This paper explores the relationship between the varying eddy field and the changes in the shape and depth of the Beaufort Gyre halocline in the past two decades. The authors used in-situ observations, altimetry, and reanalysis data, to describe the changes in the halocline structure, with an emphasis on the meridional asymmetry. They then examined eddy activities, from both individual eddies and kinetic energy perspectives, and connect them to halocline structure via available potential energy. The study found that eddies played an important role in redistributing fresh water and thus adjusted the halocline structure through analysis of eddy fluxes. Despite the paper’s intriguing ideas, the writing is unclear, making it very difficult to follow. The overall flow needs improvements. Therefore, I cannot recommend this paper to be published in its current form.

Minor points:

Line 132: Could the authors explain how eq (3) relates to this sentence?

Line 146: This is not the right place to insert citation.

Line 205-206: Mooring C ends before 2008. How can we make a conclusion about the shape in the northeast and northwest in recent years?

Figure 3: APE at a single point is not meaningful. It represents the slope of the halocline in a region, not a single point. The reference density in the APE equation is related to some mean density across the field. Authors should make it clear.

Figure 6: The lengths of data available are determined the days of open ocean. Will different data lengths impact the results?

Figure 6: The plot is from satellite data, so I assume it indicates surface EKE. How is it related to the previous paragraphs in which subsurface EKE are discussed?

Line 306: This statement made based on observations from two points. Other factors could be responsible. For example, eddies were generated near the mooring site; or the eddies simply did not pass by the mooring.

Figure 10: How is the probability estimated?

Line 399: Eady timescale is not interpreted previously.

Major point:

The authors described both surface and subsurface EKE. There could be asymmetry in the surface sea level and subsurface halocline. How does the surface EKE relate to the subsurface halocline shape?

To show how the writing could be improved, I will provide some wording issues in the first 50 lines. Please not that this list is not exhaustive, but rather serves as a guide for improving the paper’s overall clarity.

Line 10: Please specify the “varying eddies”. Number, shape, velocity?

Line 16: This long sentence is confusing. Maybe it would be clearer if breaking into several shorter sentence.

Line 16: The second “Eddy” should not start with capital letter.

Line 30: This sentence is not grammatically correct.

Line 37: It should be “paid attention” instead of “payed attention”

Line 39: “The increase of isopycnal slop with depth” is a new piece of information and it should be introduced before directly describing the mechanism.

Line 42: “Likewise” used here is confusing. The previous and this sentence do not have characteristics in common.

Line 43: Should be “identified as”.

Line 46: high resolution eddy-resolving?

Line 47: “etc” should not be used.

Line 48: “focused on” instead of “focus”.

Line 48: “Moreover” is not correctly used here. The following sentence changes the topic.

Line 50: “even they may extend”?

Review of “Long-term eddy modulation inhibited the meridional asymmetry of halocline in the Beaufort Gyre” by Lu et al.

This paper uses a combination of satellite, in-situ and reanalysis data to investigate changes to the structure of the halocline in the Canada Basin over time. The paper begins by looking at changes in halocline depth, halocline thickness, and available potential energy from CTDs and moorings in the region to determine three periods of distinct behaviour. It then analyses both EKE and individual eddy properties at the four moorings and how they have varied over the three periods, before zooming in to a small region that experiences significant changes to EKE over the time period. The paper ends with an analysis of an eddy streamfunction for each period and relates it to salinity changes to explain the differences in the halocline between each period.

The paper has a lot of information contained within it, but I found it very hard to follow in many places. Some of the figures had long descriptions of details without reference to the relevant subpanel or feature being described, and the main take-home message of what the reader is meant to gain from the figure is often missing. This makes it difficult to understand the context. There are also a number of typos and grammatical errors which need to be corrected – I have not addressed them in the comments below but the manuscript should be checked thoroughly and rephrased in a number of places. I appreciated the instances where the authors stated the question they were addressing in the upcoming subsection, and feel it would be beneficial to do this much more often in the text to help the reader. I do not think the paper is publishable in its current form, but do think there are some interesting ideas that could be of interest to the community if they were presented in a more coherent way.

I have some points to consider based on the paper in its current form, which I feel should be addressed before the paper is resubmitted.

• The paper aims to understand why the BG has stabilised in recent years. But the asymmetry that is described is based on a) moorings in the south versus moorings in the north of the basin (which are located at around 74-75N and 78N respectively, based on Figure 1b), and b) CTD stations which head north along a transect to 79N. The paper refers to Bertosio et al (2022) and Regan et al (2019) when describing gyre asymmetry and a northward expansion, and also notes a shift in the BG found by Moore et al. (2018). Given this acknowledgement that the gyre is not always in the same place or has the same size/strength, it is surprising that there is no discussion of how choosing a static section or static moorings that are limited in their northward extent could affect the perceived loss of asymmetry of the gyre. The gyre has deepened in the portion of the basin captured by the observations, but the deepest part of the section has also moved further north – so the sloping isopycnals are likely now further north than the section shows. It is not known from this data whether they are steeper or flatter than before, but I feel this should be acknowledged in the paper (does the SODA data show this if you look further north?)
• The discussion around EKE and individual eddies is unclear. In particular, surface and subsurface (from 50m down) EKE seem to be shown on the same plot, and surface-generated EKE from wind input is linked to gyre stability (e.g. lines 347-349) even though the eddies associated with baroclinic instability are generated in the halocline. Care should be taken when associating these.
• In terms of structure, the flow is broken after Figure 5 (Mooring-based EKE and eddy counts) to Figure 6 (maps of EKE from other datasets), then back to eddies from moorings in Figure 7, then back to the maps to identify a key region to zoom in on for Figure 8. Is there a reason that we jump between datasets and region size? It might flow better if all of the information from moorings was put together. I found the jumps from Figure 6 to 7, then 7 to 8, quite confusing, so maybe that would help to make it clearer
• Introduction: there is a lot of information that has not been fully synthesised. In particular, the second paragraph is very long and detailed with a lot of different threads. I would suggest splitting into multiple paragraphs, perhaps one describing the vertical structure and one based on eddies. In general, there is a large amount of information on eddies in the introduction paragraphs – perhaps it can be streamlined, or reordered to group similar themes together.

Line 42: The Pacific Winter Water layer is mentioned without describing how it fits into the vertical structure. It should be introduced first

Line 50: What depth range is meant by “subsurface”?

Line 132: is (3) the correct equation reference here? It hasn’t yet been introduced

Line 155: Simth, 2007 should be Smith, 2007. Eddies are only a part of the EKE which also includes deviations from a mean current. How much EKE is not attributable to eddies? I.e. how much is not due to eddy genesis? That might affect the assumptiion that it is correlated with baroclinic growth rate. How much EKE do you miss by only having SODA at ½ degree resolution?

Lines 160-161: Section 3 talks about the asymmetry of the halocline being the focus of the article, but this was only mentioned briefly amongst all of the text about eddies. I understand that EKE is being investigated to explain the asymmetry, but feel the asymmetry needs to be introduced more thoroughly first – why do we care that it’s asymmetric or not?

Line 169: I think by “void measurements” you mean “lack of measurements”?

Line 174: what do you mean by “30m company”?

Line 182: Does “in final” mean “in the final period”? Or “finally”, as in the final point being made? I am not sure what is meant by “homogeneously distributed”, or what differences are being described as reduced compared to what.

Line 191: “improving” is not correct here. “Increasing”?

Lines 194-195: what are partial variables?

Table 1, and related text: what is the signifiicance of these trends? Some are very small, and there is clear variabiliity in the timeseries. For example, lines 177-179 state “A negative trend of halocline depth is clearly during 2008– 2014 in the southern sites of the basin (moorings A and D)” but in the table Mooring D only deepens by 0.35 m/yr – is it statistically significant? Is the short-lived deepening in early 2009 having an effect on this trend?

Lines 205-206: “According to section 3.1, we find the main differences of evolution only between northern and southern basin are obvious, which is not completely identical with previous findings.” What specifically is different from previous studies?

Figure 3: Are these the average values?

Lines 220-235, Figure 4: See one of the major points - this analysis does not consider that the gyre centre moves and area it covers expands/contracts over time. Given that the northern limit is only 79N, perhaps the stationary section is seeing a different part of the gyre/not capturing all of the northern extent in later years? You might see the same “equilibrium” if you took just the 73-76N range of “before 2008” plot, for example.

Line 256: “The cold-core anticyclones are popular in the BG region due to oceanic stratification and large-scale dominated circulation.”. Why is this? Also, the word “popular” should not be used here – maybe “common”?

Figure 5, section 4.1: There are some interesting features here. However, it would be nice to have a paragraph relating the individual eddy counts with the EKE profiles. For example, why does Mooring D have a similar profile of EKE in 2003-2007 and 2008-2014, but more eddies identified in 2008-2014 than 2003-2007? Does this mean that the deviation of velocities from the mean is contributing a lot to the EKE profile in 2008-2014?

Lines 326, Figure 8: you have spent much of the paper describing the differences between the moorings (halocline properties and EKE). So you need to justify more why you are choosing to combine the mooring data here.

Lines 333-335: A fluctuation of both datasets doesn’t seem to be the case between 2010 and 2015?

Lines 339-341: Which datasets are you talking about here? MMP data seems to be higher since 2014. “Recently” should be specified, since oscillations occur at different times in each dataset.

Lines 334-335: which halocline variables? Do you mean depth and thickness from the first few figures in the paper? If so, refer to that here. The “plateauing” is only relevant for SODA and altimetry – MMP seems to decrease over this time period.

Lines 357-358: it would help to guide the reader to the relevant part of the figure here (where the Alaska box is) as this is a new way of looking at the information.

Lines 359-362: This is the second time MKE is referred to. Since it is not shown, it should not be described as though it is referring to a figure unless it is of relevance to the discussion. What is the main point of talking about MKE here?

Figure 10:  I would recommend putting the Alaska box on this map to help the reader.

Lines 390-391: see major point about asymmetry along the section

lines 397-413: I found this paragraph hard to follow. It might help if figure was referred to more. Perhaps remind the reader what a positive value in A means, as you described in the methods. Why can’t the salinity anomaly can be related to changes in freshwater rather than eddy transport?

Line 450: The proposed relationship between changes in the mixed layer and the tilt of the halocline should be explained much more clearly here.

Several instances:

• “abnormal” or “anormal” salinity should be clarified
• “mean time” is used a lot – is it meant to mean “average state”? Or “same time”?