Probabilistic Modeling of Variability and Uncertainty in Urban Air Toxics EmissionsEPA Grant Number: R826790
Title: Probabilistic Modeling of Variability and Uncertainty in Urban Air Toxics Emissions
Investigators: Frey, H. Christopher
Institution: North Carolina State University
EPA Project Officer: Chung, Serena
Project Period: October 1, 1998 through September 30, 2001 (Extended to March 31, 2004)
Project Amount: $329,425
RFA: Urban Air Toxics (1998) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air
Description:Information regarding variability in urban air toxics emissions is needed to identify high emitters or highly exposed populations. Information regarding uncertainty is needed to characterize the quality of an emissions inventory and to target data collection to reduce uncertainty. Our objectives are: (1) to develop methods for quantifying variability and uncertainty in urban air toxics emissions; (2) to develop methods for identifying key sources of variability and uncertainty in assessments of urban air toxic emissions and exposures; (3) to develop probabilistic process engineering models for making realistic estimates of emissions of, and the effects of control measures for, urban air toxics; (4) to demonstrate the methods via a detailed case study of urban air toxics emissions and exposures; and (5) to characterize the benefits of the methods with respect to environmental and research management.
Approach:We will focus on two major types of activities: (1) development of new methods for quantitative analysis of variability and uncertainty; and (2) demonstration of the methods through detailed case studies. We will develop methods for quantifying variability and uncertainty for specific inputs in an urban air toxics emission inventory. These methods will address small data sets, averaging times, selection of empirical versus parametric distributions, measurement and random sampling error, censored data sets, and dependencies between variable and/or uncertain quantities. A two-dimensional numerical method for propagation of both variability and uncertainty through an inventory will be a key innovation of this work. We will refine methods for identifying key sources of both variability and uncertainty in the inventory using statistical and graphical techniques. We will develop probabilistic engineering models for the purpose of predicting variability and uncertainty in emissions for significant urban air toxic emissions sources, such as combustion and evaporative sources. The new analysis techniques and source- specific emissions models will be applied to detailed development of a probabilistic city-specific emission inventory. The benefits of probabilistic analysis will be identified and illustrated using case studies of the propagation of variability and uncertainty in emissions through an exposure model. Initially, we will focus on inventories for benzene, 1,3-butadiene, formaldehyde, hexavalent chromium, and polycyclic organic matter (POM).
The benefits which we will demonstrate include: (a) development of more scientifically defensible estimates of emissions; (b) more robust identification of the largest emitters of a given pollutant; (c) more realistic estimates of the effect of control strategies on emissions; (d) development of emissions inventories based upon averaging times that are relevant to a given health effect end-point; (e) identification of key sources of variability that can be used to identify highly exposed subpopulations; (f) identification of key sources of uncertainty that can be used to prioritize additional data collection and research. Thus, using the methods developed in this research, environmental managers will have better information regarding the quality of emissions inventories as well as insights regarding where to target resources to achieve the greatest reduction in uncertainty regarding the emissions estimates.
Improvement in Risk Assessment or Management: For this work, there are two types of risks: (1) the probability that air quality management strategies will be ineffective; and (2) the probability of adverse effects due to exposures to urban air toxics. We will develop methods that reduce the risk of ineffective air quality management strategies by allowing decision makers to assess the quality of their decisions and to decide on whether and how to reduce the uncertainties that most significantly affect those decisions. The results of this work will improve human health risk management by providing exposure and risk assessors with both methods and case studies regarding quantification of variability and uncertainty in urban air toxics emissions and exposures.