Constraining the uncertainty associated with sea salt aerosol parameterizations in global models using nudged UKESM1-AMIP simulations

Sea salt is the largest source of natural aerosol in the atmosphere by mass. Formed when ocean waves break and bubbles burst, sea salt aerosols (SSA) influence Earth’s climate via direct and indirect processes. Models participating in the sixth Coupled Model Intercomparison project (CMIP6) demonstrate a negative effective radiative forcing when SSA emissions are doubled. However, the magnitude of the effective radiative forcing ranges widely from -0.35 +/- 0.04 W/m^2 to -2.28 +/- 0.07 W/m^2, with the largest difference over the Southern Ocean. Differences in the response to doubled SSA emissions arise from model uncertainty (e.g. individual model physics, aerosol size distribution) and parameterization uncertainty (e.g. how SSA is produced in the model). Here, we perform single-model experiments with UKESM1-AMIP incorporating all of the SSA parameterizations used by the current generation of CMIP6 Earth system models. Using a fixed SSA size distribution, our experiments show that the parameterization uncertainty causes large inter-model diversity in SSA emissions in the models, particularly over the tropics and the Southern Ocean. The choice of parameterization influences the ambient aerosol size distribution, cloud condensation nuclei and cloud droplet number concentrations, and therefore direct and indirect radiative forcing. We recommend that modelling groups evaluate their SSA parameterizations and update them where necessary in preparation for future model intercomparison activities