Event Date
ABSTRACT
In recent years, growing concerns over air quality have been driven by both increased wildfire
activity and the complex processes related to pollutants such as ozone. Addressing these evolving challenges requires advanced tools capable of integrating diverse environmental data and numerical models to deliver deeper insights for public health and regulatory decisions. This seminar presents research that leverages artificial intelligence (AI) and machine learning (ML) to better understand the underlying mechanisms and support data-driven decision-making in the context of wildland fire-weather and air quality management.
The presentation highlights an integrated approach to understanding and addressing wildfire smoke emissions. This includes the development of a web-based interface that visualizes the probability of smoke transport, aiding prescribed fire planning and exposure management. To complement this short-term decision support, long-term ML-based projections of surface-level smoke concentrations through future decades are introduced, providing insight into potential future exposure scenarios. Using AI/ML techniques, variations in atmospheric and vegetation characteristics are analyzed across various time scales to better understand their dynamic interactions with wildland fire-weather patterns and the resulting smoke concentrations and transport in the Western U.S.
To further demonstrate the potential of AI/ML in interpreting complex atmospheric processes and supporting air quality decision-making, the presentation also explores the use of machine learning models to improve estimates of background ozone concentrations. Since observations capture only total ozone, isolating the background component—often shaped by natural sources and long-range transport—remains a significant challenge. By fusing observational data with outputs from physical models using ML techniques, this approach provides a better estimation of background ozone, which is critical for accurately assessing ozone exceedances under the National Ambient Air Quality Standards (NAAQS).