An Ensemble Model-Based Estimation of Nitrogen Dioxide in a Southeastern Coastal Region of China

Abstract

NO2 (nitrogen dioxide) is a common pollutant in the atmosphere that can have serious adverse effects on the health of residents. However, the existing satellite and ground observation methods are not enough to effectively monitor the spatiotemporal heterogeneity of near-surface NO2 concentrations, which limits the development of pollutant remediation work and medical health research. Based on TROPOMI-NO2 tropospheric column concentration data, supplemented by meteorological data, atmospheric condition reanalysis data and other geographic parameters, combined with classic machine learning models and deep learning networks, we constructed an ensemble model that achieved a daily average near-surface NO2 of 0.03° exposure. In this article, a meteorological hysteretic effects term and a spatiotemporal term were designed, which considerably improved the performance of the model. Overall, our ensemble model performed better, with a 10-fold CV R2 of 0.89, an RMSE of 5.62 µg/m3, and an MAE of 4.04 µg/m3. The model also had good temporal and spatial generalization capability, with a temporal prediction R2 and a spatial prediction R2 of 0.71 and 0.81, respectively, which can be applied to a wider range of time and space. Finally, we used an ensemble model to estimate the spatiotemporal distribution of NO2 in a coastal region of southeastern China from May 2018 to December 2020. Compared with satellite observations, the model output results showed richer details of the spatiotemporal heterogeneity of NO2 concentrations. Due to the advantages of using multi-source data, this model framework has the potential to output products with a higher spatial resolution and can provide a reference for downscaling work on other pollutants.

Satellite-based long-term spatiotemporal trends in ambient NO2 concentrations and attributable health burdens in China from 2005 to 2020

“…Third, we proposed a gap filling method to impute the missing OMI VCD values and achieved a 100% spatiotemporal coverage in NO2 estimation. Most existing studies simply excluded or use the nearby observations to interpolate the missing values (He et al, 2022;S Wu et al, 2021a;Xu et al, 2021). This may lead to significant exposure bias in epimiological studies when there is a high missing rate.…”

“…Without considering the non-random missing values, it may lead to exposure misclassification and bias the health effects of NO2 exposure in epidemiological studies. However, most existing studies simply excluded or use the nearby observations to interpolate the missing values (He et al, 2022;Xu et al, 2021).…”

FY‐4A/AGRI Infrared Brightness Temperature Estimation of Precipitation Based on Multi‐Model Ensemble Learning

“…For an introduction to random forest, XGBoost, and decision tree models, readers can refer to Natras et al (2022); for LightGBM and extra tree models, see Feng et al (2022) and He et al (2022), respectively.…”

High-Spatiotemporal-Resolution Estimation of Ground-Level Ozone in China Based on Machine Learning

“…Compared to ground-based station monitoring, satellite remote sensing monitoring is not constrained by time, climate, or geographical limitations, facilitating large-scale synchronous observations and providing extensive spatial coverage [12][13][14]. For instance, in 2015, the Japan Aerospace Exploration Agency (JAXA) successfully launched the Himawari-8 satellite, which has a 10 min observation frequency [15].…”