Determination of the spatial and temporal distribution of population for air toxics exposure assessments

Assessment of the effects of air toxics releases in urban areas requires accurate definition of population exposure. There are two critical components in such population exposure assessments: 1) the quantification of the plume dispersal and 2) the quantification of the spatial and temporal distribution of population underlying that plume. Urban buildings in downtown areas complicate both of these issues. Plume dispersal around urban buildings is highly complex and the occupancy status of those buildings changes with people migrating from indoors to outdoors and from residence to workplace. Population datasets used for urban exposure assessments need to reflect urban diurnal population migrations. Currently, the vast majority of available population datasets are based on residential populations with no accounting for daily migrations from residence to workplace or the indoor/outdoor status of those people. The use of these residential population databases in exposure assessments can produce significant inaccuracies in morbidity and mortality estimates especially when considering daytime releases in downtown areas. In this research, we demonstrate a method for estimating urban daytime and nighttime population using US Census, infrastructure, and business demographic data in a GIS. We also develop techniques for estimating the fraction of those populations that are within buildings versus outdoors during the day and night using the US Environmental Protection Agency Consolidated Human Activity Database (CHAD). We then demonstrate the value of these datasets by estimating the population exposed during a hypothetical chemical spill in Houston, TX.