Uncertainty of CYGNSS-Derived Heat Flux Variations at Diurnal to Seasonal Time Scales over the Tropical Oceans

Round 1

Reviewer 1 Report

This study provides a comprehensive and detailed analysis of the accuracy and limitations of the CYGNSS derived turbulent heat fluxes for the global tropical oceans. The study is well-conducted and provides valuable insights into the accuracy and limitations of the CYGNSS heat fluxes product for the global tropical oceans. With some revisions and clarifications, the study could make a significant contribution to the field of oceanography. 

Overall Comments:

1. The results of the study indicate that while there is overall consistency between the turbulent heat fluxes derived from CYGNSS and MERRA-2 wind speed and those calculated from buoy wind speed, there are differences in bias and RMSE for sites located near the equator in the central-eastern Pacific Ocean, the Arabian Sea and the Bay of Bengal, and near the Gulf of Guinea. The study should provide a more detailed discussion of the potential causes of these differences and their implications for future research.

2. The authors should provide more detailed information on the methods you used to calculate the difference between the heat fluxes calculated by the CYGNSS wind speed and MERRA-2 wind speed and the heat flux calculated by the buoy wind speed within the three tropical oceans, especially for the experiment design. This would help readers to better understand the procedures you used to obtain your results.

Specific Comments:

1. Introduction

-Line 49-50, It is better may be as: However, the turbulent heat fluxes obtained from in-situ observations are limited in spatial and temporal coverage.

-Line 81-83, do the authors mean that the two versions of CYGNSS heat flux products are based on the CYGNSS wind products provided by NASA and NOAA?

-Line 88-89, what are the main points of the “6 experiments”?

-Line 89, the full name of “COARE” should be given for the first time.

-Line 90-91, “error” -> “errors”, by the way, the authors use the words error and bias in their paper, are the two words mean the same thing?

2. Data methods and experimental design

-Line 93, usually we use “Data and methods”

-The constant parameters used in the equations should be given its values.

-The authors seem calculated the LHF and SHF based on the COARE3.5 formulas, by using different input variables from the buoy, CYGNSS and MERRA-2 dataset, as we known, the buoy observed air and sea temperature are not exactly equivalent to the satellite and reanalysis data. How do the authors treat about this?

3. Results

-Line 200, I suggest to give the biases and correlation coefficients quantitatively, rather than to descript by using the words small and high.

-Line 241, please explain what is heat flux interval?

-Line 270-272, why the class B experiments show smaller values than those of class A?

-Line 288-289, here the authors indicate larger bias in the western Pacific, but in the abstract is in the central-eastern Pacific, this should be explained.

-Line 331-347, I can not understand Figure 8.

-Fig.9 and Fig.10, the authors pointed out large biases in the Atlantic Ocean, discussion about the reasons why the biases are large in Atlantic but small in Pacific and Indian Ocean will improve our understanding.

4. Discussion and conclusions

- I suggest the authors to provide a more detailed discussion of the potential causes of these differences and their implications for future research

It is recommended to check the full text carefully to avoid some syntax errors。

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

This study evaluates the CYGNSS derived sea surface latent and sensible heat fluxes by comparing with those derived from buoy observation and reanalysis data. This study gives an assessment of uncertainty of the CYGNSS heat fluxes at different time scale, providing important information for the application of the CYGNSS heat flux product. The paper is basically well organized and written. I would like to recommend it for publication in Remote Sensing after minor revision. Comments are listed below.

 

Major comments:

1. The chapter titles are not accurate enough to summarize the content, e.g., Section 3.1 “Overall comparison of the experiments”, it is not clear for what comparison? and what experiment?” Section 3.2 “Impact of Input Wind Speed Dataset on the Calculation”, it should be “Sensitivity of different input wind speed dataset for the heat flux calculation”. Overall, the chapter titles should be revised throughout the manuscript.

2. Section 3.4, the wavelet coherence analysis of SHF (Figure A2) seems show different features compared with the LHF (Figure 7), therefore, I would like to suggest move the Figure A2 to the main text as Figure 8.

3. The authors tried to assess the biases of CYGNSS heat flux at different time scales, this is different from that to assess the biases of hourly, daily, and monthly CYGNSS heat flux data as they shown in their sections 3.4, 3.5 and 3.6. These should be clearly clarified.

 

Specific comments:

1. Abstract, L25, “… CYGNSS Derived turbulent …”, to “CYGNSS derived turbulent”.

2. Introduction, L40, “because of”, delete “of”.

3. Section 2.3, L148-149, please confirm whether it is surface skin temperature or sea surface temperature.

4. Section 3.1, L219-224, it is confused, and should be rewritten.

5. Section 3.1, L241-243, and L243-245, I think these two sentences are repeated and please combine them into one sentence.

6. Section 3.2, L255-257, the sentence is confused, I guess the authors means “There are significant differences of LHF derived from different combination of input wind speed and air-sea conditions in tropical oceans”.

7. Section 3.2, L257-259, maybe rewritten the sentence as “The experiment A2 gives the smallest LHF, and experiments A0, B0 and B1 gives the largest LHF”.

8. Section 3.4, L302, Why the wavelet coherence analysis is at “hourly scales?”, what means hourly scales?

9. Section 3.5, L373-374, what means “not significant enough”?

10. Section 3.6, L400, what means “at all time periods”? It should be given clearly.

11. Section 4, L448, “high wind speed region”, should be “high wind speed condition”.

12. There are still many inappropriate English expression in the manuscript, above is only few that I can easily find, the whole manuscript should be checked thoroughly.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript is well written on the comparison of satellite datasets and in-situ observations. Such research is required to work on specific algorithms. Some of my comments are as follows:

1. The abstract could carry some skill scores, otherwise, it seems to be very plain.

2. Why was it difficult to use these datasets for synoptic scale processes? Satellite datasets are useful for data assimilation and comparison of model forecasts. Thus, this is a great question on the ability of satellites even though they have great spatial and temporal coverage.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The recently deployed CYGNSS satellite constellation facilitates an unprecedented opportunity to ascertain the sea surface turbulent heat flux with exceptional temporal and spatial resolution, utilizing its measurements of sea surface wind velocity. Prior research endeavours, such as those conducted by Crespo et al. (2019) and Li et al. (2022), have underscored the merits of employing CYGNSS for evaluating sea surface heat flux. These investigations imply a strong correlation between the CYGNSS sea surface heat flux product and estimates extrapolated from buoy data, as corroborated by scatterplot analyses and calculations of bias and root mean square errors (RMSEs).

In the paper under consideration, the authors have made contributions to the appraisal of the CYGNSS heat flux in two pivotal aspects: 1) they have scrutinized the uncertainty inherent in the variability of CYGNSS heat flux across different temporal scales, revealing a consistency in heat fluxes among CYGNSS, MERRA-2, and buoy products at lower frequency bands with periods exceeding 10 days. However, they found an inconsistency at higher frequency bands with periods less than 5 days; 2) they have conducted a rigorous quantitative analysis of the biases in heat flux that are associated with wind speed and other variables, using the sophisticated Ensemble Empirical Mode Decomposition (EEMD) techniques.

The manuscript is meticulously composed and I endorse its publication in Remote Sensing, subject to minor rectifications to correct a few discrepancies.

On line 41, replace 'are also refer to' with 'are also referred to as'.

On line 44, omit 'the' preceding 'wind speed'.

On line 53, alter 'satellite derived' to 'satellite-derived'.

On line 330, modify 'generally are ' to 'are generally'.

On line 525, substitute 'the study' with 'our investigation'.

Author Response

We would like to express our great appreciate for you encouragement and comments. The revised manuscript has been refined according to your suggestions. These comments and suggestions greatly help us in improving the quality of this manuscript. Once again, thank you for your time, expertise, and for acknowledging the merits of our work. Revision please see the attachment.

Author Response File: Author Response.pdf