Incorporating Uncertainty Analysis into Integrated Air Quality PlanningEPA Grant Number: R833665
Title: Incorporating Uncertainty Analysis into Integrated Air Quality Planning
Investigators: Cohan, Daniel , Bell, Michelle L. , Bergin, Michelle S. , Boylan, James , Cox, Dennis , Marmur, Amit
Institution: Rice University , Environmental Protection Division of the Georgia Department of Natural Resources , Yale University
EPA Project Officer: Pascual, Pasky
Project Period: October 28, 2007 through October 27, 2009
Project Amount: $229,770
RFA: Uncertainty Analyses of Models in Integrated Environmental Assessments (2006) RFA Text | Recipients Lists
Research Category: Ecological Indicators/Assessment/Restoration , Economics and Decision Sciences , Ecosystems
As states finalize implementation plans to achieve current ozone and particulate matter (PM) standards, more stringent standards now under consideration may prompt a new wave of attainment planning. Although states have begun to explore the linkage of cost, photochemical, and health effect models to inform strategy selection, model uncertainties have typically remained unexamined. The proposed research will explore how formal examination and communication of model uncertainty might enhance future air quality planning, using Georgia as a case study to develop broadly applicable methodologies. We hypothesize that (1) high-order direct sensitivity analysis of a photochemical model can efficiently quantify uncertainty in atmospheric responsiveness resulting from uncertain model inputs, (2) alternate exposure metrics for epidemiological concentration-response functions may dramatically impact prioritization of control options, and (3) joint consideration of uncertainty across linked models, if properly communicated, may enhance the attainment planning process.
A diverse team of scientists will examine how integrated attainment modeling can be reconsidered from a stochastic perspective. Integrated control cost, photochemical sensitivity, and health effect assessments already conducted by the Georgia Environmental Protection Division (GA EPD) for attainment planning will provide the basis for our work. Whereas GA EPD deterministically estimated the costs, ambient air impacts, and health benefits associated with various control options, we will investigate how those estimates depend upon uncertain assumptions within the linked components. We will apply a recently introduced high-order sensitivity analysis technique to probe how modeled pollutant responsiveness to emission controls depends upon uncertain assumptions regarding chemical reaction rates, emissions inventories, and boundary conditions. Epidemiological literature will be analyzed to examine uncertainty both in the magnitude and exposure metric of concentration-response functions for ozone and PM health impacts. Stochastic modeling will then be conducted to characterize how uncertainty across the linked components impacts the prioritization of control measures for health and attainment objectives. Workshops at GA EPD will provide a forum for project scientists and state officials to jointly explore how to improve the communication of uncertainty in the attainment planning process.
The proposed research will develop methods by which air quality planners could formally consider the uncertainty of models that inform control strategy development. We will identify key photochemical model inputs, epidemiological parameters, and other assumptions that most induce uncertainty in strategy assessment. Close interaction with state officials will foster development of explanatory techniques for communicating uncertainty.