A maximum entropy approach to estimating emissions

A maximum information entropy method of calculating probabilistic estimates of volatile organic compound (VOC) emissions by the wood furniture and fix ture coating industry is presented. The maximum entropy approach is used to produce minimally biased probability distributions for number of firms, co ating use, and coating emission factors from existing summary statistics. T hese distributions are combined to estimate VOC emissions. The maximum entr opy emissions estimate provides information to support probabilistic modeli ng of regional air quality, probabilistic assessment of emission reduction strategies, and risk assessments. Accurate estimation of emission distribut ions produces more informed regulatory decisionmaking, risk comparisons, an d regulatory and scientific priority setting.