Dear authors,
Apologies for the delay, a second review never came back. I have reviewed your responses, but i remain unconvinced about the reasons you outline for not comparing CPv1.0 fluxes with CLM stand alone fluxes. These are of course different models, making different assumptions. But this is not a valid reason for not explaining the differences. You simply refer the reader to Figure S4, and leave the reader to make their own conclusions. The differences in latent heat flux between CPv1.0 and CLM stand alone are very large. You need to explain why these differences are so large. And which is likely to be more "correct/realistic"???
If you were to use CLM versus CPv1.0 coupled to an atmospheric model, such large differences in latent heat flux would have a very large influence on near surface temperature, humidity, boundary-layer structure etc. I do not think you can leave this analysis out. You have added/coupled an additional component to CLM, it makes sense to me, that you should compare simulations with the additional component, to the original model. That would generally be expected for any model development. A reader of this manuscript will be left very confused as to why the latent heat fluxes are so different between the two. You have to address this explicitly in the manuscript.
Kind regards,
Jatin

Dear authors,
Thanks for addressing my additional comments. Your amended text now provides the proper context to help the reader better understand the differences in water and energy budgets between the two models, as it relates to spatial scale. Congratulations on getting your work accepted for publication in GMD.
Kind regards,
Jatin