This session welcomes new methodological innovations in frameworks for predicting atmospheric pollutants across urban, regional, and global scales. Because emission drivers and atmospheric chemistry-transport processes are often shared, advanced tools are commonly applicable across a wide range of species with varied lifetimes. We encourage studies demonstrating how methods applied to specific pollutants can be adapted to others sharing similar environmental drivers.

We invite contributions spanning AI/ML approaches, coupled chemistry–climate systems, process-based modeling, data assimilation, inverse modeling, digital twins, and ensemble prediction. Applications may cover greenhouse gases (e.g., CO2, CH4, N2O), reactive gases (e.g., O3, NO2, CO, VOCs), and aerosols (e.g., PM2.5, black carbon, and ultrafine particles). Submissions should emphasise methodological innovation, evaluation, uncertainty quantification, and operational implementation.

We also encourage integrating multi-platform data streams to improve source attribution and forecast skill. Contributions can address predictions for high-impact contexts like urban pollution, wildfire smoke, and climate extremes.

• B - Biogeosciences
• GC - Global Environmental Change
• IN - Informatics