In this manuscript, the authors present a focal depth re-analysation 35 earthquakes on the St. Paul Transform System using teleseismic stations in Africa and Brazil. They find that the earthquakes are generally located deeper towards the centre areas of the transform segments and shallower at the ridge-transform intersections. Placed within half space cooling models, they find cooler mantle at the centre of the transform faults. The results and implications should be interesting for the readers.

In general, this manuscript has a good structure and the setup, progression of points, as well as the thoroughness of the study are logical and easy to follow. Likewise, the figures are relevant to the text and mostly of good quality, but I will point out a few minor to moderate issues below. My comments are primarily of technical and typographical nature, therefore I recommend publication after technical corrections.

I will start with two broader issues.

1. Use of a rainbow colour scale in multiple figures: it would be better to not use rainbow scales, as these can be very hard for colourblind people to understand.

2. Depth “below seafloor (bsf)” – I might be wrong here, but I think that this is not used consistently throughout the manuscript (e.g., l. 269 or l. 304). If all of the depths are given below seafloor, it would be easier to state that in the beginning and then not having to use the “bsf” every time.

Below I will list my specific corrections.

l. 16: “5.3³Mw³6.9” – there seems to be some formatting error here.

l. 75: “Saint Peter and Saint Paul Archipelago” – it would be good to show (and annotate) this on the map.

l. 101: “2.1” – it would be better to have at least two sub chapters in a chapter. Therefore, the first part of “2. Study area” should be a sub chapter as well.

l. 197: “Mw” – this should be in italics with a “W” subscript.

l. 233: “neither comparison” – “neither the comparison”

l. 253: “focal depth my vary” – “focal depth may vary”

l. 269: “10 kilometers” – “10 km”

l. 271: “beneath the seafloor” – “bsf”

l. 286: “3-dimmensional” – “3-dimensional”

l. 330: “are strongly segments” – “are strongly segmented”

l. 339: “3-Compared” – “3. Compared”

l. 340: “Furthermore, the shallower earthquakes hotter areas close to the RTI.” – This is not a complete sentence.

l. 348: “GWSdM […]” – that whole sentence has multiple grammatical errors.

l. 352: “Thank captain” – grammar

l. 353: “Funding COLMEIA ship time was funded” – this sounds strange to me.

l. 360: “were downloaded by EarthScope” – do you mean “were downloaded from EarthScope”?

Figure 1:

- The text "a)" and "b)" seem to be cut off.

- Details in (a) , such as annotations, are too small. Maybe make the histogram inset an extra subplot, there would be space next to the map in the top row.

- Is there a better way to visualise the transform fault and MAR sections? Because the red and black lines are almost invisible, since they are covered by event circles.

- The black circles in the legend should be white.

l. 577: “White square” – it is not a square, so I would write “The white box”

l. 583: “(doi:[…]”) – I am not sure if this is needed here, it could be in the data availability statement.

Figure 2: The colour coding is opposite to the previous figure (red and black lines).

Figure 3: The magnitude is given by 6.9 in the top left and the caption of the figure, but as 6.68 on the right of the figure. Maybe a short explanation is needed as to why the difference is larger here.

Figure 6:

l. 811: “The a) Comparison” – remove “The”

Figure S04: “Map with the locations of the 21 earthquakes” – “Map with the locations of the 21 stations”

David Schlaphorst

Lisbon, 25/06/2024