Abstract:Significance Ozone is an important trace gas in the atmosphere and can affect the state and process of the troposphere and stratosphere. About 90% of ozone is concentrated in the stratosphere (10-50 km) and can absorb ultraviolet radiation from the sun, thus affecting the atmospheric circulation and the earth's climate, and protecting the earth's life system. 10% of ozone is located in the troposphere, which exerts an important influence on atmospheric chemistry, air quality, and climate change, and its spatial distribution is affected by both cross-regional transport and regional production. The main source of near- surface ozone is a photochemical reaction, and its main precursors are carbon monoxide (CO), nitrogen oxides (NOx), and volatile organic compounds (VOCs). In addition, near-surface ozone concentration is also affected by meteorological conditions and regional transport. In recent years, ozone has become the primary pollutant after PM2.5 in China and even the world, especially in summer and autumn. Correspondingly, ozone pollution prevention and control have been the focus of air pollution control in the future. Ozone data can be obtained by ground-based, sounding, airborne, and space-borne observations. The ground-based observation stations can provide spatial-temporal distribution information of ozone. The data at each site are of high accuracy and good stability with the insufficient spatial representation of the sites, and the ozone concentration in the whole troposphere is not well- reflected. The vertical distribution characteristics of atmospheric ozone can be obtained by sounding and airborne observations, which can be employed to verify the satellite observation accuracy. However, the lack of spatial- temporal continuity makes it difficult to obtain the ozone distribution in a large area. As space-borne observations are not subject to geographical restrictions, it is possible to acquire global ozone spatial-temporal distribution information with all-weather coverage and provide hyperspectral and high-precision data. Therefore, high-precision, global, and all-weather ozone information can be obtained based on multiple satellite detection payloads. Progress Currently, the global ozone detection instruments are divided into three detection methods of nadir observation, occultation observation, and limb-viewing (Fig. 1). The total ozone column with high precision and ozone profiles with low vertical resolution can be obtained by the nadir observation. The ozone profile can be detected by limb-viewing and occultation observation. Occultation observation features high vertical resolution and precision, but with limited sampling frequency and small data volume. In contrast, limb-viewing can detect ultraviolet, infrared, and microwave bands, and it has high sampling frequency and can realize all-weather sampling. According to the detection spectrum and detection principles, global ozone detection instruments can be divided into ultraviolet spectral detection sensors and infrared spectral detection sensors. Based on the satellite development technologies, the inversion algorithms of the total ozone column and ozone profile are proposed (Figs. 3 and 4), and the estimation method of near-surface ozone is developed by integrating multi-source data. The whole layer ozone information and the vertical ozone distribution information can be obtained from the ultraviolet spectrum and infrared spectrum of satellites respectively. The monitoring accuracy of the total ozone column has currently reached 90%, but the inversion accuracy of the ozone concentration in the middle and lower troposphere and near the surface needs to be improved. According to the current level of inversion technology, the combination of various technical methods can be adopted to improve the detection capability of the middle and lower ozone. The application of various ozone satellite remote sensing can be carried out in the technology of atmospheric ozone detection and inversion. Our study focuses on ozone pollution progress, including the analysis of spatial-temporal characteristics of ozone pollution and typical pollution events, and the interaction between ozone pollution and meteorological conditions. The different meteorological factors can affect ozone pollution precursors. Quantifying the influence of meteorological conditions on the photochemical reaction process of ozone is an important prerequisite for formulating scientific emission reduction schemes to improve air quality. The analysis of typical ozone pollution processes can clarify the formation mechanism, development process, and subsequent evolution of near-surface ozone pollution. Conclusions and Prospects The continuous development of instrument design and inversion technology of various satellite detection payloads makes it possible for satellite remote sensing inversion and monitoring applications of ozone. The supervision and control of ozone pollution need to find out the source and accurately evaluate the pollution cases, which can be gradually analyzed in precursor emissions, chemical conversion, meteorological influence, and three-dimensional transport. The synergistic emission reduction of VOCs and NOx is the ozone treatment fundamental in China, and it is also the major research direction in the next step.

Research on satellite data-driven algorithm for ground-level ozone concentration inversion: case of Yunnan, China

High-spatial resolution ground-level ozone in Yunnan, China: A spatiotemporal estimation based on comparative analyses of machine learning models

Combining physical mechanisms and deep learning models for hourly surface ozone retrieval in China

A Generic Model to Estimate Ozone Concentration From Landsat 8 Satellite Data Based on Machine Learning Technique

A Deep Forest Algorithm Based on TropOMI Satellite Data to Estimate Near-Ground Ozone Concentration

A High-Performance Convolutional Neural Network for Ground-Level Ozone Estimation in Eastern China

First estimation of hourly full-coverage ground-level ozone from Fengyun-4A Satellite using machine learning

Estimating Daily Ground-Level NO2 Concentrations over China Based on TROPOMI Observations and Machine Learning Approach

Understanding the spatial and seasonal variation of the ground-level ozone in Southeast China with an interpretable machine learning and multi-source remote sensing

Estimation of near-surface ozone concentration and analysis of main weather situation in China based on machine learning model and Himawari-8 TOAR data

Estimation of high spatial resolution ground-level ozone concentrations based on Landsat 8 TIR bands with deep forest model

Spatiotemporal distributions of surface ozone levels in China from 2005 to 2017: A machine learning approach

Estimating Monthly Surface Ozone Using Multi-Source Satellite Products in China Based on Deep Forest Model

Progress and Challenges of Ozone Satellite Remote Sensing Inversion

Estimates of daily ground-level NO2 concentrations in China based on big data and machine learning approaches

Predicting plateau atmospheric ozone concentrations by a machine learning approach: A case study of a typical city on the southwestern plateau of China

Diurnal hourly near-surface ozone concentration derived from geostationary satellite in China

Unraveling the Influence of Satellite-Observed Land Surface Temperature on High-Resolution Mapping of Ground-Level Ozone Using Interpretable Machine Learning

Estimating ground-level high-resolution ozone concentration across China using a stacked machine-learning method

Machine Learning-Based Estimation of Ground-Level NO2 Concentrations over China.

Estimation of Near-Ground Ozone With High Spatio-Temporal Resolution in the Yangtze River Delta Region of China Based on a Temporally Ensemble Model