Atmospher Sci & Global Chg
Research Highlights

November 2009

PNNL Testbed Gives Science Community a Better Tool for Climate Modeling

Objective, quantifiable "reality check" for evaluating new atmospheric processes in models

The Aerosol Modeling Testbed consists of the chemistry version of the Weather Research and Forecasting community model, testbed cases, and the analysis toolkit. Data from field campaigns (cases) are used to evaluate candidate aerosol process modules over a wide range of meteorological conditions. Enlarge Image

Results: A national research team led by scientists at Pacific Northwest National Laboratory has created an Aerosol Modeling Testbed—a framework where the worldwide science community can test, evaluate, and compare new treatments for aerosol processes in models. This is the first testbed of its kind designed to quantify the performance of aerosol process modules over spatial scales consistent with measurements collected during field campaigns.

Why it matters: Aerosol processes—how they behave and interact with clouds in the atmosphere—are among the most important—yet most difficult to simulate, aspects of climate modeling. Differences in how a model represents space and time scales, emission rates, meteorology, gas-phase chemistry, and other aerosol processes all affect model predictions.  Current methods for improving aerosol processes in models are haphazard and slow. Typically, scientists incorporate new aerosol process modules into models and then evaluate them using limited laboratory or field data.  Each evaluation study often uses a different model and/or data set, making it impossible to directly compare the performance and computational efficiency of various approaches that simulate the same aerosol process.  In addition, model intercomparison studies do not quantify the range of uncertainty associated with a specific aerosol process, nor does this type of uncertainty analysis provide much information on which aerosol process needs improving the most. 

Now, with the Aerosol Modeling Testbed, scientists can systematically and objectively evaluate new aerosol process modules over a wide range of spatial and temporal scales. The new tool complements existing modeling activities and reduces uncertainties in climate models, ultimately making climate projections more accurate.

Methods: The Aerosol Modeling Testbed consists of the chemistry version of the Weather Research and Forecasting model (WRF-Chem) that simulates meteorology, chemistry, and aerosol interactions, a series of testbed cases that use U.S. Department of Energy field campaign measurements, and a suite of tools. The tool suite documents the performance of specific aerosol process modules.

The testbed facilitates science by minimizing redundant tedious tasks, quantifying how multiple treatments for a specific aerosol process perform for real-world conditions, and assessing the performance of aerosol treatments in relation to their computational expense.  Other important benefits of a testbed approach include the ability to document improvements to treatments of aerosol processes over time and foster collaboration among aerosol modelers, experimentalists, and field researchers.

What's next: PNNL is collaborating with several investigators that are using a suite of evaluation tools for their research.  Their feedback will be used to enhance the Aerosol Modeling Testbed to address the needs of the scientific user community.  Data from a number of field campaigns are being added as test cases, enabling evaluation of a wide variety of cloud-aerosol interactions, secondary organic aerosols, and other aerosol conditions. Eventually, the website will enable users to download parts of the code, after registering and describing the aerosol process they are interested in testing.  A paper describing the Aerosol Modeling Testbed and a comparison of two aerosol models will be submitted to the Bulletin of the American Meteorological Society in 2009.

Acknowledgments: PNNL is transforming the nation's ability to predict climate change and its impacts. The development of the beta version of other Aerosol Modeling Testbed was sponsored by PNNL's Laboratory Directed Research and Development program through the Aerosol Climate Initiative.

Research Team: The research team includes Jerome Fast, William Gustafson Jr., Elaine Chapman, Richard Easter, and Jeremy Rishel from PNNL; Georg Grell, National Oceanic and Atmospheric Administration Earth System Research Laboratory; and Mary Barth, National Center for Atmospheric Research.

Reference: Fast JD, WI Gustafson, EG Chapman, GA Grell, SE Peckham, and R Schmitz. 2009. "Applying the Aerosol Modeling Testbed to Assess Modal and Sectional Predictions of Particulates and Aerosol Radiative Forcing Downwind of a Megacity." Presented at the 11th Conference on Atmospheric Chemistry, 89th American Meteorological Society Annual Meeting, January 10-16, 2009, Phoenix, AZ.  Contact PNNL's Dr. Jerome Fast to obtain the analysis toolkit software and related components.