Correcting Aerosol Extinction Coefficient Vertical Structure Biases in GEOS-Chem via a Physics-Informed Transformer with Physical Mechanism Diagnosis

Abstract. We propose a physics-informed Transformer framework to correct biases in the Aerosol Extinction Coefficient (AEC, km^-1) profiles simulated by GEOS-Chem. Unlike standard Transformer, our framework features a dual-stream architecture with explicit physical constraints. It employs Gated Feature Fusion to integrate vertical structures (combining GEOS-Chem priors with MERRA-2 profiles) by dynamically identifying height-dependent drivers, and leverages Cross-Attention to incorporate MERRA-2 surface environmental constraints for modulating AEC vertical reconstruction with synoptic contexts. This approach effectively predicts systematic biases relative to Cloud-Aerosol Lidar with Orthogonal Polarization satellite observations and resolves AEC profiles, surpassing methods retrieving only aerosol layer heights. "Leave-One-Year-Out" validation over East Asia during 2017–2019 demonstrates significant AEC fidelity improvements, increasing R from 0.49–0.53 in the GEOS-Chem simulations to 0.66–0.73 and reducing RMSE by approximately 25 %. The model effectively mitigates over-diffusion, significantly reducing AEC simulation biases in the critical near-surface layer while restoring smoothed biomass burning and dust plumes. Additionally, it exhibits robust cross-continental transferability, reproducing bias patterns over North American domain (R=0.70) without retraining, confirming the internalization of universal physicochemical relationships linking atmospheric states to simulation biases. Furthermore, interpretability analysis establishes a feedback loop from data-driven correction to physical model improvement. The model identifies temperature and sensible heat flux as primary drivers to constrain boundary layer mixing, and uses environmental proxies (e.g., vegetation indices) to diagnose deficiencies in dust uplift and secondary aerosol formation. These insights provide a physical basis for refining parameterization schemes in chemical transport models.