Simulating Sources and Sinks of Aerosol to predict Southern Ocean Cloud Droplet Concentrations in the UK Met Office Climate Model

Most or all climate models struggle to accurately simulate clouds and radiation balance in the Southern Ocean. The Met Office Unified Model is no exception. In this study, we attempt to better understand and correct some of these biases by focusing on aerosol-cloud interactions in liquid clouds. We use a convection-permitting regional nested model configuration to simulate realistically air motion in extra-tropical cyclones sampled by the MARCUS and SOCRATES field campaigns. We use prognostic double-moment bulk aerosol and cloud microphysics schemes to simulate aerosol and cloud processes as realistically as possible.

We find that the model substantially under-predicts the concentration of aerosols capable of acting as CCN, and correspondingly under-predicts cloud droplet concentrations measured by the MODIS satellite instruments and aircraft droplet probes. We hypothesize that this under-prediction is primarily the result of missing aerosol sources, but it could also be exacerbated by over-active aerosol sinks. Using sensitivity studies, we examine the potential roles of new particle formation from sulfur-containing species, condensation of vapors on newly formed aerosol, and primary aerosol emissions, in contributing to the CCN budget. We also examine the strength of precipitation scavenging.

As expected, we find new particle formation may contribute a high fraction of CCN concentrations. Using tracer species in our model to understand the transport of precursors and newly formed aerosols, we test current understanding of where this new particle formation frequently occurs and of the timescales for mixing of CCN formed in the free troposphere into the boundary layer. We investigate what changes to the model are needed in order for Southern Ocean cloud droplet concentrations to be simulated more accurately, and the implications for radiative fluxes.