This manuscript presents a well-executed study on regional wave simulations along the Australian coast using the WAVEWATCH III (WW3) model, with particular emphasis on improving model performance over the Great Barrier Reef (GBR) through a two-step subgrid parameterization approach. The study is timely and relevant, demonstrating both technical rigor and practical applicability in modeling ocean waves in complex coastal environments.

I believe this manuscript is suitable for publication after minor revisions. I recommend the authors consider the following suggestions to further enhance the clarity and completeness of the work:

1. Brief description of the prevailing wave conditions along the northeastern Australian coast would help contextualize the results. Based on the manuscript, the region appears to be predominantly swell-influenced. It would be helpful to comment on model performance under wind-sea–dominated conditions.
2. The authors have carefully included several important coastal processes, such as tidal variations and wave–current interactions. However, the representation of wind stress could be further improved. The drag coefficient used in ST6 is originally developed for open-ocean conditions; a brief discussion of its applicability and limitations in reef-dense or shallow water environments would be valuable.
3. While the manuscript addresses the challenges of wave modeling over complex reef geometries, it would strengthen the discussion to acknowledge potential limitations in applying the UOST scheme to other reef-rich coastal systems globally.

The authors present a numerical simulation study of ocean waves in the Great Barrier Reef (GBR) region of Australia. This research incorporates state-of-the-art physics and numerical schemes, with principal methodologies comprising: implementation of an unstructured mesh to accurately resolve Australia's extensive coastline, and development of a two-step modeling strategy to address unresolved individual coral reefs and their dissipative effects. This two-step strategy treats discrete reefs as unresolved obstacles that act as complete barriers to wave energy, and parameterizes subgrid-scale reef-induced dissipation through a novel source term to represent the effects of unresolved obstacles. Critically, the experimental design features comprehensive controls: appropriate specification of open boundary conditions; isolation of interference from wind forcing errors, tidal currents, and surface circulation on GBR wavefield simulations. Validation against satellite altimetry and buoy observations demonstrates reduction of wave height bias from >100% to <20%, providing compelling evidence of substantially enhanced model performance following strategy implementation.

This methodology offers valuable insights for simulating wave fields in archipelago-fringed marginal and regional seas where dense reef systems exist. I recommend acceptance of this work for publication in your esteemed journal. However, I think a minor revision is necessary before the acceptance. Below are a few comments and suggestions for the authors’ consideration.

    1. The coral reef location data used in this study were extracted from the global dataset provided by UNEP-WCMC et al. (2010), which defines the outer polygons of reef structures. These polygons may represent reef platforms rather than the actual outlines of reef canopies, and such discrepancies could introduce uncertainties into the model results. A comment on the possible effect of the accuracy of the reef outline on the model performance is desired in the discussion or conclusion section. 

    2. The application of UOST has clearly improved the model’s performance in simulating significant wave height and peak period in the Great Barrier Reef region. However, the simulated values of T02 are underestimated. This may be related to the choice of the empirical coefficient ψ in Eq. (14). I would appreciate if the authors could stress the limitation of this engineering scaling on the model results. 

    3. In line 551 of the Appendix, the authors mention a comparison between model results and satellite altimeter data before and after accounting for tidal effects (Run5 vs. Run8). However, the results of this comparison are not presented. It is recommended that the authors include these results, as they could provide useful reference for future research.