Thanks for the comments and suggestions!

We have now created and included in the repository an example of a composite analysis of a unique eddy that had many profiles while trapping an Argo float.

https://github.com/iuryt/vortex_profile_matching/blob/main/examples/unique_eddy.ipynb

Regarding updates, we are planning to create an NRT combined dataset in the future. The code is also available for other users to collaborate on future enhancements and requests. Feel free to create new issues on the GitHub repo.

https://github.com/iuryt/vortex_profile_matching

We thank the reviewer for the comments, and we are glad the manuscript and dataset are seen as useful and well-documented. This positive assessment supports our main goal to make this dataset broadly accessible and relevant to the oceanographic community.

General Comments:

We would like to thank the reviewer for the valuable comments and general suggestions. These recommendations definitely improved the quality of our work. We are also pleased that the reviewer found the manuscript to be well-presented and well-written, and that the figures were considered compelling. Below, we provide detailed responses to the main points raised in the general comments section. We also attached the latexdiff document to make it easier to track changes from the last version.

We agree that “Eulerian” is traditionally associated with fixed-point observations, such as moorings. However, in the context of mesoscale eddy detection, the term has been previously used in the literature (Abernathey and Haller, 2018), including in a publication in Earth System Science Data (Liu and Abernathey, 2023). This term seems appropriate to describe eddies identified from methods that do not take into account the evolution of the velocity field. In other words, Eulerian eddies can be identified with a single SSH snapshot, while Lagrangian eddies are identified by methods that integrate the velocity into trajectories and thus are defined by material coherence. We use “Eulerian eddies” specifically to distinguish the approach used by Pegliasco et al (2022), which is based on SSH extrema and contours obtained AVISO+ interpolated fields, from the method used by Liu and Abernathey (2023). This distinction is important for readers comparing different eddy identification frameworks and also because we plan to include Lagrangian eddies in future versions of this dataset. We also find the term pedagogically useful, as it provides a concise and internally consistent way to emphasize the differences in the assumptions taken by each method while avoiding the misleading implication of material coherence. We therefore prefer to retain the use of this term.

We appreciate the reviewer’s observation regarding the distinction between daily eddy observations and unique eddies tracked over their full lifetimes. To clarify this point, we have revised the text around page 10 to define the terminology more explicitly. [L27-31] 

Regarding the eddy-tracking algorithm, we have now added a short explanation in Section 3, clarifying that eddies are tracked based on spatial overlap and similarity of geometric properties between consecutive days of closed sea-surface height contours around local extrema (Pegliasco et al., 2022). [L164-167]

We also agree that the level of detail available in each profile–eddy match is an important feature of the dataset and needs a more explicit description. To address this, we have added a clarifying paragraph at the beginning of the “Combining datasets” section (Section 3), where we now specify the full set of eddy and profile properties stored in the matched dataset. This includes the eddy center coordinates, effective and speed radii, amplitude, Rossby number, and a persistent eddy identifier that allows users to track the full eddy trajectory over time. The profile metadata (e.g., location, time, depth, instrument type) is also preserved, enabling composite and evolutionary analyses of eddy structures. [L227-L231]

Minor Comments (following the order):

• We deleted the word “Data” from the title.
• Yes, each eddy is assigned a unique identification number (track), which is maintained consistently throughout its lifetime. This allows eddies to be connected across different monthly files. The 120 "chunks" refer to divisions of the full dataset used for efficient parallel processing. These chunks are independent of the monthly separation and are used during the filtering stage, for example, to remove eddies with short lifetimes or low amplitudes. After filtering, the resulting dataset is loaded into memory as a whole, and only then is it split into individual monthly NetCDF files for output.
• We made the suggested changes in the caption of Figure 5.
• We made the suggested changes in the caption of Figures 7-10 and the reference to the panels in the text.

Following a comment from the open discussion, we have now added an example of a composite analysis of a unique eddy in the repository and included an alternative way to access the data through AWS S3 bucket.

https://github.com/iuryt/vortex_profile_matching/blob/main/examples/unique_eddy.ipynb