This is an interesting and well-written manuscript looking at the segregation, in terms of spatial migratory connectivity and habitat use, of three closely related shearwater species in the Mediterranean Sea.

Abstract
Aim - this section sets out the introduction of the manuscript well, but it does not set out a clear aim. In addition, there isn't a clear link between the use of migratory connectivity to promote population divergence and then the next sentence which focuses on environmental preferences, although I appreciate that the latter can result in migratory connectivity.
Lines 20-21. I recommend including strong MC here where the authors mention 'more structure' to emphasis the ecological meaning of this to readers given your introduction to MC on line 5. Furthermore, I appreciate that space is limited in the abstract but I recommend changing 'where MC was close to zero' to 'where MC was weak' as the authors have not yet introduced what an MC value of close to zero means.

Introduction
Line 42. Although for some seabird species, i.e. Kittiwakes, natal philopatry can be weak and therefore there is connectivity/gene flow between even distant populations. Recommend clarifying that this is the case in some populations on line 40.
Line 55. It would be useful here (or in the maybe the discussion would be more appropriate) to expand on why this has consequences for conservation and management actions an provide specific examples.
Line 83. Please also include in the methods / results the number of individuals tracked per taxa.
Line 86. Be consistent throughout as sometimes the authors state (colony, population, taxa) and other times (colony, population, taxon).
Line 89-90. Change low and high, to weak and strong as previously used in the manuscript to describe MC.

Methods
Line 97. Does this strong philopatry refer to both natal and breeding philopatry? Please include that here.
Line 142. It is not immediately obvious here what the authors mean by ' minimum representativeness valus of 70%)?
Line 167. How were the boundaries of these three offshore areas decided / defined?
Line 194. It would be useful for the authors to report (i.e. in a supplementary Table) the proportion of fixes that fell within each assigned non-breeding area for each individual.
Line 197. Please also add a supplementary Table which includes the population size of each colony included in your analysis. And the number of individuals that were tracked from each colony.

Results
Line 266 and throughout the results. It would be useful to include some ecological meaning to the MC results i.e. an indication of whether reported MC values are considered strong / weak / intermediate? One line 67 and 74, adding this context will help improve the link between the following text highlighting the mixing of individuals from different colonies within the non-breeding regions.
Lines 289-304. If all models fitted well with high AUC values, it is not clear why the authors then describe the AUC values in the text here - I would remove this text unless I have missed something here? Maybe adding some ecological context to this section might help understand its importance?

Discussion
Line 337. Change 'high' to 'strong' MC values.
Line 432. Adapted to local conditions in what why? In terms of foraging strategies? Please include more clarification, as well as being more specific in terms of what local conditions? Your results indicate that SST and Chla were important in the habitat models, but these are only proxies for prey availability / marine productivity?
Line 439. What do the authors mean by 'This' here?
Line 442. What kind of behaviour? Migratory behaviour? Please be more specific.
Line 480. Please expand here on why separate management and conservation plans may therefore be required?
Figure 1b. It would be useful to number these regions so they can be easily matched / compared with the numbers in Figures 2 and 3.