A 15-year full-coverage daily PM1 dataset across China based on a data-fusion model

<p>The long-term (2004-2018) full-coverage daily PM₁ dataset across China was developed based on a stacking decision tree model assimilating satellite data, meteorological variables, and other geographical covariates. The PM₁ dataset captured a strong prediction capability with a high cross-validation (CV) R² value (0.64), and the low root-mean-square error (RMSE: 18.60 &mu;g/m³) and moderate absolute error (MAE: 11.96 &mu;g/m³). The long-term PM₁ dataset obtained here provide a key scientific basis and data support for epidemiological research and air pollution mitigation.</p>