You are here

From 100,000 to 8: Representing Complex Aerosol Patterns with Far Fewer Particles
A sparse set of weighted particle (stems) accurately and efficiently represents realistically complex aerosol distributions simulated by a detailed particle-resolved model (number density distributions). Number density distributions with respect to particles' dry diameter and hygroscopicity parameter are a projection of the 20-dimensional aerosol mixing state distribution, sampled at three time steps from the particle-resolved model PartMC-MOSAIC. The particle-resolved model requires simulation of 10,000 to 100,000 particles; the new sparse-particle framework represents cloud condensation nuclei activation of these complex distributions using only 8 particles.

The Science

Capturing processes and properties across multiple scales is a big challenge in simulating aerosols interacting with clouds. How aerosols impact clouds depends on particle-level variation in size and composition. However, small-scale complexity is not easily captured in large-scale atmospheric models. Today's models of large cities, ecosystems, and the global atmosphere simplify the aerosols. The result? Errors in the presentation of aerosol effects on clouds and energy transfer. Researchers created a framework that represents the different sizes and compositions of aerosols using only 8 weighted particles. This simple substitution adequately captures the complexity of the system, but can more realistically be used for calculations.

The Impact

Current simulations are either too simple to really study climate-relevant aerosols or too complex to give answers about large areas. The study is a first step in creating a simulation that is neither too simple nor too complex. The new method enables accurately and efficiently representing key features in large-scale atmospheric models.


Scientists describe a new technique for constructing sparse representations of realistically complex aerosol populations from distribution moments. The study shows that cloud condensation nuclei activity of particle-resolved simulations, which track tens to hundreds of thousands of computational particles, are accurately represented using only a few sparse particles. This sparse representation of the aerosol mixing state, designed for use in quadrature-based moment models, is constructed from a linear program constrained by low-order moments and combined with an entropy-inspired cost function. The critical supersaturation at which each sparse particle becomes an active cloud condensation nucleus is computed as a function of its size and composition. Continuous cloud condensation nuclei activation spectra are then computed from the sparse critical supersaturation values using constrained maximum entropy distributions. Unlike reduced representations common to large-scale atmospheric models, such as modal and sectional schemes, the researchers' approach is not confined to pre-determined size bins or assumed distribution shapes. This study is a first step towards a quadrature-based aerosol scheme that will track multivariate aerosol distributions with both reliable accuracy and sufficient computational efficiency for large-scale simulations.


BER Program Managers
Ashley Williamson
Atmospheric System Research Program Manager
Department of Energy

Shaima Nasiri
Atmospheric System Research Program Manager
Department of Energy

Principal Investigator
Laura Fierce
Brookhaven National Laboratory


L.M.F. is supported by the University Corporation for Atmospheric Research (UCAR) through a National Oceanic and Atmospheric Administration Climate & Global Change Postdoctoral Fellowship and the Department of Energy, Office of Science, Office of Biological and Environmental Research, Atmospheric System Research program. R.L.M. is supported by the Department of Energy, Office of Science, Office of Biological and Environmental Research, Atmospheric System Research program.


L. Fierce and R.L. McGraw, "Multivariate quadrature for representing cloud condensation nuclei activity of aerosol populationsExternal link." Journal of Geophysical Research: Atmospheres 122, 9867 (2017). [DOI: 10.1002/2016JD026335]

Related Links

Department of Energy Atmospheric System Research: New Method for Efficiently Representing Complex Aerosol Distributions

Highlight Categories

Program: BER, CESD

Performer/Facility: DOE Laboratory

Additional: Collaborations, Non-DOE Interagency Collaboration