On the Dependence of Cirrus Parametrizations on the Cloud Origin

Particle size distributions (PSDs) for cirrus clouds are important for both climate models as well as many remote sensing retrieval methods. Therefore, PSD parametrizations are required. This study presents parametrizations of Arctic cirrus PSDs. The dataset used for this purpose originates from balloon-borne measurements carried out during winter above Kiruna (Sweden), i.e. north of the Arctic circle. The observations are sorted into two types of cirrus cloud origin, either in-situ or liquid. The cloud origin describes the formation pathway of the ice particles. At temperatures below −38 °C, ice particles form in-situ from solution or ice nucleating-aerosol particles. Liquid origin ice particles have formed at temperatures warmer than or equal to −38 °C, either via ice-nucleating particles embedded in liquid drops or via homogeneous drop freezing, and are then further uplifted to the cirrus temperature regime.

In order to derive parametrizations for each cloud origin, the observed PSDs are represented by gamma functions. The gamma coefficients exhibit large differences with regard to cloud origin. Functions describing the relationships in between the gamma coefficients and with temperature are fitted. These functions for Arctic cirrus confirm established parametrizations for continental cirrus sorted by two particle size modes but differ from others depending only on temperature. We suppose that the agreement between the parametrizations of the geographically different cirrus is because in-situ and liquid origin cirrus also distinguish by particle size modes. Since cloud sorting by their origin is based on physical processes that are independent of geographical region, we further hypothesize that these cloud-type-based parametrizations might be generally valid for use in global models and satellite retrievals, given the distribution of the cloud types is known.