surements are difficult to make. Emission models are used for estimating on-road emissions and air emissions from landfills, storage tanks, wastewater-collection and -treatment systems, wind erosion, fugitive dust from roads, material handling, agricultural tilling, and construction and demolition. These models generally estimate Ei(s), the emission rate (in, say, tons per day) of pollutant i from source s, as


where EFi(s) is the emission factor (in units of tons of emissions per unit of activity) and A(s) is the activity level for that specific emission. Depending on the source, various types of activity levels can be chosen—for example, the total amount of fuel used by the source, the amount of product produced or consumed, the population density, or the vehicle miles traveled. In some cases, an emission factor is derived for an uncontrolled source. In this case, the emission rate for a source with controls is corrected to allow for the fractional control of the emissions. The control efficiency (CE) may be further modified by correction factors to take into account the rule effectiveness (RE) and the rule penetration (RP) (EPA 2001b). In this case, Equation 3-1 is modified as follows:


where CE is the fraction of a source category’s emissions that are controlled (for example, controlled by a control device or process change). RE is an adjustment to the CE to account for failures or uncertainties in the performance of the control. RP is the fraction of the source category that is covered by the regulation or is expected to comply.

EPA compiles and periodically updates emission factors for a large number of sources (EPA 2002g). Emission factors are simply averages of all available data of acceptable quality and are generally assumed to be representative of long-term averages for all facilities in the source category (that is, a population average) (EPA 1995). For example, for mobile sources, emission factors are derived from measurements made on a selected set of vehicle types and ages deemed to be representative of the fleet of vehicles in use. In other cases, emission factors may be based on the properties of the fuel used. For example, sulfur emissions from vehicles are often estimated from fuel consumption and fuel sulfur concentration, which is determined by measurements made on a sampling of the fuel in use at the time.

An activity level can be estimated from a wide range of data and measurements. For area sources involving consumer products, it can simply be the population density or other relevant socioeconomic indicator (for example, the amount of a given item sold or consumed). Activity levels for

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement