One of the reviewers mentions that it is not still clear that what the novel results are as shown below. The authors have to address these points.

-----
Thank you to the authors for responding to my questions/concerns (Reviewer #2).

Comment #1: I’m still concerned that the results presented are very limited. The model is designed to study the relative importance of aerosols with different chemical compositions but only the most basic studies were done (only dust can form ice). It is not a new result that allowing ice to precipitate will reduce the IWC and allow LWC to persist longer or that smaller aerosols that are not activated at cloud top can fall through the cloud layer and be transported into the cloud layer at cloud base. At the very least, a complete analysis of the freezing of smaller aerosols in supersaturated updrafts would provide support for this statement (lines 285-287).

Comment #2: This model is glaciating much earlier than all 11 models included in the ISDAC intercomparison for fixed ice number concentration. It would be useful to understand why in order to put the prognostic ice simulations into context. The authors say this is because the model is tracking the dry aerosol size. Why would this produce this result? Do the authors believe that this model is producing more realistic sensitivity that is not in the other models?

Comment #3: Are the aerosols tracked after they form ice? After the ice sublimates, are the aerosols added back into the population? It looks like this is not taken into account in Figure 1. Including this process will greatly impact aerosol concentrations below the cloud layer and entrainment of aerosols at cloud base.

Comment #4: Figure 2 legend: “bulk” and “bin” should be switched.

The revised manuscript is fully addressed to reviewers' comments. Please revise Figures 6 and 8 because it is difficult to distinguish between blue and black lines.