General Comments

Overall, “Decadal sea level variability in the Australasian Mediterranean Sea” is very well constructed, explained, and is an important contribution to our understanding of how climate variability impacts regional sea level. Specifically, the authors use an ocean model to determine how ENSO and PDO impact decadal regional sea-level variability in the Australasian Mediterranean Sea (AMS), determine the contributions of momentum and buoyancy fluxes to sea-surface height (SSH) variability, and further disentangle the steric variability into thermo and saline components. I do not see any major flaws in methodology or reasoning, and as such, recommend acceptance after addressing some minor issues mainly related to presentation.

Specific Comments

1. It would be very helpful to include an overview map of the study area, with the different seas and currents discussed clearly marked. The labels in Fig. 1a were helpful but not sufficient in this regard.
2. Some of the markings on the figures are difficult to see, e.g. the labels and contours in Fig. 1, current vectors in Fig. 6. Switching grid colors to black and others to white may be one way to remedy this.
3. There are some repeated panels between figures. Perhaps some of the redundancy can be removed (though admittedly, I don’t see an immediately obvious way). If not, please clearly mark where repeated. 
4. As you indicate they are not the same, what time step do the coarse and nested grid run at? 
5. I’m glad you talk about the uncertainty in the atmospheric forcing product at the end. Perhaps you could add a justification as to why JRA55-do was chosen relative to another reanalysis product?
6. I agree that the linear separation of WIND and BUOY is justified on the spatiotemporal scales you are looking at, but a further note clarifying what scales you expect this to break down would be useful to a more general readership (which this manuscript attracts, through the implications for sea-level projections).
7. Along these lines, I find the CLIM experiment quite interesting, and a nice way to capture the remaining variability. What’s not clear to me is the relative role of the seasonal cycle vs nonlinear interactions in ‘intrinsic variability. The discussion of eddy kinetic energy seems useful to this end, and I wonder if it could be quantified in the important regions (e.g., the central South China Sea). I freely admit that this discussion is outside my area of expertise, so apologies if this is not a well-posed comment.
8. Can you elaborate on the methodology around line 175? Why did you not just calculate the variability contribution from BUOY? If you do that (assuming you can), does it compare well with the REF025-WIND contribution?
9. In your concluding sentence, you comment on the resolution of coupled circulation models. My initial thought would be that yes, coupling would indeed be important for sea-level projections as ENSO/PDO are subject to change, but that your findings would be largely robust. A comment to this effect would be good to see.
   • It would also be nice to wrap the paper up on a positive note, rather than the warning.
     
     


Technical Corrections

• When used as an adjective, e.g. "sea-level variability", sea level should be hyphenated (sea-level).
• In the first paragraph of the results, references to Fig. 1 are incorrectly marked as Fig. 3.
• On line 178, I think you mean sea-level variability (not just sea level).
• L. 247, missing parenthesis before Fig 3.

This is a nice modeling study of the decadal variations of the sea-level variability in the Australasian Mediterranean Sea. The study identified ENSO and PDO as the dominant climate drivers of the sea level variability in the marginal seas in this complex region, predominantly due to wind forcing and planetary wave propagations. The study also pointed out that buoyancy forcing are important in certain regions in the AMS. The manuscript is well structured so I suggest publications after some revisions. 

I wonder if it is necessary to use the PDO index, as suggested by the authors, there are little differences in the result when just using the ENSO index. Maybe just use two frequency bands of the ENSO index.

As I understand both the ENSO and PDO indices are derived from the model results. Maybe it is necessary to show how the modeled indices compare with observations. It is not explained how the PDO index is derived.

Do different flavors of ENSO have an influence? Such as the so-called central Pacific ENSO.

Does the Indian Ocean have a decadal mode that could influence the region through oceanic Kelvin wave propagation or some teleconnection? 

In Results, the first paragraph refers to Fig. 3, which in fact should be Fig. 1.

Line 198: "off the northwest coast of Australia" - it is not clear what the authors refer to in the figure.

Line 236: fix the sentence.

Fig.7: the legend doesn't match the caption. 

Line 251: explain the meaning of the "advective nature".

The article Decadal sea level variability in the Australasian Mediterranean Sea aims to disentangle the sea-level variability in the Australasian Mediterranean Sea, by means of ocean model experiments. Through sensitivity experiments, the authors are able to see how wind and buoyancy fluxes contribute to the sea-level variability in the region. Furthermore, they use climate indices (such as ENSO and PDO), to see how atmospheric-ocean events forces sea level. Finally, they look at the influence of temperature and salinity (thermo and halosteric) on sea-level change. They find that ENSO explains most of the decadal variability in the region (with exception of the South China Sea, which they hypothesize to be connected to intrinsic variability and the Kuroshio Current). PDO events seems to have a minor effect in the region.

The paper is relevant to the scientific community and doesn’t present any major flaws. I recommend the article to be accepted, given some minor corrections.

General comments:

1. A study area map, with the name of the main places (which come up a lot during the text) would help to locate the reader. Just as presented in Figure 1 is not enough.
2. On line 89-91, the authors present the different nomenclatures for the region, and explain why they use ‘Australasian Mediterranean Sea’. I must admit that I had never heard of such nomenclature, and found the word ‘Mediterranean’ in specific a bit puzzling at first. I wonder if using a more common nomenclature for the region, such as ‘Tropical Asian Seas’ and ‘Southeastern Asian Sea’, won’t attract more readers.
3. The results section is relatively long, and includes different things. I recommend the authors to create subsections to guide the reader better. For example, on Line 157 and L221 would be good places to start subsections.
4. Section 4 (Summary and conclusion), is more of a discussion section. The main findings are very clear in the abstract, but not so much in Section 4 (which is very long for a summary/conclusion). Some re-structuring of this Section would improve the text.
5. The final ‘general’ conclusion of the paper is about the models’ resolutions (from L332). However, while the effect of the resolution is clear for the temperature and salinity, it didn’t seem to me so strong on the SSH variability itself. I don’t think that these results have enough evidence to back the statements on L335,336.
6. The observational dataset and variables used should be presented before the Results Section. Furthermore, it’s important to mention that you use altimetry observations to validate the model, and that they are only available from 1993 (in contrast with your model that runs since 1958). Also how was the thermosteric and halosteric components computed? Or are they output of the model? This should be clarified. Also, where did the ENSO and PDO indices come from (from figure A1 it seems you computed it based on the model results, so would be good to say how you computed them. And does it match with the observations indices?)
7. Did you do any analysis looking at ‘specific’ ENSO and PDO events? Does the explained variability change regarding an El Niño instead of a La Niña? Or does the position of the ENSO have an influence on your results? Maybe such analysis would enrich the paper even more.

Technical comments:

1. Sea level should be hyphened when used as an adjective (sea-level change Vs. change in sea level).
2. Verbal times: In the introduction the authors use present (‘’the objectives are “ (L75)); they use past tense in Section 2 (‘we used a global ocean model” (L97)); future tense when presenting what is showed in the results (“First we will compare” (L139)); and back to present tense in Section 4. I recommend the authors to either keep it constant (i.e., using always present or past tense), or change the tenses accordingly to the Section, but keep it constant throughout the Section (i.e., present in the introduction, past in the methods (since it’s something they already did), and present in the results and discussion, for example).
3. The terms ‘i.e.’ and ‘e.g.’ are usually followed by a comma.
4. Use of one-sentence paragraphs (for example, L258-259, L289-290) or very short paragraphs (e.g., L71-73, L84-86) should be avoided.

Please find my specific/minor comments on the attached file.