| Ozone is one of the trace gases in the Earth’s atmosphere.The stratosphere distributes about 90 percent of ozone,which protects humans and the environment from the sun’s ultraviolet radiation.Stratospheric ozone is the protective umbrella of life on earth,but ground-level ozone is an important pollutant affecting human health and ecological environment.In recent years,the PM2.5concentration in China has shown a decreasing trend,but the problem of ground-level ozone pollution is becoming increasingly serious,requiring urgent attention and solution.As the Chinese economy continues to grow,coal burning,petrochemical burning and vehicle exhaust emissions have increased substantially.Chinese ground-level ozone concentration is on the rise,making it one of the most polluted regions in the world.At present,most studies on ground-level ozone concentration are based on the data of observation stations,and the results only reflect the situation of the area near the stations.Satellite-based estimates of ozone concentration tend to have low spatial resolution and are not sufficiently detailed.Therefore,how to use remote sensing data to estimate high resolution and high precision is an important problem in the study of ground-level ozone.In this paper,the spatial and temporal effects of atmospheric pollutants are considered,so the GTWR-Geoi RF model is used.Sentinel-5P data is also used in this paper,combining meteorological data,emission inventory data,site data,population data,elevation data,etc.,for modelling.Finally,the high precision spatial and temporal distribution of ground-level ozone concentration in China from March 2020to February 2021 was estimated.On this basis,the concentration,pollution and population exposure of ground-level ozone in China were analyzed.The main conclusions of this paper are as follows:(1)The estimation results of the GTWR-Geoi RF model are more accurate than those of the non-mixed model.The validation results show that,among the four ozone estimation models GTWR,RF,Geoi RF and GTWR-Geoi RF,the GTWR-Geoi RF model has the highest estimation accuracy,followed by the Geoi RF model,and the RF model and the GTWR model have relatively poor estimation results.The determination coefficient R2of the final estimation results of the this paper reached0.901,MAE and RMSE were 9.51μg/m3and 13.38μg/m3,respectively,and only13.62%of the absolute value of the error at the verification point was greater than 20μg/m3.Therefore,the characteristics of ground-level ozone concentration in China can be evaluated objectively.(2)The spatial distribution of ground-level ozone concentration in China shows obvious regional and seasonal characteristics.According to the annual average ozone concentration distribution,the ground-level ozone concentration in China was the highest in the eastern region,followed by the western region,relatively low in the southwest region,and the lowest in the northeast region.The seasonal mean distribution showed that the ozone concentration was less than 105μg/m3in most parts of western China and more than 110μg/m3in most parts of eastern China in spring.In summer,the highest ozone concentration can reach 159.5μg/m3,mainly in the North China Plain.In autumn,the ground-level ozone concentration was low,and the high ozone concentration was mainly in the North China Plain and the middle and lower Yangtze River plain.The overall ozone concentration is the lowest in winter,and the relative high ozone concentration is over the Qinghai-Tibet Plateau and Guangdong Province.(3)Ozone concentration changes significantly within a year.The ozone concentration presents a wavy curve throughout the year,with the monthly mean ozone concentration reaching a maximum in June and a minimum in December.The daily scale time series shows that the maximum concentration of ground-level ozone ranges from 131.8 to 276.3μg/m3,the minimum from 4.3 to 51.6μg/m3,and the average from 60.3 to 117.5μg/m3.The coefficient of variation shows that the regions with the most dramatic changes are the western part of Shandong Province,the southern part of Henan Province,the central and southern parts of the Beijing-Tianjin-Hebei region,the Xinzhou Basin and the Fenhe Plain.The regions with dramatic changes are the western part of Liaoning Province,the border of the Inner Mongolia Autonomous Region,Jilin Province and Heilongjiang Province,the southern part of Henan Province,the northern part of Anhui Province,the western part of Jiangsu Province,the northern part of Xinjiang Uygur Autonomous Region,and the border of Sichuan and Chongqing.The Tibetan Plateau has seen the most gradual changes.(4)Ground-level ozone pollution is serious in eastern China.The number of days exceeding the first-level limit was mainly in North China Plain,Southeast China,Xinjiang,Gansu and Inner Mongolia.The spatial distribution of ozone is similar to that of ground-level ozone.Ozone pollution mainly occurs in the eastern part of China.The most polluted areas were in northwestern Shandong Province,central and southern Hebei Province,and northern Henan province,with ozone pollution exceeding 50 days.The southeast,northeast and Sichuan Basin had fewer days of pollution.The western region,except Ningxia Hui Autonomous Region,Inner Mongolia Autonomous Region,Shaanxi Province and Sichuan Province,has virtually no ozone pollution.Ozone pollution is mainly concentrated in summer,but short-term ozone pollution events may occur at other times.(5)The risk of ground-level ozone population exposure varies greatly.The areas with high ozone exposure risk are mainly distributed in eastern China.The exposure risk of population in this area is mostly between 0.8 and 5,and the risk level is relatively high.The exposure risk of ozone population in the central areas of the Beijing-Tianjin-Hebei,Central Plains,Chengdu-Chongqing,Yangtze River Delta and Guangdong-hong Kong-Macao Greater Bay Area is over 20,which is extremely high.The ozone exposure risk of population in western China is mostly below 0.2.Especially in the Qinghai-Tibet Plateau,Kunlun Mountains,Tianshan Mountains and Inner Mongolia Plateau,ozone population exposure risk value maintained at a very low level. |