Abstract:Ambient exposure to fine particulate matter (PM2.5) is known to harm public health in China. Satellite remote sensing measurements of aerosol optical depth (AOD) were statistically associated with in-situ observations after 2013 to predict PM2.5 concentrations nationwide, while the lack of surface monitoring data before 2013 have created difficulties in historical PM2.5 exposure estimates. Hindcast approaches using statistical models or chemical transport models (CTMs) were developed to overcome this limitation, while those approaches still suffer from incomplete daily coverage due to missing AOD data or limited accuracy due to uncertainties of CTMs. Here we developed a new machine learning (ML) model with high-dimensional expansion (HD-expansion) of numerous predictors (including AOD and other satellite covariates, meteorological variables and CTM simulations). Through comprehensive characterization of the nonlinear effects of, and interactions among different predictors, the HD-expansion parameterized the association between PM2.5 and AOD as a nonlinear function of space and time covariates (e.g., planetary boundary layer height and relative humidity). In this way, the PM2.5-AOD association can vary spatiotemporally. We trained the model with data from 2013 to 2016 and evaluated its performance using annually-iterated cross-validation, which iteratively held out the in-situ observations for a whole calendar year (as testing data) to examine the predictions from a model trained by the rest of the observations. Our estimates were found to be in good agreement with in-situ observations, with correlation coefficients (R2) of 0.61, 0.68, and 0.75 for daily, monthly and annual averages, respectively. To interpolate the missing predictions due to incomplete AOD data, we incorporated a generalized additive model into the ML model. The two-stage estimates of PM2.5 sacrificed the prediction accuracy on a daily timescale (R2 = 0.55), but achieved complete spatiotemporal coverage and improved the accuracy of monthly (R2 = 0.71) and annual (R2 = 0.77) averages. The model was then used to predict daily PM2.5 concentrations during 2000-2016 across China and estimate long-term trends in PM2.5 for the period. We found that population-weighted concentrations of PM2.5 significantly increased, by 2.10 (95% confidence interval (CI): 1.74, 2.46) μg/m3/year during 2000-2007, and rapidly decreased by 4.51 (3.12, 5.90) μg/m3/year during 2013-2016. In this study, we produced AOD-based estimates of historical PM2.5 with complete spatiotemporal coverage, which were evidenced as accurate, particularly in middle and long term. The products could support large-scale epidemiological studies and risk assessments of ambient PM2.5 in China and can be accessed via the website (http://www.meicmodel.org/dataset-phd.html).

Crowdsourced Image Driven PM2.5 Estimation Based on Hybrid 3-Channel Feature Map

PM2.5 concentration estimation using convolutional neural network and gradient boosting machine

Effective and Efficient Photo-Based PM2.5 Concentration Estimation

A PM2.5 concentration estimation method based on multi-feature combination of image patches

Full Coverage Estimation of the PM Concentration Across China Based on an Adaptive Spatiotemporal Approach

Image-Based PM2.5 Estimation From Imbalanced Data Distribution Using Prior-Enhanced Neural Networks

Applying Traffic Camera and Deep Learning-Based Image Analysis to Predict PM2.5 Concentrations

Densely Connected Convolutional Networks with Attention Long Short-Term Memory for Estimating PM2.5 Values from Images

Tracking Hourly PM2.5 Using Geostationary Satellite Sensor Images and Multiscale Spatiotemporal Deep Learning

Estimation and Analysis of PM2.5 Concentrations with NPP-VIIRS Nighttime Light Images: A Case Study in the Chang-Zhu-Tan Urban Agglomeration of China

Real-time air quality estimation based on color image processing

Estimating High-Resolution PM2.5 Concentrations by Fusing Satellite AOD and Smartphone Photographs Using a Convolutional Neural Network and Ensemble Learning

Estimating Hourly PM2.5 Concentrations Using Himawari-8 AOD and a DBSCAN-modified Deep Learning Model over the YRDUA, China

Full-coverage spatiotemporal mapping of ambient PM<sub>2.5</sub> and PM<sub>10</sub> over China from Sentinel-5P and assimilated datasets: Considering the precursors and chemical compositions

PM<sub>2.5</sub> Monitoring: Use Information Abundance Measurement and Wide and Deep Learning

Estimation of PM2.5 Concentration across China Based on Multi-Source Remote Sensing Data and Machine Learning Methods

Fusion Method Combining Ground-Level Observations with Chemical Transport Model Predictions Using an Ensemble Deep Learning Framework: Application in China to Estimate Spatiotemporally-Resolved PM2.5 Exposure Fields in 2014-2017.

Estimation of Regional Ground-Level PM2.5 Concentrations Directly from Satellite Top-of-Atmosphere Reflectance Using A Hybrid Learning Model

Improving the Estimation of PM2.5 Concentration in the North China Area by Introducing an Attention Mechanism into Random Forest

Spatiotemporal Continuous Estimates of PM2.5 Concentrations in China, 2000-2016: A Machine Learning Method with Inputs from Satellites, Chemical Transport Model, and Ground Observations.

Retrieving Hourly Seamless PM2.5 Concentration Across China with Physically Informed Spatiotemporal Connection