation is for 2018 and earlier data were not available. However, emission inventory data (MEIC and S/IVOC) for 2018 were not available and the most recent year was 2017. We agree with the reviewer that it would be best to have consistent emission inventory for the modeling year. While recognizing that using the same emission inventory and modeling years would be ideal, the uncertainties in the emission inventory itself are almost certainly larger changes in emissions from 2017 to 2018. Therefore, we conclude that this difference does not substantially influence our findings.
References
Lu, Q., Zhao, Y., & Robinson, A. L. (2018). Comprehensive organic emission profiles for gasoline, diesel, and gas-turbine engines including intermediate and semi-volatile<? xmltex\break?> organic compound emissions. Atmospheric Chemistry and Physics, 18(23), 17637-17654.
Pye, H. O., & Seinfeld, J. H. (2010). A global perspective on aerosol from low-volatility organic compounds. Atmospheric Chemistry and Physics, 10(9), 4377-4401.
Woody, M. C., West, J. J., Jathar, S. H., Robinson, A. L., & Arunachalam, S. (2015). Estimates of non-traditional secondary organic aerosols from aircraft SVOC and IVOC emissions using CMAQ. Atmospheric Chemistry and Physics, 15(12), 6929-6942.
Robinson, A. L., Donahue, N. M., Shrivastava, M. K., Weitkamp, E. A., Sage, A. M., Grieshop, A. P., ... & Pandis, S. N. (2007). Rethinking organic aerosols: Semivolatile emissions and photochemical aging. Science, 315(5816), 1259-1262.
Wu, L., Ling, Z., Liu, H., Shao, M., Lu, S., Wu, L., & Wang, X. (2021). A gridded emission inventory of semi-volatile and intermediate volatility organic compounds in China. Science of the total environment, 761, 143295.
Citation: https://doi.org/10.5194/egusphere-2022-1502-CC1
-
RC2: 'Comment on egusphere-2022-1502', Anonymous Referee #2, 24 Feb 2023
This manuscript provides insight into model predictions of SOA over China using two common regional models: CAMx and CMAQ. The ability to find robust messages, such as the important role of IVOCs, across diverse model representations is useful. However, the current manuscript insufficiently describes the base models and should go farther in providing insight into what is well represented vs not.
Major comments:
- Provide a better description of CMAQ SOA model with appropriate citations.
- CMAQ AERO7 uses a VBS treatment for some systems, but not all. Specifically, many aqueous pathways and the “pcSOA” approach are not VBS style. Relabel the CMAQ treatment from “CMAQ VBS” to “CMAQ AERO7” throughout the manuscript to be more complete.
- Several SOA pathways in CMAQ are not mentioned and the current description does not adequately cite CMAQ developments (none of the biogenic SOA articles are cited, for example). Provide a more complete description of SOA in CMAQ vs CAMx and add references as appropriate (See https://www.epa.gov/cmaq/how-cite-cmaq). Consider:
- CMAQ IEPOX SOA approach (not cited)
- CMAQ oligomer approach (not cited)
- CMAQ organic nitrate SOA (not mentioned nor cited)
- CMAQ monoterpene photooxidation SOA (mentioned but not cited)
- CMAQ semivolatile POA approach (could be better cited)
- CMAQ glyoxal/methylglyoxal SOA approach (neither mentioned nor cited)
- One of the conclusions is: “CMAQ tends to estimate higher BSOA concentration, while CAMx generates more ASOA.” You could mention that CMAQ has more pathways to SOA from biogenic precursors, including aqueous pathways that are not present in CAMx.
- Emissions and model choices (e.g., MEGAN) on lines 151-160 could also use literature citations.
- Is biogenic SOA in CAMx subject to VBS aging?
- Can the authors go further in determining whether parametrizations are realistic and what improvements might be needed?
- Figure 6: Is there experimental data to support one parameterization over another? Have you compared the CAMx aging scheme to more recent 2-D VBS parameterizations such as the work of Zhao et al. (2016)? Given 5.75 hours of aging could be captured in a chamber experiment is there data to confirm a 30x increase in yield? Is the aging scheme in CAMx plausible?
- The text on line 401 through 417 (differences in the CMAQ vs CAMx benzene system) should be refocused. The experimental data (Ng et al., 2007) demonstrates why there is only mass in the lowest C* bin for CMAQ—the SOA was observed to be nonvolatile in terms of yield behavior. Providing the total yield of C* 1000 ug/m3 and lower species in CAMx vs CMAQ confuses the story which is better captured on the following page. Table S9 indicates CMAQ would predict an SOA yield from benzene of 0.146 for all atmospherically relevant conditions. CAMx would predict an SOA yield of about 10% (loading of 10ug/m3 assumed) so the CAMx SOA yield for benzene before any multigenerational aging is lower than CMAQ (later shown in Fig 6). This suggests that it isn’t the initial benzene SOA yields (whether wall loss corrected are not) that are driving differences.
- Reword sentences to bring clarity and specificity. For example: “Our results indicate that large uncertainties still exist in the simulation of SOA in current air quality models due to the aging schemes as well as uncertainties of the emission inventory” can be reworded to: “Our results indicate aging schemes are the major driver in CMAQ vs CAMx treatments of ASOA and their resulting predicted mass.” (The role of emission inventories wasn’t specifically addressed and could be removed.)
- Can the regional model bias be used to help inform which representations are plausible? For example, what is the spatial pattern of bias? Since BSOA and ASOA have some spatial separation, how does performance very by model species? Observations could be added to Fig 4.
- The authors map IVOC emissions in CMAQ to the pcSOA precursor (pcVOC). From Murphy et al. (2017): “We further introduce a new surrogate species, potential SOA from combustion emissions (pcSOA) to account for missing mass from IVOC oxidation, multigenerational aging of (anthropogenic) secondary organic vapors (from IVOC and VOC precursors), biases in SOA yields from vapor wall losses, and enhanced organic partitioning to the condensed aqueous phase. In addition to these sources, pcSOA could account for mass from oxidation of as-yet unidentified sources of SOA precursors.” Are IVOCs a good fit for the pcSOA precursor? How much does the emission magnitude of IVOCs differ from what Murphy et al. proposed as the emission rate (which was not IVOC specific)? How does the yield of SOA from pcSOA compare to that expected for IVOCs?
Minor comments:
- Clarify your definition of biogenic (BSOA and BVOC). Are BVOCs strictly from vegetation or defined as specific VOCs such as isoprene and monoterpenes? If BVOCs are defined based on isoprene or monoterpene identity, please highlight that anthropogenic monoterpene emissions can be substantial (Coggon et al., 2021) and anthropogenic NOx modulates monoterpene SOA (Pye et al., 2015) and thus biogenic does not mean the SOA is entirely biogenic.
- Line 101: What is meant by brute-force SOA estimation? Is that a zero out?
- Line 125: I recommend removing “No clear conclusions can be drawn.” Often, the different results reflect different model parameterizations. The reason they are giving different answers can be (at least partially) identified.
- Section 2.2: Include a brief overview of how emissions from previous were developed (were they scaled to POA)?
- Line 193: Are these percents of total S/IVOC or total VOC?
- The 1.6 OM/OC ratio attributed to Feng et al. is actually from Turpin and Lim 2001. The value seems a bit low considering primary wood burning emissions often have OM/OC ratios of 1.7. Consider updating the OM/OC from 1.6 to a more recent value. Alternatively, model output can be converted to OC as the model often has a specific molecular weight and other properties assigned to the species. CMAQ specifies OM/OC ratios in the Species Definition files supplied with the model (https://github.com/USEPA/CMAQ/blob/5.3.2/CCTM/src/MECHS/cb6r3_ae7_aq/SpecDef_cb6r3_ae7_aq.txt).
- Section 3.1: Experimental data to feed parameterizations has increased in concentration range over time which is what allows a greater range of volatility to be fit in the VBS vs older data sets. Similarly, older data tended to be from experiments from very high loading which made extrapolation to ambient atmospheres more difficult and likely drove errors. Consider adding this context.
- Line 364: Cite peer-reviewed original references rather than model release notes.
- Line 417: Remove personal communication citation. The CMAQ benzene yields can be traced back to experimental data which indicates if vapor wall loss was performed.
- Figure 6: Add aging time of 5.75 hours to caption.
- At least one critical reference is missing from Table 1 (Zhao et al., 2016).
- Reword the citation on line 64-65—the reason for the association with SOA and mortality has not been determined.
- Table 1 could be moved to the SI. Also consider relabeling as some figure labels use the Table 1 labels (Fig 2) and others do not (Fig 3).
References
Coggon, M. M., Gkatzelis, G. I., McDonald, B. C., Gilman, J. B., Schwantes, R. H., Abuhassan, N., Aikin, K. C., Arend, M. F., Berkoff, T. A., Brown, S. S., Campos, T. L., Dickerson, R. R., Gronoff, G., Hurley, J. F., Isaacman-VanWertz, G., Koss, A. R., Li, M., McKeen, S. A., Moshary, F., Peischl, J., Pospisilova, V., Ren, X., Wilson, A., Wu, Y., Trainer, M., and Warneke, C.: Volatile chemical product emissions enhance ozone and modulate urban chemistry, P. Natl. Acad. Sci. USA, 118, e2026653118, https://doi.org/10.1073/pnas.2026653118, 2021.
Ng, N. L., Kroll, J. H., Chan, A. W. H., Chhabra, P. S., Flagan, R. C., and Seinfeld, J. H.: Secondary organic aerosol formation from m-xylene, toluene, and benzene, Atmos. Chem. Phys., 7, 3909-3922, https://doi.org/10.5194/acp-7-3909-2007, 2007.
Pye, H. O. T., Luecken, D. J., Xu, L., Boyd, C. M., Ng, N. L., Baker, K. R., Ayres, B. R., Bash, J. O., Baumann, K., Carter, W. P. L., Edgerton, E., Fry, J. L., Hutzell, W. T., Schwede, D. B., and Shepson, P. B.: Modeling the current and future roles of particulate organic nitrates in the southeastern United States, Environ. Sci. Technol., 49, 14195-14203, https://doi.org/10.1021/acs.est.5b03738, 2015.
Zhao, B., Wang, S., Donahue, N. M., Jathar, S. H., Huang, X., Wu, W., Hao, J., and Robinson, A. L.: Quantifying the effect of organic aerosol aging and intermediate-volatility emissions on regional-scale aerosol pollution in China, Sci. Rep., 6, 28815-28815, https://doi.org/10.1038/srep28815, 2016.
Citation: https://doi.org/10.5194/egusphere-2022-1502-RC2 - Provide a better description of CMAQ SOA model with appropriate citations.
-
AC1: 'Comment on egusphere-2022-1502', Li Li, 16 Apr 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2022-1502/egusphere-2022-1502-AC1-supplement.pdf
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 858 | 353 | 38 | 1,249 | 119 | 29 | 37 |
- HTML: 858
- PDF: 353
- XML: 38
- Total: 1,249
- Supplement: 119
- BibTeX: 29
- EndNote: 37
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1