When a Generalized Linear Model Meets Bayesian Maximum Entropy: A Novel Spatiotemporal Ground-Level Ozone Concentration Retrieval Method

In China, ground-level ozone has shown an increasing trend and has become a serious ambient pollutant. An accurate spatiotemporal distribution of ground-level ozone concentrations (GOCs) is urgently needed. Generalized linear models (GLMs) and Bayesian maximum entropy (BME) models are practical for predicting GOCs. However, GLMs have limited capacity to capture temporal variations and can miss some short-term and regional patterns, while the performance of BME models may degrade in cases of sparse or imperfect monitoring networks. Thus, to predict nationwide 1 km monthly average GOCs for China, we designed a novel hybrid model containing three modules. (1) A GLM was established to accurately describe the variability in GOCs in the space domain. (2) A BME model incorporating GLM residuals was employed to capture the temporal variability of GOCs in detail. (3) A combination of GLM and BME models was developed based on the specific broad range of each submodel. According to the cross-validation results, the hybrid model exhibited superior performance, with coefficient of determination (R2) values of 0.67. The predictive performance of the large-scale and high-resolution hybrid model is superior to that in previous studies. The nationwide spatiotemporal variability of the GOCs derived from the hybrid model shows that they are valuable indicators for ground-level ozone pollution control and prevention in China.

[1]  A. Anav,et al.  Projected global ground-level ozone impacts on vegetation under different emission and climate scenarios , 2017 .

[2]  F. Jiang,et al.  Impacts of O3 on premature mortality and crop yield loss across China , 2018, Atmospheric Environment.

[3]  M. Zunckel,et al.  Modelled surface ozone over southern Africa during the Cross Border Air Pollution Impact Assessment Project , 2006, Environ. Model. Softw..

[4]  Spatiotemporal distribution of ground-level ozone in China at a city level , 2020, Scientific Reports.

[5]  Jonathan H. Jiang,et al.  From COVID-19 to future electrification: Assessing traffic impacts on air quality by a machine-learning model , 2021, Proceedings of the National Academy of Sciences.

[6]  Gülsen Aydin Keskin,et al.  Predicting tropospheric ozone concentrations in different temporal scales by using multilayer perceptron models , 2011, Ecol. Informatics.

[7]  Jenise L. Swall,et al.  Interannual variation in meteorologically adjusted ozone levels in the eastern United States : A comparison of two approaches , 2007 .

[8]  M. Schultz,et al.  Severe Surface Ozone Pollution in China: A Global Perspective , 2018, Environmental Science & Technology Letters.

[9]  Youfei Zheng,et al.  Temporal and Spatial Variation in, and Population Exposure to, Summertime Ground-Level Ozone in Beijing , 2018, International journal of environmental research and public health.

[10]  Enrique Roca,et al.  Identification of NOx and Ozone Episodes and Estimation of Ozone by Statistical Analysis , 2009 .

[11]  Paulin Coulibaly,et al.  Ground-level ozone forecasting using data-driven methods , 2008 .

[12]  Wei Sun,et al.  Prediction of 8 h-average ozone concentration using a supervised hidden Markov model combined with generalized linear models , 2013 .

[13]  B. Tomassetti,et al.  Analysis of surface ozone using a recurrent neural network. , 2015, The Science of the total environment.

[14]  Saleh M. Al-Alawi,et al.  Combining principal component regression and artificial neural networks for more accurate predictions of ground-level ozone , 2008, Environ. Model. Softw..

[15]  F. Solmon,et al.  Geographical distribution of ozone seasonality over China. , 2019, The Science of the total environment.

[16]  Pat Dolwick,et al.  The effects of meteorology on ozone in urban areas and their use in assessing ozone trends , 2007 .

[17]  Xiang Ren,et al.  Comparison of Machine Learning and Land Use Regression for fine scale spatiotemporal estimation of ambient air pollution: Modeling ozone concentrations across the contiguous United States. , 2020, Environment international.

[18]  Hamed Karimian,et al.  Daily spatiotemporal prediction of surface ozone at the national level in China: An improvement of CAMS ozone product , 2020 .

[19]  Alexander Kolovos,et al.  Interactive spatiotemporal modelling of health systems: the SEKS–GUI framework , 2007 .

[20]  Cheng Huang,et al.  Source apportionment of surface ozone in the Yangtze River Delta, China in the summer of 2013 , 2016 .

[21]  Kyongjin Sok,et al.  A hybrid model based on convolutional neural networks and long short-term memory for ozone concentration prediction , 2018, Air Quality, Atmosphere & Health.

[22]  J. Jason West,et al.  An Estimate of the Global Burden of Anthropogenic Ozone and Fine Particulate Matter on Premature Human Mortality Using Atmospheric Modeling , 2010, Environmental health perspectives.

[23]  Kebin He,et al.  Ground-level ozone pollution and its health impacts in China , 2018 .

[24]  S. Xie,et al.  Spatial Distribution of Ozone Formation in China Derived from Emissions of Speciated Volatile Organic Compounds. , 2017, Environmental science & technology.

[25]  G. Grigoras,et al.  Assessing of surface-ozone concentration in Bucharest, Romania, using OML and satellite data , 2016 .

[26]  George Christakos,et al.  Spatiotemporal Characterization of Ambient PM2.5 Concentrations in Shandong Province (China). , 2015, Environmental science & technology.

[27]  Xiaoke Wang,et al.  Ground-level ozone in China: distribution and effects on crop yields. , 2007, Environmental pollution.

[28]  Yeyao Wang,et al.  Regionalization based on spatial and seasonal variation in ground-level ozone concentrations across China. , 2017, Journal of environmental sciences.

[29]  Surajit Chattopadhyay,et al.  Modeling and Prediction of Monthly Total Ozone Concentrations by Use of an Artificial Neural Network Based on Principal Component Analysis , 2012, Pure and Applied Geophysics.

[30]  P. Tchounwou,et al.  Spatial Variation of Ground Level Ozone Concentrations and its Health Impacts in an Urban Area in India. , 2017, Aerosol and air quality research.

[31]  Michael Jerrett,et al.  Spatiotemporal Modeling of Ozone Levels in Quebec (Canada): A Comparison of Kriging, Land-Use Regression (LUR), and Combined Bayesian Maximum Entropy–LUR Approaches , 2014, Environmental health perspectives.

[32]  Marc L. Serre,et al.  Comparison of Geostatistical Interpolation and Remote Sensing Techniques for Estimating Long-Term Exposure to Ambient PM2.5 Concentrations across the Continental United States , 2012, Environmental health perspectives.

[33]  S. Jaber,et al.  Mapping the spatial distribution of tropospheric ozone and exploring its association with elevation and land cover over North Jordan , 2017 .

[34]  M. Luo,et al.  Seasonal and spatial variability of surface ozone over China: contributions from background and domestic pollution , 2010 .

[35]  Ozone pollution over China and India: seasonality and sources , 2020 .

[36]  U. Nair,et al.  Quantifying O3 Impacts in Urban Areas Due to Wildfires Using a Generalized Additive Model. , 2017, Environmental science & technology.

[37]  Zhao-Liang Li,et al.  A framework for the retrieval of all-weather land surface temperature at a high spatial resolution from polar-orbiting thermal infrared and passive microwave data , 2017 .

[38]  Jonathan I. Levy,et al.  Ozone Exposure and Mortality: An Empiric Bayes Metaregression Analysis , 2005, Epidemiology.

[39]  Yang Liu,et al.  Spatiotemporal distributions of surface ozone levels in China from 2005 to 2017: A machine learning approach. , 2020, Environment international.

[40]  Patrick Bogaert,et al.  Spatiotemporal modelling of ozone distribution in the State of California , 2009 .

[41]  Hui Zhang,et al.  Spatiotemporal modeling of PM2.5 concentrations at the national scale combining land use regression and Bayesian maximum entropy in China. , 2018, Environment international.

[42]  Baofeng Di,et al.  Spatiotemporal prediction of daily ambient ozone levels across China using random forest for human exposure assessment. , 2018, Environmental pollution.