Integrated Remote Sensing Observations of Radiative Properties and Sources of the Aerosols in Southeast Asia: The Case of Thailand
Round 1
Reviewer 1 Report
Please see attached review
Comments for author File: 
Comments.pdf
Please see attachment.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The authors conducted a study to understand the complex behaviors of aerosols in Southeast Asia (SEA), focusing on their optical properties and radiative forcing. To gather data, the study employed remote sensing instruments from Earth-observing satellites, including MODIS/Terra, MODIS/Aqua, and CALIOP/CALIPSO. Ground-based observations from AERONET were also utilized. Through cluster analysis, the study identified seven dominant types of aerosols across the region, each with varying impacts on radiative forcing. The study revealed that aerosols with positive radiative forcing (warming) were primarily found in mainland SEA during both regular and high-aerosol events. Conversely, aerosols with negative radiative forcing (cooling) were associated with aging processes. In insular SEA, background aerosols with neutral radiative forcing were most prevalent. These were primarily a mix of oceanic and local anthropogenic aerosols. The author suggests that future research should focus on carbonaceous aerosols, such as organic carbons, black carbon, and brown carbon, as well as the aging processes of these aerosols since they play a pivotal role in regional aerosol optical properties.
This manuscript delves into a captivating topic and is generally well-organized and informative. I recommend this manuscript for publication, provided that one concern is addressed. My main critique of the manuscript is that it does not offer comprehensive descriptions of the clustering approaches. I would advise the authors to provide the following information:
1) Detailed specifics regarding clustering inputs and methods, especially data processing for the clustering inputs.
2) An explanation of how the clustering number, seven, was determined.
3) Descriptions of how the HYSPLIT model was utilized for the study, which is missing from current manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Major Comments:
1. Suggest adding content to explain why the seven aerosol clusters were chosen. Although I guess you are trying to be able to correspond to the seven aerosol subtypes of CALIOP, there is still a need to demonstrate the stability of the cluster results. The uncertainty in some of the aerosol clusters (e.g. L3 and L6) may be due to not choosing the right number of clusters.
2. The analysis of satellite remote sensing (CALIOP) in the manuscript is insufficient, and currently only a simple selection of one CALIOP aerosol subtype corresponds to one aerosol cluster. It is hoped that the authors will explain in detail the correspondence between the seven CALIOP aerosol subtypes and the seven aerosol clusters calculated, and how the CALIOP aerosol subtype profiles contribute to the results of this manuscript.
3. Lines 287-337: It is important to note that while the SSA and the imaginary part of refractive indices are strongly correlated with aerosol radiative forcing, the values of the SSA and the imaginary part of refractive indices cannot be used directly to determine positive or negative radiative forcing. It is not rigorous to use SSA and the imaginary part of refractive indices directly to classify aerosol clusters as positive or negative radiative forcing aerosols in this manuscript.
4. Lines 306 to 309: The correlation between aerosol effective size and precipitable water may not directly represent aerosol hygroscopicity, and it's essential to understand why this correlation exists and the limitations associated with using it as a measure of aerosol hygroscopicity. Please give reasons and provide references.
5. Lines 350-353: An accuracy of only 48% would appear to make the result less reliable. In fact, most of the AERONET-based aerosol clusters are more than 50% accurate (Table 5), and the results are still informative, but the authors are advised to state it differently.
6. Lines 426-489: The manuscript infers that the L3 aerosol is an oceanic aerosol mixed with urban aerosols based on optical properties and environmental parameters, but the CALIOP results suggest it may be dust, polluted dust, and polluted continental/smoke aerosol subtypes, and the difference seems to be too great. Possibly because of the low accuracy of L3 aerosols (Table 5)? Or because of the limited number of CALIOP data points (Figure 4)? It is recommended to explore which of the AERONET and CALIOP results is more accurate.
Minor Comments:
1. Lines 59 to 61: The cited references are missing here. In terms of cloud formation, the author can refer to Wang et al. (2022, DOI: 10.5194/acp-22-15943-2022).
2. AERONET data are the most important data in this manuscript. However, in the introduction, too much space is devoted to satellite remote sensing and not enough to the advantages and current research progress of ground-based remote sensing. It is recommended that the introduction be reorganized.
3. The MODIS AOD-derived PM2.5 AQI has not been used in the main text and does not seem to be helpful for data analysis.
4. Prevailing winds and backward trajectories are often used in the manuscript to illustrate the long-range transported aerosols. It is recommended that pictures or data be given.
5. Some aerosol clusters (e.g. L1 and L7) have different names in section 3.2 and in the conclusion, and I think the names in the Conclusion are more appropriate.
6. Lines 225-226: If I remember correctly, the spatial resolution of the SRTM data is 30 m only for the US range, and 90 m elsewhere.
7. Lines 309-310 and 350-353: R2 of 0.2 or 0.3 is not significant, and "significant" needs to be used with more caution.
8. Lines 676-683: there is an error in the numbering here.
Minor editing of English language required
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Please see attached review
Comments for author File: Comments.pdf
Please see attached review
Author Response
We greatly appreciate the time and dedication you've invested in enhancing the quality of this manuscript. Your valuable comments have provided us with a wealth of knowledge.
- L159: Do not specify a value of “about 0.85”. This is not supported by Table 2, and also this seems like an introductory statement about the asymmetry parameter, so specific values are not needed. It is sufficient to state that forward scattering aerosols have a “positive “asymmetry parameter.
Response We are thankful for your comment. We agree with you and the value had been deleted.
- P6: The response on the date range for CALIPSO data was insufficient.
Response We appreciate your concern. We have made an effort to provide a detailed explanation of the temporal availability gap in CALIPSO data. This helps clarify why a limited number of profiles were utilized in this study.
“The availability of CALIOP aerosol profiles is restricted due to two main factors: 1) CALIPSO revisited the same locations only every 16 days and 2) cloud interference in aerosol observations, leading to limited data during rainy months.”
- L218: “spatial corresponding” > “proximity”
Response Thank you so much. The change has been made accordingly.
- P8 (Table 2): Imaginarry > Imaginary
Response It is our mistake. Thank for your correction. The change has been made accordingly.
- P8 (Table 2): Asymmetry parameters at 440 nm are >1.00 and give evidence of a large bias.
Response Thank you for your valuable feedback. We have thoroughly reviewed the raw datasets and noted that the asymmetry parameters at 440 nm frequently exceeded 1. This suggests that the discrepancy likely arises from the data processing stage rather than our calculations. Importantly, this oversight does not impact the conclusions drawn in our study, as our analysis did not hinge on data from this specific wavelength. In response, we have augmented the 'methodology' section to include detailed information on the range of asymmetry parameter data. Additionally, we have included a note below Table 2, acknowledging the measurement error.
Section 2.2. Data description
“Asymmetry parameter represents the average cosine of light scattering angles by aerosol particles, and consequently, it falls within the range of -1 to 1.”
Table 2:
“Note: The asymmetry factor is expected to fall within the range of −1 to 1. However, its value often exceeded unity at 440 nm, indicating a potential error in the measurement process.”
- L305 (and L381 and L559): I think the authors should not use “negative”. I strongly suggest that they simply replace with “weak”.
Response We agree with you. The weak correlation aligns well with the following statement regarding the less hygroscopic properties. We have changed it accordingly.
- P10 (Table 5): There should not be a horizontal line under “1” in the Aerosol Cluster column. Also, “1” should not be in bold. I realize this is not the fault of the authors, but these corrections should hopefully be made by the journal.
Response We have informed the Editor about this fault.
- L420: The authors responded: “In this sentence, the L2 aerosol has more water soluble organic content in both fine and coarse model. Thus we maintain it in its original format.” But I think that only the coarse model has a stronger correlation than the L1 aerosol according to Table 3.
Response We acknowledge and greatly appreciate your comment and suggestion. We would like to address the original statement of “Compared with the L1 forest fire aerosol, the L2 aerosol had a lower real-part refractive index and overall water-absorbing property.”
In this context, it is crucial to differentiate between weak and significant correlations, as they convey distinct meanings. Specifically, for fine aerosol, we observe a correlation coefficient (r) of 0.201 with a p-value less than 0.001, and for coarse aerosol, a correlation coefficient of 0.28 with a p-value < 0.001. A p-value below 0.001 signifies a statistically significant correlation at the 99.9% confidence level.
Despite the r values being approximately 0.2, which indicates a weak correlation and implies that the dependent variable alone does not fully account for the variation in the independent variable, we maintain that the significant positive correlation, as suggested by the p-value, strongly supports the assertion that both aerosol sizes exhibit water-absorbing properties. This holds true regardless of the strength of the correlation."
- L462: Reverse the order for clarity: “light-absorbing capacity and poor hygroscopicity.”
Response The order has been changed accordingly.
- L498: E > East
Response The change has been made accordingly
- L513: “Appendix2” > “Appendix 2”
Response The change has been made accordingly
- L582: “agree” is not correct. There is an order of magnitude of difference in the imaginary component of the refractive index.
Response Thank you for your suggestion. We agree that the sentence conveys a vague meaning. We have rephrased the sentence to: “The L7 cluster displays the lowest complex refractive indices (1.42+0.01i, at 675 nm) among all aerosol clusters. This observation aligns with maritime aerosols, known for their characteristics lower refractive indices as reported by Levoni et al. (1997) at 1.381+ <0.001i.”
- L583: “a” > “is a”
Response Thanks a bunch for your improvement. The change has been made accordingly.
Reviewer 2 Report
I appreciate the authors' efforts to revise the manuscript. I recommend the manuscript for publicaton.
Author Response
We greatly appreciate the time and dedication you've invested in enhancing the quality of this manuscript. Your valuable comments have provided us with a wealth of knowledge.
Reviewer 3 Report
All my questions have been addressed. I recommand to publish it on Remote Sensing.
Author Response
We greatly appreciate the time and dedication you've invested in enhancing the quality of this manuscript. Your valuable comments have provided us with a wealth of knowledge.
