Modeling indoor air concentrations near emission sources in imperfectly mixed rooms.

Assessments of exposure to indoor air pollutants usually employ spatially well-mixed models which assume homogeneous concentrations throughout a building or room. However, practical experience and experimental data indicate that concentrations are not uniform in rooms containing point sources of emissions; concentrations tend to be greater in close proximity to the source than they are further from it. This phenomenon could account for the observation that "personal air" monitors frequently yield higher concentrations than nearby microenvironmental monitors (i.e., the so-called "personal cloud" effect). In this project, we systematically studied the concentrations of a tracer gas at various distances from its emission source in a controlled-environment, room-size chamber under a variety of ventilation conditions. Measured concentrations in the proximity of the source deviated significantly above the predictions of a conventional well-mixed single-compartment mass balance model. The deviation was found to be a function of distance from the source and total room air flow rate. At typical air flow rates, the average concentration at arm's length (approximately 0.4 meters) from the source exceeds the theoretical well-mixed concentration by a ratio of about 2:1. However, this ratio is not constant; the monitored concentration appears to vary randomly from near the theoretical value to several times above it. Concentration data were fitted to a two-compartment model with the source located in a small virtual compartment within the room compartment. These two compartments were linked with a stochastic air transfer rate parameter. The resulting model provides a more realistic simulation of exposure concentrations than does the well-mixed model for assessing exposure to emissions from active sources. Parameter values are presented for using the enhanced model in a variety of typical situations.