Zeeman: A Deep Learning Regional Atmospheric Chemistry Transport Model

2025/10/07 by Mijie Pang, Pang, Mijie, Jianbing Jin +11
Environmental Science · #Air Quality Monitoring and Forecasting #Atmospheric and Environmental Gas Dynamics #Atmospheric and Oceanic Physics (physics.ao-ph) #FOS: Physical sciences #Water Quality Monitoring and Analysis

paper · pdf · doi:10.48550/arxiv.2510.06140

openalex publication_date 2025/10/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Atmospheric chemistry encapsulates the emission of various pollutants, the complex chemistry reactions, and the meteorology dominant transport, which form a dynamic system that governs air quality. While deep learning (DL) models have shown promise in capturing intricate patterns for forecasting individual atmospheric component - such as PM2.5 and ozone - the critical interactions among multiple pollutants and the combined influence of emissions and meteorology are often overlook. This study introduces an advanced DL-based atmospheric chemistry transport model Zeeman for multi-component atmospheric chemistry simulation. Leveraging an attention mechanism, our model effectively captures the nuanced relationships among these constituents. Performance metrics demonstrate that our approach rivals numerical models, offering an efficient solution for atmospheric chemistry. In the future, this model could be further integrated with data assimilation techniques to facilitate efficient and accurate atmospheric emission estimation and concentration forecast.

Citations

• FuXi-ENS: A machine learning model for medium-range ensemble weather forecasting
• Do AI models produce better weather forecasts than physics-based models? A quantitative evaluation case study of Storm Ciarán
• Learning skillful medium-range global weather forecasting
• Accurate medium-range global weather forecasting with 3D neural networks
• FuXi: A cascade machine learning forecasting system for 15-day global weather forecast
• Swin Transformer V2: Scaling Up Capacity and Resolution
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows

Related

• A deep-learning model for predicting daily PM2.5 concentration in response to emission reduction
• Offline Meteorology-Pollution Coupling Global Air Pollution Forecasting Model with Bilinear Pooling
• Atmospheric chemistry surrogate modeling with sparse identification of nonlinear dynamics
• PCDCNet: A Surrogate Model for Air Quality Forecasting with Physical-Chemical Dynamics and Constraints
• Toward Mechanistic Interpretability of an AI Foundation Model Fine-Tuned for Atmospheric Chemistry