the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 13 Oct 2023
Contribution of brown carbon on light absorption in emissions of European residential biomass combustion appliances
Aki Virkkula
Pasi Yli-Pirilä
Miika Kortelainen
Heikki Suhonen
Laura Kilpeläinen
Mika Ihalainen
Sampsa Väätäinen
Juho Louhisalmi
Markus Somero
Jarkko Tissari
Gert Jakobi
Ralf Zimmermann
Antti Kilpeläinen
Olli Sippula
Abstract. Residential biomass combustion significantly contributes to light-absorbing carbonaceous aerosols in the atmosphere, impacting the Earth's radiative balance at regional and global levels. This study investigates the contribution of brown carbon to the total particulate light absorption in the wavelength range of 370 to 950 nm (BrC370-950) and the particulate absorption Ångström exponents (AAE470/950) in 15 different European residential combustion appliances using a variety of wood-based fuels. BrC370-950 was estimated to be from 1 % to 21 % for wood log stoves and was primarily influenced by fuel moisture content and minimally affected by combustion appliance type. The BrC370-950 contribution was 10 % for a fully automatized residential pellet boiler. Correlations between the ratio of organic to elemental carbon (OC/EC) and BrC indicated that a one unit increase in OC/EC corresponded to approximately a 14 % increase in BrC370-950. Additionally, BrC370-950 increased with decreasing combustion efficiency. AAE470/950 of wood log combustion aerosols ranged from 1.06 to 1.61. By examining the correlation between AAE470/950 and OC/EC, an AAE470/950 close to unity was found for pure black carbon (BC) particles originating from residential wood combustion. This supports the common assumption used to differentiate light absorption caused by BC and BrC. Moreover, diesel aerosols exhibited an AAE470/950 of 1.02, with BrC370-950 contributing only 0.66 % to the total absorption, aligning with the assumption employed in source apportionment. These findings provide important data to assess the BrC370-950 of RWC emissions with different emission characteristics and confirm that BrC can be a major contributor to particulate UV and near-UV light absorption for Northern European wood stove emissions with relatively high OC/EC ratios.
- Preprint
(1388 KB) - Metadata XML
-
Supplement
(1123 KB) - BibTeX
- EndNote
Satish Basnet et al.
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2023-2228', Anonymous Referee #1, 02 Nov 2023
This manuscript by Basnet et al. reports results of emission factors and light-absorption properties of carbonaceous aerosol emissions from residential biomass combustion appliances. The experiments involved an extensive set of fuels (7) and appliances (15). The major measurements involved (i) offline thermal-optical analysis to determine emission factors of elemental carbon (EC) and organic carbon (OC), (ii) online measurements of light-absorption at 7 wavelengths using an aethalometer, and (iii) online measurements of size distributions using a low-pressure impactor. The major analysis involved apportioning absorption to either black carbon (BC) or brown carbon (BrC) based on the assumptions that (i) only BC absorbs at 880 nm and (ii) BC absorptions exhibits a wavelength dependence with AAE = 1. The results show variable contributions to absorption by BC and BrC, with fuel moisture content playing an important role.
Major comments:
1) There are a lot of previous studies that quantified BrC and BC absorption from residential wood burning. It is not clear if/how this study provides any new insights beyond what is already in the literature. For this paper to be suitable for publication in ACP, it needs to clearly identify the new knowledge generated from the experiments. Given the large data set in this study, there is probably potential for deriving new useful knowledge. However, this is not clear in the current version of the paper. A couple of examples of how the paper could potentially highlight new/important/useful results:
1.1) Discuss how emission factors from European residential combustion is currently quantified in emission inventories. Can the results obtained here (Figure 1) help improve these emission inventories?
1.2) The paper mentions that moisture content is more important than the type of appliance in dictating BrC emissions. However, the results (Figures) are not formulated in a way to make use of this finding, or to clearly show that this assertion is valid to begin with. I think that the paper can possibly be restructured to focus on the effect of moisture content versus appliance type. In order to do that, the paper needs to establish the significance of the ranges of moisture contents used in the experiments: Is this variability typical in residential appliances? Data (e.g. BrC absorption) needs to be plotted as a function of moisture content to actually show that moisture content is indeed important and that the data does not cluster based on appliance type.
2) The figures are often hard to follow and are not discussed well in the text. For example:
2.1) Most of the details in Figure 4 are not discussed in the text. Why were these specific 8 experiments chosen?
2.2) It is not clear what the purpose of Figure 5 is. Figure 5, which has 9 panels, is referred to only once in the text, rather in passing.
2.3) What does Figure 7 signify? And what are the data points?
3) What is the physical significance of the dimensionless integrated absorption? If the goal is to quantify the overall contribution to absorption in the atmosphere, the integration has to be performed with respect to the wavelength-dependent intensity of solar radiation. Otherwise, the integration artificially skews the contribution to absorption to shorter wavelengths (because AAE > 0). Usually, experimental results provide wavelength-dependent optical properties, with the understanding that those can be used within radiative transfer calculations that account for the wavelength-dependent solar spectrum. If the authors wish to estimate the integrated contribution to radiative effect by BC and BrC, they could perform ‘simple forcing efficiency’ calculations (e.g. Chen, Y. & Bond, T. C. Light absorption by organic carbon from wood combustion. Atmos. Chem. Phys. 10, 1773–1787 (2010)).
Minor comments:
1) Why not use absorption at 950 nm instead of 880 nm to estimate BC absorption?
2) Line 395-396: The study defines BrC based on the assumption that absorption above the extrapolated BC absorption with AAE = 1 is attributed to BrC. Therefore, it is no surprise that BrC absorption is correlated with AAE. This statement is circular.
Citation: https://doi.org/10.5194/egusphere-2023-2228-RC1 -
RC2: 'Comment on egusphere-2023-2228', Anonymous Referee #2, 03 Nov 2023
This paper presents optical properties of fresh smoke measured with a 7-wavelength aethalometer (AE33) from various wood burning devices using different fuels. Emissions from a diesel engine are also included. Additional measurements include the smoke OC, EC concentrations, CO and CO2 to determine the MCE (and NOx). A significant amount of data is presented. The results are generally as expected, when the fuel is wet (or when fresh wood is placed on the fire) the combustion is more smoldering, MCE drops and OC/EC and BrC increases. Many of the graphs need more explanation or improved clarity. More discussion on the effect of coatings over BC on the overall AAE is needed. Possible effects of tar balls or dark BrC is never discussed. Overall, the paper is interesting and the detailed analysis adds to the data base on residential wood heating emissions, but improvements could be made.
Specific Comments;
Line 41-42 & 46. What is small scale residential combustion? Is it different than low-T biomass/biofuel combustion? Is it RWC?
Line 250, what is evaporation heat of water. I think the term is water latent heat of vaporization.
It is unfortunate that lambda has two separate meanings in this paper. Why not use a different variable for the air to fuel ratio (line 246)?
Line 260, what is the meaning of the ‘ in the denominator in equation 12?
Line 178 and line 312 and table S4, are the particle size distribution data for number or mass distributions? Please specify throughout, but it should be stated clearly in line 178 when discussing the electric low pressure impactor.
Line 332, what exactly is the meaning of indicative? Is the meaning that the magnitude could be biased relative to the real value (whatever that is), but the comparisons between stoves-fuels are ok (ie, like the comparisons shown in Fig 2b)?
Could the variability in MAC880 also be due to the presence of tar balls or dark BrC which does absorb at the highest wavelengths? The possible role of these forms of BrC should be discussed.
Line 343, GMD of what, number or mass? Isn’t mass distribution what matters here?
Fig 2b is not clear. What is the meaning of the overall bar height? Eg, is the MAC with C=3 the sum of the gray and red bars? Needs more explanation.
In Fig 3a the AAE vs OC/EC data looks more correlate then BrC vs OC/EC, opposite to the regression (r2) results given in the plot.
Line 396-399. Is this statement correct, ie does lensing change the AAE to a value different than that of pure BC and is this statement supported by references Virkkula et al and He et al. For example, does a clear coating (could be OA, or other species) on BC result in an AAE different from the AAE of pure BC? If the coating is OA containing chromophores, is the AAE different from pure BC AAE only due to those chromophores absorbing light (the BC core has no effect) or does the coating-BC combination change the overall AAE? This needs clarification, possible by adding more details.
This argument is made in many places, so the question also applies to line 429, line 441 and line 489.
Fig 4 is not very useful, maybe better to extract the important information and move it to the supplement.
Line 458 and 459. This line seems incorrect, or at least the meaning is not clear. The contribution of BrC (ie, b(abs) just due to BrC) at 370 nm is always higher than at 470 nm. The issue here is more related to what wavelength ranges are used for the fit. I believe what is being noted here is that when using the lowest wavelength, the fit produces a lower AAE.
Citation: https://doi.org/10.5194/egusphere-2023-2228-RC2
Satish Basnet et al.
Satish Basnet et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 246 | 78 | 17 | 341 | 31 | 6 | 7 |
- HTML: 246
- PDF: 78
- XML: 17
- Total: 341
- Supplement: 31
- BibTeX: 6
- EndNote: 7
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1