A computer model used by the U.S. Environmental Protection Agency to estimate estimate pollutants from motor vehicles is underestimating diesel truck emissions, according to a new report by the National Research Council of the National Academies.

The study, funded by EPA and the U.S. Department of Transportation, takes a broad look at the model, known as Mobile Source Emissions Factor Model or MOBILE, and finds that considerable updating is needed. The review committee, made up of scientists from around the country, also urged EPA to begin using additional data sources and tools as soon as possible to more accurately predict vehicle emissions and the pollutants they produce.
"EPA uses these estimates to develop regulations and programs for protecting air quality," said Armistead Russel, chairman of the committee and professor of environmental engineering, Georgia Institute of Technology. "These estimates need to be as accurate as possible so that pollution control programs, which cost many millions of dollars, are effective in protecting the environment and public health."
MOBILE was originally developed in 1978 and has since become a central tool of environmental and transportation agencies to assess national, state and local programs aimed at controlling vehicle emissions. There have been some technical improvements to MOBILE over the years, but studies indicate that the model is underestimating vehicle emissions of volatile organic compounds which form ozone that is damaging to public health and the environment. Estimates of nitrogen oxides and fine particles in vehicle emissions are also inaccurate, the report says.
The committee said that emissions from diesel trucks may be underestimated in the current model. It urged EPA to improve estimates quickly because states are developing pollution-control plans based on MOBILE predictions. It recommended that the model be refined to include a broad range of engine technologies and draw on data that indicate when and where trucks are most likely to be on the road.
It also noted that only a selected number of vehicles (cars and trucks) are used in the model and those most likely to have higher emissions, such as older vehicles with malfunctioning exhaust systems, are not adequately represented in the model. Other factors that affect emissions, such as how the vehicle is driven and maintained, are not fully accounted for in the model.
Some other recommendations:
* Incorporate new information about emissions-control measures such as the effectiveness of vehicle inspections and maintenance programs and of oxygenated fuels.
* Improve estimates of fine particles in exhaust.
* Determine the number of high emitting vehicles ont the road, their emission rates, and travel patterns.
* Review predictions on the effectiveness of emissions-control inspection programs and equipment. The committee noted that the current model assumes that these programs help reduce emissions, but existing data do not support the level of reduction estimated. The model also predicts that new technologies, such as warning lights to signal equipment failures, will reduce vehicle emissions but very little information is available to assess whether motorists actually repair their vehicles in response to the prompts.
The National Research Council is the principal operating agency of the National Academy of Sciences and the National Academy of Engineering. Copies of the report, Modeling Mobile-Source Emissions, will be available in June from the National Academy Press. For ordering information call (202) 334-3313 or (800) 624-6242.