Lunch Talk Series by Earth, Ocean and Atmospheric Sciences (EOAS) Thrust, HKUST (GZ)  - 空气污染-气象相互作用的模拟与预报 The Simulation and Forecast of Air Pollution-Meteorology Interactions

Air pollution has a significant impact on public health, but predicting air pollution events and their climatic effects involves considerable uncertainty. On a daily time scale, variations in weather conditions dominate air pollution fluctuations, making the understanding of air pollution-meteorology interactions critical for accurate air quality predictions and forecasts. This talk will present our findings from the past five years aimed at filling the knowledge gap regarding these interactions and improving air quality forecasting. Firstly, we have identified various mechanisms through which forest wildfires lead to fine particulate matter pollution and interact with meteorological factors across the world. The large quantities of particulate matter generated by wildfires affect radiation balance, cloud formation, and precipitation, which in turn influence the formation and transport of fire, creating a feedback loop among fire emissions, fire weather, and fire impacts. Next, the talk will introduce our operational air quality forecasts based on the WRF-GC regional air quality model and an AI-powered ozone ensemble forecast using deep convolutional neural networks. Based on this, we will also discuss the model’s performance in simulating ozone levels in China, particularly in the Pearl River Delta region, and its relationship with meteorological factors.