This is an interesting and timely study with some novel (and unexpected) results, and as such I think it is a good fit for the journal. I do have some reservations though and feel that there are some areas where things could be tightened up prior to acceptance.

General observations

Mix of trait and functional is used throughout the paper, I feel that it would be better to be consistent with this, maybe mention both initially then choose one and go with that for the remainder of the text

Is there any code available for analyses carried out?

Traits - was there any sensitivity analysis undertaken here or did the authors look at how closely or otherwise the traits were correlated with each other?

I feel as though 99 permutations isn't all that many - isn't it more usual to use 999?

Is functional distinctiveness done with or without abundance weighting? Given that (in my opinion) one of the most interesting findings is that abundance plays an important role in the changes uncovered, it might be interesting to calculate functional distinctiveness both with and without the weighting for abundance [1,2].

Generally, the paper reads very well with good attention to grammar/spelling but I feel there are a few places where greater attention to plurals/no plurals could be paid. I have identified a few places where this could be improved but I think there are still others.

line 182 suggest represents rather than represent

line 301 suggest evidence rather than evidences

line 316 suggest 'fishing activity has ...' or 'fishing activities have ...' rather than 'activity have'

line 319 suggest corresponds rather than correspond

line 342 suggest supports in place of support

line 364 suggest shows instead of show

lines 367-369 suggest rewording this sentence as it doesn't read well at the moment and seems an important take home message of the paper

line 698 - suggest represents instead of represent

Tables & Figures

Table 1 I think the table could be better aligned text wise as some columns seem to be spaced oddly

Figure 1 it's hard to understand where in the inset map of South America the islands appear - perhaps a box around the area would help to clarify this for the reader. I don't think that the colours on the SST plots are colourblind friendly.

Figure 2 without going back to the main text I wasn't clear about whether the absence of grey lines in the North (a and c) and at all in b meant that there was just no significant regression - also in d I wasn't sure if the line pertained to the North or the South. Maybe adding regressions for them all with dashed for non-significant (and coloured by region too) would be easier to understand.

Figure 3 although it is much clearer which is which here compared to figure 2 maybe all regressions could be added here too for consistency

Supplementary materials

Figure S1 suggest that the island names are capitalised and the underscores removed

Figure S2 were there no UVCs performed at all in the south in 2017 and 2020?

Figures S5 & S6 suggest that the species names be capitalised, italicised and the underscore removed (as with the main figure). I also wonder if it is of potential interest to see these figures combined in some way so that for those species featuring in both the Northern and Southern regions it would be possible to see a comparison between the changes found.

Figure S7 suggest using the same y axes for both plots

References

1. Grenié, M., P. Denelle, C. M. Tucker, F. Munoz and C. Violle (2017). "funrar: An R package to characterize functional rarity." Diversity and Distributions 23(12): 1365-1371.

2. Moyes, F., I. Trindade-Santos and A. E. Magurran (2023). "Temporal change in functional rarity in marine fish assemblages." Proceedings of the Royal Society B: Biological Sciences 290(1993): 20222273.

I am happy that the authors have addressed my concerns where possible (and have explained their reasoning where this was not possible). I feel that the addition of the sensitivity analyses and null model strenghten the overall results and enhance the robustness of the findings. I am also glad to see the inclusion of the abundance weighted version of distinctiveness and feel it adds a level of interest that the results were similar.

Overall, I feel that the paper now reads better and the figures are clearer. I think the authors have done a good job in addressing reviewer concerns and as such feel that the paper should be accepted for publication.

I was happy with the earlier changes the authors made and now feel that they have also addressed the concerns raised by other reviewers very well. The inclusion of Figure 2 really aids understanding of the steps in the analyses and adds to the overall appeal of the paper. I feel that the authors have done an extremely good job of addressing all concerns and as such would recommend publication.

I had a couple of very minor comments though -

Lines 88-90 'due' is used twice in this sentence, maybe consider rewording
Lines 208-212 would this be better expressed as an equation? It would make it clearer to the reader.