Hi, this comment is a bit of a cop-out I'm afraid.  Having read the paper, and realised that is 95% computer modelling-based, I feel I am probably not the right person to see this one through the whole review process.  I've worked with computer tectonic modellers, but never done it myself.  Therefore you need someone who can properly examine the input variables, boundary conditions, iterations, etc.  May I suggest John Naliboff if you haven't already asked him?  I worked with him on a recent paper and he is excellent at this stuff.  He's at New Mexico Tech, john.naliboff@nmt.edu.

However, as this is an open review anyway, a few general comments:

1)  Most importantly, you need to describe why this is an important (or at least useful) contribution to the discussion on the structure of highly extended margins.  How does it differ from the models already published, or does it just confirm and add weight to them via lithospheric modelling?

2)  It would be useful to define in simple terms, up front, what "in sequence" and "out of sequence" faulting mean.  Not everyone reading this will have been as deeply immersed in the debate as you are, so some definition of terms would help the general reader.

3)  In my opinion, there isn't enough actual seismic in this paper.  It's almost all computer output, with nothing to tie it to.  Apart from Fig.  6 (which actually describes a special case) it's just assumed that the reader has seen the seismic elsewhere.  For completeness - and real-world grounding, please consider at least one more general seismic line over the margin.

All general comments, I know.  You need an expert modeller for the rest.

Best wishes and good luck with the process,

Tony Doré

Energy & Geoscience Institute (U. Utah affiliated, but based in London)

October 2023

The paper by Gomez-Romeu and Kusznir presents modelled fault geometries, whose sequential restoration includes the effects of flexural isostacy. The main aim is to demonstrate how different starting fault geometries (listric vs planar faults, fault angle) evolve with time, and respond to isostatic effects, and consequently, what criteria might be identified to discriminate the different starting geometries on seismic reflection data. The paper is well written and referenced, and the illustrations are good. The examples presented usefully illustrate the basic range of geometries that can be expected for large displacement faults on hyper extended margins, and how the sequence of deformation is important. I only have minor comments – below.

Lines 111-113 Perhaps could be more specific about the starting configuration. Fig. 3 – was a) and b) modelled by pure shear and the conjugate faults are just schematically illustrated?

While Fig. 1 is a useful summary, it would be good to have at least one regional seismic line that illustrates the full regional picture of what you are modelling. Because Iberia is referred to numerous times in the text, it got me wondering if you were actually trying to   model a particular part of the margin. I think it would be helpful to more clearly state somewhere around lines 52-59 that you are not modelling a specific margin or seismic line, but you have created generic models to address the types of feature found along the Iberia margin (and perhaps cite other examples too).

Model formulation – although the model can include sedimentation, it does not appear that sedimentation has been built into the modeling, and I did not see in section 2 any mention about why this has been omitted. I presume it is because the syn-kinematic sediments are likely to be thin, and not much denser than the water column. But I think this simplification of the model should be addressed. It also provides the opportunity to address your assumptions about how rapidly you think the extension would proceed (slow extension would acquire more syn-kinematic sediment).

 I accept that these margins can largely evolve from normal faults that dip at typical normal fault angles (i.e. 45-60 degrees), so have no issue with the scenarios presented here. There is an implication that variations in starting fault angle do not need to be considered. However, I think it would be interesting to look at normal faults with lower starting angles (20-30 degrees) as well, just to see what differences may arise, and are features produced that are incompatible with the seismic data. Reactivation of low-angled zones of weakness, such as thrusts, may sometimes produce initially low-angle normal faults.

One simple way to estimate the initial angle of a fault is to assume bedding was initially horizontal, and use the bedding-fault cutoff angle to determine the initial fault angle. It would be useful if you could comment on whether there are significant changes in cutoff angle as your model increases displacement, and what magnitude of variations occur.

 Summary, perhaps being more specific about the nature of the S reflection and how your modelling of planar and listric faults differ would be useful.

I enjoyed reading the manuscript.

Chris Morley