Towards a particle-resolved aerosol representation for the simulation of the aerosol impact on regional scales

Poster PDF

Description

Individual aerosol particles are a complex mixture of a wide variety of species, such as soluble inorganic salts and acids, insoluble crustal materials, trace metals, and carbonaceous materials, so modeling them accurately requires a high-dimensional representation. The capabilities of traditional models to treat this high dimensionality are currently very limited, and this introduces serious shortcomings in our understanding of the impact of aerosol particles on climate. To improve this, we have recently introduced the particle-resolved model PartMC-MOSAIC (Riemer et al. 2009), a new approach to examining the evolution of aerosol properties without making a priori assumptions about the evolution of particle composition. While particle-resolved aerosol simulations offer unprecedented resolution of the aerosol mixing state, they are significantly more expensive than conventional models. In this project we explore a number of strategies to improve the efficiency of PartMC-MOSAIC to enable its extension from a Lagrangian box model to a model that is coupled to a 1D and eventually 3D transport code. Here we will present advances regarding two specific aspects: (1) the development of new algorithms for parallel particle simulation, and (2) the introduction of “superparticles” as a coarse-graining method, each superparticle representing a distribution of particles. We found such a modified PartMC-MOSAIC scheme to be more accurate and efficient than the original method by several orders of magnitude when applied to an urban plume scenario where we simulated the evolution of the mixing state of an aerosol population in a polluted environment.