Retrieval of Daily PM2.5 Concentrations Using Nonlinear Methods: A Case Study of the Beijing–Tianjin–Hebei Region, China

Abstract

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/m 3 ; Aqua: R = 0.85, RMSE = 33.90 µg/m 3 ). 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.

Estimating PM2.5 Concentrations in Contiguous Eastern Coastal Zone of China Using MODIS AOD and a Two-Stage Random Forest Model

“…The YRD-FJ-PRD region has a wide range of altitudes (0-2153 m) and represents two different levels of air pollution in China. The air quality report released by the Department of Environmental Protection of China showed that YRD and PRD region are the most air-polluted developed urban agglomerations in China, which has also been validated in previous studies (Li et al 2015;Li et al 2018;Yang et al 2020). Fujian Province, located between YRD to the north and PRD to the south, is a developed eastern coastal area predominated by mountainous and hilly regions.…”

Urban form, traffic volume, and air quality: A spatiotemporal stratified approach

“…Therefore, it could be helpful to use satellite-derived data to simulate the distribution of air pollutants (e.g. NO 2 (Bechle et al., 2011; Li et al., 2018) and PM 2.5 (Wu et al., 2018)). Finally, the traffic volume data were estimated with original data collected in six fixed periods.…”

Evaluation of Machine Learning Models for Estimating PM2.5 Concentrations across Malaysia

“…The data splitting, pre-processing, model tuning and variable importance analysis were executed in a R environment. SVR depends on the kernel function and due to its excellent generalization capability, it is able to minimize the overfitting [120], and therefore, it has been used for PM estimations [69,108,121]. SVR can fit the errors within a certain threshold by finding an appropriate boundary line (between hyperplane) to suit the data.…”

“…The flexibility of SVR depends on the selection of the parameter such as kernel function, cost function and epsilon value. There are four types of kernel functions namely linear, polynomial, sigmoid and radial basis function (RBF) that were used in this study for capturing the non-linear dynamics [69], whilst cost function was used to avoid any overfitting of the data, as small cost value leads to large margin (or wide boundary line) and causes overfitting in the model. The epsilon value controls the number of support vectors used to develop the regression function, while the smaller epsilon value indicates an optimum accuracy.…”

“…Consequently, complex techniques have been developed by combining two or more statistical techniques; for instance, merging MEM and GWR [67] or incorporating MEM into GAM [68]. Nowadays, Machine Learning (ML) techniques such as deep neural network (DNN), support vector regression (SVR) and random forest (RF) enable to capture the complex relationships between parameters, exhibiting greater performance in estimating PM 2.5 [69,70] and are increasingly used in air quality studies [71][72][73][74]. Furthermore, some studies have incorporated meteorological factors with land use variables to predict the spatial and temporal variation of aeolian erosion and PM [75][76][77][78][79].…”