Please see the attached PDF supplement.

The paper examines summer-time Arctic boundary layer clouds, observed during a shipborne campaign. For a few hours clouds showed a two-layer vertical structure. The paper uses LES to reproduce the observations and to also perform a sensitivity analysis. The authors find that decreased aerosol concentrations and greater windspeed would erode one or the other layer.

The paper is well written and the figures are of good quality. I have one major and a few minor recommendations that the authors should resolve before publication.

Major comment

Section 4.2.3: Given the great sensitivity to windspeed, the topic appears under-explored. Perhaps the authors could provide additional information that helps to understand how erosion of the lower cloud layer was facilitated:

• How much of increase in surface turbulent heat fluxes was found for greater windspeed?
• How did the TKE profiles change across simulations?
• What explains the increase in column total water?

Minor comments

Fig. 1: For geographical context either Latitude lines or labels of nearby landmasses would be helpful.  

Fig. 5: Perhaps changed y label to dN/dlogDp.

Section 2.2: A few details should be added here:

• How was temperature and moisture advection handled?
• Was aerosol treated diagnostically or prognostically?
• Does the simulation consider the water vapor and ozone column above the model domain?
• Was the diurnal cycle of solar insolation taken into account?

ll. 308-309: Please explain which instrument or technique was involved in estimating hygroscopicities.