Vertical Aerosol Structure Matters: Improving the AOD–PM _2.5 Link for Air Quality and Exposure

High Resolution Image Download MS PowerPoint Slide Relating satellite-derived aerosol optical depth (AOD) to surface PM 2.5 is a primary challenge in global air quality assessment, hindered by variable vertical conditions and aerosol physicochemical properties. By conditioning five years of observational data on 10 distinct layering types, we demonstrate that the AOD-PM 2.5 relationship is strongly regime-dependent. We identified specific conditions, such as deep dust with overlying anthropogenic layers, where column-to-surface coupling is robust (up to R 2 ≈ 0.7 under optimal configurations). Conversely, mixed-layer regimes exhibit physical decoupling, rendering the satellite signal nonrepresentative ( R 2 ≈ 0). For elevated transport layers, the satellite signal often acts as a temporal precursor; “previous-day” lag variance outperforms concurrent data due to downward dust mixing. Lidar-derived relative humidity profiles further differentiate the boundary layer from the free troposphere, highlighting the effect of vertical humidity gradients. Testing these findings via machine learning (Gradient Boosting, LightGBM, Random Forest) showed that incorporating lidar-derived vertical features reduces PM 2.5 prediction error. While MAIAC AOD remains a critical predictor, model performance degrades in mixed-layer regimes compared with deep-dust environments. We conclude that reliable satellite-based monitoring requires transitioning from bulk column-integrated assumptions to frameworks incorporating vertical aerosol layering and atmospheric residence time.