UNC-led team wins NOAA grant to develop better air quality forecasting
June 30, 2019A team of air quality monitoring experts will soon kickstart the development of a state-of-the-art emissions testbed to enhance current air quality forecasting systems produced by the National Oceanic and Atmospheric Association (NOAA). Leading this team is B.H. Baek, a research assistant professor at the UNC Institute for the Environment and an experienced developer of air quality modeling systems. The $750,000 grant will span three years.
The testbed will enhance the functionality and accuracy of NOAA’s current air quality forecasting system, the National Air Quality Forecast Capability (NAQFC). Referred to as the NAQFC Community Emission Testbed (NCET), this tool will expedite the interpretation of regional emissions data and help improve national air quality forecasts that better reflect nuanced changes in atmospheric conditions at the local level. This asset will make it easier for state and local agencies to protect the public from adverse health effects caused by poor air quality.
“Once the air changes, everything changes,” says Baek.
As a key developer of the Sparse Matrix Operator Kerner Emissions (SMOKE) modeling system, which is a key component of air quality modeling systems, Baek understands the difficulties of accurately representing the fluctuating emissions outputs in local areas from sources like cars, residential heating systems and trees.
“Getting the proper understanding of the amount of emissions coming from individual sources has been very challenging,” Baek explains. “I am working to use my expertise in interpreting actual, real-time emissions data to understand how to better represent them.”
The U.S. EPA releases the official National Emissions Inventory (NEI) database provided by each state for various air quality regulatory applications. Air quality forecasting models crafted by NOAA draw from the information included in this database. As noted by Baek, this system is often challenged to account for real-time air quality patterns and emissions trends that occur on a much smaller scale, such as at the street level. To address this, the NCET testbed developed by this grant will supply national forecasting models with fine-tuned emissions data at the local level for a better air quality forecasting.
Once NCET is implemented and an updated emissions dataset is compiled, Baek and his team will incorporate “meteorological coupling,” a vital component to the development of enhanced air quality forecasting. Baek anticipates that this aspect of the grant will enable his team to develop localized, detail-oriented emissions data that is documented hourly.
“This grant is going to allow us to enhance the representation of meteorology-driven emissions sources to make it more sensitive and responsive to its local meteorology,” Baek says.
The meteorological coupling component is primarily concerned with three local emission sources that contribute heavily to regional air emissions: (1) mobile units—specifically automobiles, which are significant emitters of nitrogen oxides (NOx), atmospheric particulate matter with diameters less than 2.5 micrometers (PM2.5), and carbon monoxide (CO); (2) residential heating systems, which produce PM2.5; and (3) agricultural sites, in which livestock waste harbors harmful levels of ammonia (NH3), which is a precursor of ambient PM2.5.
Baek recognizes the difficulties faced by air quality modelers in properly representing these emissions from individual units like automobiles. To address these difficulties, Baek proposes that using meteorological tools in tandem with the NCET testbed will account for subtle characteristics that can impact local air quality, such as vehicle and fuel types as well as local weather patterns. On warmer days, for example, diesel emissions will emit larger amounts of NOx than on cooler days.
“Using the meteorology data on that local scale of interest, we can actually estimate the emissions on the fly with hourly ambient temperature and humidity to compute the emissions from onroad mobile vehicles,” Baek explains.
Among Baek’s team is Daniel Tong, a research professor at George Mason University and emissions model developer for NOAA.
Baek anticipates the long-term benefits of his grant will be seen internationally. The most up-to-date emissions modeling system by the EPA, the Next Generation Air Quality Model, will utilize a wide variety of meteorologically-senstive emission sources to create a system that will bridge the gap between climate modeling and regional air quality modeling. Coupling meteorology-sensitive sources is expected to increase the computation speed and accuracy of regional emissions reports.
“The ‘Next Generation’ model is where it’s going,” Baek says. “And this is the first step that could be used as input for NOAA’s air quality forecasting modeling system.”
Story by Dylan Morgan ’22
Dylan Morgan is an environmental science major at UNC and is part of the graduating class of 2022. This summer, he is working as a communications intern at the UNC Institute for the Environment. Morgan’s career aspirations combine environmental research with journalistic storytelling, particularly in the fields of ecological restoration and natural resource conservation.