the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 11 Dec 2024
Source-explicit estimation of brown carbon in the polluted atmosphere over North China Plain: implications for distribution, absorption and direct radiative effect
Abstract. Brown carbon (BrC) is recognized as a considerable factor changing the atmospheric radiation balance. In addition to the biomass and biofuel sources, both field observations and laboratory studies suggest that fossil fuel combustion is an important contributor to BrC. This highlights a critical gap in the treatment of BrC in climate models, which typically categorize organic aerosols (OA) from fossil fuels as non-absorbing or simplistically assume that all OA are light-scattering. Here we present a regional simulation of BrC during a highly polluted winter in North China Plain (NCP) by using the WRF-Chem model incorporating currently known BrC sources with explicit absorption properties. The modified model generally performs well in simulating air pollutants and aerosols species against observations. Our simulations show that the average near-surface mass concentration of BrC in the NCP is 4.8 μg m-3 and its contribution to the aerosol absorption optical depth at 365 nm is 11.2 %. A diagnostic adjoint method has been used to quantify the overall direct radiative effect (DRE) of BrC and contributions from various sources. We find that the DRE of BrC is predominantly negative with an average of -0.10 W m-2 at the top of the atmosphere (TOA) over the NCP, and consequently decreases the direct radiative cooling effect of OA by 24.0 % with a TOA warming of up to +0.34 W m-2. Our findings reveal that residential coal combustion is the principal contributor to the DRE of BrC in the NCP, and a noteworthy contribution from secondary BrC.
- Preprint
(2205 KB) - Metadata XML
-
Supplement
(1796 KB) - BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2024-3468', Anonymous Referee #1, 06 Jan 2025
Zhou et al. present results from WRF-Chem simulations over the North China Plain, with a focus on brown carbon from different sources. They find residential coal combustion and biomass burning dominate brown carbon emissions in the region, and their results show that the inclusion of brown carbon reduces organic aerosols' cooling by 24%, in terms of DRE. While their investigation of brown carbon from fossil fuels looks interesting, the paper lacks sufficient scientific justification for their methods and model parameters. I believe this manuscript requires substantial revision, including new simulations and sensitivity analyses, before it can be considered suitable for publication in ACP. My specific comments are detailed below.
1) L131-135: A major methodological concern is the arbitrary assumption that 10% of total SOA contributes to BrC, which lacks scientific justification. While the authors cited Liu et al. (2015, 2016) and Xiu et al. (2017b) to support their treatment of SOA-derived BrC, these studies actually demonstrated that biogenic SOA showed significantly lower light absorption compared to aromatic VOC-derived SOA. Indeed, this finding explains why previous studies, including Wang et al. (2014) and others cited in the introduction, specifically considered only aromatic SOA in their BrC calculations. Given the distinct spatial distribution patterns of biogenic and anthropogenic SOAs, the model simulations should be revised to treat only aromatic SOA as light-absorbing while explicitly treating biogenic SOA as non-absorbing.
2) L121-130: The methodology for deriving BrC emissions from primary sources lacks clear documentation. The treatment of RCC emissions is particularly unclear - the authors do not specify whether they assumed all RCC organic aerosols contribute to BrC. Similarly, their apparent assumption that all BB OA acts as BrC represents an oversimplification. Even in a study more than 10 years ago, Feng et al. (2013) used a non-unity fraction (92%) based on scientific justification (Chen and Bond, 2010). The current study requires similar scientific justification for these critical assumptions about BrC source contributions.
3) L148-154 & Table 1: The selection criteria for MAE and k of BrC require clarification. The authors inconsistently adopted MAE values from Zhang et al. (2022) for primary BrC while using values from Ni et al. (2021) for secondary BrC, although Zhang et al. (2022) reported both. Furthermore, the refractive indices from Ni et al. (2021) are notably lower than those reported in Zhang et al. (2022) (e.g., MAC at 370 nm: 2.39 for BBOA vs. 2.08 for LO-OOA) and other cited studies such as Xie et al. (2017) and Liu et al. (2016), where k365 values for toluene SOA range from 0.008 (H2O2) to 0.025 (NOx). For consistency and completeness, the study should adopt values from Zhang et al. (2022) for both primary and secondary BrC, which would also ensure consistent wavelength ranges (370-660 nm) across all BrC components.
4) L55-57: The claim of "a growing number of studies" is not supported by the cited references. One reference is from 2001, and the 2020 reference actually states that "fossil-fuel combustion is not an important BrC emitter". Given that fossil-fuel BrC is an important component of this study, the introduction requires a more comprehensive discussion of how the scientific understanding of fossil-fuel BrC has evolved. This should include an explanation of why fossil-fuel BrC was historically overlooked, and how recent observational and modeling studies have revealed its importance. While some of these aspects are mentioned in L57-67, the current discussion lacks clear organization and logical flow, and should be restructured to present this evolution more coherently.
5) L54-55: This limitation (BrC from FFs) usually extends to chemical transport models and atmospheric chemistry models as well. Climate models present an even greater concern, as they typically do not consider brown carbon at all. And these statements require supporting references from the literature.
6) L101: A reference for the density values (1.2 g cm-3 for primary and 1.0 g cm-3 for secondary BrC) is needed.
7) L113: I would recommend rephrasing this sentence, such as "To date, studies on BrC emissions have been limited"
8) L124: The manuscript contains numerous grammatical errors that need attention. Professional language editing is recommended. Some examples include:
- L125: "which resulting" should be "resulting in~"
- L145: "it follow the study" should be "this study follows~"
- L170: "A adjoint" should be "An adjoint~"
- etc.
9) L191: The reference "Bai et al. (2022)" is missing from the reference list. Additionally, the LGHAP dataset requires further explanation regarding its derivation methodology and validation.
10) L204: The authors need to provide details on how the model calculates hygroscopic growth for different aerosol components. This is particularly important as hygroscopic growth significantly affects the model's AOD calculations and subsequent comparisons with satellite observations.
11) L215-223: SSA measurements typically have larger uncertainties at low AOD values. Did the authors apply any screening criteria for SSA values associated with low AOD conditions? If not, the authors may want to do a screening to understand the SSA simulation uncertainty.
12) L241-252: The reported values in this section are meaningless without proper scientific justification for the key assumptions as mentioned in earlier comments:
- The BrC ratios from different primary sources
- The arbitrary 10% SOA contribution to BrC
These fundamental parameters require thorough justification before the resulting calculations can be considered reliable.
13) L254-273: The SOA-BrC values reported here should be higher if k values from other studies were adopted, as discussed in the previous comments. While sensitivity simulations were mentioned in the introduction, no such analyses appear in the manuscript. This section would benefit from sensitivity tests exploring the impact of different k values on the results.
14) L269-271: The attribution of differences to aerosol density and mixing state lacks sufficient discussion and scientific justification. A more likely explanation involves the vertical distribution of aerosols: while the authors compare surface concentrations of primary versus secondary BrC, AOD represents column-integrated concentrations. Given that SOA/POA ratios typically increase with altitude, analysis of vertical profiles should help clarify these differences.
15) L310-313: This pattern likely stems from the authors' arbitrary assumption that 10% of total SOA contributes to BrC. This assumption incorrectly attributes a significant portion of biogenic SOA from southern China as BrC, which does not reflect actual atmospheric conditions.
16) References: Several references are incomplete. For example, the journal name and DOI are not available in Feng et al. (2013). The authors should double-check the reference list.
Citation: https://doi.org/10.5194/egusphere-2024-3468-RC1 -
RC2: 'Comment on egusphere-2024-3468', Anonymous Referee #2, 29 Jan 2025
General overview:
The manuscript by Zhou et al. estimates brown carbon (BrC) emissions from multiple sources and investigates the effects on aerosol optical properties and direct radiative effect (DRE) over the North China plain. The authors used WRF-Chem and observational data to simulate BrC emissions and impacts. They found that the net radiative effect of BrC is negative (cooling effect). However, accounting for light absorption by BrC results in a positive (warming) radiative effect, compared with scattering-only BrC. Although this study could be potentially useful for the scientific community, I believe that this manuscript needs major revisions before being published. In the revision, the authors should clarify the methodology and results, and also improve the writing of the manuscript (grammatically). I have provided some major and minor comments below.Major comments:
- While the authors provided many details in the methodology, there are still some missing important information. For example, BrC refractive index and DRE calculations are well explained, but AOD calculations are not mentioned in the methodology. Since the modeled AOD is discussed and compared with observations, the author should briefly describe the process of calculating AOD (e.g., Mie theory uses refractive indices as inputs?)
- Any assumption made in the study needs either a reference or an explanation. There are multiple assumptions in your study that are not justified. For example, “In this work, a proportion of 10% of the total SOA is included as a part of BrC” or “The BrC in the model has an effective density of 1.2 g cm-3 for primary BrC and of 1.0 g cm-3 for secondary BrC”.
Minor comments:
- Page 6, line 125: “RCC is responsible for about 45% of primary BrC emissions”. Is this percentage based on your calculations? Or from other studies or data? Please use proper citations or discuss the calculations further.
- Page 7, line 148: Expand MAE when it appears for the first time. This applies to all other abbreviations used in the manuscript.
- Page 9: Give more details on LGHAP and OMI satellites (e.g., horizontal resolution, local time overpass, etc.). Then discuss those further. For example, are you comparing model results at the satellite overpass time?
- Page 10, line 204: Change “hydroscopic” to “hygroscopic”. Hygroscopic growth is the size increase of aerosols due to water vapor absorption.
- Page 11, line 221: Change “ration” to “ratio”.
- Page 13, line 259: Citation is needed for “Compared with the global mean ratio of BrC to BC which is 1.24”.
- Page 15, line 288: “the increased DRETOA induced by BrC which is usually considered as its scattering effect is up to an average of +0.37 W m-2 …”. This sentence is confusing a little. The authors should clarify if they are talking about BrC's net radiative effect or just absorption. As mentioned in Page 15, line 285: “the BrC populations have a net cooling effect”. So, the net BrC DRE could not increase OA DRE.
- Figures 8 and 9: Panel letter C is missing.
- Figure S1: Adding error bars to the timeseries would be helpful since the authors are demonstrating average concentrations over all the monitors in NCP.
- Figure S4: The units inside the charts (MB and RMSE) should be updated.
Citation: https://doi.org/10.5194/egusphere-2024-3468-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 164 | 31 | 12 | 207 | 20 | 11 | 13 |
- HTML: 164
- PDF: 31
- XML: 12
- Total: 207
- Supplement: 20
- BibTeX: 11
- EndNote: 13
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1