Impacts of Subgrid Heterogeneous Mixing between Cloud Liquid and Cloud Ice on Wegner-Bergeron-Findeisen Process and Mixed-phase Clouds in CAM5

Description

Mixed-phase clouds are persistently observed in the Arctic and the phase partition of cloud liquid and ice in mixed-phase clouds has the important impacts on the surface energy budget and Arctic climate. Our model validations against ARM ground-based remote sensing retrievals as well as global satellite measurements show that the Community Atmosphere Model Version 5 (CAM5), like other global climate models, poorly simulates the phase partition in the mixed-phase clouds by significantly underestimating the supercooled liquid water content. In this study, to reduce modeled mixed-phase cloud biases, we improve the representations of two critical processes in CAM5 determining the phase partitioning in mixed-phase clouds by (1) slowing down the Wegner-Bergeron-Findeisen (WBF) process by accounting for pocket structures in the distribution of cloud liquid and cloud ice inside mixed-phase clouds; and (2) improving the treatment of ice nucleating particles (INPs) by accounting for scavenging of INPs due to ice nucleation and regeneration of INPs due to evaporation of ice crystals. The modification of WBF process in CAM5 has been achieved by applying a random number in the parameters controlling the time scale of WBF relevant to both ice and snow, by assuming the heterogeneous distributions of cloud liquid and ice in mixed-phase clouds. Our results show that this modification of WBF process improves the modeled phase partition in mixed-phase clouds. The seasonality of mixed-phase cloud properties is also better captured in the model when comparing with the long-term ground-based remote sensing observations. Treating the scavenging of INPs further increases the liquid water content and improves the vertical structure of clouds.