A methodology for estimating time-of-day variations in the size of a population exposed to risk.

Consequence models for the risk assessment of man-made or natural disasters do not ordinarily take into account time-of-day variations in the size of the exposed population. Residential census population statistics are used instead. This paper proposes and illustrates a methodology for using metropolitan travel survey data to estimate the variations in question. Variations are computed from the Washington, D.C. area sample survey statistics on the number of trips taken in and out of different census tracts throughout each workday. Four principal patterns of population variation are identified, corresponding to four types of land use: commercial, residential, shopping/entertainment, and mixed use. Some general implications for consequence analysis are discussed.