HR: 0800h
AN: A51D-0094    [Abstracts]
TI: Adjoint Sensitivity Analysis of Ozone Over Continental United States
AU: * Hakami, A
EM: amir@cheme.caltech.edu
AF: Departments of Chemical Engineering and Environmental Science and Engineering, California Institute of Technology, MC 210-41 1200 E California Blvd, Pasadena, CA 91125 United States
AU: Seinfeld, J H
EM: seinfeld@caltech.edu
AF: Departments of Chemical Engineering and Environmental Science and Engineering, California Institute of Technology, MC 210-41 1200 E California Blvd, Pasadena, CA 91125 United States
AU: Tang, Y
EM: ytang@cgrer.uiowa.edu
AF: Center for Global and Regional Environmental Research, University of Iowa, Iowa City, IA 52242 United States
AU: Chai, T
EM: tchai@cgrer.uiowa.edu
AF: Center for Global and Regional Environmental Research, University of Iowa, Iowa City, IA 52242 United States
AU: Carmichael, G R
EM: gcarmich@engineering.uiowa.edu
AF: Center for Global and Regional Environmental Research, University of Iowa, Iowa City, IA 52242 United States
AU: Sandu, A
EM: sandu@cs.vt.edu
AF: Department of Computer Science, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061 United States
AB: An application of the adjoint method in air quality management is demonstrated. In the adjoint applications (e.g., variational data assimilation), sensitivities of a receptor-based metric to a multitude of parameters can be efficiently calculated. Here, we calculate the sensitivities of a nationwide U.S. ozone National Ambient Air Quality Standard (NAAQS) non-attainment metric to precursor emissions during the month of July 2004. We use the adjoint version of a continental scale chemical transport model (STEM) with 60 km horizontal grid resolution. The model shows low bias and error (3, and 20 percent, respectively), particularly for areas with high ozone concentrations that are subject of this study. The non-attainment metric is integrated over the month and entire domain, and accounts for both 1-hour and 8-hour ozone standards. The metric is quadratic, and therefore, higher concentrations are given larger overall weights in the calculations. The adjoint system is driven by exceedances above the standards, and links the non-attainment metric with the emissions of various species at each location. The non-attainment metric for the month of July is heavily dominated by the 8-hour standard (98%). Using these spatial distributions of sensitivities we identify the regions and source categories with significant contribution/effect on the overall non-attainment metric. Largest values of sensitivities are found to be with respect to emissions in the southeast and along the Ohio River Valley. When non-attainment sensitivities are integrated domainwide, NOx emissions account for the largest contribution (64% of the total), followed by the biogenic and anthropogenic VOCs (23% and 13%, respectively). For NOx emissions, point/area, mobile, and non-road mobile sources account for 51, 39, and 10 percent of the total sensitivities, respectively. Point/area sources also have the largest effect among the VOC source categories (62%). No significant difference is seen between weekday and weekend emission contribution patterns. We also provide a regional comparison for the non-attainment magnitude, non-attainment sensitivity, and emission magnitudes to explore the influence of interstate transport of ozone and its precursors.
DE: 0345 Pollution: urban and regional (0305, 0478, 4251)
DE: 0368 Troposphere: constituent transport and chemistry
DE: 3355 Regional modeling
SC: Atmospheric Sciences [A]
MN: Fall Meeting 2005