Summary

Results from two large eddy simulations of the 18 May 1998 SHEBA/FIRE case have been presented. One simulation uses the thermodynamic and CCN sounding data taken from the SHEBA/FIRE research flight RF05 (May 18, 1998). The other uses a CCN profile of smaller constant value.

The major results of the present study may be summarized as follows. The simulation with the observed CCN profile (high CCN) shows that the increase in CCN concentration resulting from entrainment results in a a higher droplet concentration, smaller drop sizes, more liquid water retained in the cloud layer, and less drizzle reaching the surface. The smaller drops also cause a decrease in the surface shortwave flux.

Although the cloud optical properties respond to the change in CCN concentration from the beginning of the simulations with higher values of cloud albedo A and cloud optical depth $\tau$, liquid water path does not show significant variations until 4 h into the simulations.

The differences between the high CCN run and the low CCN run show characteristics of the differences between drizzling and non-drizzling clouds [e.g. Steven et al., 1998], namely that drizzle redistributes heat and vapor in the vertical in a manner that stabilizes the boundary layer. In the case under consideration we see a significant dynamical response to the entrained CCN concentration as reflected in more vigorous eddies.

Higher entrainment rates are associated with stronger cloud-top cooling. This is consistent with the study of Deardorff [1981], who showed that large values of cloud-top cooling are more favorable for entrainment and the production of mixed-layer convection.

We have shown that the contamination of the cloud is much less than the large jump in CCN at the inversion might imply because the contamination of the cloud by the polluted air aloft is limited by the rate at which the boundary layer top z_i rises. In spite of the more modest contamination there are still noticeable differences in cloud optical properties, and suppression of drizzle makes these differences even stronger.

This work underscores the fact that knowledge of boundary layer deepening is critical to prediction of cloud optical properties. This is true both from the thermodynamic point of view where the properties of the entrained air affect bulk cloud features such as LWP; as well as from the microscale point of view where aerosol gradients across the top of the boundary layer can alter microphysical processes, and in turn, cloud optical properties. This issue is even more critical for mixed phase clouds given the sensitivity of such clouds to even small concentrations (order 1 l^-1) of ice forming nuclei [Jiang et al., 1999].
 
 

Acknowledgments

B. Stevens is thanked for useful discussions. D. Baumgardner and K. Laursen are thanked for providing the initial sounding information. D. Rogers is thanked for providing the Ice Nuclei data used to generate Figure 9. This research was supported by grants from the NASA/FIRE III (NAG-1-2045). G. Feingold was partially supported by a NOAA Office of Global Programs Grant. Peter Duynkerke was supported for the participation in the SHEBA experiment by the Netherlands Geosciences Foundation (GOA) with financial aid (GRANT 750.295.03) from the Netherlands Organization for Scientific Research (NWO). The last author thanks NCAR and its sponsor, the National Science Foundation, for allowing him to use the C130 observational data. Finally the IMAU wishes to thank the many people of the FIRE/SHEBA community for giving logistical support by bringing the equipment to the SHEBA ice camp and for assisting on the site.