Dust storms. Wildfires. Industrial chemical leaks.
These events pose a significant risk to human health in the form of increased concentrations of ozone and particulate matter. Public health officials need to forecast the impact these events will have on air quality and human health, as well as recommend the best preventive measures for reducing exposure.
Yunsoo Choi, assistant professor in the Department of Earth and Atmospheric Sciences, has developed a new air quality forecasting tool, called the Screening Trajectory Ozone Prediction System, or STOPS for short.
STOPS was developed and applied to the prediction of dust storms in South Korea. In a recent paper, published in the journal Geoscientific Model Development, STOPS forecast a dust storm in South Korea at a significantly faster and more accurate rate than current approaches.
“Our approach offers forecasting capabilities for unexpected disturbances, which require very fast calculations,” Choi said. “This approach also provides a comprehensive forecast of chemical reactions that are going on within these air masses.”
This new tool will be expanded to predict the effect of other disturbances such as wildfires, oil and gas “upset” emissions such as the Deepwater Horizon oil spill, and nuclear fallouts. Its speed and accuracy offers the potential for real-time forecasting capabilities.
Particulate matter is made up of a mixture of solid particles and liquid droplets. Some of these particles are so small they can enter the tiny alveoli of the lungs, irritating the tissue and impairing lung function. This leads to respiratory problems, as well as cardiovascular diseases such as coronary artery disease.
Some particulate matter is emitted directly from wildfires, dust storms or construction sites. Other particulate matter results from pollutants, such as emissions from power plants or cars, combining in chemical reactions.
Forecasting air quality is challenging for two reasons. First, the chemical reactions of pollutants to form particulate matter adds extra complexity to these calculations. Second, air mass movements are unpredictable. Current approaches are both time-consuming and overly simplistic.
“STOPS as a forecasting tool is fast and comprehensive,” Choi said.
The STOPS tool uses remote sensing data, called aerosol optical depth measurements, which provide measurements of the levels of particulate matter in the atmosphere, and applies these within a hybrid computational framework.
This computational framework takes the Eulerian method, which calculates air quality according to a grid, and combines it with the Lagrangian method, which calculates the trajectory of large air masses. Unlike other methods, STOPS can quickly predict where an air mass will end up without needing to know the origin.
In addition to providing real-time capabilities, STOPS also has the potential to make long-term predictions that can account for factors such as climate change.
“In the next 100 years, how will ozone and particulate matter levels change as a result of climate change? Modeling these effects using other Eulerian methods would be time-consuming and expensive. Using STOPS, we can make these long-term sensitivity predictions,” Choi said.
Rachel Fairbank, College of Natural Sciences and Mathematics