The manuscript authored by Yuan et al studies rip current variability, real-time rip hazard identification, and the optimal swimmer escape strategies at Dadonghai Beach, China considering two bathymetric scenarios between 2018 and 2019.

Overall, the document is very well structured, details each section adequately, and is written in fluent English.

General comments are indicated below and more detailed comments are in the attached document:

• How accurate are the bathymetries derived from satellite imagery concerning measurements? Have any checks been made in the study case?
• Is the beach in dynamic equilibrium, preserving sediment balance? Are the scenarios analyzed from the 2018 and 2019 satellite-derived bathymetries expected to be representative of all summers?
• The model used to obtain the currents has been validated with field measurements? this is essential to validate the directions and magnitudes of the velocities.
• Where is the wave point located? It is essential to know its location in order to describe the wave conditions.

The paper presents a Boussinesq model study of rip current variability and the evaluation of hazards induced by rip currents. The study showcases the trajectory tracking method used in an embayed beach area for swimmer escape strategies. It is an interesting study.

Because the first author was the developer of FUNWAVE-GPU, I don’t have many concerns about modeling details. However, I feel it may need logical consistency across the entire article, from the introduction to the conclusions. The authors expressed the importance of a wave-resolving model for such a trajectory-tracking study due to the random and dynamic nature of rip currents. But the paper concluded that the results from the Boussinesq model are comparable to that from the wave-averaged method. If the swimmer escape strategies are the main objective of the study, using the expensive Boussinesq model seems to be an overkill. For this reason, I suggest that the authors may emphasize more about the effects of IG and VLF bands because a regular wave averaged model cannot predict IG motions without using a non-stationary wave condition. The VLF motions in the model results are interesting. It is good to check if those are the shear wave mode as said or long gravity wave oscillations. A plot of a wavenumber spectrum would be helpful.       

In general, the paper was well-written and easy to follow. It should be published after addressing the issue mentioned above.

Details:

Figure 1(f), please provide the vertical datum for 2018 and 2019 measurements. Provide an accuracy estimate for the satellite-derived bathymetric data.

How does a periodic boundary condition set up in such a bay-like domain?

Line 112, both ends is exposed, grammar

Line 139, FUNWAVE-GPU add the reference here

Line 142, CFL, add the complete terminology

Line 144, friction coefficient, need to clarify bottom friction form, manning formula? or provide a reference

Section 3.3. It’s interesting to make a definition for the hazard levels. Any reference for this definition, or just created by the authors?

Line 181, arbitrary factor of 0.8.  need a sensitivity test on this number