Evaluation of Terrestrial Water Storage Changes over China Based on GRACE Solutions and Water Balance Method
Round 1
Reviewer 1 Report
This is the review for the manuscript titled “Evaluation of terrestrial water storage changes over China based on GRACE solutions and water balance method”. The authors attempt to evaluate the water balance method-based TWSA series based on GRACE and GLDAS solutions over 10 river basins within China. Generally speaking, this manuscript is meaningful and well prepared. Therefore, I suggest this manuscript be accepted after a minor revision. Following are some suggestions to improve this manuscript:
(1) With respect to “3.1.2 GLDAS simulations”, the word “in this study” emerges twice in one sentence, and suggest deleting the duplicate contents.
(2) In section 3.2.2, please give a detailed explanation about the H0 hypothesis, which emerges in the sentence “the original hypothesis H0 is rejected”.
(3) In section 4.1, “in SERB” should be changed to “in the SERB”, and modify the similar mistakes throughout the manuscript.
(4) In section 4.3, the range of GRACE and GLDAS varies from -200 mm to 200 mm. However, the manuscript shows the range from -200 mm to 2000 mm, please correct it.
Author Response
This is the review for the manuscript titled “Evaluation of terrestrial water storage changes over China based on GRACE solutions and water balance method”. The authors attempt to evaluate the water balance method-based TWSA series based on GRACE and GLDAS solutions over 10 river basins within China. Generally speaking, this manuscript is meaningful and well prepared. Therefore, I suggest this manuscript be accepted after a minor revision. Following are some suggestions to improve this manuscript:
Responses: Thanks for spending time on my manuscript. We have carefully considered every comment, and tried our best to make responses and modifications. We sincerely hope that our reply can make you satisfied.
(1) With respect to “3.1.2 GLDAS simulations”, the word “in this study” emerges twice in one sentence, and suggest deleting the duplicate contents.
Responses: Thanks for your kind suggestions. Change is made.
We have deleted the duplicate words “in this study”, and this sentence has been changed to “In this study, the GLDAS Noahv2.1 model is utilized to calculate TWSA estimates, including soil moisture, snow water equivalent, and plant canopy water at a 0.25° resolution.”
(2) In section 3.2.2, please give a detailed explanation about the H0 hypothesis, which emerges in the sentence “the original hypothesis H0 is rejected”.
Responses: Thanks for your kind suggestions. Change is made.
The null hypothesis H0 assumes that the sample data is random, independent, and no linear trend. The detailed explanation about the H0 hypothesis has been given in the revised manuscript.
(3) In section 4.1, “in SERB” should be changed to “in the SERB”, and modify the similar mistakes throughout the manuscript.
Responses: Thanks for your kind suggestions. Changes are made. We have replaced “in SERB” by “in the SERB” throughout the manuscript.
(4) In section 4.3, the range of GRACE and GLDAS varies from -200 mm to 200 mm. However, the manuscript shows the range from -200 mm to 2000 mm, please correct it.
Responses: Thanks for your kind suggestions. Change is made.
We have corrected the mistake number in the revised manuscript, and this sentence has been changed to “WB-based TWSA range from -2000 mm to 4000 mm in the HHRB, while the range of GRACE and GLDAS is from -200 mm to 200 mm”.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This paper reconstructed long-term TWSA estimates to apply to 10 river basins (RBs) within China and to validate against GRACE observations and GLDAS simulations. It has a good potential to be published in the journal. However, there are some issues which need to be addressed before it is ready for publication.
1. In the Abstract, what are CC, NSE, RMSE and MAE metrics? Please write the full name.
2. “Terrestrial water storage anomalies (TWSA)” is an important concept. It is necessary to explain it in the first paragraph of the Introduction.
3. Why use CC, NSE, RMSE and MAE metrics rather than one or two in the Methods? Please clarify.
4. The lack of Discussion is the largest flaw of this manuscript. I suggest that authors should adjust structure as independent Results and Discussion. Authors add the discussion to explain the contribution and attribution of your article.
Author Response
This paper reconstructed long-term TWSA estimates to apply to 10 river basins (RBs) within China and to validate against GRACE observations and GLDAS simulations. It has a good potential to be published in the journal. However, there are some issues which need to be addressed before it is ready for publication.
Responses: Thank you very much for your time in this review and your great help in improving our paper. We really appreciate your comments, which have inspired us a lot. We considered each comment raised from you carefully, and sincerely hope that our reply can make you satisfied.
1. In the Abstract, what are CC, NSE, RMSE and MAE metrics? Please write the full name.
Responses: Thanks for your kind suggestions. Changes are made.
The full names of CC, NSE, RMSE, and MAE are correlation coefficient, Nash-Sutcliffe efficiency coefficient, root mean square error, and mean absolute error. We have given full names of CC, NSE, RMSE, and MAE metrics in the abstract.
2. “Terrestrial water storage anomalies (TWSA)” is an important concept. It is necessary to explain it in the first paragraph of the Introduction.
Responses: Thanks for your kind suggestions. Changes are made.
The detailed concepts about terrestrial water storage anomalies has been added in the first paragraph of the Introduction.
3. Why use CC, NSE, RMSE and MAE metrics rather than one or two in the Methods? Please clarify.
Responses: Thanks for your kind suggestions. Changes are made.
The CC can only reflect the correlation between two datasets, while the NSE also takes into account the differences of specific values. The MAE and RMSE metrics are used to quantify the discrepancies between simulated and observed values, and the abnormal values can also be detected based on the RMSE. We have clarified the reasons why four metrics, including CC, NSE, RMSE, and MAE are utilized in this study.
4. The lack of Discussion is the largest flaw of this manuscript. I suggest that authors should adjust structure as independent Results and Discussion. Authors add the discussion to explain the contribution and attribution of your article.
Responses: Thanks for your kind suggestions. Changes are made.
We have adjusted the structure as independent Results and Discussions based on the reviewer’s suggestions. Please see the modifications in sections 5.1 and 5.2 of the revised manuscript.
Author Response File: Author Response.pdf
Reviewer 3 Report
This manuscript estimated TWSA by using remotely-sensed precipitation and ET products, GLDAS-based runoff and the water balance method over 10 river basins in China from 2003 to 2020. My comments are presented in the following:
1. General comment: Unfortunately, there is no line number to make it easy present specific comments. Nevertheless, I tried to provide detailed comments.
2. General comment: What is the relationship between your work and the “Sustainability” journal? In other words, how did you choose this journal for submitting your manuscript? Clarify.
3. General comment: Add the spatial and temporal scale of your study to the abstract.
4. Abstract: I suggest removing abbreviations of phrases that only used once in the abstract from the text of the abstract. These abbreviations include NSE, RMSE, and MAE. Also, when you define WB, you should use it (refer to the last sentence of the abstract).
5. Abstract: What do you mean by “overall correlation coefficient”? DO you mean the average value or the minimum value of CC?
6. Abstract: What do you mean by “Moreover, the optimal solutions are provided based on different cases for these above 4 RBs, e.g., Liahe RB (Case 2), Haihe RB (Case 1), Yellow RB (Case 4), and Northwest RB (Case 5).” This sentence is not clear, while there is a gap between this specific sentence and last sentence of the abstract. So, I suggest removing it.
7. Abstract: Since you did not define what each case is in the abstract, a future reader may not understand what you mean by “Moreover, the optimal solutions are provided based on different cases for these above 4 RBs, e.g., Liahe RB (Case 2), Haihe RB (Case 1), Yellow RB (Case 4), and Northwest RB (Case 5).” So, it is suggested that you mention which case yielded the best estimation results.
8. Keywords: Use “river basins” instead of “10 river basins”.
9. Introduction: What is the difference between “TWS changes” and “TWSA”? If they are the same, why did you use two terms?
10. Introduction: This sentence requires citations: “To the best of our knowledge, few studies have been conducted to comprehensively evaluate the performance of WB-based TWSA estimates based on GRACE observations and GLDAS simulations.”
11. Introduction: Similar studies that “evaluated the performance of WB-based TWSA estimates based on GRACE observations and GLDAS simulations” should be reviewed.
12. Introduction: Each paragraph should focus on a single and clear topic. However, some paragraphs in the current introduction has more than one general topic. For example, check the second paragraph of the introduction section, which has several topics.
13. Introduction: There should be a rational consistency of materials between two successive paragraphs in the introduction. However, sometimes there is a gap between the second and third paragraphs of the introduction section.
14. Study area: Present this section as the first sub-section of methods and materials.
15. Figure 1: Each figure should be self-explanatory. However, this figure shows some regions with abbreviations that only introduced in the text (above the figure). I suggest adding the long version of each region to the legend, caption or on the figure itself. The current version of the figure is not appropriate.
16. Table 1: I suggest merging the information provided in Table 1 with Figure 1 by showing the area and full name of each region on the map or adding a new legend.
17. Datasets and Methods: Change the section title to “materials and methods”.
18. General comment: There are a lot of abbreviations in the text. So, I suggest adding a list of abbreviations to the article.
19. Table 2: Add year to the table.
20. Methods: Change the section title to “Methods for estimating TWSA”.
21. Eq. 1: Add an appropriate reference for each equation that is not from your work.
22. Mann-Kendall trend test algorithm: Add a recently-published reference to the following sentence (My suggestion: 10.1007/s00704-021-03837-0): “It has been widely used to detect trends in hydro-meteorological sequences.”
23. Evaluation index: Add an appropriate reference for the metrics you used (My suggestion: 10.1155/2021/5547889).
24. Figure 2: I suggest adding where you used Mann-Kendall trend test algorithm.
25. Figure 2: Improve the flowchart for your study by adding input data, and output data.
26. Figure 2: I suggest putting the definition of 12 cases out of the figure and present it table 2 and in the text as well.
27. Table 2: Add what which case refers to in the caption.
28. Figures 3-8: Add more descriptions for figures to the text.
29. General comment: I suggest discussing why some cases yielded good estimations and some did not. This needs to be discussed more in the discussion section.
30. Insights and limitations of this research: The following sentences do not belong to this sub-section and should move to its correct place: “For example, the optimal TWSA are detected in Case 2, Case 1, Case 4, and Case 5 for the LRB, HRB, YRB, and NWRB, respectively. Particularly, WB-based TWSA are unsatisfactory in 6 out of 10 RBs, and this situation is caused by multiple factors.”
31. Insights and limitations of this research: Add a few suggestions for future works in line with your article.
32. Discussion: Compare your findings with the similar studies that used satellite-based products for estimating TWSA.
33. Conclusions: This section should be improved by removing sentences that are not conclusion materials and adding appropriate sentences related to this section.
Author Response
This manuscript estimated TWSA by using remotely-sensed precipitation and ET products, GLDAS-based runoff and the water balance method over 10 river basins in China from 2003 to 2020. My comments are presented in the following:
Responses: Thanks for spending time on my manuscript. We have carefully considered every comment, and tried our best to make responses and modifications. We sincerely hope that our reply can make you satisfied.
Detailed responses to reviewer#1 comments can be seen in attached pdf file.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
The authors have addressed the earlier comments properly. I believe it has reached the quality required by the journal, and I recommend publishing it.
Author Response
Thanks for your approval of this manuscript. I greatly thank for your insightful suggestions.
Reviewer 3 Report
The manuscript has been improved significantly. My minor comments are presented in the following:
1. In your response, it was stated that “in your opinion, this manuscript is suitable for the topic of the Special Issue “Sustainable Assessment and Management of Groundwater Resources”. Thus, you decided to submit it to the “Sustainability” journal.” But, did you do any analysis/management on groundwater? Or, at least, did you consider or analysis groundwater data in your work? Clarify in the introduction why this study may be related to the theme of the “Sustainability” journal.
2. Add the difference between “TWS changes” and “TWSA” to the text.
3. Add a new figure or a table to show input and output data used in your study.
4. The gridded runoff data were produced by the GLDAS Noahv2.1 model and not evaluated against in-situ measurements. First, why did not you compare the simulated values with the observed ones? Also, is this the reason why some cases yielded not good estimations. Clarify in the text.
5. Conclusion: The last paragraph does not contain conclusion materials. Remove it and add a take-home message to the end of the conclusions.
Author Response
The responses to the reviewer's comments have been shown in the attached pdf file.
Author Response File: Author Response.pdf
