Large‐eddy simulation of three mixed‐phase cloud events during ISDAC: Conditions for persistent heterogeneous ice formation

Abstract

AbstractA Classical‐Nucleation‐Theory‐based parameterization for heterogenous ice nucleation, including explicit dependencies of the nucleation rates on the number concentration, size, and composition of the ambient aerosol population, is implemented in a cloud‐scale, large‐eddy simulation model and evaluated against Arctic mixed‐phase cloud events observed during Indirect and Semi‐Direct Aerosol Campaign (ISDAC). An important feature of the parameterization is that the ice nucleation efficiency of each considered aerosol type is described using a contact angle distribution which evolves with time so that the model accounts for the inhibition of ice nucleation as the most efficient ice‐forming particles are nucleated and scavenged. The model gives a reasonable representation of first‐order (ice water paths) and second‐order (ice crystal size distributions) ice microphysical properties. The production of new ice crystals in the upper part of the cloud, essential to guarantee sustained mixed‐phase conditions, is found to be controlled mostly by the competition between radiative cooling (resulting in more aerosol particles becoming efficient ice nuclei as the temperature decreases), cloud‐top entrainment (entraining fresh particles into the cloud), and nucleation scavenging of the ice+forming aerosol particles. The relative contribution of each process is mostly determined by the cloud‐top temperature and the entrainment rates. Accounting for the evolution of the contact angle probability density function with time seems to be essential to capture the persistence of in‐cloud ice production without having to, for example, increase the free tropospheric aerosol concentration. Although limited to only three cases and despite important limitations of the parameterization (e.g., the present version only considers dust and black carbon as potential ice nuclei), the results suggest that modeling the time evolution of the ice nuclei population ability to form ice is required to accurately model Arctic mixed‐phase cloud processes.