1Department of Meteorology, University of Utah, Salt Lake City, Utah, USA
2Atmospheric Chemistry and Dynamics Branch, NASA Ames Research Center, Moffett Field, California, USA
3WB-57 Program Offices, NASA Johnson Space Center,Ellington Field, Houston, Texas, USA
4Universidad Nacional Autonoma de Mexico, Mexico City, Mexico
5Jet Propulsion Laboratory, 4800 Oak Grove Drive, Pasadena, California, USA
6NASA Langley Research Center, Hampton, Virginia, USA
Abstract. Pileus clouds form where humid, vertically stratified air is mechanically displaced ahead of rising convection. This paper describes convective formation of pileus cloud in the tropopause transition layer (TTL), and explores a possible link to the formation of long-lasting cirrus at cold temperatures. The study examines in detail in-situ measurements from off the coast of Honduras during the July 2002 CRYSTAL-FACE experiment that showed an example of TTL cirrus associated with, and penetrated by, deep convection. The TTL cirrus was enriched with total water compared to its surroundings, but was composed of extremely small ice crystals with effective radii between 2 and 4 μm. Through gravity wave analysis, and intercomparison of measured and simulated cloud microphysics, it is argued that the TTL cirrus originated neither from convectively-forced gravity wave motions nor environmental mixing alone. Rather, it is hypothesized that a combination of these two processes was involved in which, first, a pulse of convection forced pileus cloud to form from TTL air; second, the pileus layer was punctured by the convective pulse and received larger ice crystals through interfacial mixing; third, the addition of this condensate inhibited evaporation of the original pileus ice crystals where a convectively forced gravity wave entered its warm phase; fourth, through successive pulses of convection, a sheet of TTL cirrus formed. While the general incidence and longevity of pileus cloud remains unknown, in-situ measurements, and satellite-based Microwave Limb Sounder retrievals, suggest that much of the tropical TTL is sufficiently humid to be susceptible to its formation. Where these clouds form and persist, there is potential for an irreversible repartition from water vapor to ice at cold temperatures.