Department of Environmental Science and Engineering, Fudan University, Shanghai, 200433, China; Shanghai Key Laboratory of Atmospheric Particle Pollution and Prevention (LAP3), Fudan University, Shanghai, 200433, China.
Department of Environmental Science and Engineering, Fudan University, Shanghai, 200433, China.
Environ Res. 2022 Jan;203:111799. doi: 10.1016/j.envres.2021.111799. Epub 2021 Jul 31.
In spite of the state-of-the-art performances of machine learning in the PM estimation, the high-value PM underestimation and non-random aerosol optical depth (AOD) missing are still huge obstacles. By incorporating wavelet decomposition (WD) into the extreme gradient boosting (XGBoost), a hybrid XGBoost-WD model was established to obtain the full-coverage PM estimation at 3-km spatial resolution in the Yangtze River Delta Urban Agglomeration (YRDUA). In this study, 3-km-resolution meteorological fields simulated by WRF along with AOD derived from Moderate Resolution Imaging Spectroradiometer (MODIS) were served as explanatory variables. Model MW and Model NW were developed using XGBoost-WD for the areas with and without AOD respectively to obtain a full-coverage PM mapping in the YRDUA. The XGBoost-WD model showed good performances in estimating PM with R of 0.80 in the Model MW and 0.87 in the Model NW. Moreover, the K-value of Model MW increased from 0.77 to 0.79 and that of Model NM increased from 0.81 to 0.86 compared with the model without the step of WD, indicating an improvement on the problem of PM underestimation. Due to a better ability of capturing abrupt changes in the PM concentrations, the spatial evolution of PM during a typical pollution event could be mapped more accurately. Finally, the analysis of variable importance showed that the three most important variables in the estimation of the low-frequency coefficients of PM (PM_A4) were temperature at 2 m (T2), day of year (DOY) and longitude (LON), while that in the high-frequency coefficients of PM (PM_D) were CO, AOD and NO. This study not only provided an effective solution to the PM underestimation and AOD missing problems in the PM estimation, but also proposed a new method to further refine the sophisticated correlations between PM and some spatiotemporal variables.
尽管机器学习在 PM 估算方面表现出色，但仍存在高价值 PM 低估和非随机气溶胶光学深度 (AOD) 缺失等巨大障碍。本研究通过将小波分解 (WD) 纳入极端梯度提升 (XGBoost) 中，建立了一个混合 XGBoost-WD 模型，以获得长江三角洲城市群 (YRDUA) 3 公里空间分辨率的全覆盖 PM 估算。在这项研究中，WRF 模拟的 3 公里分辨率气象场以及来自中分辨率成像光谱仪 (MODIS) 的 AOD 被用作解释变量。使用 XGBoost-WD 为有和没有 AOD 的区域分别开发了模型 MW 和模型 NW，以获得 YRDUA 的全覆盖 PM 映射。XGBoost-WD 模型在模型 MW 中的 PM 估计中表现出良好的性能，R 值为 0.80，在模型 NW 中的 R 值为 0.87。此外，与没有 WD 步骤的模型相比，模型 MW 的 K 值从 0.77 增加到 0.79，模型 NM 的 K 值从 0.81 增加到 0.86，表明 PM 低估问题得到了改善。由于捕捉 PM 浓度突然变化的能力更强，因此可以更准确地绘制典型污染事件期间 PM 的空间演变。最后，变量重要性分析表明，PM 低频系数 (PM_A4) 估计中最重要的三个变量是 2 米温度 (T2)、年日 (DOY) 和经度 (LON)，而 PM 高频系数 (PM_D) 则是 CO、AOD 和 NO。本研究不仅为 PM 估算中的 PM 低估和 AOD 缺失问题提供了有效的解决方案，还提出了一种新方法，以进一步细化 PM 与一些时空变量之间的复杂关系。
