A Hybrid Coupled Model for Groundwater-Level Simulation and Prediction: A Case Study of Yancheng City in Eastern China
Round 1
Reviewer 1 Report
Please find the attached pdf for detailed comments and/or suggestions.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
This is the review comments on the manuscript “Construction of Prediction Model of Groundwater Level Spatio-temporal Series Based on BP-STARMA Coupling Model” by Hou et al. submitted to Water for publication. The paper tries to address the efficiency of a developed coupled model for predicting groundwater levels in space and time. Slight differences between the original models and the developed model have been presented. The study may be of interest to the readers of Water but the quality is still of concern regarding the language and the literature review, as well as the demonstration of the outstanding advantage of recommending the developed method. I suggest major revisions before submitting it for a re-consideration.
Several specific comments:
1. The title contains the name of a very specific model (BP-STARMA), which is not recommended. The term “construction of prediction model…based on … model” is also redundant.
2. Statements on the background of the paper, such as ground subsidence and ground cracks are not sufficient. The authors have not done a good literature review.
3. Language issue. Too many vague and irrelevant descriptions, such as “we can grasp the dynamic changes…”. The paper needs to be edited by a native speaker.
4. The BP and STARMA models need to be carefully defined at the beginning of the abstract. In the introduction as well, there is no rigorous definition of these abbreviations.
5. The four error criteria do not necessarily need to be listed in the abstract. They essentially play the same role in the paper.
6. The term “spatio-temporal prediction” is almost redundant. The authors can make a note of the spatiotemporal freatures and then simply mention their work as a “prediction”.
7. The paper does not work on geographic phenomena. Instead, it can be described as geological.
8. The selection of the BP, STARMA, as well as the coupled version, has not been fully addressed as a research need.
9. Please remove some of the figures that look similar and cannot provide straightforward information to the readers. For example, Figures 5, 6, 7, and 8. They look tedious.
10. Figure 4 has a subplot in another language rather than in English.
11. Please review other alternative methods that are not used in this paper, and demonstrate the necessity of using the selected approaches.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
I have no more comments and recommend accepting the manuscript for publication.
