Disaggregate Distributions of Environmental Exposure and Risk: Synthesis and Implications for Environmental Equity. M. J. Small, E. S. Siegel, and R. L. Axtell, Engineering & Public Policy, Carnegie Mellon University, Pittsburgh, PA 15213; and The Brookings Institution, Washington, DC 20036
Full and effective characterization of risk requires an understanding and assessment of variations in exposure and risk among subgroups of the population, considering age, gender, and various socioeconomic and ethnic groups. This disaggregate characterization is important for both health assessment and policy evaluation for issues of environmental equity and justice. This paper presents a framework and methods for evaluating the causes and implications of variations in environmental exposure and risk among the subgroups of a population. The framework treats the overall population distribution as a mixture of the individual subgroups, and presents analytical and simulation methods for aggregating (or disaggregating) the overall distribution from (or into) its components. These methods consider both correlative and causative factors which can lead to variation within and between subgroups. A hierarchy of idealized conditions is considered, ranging from a simple correlative model where exposure is related to income and subgroups have different distributions of income, to more complex spatial systems where individual locations relative to centers of production and pollution are dependent on income, and income distributions again vary between subgroups. These relationships are further examined using the agent-based, artificial life ("Sugarscape") simulation model of Epstein and Axtell, which includes economic activity, pollution generation associated with production and consumption, and socioeconomic evolution and division of groups of agents over time and space. The analysis will attempt to identify patterns of exposure and risk which arise from both individual preferences and from societal policy and choices which could lead to purposeful, preferential location of pollution sources.