This paper aims to create a precise bathymetric model for the Philippine Sea. It introduces the Improved Gravity-Geologic Method (IGGM) to enhance accuracy, considering regional seafloor topography effects. Shipborne bathymetric data and SIO V32.1 gravity anomaly are used for modeling. The paper introduces a compelling method, the IGGM method, for inverting the bathymetric model of the Philippine Sea, a region characterized by a complex marine environment and significant strategic importance. I recommend a moderate revision and the English expression needs to be polished. Please find detailed comments on the current MS below.

Specific Comments

1. Line30: why are there two trench-like terrains at (140°E-150°E, 30°N-35°N) and (125°E-127°E, 27°N-30°N)? This is clearly not true.
2. Line 43, please accurately list several GEBCO models.
3. Line 126, in Eq. (2), how are parameters Δρ and D obtained?
4. Line 139, in Eq. (3), what does the letter "k" represent here? What is the influence of this parameter on the accuracy of bathymetric model inversion?
5. The elimination rate of data in Table 1 exceeds 1%, which will have an impact on the results of submarine terrain inversion.
6. When calculating the short-wave gravity at any point i, the long-wave gravity is derived from the long-wave gravity field, calculated using the tension spline function, and subjected to cubic spline interpolation. This entire calculation process inevitably introduces calculation errors. Does the paper address any relevant methods to mitigate or handle these errors?
7. Fig7: The image is low resolution and I can't see the distribution of the black pints.
8. Fig9 (b) and (c): Unlike Fig8(c), the comparison between the GEBCO model and the GGM/IGGM model should be described as the absolute value of the difference.
9. For the BAT_PS model and other reference models, there are still very large errors at some ship measurement points (Table 5). Can shipborne bathymetric data be used to further improve the accuracy of the model? For example, the difference at the ship measurement point can be added to the BAT model as a correction.
10. Is it reasonable to determine the computational radius to be 2′ for the entire Philippine Sea area? In general, re-selecting the optimal value for each sub-region is required to obtain a locally optimal model.
11. Line 287: How are the boundary points of each region treated? How are the subregions stitched together?

Technical corrections

1. In line 47, the sentence "as compared to other observation methods" can be changed to "compared to alternative observation methods".
2. In line 59, the first occurrence of 'SIO' should be provided in full.
3. In lines 74-77, the sentence is too long to convey its meaning accurately.
4. In line 100, please provide an accurate explanation for '3b' to ensure understanding for all readers.
5. Line105： “is a continuous global terrain model”.
6. Figure 3 needs to be redrawn, the text in the figure is not clear, and the flowchart logic is confused.
7. Line218：Does “the total number of surrounding points within a 2′ radius centered on each shipborne point” include the centre point? The minimum value of colorbar in Fig7 starts at 2.
8. Check the formatting of references, e.g. Lines 394, 402, 429, 437, 460.
9. Line 153: “value” should be “values”.
10. Line345： “were not utilized” should be deleted.

See the attachment for comments