Coupling a Parametric Wave Solver into a Hydrodynamic Circulation Model to Improve Efficiency of Nested Estuarine Storm Surge Predictions

Round 1

Reviewer 1 Report

This work presents some results of a coupled numerical model (ADCIRC+a wave model from the authors). The manuscript is well structured and very well written. The research topic is interesting for a general audience but probably not so much for scientific modellers. This reviewer has some concerns that may help the authors improve the presentation of their work.

- The introduction seems too basic; maybe the authors would prefer reviewing the current state of the art in wave propagation numerical models to set a framework for their work.

- The manuscript lacks an accurate description of the study site. No georeferenced images are provided, nor an image in which the reader can foresee the expected hydrodynamics.

- This reviewer would encourage the authors to justify using such a simple wave model; computing time is a valid argument but somehow weak.

- The manuscript seems to state that the coupled model ADCparam can be used in any water depth and domain. The authors should be more precise in limiting the model's applicability, which, due to its governing equations, cannot provide accurate results in shallow waters (lower than 20 m depth) and on complex domains such as steep bottoms and in the presence of obstacles. In turn, this coupled model cannot accurately solve phenomena such as wave diffraction or reflection.

- The preliminary validation is performed only over first-order variables, i.e. elevation, wave height and wave period. In estuaries, the velocity fields are probably more important. Authors should, at least, provide some comment regarding the physics of estuaries.

The above concerns do not demerit the great work performed by the authors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The study presented an approach to using a parametric wave model and implemented it in the ADCIRC storm surge model to improve the efficiency of storm surge prediction. The study built on the previous study results and presented a case study to demonstrate the model's capability. The authors did a comparison of using the parametric model against the results using the SWAN model. The results suggest that it has sufficient accuracy to simulate storm surges combining wind-wave effects while shortening simulation time. Overall, the paper is well written and includes all the details. I only have a couple of minor comments.

The authors presented 4 parametric wave models and used the assembly of these models to be input for the storm surge model. It will be good to provide some explanations of the difference among these models. It will be great if a comparison of these models for a specific case can be presented. It will help readers to get an idea of the differences between these models.  

The author indicates that the wave is input to ADCIRC to compute radiation stress gradients. It will be good to show the equations (or provide references). It will help readers to see how important the wave effect is.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Review of "Coupling a Parametric Wave Solver into A Hydrodynamic Circulation Model to Improve Efficiency of Nested Estuarine Storm Surge Predictions"

In this manuscript, the authors seek to improve the computational efficiency of storm surge modelling in estuarine environments by substituting use of SWAN with a "parametric wave solver" (PARAM) in the ADCIRC numerical circulation model. This is subsequently integrated into an ensemble modeling system previously reported on by the second author.
The principal idea behind PARAM is replacing costly computations of the wave field via SWAN by fetch-dependent expressions for significant wave-height and peak period based on the TMA spectrum, the Bretschneider spectrum, and others. This simplification leads to a significant reduction of over 50% in computational time versus using SWAN in ADCIRC, and to a 33% reduction when this is implemented in the ensemble modeling system. The expected discrepancies are mostly observed in the wave height, while water level and period show reasonable agreement.
Overall this is an interesting study, and I am happy to recommend publication in the Journal of Marine Science and Engineering. I have some minor comments below which should be addressed. I also believe that the community could see wider benefit from making the source code openly available, e.g. via GitHub, but realise this may not be easily possible/desirable.

Minor issues:

1) Some of the citations which appear in the text do not seem to be in the list of references. E.g. line 159, the citation to Taeb et al 2019, or line 180 Malhotra and Fonseca. These should be checked carefully.

2) The authors state (line 208ff) that the CEM formulation assumes deep-water conditions, and further (line 814ff) that it appears to degrade the ensemble average. Please clarify why it is included.

3) Line 266, 282  "coarse" is misspelled.

4) Line 380, 474, 475: "ran" -> "run"

5) Line 433: please explain why a constant Manning's n friction coefficient is employed, given unique values are provided by ADCIRC.

6) Line 616: "The" -> "the"

7) Line 680: this line gives the impression that only eq. 5, from the SPM, is used for the spectral peak. However in line 184 it is stated that an average of eqns. 5-8 is employed. Please clarify.

8) The appendix contains no text and the figures have no captions. In fact, the appendix is only referred to once in the main text on line 688. To assist the reader, please add descriptions to tie the appendix to the remaining text.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I appreciate that all of my concerns were accurately addressed. I recommend the manuscript for publication.