Lagrangian condensation microphysics with Twomey CCN activation

Abstract

Abstract. We report the development of a novel Lagrangian microphysics methodology for simulations of warm icefree clouds. The approach applies the traditional Eulerian method for the momentum and continuous thermodynamic fields such as the temperature and water vapor mixing ratio, and uses Lagrangian "super-droplets" to represent condensed phase such as cloud droplets and drizzle or rain drops. In other applications of the Lagrangian warm-rain microphysics, the super-droplets outside clouds represent unactivated cloud condensation nuclei (CCN) that become activated upon entering a cloud and can further grow through diffusional and collisional processes. The original methodology allows for the detailed study of not only effects of CCN on cloud microphysics and dynamics, but also CCN processing by a cloud. However, when cloud processing is not of interest, a simpler and computationally more efficient approach can be used with super-droplets forming only when CCN is activated and no super-droplet existing outside a cloud. This is possible by applying the Twomey activation scheme where the local supersaturation dictates the concentration of cloud droplets that need to be present inside a cloudy volume, as typically used in Eulerian bin microphysics schemes. Since a cloud volume is a small fraction of the computational domain volume, the Twomey super-droplets provide significant computational advantage when compared to the original superdroplet methodology. Additional advantage comes from significantly longer time steps that can be used when modeling of CCN deliquescence is avoided. Moreover, other formulation of the droplet activation can be applied in case of low vertical resolution of the host model, for instance, linking the concentration of activated cloud droplets to the local updraft speed. This paper discusses the development and testing of the Twomey super-droplet methodology, focusing on the activation and diffusional growth. Details of the activation implementation, transport of super-droplets in the physical space, and the coupling between super-droplets and the Eulerian temperature and water vapor field are discussed in detail. Some of these are relevant to the original superdroplet methodology as well and to the ice phase modeling using the Lagrangian approach. As a computational example, the scheme is applied to an idealized moist thermal rising in a stratified environment, with the original superdroplet methodology providing a benchmark to which the new scheme is compared.

Cloud droplet size distribution broadening during diffusional growth: ripening amplified by deactivation and reactivation

“…Our results suggest the importance of solute and curvature effects to the deactivation and reactivation processes, which are consistent with previous studies (e.g., Andrejczuk et al, 2008;Hoffmann et al, 2015;Hoffmann, 2017;Chen et al, 2018). However the results are counter to some other studies where details of activation and deactivation are argued to be unimportant in the cloud simulation (e.g., Srivastava, 1991;Chuang et al, 1997;Grabowski et al, 2018). Large eddy simulations with a similar microphysical treatment would be useful to investigate how important this mechanism is to CDSD broadening in more realistic clouds.…”

“…First, turbulenceinduced mixing and entrainment can trigger in-cloud activation of haze particles, which can broaden the left branch of the size distribution (e.g., Khain et al, 2000;Devenish et al, 2012;Yang et al, 2016;Grabowski et al, 2018). Secondly, giant cloud condensational nuclei (GCCN, usually defined as aerosols with a dry diameter larger than a few µm) provides an embryo for large droplets, which can broaden the right branch of the size distribution and can be important for warm rain initiation (e.g., Johnson, 1982;Feingold et al, 1999;Yin et al, 2000;Jensen and Lee, 2008;Cheng et al, 2009).…”

University of Warsaw Lagrangian Cloud Model (UWLCM) 1.0: a modern Large-Eddy Simulation tool for warm cloud modeling with Lagrangian microphysics

“…To calculate the 5 velocity of air that advects a given SD, velocities, which reside at edges of the dual grid, are interpolated to the position of the SD. The interpolation is done linearly, separately in each dimension, as advocated by Grabowski et al (2018a). Spatial discretization is also necessary in the algorithm for modeling collision-coalescence (cf.…”

“…Advection of SDs is modeled with a predictor-corrector algorithm described in Grabowski et al (2018a). Simpler, first-order algorithms for advection were found to cause inhomogeneous spatial distributions of SDs, with less SDs in regions with high vorticity.…”

Improving collisional growth in Lagrangian cloud models: development and verification of a new splitting algorithm

“…where t is the length of the collection time step, V the volume of the grid box, r n the radius of a droplet represented by superdroplet n, and K is the collection kernel (based on Hall, 1980, for this study). Since p mn is usually smaller than 1, collections only occur if p mn > ξ , where ξ is a random number uniformly chosen from the interval [0, 1].…”