Hyperspectral Infrared Atmospheric Sounder (HIRAS) Atmospheric Sounding System
, Chenggege Fang 2,3, Sichen Wang 4,5,6,7, Shangpei Xun 4, Binfang He 4, Wenyu Wu 4 and Yanfeng Huo 4,6,7,*
Round 1
Reviewer 1 Report
Comments for author File: 
Comments.pdf
Author Response
We have read your comments and criticisms seriously. Thank you very much for your affirmation of our work in this paper
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Authors, the manuscript is well written and presents your Algorithm with clear and exhaustive results.
In the 4, where you present your algorithm, as well as the results of your analysis, I was expecting also a summary of the comparison of all the cases: Figure 15, Figure 14, Figure 12, Figure 11. Maybe in this case, you just select one profile and plot the same profile for each algorithm.
I would also extend a bit the conclusion.
I spotted some pdf issues on Table 3.
Best regards.
Author Response
We have read your comments and criticisms seriously. Your constructive suggestions are helpful for further improving the quality of this work. Accordingly, we have seriously modified this manuscript. The detailed replies to the comments are presented in a point-to-point manner as follows and the changes made have been highlighted in red in the revised manuscript.
-----------------------------------------------------------------------
- The validation is only done with ERA and GNOS. This reduces the value of the publication. Why no data from radiosondes or other sounders were used? Please explain and add the explanation to the manuscript.
Reply: Thanks for your constructive suggestions. It is an effective method to verify the accuracy of the algorithm by using radiosondes profiles. Unfortunately, conventional radiosondes usually occur at 08:00 UTC and 20:00 UTC, which doesn't match the time of satellite passing in the Western Pacific. We have add the explanation to the manuscript in line 315.
- The retrieval uses DA for one satellite, however, NWP uses DA for many satellites as well as in-situ data. Hence, please explain the add-on of the retrieved profiles compared to temperature and moisture profiles from ECMWF or another NWP and add it to the manuscript.
Reply: Thanks for your constructive suggestions. Although the algorithm uses DA for retrieval in this paper uses, it does not depend on the precision of the prior profile, which is different from temperature and moisture profiles from ECMWF or another NWP and can reflect the more information of satellite observation. We have added related content in section 5.
- Use of background profiles. How sensitive is the method to the choice of background profiles? Please discuss and add the discussion to the manuscript
Reply: Thanks for your constructive suggestions. The algorithm in this paper is insensitive to background profiles and it is one of the advantages of our algorithm. We have added related content in section 5.
- Please separate the conclusion from the discussion
Reply: Thanks for your constructive suggestions. The discussion in this section is only the future plan for current work. We have revised the title of section 5.
- L113-114: “These interferograms will be converted through FFT to the complex spectral energy and then to the radiances through a two-point radiometric calibration.” Please explain the steps in more detail
Reply: Thanks for your constructive suggestions. The conversion from interferograms to radiances is not the focus of this paper and we have cited the related work in Qi C. et al, 2020.
- L151 "the information entropy of HIRAS channels is analysed"
Please consider to add, as explained in more detail below
Reply: Thanks for your constructive suggestions. The specific detail of information entropy method can be seen in Rodgers, C. D., 1998, Rabier, F., 2002, and Collard, A. D., 2007.
- Figures 2,4,5,9: Please increase the size of the figures.
Reply: Thanks for your constructive suggestions. We have increase the size of Figures 2,4,5,9.
- L184 “A cloud detection module using MERSI-II visible and infrared observation” Also at other paragraphs MERSI-II is mentioned but it is not described in more detail. Cloud screening is a central error source for atmospheric soundings. Thus, could you please add more information on the MERSI-II retrieval method or a reference to a respective publication.
Reply: Thanks for your constructive suggestions. In fact, References 38 introduced the cloud detection method of FY3D/ MERSI-II. We have changed its cited position in paper.
- L 264-265 “… unbiased and Gaussian [40]. In fact, the bias between observed radiance and simulated radiance is always existed because of approximate calculation. “I think there is something wrong with the grammar, please rephrase.
Reply: Thanks for your constructive suggestions. We have rephrased the related sentence in section 3.4.
- L266 Saunders et al. The year is missing. Please correct it.
Reply: Thanks for your constructive suggestions, which are quite helpful for further improving the quality of this paper. This question is caused by the wrong format of references. We have corrected all similar mistakes in this paper.
- L 268 “Currently, we only consider the angle-dependent biases in this study” Please explain why
Reply: Thanks for your constructive suggestions. It is very complicated to consider the influence of time, geographic location, satellite zenith angle, and scene temperature on the radiance biases. Currently, we only consider the angle-dependent biases in this study. More influences will be considered in our future work.
- "Due to the lack of sounding data..." Why is there a lack of sounding data (as there are other sat sounders) and why are no radiosonde data available? Please explain.
Reply: Thanks for your constructive suggestions. This question is similar to question 1 and we have answered in question 1.
Author Response File: Author Response.docx
Reviewer 3 Report
General items:
-The validation is only done with ERA and GNOS. This reduces the value of the publication. Why no data from radiosondes or other sounders were used ? Please explain and add the explanation to the manuscript.
-The retrieval uses DA for one satellite, however, NWP uses DA for many satellites as well as in-situ data. Hence, please explain the add-on of the retrieved profiles compared to temperature and moisture profiles from ECMWF or another NWP and add it to the manuscript.
-Use of background profiles. How sensitive is the method to the choice of background profiles ? Please discuss and add the discussion to the manuscript.
-Please separate the conclusion from the discussion.
Specific comments:
L 43: Please add clear sky
L113-114: “These interferograms will be converted through FFT to the complex spectral energy and then to the radiances through a two-point radiometric calibration.”
Please explain the steps in more detail
L151 "the information entropy of HIRAS channels is analysed"
Please consider to add, as explained in more detail below
Figures 2,4,5,9: Please increase the size of the figures.
L184 “A cloud detection module using MERSI-II visible and infrared observation”
Also at other paragraphs MERSI-II is mentioned but it is not described in more detail. Cloud screening is a central error source for atmospheric soundings. Thus, could you please add more information on the MERSI-II retrieval method or a reference to a respective publication.
L 264-265 “… unbiased and Gaussian [40]. In fact, the bias between observed radiance and simulated radiance is always existed because of approximate calculation. “
I think there is something wrong with the grammar, please rephrase.
L266 Saunders et al.
The year is missing. Please correct it.
L 268 “Currently, we only consider the angle-dependent biases in this study”
Please explain why
L315: "Due to the lack of sounding data..."
Why is there a lack of sounding data (as there are other sat sounders) and why are no radiosonde data available ? Please explain.
Author Response
We have read your comments and criticisms seriously. Your constructive suggestions are helpful for further improving the quality of this work. Accordingly, we have seriously modified this manuscript. The detailed replies to the comments are presented in a point-to-point manner as follows and the changes made have been highlighted in red in the revised manuscript.
-----------------------------------------------------------------------
In the 4, where you present your algorithm, as well as the results of your analysis, I was expecting also a summary of the comparison of all the cases: Figure 15, Figure 14, Figure 12, Figure 11. Maybe in this case, you just select one profile and plot the same profile for each algorithm.
Reply: Thank you for your comment. I’m very sorry for the misunderstanding. In fact, the RMSEs and biases of the temperature and relative humidity profiles in Figure 15, Figure 14, Figure 12 and Figure 11 are the mean value of all valid profiles.
Author Response File: Author Response.docx
