Dear Johannes,

I have now read the two reviews, your response to each one and the new version of your manuscript. I think the new version has been improved by accounting for most of the remarks made by both reviewers so that your paper can be accepted in TC. I have anyway found few errors/typos in the new manuscript that you should correct before it goes to TC.

Thanks for having choosing TC for this nice piece of work.
Best regards
Olivier Gagliardini

Remarks:
- page 3, line 15: (LSM Osher and Sethian, 1988) -> (LSM; Osher and Sethian, 1988)

- page 3, lines 22-23: Be more precise here. In Pralong and Funk (2014), was it to follow the calving front or the upper surface as in the work by Jouvet work? Add Jouvet references?

- page 3, line 25: you might add melange as a factor that control calving (see e.g. Krug et al, TC 2015).

- page 4, line 1: in the LSM is straightforward -> in the LSM should be straightforward

- page 4, lines 15-17: I don't understand what is really done to distinguish these two areas? Different sliding law? And how the shear margins are accounted for in the model?

- page 7, line 9: to solve eq. (4). -> to solve Eq. (4).

- page 8, line 2: you should specify here that a^{\perp} = c^{\perp} + m, the sum of a calving rate and frontal melting? And then later assume that m=0 in what follow.

- Eq. (6): I would present it the other way : the level set velocity is the sum of ice velocity and ablation velo. I don't see the interest of equation (6), i.e. the expression of a^{\perp} as a function of v and w will never be used, it is the equation that gives w that you need.

- page 8, line 12: For SSA, there is no need to extrude the mesh. A 2D mesh in the horizontal plane is enough?

- page 8, line 14-16: you should mention that, in an other hand, using the LSM in the vertical dimension would allow for example to account to non-uniform ice front melt and undercutting effects? but would also require to move to a 3d ice flow model (Stokes).

- page 9, line 4: I would suggest to be exhaustive here and to give all the variables that need to be extrapolated outside of the ice domain. For the velocity, which is not a scalar field, you should aslo be more precise and specify how it is extrapolated (component by component I guess?).

- page 11, line 21: not all equations are concerned by the CFL criteria. You should specify which equations require this criteria to be fulfilled in order to have a stable numerical scheme.

- page 12, line 4: Knowing surface temperature doesn't allow to evaluate B. You would need to know the ice temperature for that. Which assumption is further made for ice temperature?

- page 13, line 5: w.n = 0 is true only if the norm of v and c are also equal? The fact that the two velocities are aligned along n is not sufficient?

- page 13, line 22: equation is out of the page.

- page 14, line 9-13: Isn't it A which is the constant calving rate forcing? Then I don't understand the sentence : "In experiment A, the calving front undergoes gradual
retreat over time". Is there an error here. Are you really speaking about the A experiment?

- page 15, line 14: Fig. 7 shows -> Figure 7 shows

- page 15, lines 21-23: over which time period should be specified here.

- page 21, line 7: I think this conclusion is highly dependent on the very simple test used here which is really difficult to link to a real calving front? I would be less affirmative in suggesting a grid size to be applied for a real glacier.

- Legend Fig. 6: along-trough. c) ice geometry -> along-trough; c) ice geometry

Dear Johannes,

Thank you for this new version which can be now published in TC.

Regards
Olivier Gagliardini.