Messenger - Vol. 2, No. 3, Page 9
Summer 1993
On Research
Models predict role of gases, particles and droplets in air pollution

     Anthony S. Wexler, assistant professor of mechanical engineering, is
undertaking a five-year program that should improve the way air quality is
     Under a $1.24 million grant from the IBM Corp., Wexler and John H.
Seinfeld of the California Institute of Technology will use advanced
computer technology to develop the next major step in air quality modeling,
thus enhancing the tools available to those trying to control air
     The project is divided into two primary parts: model development and
testing, which will be Wexler's main activity; and visualization, or
developing ways to present the huge amount of data developed by the model
in an understandable manner, which will be Seinfeld's emphasis.
     Air pollution is one of the most serious and pervasive environmental
problems in the world, Wexler says, and regulators attempting to control
the problem use mathematical models to relate the emission of pollutants to
their effect on air quality.
     "These models enable them to investigate how different courses of
action may affect the environment and to devise strategies for achieving
optimal cost-effective reductions in air pollution levels," he explains.
     Such models are crucial to the process, Wexler says, because an
effective response to one problem may exacerbate or disproportionately
influence another in seemingly unpredictable ways. For example, reducing
sulfur dioxide emissions to limit the problem of acid rain could affect
cloud and aerosol dynamics and ultimately accelerate the greenhouse effect.
     Because of computing constraints, the mathematical models now in use
have been limited to gaseous species, such as carbon monoxide, sulfur
dioxide, nitrogen oxides and hydrocarbons, and secondary pollutants such as
ozone and nitric acid, Wexler says, but the new model will provide an
integrated treatment of gases, particles (aerosols) and droplets.
     The addition of particles and droplets to the model will require about
10 times the computing power used in current models.
     Developing the model is particularly complex because of the number of
variables that might be involved, according to the researchers. Once a
chemical is released into the air, there are many processes that can affect
     When the model is developed, Wexler and Seinfeld will work with
scientists in Mexico City and Athens-cities with serious urban air
pollution problems. Humberto Bravo of the University of Mexico and Spyros
Pandis of the University of Patras will prepare data bases on their
respective cities and then use the new model to try to solve problems
                                   -John Brennan