Estimation of PM<sub>2.5</sub> concentration in China using linear hybrid machine learning model

Abstract

Abstract. Satellite remote sensing aerosol optical depth (AOD) and meteorological elements were employed to invert PM2.5 (the fine particulate matter with a diameter below 2.5 µm) in order to control air pollution more effectively. This paper proposes a restricted gradient-descent linear hybrid machine learning model (RGD-LHMLM) by integrating a random forest (RF), a gradient boosting regression tree (GBRT), and a deep neural network (DNN) to estimate the concentration of PM2.5 in China in 2019. The research data included Himawari-8 AOD with high spatiotemporal resolution, ERA5 meteorological data, and geographic information. The results showed that, in the hybrid model developed by linear fitting, the DNN accounted for the largest proportion, and the weight coefficient was 0.62. The R2 values of RF, GBRT, and DNN were reported as 0.79, 0.81, and 0.8, respectively. Preferably, the generalization ability of the mixed model was better than that of each sub-model, and R2 (determination coefficient) reached 0.84, and RMSE (root mean square error) and MAE (mean absolute error) were reported as 12.92 and 8.01 µg m−3, respectively. For the RGD-LHMLM, R2 was above 0.7 in more than 70 % of the sites and RMSE and MAE were below 20 and 15 µg m−3, respectively, in more than 70 % of the sites due to the correlation coefficient having a seasonal difference between the meteorological factor and PM2.5. Furthermore, the hybrid model performed best in winter (mean R2 was 0.84) and worst in summer (mean R2 was 0.71). The spatiotemporal distribution characteristics of PM2.5 in China were then estimated and analyzed. According to the results, there was severe pollution in winter with an average concentration of PM2.5 being reported as 62.10 µg m−3. However, there was only slight pollution in summer with an average concentration of PM2.5 being reported as 47.39 µg m−3. The period from 10:00 to 15:00 LT (Beijing time, UTC+8 every day is the best time for model inversion; at this time the pollution is also high. The findings also indicate that North China and East China are more polluted than other areas, and their average annual concentration of PM2.5 was reported as 82.68 µg m−3. Moreover, there was relatively low pollution in Inner Mongolia, Qinghai, and Tibet, for their average PM2.5 concentrations were reported below 40 µg m−3.

Using satellite data on remote transportation of air pollutants for PM2.5 prediction in northern Taiwan

“…Fourth, we evaluated the performance of the RTP model in comparison with the Base model [27] and other state-of-the-art ensemble models with the same settings: linear regression (LR) [46], AB [32], BG, RF [34,35], XGB [36][37][38], and a GAM [33,39]. Note XGB yields better performance than gradient boosting machine [47] because of using more regularized model formalization to control over-fitting [48].…”

“…In an ensemble model, a linear combination of the outputs of different individual deep learning models is used for PM prediction results. These are the popular ensemble machine learning models, AdaBoost (AB) [32], bagging regression (BG), random forest (RF) [34,35], extreme gradient boosting (XGB) [36][37][38], and a generalized additive model (GAM) [33,39]. In this work, we used the composite neural network that outperforms those ensemble models.…”

Estimation of Atmospheric PM10 Concentration in China Using an Interpretable Deep Learning Model and Top‐of‐the‐Atmosphere Reflectance Data From China’s New Generation Geostationary Meteorological Satellite, FY‐4A

“…Since the Himawari-8 satellite cannot cover Xinjiang, China (Song et al, 2021;Wei, Li, Pinker, et al, 2021), China's second-generation geostationary meteorological satellite FY-4A successfully launched on 11 December 2016, can cover the entire territory of China. Its Advanced Geosynchronous Radiation Imager (AGRI) imager can provide multi-band full-disk images with a time resolution of 15 min (Y.…”

Using Lidar and Historical Similar Meteorological Fields to Evaluate the Impact of Anthropogenic Control on Dust Weather During COVID-19

“…Since 2013, after implementing the "Ambient Air Quality Standards" (GB3095-2012), the Ministry of Environmental Protection has published air quality data including PM 2.5 and PM 10 in more than 90 cities (about 1,600 observation sites showed in Figure 10) on its official website. The datasets can provide a comprehensive analysis of the characteristics and temporal trends of air pollution in China (Fan et al, 2020;Song et al, 2021;Chen et al, 2022).…”