Revisiting the definition of mixed-phase clouds in satellite remote sensing

Satellite observations with passive sensors generally classify cloud phase – liquid water, mixed-phase, or ice – based on snap-shots of polar-orbiting observations or individual time steps of geostationary observations. In that context, mixed-phase clouds are defined as cloud objects that contain both pixels that are classified as liquid water and as ice.

In contrast to polar-orbiting satellites, observations with geostationary satellites provide the data needed for tracking clouds over their lifetime. This Lagrangian perspective allows for quantifying the evolution of cloud physical properties, including cloud phase. The temporally-resolved view of a cloud as a sequence of subsequent observations provides a refined perspective of mixed-phase clouds as objects that evolve over time. There are three straightforward options. First, a cloud starts as all-liquid pixels, contains at least one time step that feature both liquid and ice pixels, and ends as ice-only pixels. Second, the cloud features only time steps that are all liquid or all ice pixels (the former option without the mixed-phase state as defined in snap-shots). Third, the cloud contains both liquid and ice pixels at any time step throughout its lifetime. This revised perspective of mixed-phase clouds challenges the static snap-shot view which would identify a cloud as mixed-phase only if it was at the mid-phase of option one or an option-three cloud.

The purpose of this poster is to stimulate a discussion on the need for a time-resolved definition of mixed-phase clouds, on how to reconcile such a definition with snap-shot-based observations, and on what can be learned from the time-resolved definition of mixed-phase clouds.