Cloud resolving modelling of cirrus clouds

Recently, a state-of-the-art bulk ice microphysics scheme was developed, validated and implemented (Spichtinger & Gierens, 2009a) into the anelastic non-hydrostatic model EULAG (Prusa et al., 2008). The ice microphysics is based on a consistent two-moment approach (i.e. prognostic equations for ice crystal mass and number concentration) with an underlying ice crystal mass distribution of log-normal type. The following processes are included: 

• Nucleation of ice crystals (different mechanisms, e.g. homogeneous freezing of aqueous solution droplets, heterogeneous nucleation)
• Diffusional growth/evaporation of ice crystals, corrections for kinetic regime and ventilation are applied
• Sedimentation of ice crystals (number/mass weighted terminal velocities)

In the model, arbitrary many classes of ice can be used, only discriminated by the formation mechanism. Each class of ice is dual to a background aerosol, i.e. each class consists of the following quadruple of variables:

1. ice crystal mass concentration
2. ice crystal number concentration
3. aerosol mass concentration
4. aerosol number concentration

This approach allows us to parametrise explicit impact of aerosols on the formation mechanisms. Additionally, we are able to discriminate between the effects of different formation mechanisms. The aerosols are trapped in the cloud particles after nucleation and are released after evaporation of the crystals.
For homogeneous nucleation we use the water activity based nucleation rates as described by Koop et al. (2000). Here, the background aerosol represents sulphuric acid droplets, which can grow by collecting water molecules due to Köhler's theory. For heterogeneous nucleation several empirical parametrisations are used. For instance, we use a fixed relative humidity threshold; by surpassing the threshold all aerosol particles act as ice nuclei. This new cirrus cloud model is currently used for several studies on the formation and evolution of cirrus clouds. The main focus is here on the impact of large scale dynamics (e.g. Spichtinger & Gierens, 2009b) or even gravity waves (Joos et al., 2009). However, also the competition of different nucleation mechanisms was investigated recently (Spichtinger & Gierens, 2009c). Additionally, the model is used for investigations of contrails in a cirrus cloud forming environment (Unterstrasser et al., 2008).