EPA to support the Maximum Achievable Control Technology (MACT) standards as proposed in April, 1996 for hazardous-waste incinerators and cement kilns using hazardous waste as fuel. An approximate estimate of the impact of the proposed MACT on ambient air quality is also presented for illustrative purposes only. The emission rates were all summarized and analyzed for the shape of their statistical distributions. In all cases, there was a reasonable fit to a lognormal model. These distributions were multiplied together along with the probability distribution for unit air concentrations from Cullen (1995) using a Monte Carlo routine. The atmospheric concentrations were calculated by using 5,000 trials for each chemical. Results (rounded to two significant figures) are summarized in Table 4-7, Table 4-8, Table 4-9 through Table 4-10.
The primary uncertainty associated with this analysis is in Cullen 's unit air concentration distribution. This was apparently developed for a municipal solid-waste incinerator in Bridgeport, CT, for a radial area located between 0.4 km and 30 km from the stack—a distance that included the point of maximum ambient concentration. Thus, these results will reflect ambient air concentrations expected from the incinerators and cement kilns considered in this exercise only to the extent that the emission characteristics and the local meteorology of the Bridgeport, CT, incinerator are similar to others. It must be emphasized that a person could be exposed to higher or lower concentrations depending on where he or she was located with respect to any particular combustor, and depending on characteristics of that particular combustor.
In these calculations, the particulate-matter concentrations reflect total rather than respirable particles. It is not possible to generalize, with any certainty, regarding respirable particulate-matter concentration, because there is a lack of data concerning particle-size distributions in emissions. EPA (1992a) states that approximately 62% of the mass of the particles escaping a fabric filter for a municipal solid-waste incinerator are less than or equal to 12 µm in diameter and approximately 54% of the particles are less than 2.3 µm in diameter. These factors could be applied to the concentrations given for total particulate to yield an estimate of respirable particulate.
This analysis should be viewed as illustrative rather than comprehensive. One primary source of uncertainty is the assumption that one single probability distribution of dispersion coefficients applies to all incinerators in the database. The committee was unable to be more comprehensive because of the lack of any database that linked emissions estimates with dispersion estimates for individual facilities.
Post-MACT data were taken from information presented at the U.S. EPA Boiler and Industrial Furnace and Hazardous Waste Incinerator Technical Meeting, Kansas City, KS., March 7, 1996. The database presented at this meeting included mass flow (in pounds per year) in addition to gas flow in dry standard cubic feet per minute (dscfm) for 20 commercial hazardous-waste incineration units and 30 cement kilns using hazardous waste as fuel. It should be kept in