Sensitivities of Large Eddy Simulations of Aerosol Plume Transport and Cloud Response

Abstract

Cloud responses to surface-based sources of aerosol perturbation partially depend on how turbulent transport of the aerosol to cloud base affects the spatial and temporal distribution of aerosol. Here, scenarios of plume injection below a marine stratocumulus cloud are modeled using large eddy simulations coupled to a prognostic bulk aerosol and cloud microphysics scheme. Both passive plumes, consisting of an inert tracer, and active plumes are investigated, where the latter are representative of saltwater droplet plumes such as have been proposed for marine cloud brightening. Passive plume scenarios show higher in-plume cloud brightness (relative to out-of-plume) due to the predominant transport of the passive plume tracer from the near-surface to the cloud layer within updrafts. These updrafts rise into brighter areas within the cloud deck, even in the absence of an aerosol perturbation associated with an active plume. Comparing albedo at in-plume to out-of-plume locations associates the inert plume with the brightest cloud locations, without any causal effect of the plume on the cloud. Numerical sensitivities are first assessed to establish a suitable model configuration. Then sensitivity to particle injection rate is investigated. Trade-offs are identified between the number of injected particles and the suppressive effect of droplet evaporation on plume loft and spread. Furthermore, as the near-field in-plume brightening effect does not depend significantly on injection rate given a suitable definition of perturbed versus unperturbed regions of the flow, plume area is a key controlling factor on the overall cloud brightening effect of an aerosol perturbation.