Abstract: Aircraft icing poses a significant threat to aviation safety, particularly under supercooled large droplet (SLD) conditions, where larger droplet sizes can lead to refreezing and thrust loss in components such as engine inlet lips and fan blades. Addressing the insufficient research on the vertical structure and evolution of cloud systems under SLD icing conditions, this study selects three typical flight experiments - F06, F18, and F25 - from the ICICLE project as research subjects. First, based on ICICLE flight observation data, a comparative analysis is conducted on the distribution characteristics of key icing parameters, including liquid water content (LWC), median volume diameter (MVD), and number concentration (N). The results indicate that the occurrence of SLD is closely related to the distributions of LWC, droplet size scales, and N, exhibiting evident height dependency. Second, the WRF model is employed to numerically simulate the icing meteoro-logical environments during the three flight experiments. The findings show that the model can accurately reproduce the vertical distributions of temperature and relative humidity (RH) within the flight altitude range and demonstrates good sim-ulation capability for liquid water in icing cloud layers. However, uncertainties persist in predicting the peak positions and fine-scale structures of liquid water, primarily due to simplified assumptions in cloud microphysics schemes. Further anal-ysis of cloud system structures and their temporal evolutions reveals significant differences among the experiments: F06 aligns with the typical development path from warm clouds to freezing precipitation, facilitating SLD formation, including freezing rain and freezing drizzle conditions; F18 exhibits stable mixed-phase cloud system characteristics, indicating prolonged icing risks; while F25 features rapid ice-phase development and significant liquid water consumption, which is unfavorable for sustained icing conditions. Finally, by integrating the distribution structures and evolutionary characteris-tics of cloud hydrometeors, the influence mechanisms of different cloud system types on icing risks under Appendix C and Appendix O conditions are elucidated, providing valuable references for optimizing aircraft engine anti/de-icing sys-tems and airworthiness certification.

Key words: aircraft icing, supercooled large drop, cloud system evolution, WRF model, ICICLE observations