Responses of Boundary Layer Cloud Structure and Particle Budget to Aerosol Sources During Arctic Marine Cold‐Air Outbreaks

Abstract Marine cold‐air outbreaks (MCAOs) strongly modulate Arctic air‐sea exchange, yet their mixed‐phase clouds remain difficult to represent in Earth system models. We simulate an MCAO case using Lagrangian large‐eddy simulations that couple double‐moment microphysics with prognostic aerosol and ice formation parameterizations. We then apply a self‐supervised machine‐learning approach to objectively quantify how long cloud rolls persist before transitioning to convective cells. The prognostic treatments yield more realistic initial cloud development over a period of hours. Higher ice number concentrations consistently shorten roll persistence through boundary layer (BL) decoupling caused by snow sublimation. Snow accretion is the dominant aerosol sink. Single‐source experiments reveal state‐dependent aerosol transport: bottom‐up supply is effective only while the BL remains coupled, whereas top‐down transport via entrainment mixing remains effective after decoupling. These results emphasize the need for aerosol‐aware, ice‐coupled parameterizations that represent both source pathways and wet removal in MCAOs to accurately predict roll‐to‐cell transitions.