Simultaneous Variation of INP and CCN in a Simulated Thunderstorm Using a New Immersion Freezing Algorithm

Ice nucleating particles (INPs) and cloud condensation nuclei (CCN) influence precipitation by modifying the numbers of frozen and liquid hydrometeors, which, in turn, affect microphysical process rates. Numerous studies have investigated the sensitivity of deep convective clouds to CCN number concentrations. However, few have considered their sensitivity to INP concentrations, and far fewer have examined their sensitivity to simultaneous variations in both CCN and INP. The handful of studies that have varied both quantities suffer from poor observational constraints regarding the extent of their variation. By retaining volume-dependent (and aerosol-independent) raindrop freezing relations, such as that of Bigg (1953), they also fail to consider the role of aerosols in raindrop freezing.

A new dataset collected during the Clouds, Aerosols, and Complex Terrain Interactions (CACTI) field campaign includes both CCN number concentrations and immersion-freezing INP (IF-INP) spectra, providing an opportunity to examine their relative importance in simulations of deep convection in a manner consistent with observations. To this end, a new immersion freezing algorithm was developed, utilizing observed IF-INP and replacing the commonly used Bigg immersion freezing algorithm in most microphysics schemes. With the new algorithm, the immersion freezing of raindrops and cloud drops can both be described using a single IF-INP spectrum. A matrix of idealized simulations of a single case of deep convection observed during CACTI was created, where IF-INP temperature spectra and CCN number concentrations were independently altered from their measured values.

In all simulations, surface rainfall was primarily generated through the freezing of raindrops within the updraft, leading to riming graupel (and, to a far lesser extent, hail) that ultimately melted during its descent to the surface. As expected, rainfall accumulations and surface cold pools were substantially and systematically weakened with increased CCN, but both storm attributes were relatively insensitive to variations in both the magnitude and slope of the IF-INP spectra. Having more IF-INP active at higher temperatures increased graupel formation by freezing raindrops at lower altitudes in the updraft, irrespective of CCN, but minimally affected rainfall amounts due to other compensating factors. Further experimentation comparing the relative sensitivities of deep convective clouds in different thermodynamic environments and of different morphologies to IF-INP and CCN will also be presented.