Atmospheric models with aerosols and cloud-aerosol interactions are tasked with making accurate predictions about Earth’s evolving cloud coverage. Their numerical solution is challenging in part due to the disaggregation of physical processes into separate model components, which are evaluated separately and then recombined via a coupling strategy. Predictions of cloud coverage in global models of Earth’s climate have been shown to exhibit high sensitivities to the choice of coupling, corresponding to different forms of numerical error. Here we conduct an error analysis of temporal coupling approaches in a simplified aerosol system that retains the key processes of emission, gravitational settling, dry removal, and turbulent mixing, which are found in large-scale atmospheric aerosol models. We present results comparing the accuracy and efficiency of both single and multi-rate coupling strategies in this idealized setting, including methods based on sequential and parallel splittings as well as forcing methods. We further consider when the coupling errors are expected to dominate and when alternative or higher-order coupling strategies may be more beneficial than standard couplings used in large-scale models.

This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. LLNL-ABS-867886.