Estimating high-resolution PM₁ concentration from Himawari-8 combining extreme gradient boosting-geographically and temporally weighted regression (XGBoost-GTWR)

As a much finer particle, particulate matter less than 1 μm (PM₁) plays an important role on the haze formation and human health. However, the capability of mapping PM₁ concentration is severely impaired by coarse temporal resolution and low estimation accuracy, largely due to the neglect of spatial or temporal autocorrelation of PM₁. In order to improve the estimation of high-resolution PM₁, here we developed a novel spatiotemporal model named extreme gradient boosting (XGBoost)-geographically and temporally weighted regression (GTWR) using Himawari-8 aerosol optical depth (AOD), meteorological factors, and geographical covariates. The estimation of PM₁ over Zhejiang province showed that XGBoost-GTWR method was characterized by greater predictive ability (10-fold cross-validation R² = 0.83, root mean squared error (RMSE) = 10.72 μg/m³) compared with other 11 models. Additionally, the extrapolation test was performed to validate the robustness of the hybrid model and the result demonstrated that XGBoost-GTWR can accurately predict the out-of-band PM₁ concentration (R² = 0.75 (0.60), RMSE = 12.71 (12.58) μg/m³). The PM₁ concentration displayed pronounced spatial heterogeneity, with the highest value in Quzhou (34.72 ± 1.77 μg/m³) and the lowest in Zhoushan (26.39 ± 1.56 μg/m³), respectively. In terms of the seasonality, the highest PM₁ concentration was observed in winter (39.06 ± 3.08 μg/m³), followed by those in spring (32.54 ± 3.09 μg/m³) and autumn (30.97 ± 4.50 μg/m³), and the lowest one in summer (25.57 ± 5.22 μg/m³). The high aerosol emission and adverse meteorological conditions (e.g., low boundary layer height and lack of precipitation) were key factors accounting for the peak PM₁ concentration observed in winter. Also, the PM₁ concentration exhibited significant diurnal variation, peaking at 1500 local solar time (LST) but reaching the lowest value at 1000 LST. This method enhances our capability of estimating hourly PM₁ from space, and lays a solid data foundation for improving the assessment of the fine particle-related health effect.