the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 28 Mar 2022
Aerosol Optical Properties of Extreme Global Wildfires and Estimated Radiative Forcing with GCOM-C SGLI
Abstract. Wildfires damage land ecosystems and significantly impact the atmosphere by releasing large amounts of CO2 and aerosols. Aerosols, in particular, have a radiative forcing potential that perturbs the global radiation balance. Therefore, understanding their optical properties is essential to estimate the radiative forcing for predicting the climatic impact due to burning. The Global Change Observation Mission—Climate, or GCOM-C (SHIKISAI), is a polar-orbit satellite launched by JAXA on 23 December 2017. This study analyzed wildfires in the Amazon, Angola, Australia, California, Siberia, and Southeast Asia that occurred after 2018 using data from the GCOM-C satellite obtained with the Second-generation Global Imager (SGLI), a multi-band optical imaging radiometer. We compared the aerosol optical properties of Ångström Exponent (AE) and Single Scattering Albedo (SSA) and found that their distributions are different for each region. In addition, the negative correlations are found in the Amazon, Angola, Siberia, and Southeast Asia when we locally fitted the AE and SSA relationship for each wildfire by a linear function. This is likely due to the effect of hygroscopic growth (SSA becomes larger with water uptake), considering the difference in relative humidity in the six regions, and the significant behavior of the aerosol aging. We also found that the relationship between SSA and relative humidity varies depending on the type of the burned vegetation. For the needle-forest-dominated group, a higher SSA is observed in the range from 40 % to 70 % of relative humidity than the broad forest-dominated group. This global investigation of the aerosol optical properties reveals that their characteristics differ due to regional differences such as relative humidity and vegetation type. Thus, the regional characteristics and the aging effect of biomass-burning aerosols must be considered in model predicting. Moreover, we estimated the net incoming radiation at the top of the atmosphere during the intense fire periods to understand the impact of the direct effect of biomass burning aerosols on radiative forcing. The estimates show a significant negative radiative forcing (i.e., a cooling effect) over the ocean (−78 and −96 Wm−2 in Australia and California, respectively). In contrast, small values are observed over land (∼ −10 Wm−2 for all six regions). This suggests that the radiative forcing depends on the region of the wildfire plumes. Therefore, taking the regional characteristics of the optical property and surface reflectance into account is necessary to estimate the effect on radiative forcing and future impacts of a short-lived climate forcer from wildfires.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(2169 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-21', Anonymous Referee #1, 05 May 2022
Review of “Aerosol Optical Properties of Extreme Global Wildfires and Estimated Radiative Forcing with GCOM-C SGLI” by Tanada et al.
The authors presented an analysis of “wildfire-emitted” aerosol optical properties retrieved from the GCOM-C SGLI satellite instrument and representative of six major biomass burning regions of the Earth. The manuscript further attempts to estimate the changes in TOA radiative forcing due to the extreme wildfire events during the 2018-21 period, separately for the six regions. There is no denying the significance of the motivation behind this work, along with the excitement of having a newer dataset available for constraining the aerosol properties in our global models.
In my opinion, however, the manuscript is unsuitable for publication in its current form. Major additional work and analysis are needed before resubmitting. Moreover, the findings are not revealing, and the discussions do not add a lot to our existing knowledge about aerosol optical properties. My most significant concern is that often too much is interpreted from the retrievals that are not thoroughly vetted. The flaws in the analysis are not helped by challenging readability (and some errors/inconsistencies). Please consider the points made below as grounds for this assessment.
Major comments:
- It is important to understand that satellite observations of AOD (and more importantly of derived properties such as SSA) are after all retrievals and require a thorough validation/evaluation, at least using the aerosol products from other existing passive sensors (such as MODIS/VIIRS) and ground-based observations from AERONET sun photometers wherever available. Unless there is an existing publication demonstrating the robustness and uncertainty limits of the data set (other than the ATBD document that the authors point to, which is not peer-reviewed), authors should be cautious before using the retrievals for making detailed interpretations about aerosol properties.
- The title and manuscript suggest that the analysis focuses on the aerosol properties emitted from extreme wildfire events and the radiative impacts due to those. However, the claim is kind of misleading for the following reasons: (1) it is not very clear how are extreme wildfires detected? The authors do mention vaguely in section 2.1 that SGLI’s WFRP research product is being used to detect the hotspot locations of wildfires, but what criterion is being used to distinguish between extreme wildfires or other fires is not explained. Also, it is well-known that satellite imagers have issues detecting the accurate magnitudes of fire radiative power when thick plumes of smoke emitted from the wildfire itself cover the fire locations, therefore the uncertainty in detecting extreme wildfires using this technique is not discussed. (2) Even though the focus is on the wildfire-emitted aerosols, throughout the manuscript (barring section 3.4) the aerosol properties are discussed in the context of monthly means (including the calculations of direct radiative forcing). So, considering Angola for example (and it might be true for other regions also), where August month sees significant biomass burning from savannas and grasslands as well, how are the authors secluding the impact of aerosols emitted from wildfires only?
- Changes in SSA and AE with respect to relative humidity and vegetation: the entire section 3.3. discusses how SSA increases and AE decreases with increasing relative humidity. Firstly, the analysis does not add anything new to our current understanding because it is well understood in the literature that swelling of particles leads to enhanced scattering and hence higher SSA. Similarly, AE would obviously be lower for swollen/humidity-grown particles. Another conclusion from this analysis is that SSA varies depending on vegetation types, which again I think is well known and which is why models consider different emission factors for emission constituents depending on different biomes (e.g., boreal forests versus tropical forests, see Akagi et al. (2011)). In fact, apart from vegetation, the differences in the composition (or SSA) of the freshly emitted aerosol mixture could also arise from the type of burning i.e., flaming versus smoldering, which is not even discussed here.
- Finally, the discussion on aerosol aging effects is again limited to the humification effects only. The section lacks a discussion on changes in SSA due to mixing and chemical processes, such as lensing (where BC is coated with organic material) or photobleaching (or photo-oxidation) of brown carbon. When considered together, the impacts on aerosol absorption might not be straightforward to infer based on mere relative humidity justifications.
Following are some examples of specific errors and inconsistencies:
Line 34-35: “The remaining 30 % are water, CO, and various particles (Council, 2004)”. I don’t think this statement is correct because apart from the things listed here, there are other GHG emitted such as methane, nitrous oxide, as well as volatile organic compounds (VOCs), that can act as precursors for secondary organic aerosols.
Line 39: “Organic carbon reflects solar radiation and causes a negative forcing.” Again, not completely true because brown carbon (a subset of the organic carbon) has significant absorption in the near-UV wavelengths.
Line 44: what is the reference for the estimate of indirect forcing listed here, please clarify and if possible demonstrate the variability of these estimates based on more recent studies in the literature. Also, what about aerosol semi-direct effects?
Line 47: Not just particle size distribution and chemical composition, optical properties are also largely dependent on aerosol hygroscopicity and mixing state.
Line 48: This sentence is wrong, and the in-text citation is not even included in the bibliography. Most of the smoke particles are in fact dominated by fine or accumulation mode (Eck et al. 1999), while dust and sea salt would be coarse mode dominated.
Line 54: What is direct radiative effect efficiency? Please be consistent with the use of radiative forcing, direct radiative effect, indirect forcing, indirect effect, etc. in the text throughout. There are subtle differences between each of these definitions.
Line 71: It mentions that errors correspond to a 1 sigma confidence level throughout the manuscript, yet error depiction is missing for several critical figures such as Fig. 1 and 7.
Other discrepancies: It is not mentioned how the cloud presence/contamination is handled in the dataset? Unless we are only looking at clear-sky data, but if so, it’s nowhere mentioned until we reach the discussions specific to radiative forcing calculations.
Fig. 9 shows Angola base-period/no fire period as July 2020 and Table 3. Suggests base period is 11-20 Sep 2020. Which one is it? Moreover, a quick look at MODIS/VIIRS observed fire pixels will tell you that there are enough fire pixels during both the base periods mentioned here.
References:
Eck, T. F., Holben, B. N., Reid, J. S., Dubovik, O., Smirnov, A., O’Neill, N. T., Slutsker, I., and Kinne, S.: Wavelength dependence of the optical depth of biomass burning, urban, and desert dust aerosols, J. Geophys. Res., 104, 31333–31349, 1999.
Akagi, S. K., Yokelson, R. J., Wiedinmyer, C., Alvarado, M. J., Reid, J. S., Karl, T., Crounse, J. D., and Wennberg, P. O.: Emission factors for open and domestic biomass burning for use in atmospheric models, Atmos. Chem. Phys., 11, 4039–4072, https://doi.org/10.5194/acp-11-4039-2011, 2011.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC1 -
RC2: 'Comment on egusphere-2022-21', Anonymous Referee #2, 19 May 2022
I broadly agree with the other review comment from May 5 2022 so will not repeat those comments here. My main concern is that this is framed as an analysis to better understand biomass burning aerosol properties in different regions. But as the other reviewer notes the material here is more or less known already from previous laboratory, ground, airborne, and satellite studies. So it feels like the authors decided to do an exploratory analysis with their satellite product (they are part of the SGLI algorithm development team) and report what they found, rather than setting out to design an analysis aimed to improve our understanding of biomass burning using the most relevant data sets possible.
To me this puts the paper in a difficult place. I can see two directions. As the other reviewed noted, this data product does not appear to have been thoroughly validated (the ATBD linked is not enough as it does not provide evaluation) which casts all of the results into doubt. One path would be to use the data collected, together with AERONET sites, to do an evaluation of the SGLI aerosol data product in biomass burning conditions. This would be valuable for the data user community. However such a manuscript would be more suitable for AMT than ACP so in that case I would recommend the authors withdraw and resubmit a revised version there.
The other possible direction for the paper would be to try to learn something new about biomass burning aerosols. That probably needs a much deeper analysis using multiple satellite products together with ground or model data. To me it feels not enough to just do an analysis using only your own satellite product and ignoring the many other useful data sets that can complement it, if the real motivation here is scientific study as opposed to promoting your own data product. As it is I am afraid I am not sure I learned very much from this study and don’t see a situation where I might cite the work on my own research.
Either way to me this feels like a paper that should be withdrawn and resubmitted as the changes would go beyond the scope of major revisions.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC2 -
RC3: 'Comment on egusphere-2022-21', Anonymous Referee #3, 23 May 2022
Review of “Aerosol Optical Properties of Extreme Global Wildfires and Estimated Radiative Forcing with GCOM-C SGLI” by Tanada et al.
This paper showed an analysis of optical properties of aerosols emitted from wildfires by retrieved product from the GCOM-C/SGLI satellite imager. The authors tried to estimate the impacts in aerosol optical properties and radiative forcing by wildfires and the difference in those between six major biomass burning regions. I think this paper will be of interest to readers, but there are some concerns that need to be resolved before publication.
Major comments,
1)
One of my major concerns is that any validation for the satellite retrievals is not included in this paper. It would be quantitatively difficult to estimate aerosol properties for a thick aerosol layer such as a wildfire. The data should contain a large uncertainty. The author made comparisons between different sources and quantitative estimates from the satellite retrievals, but no mention is made for the uncertainty.2)
The conclusion is unclear. Particularly, it was not clear what Section 3.5 or the RF_SW column of Table
3 were trying to say.3)
The authors showed how the optical properties change with time (or distance). However, they discuss the cause of this change only in humidification effects. As is well known, other processes also have a significant effect on aerosol aging. This point should be included in the discussion.Specific comments
Line 127: It seems to me that the differences in vegetation just made it look like there is a correlation.Line 191-193: Is this part discussing the effect of black carbon deposited on snow? If so, it should be clarified as such. But isn't the effect of deposition irrelevant to the discussion that follows?
Line 248: Period dropped from the sentence.
Line 274: "2010" is duplicated.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC3 -
AC1: 'Comment on egusphere-2022-21', Kazuhisa Tanada, 24 Jun 2022
Dear All Referees,
My colleagues and I appreciate your sincerely comments and pointing out.
We are going to take your suggestions seriously.In conclusion, we have decided to withdraw it and resubmit to another journal close to the remote sensing field such as AMT (Atmospheric Measurement Techniques) after making corrections and improvements (also including minor comments).
We plan to add the evaluation results of the SGLI data with analyzing the other reliable data (e.g. AERONET, MODIS and so on) in biomass buring conditions.
Best regards,
Kazuhisa TanadaCitation: https://doi.org/10.5194/egusphere-2022-21-AC1
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2022-21', Anonymous Referee #1, 05 May 2022
Review of “Aerosol Optical Properties of Extreme Global Wildfires and Estimated Radiative Forcing with GCOM-C SGLI” by Tanada et al.
The authors presented an analysis of “wildfire-emitted” aerosol optical properties retrieved from the GCOM-C SGLI satellite instrument and representative of six major biomass burning regions of the Earth. The manuscript further attempts to estimate the changes in TOA radiative forcing due to the extreme wildfire events during the 2018-21 period, separately for the six regions. There is no denying the significance of the motivation behind this work, along with the excitement of having a newer dataset available for constraining the aerosol properties in our global models.
In my opinion, however, the manuscript is unsuitable for publication in its current form. Major additional work and analysis are needed before resubmitting. Moreover, the findings are not revealing, and the discussions do not add a lot to our existing knowledge about aerosol optical properties. My most significant concern is that often too much is interpreted from the retrievals that are not thoroughly vetted. The flaws in the analysis are not helped by challenging readability (and some errors/inconsistencies). Please consider the points made below as grounds for this assessment.
Major comments:
- It is important to understand that satellite observations of AOD (and more importantly of derived properties such as SSA) are after all retrievals and require a thorough validation/evaluation, at least using the aerosol products from other existing passive sensors (such as MODIS/VIIRS) and ground-based observations from AERONET sun photometers wherever available. Unless there is an existing publication demonstrating the robustness and uncertainty limits of the data set (other than the ATBD document that the authors point to, which is not peer-reviewed), authors should be cautious before using the retrievals for making detailed interpretations about aerosol properties.
- The title and manuscript suggest that the analysis focuses on the aerosol properties emitted from extreme wildfire events and the radiative impacts due to those. However, the claim is kind of misleading for the following reasons: (1) it is not very clear how are extreme wildfires detected? The authors do mention vaguely in section 2.1 that SGLI’s WFRP research product is being used to detect the hotspot locations of wildfires, but what criterion is being used to distinguish between extreme wildfires or other fires is not explained. Also, it is well-known that satellite imagers have issues detecting the accurate magnitudes of fire radiative power when thick plumes of smoke emitted from the wildfire itself cover the fire locations, therefore the uncertainty in detecting extreme wildfires using this technique is not discussed. (2) Even though the focus is on the wildfire-emitted aerosols, throughout the manuscript (barring section 3.4) the aerosol properties are discussed in the context of monthly means (including the calculations of direct radiative forcing). So, considering Angola for example (and it might be true for other regions also), where August month sees significant biomass burning from savannas and grasslands as well, how are the authors secluding the impact of aerosols emitted from wildfires only?
- Changes in SSA and AE with respect to relative humidity and vegetation: the entire section 3.3. discusses how SSA increases and AE decreases with increasing relative humidity. Firstly, the analysis does not add anything new to our current understanding because it is well understood in the literature that swelling of particles leads to enhanced scattering and hence higher SSA. Similarly, AE would obviously be lower for swollen/humidity-grown particles. Another conclusion from this analysis is that SSA varies depending on vegetation types, which again I think is well known and which is why models consider different emission factors for emission constituents depending on different biomes (e.g., boreal forests versus tropical forests, see Akagi et al. (2011)). In fact, apart from vegetation, the differences in the composition (or SSA) of the freshly emitted aerosol mixture could also arise from the type of burning i.e., flaming versus smoldering, which is not even discussed here.
- Finally, the discussion on aerosol aging effects is again limited to the humification effects only. The section lacks a discussion on changes in SSA due to mixing and chemical processes, such as lensing (where BC is coated with organic material) or photobleaching (or photo-oxidation) of brown carbon. When considered together, the impacts on aerosol absorption might not be straightforward to infer based on mere relative humidity justifications.
Following are some examples of specific errors and inconsistencies:
Line 34-35: “The remaining 30 % are water, CO, and various particles (Council, 2004)”. I don’t think this statement is correct because apart from the things listed here, there are other GHG emitted such as methane, nitrous oxide, as well as volatile organic compounds (VOCs), that can act as precursors for secondary organic aerosols.
Line 39: “Organic carbon reflects solar radiation and causes a negative forcing.” Again, not completely true because brown carbon (a subset of the organic carbon) has significant absorption in the near-UV wavelengths.
Line 44: what is the reference for the estimate of indirect forcing listed here, please clarify and if possible demonstrate the variability of these estimates based on more recent studies in the literature. Also, what about aerosol semi-direct effects?
Line 47: Not just particle size distribution and chemical composition, optical properties are also largely dependent on aerosol hygroscopicity and mixing state.
Line 48: This sentence is wrong, and the in-text citation is not even included in the bibliography. Most of the smoke particles are in fact dominated by fine or accumulation mode (Eck et al. 1999), while dust and sea salt would be coarse mode dominated.
Line 54: What is direct radiative effect efficiency? Please be consistent with the use of radiative forcing, direct radiative effect, indirect forcing, indirect effect, etc. in the text throughout. There are subtle differences between each of these definitions.
Line 71: It mentions that errors correspond to a 1 sigma confidence level throughout the manuscript, yet error depiction is missing for several critical figures such as Fig. 1 and 7.
Other discrepancies: It is not mentioned how the cloud presence/contamination is handled in the dataset? Unless we are only looking at clear-sky data, but if so, it’s nowhere mentioned until we reach the discussions specific to radiative forcing calculations.
Fig. 9 shows Angola base-period/no fire period as July 2020 and Table 3. Suggests base period is 11-20 Sep 2020. Which one is it? Moreover, a quick look at MODIS/VIIRS observed fire pixels will tell you that there are enough fire pixels during both the base periods mentioned here.
References:
Eck, T. F., Holben, B. N., Reid, J. S., Dubovik, O., Smirnov, A., O’Neill, N. T., Slutsker, I., and Kinne, S.: Wavelength dependence of the optical depth of biomass burning, urban, and desert dust aerosols, J. Geophys. Res., 104, 31333–31349, 1999.
Akagi, S. K., Yokelson, R. J., Wiedinmyer, C., Alvarado, M. J., Reid, J. S., Karl, T., Crounse, J. D., and Wennberg, P. O.: Emission factors for open and domestic biomass burning for use in atmospheric models, Atmos. Chem. Phys., 11, 4039–4072, https://doi.org/10.5194/acp-11-4039-2011, 2011.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC1 -
RC2: 'Comment on egusphere-2022-21', Anonymous Referee #2, 19 May 2022
I broadly agree with the other review comment from May 5 2022 so will not repeat those comments here. My main concern is that this is framed as an analysis to better understand biomass burning aerosol properties in different regions. But as the other reviewer notes the material here is more or less known already from previous laboratory, ground, airborne, and satellite studies. So it feels like the authors decided to do an exploratory analysis with their satellite product (they are part of the SGLI algorithm development team) and report what they found, rather than setting out to design an analysis aimed to improve our understanding of biomass burning using the most relevant data sets possible.
To me this puts the paper in a difficult place. I can see two directions. As the other reviewed noted, this data product does not appear to have been thoroughly validated (the ATBD linked is not enough as it does not provide evaluation) which casts all of the results into doubt. One path would be to use the data collected, together with AERONET sites, to do an evaluation of the SGLI aerosol data product in biomass burning conditions. This would be valuable for the data user community. However such a manuscript would be more suitable for AMT than ACP so in that case I would recommend the authors withdraw and resubmit a revised version there.
The other possible direction for the paper would be to try to learn something new about biomass burning aerosols. That probably needs a much deeper analysis using multiple satellite products together with ground or model data. To me it feels not enough to just do an analysis using only your own satellite product and ignoring the many other useful data sets that can complement it, if the real motivation here is scientific study as opposed to promoting your own data product. As it is I am afraid I am not sure I learned very much from this study and don’t see a situation where I might cite the work on my own research.
Either way to me this feels like a paper that should be withdrawn and resubmitted as the changes would go beyond the scope of major revisions.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC2 -
RC3: 'Comment on egusphere-2022-21', Anonymous Referee #3, 23 May 2022
Review of “Aerosol Optical Properties of Extreme Global Wildfires and Estimated Radiative Forcing with GCOM-C SGLI” by Tanada et al.
This paper showed an analysis of optical properties of aerosols emitted from wildfires by retrieved product from the GCOM-C/SGLI satellite imager. The authors tried to estimate the impacts in aerosol optical properties and radiative forcing by wildfires and the difference in those between six major biomass burning regions. I think this paper will be of interest to readers, but there are some concerns that need to be resolved before publication.
Major comments,
1)
One of my major concerns is that any validation for the satellite retrievals is not included in this paper. It would be quantitatively difficult to estimate aerosol properties for a thick aerosol layer such as a wildfire. The data should contain a large uncertainty. The author made comparisons between different sources and quantitative estimates from the satellite retrievals, but no mention is made for the uncertainty.2)
The conclusion is unclear. Particularly, it was not clear what Section 3.5 or the RF_SW column of Table
3 were trying to say.3)
The authors showed how the optical properties change with time (or distance). However, they discuss the cause of this change only in humidification effects. As is well known, other processes also have a significant effect on aerosol aging. This point should be included in the discussion.Specific comments
Line 127: It seems to me that the differences in vegetation just made it look like there is a correlation.Line 191-193: Is this part discussing the effect of black carbon deposited on snow? If so, it should be clarified as such. But isn't the effect of deposition irrelevant to the discussion that follows?
Line 248: Period dropped from the sentence.
Line 274: "2010" is duplicated.
Citation: https://doi.org/10.5194/egusphere-2022-21-RC3 -
AC1: 'Comment on egusphere-2022-21', Kazuhisa Tanada, 24 Jun 2022
Dear All Referees,
My colleagues and I appreciate your sincerely comments and pointing out.
We are going to take your suggestions seriously.In conclusion, we have decided to withdraw it and resubmit to another journal close to the remote sensing field such as AMT (Atmospheric Measurement Techniques) after making corrections and improvements (also including minor comments).
We plan to add the evaluation results of the SGLI data with analyzing the other reliable data (e.g. AERONET, MODIS and so on) in biomass buring conditions.
Best regards,
Kazuhisa TanadaCitation: https://doi.org/10.5194/egusphere-2022-21-AC1
Data sets
CGLS-LC100 Copernicus Land Service https://zenodo.org/record/3939050#.Yh2VuZPP3-a
Global objective analysis data Japan Meteorological Agency http://gpvjma.ccs.hpcc.jp/~gpvjma/#
GCOM-C/SGLI research products Japan Aerospace Exploration Agency https://kuroshio.eorc.jaxa.jp/JASMES/index.html
GCOM-C/SGLI standard products Japan Aerospace Exploration Agency https://gportal.jaxa.jp/gpr/index/eula
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 853 | 214 | 43 | 1,110 | 30 | 30 |
- HTML: 853
- PDF: 214
- XML: 43
- Total: 1,110
- BibTeX: 30
- EndNote: 30
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1