Dispersion modelers have long faced challenges estimating ambient pollutant concentrations caused by releases from “fugitive” sources of particulate matter, such as paved and unpaved roadways, raw material storage piles, outdoor material processing operations, agricultural activities, or windblown dust in general. Fugitive emissions are commonly defined as those that could not reasonably pass through a stack, chimney, vent, or other functionally equivalent opening. Aside from their non-point release characteristics, the unsteady state nature of most fugitive emitting activities is what makes them particularly problematic when simulated by steady-state dispersion models. Further, there has been limited field testing completed to provide performance evaluations that would support the models for these types of releases.
The primary regulatory guidance from the Environmental Protection Agency for modeling fugitive emissions is given in the Guideline on Air Quality Models (40 CFR 51, Appendix W).1 Section 188.8.131.52 of the Guideline, specific to PM10 modeling, refers the user to Section 4.2.2 “for source-specific analyses of complicated sources”, but that section says little concerning fugitive sources. In the AERMOD user’s manual2, methodologies are offered for modeling fugitive sources. Many state air regulatory agencies have also prescribed specific protocols for modeling fugitive PM sources. However, application of many of the general and/or prescribed techniques can yield unrealistically high air concentrations relative to the nature and magnitude of emissions, particularly when receptors are located close to fugitive sources. This paper explores common presumptions about fugitive source modeling techniques by examining the sensitivity of predicted PM ambient concentrations to the choice of model (AERMOD versus ISCST3), changes in source representation (volume versus area source), and variations in chosen source dimensions. The affect of key meteorological data parameters, such as wind speed and land use, are also reviewed.