Observations and modelling of microphysical variability, aggregation and sedimentation in tropical anvil cirrus outflow regions

Abstract

Abstract. Aircraft measurements of the microphysics of a tropical convective anvil (at temperatures ~−60 °C) forming above the Hector storm, over the Tiwi Islands, Northern Australia, have been conducted with a view to determining ice crystal aggregation efficiencies from in situ measurements. The observed microphysics have been compared to an explicit bin-microphysical model of the anvil region, which includes crystal growth by vapour diffusion and aggregation and the process of differential sedimentation. It has been found in flights made using straight and level runs perpendicular to the storm that the number of ice crystals initially decreased with distance from the storm as aggregation took place resulting in larger crystals, followed by their loss from the cloud layer due to sedimentation. The net result was that the mass (i.e. Ice Water Content) in the anvil Ci cloud decreased, but also that the average particle size (weighted by number) remained relatively constant along the length of the anvil outflow. Comparisons with the explicit microphysics model showed that the changes in the shapes of the ice crystal spectra as a function of distance from the storm could be explained by the model if the aggregation efficiency was set to values of Eagg~0.5 and higher. This result is supported by recent literature on aggregation efficiencies for complex ice particles and suggests that either the mechanism of particle interlocking is important to the aggregation process, or that other effects are occuring, such as enhancement of ice-aggregation by high electric fields that arise as a consequence of charge separation within the storm. It is noteworthy that this value of the ice crystal aggregation efficiency is much larger than values used in cloud resolving models at these temperatures, which typically use E~0.0016. These results are important to understanding how cold clouds evolve in time and for the treatment of the evolution of tropical Ci in numerical models.