Dear authors,

Thank you for your very interesting and important study on the formation and evolution of transform faults in the early Southern Atlantic Ocean. The manuscript provides many new insights into the structure of transform faults based on various kinds of geophysical data. The study builds on a previous study by Thomas et al. (2022) that presented 3D broadband seismic reflection data. Using the structural information from the seismic reflection data, the new study analyses and models potential field data to get a better understanding on the lithology and potential metamorphic processes in the lower crust. The study is of high interest and provides many original aspects. However, before the manuscript could get published I would recommend some moderate to major revisions.

The data and methods chapter is not yet well elaborated and needs substantially more details about the actual pre-processing and processing of the potential field data. At least proper references should be given, that it is possible to understand, which corrections were applied.  E.g., how the ship-borne data were tied into the global reference net? What is the actual resolution of the data, and what are the uncertainties? Did you run any resolution tests? In some cases, it seems to me that you try to overfit the data. Also, the ERR values do not really represent the uncertainties as can be seen from various figures.

I wonder a bit that only the lower crust is considered as a source for magnetic anomalies. To me it is not yet fully clear, how you can rule out differences also in the shallower crust. What are typical magnetic susceptibilities for the various rock types (shallow and lower crust) from literature data? The same is about the main “gravity sources”, in my opinion seafloor, basement and Moho topographies are major sources for gravity anomalies beside density variations in the individual layers or bodies.

For your modelling and inversion, you only allow changes to values within one standard deviation. Is that meaningful at all?

Regarding your results, why TNDR 3 and TNDR5 are that different?

In Figures 8 and 9 you show seismic reflections (or migration artefacts?) in the lower crust. They seem to spatially correlated with the positive magnetic anomalies. Did you try to model specific bodies within the lower crust that are different in the reflection characteristics? If these reflections are real and not artefacts, can you rule out that they are not related to later magmatic phases (e.g., hot spot magmatism)?

In my opinion, you are not yet convincing in the discussion about the metamorphic processes. You should discuss it in a better way to support your preference for metamorphic processes and why it cannot be related to serpentinization or later magmatic activity. Maybe, a schematic sketch could also help to illustrate your interpretations.

How do oceanic transform faults compare to other strike slip faults? Can you identify flower structures?

Are there any heat flow data (studies), supporting your interpretations and conclusions?

There are still many sentences that could be formulated more clearly. Sometimes, strange terms like “proxy” are used (e.g., line 211). The figures have overall a very good quality, but font sizes have to be enlarged. Abbreviations should be explained in the figure captions. Some figures should be enlarged (e.g., fig. 7).

How does your study differ and compare to classical (e.g., Lin et al. 1990/Nature or Prince & Forsyth 1988/JGR) or more recent studies? Did you also try to calculate derivatives of the potential field data to better localize the source of variations of density and magnetic susceptibility? Did you try to calculate Bouguer anomalies from gravity data or pseudo-gravity from the magnetic data?

A final technical comment, maybe more towards the journal than to the authors: I find the font size of the main text too small. It is very difficult to read.

I hope, you will find my comments and questions constructive, and that they will help to improve your manuscript.

With best regards, Wolfram Geissler

In the manuscript by Hass et al., authors perform a series of seismic and shipborn potential field data inversion to model the density and susceptibility of the oceanic crust and lithosphere in the ancient transform fault zones of São Tomé and Principe in the eastern Gulf of Guinea. Results show that these transform fault zones present low susceptibility and medium to high density, reflecting increasing pressure and temperature, what potentially generates metamorphism of the oceanic crust up to greenscist facies. The manuscript is well written and conceptualized, and both the data and the modeling are consistent and relevant for the study of plate tectonics, especially in the oceanic domain. I recommend the publication of this manuscript after a few minor modifications.

Comments by lines:

1. l.21: Please specify the facies change (prehnite-pumpellyite to greenschist?).
2. l.30-33: This sentence is unclear, what can be located within the oceanic crust?
3. l.33-35: For me it’s not clear what are the “sources”, and how they can change. I encourage a conceptulization of this terminology.
4. The addition of sediments increases the density and susceptibility of the crust? Why?
5. l.38: “generated by spreading ridges”, the strike slip movement can happen far from the spreading ridge itself.
6. l.41-43: mass deficit meaning crustal thickening? Seems contradictory, please explain.
7. l.43-44: Please improve grammar, confusing.
8. l-50-56: Can you provide a schematic figure of these potential TNDR-B processes?
9. l.63. “São Tomé and Principe transform fault zones” I suppose.
10. l.70: Please label the CVL in Fig. 1.
11. l.78: Please label the Central Fracture Zone in Fig. 1.
12. l.91: the high rugosity?
13. Fig. 1: Black bold line is the LaLOC? The C34y could be clearer, try yellow if this is an important feature to be shown. Can you also highlight the spreading ridges?
14. l.106: CGG Multi-physics is a software? Or an instrument?
15. l.113: removal of what?
16. l.180-181: “Here we use this software to invert…”
17. l.184-185: are you talking about a detachment fault? Or some other concept? I don’t understand how inverted data can “provide space for detachment”, but I might be confused by the geophysics jargon, can you explain better in the text what is this detachment?
18. l.221-223: density is high also to the right of your delimited TNDRB 3. There is some explanation for that?
19. In Fig. 8 d the dashed circle for TNDR 3 is bigger than the polygon in c. In c, from TNDR 5 to the red polygon to the right, there is not such a contrast in density or susceptibility, do you think they could represent the same structure?
20. l.398-400: The density and magnetic properties are caused by the mineral composition, I think you mean “changes in the density and magnetic properties” which can be caused by the factors you describe.
21. l.406: “which are often” or similar term, hydrothermal alteration is not the only reason for this metamorphic conditions.
22. l-410-413: In the last phrase, add that it’s unlikely in your study area, not in any all lower crust.
23. 418-420: This sentences seem contradictory with the first ones. So, you suggest greenschist was reached in the segment of crust you are studying or not?