Comparison of Dispersion and Ambient Monitoring Techniques for Demonstrating Compliance with the 1-Hour NO2 NAAQS

In 2010, EPA revised the form of the National Ambient Air Quality Standard (NAAQS) for NO2 from an annual-average to a 1-hour averaging period.  Many facilities that historically modeled compliance with the old annual-average standard have not been able to demonstrate compliance with the new 100ppb NAAQS using traditional dispersion modeling techniques.  Even the use of more advanced techniques such as the Ozone Limiting Method (OLM) are often not sufficient to demonstrate compliance with the NAAQS.  These modeling results are often at odds with ambient monitoring data indicating no exceedances of the NAAQS.  

Model overpredictions of NO2 are primarily related to the models' characterization of conversion of NO to NO2 in the atmosphere.  Data to accurately describe NO to NO2 conversion, such as ozone concentrations at the point of release, are often not available, requiring the use of assumptions that affect model accuracy.  Current dispersion modeling guidance from EPA requires the use of additional conservative assumptions that cause further overprediction, such as specifying minimum "in-stack ratios" of NO2 to NOx.  

We present our experience in modeling NO2 emissions from BASF's Freeport, TX site, including a progression of more refined modeling techniques.  We also present the results of NO2 monitoring over a multi-year period, including a comparison of model predicted concentrations with ambient monitoring data.