Skip to main content
U.S. flag

An official website of the United States government

Publication Date
20 July 2018

New Aerosol Modeling Approach Takes Cues from Lab Experiments

A dynamic empirical framework for secondary organic aerosols could help bridge a significant gap between measurements and model predictions.
Print / PDF
Powerpoint Slide

Atmospheric secondary organic aerosols (SOA) particles affect air quality, visibility, human health, clouds, and radiation. Unseen by the naked eye, large quantities of carbon-containing vapors enter the atmosphere as they escape from trees, fossil-fuel burning, and forest fires. The atmosphere acts as a large chemical reactor, "cooking" these emissions to form millions of new carbon-containing molecules. Some of these molecules then condense into SOA particles that can change clouds, precipitation, and the amount of solar energy reaching the Earth. SOA particles are extremely complex: they comprise thousands of organic molecules, many of which have not been explicitly identified. Models use simplified SOA yields derived by fitting laboratory measurements of SOA made within environmental chambers. However, these yields are theory-specific and depend on dynamic SOA processes, which are not routinely included when these yields are derived from chamber measurements. Researchers at the U.S. Department of Energy’s Pacific Northwest National Laboratory (PNNL) led the development of a new modeling approach to derive isoprene SOA yields while explicitly including complex nitrogen-oxides-dependent multigenerational chemistry processes of SOA precursors and the losses of SOA vapors and particles to the chamber walls, which were represented by a few tunable parameters. Isoprene is the dominant volatile organic carbon emitted from trees on a global basis, and is ideal for studying SOA formation due to its potential importance. Their study showed that SOA yields are highly sensitive to processes that are included during fitting of these yields. The findings from this study help fill in important details of SOA formation processes that are still not fully understood.


This modeling approach is an important step in  parameterizing complex, dynamic SOA chemistry processes based on laboratory measurements for use in three-dimensional chemical transport models in a self-consistent way. Ultimately, this research will improve the representation of complex aerosol processes using empirical approaches within atmospheric chemical transport models and increase confidence in the models’ ability to represent aerosol-cloud-radiation interactions.


In laboratory experiments designed to study SOA formation and properties, SOA yields are shown to vary as a function of experimental conditions (e.g., nitrogen oxide levels and specific SOA chemistry processes included during derivation of these yields). In this work, researchers developed a self-consistent and computationally efficient framework that captures the dynamic evolution of SOA particles observed in the PNNL environmental chamber. The framework explicitly accounts for complex multigenerational chemistry of SOA precursors, including gas-phase fragmentation (i.e. breaking of carbon-carbon bonds in the gas phase), particle-phase oligomerization (a process in which smaller molecules bond together to form bigger molecules), and losses of gases and particles within the PNNL environmental chamber. Because SOA chemistry and loss processes are explicitly included during fitting (derivation of SOA formation yields from the observed evolution of SOA in the environmental chamber), they are useful in linking these processes to atmospheric conditions. Most previous fitting approaches do not include several of these processes explicitly. The new framework includes just a few tunable parameters, which are determined by fitting the entire time series of SOA formation and evolution in an environmental chamber and could be used in 3D chemical transport models to represent complex physical and chemical processes governing SOA evolution. 

Point of Contact
ManishKumar Shrivastava
Pacific Northwest National Laboratory (PNNL)
Funding Program Area(s)