Retrieval of Aerosol Optical Properties over Land Using an Optimized Retrieval Algorithm Based on the Directional Polarimetric Camera
Round 1
Reviewer 1 Report (New Reviewer)
In this paper, the authors applied the SRON full retrieval algorithm to Directional Polarization Camera (DPC) onboard the Chinese GF-5 satellite to retrieve aerosol optical properties over land. The validation against AERONET indicates the good performance of the aerosol retrievals with DPC data. The case studies which show the SSA retrievals over land with DPC data can provide the aerosol source information such as biomass burning or dust. This work is within the scrope of the Remote Sensing and should be interesting to the audiences. I suggest it published after minor revision.
1. Line 71, the author should give the specific meaning of CM, DQ-1 and DQ-2
2. The author should give more description on the retrieval algorithm and the reason why the retrieval accuracy of SSA are relatively higher than other studies.
3. For case studies, the authors should prepare more results to show the aerosol retrievals with DPC are capable of providing valuable information for aerosol sources.
Author Response
Dear Reviewer,
Thank you for your comments concerning our manuscript entitled “Retrieval of aerosol optical properties over land using an optimized retrieval algorithm based on the Directional Polarimetric Camera”. We appreciate your comments and suggestions very much, which are valuable in improving the quality of our manuscript. Our manuscript has been carefully revised according to the comments. The main corrections in the paper and the response to the comments are as flowing:
Review report 1:
In this paper, the authors applied the SRON full retrieval algorithm to Directional Polarization Camera (DPC) onboard the Chinese GF-5 satellite to retrieve aerosol optical properties over land. The validation against AERONET indicates the good performance of the aerosol retrievals with DPC data. The case studies which show the SSA retrievals over land with DPC data can provide the aerosol source information such as biomass burning or dust. This work is within the scrope of the Remote Sensing and should be interesting to the audiences. I suggest it published after minor revision.
- Line 71, the author should give the specific meaning of CM, DQ-1 and DQ-2
Reply: Thank you for reminding. CM, DQ-1 and DQ-2 refer to Carbon dioxide Monitoring satellite, DaQi-1, also named AEMS (Atmospheric Environment Monitoring Satellite)) and DaQi-2, respectively. The specific spelling has been included in the manuscript.
- The author should give more description on the retrieval algorithm and the reason why the retrieval accuracy of SSA are relatively higher than other studies.
Reply: More description about the RemoTAP retrieval algorithm, including the formula to approximate the forward model and the iteration solution to solve the minimization–optimization problem, has been added in the manuscript in Line 224-242 of page 6. The general algorithm description can also be found in papers Fu and Hasekamp (2018), Fu et al., (2020), Fan et al (2019), Lu et al, (2022).
Better retrieval performance on SSA can probably be attributed to the following 2 aspects: 1) Multi-spectral, multidirectional and polarized satellite observations performed by DPC contain richer information of aerosols in our atmosphere from a passive remote sensing perspective. 2) The application of RemoTAP full retrieval algorithm to DPC measurements. The fine mode and the coarse mode are assumed to be composed by INORG+BC (inorganic matter and black carbon) and DUST+INORG (dust and inorganic matter), with weighting coefficients to combine the prescribed refractive index spectra from d’Almeida et al., 1991 for different aerosol components. The weighting coefficients of each aerosol components are included in the state vector to be determined in the retrieval, to better estimate the absorbing or scattering characteristics of aerosols.
- For case studies, the authors should prepare more results to show the aerosol retrievals with DPC are capable of providing valuable information for aerosol sources.
Reply: The dust and biomass burning scenarios have been extended to better show the pattern. For each case presented in the manuscript, we’ve performed retrieval for 6 days’ data and plotted the retrievals. Fig. 8 and Fig. 10. show the retrieved multi-temporal aerosol optical depth over northeastern China for the dust and biomass burning scenarios.
We appreciate for your warm work earnestly, and hope that the revisions will meet with approval.
Once again, thank you very much for your good comments and suggestions.
We look forward to your information about our revised manuscript.
Yours sincerely,
Li Fang
China Center for Resources Satellite Data and Application
Beijing, CHINA
Author Response File: 
Author Response.pdf
Reviewer 2 Report (Previous Reviewer 1)
Comment to “Retrieval of aerosol optical properties over land using an optimized retrieval algorithm based on the Directional Polarimetric Camera” by Fang et al. he authors have solve my concerns and I would like to recommend its acceptance for publication after a few minor changes.
L192, what’s the value of Nα. How many aerosol components are used in the study?
How many AERONET sites are used in AOD retrieval validation?
Author Response
Dear Reviewer,
Thank you again for your comments concerning our manuscript entitled “Retrieval of aerosol optical properties over land using an optimized retrieval algorithm based on the Directional Polarimetric Camera”. The main corrections in the paper and the response to the review report are as flowing:
Review report 2:
Comment to “Retrieval of aerosol optical properties over land using an optimized retrieval algorithm based on the Directional Polarimetric Camera” by Fang et al. the authors have solve my concerns and I would like to recommend its acceptance for publication after a few minor changes.
L192, what’s the value of Nα. How many aerosol components are used in the study?
Reply: Thanks for the comment. In this study, we set Nα = 2 for both fine and coarse modes. The fine mode and the coarse mode are respectively assumed to be composed by INORG+BC (inorganic matter and black carbon) and DUST+INORG (dust and inorganic matter). This assumption is flexible and can be updated according to the information content of the measurement. The value of Nα and corresponding components are also clarified in section 3.2 of the manuscript.
How many AERONET sites are used in AOD retrieval validation?
Reply: Since the stray light correction has been applied to re-process the DPC Level 1b data between December 2019 and April 2020, as stated in the manuscript, the available global AERONET stations over land with measurements during this time period are used for validation and comparison, including about 131 AERONET stations. The match method of DPC retrievals and AERONET products has been further clarified in Line 256-262 in page 7.
Once again, thank you very much for your good comments and suggestions, and hope that the revisions will meet with approval.
We look forward to your information about our revised manuscript.
Yours sincerely,
Li Fang
China Center for Resources Satellite Data and Application
Beijing, CHINA
Author Response File: Author Response.pdf
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
This paper by Fang et al. applies the SRON Remote Sensing of Trace gas and Aerosol Products (RemoTAP) full retrieval algorithm to Directional Polarization Camera (DPC) onboard the Chinese Gaofen-5 satellite to retrieve aerosol properties including AOD, the fine/coarse mode AOD and the SSA. Validation against AERONET indicates that DPC AOD retrievals perform better when fine mode aerosols dominate and SSA retrievals have difference with AERONET SSA retrievals lower than 0.05. The fine/coarse mode AOD and the SSA synergistic retrievals can provide valuable information for biomass burning aerosol and dust tracking.
1. Minor errors
Line 71, the specific spelling/meaning of CM, DQ-1 and DQ-2
Line 74, is -> are
Line 82, ‘measure radiance and radiance and polarization’, radiance?
Line 85, DoLP refers to the degree of polarization?
Line 131, pay attention to the citation style, Huang et al. (2019)
Line 138, add doi:10.5194/amt-12-169-2019 for AERONET version 3 database.
Line 392-393, ‘over Beijing’?
2. Major ones
Line 255-256, the RMSE values cannot be compared directly because they are calculated from two AOD datasets. Authors should use the normalized RMSE for comparison, same for biases.
Line 381-382, what does this sentence mean?
In table 2, are the parameters (the effective radius and the effective variance) in this table optimized for China? If not, can the authors use AERONET data to construct a table for aerosols considering high pollution period occasionally occurs in north China?
Reviewer 2 Report
This study presents aerosol retrievals derived from DPC observations by the SRON RemoTAP algorithm. The validation of DPC retrievals using the AERONET products is conducted. Two cases of dust and biomass burning are also discussed by the DPC retrievals. This study seems to indicate that the SRON RemoTAP algorithm could be applied to DPC observations for aerosol inversion. However, I suppose the current version does not meet the requirements of remote sensing journal. Several parts of this manuscript need to be improved. For example, the description of SRON RemoTAP algorithm is too simple, the authors should provide more information about the algorithm and also for the AERONET sites used for the validation. The match of DPC retrievals and AERONET products needs to be clarified. More importantly, the retrievals used for the biomass burning and dust cases are too limit, only some data in two paths of DPC observations on March 13 for dust case and some data on February 6 for biomass burning. These results seem to be not convincing. I suggest the authors add more retrievals at least for several days (a week), which can provide more information about specific case. The selected regions of these dust and biomass burning cases also can be improved. Because there are many more key regions of interests globally, such as Saharan desert in the north Africa, Middle East, Taklamakan in China for dust case and biomass burning events in southern Africa, Indo-China Peninsula. Therefore, I reject the current version of this study.
