頁籤選單縮合
題 名 | A Bayesian Approach for the Reduction of Uncertainty in the Industrial Source Complex-Short Term Model Version 3 (ISCST3) |
---|---|
作 者 | Ma, Hwong-wen; Tu, Wayne Wei-yuan; Crawford-Brown, Douglas; Chen, Chi-feng; | 書刊名 | Journal of Environmental Engineering and Management |
卷 期 | 17:6 2007.11[民96.11] |
頁 次 | 頁385-393 |
分類號 | 445.6 |
關鍵詞 | Bayesian approach; ISCST-3 model; Uncertainty analysis; Risk assessment; |
語 文 | 英文(English) |
英文摘要 | The Industrial Source Complex-Short Term Model-Version 3 (ISCST3) is a steady-state Gaussian dispersion model designed to support the Environmental Protection Agency's regulatory modeling programs. It is often employed to estimate air concentrations and depositions associated with air emissions from a source of air toxics, and thus to provide exposure information for risk assessment. Combining a prior distribution described by the ISCST3 model output concentration and the associated model uncertainty with a likelihood distribution consisting of measurement data and its associated measurement uncertainty, a posterior distribution describing the uncertainty in the pollutant concentration was produced. The resulting Bayesian methodology was used to determine 95% confidence intervals for pollutant concentrations under a variety of scenarios. The methodology can also determine if the inclusion of the modeling results could significantly reduce the uncertainty associated with estimates of air concentration used in exposure assessment. This study has found that given a large standard deviation of modeling results (geometric standard deviation 5 and above), the Bayesian approach does not result in reduction of overall uncertainty. The Bayesian technique could be very useful for determining where and how modeling and measurement can be combined to more accurately identify pollutant concentrations and in turn, more accurately determining human health risks. |
本系統中英文摘要資訊取自各篇刊載內容。