"I durst not laugh for fear of
opening my lips and receiving the bad air."
-- Julius Caesar, I, ii, 249-50

Cleaner Air in LA

Even before the Pittsburgh Supercomputing Center officially opened in mid-1986, Ted Russell and Greg McRae were using the new CRAY X-MP to model smog in Los Angeles. Extensive computations by Russell, an associate professor of mechanical engineering at Carnegie Mellon University, and his chemical engineering colleague McRae, now at MIT, showed that controlling hydrocarbon emissions alone wasn't enough to control smog. Nitrogen oxides, which seep into the air from oil refineries, paint factories, power plants and automobiles, also play a big role. This finding, which ran contrary to EPA policy at the time, provided the scientific underpinning for the Air Quality Management Plan adopted in 1988-89 for the Los Angeles Air Basin -- the most stringent such plan in the United States.

The next step for Russell was modeling the impact of alternative fuels. The California Air Resources Board (CARB) commissioned him to run an experiment: What would happen if all Los Angeles motor vehicles switched from gasoline to a "clean fuel" -- in particular, methanol? Hundreds of supercomputing hours later, Russell found that the change would significantly cut particulate emissions and formaldehyde, a known carcinogen, as well as ozone, the primary constituent of smog. "The alternative fuels," says Russell, "are smaller hydrocarbons, so they burn cleaner without forming aerosols." This research helped swing policy toward including fuel conversion as a recommended strategy in the 1990 revisions to the Federal Clean Air Act.

Russell and McRae's air-quality modeling began in the late 1970s when they worked together as CalTech graduate students, and it evolved over more than 15 years to become the most detailed, comprehensive model of smog formation available, tracking reactions over time for more than 50 chemical species. Their work has played a major role in an emerging success story: Peak ozone concentrations in Los Angeles have declined 20 percent over the last few years, and exposure to unhealthful concentrations has declined 50 percent.

More recently, Russell and McRae have applied their modeling expertise to California's alternative-fuel regulations for automobiles, and Russell has modeled air quality in Mexico City, with results that have influenced pollution control policy there.


Ozone Concentration in the Los Angeles Air Basin
These images show the results of air-quality modeling for the Los Angeles air basin with initializing data from Aug. 27, 1987. The clouded volume represents ozone concentrations exceeding federal ambient air quality standards (0.12 parts per million). At 8 a.m. (above), ozone levels are below the standard except for the easternmost region of the air basin. By 2 p.m. (below), hydrocarbons from vehicle emissions have reacted with sunlight to significantly increase ozone throughout the basin. The ozone cloud is overlaid on census maps and a LANDSAT satellite image using the Geographic Environmental Modeling System (GEMS) project being developed at Carnegie Mellon by Bernd Bruegge and Erik Riedel.

Researchers: Armistead Russell, Carnegie Mellon University, and Gregory J. McRae, Massachusetts Institute of Technology
Hardware: CRAY Y-MP C90, DEC Alpha Cluster
Software: User Developed Code
Keywords: air quality modeling, air pollution, clean fuel, smog, Los Angeles, Mexico City, environmental quality, environmental protection, environmental engineering, particulate emissions, airshed model evaluation, automobile emissions, air quality management.
Related Material on the Web:
A Distributed Computing Approach to Large-Scale Environmental Modeling
Projects in Scientific Computing, PSC's annual research report
References, Acknowledgements & Credits