Daily Streamflow Forecasting Based on the Hybrid Particle Swarm Optimization and Long Short-Term Memory Model in the Orontes Basin
Round 1
Reviewer 1 Report
Dear Author,
The submitted manuscript titled „Time Series Prediction Based on the Hybrid Particle Swarm Optimization and Long Short-Term Memory Model” contains interesting results. However I have found some mistakes, which I have listed below.
- The background of investigations is sufficiently However the goal 1-in my opinion- should be briefly stated. The sentence” .. Three flow measurement stations 126 representing different hydrological conditions of the Orontes River Basin were chosen to validate the predictive capacity of the generated model” is suitable for chapter „Material and method”. Moreover, lines 132-138 might be removed. In my opinion there is no necessity to describe the paper organisation.
- The Results section is very long. Perhaps it might be divided into two subchapters (e.g. according to goals).
- The obtained findings should be compared with literature od subject in chapter Discussion.
- The chapter Conclusions is written as Summary. I suggest to pointed out the meaning and novelty of obtained findings in context of current state of knowledge and indicate the furter directions of investigations.
Author Response
''Please see the attachment''
Author Response File: Author Response.docx
Reviewer 2 Report
I do not formulate any substantive objections to the submitted article. However, I believe that the authors can provide more information and suggestions on the practical application of the flow model they have developed. Its value does not only result from its scientific excellence, but above all from its applicability. In this regard, it should be indicated what conditions must be met in order to use it effectively and efficiently. Whether it is expensive or cheap to use. Who should use it and in what situations etc.
Author Response
''Please see the attachment''
Author Response File: Author Response.docx
Reviewer 3 Report
The manuscript entitled “Time Series Prediction Based on the Hybrid Particle Swarm Optimization and Long Short-Term Memory Model” is interesting, well organized, and well written. The methodology is clearly explained. Many modeling details have been provided. However, the manuscript needs significant improvement, especially with regards to results and discussion, before being further evaluated.
Therefore, I would suggest the author to carefully consider the following comments:
- The literature review should be improved, considering a larger number of studies on the prediction of time series of hydrological quantities based on the use of RNN.
- The choice of the case study should be properly justified.
- The results should be discussed in more detail regarding the ability of the models to represent extreme values, regarding any tendency to over or underestimate in some conditions, and regarding the presence of outliers.
- The author should discuss the possible presence of a bias in any of the considered models. A representation with box plots might help.
- Further comparison of the models using the Taylor diagram would be effective.
- A comparison with a "more classic" model of time series analysis (e.g. ARIMA) is strongly recommended.
- The case study should be mentioned in the title of the article.
To address the previous comments, and also to enhance the literature review, the following references could be considered:
Jabbari, A., & Bae, D. H. (2018). Application of Artificial Neural Networks for accuracy enhancements of real-time flood forecasting in the Imjin basin. Water, 10(11), 1626.
Mohammadi, B., Ahmadi, F., Mehdizadeh, S., Guan, Y., Pham, Q. B., Linh, N. T. T., & Tri, D. Q. (2020). Developing novel robust models to improve the accuracy of daily streamflow modeling. Water Resources Management, 34(10), 3387-3409.
Chen, Z., Zhu, Z., Jiang, H., & Sun, S. (2020). Estimating daily reference evapotranspiration based on limited meteorological data using deep learning and classical machine learning methods. Journal of Hydrology, 591, 125286.
Di Nunno, F., Granata, F., Gargano, R., & de Marinis, G. (2021). Prediction of spring flows using nonlinear autoregressive exogenous (NARX) neural network models. Environmental Monitoring and Assessment, 193(6), 1-17.
Granata, F., & Di Nunno, F. (2021). Forecasting evapotranspiration in different climates using ensembles of recurrent neural networks. Agricultural Water Management, 255, 107040.
Author Response
''Please see the attachment''
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
The author has significantly improved the manuscript, which is almost ready for publication. There are still minor grammatical errors to be corrected. It is also advisable to check that all the references cited are present in the list and vice versa.
Author Response
''Please see the attachment''
Author Response File: Author Response.docx