Dear Dr. Gutjahr,
As you see the reviewer is happy with your revision. He has some minor comments. Please resubmit a revised version.
Best wishes
Qiang

Dear Dr. Gutjahr,

Throughout your text you call your 0.1dg ocean setup as an eddy-resolving one. This is true for most parts of low to mid latitudes, but not for the very high latitudes, including the Arctic Ocean and marginal seas around Antarctic. For example, "However, our results show that an eddy-resolving resolution in the Arctic Ocean removes the warm and saline bias in the Atlantic Water layer, without changing any background values for vertical mixing", which is a misleading statement. Mesoscale-eddy resolving for the Arctic Ocean needs about 1km resolution in the deep basin and even finer resolution over the shelf. In the conclusion section, you used a subtitle "Eddy-resolving ocean", which is again misleading as you also referred to non-eddy-resolving regions such as Arctic and Weddell Sea there. The benefit does come from using high resolution, but not eddy-resolving in some cases you described. Please go through the text and correct all related places.

In section 4.2, you mentioned that Sein2018 used background diffusivity about two orders of magnitude lower than in Wang2014b. In the latter, 1e-5 was used in the upper ocean, so it should be about one order of magnitude lower in Sein2018.

You used the phrase "hybrid grids" in the revised version following the reviewer comments. This is misleading as it is usually reserved for naming vertical coordinates. I think you are referring to a case when some ocean regions are eddy resolving and some other regions are not on the same grid (due to varying Rossby radius and partly varying resolution in terms of km). Please revise the text to avoiding misleading. And actually the high resolution ocean setup also belongs to this case, because on this 0.1dg mesh some regions are not eddy resolving, or even not eddy permitting.

Best wishes
Qiang