Dear Anne,

The reviewers' comments look pretty minor. Please do include some of the descriptions of the dynamic vegetation model, model initiation procedure, and generations of high resolution data as suggested by Reviewer 1.

Looking forward to your revised manuscript!
Cheers,
Ran

Hi Anna,

Thanks for the revised manuscript! It looks good! I have a few technical suggestions:
1) The discussion of dust in the introduction may also consider more recent work (such as https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2018GL081225). The newer ESMs are now able to simulate indirect effect of dust aerosols on clouds and precipitation. Old results on this subject may need to be revisited.

2) Following 1), it would be helpful to add information on how the dust is treated in the MPI model. My suspicion is that ECHAM6 handles aerosol indirect effect. But I cannot figure out how the dust emission is handled in your simulations. Does it vary with vegetation cover? or a fixed set of emission files?

A few minor editorial corrections:
In figure 1, please label y-axis and x-axis.
The vector of moisture flux in Fig. 8 is a bit hard to discern. Would it be possible to reduce the reference vector length and extend the plotted vectors? Also a), c) and b), d) are not aligned correctly. Similar alignment issue also occurs in Fig. 10.

In fig. 11, Would you please convert the vertically integrated moisture convergence to the unit of mm/day, so it would be comparable to precipitation shown in the previous figures? Similarly, you may also consider converting the units of vertically integrated moisture flux to mm2/day. Here is one example of moisture divergence plotted in Seager et al., (2009): https://www.pnas.org/content/pnas/107/50/21277.full.pdf

Thank you!
Cheers,
Ran