Using Continuous Emission Monitoring Data for Air Quality Modeling Inventories

The Continuous Emission Monitoring (CEM) data have attracted interest for potentially providing improved point-source emissions data for air quality models. Through the CEM program, NOx and SO2 data have been collected on an hourly basis since 1995, but several features of the data must be addressed before they can be useful for air quality modeling. These features include the large volume of data, the lack of stack parameters or coordinates, and the lack of common identifiers between the CEM data and emissions inventories used for modeling. This paper describes how some of these problems can be addressed and how the CEM data can be used to benefit emissions for air quality modeling. Several analyses are presented that compare the 1995 CEM data to the 1995 Ozone Transport Assessment Group (OTAG) inventory. Both the emissions magnitude and the temporal variation are compared. The CEM data are also used to evaluate seasonal versus monthly temporal profiles. These analyses are considered with respect to their possible impact on air quality modeling.