Advancing our ability to predict weather and air quality is essential for protecting public health, supporting emergency response, and developing effective adaptation strategies. This project will bring together artificial intelligence (AI) and geoscience to create a new generation of forecasting tools that are faster, more accurate, and more responsive than current numerical model-based systems. By leveraging large volumes of observations from satellites, ground monitors, and physical models, the outcome from this research will enable real-time forecasting of air pollution and weather patterns across wide regions with high resolution. The tools to be developed in this study will help inform policymakers to respond to extreme weather and pollution events, improving public safety and risk management planning. Research outputs and educational materials will be available through online platforms. This project will support a graduate student involved in this work. Technically, this study aims to develop a machine learning-based surrogate for chemical transport models (CTMs), which are widely used in atmospheric research but often too computationally expensive for real-time forecasting. The proposed AI model will be trained on geoscientific big data, including meteorological inputs, satellite remote sensing, and ground-based pollutant measurements, and will learn to replicate the spatial-temporal behavior of CTMs. Unlike most existing AI models, this framework explicitly i