Emissions Factors Applicability

Close Window

Emissions factors published in this database and in most other such compilations typically 1) are arithmetic averages of available source test data, 2) are based on limited numbers of emissions tests, 3) represent only a few hours of process operating time per test, 4) represent limited ranges of process operating conditions, and 5) represent a limited sample of operating units within any source category. As a result, site-specific emissions estimates based on emissions factors will include significant data uncertainty. Such uncertainties can easily range over more than one order of magnitude in determining emissions from any one specific facility. Use of emissions factors should be restricted to broad area-wide and multiple source emissions cataloging applications that will tend to mitigate the uncertainty associated with quantifying site-specific emissions. For example, emissions factors are generally appropriate for use in compiling emissions estimates from multiple sources for area-wide inventories when measured emissions data (e.g., CEMS) for sources included in the inventory are scarce. Such inventories serve several purposes including supporting ambient dispersion modeling and analyses, developing control strategies, and screening to identify sources that are potentially major contributors to area environmental impacts for possible compliance investigations. Even in this context, significant uncertainties remain when you apply low quality rated emissions factors and when a few large sources dominate an emissions inventory. Emissions factors uncertainty may also cause air quality management programs to overlook a segment of the source population that may be responsible for significant emissions contributions and should be addressed.

Close Window

Because of the uncertainties inherent in the use of average emissions factors for facility-specific emissions determinations, emissions from potentially large numbers of permitted sources are characterized incorrectly in permitting and compliance applications. Further, emissions factors at best are imprecise tools for establishing emissions limits (e.g., permit limits based on best available control technology or BACT, lowest achievable emission rate or LAER, source category limitations to reduces emissions in a geographic regions or SIPís) or standards (e. g., National Emission Standard for Hazardous Air Pollutants or NESHAP, New Source Performance Standards or NSPS). The emissions reductions determined during regulatory standard setting done without regard to the uncertainty in emissions factors will be open to question. For these reasons, we recommend against use of source category emissions factors (whether derived from AP-42, FIRE, or elsewhere) for site-specific emissions determinations or regulatory development. We recommend instead the use of alternatives to emissions factors (see below).

Close Window

We recognize that emissions factors are often used in many applications including site-specific applicability determinations, establishing operating permit fees, and establishing applicable emissions limits even though such use is inappropriate. If you must apply emissions factors for site-specific applications, we strongly recommend due consideration of the uncertainty inherent in the data. Applying emissions factors without accounting for uncertainty will result in doubtful applicability determinations, ineffective emissions reductions requirements, and poorly supported compliance determinations or enforcement actions.

Close Window

Approaches to accounting for uncertainty include adjustments based on statistical assessments addressing bias and imprecision for both pollutant emissions control and process operations or activities variability. Under such options, we believe it appropriate to consider the quality and quantity of the source test data underlying the emissions factors and to consider the variations of emissions control and process operations between sources within the same category. With this information, we think it prudent to apply standard statistical adjustments in the use of emissions factors consistent with the goals of your specific application (e.g., upper confidence level in determining site-specific thresholds for applicability and fees, lower confidence level in setting emissions limits). We are developing detailed procedures and more explicit policies for site-specific and regulatory development applications of emissions factors along with recommended alternatives to emissions factors and will provide those procedures in the near future.

Close Window

Alternatives to Emissions Factors

Data from frequent and representative source-specific emissions tests or continuous emissions monitoring systems can provide measures of actual pollutant emissions from a source that are much more reliable than emissions factors. Note that site-specific measurement data from a limited number of emissions tests will improve the certainty of the emissions data but will also represent only the conditions existing at the time of the testing or monitoring. To improve the estimate of longer-term (e.g., daily, monthly, yearly) emissions, conditions under which tests occur should be numerous and representative of the sourceís expected range of operations. Data from continuous emissions monitoring systems provide the most complete assessment of a sourceís emissions in many cases. If you are unable to collect representative source-specific data, emissions information from process and control equipment vendors, particularly emissions performance guarantees or emissions test data from similar equipment, is a better source of information for most permitting decisions than source-category emissions factors.

Close Window

In other cases, data from material balances approach also may prove highly reliable. For some sources, a material balance may provide a better quantification of emissions than other measurements as material or mass balance determinations can account for fugitive emissions not easily measured otherwise. In general, material balances are appropriate for use in situations where a high percentage of material is lost to the atmosphere (e. g., sulfur in fuel, or solvent loss in an uncontrolled coating process.) In contrast, material balances may be inappropriate where material is consumed or chemically combined in the process, or where losses to the atmosphere are a small portion of the total process throughput. As the term implies, one needs to account for all the materials entering and exiting the process for such a pollutant material balance to be credible.

Close Window