Integrating satellite information, land use regression, and machine learning to estimate the spatiotemporal variation...

The growing understanding of PM_2.5 compositions has promoted the application of machine learning in exposure estimation. However, data-driven approaches may neglect the necessity for expert knowledge. This study presents a theory-guided approach combining land use regression (LUR), extreme gradient boosting (XGBoost), and Shapley additive explanations (SHAP) to estimate the spatiotemporal variation of ionic PM_2.5 in Taiwan. Ground-based monthly observations of eight PM_2.5 ions (Na⁺_, NH₄⁺, K⁺, Mg²⁺, Ca²⁺, Cl^-_, NO₃^-, and SO₄^2-) from 28 air quality monitoring stations across Taiwan, meteorological data simulated using the Weather Research and Forecasting model, and satellite-derived information (including land use, aerosol characteristics, greenness, and insolation) were collected to develop monthly predictions of ionic PM_2.5 from 2019 to 2021. LUR, XGBoost, and SHAP were applied for feature selection, model building, and the evaluation of feature impacts, respectively. The models demonstrated moderate performances for most ionic PM_2.5, except for Mg²⁺ and Cl^-, with an independent test set showing adjusted R² values of 0.69-0.87. Sensitivity analyses were conducted by removing directional selection and incorporating ground-observed anthropogenic factors; however, the model performance was not improved. This study offers insights into the interaction between contributing features and ionic PM_2.5 for climate research.