The Impact of Mixed-Phase Cloud Processes on Simulating Southern Ocean Clouds and Their Radiative Effect

Over the Southern Ocean, atmospheric and climate models have large biases in their radiative fluxes, primarily caused by the representation of supercooled liquid and mixed-phase low-level clouds, both at the macro- and micro-scale. The radiation biases lead to errors in simulated sea surface temperature, sea ice properties and large-scale atmospheric circulation. We assess the performance of a convection-permitting configuration of the Met Office Unified Model in simulating cloud over the Southern Ocean. We utilize aircraft and satellite observations from several cases in February 2023 during the Southern Ocean Clouds field experiment. We investigate the representation of three mixed-phase characteristics, namely ice nucleating particles (INP), the droplet number concentration and the spatial distribution of liquid and ice in mixed-phase clouds. A lower temperature-dependent INP concentration (based on synchronous INP measurements) results in lower ice mass and number concentrations that are closer to observations, and a higher cloud liquid water content. This reduces the net surface cloud radiative effect by up to 14 W m−2. Reducing the droplet number to the campaign average had a similar sized impact on the cloud radiative effect (up to 22 W m−2) but opposite in sign, highlighting these compensating errors. Similarly, changing how well mixed the clouds are leads to a large sensitivity in the cloud radiative effect (up to 31 W m−2). All three mixed-phase processes play a crucial role in correctly modeling mixed-phase clouds and their impact on the radiation budget over the Southern Ocean.