Geostationary aerosol retrievals of extreme biomass burning plumes during the 2019–2020 Australian bushfires

Robbins, Daniel J. V.; Poulsen, Caroline A.; Siems, Steven T.; Proud, Simon R.; Prata, Andrew T.; Grainger, Roy G.; Povey, Adam C.

doi:https://doi.org/10.5194/amt-17-3279-2024

Articles | Volume 17, issue 10

https://doi.org/10.5194/amt-17-3279-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/amt-17-3279-2024

© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 17, issue 10

Research article

|

29 May 2024

Research article |

| 29 May 2024

Geostationary aerosol retrievals of extreme biomass burning plumes during the 2019–2020 Australian bushfires

Daniel J. V. Robbins, Caroline A. Poulsen, Steven T. Siems, Simon R. Proud, Andrew T. Prata, Roy G. Grainger, and Adam C. Povey

Download

Final revised paper (published on 29 May 2024)
Preprint (discussion started on 10 Nov 2023)

Interactive discussion

Status: closed

RC1:
'Comment on egusphere-2023-2179', Antti Lipponen, 08 Dec 2023

Extreme biomass burning events are becoming more and more common and intense as the climate changes. More intense fires may produce thicker smoke plumes. Therefore, it is important that the remote sensing instruments and retrieval algorithms produce reliable estimates of high aerosol optical thicknesses (AOTs). The Robbins et al. manuscript uses the 2019-2020 Australian bushfires as a case to further develop and evaluate the Optimal Retrieval of Aerosol and Cloud (ORAC) algorithm for AOT retrievals in cases of extreme biomass burning plumes using a geostationary satellite instrument. They also compare the performances of different satellite instruments and retrieval algorithms to retrieve extreme AOT values. Furthermore, due to their development work and comparison, their algorithm produces estimates of significantly higher AOTs than previous studies. In their comparison, they also find that the AOTs from large-scale fires are likely to have been systemically underestimated by the existing approaches.

The development work, design of the experiments, and comparison of the results presented in Robbins et al. have been carried out carefully, and the authors' selections are well-justified and clear. The manuscript is very well written and easy to read. I only have some minor comments, which I have listed below.

Minor comments:

l.12 "L1.5" please write in full "level 1.5"

l.15 "...MODIS MAIC SLSTR SYN..." comma missing

l.45 "...can accurately monitor aerosol properties in the atmosphere directly above the site location.." To be precise, especially with larger solar zenith angles, the AERONET is also sampling some nearby station locations, not just the pointwise location. You may want to modify the text accordingly, to be precise.

l.64 "..in recent years. such as..." extra .

l.88 Please add the wavelength for the CALIOP AOT

l.89 Please add the wavelength for the AOT listed on this line

l.95 Please mention if the Zhuravleva et al. (2017) estimates were AERONET or satellite (which instrument/algorithm) estimates

l.130 Please confirm/clarify what exact AERONET data you use, the Direct Sun or Aerosol Inversions data? Now, it is not clear.

l.158 Please consider changing SLSTR to Sentinel-3 Synergy as the Synergy data is (mostly) based on both SLSTR and OLCI instruments' data both flying on Sentinel-3 satellites. If you decide to change, also change SLSTR throughout the manuscript to "Sentinel-3 Synergy" or similar.

l.161 Please clarify if you use the SYN surface reflectance and aerosol parameter (SY_2_SYN) or SYN AOD data (SY_2_AOD) product that are different Synergy products. If you used the SY_2_SYN data product, please justify why this data product, whose primary use is not aerosol information but surface reflectance, is used instead of the dedicated aerosol data product (SY_2_AOD).

l.400 Please define "NN"

Citation: https://doi.org/10.5194/egusphere-2023-2179-RC1
- AC2: 'Reply on RC1', Daniel Robbins, 16 Feb 2024
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2179/egusphere-2023-2179-AC2-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-2179-AC2
RC2:
'Comment on egusphere-2023-2179', Marloes Penning de Vries, 22 Jan 2024

The manuscript presents the capabilities of a version of ORAC that was enhanced to more adequately retrieve aerosol optical thickness of extremely dense smoke plumes. The AOT of such plumes is usually underestimated by satellite algorithms due to the mis-classification of such observations as clouds and subsequent exclusion from analysis; or the values are capped to a certain maximum retrievable AOT by the retrieval algorithm. The ORAC algorithm shows good agreement with AERONET L1.5 data and performs better than the extensive and diverse set of satellite algorithms to which it was compared, both in terms of number of successful retrievals as in terms of retrieval quality (with reference to the single AERONET site near Canberra, AUS, as ground truth).
The study is well designed and executed, and the manuscript is well written. I do miss some discussion on two important aspects. First: the dependence of AOT retrieval on viewing angle. The effect of parallax on a plume of moderate altitude is mentioned, but the strongly elongated light path due to large viewing (and/or solar) angles is not discussed. This, I imagine, is a significant issue for regions at the edges of the geostationary field of view. Second: the authors state that at higher AOT the retrieval depends more on the assumed aerosol parameters (particle size distribution, refractive index) than for smaller AOT. The ORAC algorithm performs very well for the Tumbarumba station, but then it makes use of the aerosol parameters derived from Tumbarumba AERONET data. How does this translate to a global algorithm? The implication appears to be that extreme AOT values can only be retrieved from satellite if the aerosol parameters are accurately known for the region (or even fire event) in question. I'd like to read the authors' view on this point. Because although they stress that their algorithm is scientific and not operational, it may be assumed that they, or others, will continue the advancement of aerosol algorithms towards routine monitoring of even extreme biomass burning cases.

Citation: https://doi.org/10.5194/egusphere-2023-2179-RC2
- AC1: 'Reply on RC2', Daniel Robbins, 16 Feb 2024
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-2179/egusphere-2023-2179-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2023-2179-AC1

Peer review completion

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

AR by Daniel Robbins on behalf of the Authors (07 Mar 2024) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (11 Mar 2024) by Marloes Penning de Vries

AR by Daniel Robbins on behalf of the Authors (16 Mar 2024) Author's response Author's tracked changes Manuscript

ED: Publish as is (19 Mar 2024) by Marloes Penning de Vries

AR by Daniel Robbins on behalf of the Authors (20 Mar 2024) Manuscript

Short summary

Extreme wildfire events are becoming more common with climate change. The smoke plumes associated with these wildfires are not captured by current operational satellite products due to their high optical thickness. We have developed a novel aerosol retrieval for the Advanced Himawari Imager to study these plumes. We find very high values of optical thickness not observed in other operational satellite products, suggesting these plumes have been missed in previous studies.