In their manuscript the authors deal with the effects of glacial isostatic adjustment (GIA) on the stress and deformation patterns in the Western European Alps. In particular they investigate whether present-day observations of strain rates measured with GNSS and earthquake mechanisms correlate with the theoretical deformation pattern that would be associated with GIA. Further they investigate whether GIA would promote or inhibit movement along some of the major fault systems in the Western Alps.

They use the LGM ice load, a simplified deglaciation history and a thin-plate model with a ranges of values for the effective elastic thickness of the lithosphere (he) and upper mantle relaxation times (tau). They find that model derived strain rates are consistent with the GNSS observations in the inner Western Alps, both, in orientation and magnitude. In the foreland regions to the west and to the north only the orientation matches the GNSS observations, whereas in the south neither orientation nor magnitude are in line with the data.

Concerning the faults, they perform a Coulomb Failure Stress analysis that includes (1) only the stress perturbations caused by GIA and (2) the full stress field (GIA + background). They present results for different fault dip angles and friction coefficients and conclude that the present influence of GIA tends to inhibit fault slip and that the observed earthquake kinematics is at odds with the deformation predicted for GIA and measured with GNNS.

Their main conclusion is that the GNSS is dominated by transients caused by GIA, whereas the seismicity reflects long-term geological forcings.

The manuscript is well written and I have no objections with it being published except that it lacks a conclusion section and the figures should be improved.

Minor comments/edits:

Line 103: flowing -> following

Line 186: high altitude

Line 191 – 200: Is there an effect on the strain rate induced just by the topographic gradient between the Alps and the foreland?

Line 222: in mind

Line 234: IMNF should be defined at its first use

Line 240: I suggest to use tau either for the relaxation time or the shear stress, not for both. Maybe add a subscript.

Line 254: delete “of”

Line 283: µ’ has been defined already in Line 241

Line 327: oriented

Figures:

The figures could be improved by using only one font style and size and a more consistent panel labeling.

• add scale bars to all the maps
• Fig. 3. omit the dot in the velocity unit
• Fig. 6 & 7: I struggled with the symbology of panels b) and c). The results for the different dip angles cannot be distinguished. Perhaps different marker symbols could be used (e.g. squares, triangles etc.). I suggest to replace the horizontal bars with markers.
• Fig. 6: “(a) Horizontal full stress (background + GIA) and faults tested in the CFS analyses.” But there is only a single fault shown in (a).

Dear Editor and authors,

Please find enclosed my review of the manuscript "Glacial isostatic adjustment strain rate - stress paradox in the Western Alps, impact on active faults and seismicity" by Grosset, Mazzotti & Vernant. The manuscript investigates the effects of the latest Alpine glaciation on current seismicity in the Western Alps, using models of glacial isostatic adjustment (GIA), current earthquake mechanisms and the current stress field as inferred from the earthquake mechanisms. The manuscript is a revision of an earlier version, which I find is significantly improved and ready for publication with some minor revision.

Comments and suggestions:

Abstract

- The authors use the term Last Glacial Maximum (LGM) as a name for the latest major European glaciation, both in the Abstract and elsewhere in the manuscript. The LGM is a specific point in time, not a name. The latest Fennoscandian ice sheet is known as the Weichselian glaciation, I am not sure if this includes the Alpine ice cap, or if that has another name, but the authors should not use LGM in a naming sense.

- I am not sure that the first sentence: "In regions formerly glaciated during the Last Glacial Maximum (LGM), Glacial Isostatic Adjustment (GIA) explains most of the measured uplift and deformation rates." is correct. Does this really apply to active regions like Alaska or Western Canada?

Line:

- 25. Deglaciation increases seismicity in regions with appropriate tectonic background stress, such as cratons or regions in a reverse stress state. It is likely to decrease seismicity in regions with normal stress, such as e.g. in Iceland, see e.g. Lund (2015). I suggest modifying the sentence with a "... in many regions formerly..."

- 29. Grollimund & Zoback, 2001 unfortunately has methodological problems, I would not refer to that.

- 44-45. Naming of the glaciation, don't use LGM.

- 70. The sentence is a bit confusing, the events can be divided into two N-S clusters, one in the west and one in the east? The cluster seem elongated in N-S, just clarify the writing a little.

- 103. Spelling "flowing" -> "following"

- Section 3 on GIA models and Figure 3. I would like to see at least a map of the modelled uplift rates, at the same locations as the GNSS data points in Fig 3a, or a map of the difference in these points. As the uplift rates are used to evaluate model fit, this would help the reader assessing how good the models are.

- 155. In the GIA modelling the authors assume isostatic equilibrium at the LGM owing to the short relaxation time of the mantle. But then they use relaxation times between 2,000 and 20,000 years in the modelling. A Maxwell relaxation time of 2,000 years corresponds to a viscosity of about 6x10^(21) Pa s (for Young's modulus 10^(11) Pa), which is a fairly high viscosity mantle. 20,000 years corresponds to lower mantle viscosities. There is therefore a discrepancy between the used relaxation times and the assumption of isostatic equilibrium at LGM. Also, I would have thought that the Alpine mantle would have lower viscosity/relaxation time. In the Supplement the authors model an ice sheet which resides on the model for 100 kyr before instantaneously melting. It is unlikely that the LGM lasted for such a long time, even the Fennoscandian ice sheet only had (almost) LGM dimensions for a few thousand years, which is not enough to reach equilibrium. The authors should at least discuss this. 

- 165. It seems the model testing in the supplement is performed on a relatively small (horisontally) model in comparison to the size of the ice sheet. This is likely to affect the stress values. It is also unclear if the GIA/viscoelastic model takes into account stress advection. In addition, it would have been advantageous with plots of the stress at the bottom of the elastic plates, in order to compare how stress is modeled there (for the discussion in line 348).

- 241. Perhaps add that you follow Bott (1959) in assuming slip in the direction of maximum resolved shear stress on the fault.

- 300 and below. Indicate how you use the coefficient of friction. Do you always have the same mu when constructing the stress state as when evaluating CFS? I think that is a good first order assumption, but one could also think that established faults have a lower mu and slip more often.

- 321. I find it interesting that you find that a normal stress background field can in some cases be pushed into failure by a reverse GIA induced field, under the former ice cap. Could you deliberate bit more on this; angle between fault and SHmax, stress magnitudes, dip and rake in these cases?

- 375. Add to the discussion, here somewhere, that the models predict that in the future, with diminishing GIA stress components, seismicity will increase as the crust comes back to more normal/strike-slip stress conditions.

- 403. Thanks for having me in the Acknowledgment! I don't have the umlaut on my surname, it is just "Lund".

Figures:

- Figs 4, 5, 6, 7. I prefer to illustrate the stress field with lines/bars of the same size showing the direction of SHmax, and colour contours of the magnitude of SHmax, instead of the double arrows used in (a) in these figures. That more clearly shows how SHmax rotates 90 degrees outside of the former ice margin, and better indicates the magnitude. Especially in Fig 4, the double arrows outside the ice margins are so small that they are virtually impossible to interpret.

- Figs 5, 6, 7. It took me a while to find the indicators of "SW", "NE" and similar on top of the CFS plots. I would remove those and write in the xlabel "Distance from SW to NE along..."

- Add explanation of the orange line in the box plots.

Reference: Lund, B. (2015) Palaeoseismology of glaciated terrain. In Beer, M., Kougioumtzoglou, I.A., Patelli, E., Au, S-.K. (eds.), Encyclopedia of Earthquake Engineering, Springer Berlin Heidelberg, 1765-1779, doi: 10.1007/978-3-642-36197-5_25-1.

Best regards,

Björn Lund