Saharan dust, convective lofting, aerosol enhancement zones, and potential impacts on ice nucleation in the tropical upper troposphere.

Dry aerosol size distributions and scattering coefficients were measured on 10 flights in 32 clear-air regions adjacent to tropical storm anvils over the eastern Atlantic Ocean. Aerosol properties in these regions were compared with those from background air in the upper troposphere at least 40 km from clouds. Median values for aerosol scattering coefficient and particle number concentration >0.3 µm diameter were higher at the anvil edges than in background air, showing that convective clouds loft particles from the lower troposphere to the upper troposphere. These differences are statistically significant. The aerosol enhancement zones extended ~10-15 km horizontally and ~0.25 km vertically below anvil cloud edges but were not due to hygroscopic growth since particles were measured under dry conditions. Number concentrations of particles >0.3 µm diameter were enhanced more for the cases where Saharan dust layers were identified below the clouds with airborne lidar. Median number concentrations in this size range increased from ~100 l-1 in background air to ~400 l-1 adjacent to cloud edges with dust below, with larger enhancements for stronger storm systems. Integration with satellite cloud frequency data indicates that this transfer of large particles from low to high altitudes by convection has little impact on dust concentrations within the Saharan Air Layer itself. However, it can lead to substantial enhancement in large dust particles and, therefore, heterogeneous ice nuclei in the upper troposphere over the Atlantic. This may induce a cloud/aerosol feedback effect that could impact cloud properties in the region and downwind.

Theory of growth by differential sedimentation, with application to snowflake formation.

A simple model of irreversible aggregation under differential sedimentation of particles in a fluid is presented. The structure of the aggregates produced by this process is found to feed back on the dynamics in such a way as to stabilize both the exponents controlling the growth rate, and the fractal dimension of the clusters produced at readily predictable values. The aggregation of ice crystals to form snowflakes is considered as a potential application of the model.

Evidence that nitric acid increases relative humidity in low-temperature cirrus clouds.

In situ measurements of the relative humidity with respect to ice (RHi) and of nitric acid (HNO3) were made in both natural and contrail cirrus clouds in the upper troposphere. At temperatures lower than 202 kelvin, RHi values show a sharp increase to average values of over 130% in both cloud types. These enhanced RHi values are attributed to the presence of a new class of HNO3-containing ice particles (Delta-ice). We propose that surface HNO3 molecules prevent the ice/vapor system from reaching equilibrium by a mechanism similar to that of freezing point depression by antifreeze proteins. Delta-ice represents a new link between global climate and natural and anthropogenic nitrogen oxide emissions. Including Delta-ice in climate models will alter simulated cirrus properties and the distribution of upper tropospheric water vapor.

Effects of ice-crystal structure on halo formation: cirrus cloud experimental and ray-tracing modeling studies.

During the 1986 Project FIRE (First International Satellite Cloud Climatology Project Regional Experiment) field campaign, four 22° halo-producing cirrus clouds were studied jointly from a groundbased polarization lidar and an instrumented aircraft. The lidar data show the vertical cloud structure and the relative position of the aircraft, which collected a total of 84 slides by impaction, preserving the ice crystals for later microscopic examination. Although many particles were too fragile to survive impaction intact, a large fraction of the identifiable crystals were columns and radial bullet rosettes, with both displaying internal cavitations, and radial plate-column combinations. Particles that were solid or displayed only a slight amount of internal structure were relatively rare, which shows that the usual model postulated by halo theorists, i.e., the randomly oriented, solid hexagonal crystal, is inappropriate for typical cirrus clouds. With the aid of new ray-tracing simulations for hexagonal hollow ended column and bullet-rosette models, we evaluate the effects of more realistic ice-crystal structures on halo formation and lidar depolarization and consider why the common halo is not more common in cirrus clouds.

Infrared transmission through cirrus clouds: a radiative model for target detection.

An IR transmission model for thin and subvisual cirrus clouds composed of hexagonal ice crystals with a specific use for target detection has been developed. The present model includes parameterizations of the ice crystal size distribution and the position of cirrus clouds in terms of ambient temperature. To facilitate the scattering and absorption calculations for hexagonal column and plate crystals in connection with transmission calculations, we have developed parameterized equations for their single scattering properties by using the results computed from a geometric ray-tracing program. The successive order-of-scattering approach has been used to account for multiple scattering of ice crystals associated with a target-detector system. The direct radiance, path radiance, and radiances produced by multiple scattering and background radiation involving cirrus clouds have been computed for 3.7- and 10-,microm wavelengths. We show that the background radiance at the 3.7-,microm wavelength is relatively small so that a high contrast may be obtained using this wavelength for the detection of airborne and ground-based objects in the presence of thin cirrus clouds. Finally, using the present model, including a simple prediction scheme for the ice crystal size distribution and cloud position, the transmission of infrared radiation through cirrus clouds can be efficiently evaluated if the target-detector geometry is defined.