Sangam: A Confluence of Knowledge Streams

Probabilistic Modeling of Variability and Uncertainty in Urban Air Toxics Emissions

Show simple item record

dc.contributor H.Christopher Frey, Committee Chair
dc.creator Zhao, Yuchao
dc.date 2010-04-02T19:01:30Z
dc.date 2010-04-02T19:01:30Z
dc.date 2003-12-21
dc.date.accessioned 2023-02-28T17:07:48Z
dc.date.available 2023-02-28T17:07:48Z
dc.identifier etd-09202003-151301
dc.identifier http://www.lib.ncsu.edu/resolver/1840.16/4812
dc.identifier.uri http://localhost:8080/xmlui/handle/CUHPOERS/265584
dc.description Air toxic emission factor data often contain one or more censored points below a single or multiple detection limits. Such data sets are referred to as "censored." Conventional methods used to deal with censored data sets include removing non-detects, or replacing the censored points with zero, half of the detection limit or the detection limit. However, the estimated means of the censored data set by conventional methods are usually biased. Here, an approach to quantification of the variability and uncertainty of censored data sets is demonstrated. Empirical bootstrap simulation is used to simulate censored bootstrap samples from the original data. Maximum Likelihood Estimation (MLE) is used to fit parametric probability distributions to each bootstrap sample, thereby specifying alternative estimates of the unknown population distribution of the censored data sets. Sampling distributions for uncertainty in statistics such as the mean, median and percentile are calculated. The robustness of the method was tested by application to different degrees of censoring, sample sizes, coefficients of variation and numbers of detection limits. Lognormal, gamma and Weibull distributions were evaluated. The reliability of using this method to estimate the mean is proved. The application of MLE/Bootstrap was compared favorably to results obtained with the non-parametric Kaplan-Meier method, which verify the accuracy of this method. The MLE/bootstrap method is applied to 16 cases of censored air toxic emission factors, including benzene, formaldehyde, Benzo(a)pyrene, mercury, arsenic, cadmium, total chromium, chromium VI and lead with single or multiple detection limits from coal, fuel oil and/or wood waste external combustion sources. The data differs regarding sample size, censoring degree, inter-unit variability and so on. The proportion of censored values in the emission factor data ranges from 4 to 80 percent. The largest range of uncertainty in the mean was obtained for the external coal combustion benzene emission factor, with a 95 percent probability range of minus 93 to plus 411 percent of the mean. Probabilistic emission inventories were developed for benzene, formaldehyde, chromium, and arsenic for Houston 1996 emission inventory and for 1, 3-butadiene, mercury, arsenic, benzene, formaldehyde and lead. Parametric distributions for inter-unit variability were fit using maximum likelihood estimation (MLE) and uncertainty in mean emission factors was estimated using parametric bootstrap simulation. For data sets containing one or more non-detected values, empirical bootstrap simulation was used to randomly sample detection limits for non-detected values and observations for sample values, and parametric distribution for variability were fit using MLE estimators for censored data. Goodness-of-fit for censored data was evaluated using the Kolmogorov-Smirnov test applied to a modified data set and by comparison of cumulative distributions of bootstrap confidence intervals and empirical data. The emission inventory 95 percent uncertainty ranges are as small as minus 25 to plus 42 percent for chromium for Houston to minus 75 to plus 224 percent for arsenic for Jacksonville. Uncertainty was dominated by only a few source categories. Recommendations are made for future improvements to the analysis.
dc.rights I hereby certify that, if appropriate, I have obtained and attached hereto a written permission statement from the owner(s) of each third party copyrighted matter to be included in my thesis, dissertation, or project report, allowing distribution as specified below. I certify that the version I submitted is the same as that approved by my advisory committee. I hereby grant to NC State University or its agents the non-exclusive license to archive and make accessible, under the conditions specified below, my thesis, dissertation, or project report in whole or in part in all forms of media, now or hereafter known. I retain all other ownership rights to the copyright of the thesis, dissertation or project report. I also retain the right to use in future works (such as articles or books) all or part of this thesis, dissertation, or project report.
dc.subject bootstrap simulation
dc.subject maximum likelihood estimation
dc.subject uncertainty
dc.subject censored data
dc.subject variabiliby
dc.subject urban air toxic
dc.title Probabilistic Modeling of Variability and Uncertainty in Urban Air Toxics Emissions


Files in this item

Files Size Format View
etd.pdf 1.231Mb application/pdf View/Open

This item appears in the following Collection(s)

Show simple item record

Search DSpace


Advanced Search

Browse