Random forest model based fine scale spatiotemporal O3 trends in the Beijing-Tianjin-Hebei region in China, 2010 to 2017

Ambient ozone (O-3) concentrations have shown an upward trend in China and its health hazards have also been recognized in recent years. High-resolution exposure data based on statistical models are needed. Our study aimed to build high-performance random forest (RF) models based on training data from 2013 to 2017 in the Beijing-Tianjin-Hebei (BTH) region in China at a 0.01 degrees x 0.01 degrees resolution, and estimated daily maximum 8h average O-3 (O-3 -8hmax) concentration, daily average O-3 (O-3 -mean) concentration, and daily maximum 1 h O-3 (O-3-1hmax) concentration from 2010 to 2017. Model features included meteorological variables, chemical transport model output variables, geographic variables, and population data. The test-R-2 of sample-based O-3-8hmax, O-3-mean and O-3-1hmax models were all greater than 0.80, while the R-2 of site-based and date-based model were 0.68-0.87. From 2010 to 2017, O-3-8hmax, O-3-mean, and O-3-1hmax concentrations in the BTH region increased by 4.18 mu g/m(3), 0.11 mu g/m(3), and 4.71 mu g/m(3), especially in more developed regions. Due to the influence of weather conditions, which showed high contribution to the model, the long-term spatial distribution of O-3 concentrations indicated a similar pattern as altitude, where high concentration levels were distributed in regions with higher altitude. (C) 2021 Elsevier Ltd. All rights reserved.

Automated summary

The study built high-performance random forest (RF) models to predict ozone concentrations in the Beijing-Tianjin-Hebei region in China from 2010 to 2017, showing an increasing trend, especially in more developed areas. Weather conditions played a significant role in the model, with high concentrations mainly distributed in regions with higher altitude.