From the journal Environmental Science: Atmospheres Peer review history

27-Sep-2021

Dear Dr Ponczek:

Manuscript ID: EA-ART-07-2021-000055
TITLE: Linking chemical composition and optical properties of biomass burning aerosols in Amazonia

Thank you for your submission to Environmental Science: Atmospheres, published by the Royal Society of Chemistry. I sent your manuscript to reviewers and I have now received their reports which are copied below.

I have carefully evaluated your manuscript and the reviewers’ reports, and the reports indicate that major revisions are necessary.

Please submit a revised manuscript which addresses all of the reviewers’ comments. Further peer review of your revised manuscript may be needed. When you submit your revised manuscript please include a point by point response to the reviewers’ comments and highlight the changes you have made. Full details of the files you need to submit are listed at the end of this email.

Please submit your revised manuscript as soon as possible using this link:

*** PLEASE NOTE: This is a two-step process. After clicking on the link, you will be directed to a webpage to confirm. ***

https://mc.manuscriptcentral.com/esatmos?link_removed

(This link goes straight to your account, without the need to log on to the system. For your account security you should not share this link with others.)

Alternatively, you can login to your account (https://mc.manuscriptcentral.com/esatmos) where you will need your case-sensitive USER ID and password.

You should submit your revised manuscript as soon as possible; please note you will receive a series of automatic reminders. If your revisions will take a significant length of time, please contact me. If I do not hear from you, I may withdraw your manuscript from consideration and you will have to resubmit. Any resubmission will receive a new submission date.

The Royal Society of Chemistry requires all submitting authors to provide their ORCID iD when they submit a revised manuscript. This is quick and easy to do as part of the revised manuscript submission process. We will publish this information with the article, and you may choose to have your ORCID record updated automatically with details of the publication.

Please also encourage your co-authors to sign up for their own ORCID account and associate it with their account on our manuscript submission system. For further information see: https://www.rsc.org/journals-books-databases/journal-authors-reviewers/processes-policies/#attribution-id

Environmental Science: Atmospheres strongly encourages authors of research articles to include an ‘Author contributions’ section in their manuscript, for publication in the final article. This should appear immediately above the ‘Conflict of interest’ and ‘Acknowledgement’ sections. I strongly recommend you use CRediT (the Contributor Roles Taxonomy from CASRAI, https://casrai.org/credit/) for standardised contribution descriptions. All authors should have agreed to their individual contributions ahead of submission and these should accurately reflect contributions to the work. Please refer to our general author guidelines http://www.rsc.org/journals-books-databases/journal-authors-reviewers/author-responsibilities/ for more information.

I look forward to receiving your revised manuscript.

Yours sincerely,
Dr Claudia Mohr

Associate Editor, Environmental Science: Atmospheres

************

Reviewer 1

The authors reported on the linkage between chemical composition and optical properties of ambient aerosols in Amazonia based on long-term observation. The Amazonia district is one of the most biomass-burning-influenced areas in the world, and the accumulated data in this study can help people better understand how organic and elemental carbon parts of aerosols contribute to light extinction. A 7-wavelength and a 3-wavelength nephelometer were used to obtain aerosol light absorption and scattering properties, respectively. To separate the light absorption contributed by a different part of aerosols, PMF and MLR methods were used, and both gave comparable results. While the topic of light-absorbing compounds in biomass burning aerosols is very timely and some of the results are excellent, the organization of the contents is a little disordered and the discussion depth to the key data can be deepened. Specifically, there are two concerns. First, section 3.1, and the diurnal profile of mass concentrations of bulk aerosol and PMF components in section 3.4, are irrelevant to the topic of the paper and to the rest of the paper, thus can be omitted for conciseness. The last two paragraphs of section 3.6 are a review of literature references but have little to do with the results shown in this study, thus should also be omitted. Second, the manuscript lacks some novelty. This problem does demise the value of this work, and hopefully can be improved. For reference, a simple direct radiative forcing model can be constructed to evaluate the environmental impact of the radiative effect of aerosols over Amazonia or just a small city. In conclusion, a major revision is suggested before publication in the journal Environmental Science: Atmosphere.
More specific comments about this manuscript are below.
Line 154: Aethalometers have known issues with loading. How do you correct for loading effectsand derive the equivalent BC and BrC contributions? For a recent review of relevant literature, and an example of correction factors, see Li et al Science of the Total Environment 777 (2021) 146143. Please explain the factors used and investigate how the results are affected by the factors used.
Line 180: what density of particles did you use for deriving PM1 and how sensitive are the results to the assumed density. Especially for fresh biomass burning aerosols.

Line 227-230: Please explain how to constrain the size distribution of BC core and the coating thickness. There are countless combinations if you do not explain the assumptions made. Besides, the imaginary part of the refractive index of coating is assumed to be 0.001i, does this assumption match your MAE of BrC? Line 377: Should be written as "interquartile range (IQR)".
Line 433-435: The presence of high m/z 44 and m/z 29 means that aerosols are highly aged, but do not necessarily indicate the source of biomass burning. Besides, P2 received less influence from BB according to back trajectories, but as shown in Figure 7, OOA-2 does not change significantly from P1 to P2. How do the authors interpret this?
Line 445-446: It is not new that primary BB has m/z 44. To confirm that this factor represents primary BB well, you can compare your f44 and f60 values with those of primary BB standard spectra in the literature. For instance, Grieshop 2009 (https://doi.org/10.5194/acp-9-2227-2009) and Fang 2017 (https://doi.org/10.5194/acp-17-14821-2017).
Line 460: Instead of "Graham and coworkers", please cite your reference in a formal way. The same problem is also found in other places, like line 453 and line 470. Please check it all through the paper.
Line 484 (Figure 6): This figure contains a very large area. Can the back-trajectories also be drawn on this map? If some areas are not covered by the trajectory at all, you can leave them out to give readers a better focus.
Line 551: Add your title for Section 3.5 before line 551.
Line 572-573: There should be constraints for this conclusion. What is the range of particle size? How absorbing are your assumed particles?
Line 593-594: Wrong values. Check them according to Figure 7 again.
Line 602: (1) Abundant significant digits; (2) Correct your table 1 value that corresponds to 4.85.
Line 661-662: This shows ignorance in this field. Just delete this sentence.
Line 750: MLR.
Line 771-772: Can you check your reference again? The point that "HULIS is responsible for BC" is highly doubtful.

Reviewer 2

See attached comments.

Disclaimer: This text has been copied from the PDF response to reviewers and does not include any figures, images or special characters.

Referee 1
Comments to the Author
Comment: The authors reported on the linkage between chemical composition and optical properties of ambient
aerosols in Amazonia based on long-term observation. The Amazonia district is one of the most biomassburning-influenced areas in the world, and the accumulated data in this study can help people better
understand how organic and elemental carbon parts of aerosols contribute to light extinction. A 7-
wavelength and a 3-wavelength nephelometer were used to obtain aerosol light absorption and scattering
properties, respectively. To separate the light absorption contributed by a different part of aerosols, PMF
and MLR methods were used, and both gave comparable results.
While the topic of light-absorbing compounds in biomass burning aerosols is very timely and some of the
results are excellent, the organization of the contents is a little disordered and the discussion depth to the
key data can be deepened. Specifically, there are two concerns.
First, section 3.1, and the diurnal profile of mass concentrations of bulk aerosol and PMF components in
section 3.4, are irrelevant to the topic of the paper and to the rest of the paper, thus can be omitted for
conciseness. The last two paragraphs of section 3.6 are a review of literature references but have little to
do with the results shown in this study, thus should also be omitted. Second, the manuscript lacks some
novelty. This problem does demise the value of this work, and hopefully can be improved. For reference,
a simple direct radiative forcing model can be constructed to evaluate the environmental impact of the
radiative effect of aerosols over Amazonia or just a small city. In conclusion, a major revision is suggested
before publication in the journal Environmental Science: Atmosphere.

Response: We thank Referee #1 for the constructive comments. You will find below in detail our answers in blue
color.
We would like to underline the innovative content the manuscript brings. We understand that those
points were not optimally emphasized in the main body of the manuscript. Therefore, we added this
content in the last paragraph of the introductory section.
Our results provide a detailed characterization of fine mode aerosols from field measurements at ground
level in the southwestern Amazonia. Moreover, we have integrated the results provided by the chemical
and physical characterization attributing optical properties to each organic component of the aerosols,
showing their spectral dependency. We would like to highlight the following novelties and original results:
● Our study reports the detailed and real-time chemical characterization of aerosols during a period
of important fire occurrences in the deforestation arc of Amazonia.
● This is the first study to report such a comprehensive analysis of the links between optical and
chemical properties of aerosols in a Southern region under intense pressure from land-use
changes through deforestation and wildfires.
● Our study reports the scattering and absorption efficiencies attributed to each chemical
component relevant to the PM1 population. Furthermore, we show the spectral dependence of
these efficiencies and the percentage contribution of each chemical component in relation to the
scattering and absorption coefficients.
● We estimate the contribution of BrC to the total absorption of the fine fraction of aerosols due to
biomass burning in the deforestation arc in Amazonia. Although this is not the first report of BrC
in the region, results of this type are still rare. In addition, we show that the optically defined BrC
is related to the measured organic component, and is, foremost, associated with biomass burning
aerosols.
● Our results for the scattering and absorption efficiencies provide important inputs for the
development of radiative balance models, and, therefore, advance towards understanding
parameters required to represent climate forcing of BB aerosols in tropical forests.
Addressing the question regarding the Radiative Forcing Model
We would like to thank reviewer #1 for the suggestion of estimating the direct radiative forcing (DRF) to
improve the novelty of this manuscript. We would like to clarify, however, that evaluating the effect of
biomass burning (BB) aerosols in Rio Branco on the radiative forcing is beyond the scope of our study. In
fact, addressing this question is indeed interesting and may be an output of the current work, being the
subject of future studies by our research group.
Instead of developing a new model to calculate the direct radiative forcing (DRF) of aerosols in Rio Branco,
we estimated it from AERONET’s inversion model for the period of the campaign. Figure X shows the time
series of DRF at the top of the atmosphere (TOA) obtained with level 2 data (AERONET’s highest quality).
In fact, the covered period is mostly regarded as period 1 of the campaign (08/22 - 09/29), comprising the
months with the highest number of fire spots nearby the experimental site. The estimated average DRF is
-16.4 ± 11.6 Wm-2
, indicating that the BB aerosols are efficient in cooling the atmospheric system. As a
comparison, Sena et al. (2013)1
, using 10 years of remote sensing data for the Amazon basin, obtained the
DRF for the BB season (August to September) of -5.6 ± 1.7 Wm-2
. In addition, Palacios et al. (2020)2
estimated the DRF using AERONET data for the dry season in Central Amazonia as -9.18 ± 2.80 Wm-2
. Both
studies agree well with the value we obtained within the standard deviation.
Figure 1: Direct radiative forcing obtained from the AERONET's inversion model at the top of the
troposphere (TOA) during the period of the campaign.

Comment: More specific comments about this manuscript are below.
1. Line 154: Aethalometers have known issues with loading. How do you correct for loading effects and
derive the equivalent BC and BrC contributions? For a recent review of relevant literature, and an
example of correction factors, see Li et al Science of the Total Environment 777 (2021) 146143.
Please explain the factors used and investigate how the results are affected by the factors used.

Response: Many thanks for the observation. Indeed, the aethalometers have known issues with the loading effects.
We corrected them based on the same methodology applied in Rizzo et al. (2011)
3
, Saturno et al., (2017)
4
,
and Saturno et al., (2018)
5
. We acknowledge that clarification about the AE33 corrections were missing.
We addressed it by describing the whole set of corrections (with the loading effects included) as follows:
“The aerosol absorption coefficients were calculated taking into account corrections for artifacts
such as multiple scattering effects, and filter-loading effects, etc. The correction applied here has
already been described in detail by Saturno et al., (2018), , Saturno et al., (2017), Rizzo et al (2011),
briefly, MAAP data was used as a reference measure to correct for multiple scattering effects, and
then, retrieving the σabs through a log – adjustment log to aethalometer data. Corrections for filter
loadings are not necessary, as the AE33 model uses dual-spot technology that already takes this
effect into account.”
Therefore, the equivalent BC was obtained from the referred correction and the BrC contribution was
obtained from the methodology described in section 2.3 “Brown Carbon Estimation”. In particular, it is
worth mentioning that the methodology used in this study for deriving BrC contribution is different from
the one used in Li et al., 2021. Moreover, the main instrumentation is also different: in this study, we used
the conventional and widely-used AE33 (Magee Scientific), while in the study from Li et al., (2021), they
used a micro-aethalometer, model MA200 (AETHLAB), to obtain the equivalent black carbon and to test
its performance to obtain the BrC fraction. It is also worth mentioning, the methodology used in our study
is already validated in the literature. It was developed by Wang et al., 2016, who derived the BrC
component using both AE33 and AERONET’s photometer data. In addition, Saturno et al., (2018)
5
, applied
the same methodology with slightly different wavelengths’ pairs to derive the BrC component in Central
Amazônia.

Comment: 2. Line 180: what density of particles did you use for deriving PM1 and how sensitive are the results to
the assumed density. Especially for fresh biomass burning aerosols.

Response: The comparison between instruments (PM1 mass closure) is shown in Figure S2 in the Supplement. The
PM1 mass concentration provided by the non-refractory-PM1 (from ACSM) + BC (from the AE33) was
compared to total mass concentration (from the SMPS) estimated using the integrated volume from the
particle size distribution, assuming particle density of 1.5 g cm−3. The value was chosen based on what
was reported by Brito et al., (2014)6
.
By Figure S2, it is possible to note that for the days when there was a greater influence of BB (blue
points), the aerosol seems to have a lower density.
Before using the value of 1.5 g/ cm3
, the density was calculated according to the chemical composition
of PM1, as follows:
1 = ( . + . + . + . + ℎ . ℎ +
. )/ 1
Eq (1)
The density of each component (in g /cm³) was considered as:
= 1.78 ; = 1.72 ; = 1.72 ; ℎ = 1.52 ; = 1.77
was attemptedly estimated based on the oxygen-to-carbon (O : C) and hydrogen-to-carbon (H : C)
ratio as proposed by Kuwata et al. ( 2012). However, H:C and O:C ranges were outside the limits
specified in the correlation proposed by Kuwata, et al., (2012)
7
, therefore, we chose not to use the
variable density.
For addressing the question issued by the reviewer, we made a sensibility analysis, considering the
following criteria:
1 = 1.0 >

1 = 1.5 <=

The values were chosen to represent a fresh aerosol (lower density) and a more processed aerosol
(denser). The BBOA factor from PMF was used as a criterion because its concentration is a proxy for the
occurrence of biomass burning.
Using this criterion, the mean value of dens was 1.44 g/cm³, and thus, represents approx. 4% variation
from the density value we have originally used in our calculations for PM1 mass closure. Thefore, we
assume that the average value 1.5 g/cm3 is reasonable for this study.

Comment: 3. Line 227-230: Please explain how to constrain the size distribution of BC core and the coating
thickness. There are countless combinations if you do not explain the assumptions made. Besides,
the imaginary part of the refractive index of coating is assumed to be 0.001i, does this assumption
match your MAE of BrC?

Response: In this study, we calculated the BrC contribution based on the methodology developed by Wang et al.,
(2016)
8 which was applied to the Central Amazonian aerosols by Saturno et al., (2018)
5
. In particular,
Saturno et al., (2018) evaluated the contribution of the BC core and coating thickness to the WDA, and
also estimated the coating for the particles in Central Amazônia, at the ATTO site, using SP2
measurements. The results are well-described in the supplementary material of the aforementioned
study. In the experimental site based at Rio Branco, Brazil, we did not have an SP2 available, so we used
similar parameters obtained for Central Amazônia. We recognize that there are limitations in this
approach, especially related to the real thickness of the aerosol coating, and further studies are required
to overcome them.
However, in the dry season, biomass burning dominates the aerosol sources in the Amazon basin (Pöhkler
et al., 20169
, Holanda et al., 202010, Rizzo et al., 201311, Artaxo et al., 201312
, Rizzo et al., 20113
). Thus, it
is a reasonable assumption to consider the sources and the aerosol physicochemical characteristics as
similar to those from Saturno et al., (2018)
5
. The assumption of the imaginary part of the coating as being
0.001i is based on the same assumptions made by Want et al., 2016 8 and Saturno et al., 2018 5
, which
applied the same methodology as we did in our study, whose value was extracted from Liu et al., 201513
.
In addition, as far as we know, our study is the only one that estimated the BrC contribution from in-situ
measurements in the Amazonian deforestation arc during the dry season.

Comment: 4. Line 377: Should be written as "interquartile range (IQR)".

Response: Thanks for the remark.

Comment: 5. Line 433-435: The presence of high m/z 44 and m/z 29 means that aerosols are highly aged, but do
not necessarily indicate the source of biomass burning. Besides, P2 received less influence from BB
according to back trajectories, but as shown in Figure 7, OOA-2 does not change significantly from
P1 to P2. How do the authors interpret this?

Response: We acknowledge the reviewer for pointing that out. We agree that the hypothesis that OOA-2 was
related to BB was weakly described, therefore, we have revised this section of the manuscript,
rephrasing the text in order to further explain this factor. In fact, the OOA-2 did not present any signal in
the m/zs 60 and 73, which are the major BBOA tracers, therefore the new description was adjusted
accordingly. The new paragraph describing the OOA-2 is below.
“The last factor obtained was the OOA-2. This factor presents prominent signals at m/zs 44 (CO2+),
18 (H2O+), 29 (COH+ and/or C2H5+) and 43 (C2H3O+ and/or C3H7+). In terms of f44 this factor
presents a lower value (19%) compared to the OOA-1 (32%), suggesting it is less oxygenated than
the OOA-1. This fact is also confirmed by the larger ratio between m/zs 44 and 43 obtained for the
OOA-1, and the larger content of a less oxygenated fragment, the m/z 29. Despite the similarities
between the OOA-2 and OOA-1 in terms of mass spectra, the time series does not present good
agreement (R=0.33). In fact, the diel profiles present very different behaviors in the P1 and P2
(Figure XX). The OOA-2 presents larger mass concentration values during the nighttime over the P1,
while over the P2 no clear pattern can be observed.”

Comment: 6. Line 445-446: It is not new that primary BB has m/z 44. To confirm that this factor represents
primary BB well, you can compare your f44 and f60 values with those of primary BB standard
spectra in the literature. For instance, Grieshop 2009 (https://doi.org/10.5194/acp-9-2227-2009)
and Fang 2017 (https://doi.org/10.5194/acp-17-14821-2017).

Response: We acknowledge the referee for this observation and for the references indicated. In fact, the BBOA
found in Rio Branco presents 14.8% of f44, indicating the presence of oxygenated species. As suggested,
the paragraph below was, therefore, insured in the text after line 451.
“In chamber experiments, it has been observed that fresh particles emitted by biomass burning show
a contribution of m/z44 in its mass spectra (Grieshop et al, 2009 14, Fang et al. 2017 15). Fang et al.,
(2017) showed that primary particles emitted by burning agricultural residues contained an average
of 23% of oxygenated compounds. In both studies, the proportion of f44 progressively increases as a
function of aging time. Accordingly, Timonen et al., (2013)16 found 2 well-defined factors related to
BB for a field study in the spring in Helsinki, one of them of local influence and the other of longdistance transport, also there, both BBOA factors had a significant contribution of oxygenated
compounds.”

comment: 7. Line 460: Instead of "Graham and coworkers", please cite your reference in a formal way. The same
problem is also found in other places, like line 453 and line 470. Please check it all through the
paper.
Response: Thanks for pointing that out. The citations were corrected.

Comment: 8. Line 484 (Figure 6): This figure contains a very large area. Can the back-trajectories also be drawn on
this map? If some areas are not covered by the trajectory at all, you can leave them out to give
readers a better focus.

Response: The figures have been updated to restrict the covered area. We chose not to overlay the backtrajectories on the same figure, as this would make it very polluted, making it difficult to visualize
important information. On the other hand, we made available an mp4 file that shows, in video format,
the trajectories on the map and the active fire spots during each back-trajectory period.
The video is available at
https://www.dropbox.com/s/4f0w201951rgn3g/trmm_winds_fires_anim.mp4?dl=0
Animation showing the simulated backward trajectories, fire spots, and precipitation, from 28th August
to 5th November 2018. Trajectories start every 30 min and are followed for 72h. Precipitation rate is
shown as contours for 4, 8, 12, and 16 mm/day, and winds at 950 hPa are shown as black arrows, averaged
between the start and end time of each trajectory. Fire spots are colored by their radiative power and are
accumulated between the start and end time of each trajectory. Red boxes indicate the 100 x 100 km
interest regions used to count the nearby fire spots around each trajectory point and are shown only at
6h steps.

Comment: 9. Line 551: Add your title for Section 3.5 before line 551.

Response: This section was restructured.

Comment: 10. Line 572-573: There should be constraints for this conclusion. What is the range of particle size?
How absorbing are your assumed particles?

Response: We thank the reviewer for the observation. Indeed, the Angstrom exponent is qualitatively used as an
indicator of particle size for aerosol measurements including both the fine and the coarse modes. This
was not the case in our set of measurements. Accordingly, we decided to exclude lines 572-573, and
simply compared the observed range of SAE values with previous studies in the Amazon region.

Comment: 11. Line 593-594: Wrong values. Check them according to Figure 7 again

Response: The names of periods (1 and 2) were wrong. Fixed for:
“which indicates the contribution of 39% of OOA-1 in Period 1, against 50% in Period 2.”
Comment: 12. Line 602: (1) Abundant significant digits; (2) Correct your table 1 value that corresponds to 4.85.

Response: The values have been corrected.

Comment: 13. Line 661-662: This shows ignorance in this field. Just delete this sentence.

Response: The sentence was removed.

Comment: 14. Line 750: MLR.

Response: The acronym was corrected.

Comment: 15. Line 771-772: Can you check your reference again? The point that "HULIS is responsible for BC" is
highly doubtful.

Response: This paragraph was restructured and the phrases in question were excluded. Now from line 769, it reads
as follows:
“Laboratory-controlled studies burning different vegetation species identified a variety of
oxygenated hydrocarbons 88 with diverse chemical characteristic such as lignin pyrolysis products
lignin-derived products, distillation products, nitroaromatics, and PAHs 91 emphasizing that most
BrC chromophores were common to samples from distinct vegetation species meanwhile their
relative contributions to total light absorption differ.
Thus, the formation of chromophore compounds in BB has mechanisms common to all types of
fuel/fire and exhibits unique properties that need to be addressed concerning local
characteristics. Regarding the overall fine particulate matter absorption, we cannot rule out the
contribution of other OA components to the absorption of light as many studies have assessed
atmospheric chemical processes that resulted in light-absorbing SOA 25,80. The presence of BrC,
therefore, should not be neglected, and studies in the Amazon basin, especially in the dry season
when massive biomass burning occurs, should investigate this characteristic of the regional
aerosol addressing its climatic effects.”

References
1 E. T. Sena, P. Artaxo and A. L. Correia, Spatial variability of the direct radiative forcing of biomass
burning aerosols and the effects of land use change in Amazonia, Atmos. Chem. Phys., 2013, 13,
1261–1275.
2 R. da S. Palácios, P. E. Artaxo, G. G. Cirino, V. Nakale, F. G. Morais, L. D. Rothmund, A. Y. Siqueira,
M. S. Biudes, N. G. Machado, L. F. A. Curado, J. B. Marques and J. de S. Nogueira, Long-term
measurements of aerosol optical properties and radiative forcing (2011-2017) over Central
Amazonia, Atmósfera, 2020, 0–27.
3 L. V. Rizzo, A. L. Correia, P. Artaxo, A. S. Procópio and M. O. Andreae, Spectral dependence of
aerosol light absorption over the Amazon Basin, Atmos. Chem. Phys., 2011, 11, 8899–8912.
4 J. Saturno, C. Pöhlker, D. Massabò, J. Brito, S. Carbone, Y. Cheng, X. Chi, F. Ditas, I. Hrabě de
Angelis, D. Morán-Zuloaga, M. L. Pöhlker, L. V. Rizzo, D. Walter, Q. Wang, P. Artaxo, P. Prati and
M. O. Andreae, Comparison of different Aethalometer correction schemes and a reference multiwavelength absorption technique for ambient aerosol data, Atmos. Meas. Tech., 2017, 10, 2837–
2850.
5 J. Saturno, B. A. Holanda, C. Pöhlker, F. Ditas, Q. Wang, D. Moran-Zuloaga, J. Brito, S. Carbone, Y.
Cheng, X. Chi, J. Ditas, T. Hoffmann, I. Hrabe De Angelis, T. Könemann, J. V. Lavrič, N. Ma, J. Ming,
H. Paulsen, M. L. Pöhlker, L. V. Rizzo, P. Schlag, H. Su, D. Walter, S. Wolff, Y. Zhang, P. Artaxo, U.
Pöschl and M. O. Andreae, Black and brown carbon over central Amazonia: Long-term aerosol
measurements at the ATTO site, Atmos. Chem. Phys., 2018, 18, 12817–12843.
6 J. Brito, L. V. Rizzo, W. T. Morgan, H. Coe, B. Johnson, J. Haywood, K. Longo, S. Freitas, M. O.
Andreae and P. Artaxo, Ground-based aerosol characterization during the South American
Biomass Burning Analysis (SAMBBA) field experiment, Atmos. Chem. Phys., 2014, 14, 12069–
12083.
7 M. Kuwata, S. R. Zorn and S. T. Martin, Using elemental ratios to predict the density of organic
material composed of carbon, hydrogen, and oxygen, Environ. Sci. Technol., 2012, 46, 787–794.
8 X. Wang, A. J. Sedlacek, S. S. de Sá, S. T. Martin, M. L. Alexander, M. L. Alexander, T. B. Watson, A.
C. Aiken, S. R. Springston and P. Artaxo, Deriving brown carbon from multiwavelength absorption
measurements: Method and application to AERONET and Aethalometer observations, Atmos.
Chem. Phys., 2016, 16, 12733–12752.
9 M. L. Pöhlker, C. Pöhlker, F. Ditas, T. Klimach, I. Hrabe de Angelis, A. Araújo, J. Brito, S. Carbone,
Y. Cheng, X. Chi, R. Ditz, S. S. Gunthe, J. Kesselmeier, T. Könemann, J. V. Lavrič, S. T. Martin, E.
Mikhailov, D. Moran-Zuloaga, D. Rose, J. Saturno, H. Su, R. Thalman, D. Walter, J. Wang, S. Wolff,
H. M. J. Barbosa, P. Artaxo, M. O. Andreae and U. Pöschl, Long-term observations of cloud
condensation nuclei in the Amazon rain forest – Part 1: Aerosol size distribution, hygroscopicity,
and new model parametrizations for CCN prediction, Atmos. Chem. Phys., 2016, 16, 15709–
15740.
10 B. A. Holanda, M. L. Pöhlker, D. Walter, J. Saturno, M. Sörgel, J. Ditas, F. Ditas, C. Schulz, M. A.
Franco, Q. Wang, T. Donth, P. Artaxo, H. M. J. Barbosa, S. Borrmann, R. Braga, J. Brito, Y. Cheng,
M. Dollner, J. W. Kaiser, T. Klimach, C. Knote, O. O. Krüger, D. Fütterer, J. V. Lavrič, N. Ma, L. A. T.
Machado, J. Ming, F. G. Morais, H. Paulsen, D. Sauer, H. Schlager, J. Schneider, H. Su, B. Weinzierl,
A. Walser, M. Wendisch, H. Ziereis, M. Zöger, U. Pöschl, M. O. Andreae and C. Pöhlker, Influx of
African biomass burning aerosol during the Amazonian dry season through layered transatlantic
transport of black carbon-rich smoke, Atmos. Chem. Phys., 2020, 20, 4757–4785.
11 L. V. Rizzo, P. Artaxo, T. Müller, A. Wiedensohler, M. Paixão, G. G. Cirino, A. Arana, E. Swietlicki, P.
Roldin, E. O. Fors, K. T. Wiedemann, L. S. M. Leal and M. Kulmala, Long term measurements of
aerosol optical properties at a primary forest site in Amazonia, Atmos. Chem. Phys., 2013, 13,
2391–2413.
12 P. Artaxo, L. Rizzo, J. Brito, H. M. J. Barbosa, A. Arana, E. T. Sena, G. G. Cirino, W. Bastos, S. T.
Martin and M. O. Andreae, Atmospheric aerosols in Amazonia and land use change: from natural
biogenic to biomass burning conditions, Faraday Discuss., 2013, 165, 203.
13 D. Liu, J. W. Taylor, D. E. Young, M. J. Flynn, H. Coe and J. D. Allan, The effect of complex black
carbon microphysics on the determination of the optical properties of brown carbon, Geophys.
Res. Lett., 2015, 42, 613–619.
14 A. P. Grieshop, N. M. Donahue and A. L. Robinson, Laboratory investigation of photochemical
oxidation of organic aerosol from wood fires 2: Analysis of aerosol mass spectrometer data,
Atmos. Chem. Phys., 2009, 9, 2227–2240.
15 Z. Fang, W. Deng, Y. Zhang, X. Ding, M. Tang, T. Liu, Q. Hu, M. Zhu, Z. Wang, W. Yang, Z. Huang,
W. Song, X. Bi, J. Chen, Y. Sun, C. George and X. Wang, Open burning of rice, corn and wheat
straws: Primary emissions, photochemical aging, and secondary organic aerosol formation,
Atmos. Chem. Phys., 2017, 17, 14821–14839.
16 H. Timonen, S. Carbone, M. Aurela, K. Saarnio, S. Saarikoski, N. L. Ng, M. R. Canagaratna, M.
Kulmala, V. M. Kerminen, D. R. Worsnop and R. Hillamo, Characteristics, sources and watersolubility of ambient submicron organic aerosol in springtime in Helsinki, Finland, J. Aerosol Sci.,
2013, 56, 61–77.
17 L. Varanda Rizzo, P. Roldin, J. Brito, J. Backman, E. Swietlicki, R. Krejci, P. Tunved, T. Petäjä, M.
Kulmala and P. Artaxo, Multi-year statistical and modeling analysis of submicrometer aerosol
number size distributions at a rain forest site in Amazonia, Atmos. Chem. Phys., 2018, 18, 10255–
10274.
18 M. Franco, F. Ditas, L. A. Kremper, L. Machado, M. Andreae, A. Araújo, H. Barbosa, J. de Brito, S.
Carbone, B. Holanda, F. Morais, J. Nascimento, M. Pöhlker, L. Rizzo, M. Sá, J. Saturno, D. Walter,
S. Wolff, U. Pöschl, P. Artaxo and C. Pöhlker, Occurrence and growth of sub-50 nm aerosol
particles in the Amazonian boundary layer, Atmos. Chem. Phys., 2021, 1–36.
19 J. V. Martins, P. Artaxo, Y. J. Kaufman, A. D. Castanho and L. A. Remer, Spectral absorption
properties of aerosol particles from 350–2500nm, Geophys. Res. Lett., 2009, 36, L13810.
20 L.-W. A. Chen, J. C. Chow, X. Wang, J. Cao, J. Mao and J. G. Watson, Brownness of Organic Aerosol
over the United States: Evidence for Seasonal Biomass Burning and Photobleaching Effects,
Environ. Sci. Technol., 2021, 55, 8561–8572.
21 M. Claeys, R. Vermeylen, F. Yasmeen, Y. Gómez-González, X. Chi, W. Maenhaut, T. Mészáros and
I. Salma, Chemical characterisation of humic-like substances from urban, rural and tropical
biomass burning environments using liquid chromatography with UV/vis photodiode array
detection and electrospray ionisation mass spectrometry, Environ. Chem., 2012, 9, 273–284.
22 G. R. McMeeking, S. M. Kreidenweis, C. M. Carrico, J. L. Collett, D. E. Day and W. C. Malm,
Observations of smoke-influenced aerosol during the Yosemite Aerosol Characterization Study: 2.
Aerosol scattering and absorbing properties, J. Geophys. Res. D Atmos., 2005, 110, 1–9.
23 J. L. Hand and W. C. Malm, Review of aerosol mass scattering efficiencies from ground-based
measurements since 1990, J. Geophys. Res. Atmos., , DOI:10.1029/2007JD008484.
24 W. C. Malm, D. E. Day, C. Carrico, S. M. Kreidenweis, J. L. Collett, G. McMeeking, T. Lee, J. Carrillo
and B. Schichtel, Intercomparison and closure calculations using measurements of aerosol
species and optical properties during the Yosemite aerosol characterization study, J. Geophys.
Res. D Atmos., 2005, 110, 1–21.
25 M. Ealo, A. Alastuey, N. Pérez, A. Ripoll, X. Querol and M. Pandolfi, Impact of aerosol particle
sources on optical properties in urban, regional and remote areas in the north-western
Mediterranean, Atmos. Chem. Phys., 2018, 18, 1149–1169.

Referee 2
General comment
The authors conducted a field campaign in Amazonian Basin measuring the aerosol chemical
composition and optical properties. By combining field measurements and advanced statistical
techniques (such as PMF and MLR), they investigated the relationships between atmospheric
aerosols' chemical and physical properties. They tried to address how the sources and
atmospheric processes influence the chemical composition and further modify the optical
properties. This is one of the hotspots in atmospheric studies. This is a good contribution to this
field. However, the organization of the manuscript can be improved, and some other issues must
be addressed before it can be published. Thus, a major revision is needed.
Major comment
Some of the materials in the manuscript are not well focused on the topic “Linking chemical
composition and optical properties of biomass burning aerosols in Amazonia”.
The authors presented in the manuscript in section 3.1 about traces gases, in section 3.4 about
Seasonal and diurnal variation associated with fire occurrence. These sections are helpful for the
understanding of the background of the study, the sources, and possibly the chemistry. But these
efforts are not well-organized in a way that supports the major points of this study. The
information on trace gases is never used by the later part of the discussion. This could go to
supporting information. The authors put the information about how they determine the campaign
to period 1 and 2 based on the fires spots in the results section. This could be introduced in the
experimental part already in the section where the authors define the sampling periods. With this
as background information, it is easier for readers to understand the time series/seasonal results
of the aerosols. Moreover, the authors present a lot of diurnal trends for the measured items. I
recognized these are interesting results. However, this information should be presented in a way
to support the source apportionment of OA, scattering, and absorption of aerosols, instead of
presenting these results in long paragraphs as a stand-alone part

Response: We would like to thank Referee #2 for constructive comments and suggestions. We took into
account the suggestions, restructuring several sections of the ‘Experimental’ and ‘Results and
Discussion’ such as sec 3.4, 3.5 and 3.6, suppressing section 3.1, among other modifications.
Also, we would like to acknowledge the referee for the suggestion to include black carbon in the
MLR for the scattering coefficient. The model was redone considering BC as a predictor variable
providing consistent results. Find below in detail our answers and comments in blue color.

Specific comments
1. Line 111. Reference needs to be formatted.

Response: The references were formatted.

Comment: 2. L153-155: Insufficient details are provided regarding all of the corrections applied for the
aethalometer measurements. It is well established in many studies that the multiple scattering
correction for the Aethalometer varies in time and space: it is not a constant, but instead
depends on the nature of the particles being sampled. Moreover, the loading effect was not
account for potential biases that can occur in filter-based instruments when the OA/BC ratio
is large (Lack et al., AS&T, 2008).

Response: Many thanks for your observations and we are very sorry that the sentences were not clear
enough. Indeed, we applied the corrections for multiple scattering issues and for loading
effects, according to Rizzo et al., 20113
, 201311. We also added more details in the manuscript
about the corrections, as below:
“The aerosol absorption coefficients were calculated taking into account corrections for
artifacts such as multiple scattering effects, and filter-loading effects. The correction
applied here has already been described in detail by Saturno et al., (2018)33, Saturno et al.,
(2017)34, Rizzo et al., (2011)35, briefly, a period sampled with MAAP was used as a
reference measure to correct for multiple scattering effects, and then, retrieving the σabs
through a log – adjustment to aethalometer data. Corrections for filter loadings are not
necessary, as the AE33 model uses dual-spot technology that already takes this effect into
account. Therefore, the black carbon equivalent concentration (herein referred to as BC)
was retrieved from the corrected Aethalometer data.”

Comment: 3. Line 162. What is the reason to calculate the SSA for 637 nm? This is not the working
wavelength for either instrument used by this study.

Response: The SSA can be calculated for any wavelength since it represents the relative effect of particle
scattering relative to extinction (scattering + absorption). The use of multiwavelength
aethalometers such as the AE33 used in this study, however, is relatively more recent, while
637 nm is the wavelength used by MAAP, the most widely used instrument for measuring
aerosol absorption coefficients. Therefore, 637 nm is the wavelength in which absorption
coefficients are commonly reported in the literature. See as examples the following studies
characterizing aerosols in Amazonia: Saturno et al, 2018 5
; Rizzo et al., 2011 3
; Rizzo et al
201311; Artaxo et al 2013 12
.
Comment: 4. Line 170-171. The author stated that “Intercomparison of SPMS and CPC particle
concentration measurements were used for data validation.” However, the results are not
provided regarding the intercomparison. Do they use the CPC data only for validation or
used it for the correction of the SMPS data?

Response: Many thanks for your observation. In fact, the comparisons between SMPS and CPC data were
a fundamental part of the data validation process. Since the SMPS we used was set to measure
particles between 10 - 400 nm, and an independent CPC was connected to a PM2.5 inlet line,
we considered that the SMPS integrated aerosol number concentration must be within 15 -
20% of the independent CPC measurement. This assumption is supported by many studies in
the literature, see for reference (Rizzo et al., 2018 17; Franco et al., 2021 18). The results were
not shown in the previous version of the manuscript, as the comparisons were performed for
instrumental data quality control purposes.
Thus, we show below in Figure 2 the linear regression intercomparison between the
independent CPC and the SMPS integrated total number concentration data.
Figure 2: linear regression between an independent CPC and the SMPS integrated total
number concentration data scatter plot. The colorbar represents the days covered in the
measurement campaign.
The linear regression with the linear coefficient set to 0 resulted in an angular coefficient of
0.948 ± 0.003, with R² = 0.98 and p-value of ~ 0 (1.2e-35). This shows a very good agreement
between the instruments, and it supports our results in the manuscript.

Comment: 5. Line 226-232. Details about the WDA calculation for BC are not clear. What are the size
distributions used for the calculation? What is the core diameter and shell diameter used?
Eq6, WDA should be replaced with "BC WDA". The equation assumes that absorption over
660 nm is only due to BC, without BrC. How reliable is this assumption? The refractive
index for BC is 1.95 – 0.79i instead of “1.95 – 0.76i”. The authors should check if this is only
a typo or the wrong value has been used in their calculation. The authors did not assume a fix
AAE of BC ~1 as many studies do. I wonder what is the AAE values for BC calculated by
this study? Is it significantly different from 1?
Line 226-232. Details about the WDA calculation for BC are not clear.

Response: We are sorry that these lacking details made the methodology difficult to understand. We
didn't detail the calculations because we used the IDL (Interactive Data Language) routines
and the Mie simulations developed by Wang et al., 2016 8 which a thorough description of the
methodology is available in the ACP paper “Deriving brown carbon from multiwavelength
absorption measurements: method and application to AERONET and Aethalometer
observations”, (DOI: doi:10.5194/acp-16-12733-2016). To make our manuscript clearer, we
modified the text as follows:
“To obtain the BrC contribution) to the light absorption coefficients, we applied the same
methodology and parameters as Wang et al.(2016) 40, who considers the total absorption
coefficient measured by the Aethalometer at 470 nm (abs, 470 nm) as a sum of the
contributions of the BC (BC σabs, 470 nm) and BrC (BrCabs, 470 nm), as follows:
,470 = ,470 + ,470 Eq. 3”

Comment: What are the size distributions used for the calculation? What is the core diameter and shell
diameter used?

Response: Wang et al., 2016 performed numerous Mie simulations varying the size distribution to have
a better estimation of the underlying uncertainties of the technique. They assumed a single
log-normal distribution with mean diameter varying from 20 to 300 nm, and standard
deviation ranging from 1.4 to 2.2. According to Saturno et al., 20185
and Holanda et al., 2020
10, these are reasonable values for the measured BC size distribution at the ATTO site as well.
On top of this core diameter, they assumed a coating of thickness 10-100% of the core radius,
with a refractive index of 1.55 - 0001i, which is typical for organic and inorganic nonabsorbing aerosols. Although they simulate this wide range of coatings, they only use the
coating simulations resulting in absorption enhancements less than a factor of 2. According to
them, this is in agreement with field measurements and laboratory experiments.
To make our manuscript clearer, we have modified lines 227 which now read:
“Theoretical values for BC WDA, based on Mie theory, to retrieve 470−880 and
660−880 , were obtained considering polydisperse coated BC particles with
different size distributions composed of an internally mixed monodisperse BC core, of
the refractive index of 1.95 – 0.76i41, a coat with the refractive index of 1.55 – 0.001i,
which is typical for organic and inorganic non-absorbing aerosols40, and BC density of
1.8 g cm-3 42. The coating thickness varied between 10 to 100% of the BC core radius,
following Wang et al., 201640.”

Comment: Eq6, WDA should be replaced with "BC WDA".

Response: That’s correct, it should be written BC WDA. The equation was corrected.

Comment: The equation assumes that absorption over 660 nm is only due to BC, without BrC. How
reliable is this assumption?

Response: BrC absorbs mostly near the UV, and typically its mass absorption efficiencies at 635 nm and
above are marginal. For instance, Martins et al., 2009 19 measured the full absorption spectrum
from 350 nm to 2.5 μm, and found that BrC MAE is less than 0.2 m2/g above 450 nm. More
recently, Chen et al., 2021 20 report on measurements in the US, and give BrC MAE of 0.5
m2/g at 635 nm and 0.2 m2/g at 780 nm.

Comment: The refractive index for BC is 1.95 – 0.79i instead of “1.95 – 0.76i”. The authors should check
if this is only a typo, or the wrong value has been used in their calculation.

Response: Yes, this is a typo, thank you for pointing that out. According to Wang et al., 2016 8
, they
performed the Mie Calculations using 1.95 - 0.79i, which is the correct value.

Comment: The authors did not assume a fixed AAE of BC ~1 as many studies do. I wonder what is the
AAE values for BC calculated by this study? Is it significantly different from 1?

Response: Mie calculations show that BC AAE can be very different from 1 for larger BC cores, and it
also depends on the thickness of the coating. Figure 3, extracted from Wang's paper, shows
exactly that. Only for very small particles (approx. smaller than 10 nm), we can actually
assume the BC AAE ~ 1. While it might be the case that in urban environments BC cores are
very small, that is not for Amazonia, where BC originates from biomass burning.
Figure 3: BC core as a function of the AAE for different coating thickness. Figure extracted
from Wang et al., (2016)8
.
As in our case we are assuming BC diameters ranging from 20 to 300 nm, we cannot assume
that AAE ~ 1. Instead, we compute the BC AAE from the Mie simulations.

Comment: 6. Line 236. BrC AAE 470 ― 880 nm should be replaced with BrC σ 470 ― 880 nm.

Response: Thanks for the remark, the equation was corrected in the new version of the manuscript.

Comment: 7. Experiment section “Air masses back trajectories vs. fires spots” can be moved before
“Sampling campaign periods” since the sampling campaign periods are determined based on
the air masses back trajectories vs fires spots. Moreover, the description in Line 474-497 is
the actual message of how these two periods are defined. Thus, these sentences can be moved
and merged into the experimental section “sampling campaign periods”. This will help with
the description of the results and help readers have a better understanding of the times series
data in the results section.

Response: The section was restructured as suggested.

Comment: 8. Line 286-289. I assume that the authors here “inorganic components” only refer to those
detected by the ACMS. We should note that BC also contributes to the scattering efficiently.
BC has a very large real part of the refractive index (1.95) as compared to OA (~1.55).
Though the BC mass fraction is low (~15%), the contribution can still be significant. Without
taking this into account, the apportionment can bias a lot from the real situation.

Response: Section 3.6 Scattering Coefficient was reviewed and modified, since the addition of BC in the
MLR model changed the results. The text is now as below:
“The mass scattering efficiency at each of the three wavelengths was estimated considering
all four PMF factors and the black carbon mass concentrations as predictors, according to
equation 9. The HOA factor presented non-significant regression coefficients (p-Values >
0.05), suggesting that this component is responsible for very little or no scattering, and was,
therefore, withdrawn from the model. Thus, the scattering coefficients for the three
wavelengths were obtained as a linear combination of OOA-1, OOA-2, BBOA, and BC.
The results of the multivariate linear regression are shown in Figure 10. The reconstructed
scattering coefficient at 532 nm from the MLR model is shown in Supplement (Figure SI
13 a). The model fits very well with the observed data at all wavelengths showing a
relatively high correlation coefficient (R² = 0.8)
Figure 10: Mass scattering efficiencies (MSE) for each component, MSEs are the
coefficients from multivariate linear regression. Error bars correspond to standard error, b)
Estimate of the contribution of each component to the scattering coefficient at each
wavelength.
The MSE of all components decreased as a function of wavelength (Figure 10a), in
agreement with the behavior expected for scattering in the Mie regime. BBOA had the
highest MSE, 9.15 ± 0.43 m² g-1 at 532 nm (Table 3). The high scattering efficiency for
the BBOA component corroborates the results of other authors who associated biomass
burning aerosol with high scattering efficiencies (Hand and Malm, (2007) 50 and
references therein, Malm et al (2005) 78).
Table 3: Mass scattering efficiencies (mean, standard error) in m² g-1
associated with each
component for the three wavelengths.
MSE OOA-1 OOA-2 BBOA BC
450 nm 6.15±0.27 6.69±0.51 13.97±0.60 7.44±0.30
532 nm 4.32±0.19 4.80±0.35 9.15±0.43 6.12±0.22
637 nm 2.86±0.13 3.38±0.24 5.73±0.29 4.73±0.16
Bond and Bergstrom (2006) 43 mention that the mass scattering efficiency of light-absorbing
carbon (LAC) particles is highly dependent on the particle size, and therefore these values
vary widely. Meanwhile, knowing that the real part of the BC refraction index is around
1.95 at 550 nm, it is expected that its contribution to the total scattering is not negligible,
especially in Rio Branco where the BC mass fraction was around 15%. The MSE of BC was
6.12 m²/g for 532 nm, very close to the MSE for EC particles of 150 nm, 5.9 m²/g found for
a suburban location in Hong Kong 79
.
When taking the mass concentration into account and calculating the contribution of each
component to the total scattering coefficient (Figure 10 b), we observe that both OOAs
presented an almost constant profile across the spectrum contributing to approx. 55% of
scattering, while BC and BBOA contributed to 20-25% and 19-23% of the scattering,
respectively. That is, the secondary organic aerosol components (OOAs) are most
responsible for the total light scattering. It is worth noting that the BBOA found in Rio
Branco presents an intermediate character, partially oxygenated.
The relationship between light scattering and the degree of oxidation of particles is complex
and not fully understood 80. Several studies show that secondary organic aerosols scatter
light more than primary organic aerosols, for instance, monitoring a wavelength range of
500-570 nm, Smith et al., (2020) 81 found that after 12 h of aging, the BB aerosol became
highly scattering. Miranda-Paredes et al., (2009) 82 estimated 75% of light scattering due to
secondary aerosol (inorganic and organic) from photochemical production. Nonetheless,
Malm et al (2005) 78 investigated the contribution of organics for the total scattering,
concluding that OA mass scattering efficiencies were higher during BB episodes and
recommended further studies for other smoke events where aerosol aging, and fuel types
are different. It is important to note that the increase in MSE and scattering coefficients as
a function of aerosol age is also associated with changes in particle size because as the plume
ages there is a shift in particle size modes 83,84. Therefore, measurements on the chemical
composition of size resolved aerosols would be enlightening to verify which components
contribute most to scattering, and which are the mechanisms responsible for the increase in
efficiency.”

Comment: 9. Line 296. Here is the first time that those OA factors appear in this study. A brief
introduction of the source factors observed by this study is necessary for section “Positive
Matrix Factorization (PMF)”. This will support why the regression was performed on these
chemical components. Moreover, BC should be included since it is a major contribution to
the scattering.

Response: We agree with the referee's observation. As the results of the PMF have not yet been presented
in the text, we preferred to modify the way the multilinear model was presented, opting for a
generic explanation and mentioning only that the factors of the PMF were included in the
model, but without naming them. In addition, as detailed in the previous answer, BC was
included in the MLR model for the scattering coefficient. The original text (lines 290 to 303)
was adapted as follows:
“Only OA and BC were considered relevant species for the multilinear regression models
since inorganic components had little contribution in PM1 (relative mass contribution <
10%) and attempts to include them in the model resulted in unrealistic mass efficiencies
and statistically non-significant regression coefficients for both scattering and absorption.
Assuming that the scattering/absorption coefficients are a linear combination of organic
aerosol components, we estimated the contribution of each PMF factor + BC for a single
wavelength as:
σ(

⁄ , λ) =
1 + 2 + 3 + 4 + + ϵ (Eq. 9)
where a, b, c, d, and e represent the mass scattering/absorption efficiency of each
component, whilst epsilon is the residual or intercept. The intercept accounts for chemical
components that were not included in the model such as refractory inorganic particles and
soil dust. The fitting was performed using the fitlm function of Matlab 2015.”

Comment: 10. Section 3.1. Gas-phase data are valuable for the understanding of the background. But The
focus point of the paper here is linking the aerosol chemical composition and the aerosol
optical properties. These can be supporting information instead of putting them here.
Moreover, the diurnal cycle is not relevant here. In line 307-309, the author stated that “The
diurnal profiles of the trace gases indicate strong atmospheric photochemical activity,
reaching its maximum during the middle of the day when there is more intense solar
radiation and higher temperature.” However, this is not true for NO2 as presented.

Response: The reviewer is correct. The statement is pertinent to the other gases, except for NO2 which
reaches its minimum at this time of day since it is consumed in photochemical ozone
production. This is clarified later in the text. However, the manuscript was restructured
moving section 3.1 gas-phase results to the supplement material.

Comment: 11. Line 377. 40% of BC? This is contradictory to the statement that BrC contribution ranges
from 0.16 to 0.25. Assuming the highest fraction of 0.25, the fraction of BrC to BC is about
33%.

Response: The sentence was poorly formulated, and it was removed from the updated manuscript. We
thank the reviewer for pointing this out.

Comment: 12. Line 428-435. The assignment of OOA-2 is not well supported. I agree with this assignment,
but the discussion can be improved. For example, somehow first discuss the factor BBOA,
which has very clear characteristics. And then discuss OOA-1 and OOA-2. For OOA-2,
besides the ion characteristics, the authors can also look at the correlation between OOA-2
and BBOA. This may also support that OOA-2 can be related to BB rather than saying
"Observing its time series, we note that this factor is associated with aerosols originating
from biomass burning."

Response: The text was restructured starting with the BBOA and HOA and later OOA-1 and OOA-2.
Regarding the OOA-2 assignment, this paragraph was rewritten in order to further explain this
factor. In fact, the OOA-2 did not present any signal in the m/zs 60 and 73, which are the major
BBOA tracers, therefore the new description was adjusted accordingly. The new paragraph
describing the OOA-2 is below.
“The last factor obtained was the OOA-2. This factor presents prominent signals at m/zs
44 (CO2+), 18 (H2O+), 29 (COH+ and/or C2H5+) and 43 (C2H3O+ and/or C3H7+). In
terms of f44 this factor presents a lower value (19%) compared to the OOA-1 (32%),
suggesting it is less oxygenated than the OOA-1. This fact is also confirmed by the larger
ratio between m/zs 44 and 43 obtained for the OOA-1, and the larger content of a less
oxygenated fragment, the m/z 29. Despite the similarities between the OOA-2 and OOA-1
in terms of mass spectra, the time series does not present good agreement (R=0.33). In
fact, the diel profiles present very different behaviors in the P1 and P2 (Figure XX). The
OOA-2 presents larger mass concentration values during the nighttime over the P1, while
over the P2 no clear pattern can be observed.”

Comment: 13. Line 455-457. This summary of the reference is not correct. The mentioned hydroxy
carboxylic acids are SOA tracers for biogenic SOA which were found in aerosol samples
from Hungry. This is clearly stated in the cited reference. Instead, Claeys et al. observed
nitrocatechols are dominant in biomass burning samples in Rondônia.

Response: Claeys et al., (2002) 21 in section ‘Characterisation of HULIS isolated from tropical BB PM2.5
aerosol’ identified unknown C8-hydroxy dicarboxylic acid, terebic acid, terpenylic acid, and
MBTCA as minor components in samples collected in Rondônia. For the sake of clarity, the
statement was altered to:
“Likewise, Claeys et al (2002) characterized HULIS components from BB samples
collected in Southwestern Amazonia identifying nitrocathechols as dominant components
and also some minor components such as hydroxy dicarboxylic acids (such as terebic acid,
terpenilyc acid, among others).”

Comment: 14. Line 551-560. I appreciate the authors providing the size distribution of the particles during
the campaign and try to link different modes of the particles to the chemical components
based on the diurnal cycles. However, this is not strong enough unless you have sizedependent chemical composition analysis (e.g., HR-Tof-AMS). Moreover, these results are
also isolated from the central topic of the manuscript, and they are not correlated to other
results in the paper. I feel these are redundant and can be moved out.

Response: The size distributions were moved out.

Comment: 15. Line 572-573 and line 686-689. The authors argued that the spectral dependence of the
scattering coefficient is related to the particle's size. This is not correct. We should keep in
mind that the scattering cross section of particles also shows a strong wavelength
dependency, especially for OA.

Response: We agree with the reviewer that the decrease in scattering coefficient with wavelength depends
both on the particle size distribution and refractive index. Besides, the scattering Angstrom
exponent is qualitatively used as an indicator of particle size for aerosol measurements
including both the fine and the coarse modes, which was not the case in our set of
measurements. Accordingly, we decided to suppress the lines 572-573 and the lines 687-688,
since they were misleading.

Comment: 16. Line 689 and table 1. The MSE values obtained for BBOA are extremely high in this study.
Malm et al. [2005] and McMeeking et al. [2005] obtained values of mixed fine mode near 6
m2g -1 in Yosemite National Park CA during periods dominated by biomass smoke at ~550
nm. A POM mass scattering efficiency near 7.2 m2g -1 would require specific conditions of
a fairly narrow lognormal size distribution with a mass mean diameter near 0.4 mm,
geometric standard deviation of 1.5, an organic aerosol density of 1.2, and refractive index of
1.55. The values obtained by this study are even larger. I wonder what are the reasons that
the MSE values are so high.

Response: The work of McMeeking et al., (2005)
22 is cited in section 3.6 where the average mass
scattering efficiencies for the PM1 aerosol population were presented. Our results for average
mass scattering efficiency are very close to the values reported by the aforementioned
references. At lines 648 to 651 we read:
“McMeeking et al (2005)70 noted, however, that MSE increased during periods dominated
by biomass smoke in Yosemite National Park, CA, USA and reported an average MSE (λ
= 530 nm) of 4.1 ± 0.7 m2g-1, that reached values as high as 6 m2 g-1 which are in
accordance with findings in Rio Branco.”
In addition, Hand and Malm, (2007)23 reported that: “one of the highest values observed for
average MSE (6 m²g-1) was in rural New England-mid-Atlantic region [Poirot and Husar,
2004] and corresponds to biomass smoke aerosol that appear to be associated with higher
efficiencies, as suggested in the previous section.”
Malm et al., (2005)
24 comment that using the recommended value of 4.0 m2/g for organic
aerosol mass scattering efficiency (in the work OA is treated as a single component)
underestimated the calculation of the fine mode scattering by at least 30% for periods with
high concentrations of organic mass. Overall, the work concludes that mass scattering
efficiencies were higher during BB episodes and recommend further studies for other smoke
events where aerosol aging, and fuel types are different.
Section 3.7, on the other hand, presents the mass scattering efficiencies assigned to each OA
component. These values are not comparable with the values presented in section 3.6, as the
MSE of the overall aerosol sample is a combination of the MSE for each of its components.
As far as we know, this is the first work that reports this type of assignment of ass scattering
(and absorption) efficiencies are estimated for chemically-resolved organic components. Other
works attribute the scattering or absorption efficiencies by source, such as the work by Ealo et
al., (2018)
25 (for PM10).
In our group, we are working with MLR models to estimate mass scattering efficiencies for
other regions. For instance, in a study conducted in Sao Paulo, Brazil (urban environment)
(Monteiro dos Santos, D. A. et al., paper in preparation) during the months of July to November
2017, when BB contribution was significant, MSE for the BBOA component was 12.54 ± 0.27,
9.50 ± 0.23 and 6.67 ± 0.19 for 450 nm, 525 nm e 635 nm, respectively.
Furthermore, we would like, once again, acknowledge the reviewer and to point out that the
multilinear regressions for the scattering coefficient should include BC. Therefore, the MLR
were redone, taking into account BC as a significantly scattering component, as previously
suggested by the reviewer.
The revised values for the MSEBBOA were lower than those previously reported in this
manuscript. As reported in the preceding answer, MSE for the BBOA = 9.15 ± 0.43 for 532
nm.

Comment: 17. Figure 10 was presented three times in the manuscript.

Response: This was an error that occurred during the pdf conversion. We thank the reviewer for pointing
that out.

Comment: 18. For comparison of MSE and MAE values, the author can also plot literature data in panel (a)
of Figure 9-10. This will give us a clear impression of how the results from this study are
compared to others.

Response: We want to highlight that to the authors’ knowledge, this is the first time that absorption and
scattering efficiencies are calculated for chemically-resolved organic components. Most
studies have used different approaches and methodologies to apportion sources. Or whether
they are presented in a single wavelength and for the average mass scattering/absorption
efficiencies, and third, the particle size covered was different (studying PM10, for instance).
Therefore, we feel that it would not be worth adding values from the literature to the graph in
Figure 9-10. Instead, we reinforce the discussions and comparison with the literature.

Comment: 19. Line 725-726. The argument regarding the role of bleaching is not well supported. OOA can
be produced by the aging of HOA, which might contain a bleaching process. However, OOA
can also be produced via oxidation of VOCs which can produce less light-absorbing species.
This could also result in a reduced MAE due to dilution with less absorbing SOA.

Response: The comment was suppressed.

Comment: 20. Line 769-772. I don't get the point why discuss the absorbing by BC here since the paragraph
is focused on the BrC chromophores

Response: The paragraphs were restructured and the phrases in question were excluded. Now from line
766, it reads as follows:
“The absorption of light at UV-Vis range can be explained by absorbing molecules such
HULIS23, nitroaromatics77, polyoxygenated compounds78, and polyaromatic hydrocarbons
(PAH)79, which are known chromophores and yet chelating agents, complexing with
transition metals such as iron80. Laboratory-controlled studies burning different vegetation
species identified a variety of oxygenated hydrocarbons (Lin et al. (2016)78) with diverse
chemical characteristics such as lignin pyrolysis products lignin-derived products,
distillation products, nitroaromatics, and PAHs (Fleming et al. (2020)81) emphasizing that
most BrC chromophores were common to samples from distinct vegetation species
meanwhile their relative contributions to total light absorption differ.
Thus, the formation of chromophore compounds in BB has mechanisms common to all
types of fuel/fire and exhibits unique properties that need to be addressed concerning local
characteristics. Regarding the overall fine particulate matter absorption, we cannot rule out
the contribution of other OA components to the absorption of light as many studies have
assessed atmospheric chemical processes that resulted in light-absorbing SOA23,82. The
presence of BrC, therefore, should not be neglected, and studies in the Amazon basin,
especially in the dry season when massive biomass burning occurs, should investigate this
characteristic of the regional aerosol addressing its climatic effects.”

References
1 E. T. Sena, P. Artaxo and A. L. Correia, Spatial variability of the direct radiative forcing
of biomass burning aerosols and the effects of land use change in Amazonia, Atmos.
Chem. Phys., 2013, 13, 1261–1275.
2 R. da S. Palácios, P. E. Artaxo, G. G. Cirino, V. Nakale, F. G. Morais, L. D. Rothmund,
A. Y. Siqueira, M. S. Biudes, N. G. Machado, L. F. A. Curado, J. B. Marques and J. de S.
Nogueira, Long-term measurements of aerosol optical properties and radiative forcing
(2011-2017) over Central Amazonia, Atmósfera, 2020, 0–27.
3 L. V. Rizzo, A. L. Correia, P. Artaxo, A. S. Procópio and M. O. Andreae, Spectral
dependence of aerosol light absorption over the Amazon Basin, Atmos. Chem. Phys.,
2011, 11, 8899–8912.
4 J. Saturno, C. Pöhlker, D. Massabò, J. Brito, S. Carbone, Y. Cheng, X. Chi, F. Ditas, I.
Hrabě de Angelis, D. Morán-Zuloaga, M. L. Pöhlker, L. V. Rizzo, D. Walter, Q. Wang, P.
Artaxo, P. Prati and M. O. Andreae, Comparison of different Aethalometer correction
schemes and a reference multi-wavelength absorption technique for ambient aerosol data,
Atmos. Meas. Tech., 2017, 10, 2837–2850.
5 J. Saturno, B. A. Holanda, C. Pöhlker, F. Ditas, Q. Wang, D. Moran-Zuloaga, J. Brito, S.
Carbone, Y. Cheng, X. Chi, J. Ditas, T. Hoffmann, I. Hrabe De Angelis, T. Könemann, J.
V. Lavrič, N. Ma, J. Ming, H. Paulsen, M. L. Pöhlker, L. V. Rizzo, P. Schlag, H. Su, D.
Walter, S. Wolff, Y. Zhang, P. Artaxo, U. Pöschl and M. O. Andreae, Black and brown
carbon over central Amazonia: Long-term aerosol measurements at the ATTO site, Atmos.
Chem. Phys., 2018, 18, 12817–12843.
6 J. Brito, L. V. Rizzo, W. T. Morgan, H. Coe, B. Johnson, J. Haywood, K. Longo, S.
Freitas, M. O. Andreae and P. Artaxo, Ground-based aerosol characterization during the
South American Biomass Burning Analysis (SAMBBA) field experiment, Atmos. Chem.
Phys., 2014, 14, 12069–12083.
7 M. Kuwata, S. R. Zorn and S. T. Martin, Using elemental ratios to predict the density of
organic material composed of carbon, hydrogen, and oxygen, Environ. Sci. Technol.,
2012, 46, 787–794.
8 X. Wang, A. J. Sedlacek, S. S. de Sá, S. T. Martin, M. L. Alexander, M. L. Alexander, T.
B. Watson, A. C. Aiken, S. R. Springston and P. Artaxo, Deriving brown carbon from
multiwavelength absorption measurements: Method and application to AERONET and
Aethalometer observations, Atmos. Chem. Phys., 2016, 16, 12733–12752.
9 M. L. Pöhlker, C. Pöhlker, F. Ditas, T. Klimach, I. Hrabe de Angelis, A. Araújo, J. Brito,
S. Carbone, Y. Cheng, X. Chi, R. Ditz, S. S. Gunthe, J. Kesselmeier, T. Könemann, J. V.
Lavrič, S. T. Martin, E. Mikhailov, D. Moran-Zuloaga, D. Rose, J. Saturno, H. Su, R.
Thalman, D. Walter, J. Wang, S. Wolff, H. M. J. Barbosa, P. Artaxo, M. O. Andreae and
U. Pöschl, Long-term observations of cloud condensation nuclei in the Amazon rain forest
– Part 1: Aerosol size distribution, hygroscopicity, and new model parametrizations for
CCN prediction, Atmos. Chem. Phys., 2016, 16, 15709–15740.
10 B. A. Holanda, M. L. Pöhlker, D. Walter, J. Saturno, M. Sörgel, J. Ditas, F. Ditas, C.
Schulz, M. A. Franco, Q. Wang, T. Donth, P. Artaxo, H. M. J. Barbosa, S. Borrmann, R.
Braga, J. Brito, Y. Cheng, M. Dollner, J. W. Kaiser, T. Klimach, C. Knote, O. O. Krüger,
D. Fütterer, J. V. Lavrič, N. Ma, L. A. T. Machado, J. Ming, F. G. Morais, H. Paulsen, D.
Sauer, H. Schlager, J. Schneider, H. Su, B. Weinzierl, A. Walser, M. Wendisch, H.
Ziereis, M. Zöger, U. Pöschl, M. O. Andreae and C. Pöhlker, Influx of African biomass
burning aerosol during the Amazonian dry season through layered transatlantic transport
of black carbon-rich smoke, Atmos. Chem. Phys., 2020, 20, 4757–4785.
11 L. V. Rizzo, P. Artaxo, T. Müller, A. Wiedensohler, M. Paixão, G. G. Cirino, A. Arana, E.
Swietlicki, P. Roldin, E. O. Fors, K. T. Wiedemann, L. S. M. Leal and M. Kulmala, Long
term measurements of aerosol optical properties at a primary forest site in Amazonia,
Atmos. Chem. Phys., 2013, 13, 2391–2413.
12 P. Artaxo, L. Rizzo, J. Brito, H. M. J. Barbosa, A. Arana, E. T. Sena, G. G. Cirino, W.
Bastos, S. T. Martin and M. O. Andreae, Atmospheric aerosols in Amazonia and land use
change: from natural biogenic to biomass burning conditions, Faraday Discuss., 2013,
165, 203.
13 D. Liu, J. W. Taylor, D. E. Young, M. J. Flynn, H. Coe and J. D. Allan, The effect of
complex black carbon microphysics on the determination of the optical properties of
brown carbon, Geophys. Res. Lett., 2015, 42, 613–619.
14 A. P. Grieshop, N. M. Donahue and A. L. Robinson, Laboratory investigation of
photochemical oxidation of organic aerosol from wood fires 2: Analysis of aerosol mass
spectrometer data, Atmos. Chem. Phys., 2009, 9, 2227–2240.
15 Z. Fang, W. Deng, Y. Zhang, X. Ding, M. Tang, T. Liu, Q. Hu, M. Zhu, Z. Wang, W.
Yang, Z. Huang, W. Song, X. Bi, J. Chen, Y. Sun, C. George and X. Wang, Open burning
of rice, corn and wheat straws: Primary emissions, photochemical aging, and secondary
organic aerosol formation, Atmos. Chem. Phys., 2017, 17, 14821–14839.
16 H. Timonen, S. Carbone, M. Aurela, K. Saarnio, S. Saarikoski, N. L. Ng, M. R.
Canagaratna, M. Kulmala, V. M. Kerminen, D. R. Worsnop and R. Hillamo,
Characteristics, sources and water-solubility of ambient submicron organic aerosol in
springtime in Helsinki, Finland, J. Aerosol Sci., 2013, 56, 61–77.
17 L. Varanda Rizzo, P. Roldin, J. Brito, J. Backman, E. Swietlicki, R. Krejci, P. Tunved, T.
Petäjä, M. Kulmala and P. Artaxo, Multi-year statistical and modeling analysis of
submicrometer aerosol number size distributions at a rain forest site in Amazonia, Atmos.
Chem. Phys., 2018, 18, 10255–10274.
18 M. Franco, F. Ditas, L. A. Kremper, L. Machado, M. Andreae, A. Araújo, H. Barbosa, J.
de Brito, S. Carbone, B. Holanda, F. Morais, J. Nascimento, M. Pöhlker, L. Rizzo, M. Sá,
J. Saturno, D. Walter, S. Wolff, U. Pöschl, P. Artaxo and C. Pöhlker, Occurrence and
growth of sub-50 nm aerosol particles in the Amazonian boundary layer, Atmos. Chem.
Phys., 2021, 1–36.
19 J. V. Martins, P. Artaxo, Y. J. Kaufman, A. D. Castanho and L. A. Remer, Spectral
absorption properties of aerosol particles from 350–2500nm, Geophys. Res. Lett., 2009,
36, L13810.
20 L.-W. A. Chen, J. C. Chow, X. Wang, J. Cao, J. Mao and J. G. Watson, Brownness of
Organic Aerosol over the United States: Evidence for Seasonal Biomass Burning and
Photobleaching Effects, Environ. Sci. Technol., 2021, 55, 8561–8572.
21 M. Claeys, R. Vermeylen, F. Yasmeen, Y. Gómez-González, X. Chi, W. Maenhaut, T.
Mészáros and I. Salma, Chemical characterisation of humic-like substances from urban,
rural and tropical biomass burning environments using liquid chromatography with
UV/vis photodiode array detection and electrospray ionisation mass spectrometry,
Environ. Chem., 2012, 9, 273–284.
22 G. R. McMeeking, S. M. Kreidenweis, C. M. Carrico, J. L. Collett, D. E. Day and W. C.
Malm, Observations of smoke-influenced aerosol during the Yosemite Aerosol
Characterization Study: 2. Aerosol scattering and absorbing properties, J. Geophys. Res. D
Atmos., 2005, 110, 1–9.
23 J. L. Hand and W. C. Malm, Review of aerosol mass scattering efficiencies from groundbased measurements since 1990, J. Geophys. Res. Atmos., , DOI:10.1029/2007JD008484.
24 W. C. Malm, D. E. Day, C. Carrico, S. M. Kreidenweis, J. L. Collett, G. McMeeking, T.
Lee, J. Carrillo and B. Schichtel, Intercomparison and closure calculations using
measurements of aerosol species and optical properties during the Yosemite aerosol
characterization study, J. Geophys. Res. D Atmos., 2005, 110, 1–21.
25 M. Ealo, A. Alastuey, N. Pérez, A. Ripoll, X. Querol and M. Pandolfi, Impact of aerosol
particle sources on optical properties in urban, regional and remote areas in the northwestern Mediterranean, Atmos. Chem. Phys., 2018, 18, 1149–1169.

06-Dec-2021

Dear Dr Ponczek:

Manuscript ID: EA-ART-07-2021-000055.R1
TITLE: Linking chemical composition and optical properties of biomass burning aerosols in Amazonia

Thank you for submitting your revised manuscript to Environmental Science: Atmospheres. After considering the changes you have made, I am pleased to accept your manuscript for publication in its current form. I have copied any final comments from the reviewer(s) below.

You will shortly receive a separate email from us requesting you to submit a licence to publish for your article, so that we can proceed with publication of your manuscript.

You can highlight your article and the work of your group on the back cover of Environmental Science: Atmospheres, if you are interested in this opportunity please contact me for more information.

We will publicise your paper on our Twitter account @EnvSciRSC – to aid our publicity of your work please fill out this form: https://form.jotform.com/211263048265047

For tips on how to publicise your research, please visit: https://www.rsc.org/journals-books-databases/about-journals/maximise-your-impact/

Discover more Royal Society of Chemistry author services and benefits here: https://www.rsc.org/journals-books-databases/about-journals/benefits-of-publishing-with-us/

Thank you for publishing with Environmental Science: Atmospheres, a journal published by the Royal Society of Chemistry – the world’s leading chemistry community, advancing excellence in the chemical sciences.

With best wishes,

Jamie Purcell

Dr Jamie Purcell MRSC
Publishing Editor, Environmental Science: Atmospheres
Royal Society of Chemistry
Thomas Graham House
Science Park, Milton Road
Cambridge, CB4 0WF, UK
Tel +44 (0) 1223 432168
www.rsc.org

Environmental Science: Atmospheres is accompanied by sister journals Environmental Science: Nano, Environmental Science: Processes and Impacts, and Environmental Science: Water Research; publishing high-impact work across all aspects of environmental science and engineering. Find out more at: http://rsc.li/envsci

Reviewer 1

Thank you for your detailed replies to my comments.