Extreme gradient boosting model to estimate PM2.5 concentrations with missing-filled satellite data in China

Chen, Zhaoyue; Zhang, Tian Hao; Zhang, Rong; Zhu, Ziyang; Yang, Jun; Chen, Ping Yan; Ou, Chun Quan; Guo, Yuming

doi:10.1016/j.atmosenv.2019.01.027

Cited by 161 publications

(76 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The same methodology has been applied elsewhere [24,44,45]. More recently, machine-learning methods, such as random forests [20], gradient boosting [46], and neural network [47], have been developed due to their flexibility in handling nonlinear and interactive relationships among predictors and PM. This is a highly valued characteristic in situations where the joint relationship between daily particulate matter and multiple spatial and spatiotemporal predictors is only marginally understood.…”

Section: Comparison With Previous Studiesmentioning

confidence: 99%

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

et al. 2020

View full text Add to dashboard Cite

Air pollution is one of the leading causes of mortality worldwide. An accurate assessment of its spatial and temporal distribution is mandatory to conduct epidemiological studies able to estimate long-term (e.g., annual) and short-term (e.g., daily) health effects. While spatiotemporal models for particulate matter (PM) have been developed in several countries, estimates of daily nitrogen dioxide (NO2) and ozone (O3) concentrations at high spatial resolution are lacking, and no such models have been developed in Sweden. We collected data on daily air pollutant concentrations from routine monitoring networks over the period 2005–2016 and matched them with satellite data, dispersion models, meteorological parameters, and land-use variables. We developed a machine-learning approach, the random forest (RF), to estimate daily concentrations of PM10 (PM<10 microns), PM2.5 (PM<2.5 microns), PM2.5–10 (PM between 2.5 and 10 microns), NO2, and O3 for each squared kilometer of Sweden over the period 2005–2016. Our models were able to describe between 64% (PM10) and 78% (O3) of air pollutant variability in held-out observations, and between 37% (NO2) and 61% (O3) in held-out monitors, with no major differences across years and seasons and better performance in larger cities such as Stockholm. These estimates will allow to investigate air pollution effects across the whole of Sweden, including suburban and rural areas, previously neglected by epidemiological investigations.

show abstract

Section: Comparison With Previous Studiesmentioning

confidence: 99%

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

et al. 2020

View full text Add to dashboard Cite

show abstract

“…2017b;Yu et al, 2017;Chen et al, 2018;Ma et al, 2019;Yao et al, 2019). Recently, with the release of MODIS 3-km DT aerosol products, the PM2.5 estimates can be improved to 3-km spatial resolution across China (You et al, 2016;Li et al, 2017a;He & Huang, 2018;Chen et al, 2019;Xue et al, 2019).…”

Section: Comparison With Related Studiesmentioning

confidence: 99%

“…For model performance, our newly developed STET model shows much higher accuracy with higher CV-R 2 values, smaller RMSE and MAE values than the statistical regression models (Table 2), e.g., the timely structure adaptive model (TSAM, Fang et al, 2016) model, the Gaussian model (Yu et al, 2017), 390 the Generalized Additive Model (GAM, Chen et al, 2018) model, and the GWR model (Ma et al, 2014;You et al, 2016), and the GTWR model (He and Huang, 2018). The STET model can also outperform most machine learning (ML) and deep learning approaches including the RF model (Chen et al, 2018;Wei et al, 2019e), the XGBoost model (Chen et al, 2019), the Geo-BPNN, GRNN and deep brief network (DBN) models (Li et al, 2017a(Li et al, , 2017b, and some optical combined models, e.g., the 395 Daily-GWR (D-GWR) model (He and Huang, 2018), the two-stage model (He and Huang, 2018;Ma et al, 2019;Yao et al, 2019), and the ML + GAM model (Xue et al, 2019). In addition, there are only a hanful of studies on the predictive power in PM2.5 concentrations across China.…”

mentioning

confidence: 99%

Improved 1-km-resolution PM_2.5 estimates across China using the space-time extremely randomized trees

Wei¹,

Li²,

Huang³

et al. 2019

Preprint

View full text Add to dashboard Cite

Abstract. Fine particulate matter with aerodynamic diameters &#8804;&#8201;2.5&#8201;&#956;m (PM2.5) shows adverse effects on human health and atmospheric environment. Satellite-derived aerosol products have been intensively adopted in estimating surface PM2.5 concentrations, but most previous studies failed to monitor air pollution over small-scale areas limited by coarse spatial-resolution (3&#8211;50&#8201;km) and low data-quality aerosol optical depth (AOD) products. Therefore, a new space-time extremely randomized trees (STET) model is developed that integrates spatiotemporal information to improve PM2.5 estimates at both spatial resolution and overall accuracy across China. To this end, the newly released MODIS MAIAC AOD product, meteorological and other auxiliary data are inputs to the STET model. Daily 1-km PM2.5 maps in 2018 across mainland China are produced. The STET model performs well with a high out-of-sample (out-of-station) cross-validation coefficient of 0.89 (0.88), a low root-mean-square error of 10.35 (10.97)&#8201;&#956;g/m3, a small mean absolute error of 6.71 (7.17)&#8201;&#956;g/m3, and a small mean relative error of 21.37&#8201;% (23.77&#8201;%), respectively. Particularly, it can well capture the PM2.5 concentrations at both regional and individual site scales. In addition, it posed a strong predictive power (e.g., monthly-R2&#8201;=&#8201;0.80) and can be used to predict the historical PM2.5 records. The North China Plain, the Sichuan Basin, and Xinjiang Province always are featured with high PM2.5 pollution, especially in winter. The STET model outperforms most models presented in previous related studies. More importantly, our study provides a new approach to obtain high-quality PM2.5 estimates, which is important for air pollution studies over urban areas.

show abstract

“…Statistical regression models, e.g., the multiple linear regression model, the linear mixed-effect model, the two-stage model and the geographically weighted regression (GWR) model, have been widely used for applications due to their simplicity and versatility (Gupta and Christopher, 2009;Ma et al, 2014;Xiao et al, 2017;Yao et al, 2019). Artificial intelligence models mainly involve machine learning and deep learning models, e.g., the random forest (RF; Brokamp et al, 2018;Chen et al, 2018;Wei et al, 2019a), the extreme gradient boosting model (XGBoost; Chen et al, 2019), and the back-propagation and generalized regression neural networks (BRNN and GRNN; T. Li et al, 2017a). PM 2.5 is jointly affected by numerous factors, e.g., meteorological conditions, human activities and topography, showing great spatial and temporal heterogeneities.…”

Section: Introductionmentioning

confidence: 99%

“…This makes it difficult for traditional physical and statistical regression approaches to accurately explain and construct PM 2.5 -AOD relationships, leading to poor PM 2.5 estimates. Despite their stronger data mining ability, most artificial intelligence approaches have been simplistically adopted in PM 2.5 predictions, neglecting the spatiotemporal characteristics of PM 2.5 (Brokamp et al, 2018;Chen et al, 2018Chen et al, , 2019T. Li et al, 2017a;Xue et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Improved 1 km resolution PM_2.5 estimates across China using enhanced space–time extremely randomized trees

et al. 2020

View full text Add to dashboard Cite

Abstract. Fine particulate matter with aerodynamic diameters ≤2.5 µm (PM2.5) has adverse effects on human health and the atmospheric environment. The estimation of surface PM2.5 concentrations has made intensive use of satellite-derived aerosol products. However, it has been a great challenge to obtain high-quality and high-resolution PM2.5 data from both ground and satellite observations, which is essential to monitor air pollution over small-scale areas such as metropolitan regions. Here, the space–time extremely randomized trees (STET) model was enhanced by integrating updated spatiotemporal information and additional auxiliary data to improve the spatial resolution and overall accuracy of PM2.5 estimates across China. To this end, the newly released Moderate Resolution Imaging Spectroradiometer Multi-Angle Implementation of Atmospheric Correction AOD product, along with meteorological, topographical and land-use data and pollution emissions, was input to the STET model, and daily 1 km PM2.5 maps for 2018 covering mainland China were produced. The STET model performed well, with a high out-of-sample (out-of-station) cross-validation coefficient of determination (R2) of 0.89 (0.88), a low root-mean-square error of 10.33 (10.93) µg m−3, a small mean absolute error of 6.69 (7.15) µg m−3 and a small mean relative error of 21.28 % (23.69 %). In particular, the model captured well the PM2.5 concentrations at both regional and individual site scales. The North China Plain, the Sichuan Basin and Xinjiang Province always featured high PM2.5 pollution levels, especially in winter. The STET model outperformed most models presented in previous related studies, with a strong predictive power (e.g., monthly R2=0.80), which can be used to estimate historical PM2.5 records. More importantly, this study provides a new approach for obtaining high-resolution and high-quality PM2.5 dataset across mainland China (i.e., ChinaHighPM2.5), important for air pollution studies focused on urban areas.

show abstract

Extreme gradient boosting model to estimate PM2.5 concentrations with missing-filled satellite data in China

Cited by 161 publications

References 44 publications

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

Improved 1-km-resolution PM_2.5 estimates across China using the space-time extremely randomized trees

Improved 1 km resolution PM_2.5 estimates across China using enhanced space–time extremely randomized trees

Contact Info

Product

Resources

About

Extreme gradient boosting model to estimate PM2.5 concentrations with missing-filled satellite data in China

Cited by 161 publications

References 44 publications

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

A Random Forest Approach to Estimate Daily Particulate Matter, Nitrogen Dioxide, and Ozone at Fine Spatial Resolution in Sweden

Improved 1-km-resolution PM2.5 estimates across China using the space-time extremely randomized trees

Improved 1 km resolution PM2.5 estimates across China using enhanced space–time extremely randomized trees

Contact Info

Product

Resources

About

Improved 1-km-resolution PM_2.5 estimates across China using the space-time extremely randomized trees

Improved 1 km resolution PM_2.5 estimates across China using enhanced space–time extremely randomized trees