Impacts of Peaking Unit Emissions on Urban Air Quality: A Modeling Study for Baltimore

Electricity Generating Units (EGUs) are important sources of anthropogenic NOx emissions, contributing ~30% of total national NOx emissions, an important precursor for tropospheric ozone. Large EGUs running 24/7, such as coal burning power plants, are usually required to operate with advanced flue gas treatment systems to reduce emissions and are closely monitored by the U.S. EPA Continuous Emissions Monitoring Systems (CEMS) program to ensure compliance with air quality standards and permit requirements. To meet electricity demand during high peak load, EGUs which can be activated quickly, peaking units, are operated for short periods, usually in hot summer days with elevated air pollution. Due to their short operation time, control technology is ineffective at startup and shutdown, or these peaking units often lack flue gas treatment technology and are not monitored by CEMS, leading to significantly higher emission rates and no real-time data for regulation. Urban air quality can be substantially impacted by these peaking unit emissions because of their locations within or close to urban centers.

To quantify the influence, we conducted a case study using WRF-CMAQ for an air quality episode (June 25 - 29, 2019) in Baltimore. The nested WRF-CMAQ modeling system has 12/4/1.33/0.44 km spatial resolution for the CONUS/Eastern U.S./Mid-Atlantic/Baltimore regions, respectively, with high-resolution inner most domain to resolve the urban area. We developed spatial surrogates and used the EPA Sparse Matrix Operator Kernel Emissions (SMOKE) model to generate high-resolution anthropogenic emissions. Measured emissions from the Westport CT5 peaking unit in downtown Baltimore, were collected for this air quality episode. This single peaking unit had 5~7 times higher emissions rate than continuously run baseload running power plants in Maryland and released 12% of the total Marylan daily NOx emissions within 3 hours, which were not included in the emissions inventory. A sensitivity experiment with the addition of the Westport emissions shows >0.5 ppbv hourly ozone enhancements over downwind EPA air quality monitors. The high-resolution modeling results also help to estimate pollutant levels in disadvantaged communities in Baltimore without air quality monitors, providing scientific guidance for future environmental justice studies and action.