the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Anthropogenic amplification of biogenic secondary organic aerosol production
Yiqi Zheng
Larry W. Horowitz
Raymond Menzel
David J. Paynter
Vaishali Naik
Jingyi Li
Abstract. Biogenic secondary organic aerosols (SOA) contribute to a large fraction of fine aerosols globally, impacting air quality and climate. The formation of biogenic SOA depends on not only emissions of biogenic volatile organic compounds (BVOCs) but also anthropogenic pollutants including primary organic aerosol, sulfur dioxide (SO2), and nitrogen oxides (NOx). However, the anthropogenic impact on biogenic SOA production (AIBS) remains unclear. Here we use the decadal trend and variability of observed OA in the southeast US, combined with a global chemistry-climate model, to better constrain AIBS. We show that the reduction in SO2 emissions can only explain 40 % of the decreasing decadal trend of OA in this region, constrained by the low summertime month-to-month variability of surface OA. We hypothesize that the rest of OA decreasing trend is largely due to reduction in NOx emissions. By implementing a scheme for monoterpene SOA with enhanced sensitivity to NOx, our model can reproduce the decadal trend and variability of OA in this region. Extending to centennial scale, our model shows that global SOA production increases by 36 % despite BVOC reductions from preindustrial period to present day, largely amplified by AIBS. Our work suggests a strong coupling between anthropogenic and biogenic emissions in biogenic SOA production that is missing from current climate models.
- Preprint
(1928 KB) -
Supplement
(15026 KB) - BibTeX
- EndNote
Yiqi Zheng et al.
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-372', Anonymous Referee #2, 11 Apr 2023
GENERAL COMMENT
In this manuscript, Zheng et al. investigate the role of anthropogenic pollutants on the production of biogenic secondary organic aerosols. Using a chemistry-climate model, they perform decadal simulations for present day and pre-industrial periods and compare three schemes that simulate the production of secondary organic aerosols (SOAs) with increasing complexity: (1) Simple, (2) CMPX and (3) CMPX_ag. Over high-source regions of biogenic emissions such as the South-eastern United States and the Amazon, the more advanced scheme, CMPX_ag, outperforms the others and reproduces the observed seasonal variability and trend of summertime organic aerosols. By accounting for different NOx regimes (low vs. high) and for photochemical aging, the CMPX_ag scheme is more sensible to anthropogenic pollution, thus simulating an increased SOA production over present-day compared to the pre-industrial period, although land-cover change has driven a decrease in biogenic emissions. Hence, this study highlights the tight link between SOA production and anthropogenic pollution.
The paper is within the scope of ACP. It examines an important topic such as the sensitivity of biogenic SOA production to anthropogenic pollution, and addresses relevant scientific questions. The paper is well written, the abstract is concise and complete, the introduction is exhaustive and clear, the methods and modeling are well laid out, the literature is thoroughly referenced, and the results are presented in good clear figures. For this reason, I recommend publication after a few minor comments, listed below, have been addressed by the authors.
SPECIFIC COMMENTS
Sect. 2, Methods
Regarding the GFDL AM4.1 model and the modeling of the SOA formation, since the CMPX scheme depends on OH, O3 and NOx abundance, in my opinion the author should show how well the GFDL AM4.1 model reproduces these gases, or at least insert a sentence/paragraph that summarizes results from previous studies that evaluated the model performance.
To allow the traceability of results, I think it is important to provide details on the spatial resolution and the time-step of the GFDL AM4.1 model. I also suggest to precise the original temporal resolution of observational datasets (IMPROVE, SEARCH, ACTRIS and ARM) and if (and how) these data have been aggregated.
Moreover, I think it is important to explain in the Methods section how relative/percent trends, presented in Sect. 3 (e.g., pag.6, line 18), have been computed. If I correctly understood, I found this information in Fig. S4, in the Supplementary Material (“Changing rates m have units of % per year relative to their 2000-2016 averages”).
Pag. 4, line 17-18: In my opinion, I think it is important to specify that LAI values follow an annual cycle (prescribed at the 1992 level, as precised by the authors).
Sect. 3, Results
Pag. 6, line 27: To avoid confusion among readers, figures should be referenced in the order they appear in the text. Here, the authors refer to the Supplementary Figure S5, while in the next paragraph they refer to Fig. S3. For this reason, I suggest the authors to i) revise the order supplementary figures are presented in the Supplementary Material, and ii) comment all the supplementary figures (or remove those that are not commented in the manuscript).
Pag. 7, ll. 21: I suggest to briefly recall the other vegetated regions that have been selected for evaluation and that have been presented in Sect. 2.3.
Sect. 4, Summary
I think that the application of the CMPX_ag scheme could be also interesting for local-regional studies on SOA production. For this reason, although Zheng et al. performed their study at the global scale, I think it could be useful for readers to know the computational cost of including the CMPX_ag scheme in their runs, compared to the cost of using the Simple or the CMPX schemes. This information could be provided in the Methods Section, or it could be commented in the Summary when discussing about perspectives.
MINOR COMMENTS
Through the whole manuscript: space is missing before parenthetical citations (e.g., pag.2, line 2: “BVOCs(Guenther et al., 2012))”. Please add these spaces.
Sect. 1, Introduction
Pag. 3, ll. 2: Definition of the acronym CMPX and CMPX_ag are missing. Please define them.
Sect. 3, Results
Pag. 8, ll. 18: I think "is" is missing before the adjective "consistent".
Figures
Fig. 1: In the figure caption, I suggest to recall region boundaries, which are precised in the text (pag. 5, ll. 37). As well, I think it could be useful to recall in the caption the meaning of the different acronyms (ISOA, TSOA, ASOA),
Citation: https://doi.org/10.5194/egusphere-2023-372-RC1 -
RC2: 'Comment on egusphere-2023-372', Anonymous Referee #3, 18 Apr 2023
General Comments
The authors tackle an important and underdeveloped scientific question (the relative influence of anthropogenic emissions on biogenic SOA) and the analysis has clearly involved a meaningful and thoughtful allocation of time and resources (model runs + observational comparisons). The authors highlight a few key results around the importance of appropriately modeling the anthropogenic impact on biogenic SOA formation (AIBS) in order to (1) better explain the trends in SOA over the past 2 decades and (2) provide insight into the anthropogenic influence on SOA (and associated climate impacts) relative to pre-industrial atmospheres. The study also highlights an important result regarding the relative influence SO2 vs NOx (and ISOA vs TSOA). The authors also use observational constraints to validate their models.
While the study provides valuable insight into this important domain, the SOA mechanisms themselves could be better validated, more varied and more detailed. Given that the central goal of the manuscript is to highlight the importance of AIBS, there could be value to exploring other model SOA mechanisms that explicitly vary in their sensitivity to NOx and SO2. A more comprehensive mechanistic comparison of such schemes would provide greater certainty to some of the central results in this paper (such as the outsized importance of NOx vs SO2 in explaining recent SOA trends). This paper could also be meaningfully strengthened if the authors deepened their analysis of the ISOA & TSOA pathways via a discussion and literature review of the current state of the science in modeling the anthropogenic impact on both species and a more detailed focus on the underlying uncertainties in their approach to modeling each of the three schemes (for instance, the simple mechanism which assumes a 10% yield for ISOA is likely overestimating this parameter based on other literature in this space. While the authors do allude to this, a more detailed discussion and yield sensitivity analysis would strengthen the model comparison that ties together their core argument).
Overall, the authors tackle an important scientific question and the resulting analysis fits well within the scope of ACP. I would recommend the publication of this manuscript following the revisions outlined below that would strengthen its contribution to the domain.
Specific Comments
• OA Mechanism Selection and Modification
The paper would benefit from a more detailed overview of the different SOA schemes and why they were chosen / modified in the manner they were (as opposed to other mechanistic changes that might also have nudged them in the right direction given the authors’ hypothesis). A more detailed overview / discussion of the impact that implementing other mechanisms (such as an explicit auto-oxidation mechanism, alternate IEPOX mechanisms, etc.) might have had on the CMPX_ag simulation would also be helpful.• Model Specifics
A more detailed overview and discussion of the general model simulation such as resolution, time-step, loss processes, pre-industrial emissions etc. would be useful in reproducing these results. More details / uncertainties regarding the modelled emissions and atmospheric fates of the anthropogenic gas-phase species that influence the SOA production (NOx, SO2, etc.) would be similarly useful.• Observations
Similarly, the manuscript would benefit from a more detailed overview of how the observations were sampled, averaged and manipulated. More detailed statistics / uncertainties on the model-observation comparison would also be useful. Lastly, a detailed discussion on the kind of observations that might prove out the core thesis of this work (e.g. the relative importance of NOx vs SO2) would help future research build on this study.• Mechanistic Learnings
Since the paper makes a point to stress the mechanistic relevance of anthropogenic pathways for SOA, it could be meaningfully strengthened by a more detailed mechanistic analysis / comparison of these pathways (either via more model runs that isolate their relative importance or via a detailed discussion and literature review that hypothesizes the relative importance of other mechanisms not considered in this analysis). For instance - Do the authors recommend the CMPX_ag as the SOA scheme of choice? What are the benefits / downsides of using this scheme relative to other complex schemes in different community models? In the authors opinion, does this analysis close much of the gap in our understanding of AIBS or are there still large undeveloped mechanisms that need to be explored? If NOx is more influential than SO2 in describing recent SOA trends, are there implications from this study on how atmospheric aerosol models should treat the emissions / fate of NOx / SO2 going forward?Citation: https://doi.org/10.5194/egusphere-2023-372-RC2
Yiqi Zheng et al.
Yiqi Zheng et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 378 | 116 | 12 | 506 | 29 | 5 | 5 |
- HTML: 378
- PDF: 116
- XML: 12
- Total: 506
- Supplement: 29
- BibTeX: 5
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1