the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 26 Jun 2024
Brown carbon aerosol in rural Germany: sources, chemistry, and diurnal variations
Abstract. Brown carbon aerosol (BrC) is one major contributor to atmospheric air pollution in Europe, especially in winter. Therefore, we studied the chemical composition, diurnal variation, and sources of BrC from 17th February to 16th March at a rural location in southwest Germany. In total, 178 potential BrC molecules (including 7 nitro aromatic compounds, NACs) were identified in the particle phase comprising on average 63 ± 32 ng m−3, and 31 potential BrC (including 4 NACs) molecules were identified in the gas phase contributing on average 6.2 ± 5.0 ng m−3 during the whole campaign. The 178 potential BrC molecules only accounted for 2.3 ± 1.5 % of the total organic mass, but can explain 11 ± 11 % of the total BrC absorption at 370 nm, assuming an average mass absorption coefficient at 370 nm (MAC370) of 9.5 m2 g−1. A few BrC molecules dominated the total BrC absorption. In addition, diurnal variations show that gas phase BrC was higher at daytime and lower at night. It was mainly controlled by secondary formation (e.g. photooxidation) and particle-to-gas partitioning. Correspondingly, the particle phase BrC was lower at daytime and higher at nighttime. Secondary formation dominates the particle-phase BrC with 61 ± 21 %, while 39 ± 21 % originated from biomass burning. Furthermore, the particle-phase BrC showed decreasing light absorption due to photochemical aging. This study extends the current understanding of real-time behaviors of brown carbon aerosol in the gas and particle phase at a location characteristic for the central Europe.
- Preprint
(982 KB) - Metadata XML
-
Supplement
(968 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-1848', Anonymous Referee #1, 07 Jul 2024
This paper describes ambient air measurements conduced for one month in winter in Germany in 2021. In addition to standard instrumentation for O3, NOx, particle sizes, etc, the key measurements were gas and particle phase composition using FIGAERO I-CIMS and an aethalometer. In addition to black carbon (BC) source apportionment, the goals of the work were to identify potential brown carbon (BrC) molecules in the gas and particle states, to study their diurnal variations, and to determine to what degree they contribute to the overall BrC measured by the aethalometer.
There is merit to this type of study because we need more molecular information about BrC molecules and their behavior, especially in regions which are not predominantly influenced by wildfire emissions, i.e., in this location the sources are presumably residential biomass burning and fossil fuel combustion. The findings are that there are both biomass burning (larger) and fossil fuel (smaller) contributions to BC, there were about 200 or so mass spectral features that may be BrC molecules, these features contribute about 10 percent to the total BrC absorption at 370 nm but a lower fraction of the total organic aerosol mass, gas phase BrC is largely photochemically generated during the day, and the ratio of gas phase BrC to particle phase BrC is much less than unity.
General comments:
These are potentially interesting new measurements in Germany of quantities that have been measured at other locations. The methods and results are not particularly novel, and the paper needs to be much more quantitatively rigorous, especially with regard to uncertainties. I wonder whether this paper should be classified as an ACP “Measurement Report”? Overall, there is quite a bit of work to be done to get this paper ready for publication.
Specific comments:
My major criticism of this paper is the quantitative uncertainties in the measurements. In particular, unless I missed mention of them, there are no calibrations for the FIGAERO I-CIMS measurements. Rather, I believe that an “average sensitivity” was used for all mass spectral features, with the value taken from a literature study, i.e., with an entirely different instrument/operator. This is quite problematic, as CIMS instruments vary widely in sensitivity from one to another, even if operated in nominally the same manner. Calibration of at least a small number of standard compounds is the minimum standard for field work, and increasingly many molecules are calibrated (or the voltage scanning method is applied) for I-CIMS work.
Moreover, calibrations for particle bound species (such as levoglucosan) can be performed with the FIGAERO by depositing known amounts of these molecules on the collecting filter. Thus, the authors have to better justify their reports of absolute amounts of BrC molecules. If they have not calibrated themselves, I do not believe they can report an absolute amount.
On a related note, the authors appear to dismiss this uncertainty after acknowledging it: Line 131 “These values have high uncertainty with several orders of magnitude. However, this is still a reasonable method to measure the organic aerosol in atmosphere.” They need to justify why this approach is “reasonable”.
In an analogous manner, aethalometer measurements require care to interpret, with corrections for on-filter scattering and loadings. Although the authors mention these uncertainties, they do not provide quantitative estimates for them.
Likewise, the paper performs BC source apportionment, and it decouples BC absorption from total absorption to arrive at BrC absorption. There are many ways to do these analyses. The paper should justify the methods chosen.
Moreover, it uses a literature value for the MAC value of BrC. How variable are these values from one site to another? The MAC value could be strongly dependent on the type of BrC being analyzed.
What uncertainties are there in the total organic aerosol mass loading given that there was no measurement of it during the campaign?
The paper does not provide a justification for how BrC molecules are identified from 1000’s of mass spectral features, aside from providing a reference. How accurate are the mass fittings and the calculation of DBE and elemental composition for each feature? In other words, are these fittings unique for only one elemental formula? The paper should identify the BrC mass spectral features identified, with some indications of their intensities. Were any mass spectral features observed in both the gas and particle phase spectra? It would be interesting to know this, and a partition coefficient could be calculated.
Can a non-parametric wind direction analysis be provided to aid source apportionment?
Line 267. Photochemical activity forms ozone and is not the only cause of aging.
Citation: https://doi.org/10.5194/egusphere-2024-1848-RC1 -
AC1: 'Reply on RC1', Feng Jiang, 17 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1848/egusphere-2024-1848-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Feng Jiang, 17 Oct 2024
-
RC2: 'Comment on egusphere-2024-1848', Anonymous Referee #2, 29 Aug 2024
Review of Jiang et al. Brown carbon aerosol in rural Germany: sources, chemistry, and diurnal variations
This manuscript presents results from concurrent measurements of aerosol chemical composition and light absorption from black carbon and light-absorbing organic aerosols (aka brown carbon) at a rural site in Germany during one month in winter 2021. The absorption apportionment and chemical speciation of brown carbon aerosols in both gas and particle phases are reported. The sources of brown carbon in rural Germany were identified based on its diurnal variability and regression analysis of brown carbon with emission tracers. In general, this study adds to the literature on the characteristics of brown carbon aerosols in rural Germany, but for the reasons outlined below I cannot recommend publication of this manuscript in its current form.
Overall, I am not convinced that the 178 molecules identified by FIGAERO-CIMS are representative of the brown carbon aerosols, not only because they contributed to a very small fraction (2%) of total organic mass as well as brown carbon absorption (11%), but also the correlation between the molecule mass and brown carbon absorption is not so good. There are a lot of scatters in Figure S6 which means most of the brown carbon absorption cannot be explained by the identified molecules. The authors also did not provide details in how these compounds are identified as brown carbon molecules rather than referencing to an earlier publication from the same group. In several places throughout the manuscript the authors simply refer the 178 molecules as particle-BrC and the 31 molecules as gas-BrC (e.g. Figure 1, Figure S6 and related text), which is not accurate given the reasons provided above.
There is lack of a discussion on the uncertainties related to absorption measurement by the aethalometer and calculations deriving BC and BrC absorption, as well as BrC source apportionment in Line 302-303. In Line 190-191, the authors stated that “During this winter campaign, the BrC absorption accounted for ~40% of total absorption caused by BC and BrC.” I could not trace back to how this number (40%) was derived, nor did the authors provide information about which absorption wavelength the calculation is based on.
The authors also tend to draw causal relationship based on correlation. For example, in Line 200, the authors stated “The levoglucosan had a good correlation (r=0.7) with BC. This also indicates that BC was mainly emitted from biomass burning during the winter campaign.” This statement is not supported by evidence. Having a r =0.7 means levoglucosan can explain less than half of the variability in BC concentration. Similar statement is in Line 322-324 “In addition, the O/C ratio of BrC had a positive correlation (r=0.8) with ozone. This indicates that the BrC was photo-oxidized leading to an increase of the O/C ratio of BrC.”
Specific comments:
In Figure 2, the contribution from nitro-aromatics absorption is only plotted at 370 nm. I wonder if the absorption profile of these compounds were measured, and if so, it would be interesting to show absorption contribution from the nitro-aromatics across the whole spectrum.
Figure S3: second panel from top: no color differentiation for the two AAE parameters plotted.
Figure S8. The correlation of gas-phase BrC and temperature is based on exponential fit (y=e(0.15*x)). How is the correlation between temperature and particle phase BrC like? The authors stated “Figure S8 shows that BrC in the gas phase had a good correlation (r=0.4) with temperature.” I recommend changing “good” to “moderate”.
Citation: https://doi.org/10.5194/egusphere-2024-1848-RC2 -
AC2: 'Reply on RC2', Feng Jiang, 17 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1848/egusphere-2024-1848-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Feng Jiang, 17 Oct 2024
Status: closed
-
RC1: 'Comment on egusphere-2024-1848', Anonymous Referee #1, 07 Jul 2024
This paper describes ambient air measurements conduced for one month in winter in Germany in 2021. In addition to standard instrumentation for O3, NOx, particle sizes, etc, the key measurements were gas and particle phase composition using FIGAERO I-CIMS and an aethalometer. In addition to black carbon (BC) source apportionment, the goals of the work were to identify potential brown carbon (BrC) molecules in the gas and particle states, to study their diurnal variations, and to determine to what degree they contribute to the overall BrC measured by the aethalometer.
There is merit to this type of study because we need more molecular information about BrC molecules and their behavior, especially in regions which are not predominantly influenced by wildfire emissions, i.e., in this location the sources are presumably residential biomass burning and fossil fuel combustion. The findings are that there are both biomass burning (larger) and fossil fuel (smaller) contributions to BC, there were about 200 or so mass spectral features that may be BrC molecules, these features contribute about 10 percent to the total BrC absorption at 370 nm but a lower fraction of the total organic aerosol mass, gas phase BrC is largely photochemically generated during the day, and the ratio of gas phase BrC to particle phase BrC is much less than unity.
General comments:
These are potentially interesting new measurements in Germany of quantities that have been measured at other locations. The methods and results are not particularly novel, and the paper needs to be much more quantitatively rigorous, especially with regard to uncertainties. I wonder whether this paper should be classified as an ACP “Measurement Report”? Overall, there is quite a bit of work to be done to get this paper ready for publication.
Specific comments:
My major criticism of this paper is the quantitative uncertainties in the measurements. In particular, unless I missed mention of them, there are no calibrations for the FIGAERO I-CIMS measurements. Rather, I believe that an “average sensitivity” was used for all mass spectral features, with the value taken from a literature study, i.e., with an entirely different instrument/operator. This is quite problematic, as CIMS instruments vary widely in sensitivity from one to another, even if operated in nominally the same manner. Calibration of at least a small number of standard compounds is the minimum standard for field work, and increasingly many molecules are calibrated (or the voltage scanning method is applied) for I-CIMS work.
Moreover, calibrations for particle bound species (such as levoglucosan) can be performed with the FIGAERO by depositing known amounts of these molecules on the collecting filter. Thus, the authors have to better justify their reports of absolute amounts of BrC molecules. If they have not calibrated themselves, I do not believe they can report an absolute amount.
On a related note, the authors appear to dismiss this uncertainty after acknowledging it: Line 131 “These values have high uncertainty with several orders of magnitude. However, this is still a reasonable method to measure the organic aerosol in atmosphere.” They need to justify why this approach is “reasonable”.
In an analogous manner, aethalometer measurements require care to interpret, with corrections for on-filter scattering and loadings. Although the authors mention these uncertainties, they do not provide quantitative estimates for them.
Likewise, the paper performs BC source apportionment, and it decouples BC absorption from total absorption to arrive at BrC absorption. There are many ways to do these analyses. The paper should justify the methods chosen.
Moreover, it uses a literature value for the MAC value of BrC. How variable are these values from one site to another? The MAC value could be strongly dependent on the type of BrC being analyzed.
What uncertainties are there in the total organic aerosol mass loading given that there was no measurement of it during the campaign?
The paper does not provide a justification for how BrC molecules are identified from 1000’s of mass spectral features, aside from providing a reference. How accurate are the mass fittings and the calculation of DBE and elemental composition for each feature? In other words, are these fittings unique for only one elemental formula? The paper should identify the BrC mass spectral features identified, with some indications of their intensities. Were any mass spectral features observed in both the gas and particle phase spectra? It would be interesting to know this, and a partition coefficient could be calculated.
Can a non-parametric wind direction analysis be provided to aid source apportionment?
Line 267. Photochemical activity forms ozone and is not the only cause of aging.
Citation: https://doi.org/10.5194/egusphere-2024-1848-RC1 -
AC1: 'Reply on RC1', Feng Jiang, 17 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1848/egusphere-2024-1848-AC1-supplement.pdf
-
AC1: 'Reply on RC1', Feng Jiang, 17 Oct 2024
-
RC2: 'Comment on egusphere-2024-1848', Anonymous Referee #2, 29 Aug 2024
Review of Jiang et al. Brown carbon aerosol in rural Germany: sources, chemistry, and diurnal variations
This manuscript presents results from concurrent measurements of aerosol chemical composition and light absorption from black carbon and light-absorbing organic aerosols (aka brown carbon) at a rural site in Germany during one month in winter 2021. The absorption apportionment and chemical speciation of brown carbon aerosols in both gas and particle phases are reported. The sources of brown carbon in rural Germany were identified based on its diurnal variability and regression analysis of brown carbon with emission tracers. In general, this study adds to the literature on the characteristics of brown carbon aerosols in rural Germany, but for the reasons outlined below I cannot recommend publication of this manuscript in its current form.
Overall, I am not convinced that the 178 molecules identified by FIGAERO-CIMS are representative of the brown carbon aerosols, not only because they contributed to a very small fraction (2%) of total organic mass as well as brown carbon absorption (11%), but also the correlation between the molecule mass and brown carbon absorption is not so good. There are a lot of scatters in Figure S6 which means most of the brown carbon absorption cannot be explained by the identified molecules. The authors also did not provide details in how these compounds are identified as brown carbon molecules rather than referencing to an earlier publication from the same group. In several places throughout the manuscript the authors simply refer the 178 molecules as particle-BrC and the 31 molecules as gas-BrC (e.g. Figure 1, Figure S6 and related text), which is not accurate given the reasons provided above.
There is lack of a discussion on the uncertainties related to absorption measurement by the aethalometer and calculations deriving BC and BrC absorption, as well as BrC source apportionment in Line 302-303. In Line 190-191, the authors stated that “During this winter campaign, the BrC absorption accounted for ~40% of total absorption caused by BC and BrC.” I could not trace back to how this number (40%) was derived, nor did the authors provide information about which absorption wavelength the calculation is based on.
The authors also tend to draw causal relationship based on correlation. For example, in Line 200, the authors stated “The levoglucosan had a good correlation (r=0.7) with BC. This also indicates that BC was mainly emitted from biomass burning during the winter campaign.” This statement is not supported by evidence. Having a r =0.7 means levoglucosan can explain less than half of the variability in BC concentration. Similar statement is in Line 322-324 “In addition, the O/C ratio of BrC had a positive correlation (r=0.8) with ozone. This indicates that the BrC was photo-oxidized leading to an increase of the O/C ratio of BrC.”
Specific comments:
In Figure 2, the contribution from nitro-aromatics absorption is only plotted at 370 nm. I wonder if the absorption profile of these compounds were measured, and if so, it would be interesting to show absorption contribution from the nitro-aromatics across the whole spectrum.
Figure S3: second panel from top: no color differentiation for the two AAE parameters plotted.
Figure S8. The correlation of gas-phase BrC and temperature is based on exponential fit (y=e(0.15*x)). How is the correlation between temperature and particle phase BrC like? The authors stated “Figure S8 shows that BrC in the gas phase had a good correlation (r=0.4) with temperature.” I recommend changing “good” to “moderate”.
Citation: https://doi.org/10.5194/egusphere-2024-1848-RC2 -
AC2: 'Reply on RC2', Feng Jiang, 17 Oct 2024
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2024/egusphere-2024-1848/egusphere-2024-1848-AC2-supplement.pdf
-
AC2: 'Reply on RC2', Feng Jiang, 17 Oct 2024
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 429 | 117 | 25 | 571 | 56 | 11 | 14 |
- HTML: 429
- PDF: 117
- XML: 25
- Total: 571
- Supplement: 56
- BibTeX: 11
- EndNote: 14
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1