AI-Driven Innovation in Atmospheric Chemistry and Composition–Climate Interactions

Artificial intelligence (AI) is transforming how atmospheric composition and chemical processes are observed, interpreted, and modeled. The complexity of atmospheric chemistry—shaped by nonlinear reactions, multiscale transport, and strong coupling with meteorology—has long challenged conventional retrievals and chemical transport models. Emerging AI methods, including physics-informed networks and multimodal data fusion, now enable more accurate satellite and in situ retrievals, improved representation of unobserved species, and refined constraints on key chemical pathways. These advances enhance our ability to characterize aerosols, ozone, and reactive gases, diagnose emissions, and resolve fine-scale chemical structures relevant to both pollution events and background atmospheric composition.

AI is also reshaping the study of chemistry–climate interactions. Hybrid AI–physical approaches can emulate computationally intensive chemical mechanisms, accelerate composition–climate models, refine emissions and boundary conditions, and support rapid prediction of chemical influences on clouds, radiation, and circulation. As aerosols and reactive gases continue to drive radiative forcing, cloud adjustments, and large-scale dynamical responses, AI offers new pathways to quantify feedbacks, reduce uncertainties, and explore chemical controls on climate variability and extreme events. This Collection highlights advances that integrate AI across the full chain of atmospheric chemistry research—from observations and chemical process understanding to composition-driven weather and climate impacts.

We invite Original Research, Reviews and Perspectives that examine how AI and data innovation can advance atmospheric chemistry, atmospheric composition research, and chemistry–climate interactions. Contributions may address, but are not limited to:

• AI-enhanced retrievals of aerosols, trace gases, and chemically reactive intermediates
• Machine learning frameworks for accelerating chemical mechanisms or emulating chemical transport processes
• AI-based methods for quantifying emissions, chemical budgets, and composition-driven radiative forcing
• Multisensor fusion and high-resolution mapping of atmospheric composition and its dynamical or chemical signatures
• AI applications to aerosol–cloud–radiation interactions, heterogeneous chemistry, or composition–weather coupling
• Approaches for uncertainty quantification, physical consistency, and domain adaptation in AI-enabled chemistry and climate models
• Integrated AI–observation–model systems for investigating chemical variability, extreme pollution episodes, or chemistry–climate feedbacks