Notes from Discussion with Michigan Department of Environmental Quality (MDEQ)
        Regarding Modeling of Sulfur Dioxide (SO2) in the Detroit Area
                               December 10, 2013
                                       
On December 10, 2013, modeling staff from EPA Region 5's Air and Radiation Division and MDEQ's Air Quality Division held a conference call to discuss the status of Michigan's planning for the Detroit SO2 nonattainment area.  Participants from MDEQ included Jim Haywood, Stephanie Hengesbach, and Bob Irvine.   Participants from EPA Region 5 included Randy Robinson, Mary Portanova, John Summerhays, and Sarah Arra.
MDEQ described a set of modeling runs it has conducted and shared with affected parties in Southeast Michigan.  (EPA Region 5 ultimately obtained a summary that had been shared with the affected parties, a summary that is publicly available on a web site of the Southeast Michigan Council of Governments, at http://www.semcog.org/WorkArea/DownloadAsset.aspx?id=91300.)  MDEQ and EPA Region 5 discussed modeling issues such as use of rural versus urban dispersion characteristics, terrain, and receptor grids.  MDEQ noted very high modeled concentrations near the U.S. Steel facility and described its efforts to evaluate scenarios for control of emissions from U.S. Steel, presumably including desulfurizing the coke oven gas it burns and potentially including at least some conversion to using natural gas.
EPA Region 5 asked various questions about the role of DTE's Monroe power plant in the analysis.  MDEQ noted that its initial modeling runs expressly included this plant as a modeled source, but subsequent modeling runs did not.  MDEQ explained that the subsequent runs were intended to be simplified runs that focused on the sources most influencing the violations and thus most warranting consideration of control, and so these runs excluded the Monroe power plant because impact of this plant at violating locations was relatively minor.  EPA Region 5 encouraged MDEQ to include this plant explicitly in its final, submitted modeling run (rather than address the plant as part of the monitored background concentration), and MDEQ said they planned to do so.

