Findings of the African Combustion Aerosol Collaborative Intercomparison Analysis (ACACIA) Pilot Project to Understand the Optical Properties of Biomass Burning Smoke

Fiddler, Marc N.; Moschos, Vaios; McRee, Megan M.; Shawon, Abu Sayeed Md; Gorkowski, Kyle; Lee, James E.; Franco, Nevil A.; Benedict, Katherine B.; Kattel, Samir; Thompson, Chelia; Dubey, Manvendra K.; Bililign, Solomon

doi:10.5194/egusphere-2025-2720

Preprints

https://doi.org/10.5194/egusphere-2025-2720

Preprints

24 Jul 2025

| 24 Jul 2025

Findings of the African Combustion Aerosol Collaborative Intercomparison Analysis (ACACIA) Pilot Project to Understand the Optical Properties of Biomass Burning Smoke

Marc N. Fiddler, Vaios Moschos, Megan M. McRee, Abu Sayeed Md Shawon, Kyle Gorkowski, James E. Lee, Nevil A. Franco, Katherine B. Benedict, Samir Kattel, Chelia Thompson, Manvendra K. Dubey, and Solomon Bililign

Abstract. Africa is a critical source of biomass burning (BB) aerosols, and its importance is increasing. The African Combustion Aerosol Collaborative Intercomparison Analysis (ACACIA) Pilot Project set to optically characterize BB aerosol generated from sub-Saharan African fuels. We used a photoacoustic spectrometer as a reference instrument to determine the multiple-scattering correction factor C_λ for an AE33 aethalometer at three wavelengths, which produced weighted mean values of C₃₇₀ = 3.69, C₄₇₀ = 5.65, and C₅₂₀ = 6.39. C_λ increased with wavelength and C₃₇₀ was statistically independent of the others, suggesting a single C_λ is insufficient, especially in BB scenarios. While a dependence of C_λ on burning state was not found, C_λ was shown to strongly relate to particle single scattering albedo (SSA, ω). When C_λ was plotted against SSA, values slowly rose at low SSA values, followed by a sharp rise around an SSA of ~0.9; indicating a larger correction needed for less absorbing aerosol. A number of functions operating on either SSA or C_λ were explored and the best function was -C_λ/(1‒C_λ) = Aω+B. This is an important parametrization of C_λ specifically geared towards BB aerosol from African fuels under different aging states, and is of particular importance for future field work in that continent. An Ångström matrix plot shows that African BB aerosol can have values more akin to dust, which demonstrates that these fuels are distinct in their wavelength dependence from more typical BB aerosol. Lastly, we examined the mass extinction and absorbance cross sections for BB aerosol generated for the same fuels with two different tube furnace setups. Not only is this combustion method flexible, it was found to be reproducible between labs.

Received: 10 Jun 2025 – Discussion started: 24 Jul 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 1465 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (1465 KB)

Supplement (741 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

15 Jan 2026

Findings of the African Combustion Aerosol Collaborative Intercomparison Analysis (ACACIA) Pilot Project to Understand the Optical Properties of Biomass Burning Smoke

Atmos. Meas. Tech., 19, 307–321, https://doi.org/10.5194/amt-19-307-2026,https://doi.org/10.5194/amt-19-307-2026, 2026

Short summary

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-2720', Anonymous Referee #1, 20 Sep 2025

The study aims to optically characterize biomass burning aerosols from sub-Saharan African fuels, focusing on accurately determining the multiple-scattering correction factor for AE33 aethalometers and its relationship with particle single scattering albedo (SSA). The research develops a parametrization of the correction factor specific to African BB aerosols under different aging conditions, highlights their distinct wavelength dependence. I have following major questions for authors:
1.If emission data from different types of fuel combustion are fitted separately using your fit function, is there a large difference in the fitting quality? Is it possible that the fit works better for one or a few fuel types even if those types are not well-suited? Has the author considered this?
2.The paper mentions that PAM was used to simulate aging experiments. Specifically, what degree of aging equivalent 3days? or 7days? did the authors simulate? During the aging simulation, did the degree of aging vary? Moreover, the authors combined fresh and aged data in the linear relationship shown in Fig. 3, which makes it difficult to see the differences in SSA correlation between fresh and aged emissions. Therefore, it is unclear whether the authors’ statement that the results also apply to aged aerosols is justified. It is recommended that the authors present separate linear fit plots of the fit function for fresh and aged data.
3.First, Fig. 4 is difficult to interpret because the dashed lines are too cluttered. Second, what do the shaded areas represent? Does the pink shading indicate aged aerosols and the grey shading indicate fresh aerosols? From my understanding, there is still considerable discrepancy between the experimental data and the reference data. Since the comparison is made for similar sources, why do the authors’ experimental results differ so much from previous studies in Africa? If AAE and ASE differ substantially, could this affect the general applicability of the fit function to African fuel data?

Citation: https://doi.org/10.5194/egusphere-2025-2720-RC1
- AC1: 'Reply on RC1', Solomon Bililign, 13 Oct 2025
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2720/egusphere-2025-2720-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2025-2720-AC1
RC2:
'Comment on egusphere-2025-2720', Anonymous Referee #2, 13 Oct 2025
This manuscript by Fiddler et al. presents the findings of the ACACIA pilot project, which aimed to characterize the optical properties of BB aerosols from African fuels and to intercompare measurement techniques (AE33 aethalometer vs. photoacoustic reference). The study is significant and timely: Africa is a major source of BB aerosol with growing impact on climate and air quality, yet its aerosol optical properties are less studied. The work fits well within the scope of Atmospheric Measurement Techniques, as it deals with instrument calibration, intercomparison, and methodological advances in aerosol optical measurements. Most of my criticisms are relatively major and should be straightforward to address. I believe the paper will be a valuable contribution to AMT after the authors revise the manuscript in response to the specific comments below.
The use of a three-wavelength photoacoustic spectrometer (PASS-3) as the reference for AE33 absorption measurements is appropriate. However, since the PASS-3 provided data at 405 nm and 532 nm (with the 781 nm channel not used) while the AE33 has channels at 370, 470, 520 nm, the authors extrapolate or interpolate the absorption to those wavelengths (likely assuming a power-law wavelength dependence via an Ångström exponent). Please clarify in the methods how this extrapolation was done and discuss the associated uncertainty. Figure S1 (burn 17 example) is mentioned for extrapolating using AAE – the manuscript should ensure the procedure is described in Section 2.4. How sensitive are the derived Cλ values to the assumed power-law? For example, if the aerosol absorption does not follow a strict power-law between 405 and 532 nm, could that bias the extrapolated α_abs at 370 or 470 nm? Since the PASS measurement uncertainty is noted as ~40%, it would also be useful to comment on how this uncertainty propagates to the reported Cλ. Right now the paper reports variability (standard deviations) of Cλ across runs, but a statement on the absolute accuracy or uncertainty of the Cλ values (accounting for instrument calibration uncertainties) would strengthen confidence in the results.

Tables 3 and 4 contain many functional forms, which makes the findings hard to follow. The manuscript should down-select and focus on the best-performing function (-Cλ/(1–Cλ) = A·ω + B), highlights its advantages, and discuss its potential physical meaning. Otherwise, it may appear overly data-driven without mechanistic insight.

The reported AAE values above 10 are unusual and exceed most prior studies. The manuscript should provide a more careful discussion of the possible causes. Without this, readers may question the reliability of these extreme values. On the other hand, negative or very low ASE might indicate presence of super-micron particles or ash that scatter more efficiently at longer wavelengths – perhaps related to soil/dust or inorganic residue in the fuel (especially dung might contain soil, and savannah grass fires could entrain dust). A short discussion linking these observations to possible physical causes would strengthen the impact. Currently the manuscript states the observation (dust-like optical values) but does not delve into why. Even if a detailed chemical analysis is outside the scope, a sentence like “These extreme Ångström exponent values may result from the high fraction of smoldering organic carbon (leading to very strong wavelength-dependent absorption) and/or the presence of coarse ash particles (leading to anomalous scattering spectra), distinguishing African BB aerosol from typical forest-fire smoke” would be helpful. Additionally, please clarify in the figure/caption how the reader should interpret negative ASE values – some readers might be unfamiliar with the idea that a negative exponent is possible. It might be worth noting that this occurs when larger particle modes cause scattering to increase with wavelength in the measured range (or note if it is within experimental uncertainty). This clarification can prevent confusion.

Table 5 shows MAC discrepancies as large as a factor of 9, which cannot be fully explained by the SMPS upper size cut alone. A dedicated subsection on measurement uncertainties and potential systematic bias would improve transparency and address concerns regarding reproducibility.

Minor coments:
In figures 2 and 3, pls add 95% confidence bands around fit lines.

Use color coding to distinguish fuels, which aid readability.

The NCAT combustion chamber description is too brief and only references McRee et al.(2024). Pls summarize key features, such as chamber volume, RH control, oxidant usage.

When referring to figures in the text, use consistent style (e.g., “Fig. 4” vs “Figure 4”). It looks like the manuscript mostly uses “Figure” spelled out, which is fine. Just ensure each figure is called out in order. I noticed Figure 1 is not explicitly referenced in the portions I read (it might be referenced in Section 2.2 or 2.3 when describing the setups – if not, please include a reference to Fig. 1 in the text so readers know to look at the schematic/configuration).

In Equation formatting, make sure all variables are defined. Equation (3) defines MAC_λ = α_abs,λ / M – earlier in the text, define “M” as the particulate mass concentration (µg m⁻³) if not already done. Likewise, if “MEC” (mass extinction cross-section) is used as a term, define it clearly on first use (I believe it is defined around Eq. 2, but just to be certain).
Citation: https://doi.org/10.5194/egusphere-2025-2720-RC2
- AC2: 'Reply on RC2', Solomon Bililign, 20 Nov 2025
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2720/egusphere-2025-2720-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2025-2720-AC2

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2025-2720', Anonymous Referee #1, 20 Sep 2025

The study aims to optically characterize biomass burning aerosols from sub-Saharan African fuels, focusing on accurately determining the multiple-scattering correction factor for AE33 aethalometers and its relationship with particle single scattering albedo (SSA). The research develops a parametrization of the correction factor specific to African BB aerosols under different aging conditions, highlights their distinct wavelength dependence. I have following major questions for authors:
1.If emission data from different types of fuel combustion are fitted separately using your fit function, is there a large difference in the fitting quality? Is it possible that the fit works better for one or a few fuel types even if those types are not well-suited? Has the author considered this?
2.The paper mentions that PAM was used to simulate aging experiments. Specifically, what degree of aging equivalent 3days? or 7days? did the authors simulate? During the aging simulation, did the degree of aging vary? Moreover, the authors combined fresh and aged data in the linear relationship shown in Fig. 3, which makes it difficult to see the differences in SSA correlation between fresh and aged emissions. Therefore, it is unclear whether the authors’ statement that the results also apply to aged aerosols is justified. It is recommended that the authors present separate linear fit plots of the fit function for fresh and aged data.
3.First, Fig. 4 is difficult to interpret because the dashed lines are too cluttered. Second, what do the shaded areas represent? Does the pink shading indicate aged aerosols and the grey shading indicate fresh aerosols? From my understanding, there is still considerable discrepancy between the experimental data and the reference data. Since the comparison is made for similar sources, why do the authors’ experimental results differ so much from previous studies in Africa? If AAE and ASE differ substantially, could this affect the general applicability of the fit function to African fuel data?

Citation: https://doi.org/10.5194/egusphere-2025-2720-RC1
- AC1: 'Reply on RC1', Solomon Bililign, 13 Oct 2025
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2720/egusphere-2025-2720-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2025-2720-AC1
RC2:
'Comment on egusphere-2025-2720', Anonymous Referee #2, 13 Oct 2025
This manuscript by Fiddler et al. presents the findings of the ACACIA pilot project, which aimed to characterize the optical properties of BB aerosols from African fuels and to intercompare measurement techniques (AE33 aethalometer vs. photoacoustic reference). The study is significant and timely: Africa is a major source of BB aerosol with growing impact on climate and air quality, yet its aerosol optical properties are less studied. The work fits well within the scope of Atmospheric Measurement Techniques, as it deals with instrument calibration, intercomparison, and methodological advances in aerosol optical measurements. Most of my criticisms are relatively major and should be straightforward to address. I believe the paper will be a valuable contribution to AMT after the authors revise the manuscript in response to the specific comments below.
The use of a three-wavelength photoacoustic spectrometer (PASS-3) as the reference for AE33 absorption measurements is appropriate. However, since the PASS-3 provided data at 405 nm and 532 nm (with the 781 nm channel not used) while the AE33 has channels at 370, 470, 520 nm, the authors extrapolate or interpolate the absorption to those wavelengths (likely assuming a power-law wavelength dependence via an Ångström exponent). Please clarify in the methods how this extrapolation was done and discuss the associated uncertainty. Figure S1 (burn 17 example) is mentioned for extrapolating using AAE – the manuscript should ensure the procedure is described in Section 2.4. How sensitive are the derived Cλ values to the assumed power-law? For example, if the aerosol absorption does not follow a strict power-law between 405 and 532 nm, could that bias the extrapolated α_abs at 370 or 470 nm? Since the PASS measurement uncertainty is noted as ~40%, it would also be useful to comment on how this uncertainty propagates to the reported Cλ. Right now the paper reports variability (standard deviations) of Cλ across runs, but a statement on the absolute accuracy or uncertainty of the Cλ values (accounting for instrument calibration uncertainties) would strengthen confidence in the results.

Tables 3 and 4 contain many functional forms, which makes the findings hard to follow. The manuscript should down-select and focus on the best-performing function (-Cλ/(1–Cλ) = A·ω + B), highlights its advantages, and discuss its potential physical meaning. Otherwise, it may appear overly data-driven without mechanistic insight.

The reported AAE values above 10 are unusual and exceed most prior studies. The manuscript should provide a more careful discussion of the possible causes. Without this, readers may question the reliability of these extreme values. On the other hand, negative or very low ASE might indicate presence of super-micron particles or ash that scatter more efficiently at longer wavelengths – perhaps related to soil/dust or inorganic residue in the fuel (especially dung might contain soil, and savannah grass fires could entrain dust). A short discussion linking these observations to possible physical causes would strengthen the impact. Currently the manuscript states the observation (dust-like optical values) but does not delve into why. Even if a detailed chemical analysis is outside the scope, a sentence like “These extreme Ångström exponent values may result from the high fraction of smoldering organic carbon (leading to very strong wavelength-dependent absorption) and/or the presence of coarse ash particles (leading to anomalous scattering spectra), distinguishing African BB aerosol from typical forest-fire smoke” would be helpful. Additionally, please clarify in the figure/caption how the reader should interpret negative ASE values – some readers might be unfamiliar with the idea that a negative exponent is possible. It might be worth noting that this occurs when larger particle modes cause scattering to increase with wavelength in the measured range (or note if it is within experimental uncertainty). This clarification can prevent confusion.

Table 5 shows MAC discrepancies as large as a factor of 9, which cannot be fully explained by the SMPS upper size cut alone. A dedicated subsection on measurement uncertainties and potential systematic bias would improve transparency and address concerns regarding reproducibility.

Minor coments:
In figures 2 and 3, pls add 95% confidence bands around fit lines.

Use color coding to distinguish fuels, which aid readability.

The NCAT combustion chamber description is too brief and only references McRee et al.(2024). Pls summarize key features, such as chamber volume, RH control, oxidant usage.

When referring to figures in the text, use consistent style (e.g., “Fig. 4” vs “Figure 4”). It looks like the manuscript mostly uses “Figure” spelled out, which is fine. Just ensure each figure is called out in order. I noticed Figure 1 is not explicitly referenced in the portions I read (it might be referenced in Section 2.2 or 2.3 when describing the setups – if not, please include a reference to Fig. 1 in the text so readers know to look at the schematic/configuration).

In Equation formatting, make sure all variables are defined. Equation (3) defines MAC_λ = α_abs,λ / M – earlier in the text, define “M” as the particulate mass concentration (µg m⁻³) if not already done. Likewise, if “MEC” (mass extinction cross-section) is used as a term, define it clearly on first use (I believe it is defined around Eq. 2, but just to be certain).
Citation: https://doi.org/10.5194/egusphere-2025-2720-RC2
- AC2: 'Reply on RC2', Solomon Bililign, 20 Nov 2025
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2720/egusphere-2025-2720-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2025-2720-AC2

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Solomon Bililign on behalf of the Authors (20 Nov 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (23 Nov 2025) by Jianhuai Ye

RR by Anonymous Referee #2 (18 Dec 2025)

ED: Publish as is (25 Dec 2025) by Jianhuai Ye

AR by Solomon Bililign on behalf of the Authors (26 Dec 2025)

Journal article(s) based on this preprint

15 Jan 2026

Findings of the African Combustion Aerosol Collaborative Intercomparison Analysis (ACACIA) Pilot Project to Understand the Optical Properties of Biomass Burning Smoke

Atmos. Meas. Tech., 19, 307–321, https://doi.org/10.5194/amt-19-307-2026,https://doi.org/10.5194/amt-19-307-2026, 2026

Short summary

Supplement

https://doi.org/10.5194/egusphere-2025-2720-supplement

Viewed

Total article views: 984 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
848	96	40	984	73	23	27

HTML: 848
PDF: 96
XML: 40
Total: 984
Supplement: 73
BibTeX: 23
EndNote: 27

Views and downloads (calculated since 24 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	97	13	2	112
Aug 2025	183	8	4	195
Sep 2025	424	12	8	444
Oct 2025	54	18	9	81
Nov 2025	33	22	7	62
Dec 2025	30	17	6	53
Jan 2026	27	6	4	37

Cumulative views and downloads (calculated since 24 Jul 2025)

Month	HTML	PDF	XML	Total
Jul 2025	97	13	2	112
Aug 2025	183	8	4	195
Sep 2025	424	12	8	444
Oct 2025	54	18	9	81
Nov 2025	33	22	7	62
Dec 2025	30	17	6	53
Jan 2026	27	6	4	37

Viewed (geographical distribution)

Total article views: 893 (including HTML, PDF, and XML) Thereof 893 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 15 Jan 2026

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (1465 KB)
Metadata XML

Short summary

The study used a photoacoustic spectrometer as a reference instrument to determine the multiple-scattering correction factor C_λ for an AE33 aethalometer at three wavelengths, which we believe is critical for aerosol absorption measurements using aethalometer. This is an important parametrization of C_λ specifically geared towards BB aerosol from African fuels under different aging states, and is of particular importance for future field work in that continent which is at present least studied.


Total:	0
HTML:	0
PDF:	0
XML:	0