Estimation Of Near-ground Nitrogen Dioxide Concentrations In China Using Random Forests Based On TropOMI

The near-ground NO₂ concentration is closely related to people's lives and health.The monitoring of near-ground NO₂ concentration in China is mainly through ground-based observations.However,the ground-based sites are limited and can't cover the entire region of China.Satellite data has the wide coverage,but there will be missing values during the scanning process.Therefore,it is necessary to integrate the two types of data to obtain the NO₂ concentration near the ground.This study is based on the random forest method to estimate China's near-surface NO₂ concentration.This study used the NO₂ data of ground stations from 2015 to 2019,and analyzed the changes in the NO₂ near the ground in China over the past five years.Next,this article studies the meteorological elements and the tropospheric NO₂ concentration data of satellite,and selects based on the relationship between these factors and the NO₂ near the ground.Finally,based on the model,combined with the selected data,a daily model for estimating the near-surface NO₂ concentration in China is established.This study draws the following conclusions:during the period 2015-2019 in China,the near-surface NO₂ concentration first increased and then decreased.The seasonal changes showed high concentrations in winter and low in summer;The temperature and the height of the boundary layer have the highest correlation with the NO₂,But the total precipitation don't put into the model;Trop OMI products have high correlation with near-surface NO₂,so Trop OMI products are selected as input data;the daily RMSE of the daily model established in this study has been significantly improved.During the COVID-19 pandemic,the near-surface NO₂ concentration in China was significantly reduced compared to 2019,so this model is suitable for estimating the daily concentration of NO₂ near-surface in China.