Abstract
Abstract. A comprehensive ice nucleation parameterization has been implemented in the global chemistry-climate model EMAC to improve the representation of ice crystal number concentrations (ICNCs). The parameterization of Barahona and Nenes (2009, hereafter BN09) allows for the treatment of ice nucleation taking into account the competition for water vapour between homogeneous and heterogeneous nucleation in cirrus clouds. Furthermore, the influence of chemically heterogeneous, polydisperse aerosols is considered by applying one of the multiple ice nucleating particle parameterizations which are included in BN09 to compute the heterogeneously formed ice crystals. BN09 has been modified in order to consider the pre-existing ice crystal effect and implemented to operate both in the cirrus and in the mixed-phase regimes. Compared to the standard EMAC parameterizations, BN09 produces fewer ice crystals in the upper troposphere but higher ICNCs in the middle troposphere, especially in the Northern Hemisphere where ice nucleating mineral dust particles are relatively abundant. Overall, ICNCs agree well with the observations, especially in cold cirrus clouds (at temperatures below 205 K), although they are underestimated between 200 and 220 K. As BN09 takes into account processes which were previously neglected by the standard version of the model, it is recommended for future EMAC simulations.
Highlights
Clouds play an important role in the Earth system by affecting the global radiative energy budget, the hydrologic cycle, the scavenging of gaseous and particulate substances, and by providing a medium for aqueous-phase chemical reactions
Sensitivity studies have shown that global means of ice crystal number concentrations (ICNCs) vary by up to a factor of 20 according to the ice nucleating particles (INPs) parameterization used and empirical-based parameterizations better agree with observations, while classical nucleation theory (CNT) overestimates the number of ice crystals (Barahona et al, 2010; Sullivan et al, 2016)
We have tested the BN09 scheme operating in the cirrus and/or in the mixed-phase regimes and compared the results with the standard configuration of the model, which assumes that cirrus clouds form via pure homogeneous nucleation (Kärcher and Lohmann, 2002) and uses the immersion nucleation parameterization of Lohmann and Diehl (2006) for mixed-phase clouds
Summary
Clouds play an important role in the Earth system by affecting the global radiative energy budget, the hydrologic cycle, the scavenging of gaseous and particulate substances, and by providing a medium for aqueous-phase chemical reactions. Barahona and Nenes (2009) introduced an ice cloud formation parameterization, based on the analytical solution of the cloud parcel model equations, which calculates the competition for water vapour between homogeneous and heterogeneous nucleation and takes into account the variability (in size and chemical composition) of different aerosol species through a variety of INP parameterizations. Since these parameterizations have been included in GCMs in order to better predict cloud phase partitioning. The paper is organized as follows: the description of the operational model and the BN09 scheme are in Sect. 2, as well as the information about the implementation work and the simulations run for this study, Sect. 3 describes the modelled ice-related products, Sect. 4 contains the evaluation of the model, and Sect. 5 presents our conclusions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
