Abstract
Abstract. The homogeneous nucleation of ice in supercooled liquid-water clouds is characterized by time-dependent freezing rates. By contrast, water phase transitions induced heterogeneously by ice-nucleating particles (INPs) are described by time-independent ice-active fractions depending on ice supersaturation (s). Laboratory studies report ice-active particle number fractions (AFs) that are cumulative in s. Cloud models budget INP and ice crystal numbers to conserve total particle number during water phase transitions. Here, we show that ice formation from INPs with time-independent nucleation behavior is overpredicted when models budget particle numbers and at the same time derive ice crystal numbers from s-cumulative AFs. This causes a bias towards heterogeneous ice formation in situations where INPs compete with homogeneous droplet freezing during cloud formation. We resolve this issue by introducing differential AFs, thereby moving us one step closer to more robust simulations of aerosol–cloud interactions.
Highlights
A wide variety of macromolecular or proteinaceous, crystalline, glassy, and solid aerosol particles act as ice-nucleating particles (INPs) in the atmosphere and participate in the formation of cirrus or in the glaciation of supercooled liquid-water clouds (Kanji et al, 2017)
In conditions below liquid-water saturation, deposition nucleation occurring in the absence of liquid water has traditionally been considered the most relevant heterogeneous ice formation mode (Vali et al, 2015)
Water phase transitions in clouds induced by INPs with deterministic ice nucleation behavior are described by timeindependent active particle number fractions (AFs) that are cumulative in ice supersaturation
Summary
A wide variety of macromolecular or proteinaceous, crystalline, glassy, and solid aerosol particles act as INPs in the atmosphere and participate in the formation of cirrus or in the glaciation of supercooled liquid-water clouds (Kanji et al, 2017). PCF is described by a deterministic parameterization as well, as ice formation in this mode is determined by the relative humidities required either for pore water condensation or ice growth with no stochastic component involved when temperatures are well below the threshold for homogeneous freezing of supercooled solution droplets, which is the case at cirrus temperatures (Marcolli, 2020; Marcolli et al, 2021). The time dependence of condensation freezing is determined by the process of cloud droplet activation, and ice nucleation can be considered immediate once the INP is immersed in water For these reasons, a formulation of AFs as φ(s) without explicit time dependence is recommended for all modes of ice formation initiated by INPs
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
