Dear Authors,

Thank you for the revisions made which improve the quality of the paper. I understandthe wish to keep hypothesis first in the paper (model based writing style mentionned byboth reviewers and I) and then I have recommendations about strengthening both yourhypothesis and interpretation. This is needed anyway in the paper wathever the syleis chosen, but actually more relevant if you chose this style. Especially, I recommendconsidering some significant references on the topic that have been ignored to reinforceboth your interpretations and hypotheses.

1) In the revised version, you mention in lines 30-31 of p7 "we conclude that foredeep-parallel extension has occurred in the foredeep of the Pyrenean belt since the Pale-ocene and until the end of convergence" Do you consider here that Sigma 3 is negativeas proposed in Figure 1b and introduction ? Extension is an unclear deformation termnot synonym to tension or extensional stresses (i.e. negative stresses). Clarify this inthe text please.

2) On this negative stresses as shown in Figure 1b, although we can agree on your in-terpretation, the paper suffers considering the significant contributions from experimen-tal tests which have been compared to natural joints from the past decade. You mentionextensional stresses (negative) but what about splitting without negative stresses (andeven with a slightly compressive sigma 3) such as demonstrated in dry axi-symmetric,oedometric, plane strain and poly-axial experiments by Chemenda et al. (JGR,2011)and Jorand et al (Tectonophysics 2012) ? These studies shows joints formed under drycontraction without negative sigma 3, which are not so far than uniaxial splitting frac-tures observed in triaxial cells (e.g. Holzhausen and Johnson, 1979), but here clearlywithout the triaxial boundary effect mentionned by Fakhimi and Hemami (2015).

3) A common species of joints show very low displacement gradients compared toother fractures (veins, faults) (Pollard and Aydin, 1998; Schultz et al., 2008), whichalso support the general fact that joint sets do not require significant amount of negativestresses perpendicular to them. Have you measured the mean opening of the observedfractures ? Can this help to discuss this point ?4) I recommend you to better support the hypothesis mentionned in lines 31-32 p2and 1-3 p3, which only relies on one reference, while others works previously de-scribed stress permutation during LPS. For example, stress permutation in forelandbasin has been proposed from field observations and stress path calulations by Solivaet al. (2013), and reused with nearly the same concept in Fossen’s book 2015 version.Addition of such references is just a fair strengthening of the hypothesis on which thework relies.

Please, consider discussing/including these elements in your paper, you have plenty ofspace since the paper is quite short.

Best regards,
Roger Soliva

Dear Authors,

Thank you for the revisions and the explanations. The paper is now ready for publications.

Regards,
Roger Soliva