Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript entitled “MNAT: A Simulation Tool for Underwater Radiated Noise” presents the design and implementation of a web-based platform for real-time simulation and prediction of underwater radiated noise. The system integrates AIS data, environmental databases, and classical acoustic propagation models (Bellhop and RAM), complemented by machine learning approaches for data cleaning, vessel type prediction, trajectory prediction, and noise interpolation. Two case studies (shipping noise propagation in the Bay of Biscay and Adriatic Sea, and seismic airgun noise in the Mediterranean) illustrate the applicability of the tool.
The topic is timely and important, as underwater noise pollution is increasingly recognized as a major environmental stressor. The manuscript is clearly structured and contains a comprehensive review of relevant literature. The tool’s open-source availability is another strength that enhances reproducibility. The strengths of this manuscript are as follows:

1. Relevance and novelty: The paper addresses an urgent issue of marine noise pollution and provides a practical tool for both technical and ecological applications.
2. Methodological breadth: The integration of AIS preprocessing, noise source modeling, acoustic propagation, and machine learning is well-conceived and technically robust.
3. Practical value: The case studies convincingly demonstrate that MNAT can generate realistic and meaningful noise distribution maps under different conditions.
4. Contribution to community: The open-source release is commendable and ensures transparency and reproducibility.

The aspects of this manuscript that warrant improvement are as follows:

1. Innovation and comparison with existing tools: Although the manuscript reviews other platforms (e.g., dBSea, MANTA, AcTUP), the comparative analysis is largely qualitative. It remains unclear in which specific aspects MNAT outperforms prior tools (e.g., speed, scalability, usability).
2. Methodological details: For the stacking regression model in noise prediction, the training dataset size, generalization ability, and limitations across different marine regions require further discussion.  Recommendation: Provide a comparative table or quantitative benchmarks to highlight MNAT’s improvements over existing solutions.
3. Validation and evaluation: The current evaluation mainly compares ML predictions with RAM/Bellhop outputs. However, no validation against in-situ acoustic measurements is reported.
   Recommendation: If observational data are unavailable, explicitly acknowledge this limitation and discuss possible validation strategies.

The manuscript makes a valuable contribution to the field of marine acoustics and underwater noise modeling. However, the issues outlined above—particularly regarding comparative validation, methodological detail—should be addressed before the paper can be considered for publication.

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

please find our reply to your comments in the attached PDF file.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

1. It would be nicer to broadly introduce about shipborne noise. Authors can refer to "https://doi.org/10.1016/j.apor.2024.104195".
2. There are lots of abbreviations. Authors can make nomenclature table.
3. The contents of Section 2 can be compressed.
4. Authors should add arrow direction and distance scale for figure 2 with brief explanation of black (ship) and red line.
5. Authors should mention for the detail of dataset including its size.
6. Authors should mention all softwares used and their versions.
7. For figure 5, authors can remove the line below "B".
8. It seems authors did not use neural network. Is there any reason to not use it?
9. Private question. What software that authors used for Figure 10? Especially worldmap and the dense trafiic lines.
10. It would be nicer to add comma(,) for every thousand points. [Ex) 1,574]
11. There should be math equation for all performance metrics.
12. The font size could be enlarged for figures 13 and 21.
13. Overall it has a high quality contents. Minor revision for it.

Author Response

Dear reviewer,

please find our answers to your comments in the attached PDF file.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper presents the design and functionalities of Marine Noise Analysis Tool (MNAT), including the design process and the optimizations applied to improve the speed of acoustic propagation modeling over basin- and sea-level scales. The tool is not tailored to just assess the impact of noise on communication systems, but rather it can help assess noise in a variety of contexts. These include, e.g., evaluating the average insonification of different sea areas at various frequencies and possibly at different times of day, to assess how various marine fauna species are affected by radiated noise.

The shortcomings and missing of the paper are the following:

1. The abbreviations SQL (Line 287) and ITU (Line 288) must be deciphered.
2. (Line 470): Mathematical formulae for “mean absolute error (MAE), root mean squared error (RMSE), and the R-squared (R²)” should be presented with corresponding links.
3. Formula (2): Arc (p) should be replaced by Arc (k), because the summation is performed on k.
4. Formula (3): n should be replaced by k according to Formula (2).
5. Figure 13: Inscriptions should be bigger.
6. (Line 577 – 587): The links for the used methods should be pointed.
7. Figure 17 should be bigger.
8. (Line 680): What does the sign @ mean?

Author Response

Dear reviewer,

please find our answers to your comments in the attached PDF file.

Author Response File: Author Response.pdf