A combination of radiosonde, wind-profiler and aircraft observations, together with large-eddy modelling, show that shading from orphaned cirrus anvils had significant effects on the boundary layer during CSIP IOP 5. The control of boundary-layer development by the cirrus shading, and the subsidence induced by the cirrus cover, led to a boundary layer that, at an aircraft altitude of around 500 m, was less turbulent, 0.8 g kg−1 drier and 0.3 K warmer under thick cirrus than in clear skies. In addition, at the top of the boundary layer, turbulence persisted for some time after the onset of cirrus shading, allowing continued entrainment and so stabilization of the previously well-mixed boundary layer. These effects, which would have inhibited convective initiation under the cirrus, are consistent with the results of the first part of this study (Marsham et al., 2007).
Power spectra show that, while vertical velocities were dominated by variations on the scale of the boundary-layer depth, for water-vapour mixing ratios and potential temperatures, mesoscale variations were comparable or dominant. In some cases a spectral gap is observed between these two scales of contributions. Radiosonde data and modelling show that, in addition to the variations in the boundary layer caused by the variable cirrus cover, pre-existing variations in lid strength and boundary-layer moisture were also significant on this day. These water-vapour variations are also detectable in retrievals of total-column water vapour from Global-Positioning-System data. Copyright © 2007 Royal Meteorological Society