the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Satellite (GOSAT-2 CAI-2) retrieval and surface (ARFINET) observations of Aerosol Black Carbon over India
Mukunda M. Gogoi
S. Suresh Babu
Ryoichi Imasu
Makiko Hashimoto
Abstract. The light-absorbing Black Carbon (BC) aerosols have very sensitive role in affecting the Earth’s radiation budget and climate. In this study, satellite-based retrieval of BC over India is presented based on observations from the Cloud and Aerosol-Imager-2 (CAI-2) on-board the Greenhouse gases Observing Satellite-2 (GOSAT-2). To evaluate and validate the satellite retrievals, near surface BC mass concentrations measured across a network of aerosol observatories (ARFINET) over India are used and the findings are extended to comprehend the global BC features. As the analysis revealed, this satellite retrieval fairly depicts the regional and seasonal features of BC over the Indian region, which are similar to those recorded by surface observations. The validation and closure studies between the two data sets show RMSE < 1 and absolute difference below 2 µg m-3 for > 60 % of simultaneous observations, possessing fairly good associations in Dec-Jan-Feb (R ~ 0.73) and Mar-Apr-May (R ~ 0.76). Over the hotspot regions of India, the satellite retrievals show soot volume fraction of ~ 5 %, columnar single scattering albedo of ~ 0.8 and BC column optical depth of ~ 0.1; which are comparable to that of other in-situ or satellite measurements. In terms of global spatio-temporal variability, satellite retrieval shows higher BC occurring mostly in areas where biomass burning is intense. Overall, this study highlights the effectiveness of satellite retrieval of BC, which could be effectively used for the regular monitoring of BC load arising out of vehicular/ industrial/ biomass burning activities across the globe.
- Preprint
(2499 KB) -
Supplement
(2426 KB) - BibTeX
- EndNote
Mukunda M. Gogoi et al.
Status: closed
-
RC1: 'Comment on acp-2022-555', Anonymous Referee #1, 08 Oct 2022
Reviewers' comments:
The authors use extensive measurements to retrieve BC mass concentrations and compare them with satellite-retrieved BC mass concentrations. This paper contributes to better understanding of spatial and temporal distributions of BC concentrations over the Indian regions. I believe the paper should be considered for publication only after addressing the concerned expressed below.
Major Issues:
1.
At first, it is difficult to read this paper because of unclear and unappropriated sentences.
For readers to understand clearly, many sentences need to be polished with clear and concise structure including appropriate English.
2.
Secondly, this paper mostly describes how much satellite and ground-based measurement agree or not. The authors need more descriptions about why they are different and what science is behind it. For example, you may investigate less consistency between satellite-retrieved and ground-based BC concentrations in JJA compared to DJF and MAM due to the cloud contamination or a deficit of data availability during monsoon.
Minor Issues and specific comments:
Page1 L11: Is the acronym of ARFINET correctly located and explained?
Aerosol Radiation Forcing over India NETwork (ARFINET)
P1 L12: revealed -> reveals
P1 L17: that of other in-situ -> those of other in-situ
P1 L18: satellite retrieval shows -> satellite retrievals show
P1 L33: However, the very challenging task is to accurately retrieve
-> However, it is challenging is to accurately retrieve
P2 L42: have not addressed so far -> have not been addressed so far
P2 L50: 10-sepctral bands -> 10-spectral bands
P2 L50-51: use full word and abbreviation of UV, VIS, and NIR
P2 L59: to develop periodic and accurate estimates of aerosol radiative forcing over India and assess their impacts on regional and global climate, taking into account their heterogeneous properties in space, time and spectral domains
->
to develop periodic and accurate estimates of aerosol radiative forcing over India, assess their impacts on regional and global climate, and take into account their heterogeneous properties in space, time and spectral domains
P2 L71: remove etc.
P3 L95: the surface albedo is derived by performing a correction removing the influence of atmospheric 95 molecular scattering (Rayleigh scattering) ->
the surface albedo is derived by removing the influence of atmospheric 95 molecular scattering (Rayleigh scattering)
P3 L98: single scattering and multiple scattering -> single- and multiple-scattering
P3 L101: inversion algorithm developed by Hashimoto and Nakajima (2017) is used.
-> inversion algorithm (Hashimoto and Nakajima, 2017) is used.
P3 L108: This sentence should be clearly stated.
aerosol light absorption (or single scattering albedo - SSA) -> aerosol light scattering (or single scattering albedo - SSA)
If you want to describe aerosol light absorption, you can use co-albedo or 1-SSA instead of SSA.
P4 L131: Does “several sensitivity studies” means studies you performed? If not, add references.
P4 L148: Detail about the aethalometer uncertainty and correction of raw-data is available in Gogoi et al., (2017). The overall uncertainty in BC mass measured by the Aethalometer is estimated at about 10%.
-> The overall uncertainty in BC mass measured by the aethalometer is estimated at about 10% and more details are available in Gogoi et al., (2017).
P5 L161: MAE = 10 m2g-1 is used. -> MAE = 10 m2g-1 is assumed (add references).
P5 L167:
You mentioned, “As the ambient BC in the atmosphere is mostly aged in nature”
It is a vague sentence for the reason of “a value of MAE = 10 m2g-1 is used” since BC is not aged in nature if it is just released from biomass burning.
P5 L175: winter, pre-monsoon, and monsoon respectively. -> winter, pre-monsoon, and monsoon, respectively.
P5 L177:
You should add one more figure or add text into Fig. 1 to indicate each region like HIM, IGP, NEI, NWI and so on. Although it is written in the supplement, this straightforward figure would help readers to understand your figure better.
P5 L195-197:
You mentioned satellite retrievals and surface observations of BC are more consistent in DJF and MAM than JJA. Is less consistency in JJA caused by cloud contamination during monsoon? Or how many data used for this analysis for each season? Are the number of data used during monsoon fewer compared to other seasons?
P6 L204:
better associations -> better agreement
P6 L205:
the association between the two data sets -> the correlation between the two data sets
P6 L205:
Thus, despite satellite retrievals during winter and pre-monsoon months showing the regional hotspots of BC over India fairly well, there appears to be a lack of consistent associations between the two datasets in winter at some of the ARFINET observational sites.
->
Thus, satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during wither and pre-monsoon months. However, there are a lack of consistency between the two datasets in winter.
P6 L208: The above observations point to
-> The discrepancies between satellite retrievals and ground-based observations can be attributed to
P6 L215: Do not use “Despite this” and use specific words
P6 L215: the satellite retrievals differ from surface measured BC -> the satellite-retrieved BC differ from surface-measured BC
P6 L228:
we will now examine simultaneous day-to-day values -> we examine simultaneous day-to-day concentrations
P6 L223-L225:
This sentence needs to be polished.
P7 L239: play important role -> play an important role
P7 L242: Add references of ERA5
P7 L246:
in all three periods of DJF, MAM and JJA -> during all periods
P7 L249:
It has also been observed that absolute differences between the two data sets ->
Absolute differences between the two data sets
P7 L249: peninsular Indian locations -> PI
P7 L251: It is further evident from the figure -> It is further evident from Fig. 6
P7 L259:
Provide the reason of the sentence “Especially, the association between the two data sets significantly improves in JJA.”
P7 L261: add references
P7 L267: show -> shows
P7 L276: more wet soils -> wetter soils
P8 L288: The forgone observation -> The prior observation
P8 L298: Based on in-situ vertical profiling of aerosol scattering and absorption properties on a research aircraft, Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period.
->
Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period using aircraft measurements.
P8 L305:
Over the oceanic regions, the values of SSA are, in general, high
-> Over the oceanic regions, the values of SSA are generally high
P8 L310-311: Mar -> Mar.
Jun -> Jun.
You may use abbreviations of the months consistently: decide full name or abbreviations of the months.
P8 L315:
duringMar/Apr/May -> during March to May
P9 L319: Figs. 8, 9 and 10 -> Figs. 8, 9 and 10, respectively
P9 L322: day time FRP -> day-time FRP
P9 L343: add references after “Several studies”
P9 L345: evaluate and validate -> evaluate
P9 L349: remove unnecessary sentence (The main findings are as follows:)
P9 L351: do not use “fairly”. It sounds informal.
P9 L354:
for > 60% of the observations (for all the locations considered in this study) the absolute difference between the two data sets is < 2 μgm-3.
->
the absolute difference between the two data sets is less than 2 μgm-3 for over 60% of the locations in this study.
P10 L365:
during times of biomass burning -> during the biomass burning season
P22 L654:
Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
P24-26:
Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Citation: https://doi.org/10.5194/acp-2022-555-RC1 -
AC1: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
We appreciate the summary evaluation of the reviewer and agree to the observations. Following the valuable comments and fruitful suggestions for improving the quality of the manuscript, we have revised it incorporating the review comments of all the reviewers. Our point wise response to each of the comment is given in bold letters, below the respective comments.
-
AC3: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
The authors use extensive measurements to retrieve BC mass concentrations and compare them with satellite-retrieved BC mass concentrations. This paper contributes to better understanding of spatial and temporal distributions of BC concentrations over the Indian regions. I believe the paper should be considered for publication only after addressing the concerned expressed below.
We appreciate the summary evaluation of the reviewer and agree to the observations. Following the valuable comments and fruitful suggestions for improving the quality of the manuscript, we have revised it incorporating the review comments of all the reviewers. Our point wise response to each of the comment is given below in bold letters, below the respective comments.
Major Issues:
- At first, it is difficult to read this paper because of unclear and unappropriated sentences. For readers to understand clearly, many sentences need to be polished with clear and concise structure including appropriate English.
Response: We are sorry for the lack of clarity in certain areas. We have revised the manuscript and taken care of the clear and concise structures as suggested by the reviewer.
- Secondly, this paper mostly describes how much satellite and ground-based measurement agree or not. The authors need more descriptions about why they are different and what science is behind it. For example, you may investigate less consistency between satellite-retrieved and ground-based BC concentrations in JJA compared to DJF and MAM due to the cloud contamination or a deficit of data availability during monsoon.
Response: We thank the reviewer for the valuable suggestion. Adequate descriptions are provided to explain the less consistency between satellite-retrieved and ground-based BC in the revised manuscript, as given below.
Line 318-334: “Based on the above observations it appears that the spatio-temporal distribution of BC as obtained from satellite retrievals show better consistency with the surface measured BC over the Indian region during DJF and MAM. As the rise in temperature caused by increased solar heating during MAM and JJA results in strong thermal convection over the Indian region (especially in the northern part), this leads to dilutions of near-surface aerosol concentrations. Depending upon the geographic position and local meteorological conditions, the strengths of convections vary from one location to the other. As the satellite retrieve BC is 1-km column average BC concentrations, the variation in the vertical distribution of BC may lead to variable associations between satellite-retrieved and surface measured BC concentrations at distinct geographic locations of India. More details on these aspects are discussed in the subsequent sections. Apart from the vertical heterogeneity, the various other factors that may lead to discrepancy in the satellite retrieval of BC include the bias caused by the cloud-screening algorithm, especially during JJA when the cloud cover over the Indian region is extensive. Moreover, CLAUDIA3 is unable to detect optically thin clouds. Lack of accurate detection of cloud shadow also can cause overestimation in the retrieve values of aerosol parameters from CAI-2 measurements. Since the revisiting time of CAI-2 is long (6 days), the minimum reflection criterion based on the consideration of 2 months data (one month prior and after the measurement days) can lead to large uncertainty in cloud-shadow detection, hence the accurate estimation of minimum surface reflectance. Subsequently, these errors can propagate and add uncertainty in the accurate estimation of aerosol parameters from CAI-2 measurements.”
Minor Issues and specific comments:
Page1 L11: Is the acronym of ARFINET correctly located and explained?
Aerosol Radiation Forcing over India NETwork (ARFINET)Response: Yes, Aerosol Radiation Forcing over India NETwork (ARFINET) of aerosols observatories is clearly mentioned.
P1 L12: revealed -> reveals
Response: Complied with
P1 L17: that of other in-situ -> those of other in-situ
Response: Complied with
P1 L18: satellite retrieval shows -> satellite retrievals show
Response: Complied with
P1 L33: However, the very challenging task is to accurately retrieve -> However, it is challenging is to accurately retrieve
Response: Complied with
P2 L42: have not addressed so far -> have not been addressed so far
Response: In the revised manuscript, the sentence is modified as
“… the retrieval of BC from satellite-based radiation measurement is very limited.”
P2 L50: 10-sepctral bands -> 10-spectral bands
Response: Complied with
P2 L50-51: use full word and abbreviation of UV, VIS, and NIR
Response: Complied with
P2 L59: to develop periodic and accurate estimates of aerosol radiative forcing over India and assess their impacts on regional and global climate, taking into account their heterogeneous properties in space, time and spectral domains -> to develop periodic and accurate estimates of aerosol radiative forcing over India, assess their impacts on regional and global climate, and take into account their heterogeneous properties in space, time and spectral domains
Response: Complied with. The sentence is modified as:
“In the ARFINET, the main objective of the measurements of various aerosol parameters (e.g., columnar aerosol optical depth, BC mass concentrations, etc.) is to characterize their heterogeneous properties in space, time and spectral domains, develop periodic and accurate estimates of aerosol radiative forcing over India, and assess their impacts on regional and global climate.”
P2 L71: remove etc.
Response: Complied with
P3 L95: the surface albedo is derived by performing a correction removing the influence of atmospheric molecular scattering (Rayleigh scattering) -> the surface albedo is derived by removing the influence of atmospheric molecular scattering (Rayleigh scattering)
Response: This section is now modified in the revised manuscript as
“After cloud and cloud shadow correction, the influence of atmospheric molecular scattering (Rayleigh scattering) is corrected from the minimum reflectance data.”
P3 L98: single scattering and multiple scattering -> single- and multiple-scattering
Response: Complied with
P3 L101: inversion algorithm developed by Hashimoto and Nakajima (2017) is used. -> inversion algorithm (Hashimoto and Nakajima, 2017) is used.
Response: Complied with
P3 L108: This sentence should be clearly stated.
aerosol light absorption (or single scattering albedo - SSA) -> aerosol light scattering (or single scattering albedo - SSA)
If you want to describe aerosol light absorption, you can use co-albedo or 1-SSA instead of SSA.
Response: Complied with. We have maintained the consistency with aerosol light scattering (or single scattering albedo - SSA).
P4 L131: Does “several sensitivity studies” means studies you performed? If not, add references.
Response: Sorry for the confusing statement. The reference “Hashimoto and Nakajima, 2017” is included.
P4 L148: Detail about the aethalometer uncertainty and correction of raw-data is available in Gogoi et al., (2017). The overall uncertainty in BC mass measured by the Aethalometer is estimated at about 10%. -> The overall uncertainty in BC mass measured by the aethalometer is estimated at about 10% and more details are available in Gogoi et al., (2017).
Response: Complied with
P5 L161: MAE = 10 m2g-1 is used. -> MAE = 10 m2g-1 is assumed (add references).
Response: Complied with. The following citation is included:
Kondo, Y., Sahu, L., Kuwata, M., Miyazaki, Y., Takegawa, N., Moteki, N., Imaru, J., Han, S., Nakayama, T., Oanh, N. T. K., Hu, M., Kim, Y. J., and Kita, K.: Stabilization of the Mass Absorption Cross Section of Black Carbon for Filter-Based Absorption Photometry by the use of a Heated Inlet, Aerosol Science and Technology, 43, 741-756, https://doi.org/10.1080/02786820902889879, 2009.
P5 L167: You mentioned, “As the ambient BC in the atmosphere is mostly aged in nature”
It is a vague sentence for the reason of “a value of MAE = 10 m2g-1 is used” since BC is not aged in nature if it is just released from biomass burning.
Response: We thank the reviewer for pointing out the vague statement. We have revised the sentence as
“For estimating σabs for the columnar content of BC, a constant value of mass absorption efficiency, MAE = 10 m2 g-1 is assumed (Kondo et al., 2009).”
P5 L175: winter, pre-monsoon, and monsoon respectively. -> winter, pre-monsoon, and monsoon, respectively.
Response: Complied with
P5 L177: You should add one more figure or add text into Fig. 1 to indicate each region like HIM, IGP, NEI, NWI and so on. Although it is written in the supplement, this straightforward figure would help readers to understand your figure better.
Response: Complied with the suggestion. We have included the following figure in the revised manuscript, clearly showing the regions of HIM, IGP, NEI, NWI, CI, PI and IL.
Figure-1: The network of aerosols observatories over India, distributed in the Indo-Gangetic Plains (IGP); North-eastern India (NEI); North-western India (NWI); Himalayan, sub-Himalayan and foothills regions (HIM), Central India (CI), Peninsular India (PI) and Island Locations (IL). More details about the ground-based observational locations in the ARFINET are provided in Supplementary Table-T1.
P5 L195-197: You mentioned satellite retrievals and surface observations of BC are more consistent in DJF and MAM than JJA. Is less consistency in JJA caused by cloud contamination during monsoon? Or how many data used for this analysis for each season? Are the number of data used during monsoon fewer compared to other seasons?
Response: We are thankful to the reviewer for the valuable suggestion. The discussions regarding the associations/ discrepancy between satellite-retrieved and surface measured BC are modified in the revised manuscript. The following section is also added to highlight the possible causes of higher discrepancy in the spatio-temporal distribution of BC over the Indian region during JJA as compared to that during DJF and MAM.
Line 318-334: “Based on the above observations it appears that the spatio-temporal distribution of BC as obtained from satellite retrievals show better consistency with the surface measured BC over the Indian region during DJF and MAM. As the rise in temperature caused by increased solar heating during MAM and JJA results in strong thermal convection over the Indian region (especially in the northern part), this leads to dilutions of near-surface aerosol concentrations. Depending upon the geographic position and local meteorological conditions, the strengths of convections vary from one location to the other. As the satellite retrieve BC is 1-km column average BC concentrations, the variation in the vertical distribution of BC may lead to variable associations between satellite-retrieved and surface measured BC concentrations at distinct geographic locations of India. More details on these aspects are discussed in the subsequent sections. Apart from the vertical heterogeneity, the various other factors that may lead to discrepancy in the satellite retrieval of BC include the bias caused by the cloud-screening algorithm, especially during JJA when the cloud cover over the Indian region is extensive. Moreover, CLAUDIA3 is unable to detect optically thin clouds. Lack of accurate detection of cloud shadow also can cause overestimation in the retrieve values of aerosol parameters from CAI-2 measurements. Since the revisiting time of CAI-2 is long (6 days), the minimum reflection criterion based on the consideration of 2 months data (one month prior and after the measurement days) can lead to large uncertainty in cloud-shadow detection, hence the accurate estimation of minimum surface reflectance. Subsequently, these errors can propagate and add uncertainty in the accurate estimation of aerosol parameters from CAI-2 measurements.”
P6 L204: better associations -> better agreement
Response: Complied with.
P6 L205: the association between the two data sets -> the correlation between the two data sets
Response: Complied with.
P6 L205: Thus, despite satellite retrievals during winter and pre-monsoon months showing the regional hotspots of BC over India fairly well, there appears to be a lack of consistent associations between the two datasets in winter at some of the ARFINET observational sites. -> Thus, satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during wither and pre-monsoon months. However, there are a lack of consistency between the two datasets in winter.
Response: Complied with. We have modified the sentence as:
“This indicates that even though satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during winter and pre-monsoon months, there is a lack of consistency between the two datasets in winter at some of the other ARFINET observational sites.”
P6 L208: The above observations point to -> The discrepancies between satellite retrievals and ground-based observations can be attributed to
Response: Complied with.
P6 L215: Do not use “Despite this” and use specific words
Response: We have removed the term “Despite this”
P6 L215: the satellite retrievals differ from surface measured BC -> the satellite-retrieved BC differ from surface-measured BC
Response: Complied with.
P6 L228: we will now examine simultaneous day-to-day values -> we examine simultaneous day-to-day concentrations
Response: Complied with.
P6 L223-L225: This sentence needs to be polished.
Response: Complied with. The sentence is modified as:
Lines 359-365: “In general, the surface measurements of BC concentrations over the entire Indian region show a gradual decline from its highest values in DJF (2.54 ± 0.11 μg m-3) through MAM (2.06 ± 0.47) to its lowest value in JJA (1.11 ± 0.17 μg m-3). Similar to this, the 1-km column average satellite retrieved BC also show highest BC concentrations over the collocated locations of India during DJF and their gradual decline in MAM. However, the satellite retrieved BC are found to be higher in JJA than in MAM, as opposed to the pattern seen in the case of surface measured BC. These observations hint again the discrepancy between satellite retrievals and surface measured BC in JJA, while their absolute magnitudes and regional distributions are nearly consistent during DJF and MAM in most locations.”
P7 L239: play important role -> play an important role
Response: Complied with.
P7 L242: Add references of ERA5
Response: Complied with. The following reference is included.
“Hersbach H., Bell, B., Berrisford P. et al.: The ERA5 global reanalysis. Quarterly Journal of Royal Meteorological Society, 146, 1999–2049, https://doi.org/10.1002/qj.3803, 2020.”
P7 L246: in all three periods of DJF, MAM and JJA -> during all periods
Response: Complied with.
P7 L249: It has also been observed that absolute differences between the two data sets -> Absolute differences between the two data sets
Response: Complied with.
P7 L249: peninsular Indian locations -> PI
Response: Complied with.
P7 L251: It is further evident from the figure -> It is further evident from Fig. 6
Response: Complied with.
P7 L259: Provide the reason of the sentence “Especially, the association between the two data sets significantly improves in JJA.”
Response: We sincerely thank the reviewer for suggesting a very important point to include in the discussions. Accordingly we have elaborated the discussion as given blow.
Line 389-405: “During winter, even though the abundance of BC is confined near to the surface due to shallow PBL condition, the noon time PBL is much extended (close to or beyond 1-km) over most of the Indian locations (the spatio-temporal variability in PBL height is shown in supplementary Fig. S6). Thus, BCSUR-N follows the same general trend as the BCSUR, indicating that noon-time surface measured BC concentrations during winter are similar to the 1-km column average BC. During MAM, the locations with PBL heights extended above 1-km are found to show good association of BCSAT with BCSUR-N than that of BCSAT with BCSUR. In JJA, the height of PBL is found to be highly region specific. At some of the locations, the PBL is much above 1-km (e.g., CHN and KDP), while some other locations show the opposite pattern (i.e., TVM, PBL height below 1 km). The locations with PBL heights below 1-km are found to show lower absolute difference between BCSAT and BCSUR-N than that between BCSAT and BCSUR. However, it is also to be noted that the simultaneous data of satellite-retrieved and surface measured BC are less in JJA as compared to DJF and MAM. Overall, it is observed that, in most of the locations, the absolute difference between BCSAT and BCSUR-N is lower than that between BCSAT and BCSUR. This leads to better correlation between BCSAT and BCSUR-N, especially during JJA where the correlation between BCSAT and BCSUR-N is much better (R ~ 0.61) than that between BCSAT and BCSUR (R ~ 0.38).”
Figure 7: Seasonal mean values of satellite-retrieved (BCSAT) and surface-measured (BCSUR and BCSUR-N) BC concentrations at different ARFINET sites (shown with respect to their latitudes) of India. The absolute difference between BCSAT and BCSUR-N are also shown. The top panel shows the seasonal values of BCSAT, BCSUR, BCSUR-N and |(BCSAT - BCSUR-N)| around each of the observational sites during December-January-February (DJF). Same parameters are shown in the middle panel for March-April-May (MAM) and bottom panel for June-July-August (JJA). The letters in the histograms represent the names of individual stations (details in supplementary Table-T1). The simultaneous data available for inter-comparison are highest in DJF (17-stations) and least in JJA (9-stations).
P7 L261: add references
Response: The following references are included in support of the seasonal changes in the incoming ground reaching solar radiation in the northern part of India.
Soni, V.K., Pandithurai, G., Pai, D.S.: Evaluation of long-term changes of solar radiation in India. International Jouirnal of Climatology, 32 (4), 540–551, https://doi.org/10.1002/joc.2294, 2012.
Subba, T., Gogoi, M. M., Moorthy, K. K., Bhuyan, P. K., Pathak, B., Guha, A., Srivastava, M. K., Vyas, B. M., Singh, K., Krishnan, J., Lakshmikumar, T. V. S., Babu, S. S.: Aerosol Radiative Effects over India from Direct Radiation Measurements and Model Estimates, Atmospheric Research, 276, 106254, https://doi.org/10.1016/j.atmosres.2022.106254, 2022.
P7 L267: show -> shows
Response: Complied with.
P7 L276: more wet soils -> wetter soils
Response: Complied with.
P8 L288: The forgone observation -> The prior observation
Response: Complied with.
P8 L298: Based on in-situ vertical profiling of aerosol scattering and absorption properties on a research aircraft, Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period. -> Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period using aircraft measurements.
Response: Complied with.
P8 L305: Over the oceanic regions, the values of SSA are, in general, high -> Over the oceanic regions, the values of SSA are generally high
Response: Complied with.
P8 L310-311: Mar -> Mar.
Jun -> Jun.
You may use abbreviations of the months consistently: decide full name or abbreviations of the months.
Response: Complied with. The consistency is maintained.
P8 L315: during Mar/Apr/May -> during March to May
Response: Complied with
P9 L319: Figs. 8, 9 and 10 -> Figs. 8, 9 and 10, respectively
Response: Complied with.
P9 L322: day time FRP -> day-time FRP
Response: Complied with.
P9 L343: add references after “Several studies”
Response: Complied with. The following references are included in the revised manuscript.
Dixon, R. K., Krankina, O. N.: Forest fires in Russia: carbon dioxide emissions to the atmosphere, Canadian Journal of Forest Research, 23, 700-705, 1993.
Leskinen, P., Lindner, M., Verkerk, P.J., Nabuurs, G.J., Van Brusselen, J., Kulikova, E., Hassegawa, M. and Lerink, B. (eds.).: Russian forests and climate change. What Science Can Tell Us 11. European Forest Institute, 2020.
P9 L345: evaluate and validate -> evaluate
Response: Complied with.
P9 L349: remove unnecessary sentence (The main findings are as follows:)
Response: Complied with.
P9 L351: do not use “fairly”. It sounds informal.
Response: Complied with.
P9 L354: for > 60% of the observations (for all the locations considered in this study) the absolute difference between the two data sets is < 2 μgm-3. -> the absolute difference between the two data sets is less than 2 μgm-3 for over 60% of the locations in this study.
Response: Complied with.
P10 L365: during times of biomass burning -> during the biomass burning season
Response: Complied with.
P22 L654: Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Response: Complied with. The figure caption is modified as
“Figure 7: Seasonal mean values of satellite-retrieved (BCSAT) and surface-measured BC (BCSUR and BCSUR-N) BC concentrations at different ARFINET sites (shown with respect to their latitudes) of India. The absolute difference between BCSAT and BCSUR-N at different locations are also shown. The top panel shows the seasonal values of BCSAT, BCSUR, BCSUR-N and |(BCSAT - BCSUR-N)| around each of the observational sites during December-January-February (DJF). Same parameters are shown in the middle panel for March-April-May (MAM) and in the bottom panel for June-July-August (JJA). The letters in the histograms represent the names of individual stations (details in supplementary Table-T1). The simultaneous data available for inter-comparison are highest in DJF (17-stations) and least in JJA (9-stations).”
P24-26: Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Response: Complied with.
-END-
-
AC1: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
-
RC2: 'Comment on acp-2022-555', Anonymous Referee #2, 09 Oct 2022
General opinion:
The authors proposed a algorithm to retrieve black carbon from GOSAT-2 CAI-2. The authors also incorporated evaluation and validation of the satellite retrievals across a network of aerosol observatories (ARFINET) over India and the findings are extended to comprehend the global BC features. Such model is in highly demand if it is proven to work effectively. However, I am more concerned about the validity of the algorithm itself because the authors did not provide enough details on the methods, equations, and uncertainties. This may prevent the readers from understanding their work. Some descriptions and discussions are sometime puzzling, and there are thus much more revisions need to be made carefully by the authors.Major Comments
1. Inadequate innovation of the MS based on the claim of Line 40-41 “ the direct retrieval of BC from satellite based radiation measurement have not addressed so far.” This is really not true. Below are some articles published in recent years, proposed similar algorithm in other countries.Bao, F., Cheng, T., Li, Y., Gu, X., Guo, H., Wu, Y., Wang, Y., & Gao, J. (2019). Retrieval of black carbon aerosol surface concentration using satellite remote sensing observations. Remote Sensing of Environment, 226, 93-108
Bao, F., Li, Y., Cheng, T., Gao, J., & Yuan, S. (2020). Estimating the Columnar Concentrations of Black Carbon Aerosols in China Using MODIS Products. Environmental Science & Technology, 54, 11025-11036
Li, L., Che, H., Derimian, Y., Dubovik, O., Schuster, G.L., Chen, C., Li, Q., Wang, Y., Guo, B., & Zhang, X. (2020). Retrievals of fine mode light-absorbing carbonaceous aerosols from POLDER/PARASOL observations over East and South Asia. Remote Sensing of Environment, 247, 111913
- Comprehensive literature review and rigorous discussionis required in the introduction. And some details about the satellite sensor and data should be removed from the introduction to make the introduction concise.
- the authors should give a clear description of their algorithm. In section 2.1 the authors seem to spend a lot of space to review some other scholar's algorithms, which is confusing for some cross-field In addition, did the authors use official unpublished products? The authors mentionedthat the algorithm cite an under-preparation version of CAI-2 L2 aerosol retrieval ATBD (L117). If an official unpublished product is used, then a detailed description of the algorithm is needed. If the MS focuses on the improvements to existing algorithms, the basis, formulas, and the updates in this paper should also be emphasized. These descriptions must be detailed and not misleading.
- I have a few doubts about the algorithm itself. Does the minimum reflectance strategy of surface reflectance correction in this MS consistent with that described in lines 81-85? What is the role of NDVI in this decision? In addition, if this strategy is used, it should be explained in detail in the flowchart (Fig. S1), as using ‘minimum’ may lead to misunderstandings.
- In addition, the authors mentioned an internal mixing model to describe the proportion of BC in the aerosol. But it is not clear which internal mixing model are used. It is necessary to state and state the formula. How is the change in absorption of BC at different wavelengths considered? How is the absorption of other non-BC particles considered? The author defines: fbc=Vsoot/Vfine. It is also necessary to discuss the reasonableness of ignoring coarse particle aerosols. As far as I know, the spectral absorption of mixing aerosol is greatly influenced by some coarse particle (like DUST), which also show significant absorption in the near UV spectrum. These seemingly unreasonable assumptions can also have a very huge impact on later application studies.
- In the validation section I note that the authors assume a uniformly columnar distributed BC, using a simple equation for the columnar concentration and near-surface conversion, but the ideal conditions are quite different from the actual observations. I would like to see a more reasonable solution. If not, I would like to see more validation, such as SSA, BCAOD, which makes the the accuracy of the product more intuitive.
- In the comparison of Satellite retrievals vs climatological surface BC concentrations, does Satellite retrievals convert to near-ground magnitudes? If so, we need to move equation 3 here, but if not, the metrics RMSE in the validation needs to be removed, because they are two parameters with different magnitudes.
- The uncertainty analyses is missing in the MS. i.e., the uncertainty of the algorithm itself; The uncertainty of interpolation; The uncertainty of internal mixing; The uncertainty of switching columnar concentration to near ground; The uncertainty of ignoring coarse particle aerosols. The uncertainty analyses is very important for those who use the product in the future.
- How are SSA and FRP calculated in the section2-3.3? It is not reasonable to extrapolate Indianretrievals to global FRP without extended validation and uncertainty analyses, and it may be more convincing to state Indian only.
Minor Comments
- Some paragraphs are too long, need to split and simplifie
- The data in T2 and S2 for January are not matched, need double-check
Citation: https://doi.org/10.5194/acp-2022-555-RC2 -
AC2: 'Reply on RC2', Mukunda M Gogoi, 17 Jan 2023
We appreciate the summary evaluation of the reviewer. We have complied with the observations and revised the manuscript incorporating valuable comments by the reviewers. In the revised manuscript, we have given more emphasis on the algorithm description, including the various steps involved in the retrieval process. The validation and uncertainties involved in this retrieval method is also elaborated. Our point wise response to each of the comment is given in bold letters, below the respective comments.
Status: closed
-
RC1: 'Comment on acp-2022-555', Anonymous Referee #1, 08 Oct 2022
Reviewers' comments:
The authors use extensive measurements to retrieve BC mass concentrations and compare them with satellite-retrieved BC mass concentrations. This paper contributes to better understanding of spatial and temporal distributions of BC concentrations over the Indian regions. I believe the paper should be considered for publication only after addressing the concerned expressed below.
Major Issues:
1.
At first, it is difficult to read this paper because of unclear and unappropriated sentences.
For readers to understand clearly, many sentences need to be polished with clear and concise structure including appropriate English.
2.
Secondly, this paper mostly describes how much satellite and ground-based measurement agree or not. The authors need more descriptions about why they are different and what science is behind it. For example, you may investigate less consistency between satellite-retrieved and ground-based BC concentrations in JJA compared to DJF and MAM due to the cloud contamination or a deficit of data availability during monsoon.
Minor Issues and specific comments:
Page1 L11: Is the acronym of ARFINET correctly located and explained?
Aerosol Radiation Forcing over India NETwork (ARFINET)
P1 L12: revealed -> reveals
P1 L17: that of other in-situ -> those of other in-situ
P1 L18: satellite retrieval shows -> satellite retrievals show
P1 L33: However, the very challenging task is to accurately retrieve
-> However, it is challenging is to accurately retrieve
P2 L42: have not addressed so far -> have not been addressed so far
P2 L50: 10-sepctral bands -> 10-spectral bands
P2 L50-51: use full word and abbreviation of UV, VIS, and NIR
P2 L59: to develop periodic and accurate estimates of aerosol radiative forcing over India and assess their impacts on regional and global climate, taking into account their heterogeneous properties in space, time and spectral domains
->
to develop periodic and accurate estimates of aerosol radiative forcing over India, assess their impacts on regional and global climate, and take into account their heterogeneous properties in space, time and spectral domains
P2 L71: remove etc.
P3 L95: the surface albedo is derived by performing a correction removing the influence of atmospheric 95 molecular scattering (Rayleigh scattering) ->
the surface albedo is derived by removing the influence of atmospheric 95 molecular scattering (Rayleigh scattering)
P3 L98: single scattering and multiple scattering -> single- and multiple-scattering
P3 L101: inversion algorithm developed by Hashimoto and Nakajima (2017) is used.
-> inversion algorithm (Hashimoto and Nakajima, 2017) is used.
P3 L108: This sentence should be clearly stated.
aerosol light absorption (or single scattering albedo - SSA) -> aerosol light scattering (or single scattering albedo - SSA)
If you want to describe aerosol light absorption, you can use co-albedo or 1-SSA instead of SSA.
P4 L131: Does “several sensitivity studies” means studies you performed? If not, add references.
P4 L148: Detail about the aethalometer uncertainty and correction of raw-data is available in Gogoi et al., (2017). The overall uncertainty in BC mass measured by the Aethalometer is estimated at about 10%.
-> The overall uncertainty in BC mass measured by the aethalometer is estimated at about 10% and more details are available in Gogoi et al., (2017).
P5 L161: MAE = 10 m2g-1 is used. -> MAE = 10 m2g-1 is assumed (add references).
P5 L167:
You mentioned, “As the ambient BC in the atmosphere is mostly aged in nature”
It is a vague sentence for the reason of “a value of MAE = 10 m2g-1 is used” since BC is not aged in nature if it is just released from biomass burning.
P5 L175: winter, pre-monsoon, and monsoon respectively. -> winter, pre-monsoon, and monsoon, respectively.
P5 L177:
You should add one more figure or add text into Fig. 1 to indicate each region like HIM, IGP, NEI, NWI and so on. Although it is written in the supplement, this straightforward figure would help readers to understand your figure better.
P5 L195-197:
You mentioned satellite retrievals and surface observations of BC are more consistent in DJF and MAM than JJA. Is less consistency in JJA caused by cloud contamination during monsoon? Or how many data used for this analysis for each season? Are the number of data used during monsoon fewer compared to other seasons?
P6 L204:
better associations -> better agreement
P6 L205:
the association between the two data sets -> the correlation between the two data sets
P6 L205:
Thus, despite satellite retrievals during winter and pre-monsoon months showing the regional hotspots of BC over India fairly well, there appears to be a lack of consistent associations between the two datasets in winter at some of the ARFINET observational sites.
->
Thus, satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during wither and pre-monsoon months. However, there are a lack of consistency between the two datasets in winter.
P6 L208: The above observations point to
-> The discrepancies between satellite retrievals and ground-based observations can be attributed to
P6 L215: Do not use “Despite this” and use specific words
P6 L215: the satellite retrievals differ from surface measured BC -> the satellite-retrieved BC differ from surface-measured BC
P6 L228:
we will now examine simultaneous day-to-day values -> we examine simultaneous day-to-day concentrations
P6 L223-L225:
This sentence needs to be polished.
P7 L239: play important role -> play an important role
P7 L242: Add references of ERA5
P7 L246:
in all three periods of DJF, MAM and JJA -> during all periods
P7 L249:
It has also been observed that absolute differences between the two data sets ->
Absolute differences between the two data sets
P7 L249: peninsular Indian locations -> PI
P7 L251: It is further evident from the figure -> It is further evident from Fig. 6
P7 L259:
Provide the reason of the sentence “Especially, the association between the two data sets significantly improves in JJA.”
P7 L261: add references
P7 L267: show -> shows
P7 L276: more wet soils -> wetter soils
P8 L288: The forgone observation -> The prior observation
P8 L298: Based on in-situ vertical profiling of aerosol scattering and absorption properties on a research aircraft, Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period.
->
Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period using aircraft measurements.
P8 L305:
Over the oceanic regions, the values of SSA are, in general, high
-> Over the oceanic regions, the values of SSA are generally high
P8 L310-311: Mar -> Mar.
Jun -> Jun.
You may use abbreviations of the months consistently: decide full name or abbreviations of the months.
P8 L315:
duringMar/Apr/May -> during March to May
P9 L319: Figs. 8, 9 and 10 -> Figs. 8, 9 and 10, respectively
P9 L322: day time FRP -> day-time FRP
P9 L343: add references after “Several studies”
P9 L345: evaluate and validate -> evaluate
P9 L349: remove unnecessary sentence (The main findings are as follows:)
P9 L351: do not use “fairly”. It sounds informal.
P9 L354:
for > 60% of the observations (for all the locations considered in this study) the absolute difference between the two data sets is < 2 μgm-3.
->
the absolute difference between the two data sets is less than 2 μgm-3 for over 60% of the locations in this study.
P10 L365:
during times of biomass burning -> during the biomass burning season
P22 L654:
Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
P24-26:
Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Citation: https://doi.org/10.5194/acp-2022-555-RC1 -
AC1: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
We appreciate the summary evaluation of the reviewer and agree to the observations. Following the valuable comments and fruitful suggestions for improving the quality of the manuscript, we have revised it incorporating the review comments of all the reviewers. Our point wise response to each of the comment is given in bold letters, below the respective comments.
-
AC3: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
The authors use extensive measurements to retrieve BC mass concentrations and compare them with satellite-retrieved BC mass concentrations. This paper contributes to better understanding of spatial and temporal distributions of BC concentrations over the Indian regions. I believe the paper should be considered for publication only after addressing the concerned expressed below.
We appreciate the summary evaluation of the reviewer and agree to the observations. Following the valuable comments and fruitful suggestions for improving the quality of the manuscript, we have revised it incorporating the review comments of all the reviewers. Our point wise response to each of the comment is given below in bold letters, below the respective comments.
Major Issues:
- At first, it is difficult to read this paper because of unclear and unappropriated sentences. For readers to understand clearly, many sentences need to be polished with clear and concise structure including appropriate English.
Response: We are sorry for the lack of clarity in certain areas. We have revised the manuscript and taken care of the clear and concise structures as suggested by the reviewer.
- Secondly, this paper mostly describes how much satellite and ground-based measurement agree or not. The authors need more descriptions about why they are different and what science is behind it. For example, you may investigate less consistency between satellite-retrieved and ground-based BC concentrations in JJA compared to DJF and MAM due to the cloud contamination or a deficit of data availability during monsoon.
Response: We thank the reviewer for the valuable suggestion. Adequate descriptions are provided to explain the less consistency between satellite-retrieved and ground-based BC in the revised manuscript, as given below.
Line 318-334: “Based on the above observations it appears that the spatio-temporal distribution of BC as obtained from satellite retrievals show better consistency with the surface measured BC over the Indian region during DJF and MAM. As the rise in temperature caused by increased solar heating during MAM and JJA results in strong thermal convection over the Indian region (especially in the northern part), this leads to dilutions of near-surface aerosol concentrations. Depending upon the geographic position and local meteorological conditions, the strengths of convections vary from one location to the other. As the satellite retrieve BC is 1-km column average BC concentrations, the variation in the vertical distribution of BC may lead to variable associations between satellite-retrieved and surface measured BC concentrations at distinct geographic locations of India. More details on these aspects are discussed in the subsequent sections. Apart from the vertical heterogeneity, the various other factors that may lead to discrepancy in the satellite retrieval of BC include the bias caused by the cloud-screening algorithm, especially during JJA when the cloud cover over the Indian region is extensive. Moreover, CLAUDIA3 is unable to detect optically thin clouds. Lack of accurate detection of cloud shadow also can cause overestimation in the retrieve values of aerosol parameters from CAI-2 measurements. Since the revisiting time of CAI-2 is long (6 days), the minimum reflection criterion based on the consideration of 2 months data (one month prior and after the measurement days) can lead to large uncertainty in cloud-shadow detection, hence the accurate estimation of minimum surface reflectance. Subsequently, these errors can propagate and add uncertainty in the accurate estimation of aerosol parameters from CAI-2 measurements.”
Minor Issues and specific comments:
Page1 L11: Is the acronym of ARFINET correctly located and explained?
Aerosol Radiation Forcing over India NETwork (ARFINET)Response: Yes, Aerosol Radiation Forcing over India NETwork (ARFINET) of aerosols observatories is clearly mentioned.
P1 L12: revealed -> reveals
Response: Complied with
P1 L17: that of other in-situ -> those of other in-situ
Response: Complied with
P1 L18: satellite retrieval shows -> satellite retrievals show
Response: Complied with
P1 L33: However, the very challenging task is to accurately retrieve -> However, it is challenging is to accurately retrieve
Response: Complied with
P2 L42: have not addressed so far -> have not been addressed so far
Response: In the revised manuscript, the sentence is modified as
“… the retrieval of BC from satellite-based radiation measurement is very limited.”
P2 L50: 10-sepctral bands -> 10-spectral bands
Response: Complied with
P2 L50-51: use full word and abbreviation of UV, VIS, and NIR
Response: Complied with
P2 L59: to develop periodic and accurate estimates of aerosol radiative forcing over India and assess their impacts on regional and global climate, taking into account their heterogeneous properties in space, time and spectral domains -> to develop periodic and accurate estimates of aerosol radiative forcing over India, assess their impacts on regional and global climate, and take into account their heterogeneous properties in space, time and spectral domains
Response: Complied with. The sentence is modified as:
“In the ARFINET, the main objective of the measurements of various aerosol parameters (e.g., columnar aerosol optical depth, BC mass concentrations, etc.) is to characterize their heterogeneous properties in space, time and spectral domains, develop periodic and accurate estimates of aerosol radiative forcing over India, and assess their impacts on regional and global climate.”
P2 L71: remove etc.
Response: Complied with
P3 L95: the surface albedo is derived by performing a correction removing the influence of atmospheric molecular scattering (Rayleigh scattering) -> the surface albedo is derived by removing the influence of atmospheric molecular scattering (Rayleigh scattering)
Response: This section is now modified in the revised manuscript as
“After cloud and cloud shadow correction, the influence of atmospheric molecular scattering (Rayleigh scattering) is corrected from the minimum reflectance data.”
P3 L98: single scattering and multiple scattering -> single- and multiple-scattering
Response: Complied with
P3 L101: inversion algorithm developed by Hashimoto and Nakajima (2017) is used. -> inversion algorithm (Hashimoto and Nakajima, 2017) is used.
Response: Complied with
P3 L108: This sentence should be clearly stated.
aerosol light absorption (or single scattering albedo - SSA) -> aerosol light scattering (or single scattering albedo - SSA)
If you want to describe aerosol light absorption, you can use co-albedo or 1-SSA instead of SSA.
Response: Complied with. We have maintained the consistency with aerosol light scattering (or single scattering albedo - SSA).
P4 L131: Does “several sensitivity studies” means studies you performed? If not, add references.
Response: Sorry for the confusing statement. The reference “Hashimoto and Nakajima, 2017” is included.
P4 L148: Detail about the aethalometer uncertainty and correction of raw-data is available in Gogoi et al., (2017). The overall uncertainty in BC mass measured by the Aethalometer is estimated at about 10%. -> The overall uncertainty in BC mass measured by the aethalometer is estimated at about 10% and more details are available in Gogoi et al., (2017).
Response: Complied with
P5 L161: MAE = 10 m2g-1 is used. -> MAE = 10 m2g-1 is assumed (add references).
Response: Complied with. The following citation is included:
Kondo, Y., Sahu, L., Kuwata, M., Miyazaki, Y., Takegawa, N., Moteki, N., Imaru, J., Han, S., Nakayama, T., Oanh, N. T. K., Hu, M., Kim, Y. J., and Kita, K.: Stabilization of the Mass Absorption Cross Section of Black Carbon for Filter-Based Absorption Photometry by the use of a Heated Inlet, Aerosol Science and Technology, 43, 741-756, https://doi.org/10.1080/02786820902889879, 2009.
P5 L167: You mentioned, “As the ambient BC in the atmosphere is mostly aged in nature”
It is a vague sentence for the reason of “a value of MAE = 10 m2g-1 is used” since BC is not aged in nature if it is just released from biomass burning.
Response: We thank the reviewer for pointing out the vague statement. We have revised the sentence as
“For estimating σabs for the columnar content of BC, a constant value of mass absorption efficiency, MAE = 10 m2 g-1 is assumed (Kondo et al., 2009).”
P5 L175: winter, pre-monsoon, and monsoon respectively. -> winter, pre-monsoon, and monsoon, respectively.
Response: Complied with
P5 L177: You should add one more figure or add text into Fig. 1 to indicate each region like HIM, IGP, NEI, NWI and so on. Although it is written in the supplement, this straightforward figure would help readers to understand your figure better.
Response: Complied with the suggestion. We have included the following figure in the revised manuscript, clearly showing the regions of HIM, IGP, NEI, NWI, CI, PI and IL.
Figure-1: The network of aerosols observatories over India, distributed in the Indo-Gangetic Plains (IGP); North-eastern India (NEI); North-western India (NWI); Himalayan, sub-Himalayan and foothills regions (HIM), Central India (CI), Peninsular India (PI) and Island Locations (IL). More details about the ground-based observational locations in the ARFINET are provided in Supplementary Table-T1.
P5 L195-197: You mentioned satellite retrievals and surface observations of BC are more consistent in DJF and MAM than JJA. Is less consistency in JJA caused by cloud contamination during monsoon? Or how many data used for this analysis for each season? Are the number of data used during monsoon fewer compared to other seasons?
Response: We are thankful to the reviewer for the valuable suggestion. The discussions regarding the associations/ discrepancy between satellite-retrieved and surface measured BC are modified in the revised manuscript. The following section is also added to highlight the possible causes of higher discrepancy in the spatio-temporal distribution of BC over the Indian region during JJA as compared to that during DJF and MAM.
Line 318-334: “Based on the above observations it appears that the spatio-temporal distribution of BC as obtained from satellite retrievals show better consistency with the surface measured BC over the Indian region during DJF and MAM. As the rise in temperature caused by increased solar heating during MAM and JJA results in strong thermal convection over the Indian region (especially in the northern part), this leads to dilutions of near-surface aerosol concentrations. Depending upon the geographic position and local meteorological conditions, the strengths of convections vary from one location to the other. As the satellite retrieve BC is 1-km column average BC concentrations, the variation in the vertical distribution of BC may lead to variable associations between satellite-retrieved and surface measured BC concentrations at distinct geographic locations of India. More details on these aspects are discussed in the subsequent sections. Apart from the vertical heterogeneity, the various other factors that may lead to discrepancy in the satellite retrieval of BC include the bias caused by the cloud-screening algorithm, especially during JJA when the cloud cover over the Indian region is extensive. Moreover, CLAUDIA3 is unable to detect optically thin clouds. Lack of accurate detection of cloud shadow also can cause overestimation in the retrieve values of aerosol parameters from CAI-2 measurements. Since the revisiting time of CAI-2 is long (6 days), the minimum reflection criterion based on the consideration of 2 months data (one month prior and after the measurement days) can lead to large uncertainty in cloud-shadow detection, hence the accurate estimation of minimum surface reflectance. Subsequently, these errors can propagate and add uncertainty in the accurate estimation of aerosol parameters from CAI-2 measurements.”
P6 L204: better associations -> better agreement
Response: Complied with.
P6 L205: the association between the two data sets -> the correlation between the two data sets
Response: Complied with.
P6 L205: Thus, despite satellite retrievals during winter and pre-monsoon months showing the regional hotspots of BC over India fairly well, there appears to be a lack of consistent associations between the two datasets in winter at some of the ARFINET observational sites. -> Thus, satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during wither and pre-monsoon months. However, there are a lack of consistency between the two datasets in winter.
Response: Complied with. We have modified the sentence as:
“This indicates that even though satellite retrievals and surface observations show good agreement at the regional hotspots of BC over India during winter and pre-monsoon months, there is a lack of consistency between the two datasets in winter at some of the other ARFINET observational sites.”
P6 L208: The above observations point to -> The discrepancies between satellite retrievals and ground-based observations can be attributed to
Response: Complied with.
P6 L215: Do not use “Despite this” and use specific words
Response: We have removed the term “Despite this”
P6 L215: the satellite retrievals differ from surface measured BC -> the satellite-retrieved BC differ from surface-measured BC
Response: Complied with.
P6 L228: we will now examine simultaneous day-to-day values -> we examine simultaneous day-to-day concentrations
Response: Complied with.
P6 L223-L225: This sentence needs to be polished.
Response: Complied with. The sentence is modified as:
Lines 359-365: “In general, the surface measurements of BC concentrations over the entire Indian region show a gradual decline from its highest values in DJF (2.54 ± 0.11 μg m-3) through MAM (2.06 ± 0.47) to its lowest value in JJA (1.11 ± 0.17 μg m-3). Similar to this, the 1-km column average satellite retrieved BC also show highest BC concentrations over the collocated locations of India during DJF and their gradual decline in MAM. However, the satellite retrieved BC are found to be higher in JJA than in MAM, as opposed to the pattern seen in the case of surface measured BC. These observations hint again the discrepancy between satellite retrievals and surface measured BC in JJA, while their absolute magnitudes and regional distributions are nearly consistent during DJF and MAM in most locations.”
P7 L239: play important role -> play an important role
Response: Complied with.
P7 L242: Add references of ERA5
Response: Complied with. The following reference is included.
“Hersbach H., Bell, B., Berrisford P. et al.: The ERA5 global reanalysis. Quarterly Journal of Royal Meteorological Society, 146, 1999–2049, https://doi.org/10.1002/qj.3803, 2020.”
P7 L246: in all three periods of DJF, MAM and JJA -> during all periods
Response: Complied with.
P7 L249: It has also been observed that absolute differences between the two data sets -> Absolute differences between the two data sets
Response: Complied with.
P7 L249: peninsular Indian locations -> PI
Response: Complied with.
P7 L251: It is further evident from the figure -> It is further evident from Fig. 6
Response: Complied with.
P7 L259: Provide the reason of the sentence “Especially, the association between the two data sets significantly improves in JJA.”
Response: We sincerely thank the reviewer for suggesting a very important point to include in the discussions. Accordingly we have elaborated the discussion as given blow.
Line 389-405: “During winter, even though the abundance of BC is confined near to the surface due to shallow PBL condition, the noon time PBL is much extended (close to or beyond 1-km) over most of the Indian locations (the spatio-temporal variability in PBL height is shown in supplementary Fig. S6). Thus, BCSUR-N follows the same general trend as the BCSUR, indicating that noon-time surface measured BC concentrations during winter are similar to the 1-km column average BC. During MAM, the locations with PBL heights extended above 1-km are found to show good association of BCSAT with BCSUR-N than that of BCSAT with BCSUR. In JJA, the height of PBL is found to be highly region specific. At some of the locations, the PBL is much above 1-km (e.g., CHN and KDP), while some other locations show the opposite pattern (i.e., TVM, PBL height below 1 km). The locations with PBL heights below 1-km are found to show lower absolute difference between BCSAT and BCSUR-N than that between BCSAT and BCSUR. However, it is also to be noted that the simultaneous data of satellite-retrieved and surface measured BC are less in JJA as compared to DJF and MAM. Overall, it is observed that, in most of the locations, the absolute difference between BCSAT and BCSUR-N is lower than that between BCSAT and BCSUR. This leads to better correlation between BCSAT and BCSUR-N, especially during JJA where the correlation between BCSAT and BCSUR-N is much better (R ~ 0.61) than that between BCSAT and BCSUR (R ~ 0.38).”
Figure 7: Seasonal mean values of satellite-retrieved (BCSAT) and surface-measured (BCSUR and BCSUR-N) BC concentrations at different ARFINET sites (shown with respect to their latitudes) of India. The absolute difference between BCSAT and BCSUR-N are also shown. The top panel shows the seasonal values of BCSAT, BCSUR, BCSUR-N and |(BCSAT - BCSUR-N)| around each of the observational sites during December-January-February (DJF). Same parameters are shown in the middle panel for March-April-May (MAM) and bottom panel for June-July-August (JJA). The letters in the histograms represent the names of individual stations (details in supplementary Table-T1). The simultaneous data available for inter-comparison are highest in DJF (17-stations) and least in JJA (9-stations).
P7 L261: add references
Response: The following references are included in support of the seasonal changes in the incoming ground reaching solar radiation in the northern part of India.
Soni, V.K., Pandithurai, G., Pai, D.S.: Evaluation of long-term changes of solar radiation in India. International Jouirnal of Climatology, 32 (4), 540–551, https://doi.org/10.1002/joc.2294, 2012.
Subba, T., Gogoi, M. M., Moorthy, K. K., Bhuyan, P. K., Pathak, B., Guha, A., Srivastava, M. K., Vyas, B. M., Singh, K., Krishnan, J., Lakshmikumar, T. V. S., Babu, S. S.: Aerosol Radiative Effects over India from Direct Radiation Measurements and Model Estimates, Atmospheric Research, 276, 106254, https://doi.org/10.1016/j.atmosres.2022.106254, 2022.
P7 L267: show -> shows
Response: Complied with.
P7 L276: more wet soils -> wetter soils
Response: Complied with.
P8 L288: The forgone observation -> The prior observation
Response: Complied with.
P8 L298: Based on in-situ vertical profiling of aerosol scattering and absorption properties on a research aircraft, Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period. -> Babu et al., (2016) have reported the values of SSA between 0.86 and 0.94 over different West Indian and IGP locations during the pre-monsoon (April-May) period using aircraft measurements.
Response: Complied with.
P8 L305: Over the oceanic regions, the values of SSA are, in general, high -> Over the oceanic regions, the values of SSA are generally high
Response: Complied with.
P8 L310-311: Mar -> Mar.
Jun -> Jun.
You may use abbreviations of the months consistently: decide full name or abbreviations of the months.
Response: Complied with. The consistency is maintained.
P8 L315: during Mar/Apr/May -> during March to May
Response: Complied with
P9 L319: Figs. 8, 9 and 10 -> Figs. 8, 9 and 10, respectively
Response: Complied with.
P9 L322: day time FRP -> day-time FRP
Response: Complied with.
P9 L343: add references after “Several studies”
Response: Complied with. The following references are included in the revised manuscript.
Dixon, R. K., Krankina, O. N.: Forest fires in Russia: carbon dioxide emissions to the atmosphere, Canadian Journal of Forest Research, 23, 700-705, 1993.
Leskinen, P., Lindner, M., Verkerk, P.J., Nabuurs, G.J., Van Brusselen, J., Kulikova, E., Hassegawa, M. and Lerink, B. (eds.).: Russian forests and climate change. What Science Can Tell Us 11. European Forest Institute, 2020.
P9 L345: evaluate and validate -> evaluate
Response: Complied with.
P9 L349: remove unnecessary sentence (The main findings are as follows:)
Response: Complied with.
P9 L351: do not use “fairly”. It sounds informal.
Response: Complied with.
P9 L354: for > 60% of the observations (for all the locations considered in this study) the absolute difference between the two data sets is < 2 μgm-3. -> the absolute difference between the two data sets is less than 2 μgm-3 for over 60% of the locations in this study.
Response: Complied with.
P10 L365: during times of biomass burning -> during the biomass burning season
Response: Complied with.
P22 L654: Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Response: Complied with. The figure caption is modified as
“Figure 7: Seasonal mean values of satellite-retrieved (BCSAT) and surface-measured BC (BCSUR and BCSUR-N) BC concentrations at different ARFINET sites (shown with respect to their latitudes) of India. The absolute difference between BCSAT and BCSUR-N at different locations are also shown. The top panel shows the seasonal values of BCSAT, BCSUR, BCSUR-N and |(BCSAT - BCSUR-N)| around each of the observational sites during December-January-February (DJF). Same parameters are shown in the middle panel for March-April-May (MAM) and in the bottom panel for June-July-August (JJA). The letters in the histograms represent the names of individual stations (details in supplementary Table-T1). The simultaneous data available for inter-comparison are highest in DJF (17-stations) and least in JJA (9-stations).”
P24-26: Need more description about the plot (e.g., upper, center, and bottom panels indicate what) in the caption
Response: Complied with.
-END-
-
AC1: 'Reply on RC1', Mukunda M Gogoi, 17 Jan 2023
-
RC2: 'Comment on acp-2022-555', Anonymous Referee #2, 09 Oct 2022
General opinion:
The authors proposed a algorithm to retrieve black carbon from GOSAT-2 CAI-2. The authors also incorporated evaluation and validation of the satellite retrievals across a network of aerosol observatories (ARFINET) over India and the findings are extended to comprehend the global BC features. Such model is in highly demand if it is proven to work effectively. However, I am more concerned about the validity of the algorithm itself because the authors did not provide enough details on the methods, equations, and uncertainties. This may prevent the readers from understanding their work. Some descriptions and discussions are sometime puzzling, and there are thus much more revisions need to be made carefully by the authors.Major Comments
1. Inadequate innovation of the MS based on the claim of Line 40-41 “ the direct retrieval of BC from satellite based radiation measurement have not addressed so far.” This is really not true. Below are some articles published in recent years, proposed similar algorithm in other countries.Bao, F., Cheng, T., Li, Y., Gu, X., Guo, H., Wu, Y., Wang, Y., & Gao, J. (2019). Retrieval of black carbon aerosol surface concentration using satellite remote sensing observations. Remote Sensing of Environment, 226, 93-108
Bao, F., Li, Y., Cheng, T., Gao, J., & Yuan, S. (2020). Estimating the Columnar Concentrations of Black Carbon Aerosols in China Using MODIS Products. Environmental Science & Technology, 54, 11025-11036
Li, L., Che, H., Derimian, Y., Dubovik, O., Schuster, G.L., Chen, C., Li, Q., Wang, Y., Guo, B., & Zhang, X. (2020). Retrievals of fine mode light-absorbing carbonaceous aerosols from POLDER/PARASOL observations over East and South Asia. Remote Sensing of Environment, 247, 111913
- Comprehensive literature review and rigorous discussionis required in the introduction. And some details about the satellite sensor and data should be removed from the introduction to make the introduction concise.
- the authors should give a clear description of their algorithm. In section 2.1 the authors seem to spend a lot of space to review some other scholar's algorithms, which is confusing for some cross-field In addition, did the authors use official unpublished products? The authors mentionedthat the algorithm cite an under-preparation version of CAI-2 L2 aerosol retrieval ATBD (L117). If an official unpublished product is used, then a detailed description of the algorithm is needed. If the MS focuses on the improvements to existing algorithms, the basis, formulas, and the updates in this paper should also be emphasized. These descriptions must be detailed and not misleading.
- I have a few doubts about the algorithm itself. Does the minimum reflectance strategy of surface reflectance correction in this MS consistent with that described in lines 81-85? What is the role of NDVI in this decision? In addition, if this strategy is used, it should be explained in detail in the flowchart (Fig. S1), as using ‘minimum’ may lead to misunderstandings.
- In addition, the authors mentioned an internal mixing model to describe the proportion of BC in the aerosol. But it is not clear which internal mixing model are used. It is necessary to state and state the formula. How is the change in absorption of BC at different wavelengths considered? How is the absorption of other non-BC particles considered? The author defines: fbc=Vsoot/Vfine. It is also necessary to discuss the reasonableness of ignoring coarse particle aerosols. As far as I know, the spectral absorption of mixing aerosol is greatly influenced by some coarse particle (like DUST), which also show significant absorption in the near UV spectrum. These seemingly unreasonable assumptions can also have a very huge impact on later application studies.
- In the validation section I note that the authors assume a uniformly columnar distributed BC, using a simple equation for the columnar concentration and near-surface conversion, but the ideal conditions are quite different from the actual observations. I would like to see a more reasonable solution. If not, I would like to see more validation, such as SSA, BCAOD, which makes the the accuracy of the product more intuitive.
- In the comparison of Satellite retrievals vs climatological surface BC concentrations, does Satellite retrievals convert to near-ground magnitudes? If so, we need to move equation 3 here, but if not, the metrics RMSE in the validation needs to be removed, because they are two parameters with different magnitudes.
- The uncertainty analyses is missing in the MS. i.e., the uncertainty of the algorithm itself; The uncertainty of interpolation; The uncertainty of internal mixing; The uncertainty of switching columnar concentration to near ground; The uncertainty of ignoring coarse particle aerosols. The uncertainty analyses is very important for those who use the product in the future.
- How are SSA and FRP calculated in the section2-3.3? It is not reasonable to extrapolate Indianretrievals to global FRP without extended validation and uncertainty analyses, and it may be more convincing to state Indian only.
Minor Comments
- Some paragraphs are too long, need to split and simplifie
- The data in T2 and S2 for January are not matched, need double-check
Citation: https://doi.org/10.5194/acp-2022-555-RC2 -
AC2: 'Reply on RC2', Mukunda M Gogoi, 17 Jan 2023
We appreciate the summary evaluation of the reviewer. We have complied with the observations and revised the manuscript incorporating valuable comments by the reviewers. In the revised manuscript, we have given more emphasis on the algorithm description, including the various steps involved in the retrieval process. The validation and uncertainties involved in this retrieval method is also elaborated. Our point wise response to each of the comment is given in bold letters, below the respective comments.
Mukunda M. Gogoi et al.
Mukunda M. Gogoi et al.
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 356 | 172 | 17 | 545 | 44 | 3 | 5 |
- HTML: 356
- PDF: 172
- XML: 17
- Total: 545
- Supplement: 44
- BibTeX: 3
- EndNote: 5
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1