Monsoon-Based Linear Regression Analysis for Filling Data Gaps in Gravity Recovery and Climate Experiment Satellite Observations

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The LRA method proposed in this article performs well for GRACE/GRACE-FO gap filling, and the method is well-designed and detailed. There are still confusion and needs for improvement in terms of content:

1. The clarity of the images is too poor, it is recommended to modify the format as well as the color of the images, otherwise, the reader will not be able to get much valid information. There are also some flaws in the writing, such as

*  line 170, "EWT" change to "EWH"

* the second equal sign in eq.(17)

2. The eq.(3)  is confusing, when m=17, Q_m=17=17?  and when n is grid points number, how to use it for the SHCs? 

3. The method you mentioned is equivalent to performing a linear regression on each point to obtain the regression coefficients, so is it true that the denser the grid the better the results? How to choose this value? 

4. The monthly gravity field product is smoothed by a 300km Gaussian filter in your description of the article, so how much influence will the spatial leakage error have in the parameter estimation? What are the results if other smoothing methods are used?

5. It is recommended to provide a triangle spectrum (like Fig.3 ) for each degree/order correlation coefficient of Figure 4.

 

 

 

 

Comments on the Quality of English Language

This paper is of medium English quality and the exposition is rather wordy and could be further streamlined and improved in terms of writing quality.

Author Response

Thank you for your valuable time and feedback. We sincerely appreciate your input. Please find our response in the attached comment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

THis paper presents another method to interpolate GRACE and GRACE FO missing data months. THe method is simply a linear interpolation at fixed month across different years. The assumption, which is really not good, is that there is NO INTERANNUAL VARIABILITY, that a simple linear trend fitted to each january, each february... is enough to interpolate a missing january, a missing february, etc. This would be true if there were no El Nin~o events, no atmospheric rivers, etc.

THe paper attempts to validate the interpolation method ('LRA') in three ways: 1) by removing the year 2009 and comparing the LRA interpolated 2009 with the GRACE-measured one; 2) by comparing the LRA with other published interpolation methods; 3) by comparing area averages in several hydrologic basins for one month (jan 2016), as well as by computing gravity anomalies.

I have many objections to this paper, and believe iot should nto published until these defects are repaired

1) A proper estimation of interpolation error consists of the following: eliminate 2004, interpolate, compute differences; eliminate 2005, interpolate, compute differences,.... repeat until 2016, interpolate, compute differences. Then compute the ROOT MEAN SQUARED (RMS) DIFFERENCES for each n,m,month across all years. 

2) Repeat 1 with the other global methods and compare LRA with them.

3) Correlation coefficients are useless in this context, you are trying to evaluate ERROR, no need to present. Equations for linear interpolation can be found in any textbook, no need to present. 

4) Figure 3 is another example of useless information: all those identical plots should be summarized in a few RMS numbers.

5) the basins calculation: a 1 month comparison is not statistically significant. Repeat for different months and produce the RMS numbers.

6) I also have a language objection: the authors use many superlative adjectives to describe their incredibly excellent method. We the readers will be the judges. Just as figure 3 has too many useless plots, the text has too many useless adjectives.

Comments on the Quality of English Language

see above

Author Response

Thank you for your valuable time and feedback. We sincerely appreciate your input. Please find our response in the attached comment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear authors,

thank you for this interesting study. I only have some minor points marked and commented in the attached PDF, which I recommend to be addressed in a revised version.

Kind regards

Reviewer

Comments for author File: Comments.pdf

Comments on the Quality of English Language

English Language is fine.

Author Response

Thank you for your valuable time and feedback. We sincerely appreciate your input. Please find our response in the attached file.

Author Response File: Author Response.pdf