the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 19 Sep 2023
Measurement report: Assessing the Impacts of Emission Uncertainty on Aerosol Optical Properties and Radiative Forcing from Biomass Burning in Peninsular Southeast Asia
Yinbao Jin
Xiaoyang Chen
Haofan Wang
Yinping Cui
Yifei Xu
Siting Li
Ming Zhang
Yingying Ma
Qi Fan
Abstract. Despite significant advancements in improving the dataset for biomass burning (BB) emissions over the past few decades, uncertainties persist in BB aerosol emissions, impeding the accurate assessment of simulated aerosol optical properties (AOPs) and direct radiative forcing (DRF) during wildfire events in global and regional models. This study assessed AOPs (including aerosol optical depth (AOD), aerosol absorption optical depth (AAOD), and aerosol extinction coefficients (AEC)) and DRF using eight independent BB emission inventories applied to the WRF-Chem model during the BB period (March 2019) in Peninsular Southeast Asia (PSEA), where the eight BB emission inventories were the Global Fire Emissions Database version 4.1s (GFED), Fire INventory from NCAR version 1.5 (FINN1.5), the Fire Inventory from NCAR version 2.5 MOS (MODIS fire detections, FINN2.5 MOS), the Fire Inventory from NCAR version 2.5 MOSVIS (MODIS+VIIRS fire detections, FINN2.5 MOSVIS), Global Fire Assimilation System version 1.2s (GFAS), Fire Energetics and Emissions Research version 1.0 (FEER), Quick Fire Emissions Dataset version 2.5 release 1 (QFED), and Integrated Monitoring and Modelling System for Wildland FIRES Project version 2.0 (IS4FIRES), respectively. The results show that in the PSEA region, organic carbon (OC) emissions in the eight BB emission inventories differ by a factor of about 9 (0.295–2.533 Tg/M), with 1.09 ± 0.83 Tg/M and a coefficient of variation (CV) of 76 %. High-concentration OC emissions occurred primarily in savanna and agricultural fires. The OC emissions from the GFED and GFAS are significantly lower than the other inventories. The OC emissions in FINN2.5 VISMOS are approximately twice as high as those in FINN1.5. Sensitivity analysis of AOD simulated by WRF-Chem to different BB emission datasets indicated that the FINN scenarios (v1.5 and 2.5) significantly overestimate AOD compared to observation (VIIRS), while the other inventories underestimate AOD in the high AOD (HAOD, AOD>1) regions range from 97–110° E, 15–22.5° N. Among the eight schemes, IS4FIRES and FINN1.5 performed better in terms of AOD simulation consistency and bias in the HAOD region when compared to AERONET sites. The AAOD in WRF-Chem during the PSEA wildfire period was assessed using satellite observations (TROPOMI) and AERONET data, and it was found that the AAOD simulated with different BB schemes did not perform as well as the AOD. The significant overestimation of AAOD by FINN (v1.5 and 2.5), FEER, and IS4FIRES schemes in the HAOD region, with the largest overestimation for FINN2.5 MOSVIS. FINN1.5 schemes performed better in representing AAOD at AERONET sites within the HAOD region. The simulated AOD and AAOD from FINN2.5 MOSVIS always show the best correlation with the observations. AEC simulated by WRF-Chem with all the eight BB schemes trends were consistent with CALIPSO in the vertical direction (0.5 km to 4 km), demonstrating the efficacy of the smoke plume rise model used in WRF-Chem to simulate smoke plume heights. However, the FINN (v1.5 and 2.5) schemes overestimated AEC, while the other schemes underestimated it. In the HAOD region, BB aerosols exhibited a daytime shortwave radiative forcing of -32.60±24.50 W/m2 at the surface, positive forcing (1.70±1.40 W/m2) in the atmosphere, and negative forcing (-30.89±23.6 W/m2) at the top of the atmosphere. Based on the analysis, FINN1.5 and IS4FIRES are recommended for accurately assessing the impact of BB on air quality and climate in the PSEA region.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(4981 KB)
-
Supplement
(1259 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4981 KB) - Metadata XML
-
Supplement
(1259 KB) - BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Yinbao Jin et al.
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1650', Anonymous Referee #1, 12 Oct 2023
Jin et al. present comparison of different biomass burning emissions inventories for the Peninsular Southeast Asia. These different biomass burning emission inventories were ran in the Weather Research and Forecasting (WRF) - Chemical (WRF-Chem) chemical transport model, using the Model for Ozone and Related Chemical Tracers (MOZART) chemistry with Model for Simulating Aerosol Interactions and Chemistry (MOSAIC), along with numerous parameterizations for atmospheric dynamics. The authors compare the WRF-Chem output of these 8 different emission inventories against themselves and satellite products of aerosol, including aerosol optical depth (AOD), absorbing aerosol optical depth (AAOD), and aerosol extinction. The paper and results would be of interest for the community and ACP, especially due to the increasing biomass burning. The authors should address the following comments to be published in ACP:
(1) To further emphasize why the month that was used for the simulations as it was one line in the introduction that may be lost, it would be good in Fig. 1 (or another figure), to show the total fire counts in Peninsular Southeast Asia.
(2) How does the model treat the inorganic aerosol? E.g., it is not clear if the inorganic aerosol is treated thermodynamically or not. This is important to better understand how the model may be treating aerosol liquid water, aerosol acidity, etc., which all impact the physicochemical properties of the aerosol and thus the aerosols' optical properties.
(3) As the results are presented, it is currently not clear what the purpose of the satellite products comparisons with the model results provides for the conclusions. E.g., the authors discuss how different emission inventories provide different agreement depending on the satellite product and/or land-based product, which indicates no emission inventory is superior. Further, the authors have not provided or discussed the following properties that would be potentially of more interest/importance in understanding the aerosol from biomass burning to compare with observations and products:
(a) What is the aerosol composition with each emission inventory? E.g., how much primary vs secondary organic carbon/aerosol? How much secondary inorganic aerosol vs organic aerosol? How much black carbon vs these other components? All these aspects impact the hygroscopicity of the aerosol, and thus how it would be retrieved by satellite and ground-based measurements.
(b) How does the size distribution change amongst the different emission inventories? Similar to the chemical composition, the sizes would impact both water uptake, scattering, and how well the satellite and ground-based observations detect the aerosol.
(c) What is the oxidation state, e.g., O/C and H/C ratio, of the primary and secondary aerosol? Similar to (a), the amount of oxidation of the organic aerosol/carbon will impact its physicochemical properties and how it would be retrieved.
(d) Besides retrieval, all these properties would impact the aerosols role in clouds and radiative forcing, making it important to understand how much these differences may impact the differences presented in the different figures.(3) Without the information provided in (2), the intercomparisons of the model and observed PM2.5 is hard to interpret, as the models may be getting PM2.5 correct for the incorrect reason. Also, it is unclear in the intercomparison of the model with observed PM2.5 for one fire emission inventory how to interpret the results as (a) it seems most of the PM2.5 was measured in urban areas, meaning the urban emissions may be driving the intercomparison more than fire emissions and (b) the emission inventory used for the intercomparison and validation of the model has mixed results (e.g., Table 2).
(4) Due to (2) and (3), the paper may be presented better as a comparison against the emission inventories without comparison with satellite and ground based products as it is not clear that there is a better emission inventory to used currently for chemical transport models. More discussion could be placed into the description in the similarity and differences in the physicochemical properties due to differences in the emission inventory, which would be of extreme interest towards the community.
Minor
(1) For all figures, please label either which emission inventory is being used or what location the observations/model is for. It is currently difficult to interpret the figures without this key information.
(2) Please check figures and tables. There are many instances of inconsistencies or typos in the labels (e.g., line 103 says red line around the study area, Fig. 4 has methanal which is formaldehyde and then an abbreviation for methylglyoxal (Mgly) and methyl vinyl ketone twice with MACR for one, etc).
(3) It is highly recommended to not use rainbow for color bars. Rainbow color bars can be difficult to interpret due to color blindness and the contrast between colors can be difficult to observe differences. Similarly, the color bar in Fig. 6c and Fig. 10c is extremely difficult to read and interpret any differences.
(4) Table S4. Please include location for each met station.
(5) Please introduce the supplemental figures and tables in numerical order. E.g., right now, one supplemental table with a higher numerical value is introduced prior to a lower numerical value table, making the reader jump between tables.
Citation: https://doi.org/10.5194/egusphere-2023-1650-RC1 -
AC2: 'Reply on RC1', Yinbao Jin, 11 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1650/egusphere-2023-1650-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Yinbao Jin, 11 Nov 2023
-
RC2: 'Comment on egusphere-2023-1650', Anonymous Referee #2, 13 Oct 2023
Jin et al., 2023 present a comparison study of 8 biomass burning (BB) inventories using the Weather Research Forecasting model coupled with Chemistry (WRF-Chem) configured with the Model for Ozone and Related chemical Tracers (MOZART) and the Model for Simulating Aerosol Interactions and Chemistry (MOSAIC). They assess how these BB inventories impact aerosol optical properties (AOPs) such as, aerosol optical depth (AOD), aerosol absorption optical depth (AAOD), and aerosol extinction coefficients (AEC). Additionally, the direct radiative forcing (DRF) of BB aerosol was assessed. The AOPs were compared against ground and satellite-based measurements. This study is valuable to the ACP community as BB events increase in frequency, furthering the need to understand the biases certain BB inventories impose on AOPs. With that said, the authors need to address a number points prior to publication.
What is the rationale behind choosing March 2019 as the study timeframe? How does the fire season (March 2019) compare to other fire seasons in the region, was it representative of the average conditions (or anomalously high/low)?
The influence of external dust aerosol on AEC is mentioned in section 3.5, 4.2, and in the summary and conclusions. Can you provide more details on how external dust (or other inorganic aerosols e.g., sea salt aerosol) impacts the AEC profiles?
Lines 553 – 556, Jin et al. mention that when direct and indirect radiation feedbacks are included in WRF-Chem they improve the representation of AOPs, but indirect radiation feedbacks are not included in their simulations. Jin et al., mention that this, “may also lead to biases in the AOPs” (Line 556), but what specifically are those biases? Please expand on this point.
The semi-direct effect from absorbing aerosols (AAs) is another important process that impacts DRF. AAs are effective at absorbing shortwave radiation in the atmosphere and can burn-off clouds (impacting DRF). Is this process included in this modelling framework? A useful study for this may be Mallet et al., 2020.
Understanding more details of the aerosol composition in the BB inventories will be useful. How are aerosol mixing processes (external and internal mixed aerosol) included in your modelling framework? These mixing processes will impact the hygroscopicity of aerosols, impacting AOPs depending on the aerosol composition of each inventory.
Minor points are below.
Figures 3, 4, and 13 should have the inventories labelled on the top of the panel. This will make it easier to interpret the results.
Line 152 – is “gas” referring to SO2 and NH3 (as it is on line 310)? If so, I might suggest just stating SO2 and NH3 explicitly as “gas” is somewhat ambiguous.
Lines 321 – 322, Jin et al., mention that QFED exhibits a lower BC to OC ratio compared to the other inventories. Do you have any comments as to why this inventory leads to a lower BC/OC compared to the other inventories?
Line 493 – 494, “(with FINN2.5 MOSVIS reaching a maximum of 70 W m-2)” Please make it clearer what maximum you are referring to.
In table 1, I suggest changing the “Main EF” label to “EF reference (s)”. Make it clear that these are references.
(As an example) Line 45 uses “W/m2”, please change all instances of this to “W m-2”.
On figure S1, please remove the “figure” label at the top left.
Figure S2, make it clearer which letter labels refer to which of the 23 cities.
References:
Mallet, M., Solmon, F., Nabat, P., Elguindi, N., Waquet, F., Bouniol, D., Sayer, A. M., Meyer, K., Roehrig, R., Michou, M., Zuidema, P., Flamant, C., Redemann, J., and Formenti, P.: Direct and semi-direct radiative forcing of biomass-burning aerosols over the southeast Atlantic (SEA) and its sensitivity to absorbing properties: a regional climate modeling study, Atmos. Chem. Phys., 20, 13191–13216, https://doi.org/10.5194/acp-20-13191-2020, 2020.
-
AC1: 'Reply on RC2', Yinbao Jin, 11 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1650/egusphere-2023-1650-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Yinbao Jin, 11 Nov 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1650', Anonymous Referee #1, 12 Oct 2023
Jin et al. present comparison of different biomass burning emissions inventories for the Peninsular Southeast Asia. These different biomass burning emission inventories were ran in the Weather Research and Forecasting (WRF) - Chemical (WRF-Chem) chemical transport model, using the Model for Ozone and Related Chemical Tracers (MOZART) chemistry with Model for Simulating Aerosol Interactions and Chemistry (MOSAIC), along with numerous parameterizations for atmospheric dynamics. The authors compare the WRF-Chem output of these 8 different emission inventories against themselves and satellite products of aerosol, including aerosol optical depth (AOD), absorbing aerosol optical depth (AAOD), and aerosol extinction. The paper and results would be of interest for the community and ACP, especially due to the increasing biomass burning. The authors should address the following comments to be published in ACP:
(1) To further emphasize why the month that was used for the simulations as it was one line in the introduction that may be lost, it would be good in Fig. 1 (or another figure), to show the total fire counts in Peninsular Southeast Asia.
(2) How does the model treat the inorganic aerosol? E.g., it is not clear if the inorganic aerosol is treated thermodynamically or not. This is important to better understand how the model may be treating aerosol liquid water, aerosol acidity, etc., which all impact the physicochemical properties of the aerosol and thus the aerosols' optical properties.
(3) As the results are presented, it is currently not clear what the purpose of the satellite products comparisons with the model results provides for the conclusions. E.g., the authors discuss how different emission inventories provide different agreement depending on the satellite product and/or land-based product, which indicates no emission inventory is superior. Further, the authors have not provided or discussed the following properties that would be potentially of more interest/importance in understanding the aerosol from biomass burning to compare with observations and products:
(a) What is the aerosol composition with each emission inventory? E.g., how much primary vs secondary organic carbon/aerosol? How much secondary inorganic aerosol vs organic aerosol? How much black carbon vs these other components? All these aspects impact the hygroscopicity of the aerosol, and thus how it would be retrieved by satellite and ground-based measurements.
(b) How does the size distribution change amongst the different emission inventories? Similar to the chemical composition, the sizes would impact both water uptake, scattering, and how well the satellite and ground-based observations detect the aerosol.
(c) What is the oxidation state, e.g., O/C and H/C ratio, of the primary and secondary aerosol? Similar to (a), the amount of oxidation of the organic aerosol/carbon will impact its physicochemical properties and how it would be retrieved.
(d) Besides retrieval, all these properties would impact the aerosols role in clouds and radiative forcing, making it important to understand how much these differences may impact the differences presented in the different figures.(3) Without the information provided in (2), the intercomparisons of the model and observed PM2.5 is hard to interpret, as the models may be getting PM2.5 correct for the incorrect reason. Also, it is unclear in the intercomparison of the model with observed PM2.5 for one fire emission inventory how to interpret the results as (a) it seems most of the PM2.5 was measured in urban areas, meaning the urban emissions may be driving the intercomparison more than fire emissions and (b) the emission inventory used for the intercomparison and validation of the model has mixed results (e.g., Table 2).
(4) Due to (2) and (3), the paper may be presented better as a comparison against the emission inventories without comparison with satellite and ground based products as it is not clear that there is a better emission inventory to used currently for chemical transport models. More discussion could be placed into the description in the similarity and differences in the physicochemical properties due to differences in the emission inventory, which would be of extreme interest towards the community.
Minor
(1) For all figures, please label either which emission inventory is being used or what location the observations/model is for. It is currently difficult to interpret the figures without this key information.
(2) Please check figures and tables. There are many instances of inconsistencies or typos in the labels (e.g., line 103 says red line around the study area, Fig. 4 has methanal which is formaldehyde and then an abbreviation for methylglyoxal (Mgly) and methyl vinyl ketone twice with MACR for one, etc).
(3) It is highly recommended to not use rainbow for color bars. Rainbow color bars can be difficult to interpret due to color blindness and the contrast between colors can be difficult to observe differences. Similarly, the color bar in Fig. 6c and Fig. 10c is extremely difficult to read and interpret any differences.
(4) Table S4. Please include location for each met station.
(5) Please introduce the supplemental figures and tables in numerical order. E.g., right now, one supplemental table with a higher numerical value is introduced prior to a lower numerical value table, making the reader jump between tables.
Citation: https://doi.org/10.5194/egusphere-2023-1650-RC1 -
AC2: 'Reply on RC1', Yinbao Jin, 11 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1650/egusphere-2023-1650-AC2-supplement.pdf
-
AC2: 'Reply on RC1', Yinbao Jin, 11 Nov 2023
-
RC2: 'Comment on egusphere-2023-1650', Anonymous Referee #2, 13 Oct 2023
Jin et al., 2023 present a comparison study of 8 biomass burning (BB) inventories using the Weather Research Forecasting model coupled with Chemistry (WRF-Chem) configured with the Model for Ozone and Related chemical Tracers (MOZART) and the Model for Simulating Aerosol Interactions and Chemistry (MOSAIC). They assess how these BB inventories impact aerosol optical properties (AOPs) such as, aerosol optical depth (AOD), aerosol absorption optical depth (AAOD), and aerosol extinction coefficients (AEC). Additionally, the direct radiative forcing (DRF) of BB aerosol was assessed. The AOPs were compared against ground and satellite-based measurements. This study is valuable to the ACP community as BB events increase in frequency, furthering the need to understand the biases certain BB inventories impose on AOPs. With that said, the authors need to address a number points prior to publication.
What is the rationale behind choosing March 2019 as the study timeframe? How does the fire season (March 2019) compare to other fire seasons in the region, was it representative of the average conditions (or anomalously high/low)?
The influence of external dust aerosol on AEC is mentioned in section 3.5, 4.2, and in the summary and conclusions. Can you provide more details on how external dust (or other inorganic aerosols e.g., sea salt aerosol) impacts the AEC profiles?
Lines 553 – 556, Jin et al. mention that when direct and indirect radiation feedbacks are included in WRF-Chem they improve the representation of AOPs, but indirect radiation feedbacks are not included in their simulations. Jin et al., mention that this, “may also lead to biases in the AOPs” (Line 556), but what specifically are those biases? Please expand on this point.
The semi-direct effect from absorbing aerosols (AAs) is another important process that impacts DRF. AAs are effective at absorbing shortwave radiation in the atmosphere and can burn-off clouds (impacting DRF). Is this process included in this modelling framework? A useful study for this may be Mallet et al., 2020.
Understanding more details of the aerosol composition in the BB inventories will be useful. How are aerosol mixing processes (external and internal mixed aerosol) included in your modelling framework? These mixing processes will impact the hygroscopicity of aerosols, impacting AOPs depending on the aerosol composition of each inventory.
Minor points are below.
Figures 3, 4, and 13 should have the inventories labelled on the top of the panel. This will make it easier to interpret the results.
Line 152 – is “gas” referring to SO2 and NH3 (as it is on line 310)? If so, I might suggest just stating SO2 and NH3 explicitly as “gas” is somewhat ambiguous.
Lines 321 – 322, Jin et al., mention that QFED exhibits a lower BC to OC ratio compared to the other inventories. Do you have any comments as to why this inventory leads to a lower BC/OC compared to the other inventories?
Line 493 – 494, “(with FINN2.5 MOSVIS reaching a maximum of 70 W m-2)” Please make it clearer what maximum you are referring to.
In table 1, I suggest changing the “Main EF” label to “EF reference (s)”. Make it clear that these are references.
(As an example) Line 45 uses “W/m2”, please change all instances of this to “W m-2”.
On figure S1, please remove the “figure” label at the top left.
Figure S2, make it clearer which letter labels refer to which of the 23 cities.
References:
Mallet, M., Solmon, F., Nabat, P., Elguindi, N., Waquet, F., Bouniol, D., Sayer, A. M., Meyer, K., Roehrig, R., Michou, M., Zuidema, P., Flamant, C., Redemann, J., and Formenti, P.: Direct and semi-direct radiative forcing of biomass-burning aerosols over the southeast Atlantic (SEA) and its sensitivity to absorbing properties: a regional climate modeling study, Atmos. Chem. Phys., 20, 13191–13216, https://doi.org/10.5194/acp-20-13191-2020, 2020.
-
AC1: 'Reply on RC2', Yinbao Jin, 11 Nov 2023
The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2023/egusphere-2023-1650/egusphere-2023-1650-AC1-supplement.pdf
-
AC1: 'Reply on RC2', Yinbao Jin, 11 Nov 2023
Peer review completion
Journal article(s) based on this preprint
Yinbao Jin et al.
Data sets
Global Fire Emissions Database, Version 4.1 (GFEDv4) Randerson, J. T., van der Werf, G. R., Giglio, L., Collatz G. J., and Kasibhatla, P. S. https://doi.org/10.3334/ORNLDAAC/1293
The Fire INventory from NCAR (FINN): a high resolution global model to estimate the emissions from open burning Wiedinmyer, C., Akagi, S. K., Yokelson, R. J., Emmons, L. K., Al-Saadi, J. A., Orlando, J. J., and Soja, A. J. https://www.acom.ucar.edu/Data/fire/
The Fire Inventory from NCAR version 2.5: an updated global fire emissions model for climate and chemistry applications Wiedinmyer, C., Kimura, Y., McDonald-Buller, E. C., Emmons, L. K., Buchholz, R. R., Tang, W., Seto, K., Joseph, M. B., Barsanti, K. C., Carloton, A. G., and Yokelson, R. https://www.acom.ucar.edu/Data/fire/
CAMS global biomass burning emissions based on fire radiative power (GFAS) ECWMF https://ads.atmosphere.copernicus.eu/cdsapp#!/dataset/cams-global-fire-emissions-gfas?tab=form
Global top-down smoke-aerosol emissions estimation using satellite fire radiative power measurements Ichoku, C. and Ellison, L. https://feer.gsfc.nasa.gov/data/emissions/
QFED - High Resolution Global Fire Emissions Darmenov, A., da Silva, A., and Govindaraju, R. https://portal.nccs.nasa.gov/datashare/iesa/aerosol/emissions/QFED/v2.5r1/
Uncertainties of wild-land fires emission in AQMEII phase 2 case study Soares, J., Sofiev, M., and Hakkarainen, J. http://silam.fmi.fi/thredds/catalog/i4f20emis-arch/catalog.html
Yinbao Jin et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 313 | 99 | 19 | 431 | 37 | 16 | 11 |
- HTML: 313
- PDF: 99
- XML: 19
- Total: 431
- Supplement: 37
- BibTeX: 16
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Cited
1 citations as recorded by crossref.
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(4981 KB) - Metadata XML
-
Supplement
(1259 KB) - BibTeX
- EndNote
- Final revised paper