Exposure to fine particulate matter (PM_2.5) is associated with adverse health impacts on the population. Satellite observations and machine learning algorithms have been applied to improve the accuracy of the prediction of PM_2.5 concentrations. In this study, we developed a PM_2.5 retrieval approach using machine-learning methods, based on aerosol products from the Moderate Resolution Imaging Spectroradiometer (MODIS) aboard the NASA Earth Observation System (EOS) Terra and Aqua polar-orbiting satellites, near-ground meteorological variables from the NASA Goddard Earth Observing System (GEOS), and ground-based PM_2.5 observation data. Four models, which are orthogonal regression (OR), regression tree (Rpart), random forests (RF), and support vector machine (SVM), were tested and compared in the Beijing–Tianjin–Hebei (BTH) region of China in 2015. Aerosol products derived from the Terra and Aqua satellite sensors were also compared. The 10-repeat 5-fold cross-validation (10 × 5 CV) method was subsequently used to evaluate the performance of the different aerosol products and the four models. The results show that the performance of the Aqua dataset was better than that of the Terra dataset, and that the RF algorithm has the best predictive performance (Terra: R = 0.77, RMSE = 43.51 μg/m³; Aqua: R = 0.85, RMSE = 33.90 μg/m³). This study shows promise for predicting the spatiotemporal distribution of PM_2.5 using the RF model and Aqua aerosol product with the assistance of PM_2.5 site data.

In this study, we evaluated estimates and predictions of the PM_2.5 (fine particulate matter) concentrations and emissions in Xuzhou, China, using a coupled Lagrangian particle dispersion modeling system (FLEXPART-WRF). A Bayesian inversion method was used in FLEXPART-WRF to improve the emission calculation and mixing ratio estimation for PM_2.5. We first examined the inversion modeling performance by comparing the model predictions with PM_2.5 concentration observations from four stations in Xuzhou. The linear correlation analysis between the predicted PM_2.5 concentrations and the observations shows that our inversion forecast system is much better than the system before calibration (with correlation coefficients of R = 0.639 vs. 0.459, respectively, and root mean square errors of RMSE = 7.407 vs. 9.805 µg/m³, respectively). We also estimated the monthly average emission flux in Xuzhou to be 4188.26 Mg/month, which is much higher (by ~10.12%) than the emission flux predicted by the multiscale emission inventory data (MEIC) (3803.5 Mg/month). In addition, the monthly average emission flux shows obvious seasonal variation, with the lowest PM_2.5 flux in summer and the highest flux in winter. This pattern is mainly due to the additional heating fuels used in the cold season, resulting in many fine particulates in the atmosphere. Although the inversion and forecast results were improved to some extent, the inversion system can be improved further, e.g., by increasing the number of observation values and improving the accuracy of the a priori emission values. Further research and analysis are recommended to help improve the forecast precision of real-time PM_2.5 concentrations and the corresponding monthly emission fluxes.

A fully coupled simulation of ecophysiological, hydrological and biochemical processes is significant for better understanding the individual and interactional impact of sophisticated land surface processes under future disturbances from nature and human beings. In this study, we spatially explicitly modelled evapotranspiration (ET) and photosynthesis (GPP) using a distributed hydrological model, Dynamic Land Model DLM-Ecohydro, over the Upper Heihe watershed for the years of 2013 and 2014. After considering the lateral water movements, the model fairly captured the variations in ET (R² = 0.82, RMSE = 1.66 mm/day for 2013; R² = 0.83, RMSE = 1.53 mm/day for 2014) and GPP (R² = 0.71, RMSE = 5.25 gC/m²/day for 2013; R² = 0.81, RMSE = 3.38 gC/m²/day for 2014) compared with the measurements from the Arou monitoring station. Vegetation transpiration accounted for total ET of around 65% and 64% in 2013 and 2014, respectively. A large spatial variability was found in these two indicators (14.30–885.36 mm/year for annual ET and 0–2174 gC/m²/day for annual GPP) over the watershed. Soil texture and vegetation functional types were the major factors affecting ET and GPP spatial variability, respectively. The study manifested a coupled water–carbon mechanism through the strong linear relationship between the variations in ET and GPP and the control of hydrological processes on the carbon cycle at the watershed scale. Although the model had a reasonable performance during most parts of the growing seasons, the lack of a soil freezing–thawing scenario caused inevitable discrepancies for the simulation of soil water and heat transfer mechanisms, hence inaccurately estimating the biophysiological processes in the transition period of winter to spring, which should be further improved especially for alpine regions.