Correcting Measurement Error in Satellite Aerosol Optical Depth with Machine Learning for Modeling PM2.5 in the Northeastern USA

Abstract

Satellite-derived estimates of aerosol optical depth (AOD) are key predictors in particulate air pollution models. The multi-step retrieval algorithms that estimate AOD also produce quality control variables but these have not been systematically used to address the measurement error in AOD. We compare three machine-learning methods: random forests, gradient boosting, and extreme gradient boosting (XGBoost) to characterize and correct measurement error in the Multi-Angle Implementation of Atmospheric Correction (MAIAC) 1 × 1 km AOD product for Aqua and Terra satellites across the Northeastern/Mid-Atlantic USA versus collocated measures from 79 ground-based AERONET stations over 14 years. Models included 52 quality control, land use, meteorology, and spatially-derived features. Variable importance measures suggest relative azimuth, AOD uncertainty, and the AOD difference in 30–210 km moving windows are among the most important features for predicting measurement error. XGBoost outperformed the other machine-learning approaches, decreasing the root mean squared error in withheld testing data by 43% and 44% for Aqua and Terra. After correction using XGBoost, the correlation of collocated AOD and daily PM2.5 monitors across the region increased by 10 and 9 percentage points for Aqua and Terra. We demonstrate how machine learning with quality control and spatial features substantially improves satellite-derived AOD products for air pollution modeling.

Predicting Fine Particulate Matter (PM2.5) in the Greater London Area: An Ensemble Approach using Machine Learning Methods

“…Many PM 2.5 prediction models rely on AOD as the main predictor of PM 2.5 concentrations. Yet, there is error associated with satellite measurements of AOD as well [61,62]. However, given that we are predicting a continuous outcome, this measurement error should not cause bias and it is accounted for in the residuals of the prediction model.…”

Mapping and Understanding Patterns of Air Quality Using Satellite Data and Machine Learning

“…An advantage of tree-based methods such as RF or GBRT is that, compared to deep learning methods, (Elith et al, 2008). GBRT have shown to have good predictive power in previous studies (Elith et al, 2008;Fuchs et al, 2018;Just et al, 2018). The general framework of setting up the model is shown in Figure 2 and includes input feature selection, hyperparameter tuning, model training, and model validation.…”

A Robust Deep Learning Approach for Spatiotemporal Estimation of Satellite AOD and PM2.5

“…To fairly evaluate the proposed approach, GAM, regular feed-forward neural network, and XGBoost were also trained as base models using the same dataset, and the results were compared with those of the proposed method. As a scalable end-to-end tree boosting learning system, XGBoost (https://xgboost.readthedocs.io) (see Supplementary Section S5 for details) is widely used to achieve state-of-the-art results in many domains [65], and Just et al (2018) [96] used it to correct the measurement errors of MAIAC AOD. Here, as a state-of-the-art method, XGBoost was tested for spatiotemporal estimation of PM 2.5 and compared with the proposed approach.…”