Abstract:Although the ground-level NO2 measurement from air quality monitoring sites is relatively accurate, it is a challenge to obtain continuous spatial coverage due to the discrete distribution of sites. Thus, the tropospheric column NO2 amount from satellites with wide spatial and temporal coverage and higher resolution has been increasingly used to estimate ground-level NO2. However, most estimation methods were performed spontaneously using a simple linear model throughout the study period. These simplified models improve the efficiency of model development and enhance the generality of the model application, but they ignore the fact that contributors to changes of ground-level NO2 are not always consistent with time. This study considered the fixed and random effects of influencing factors and developed a mixed effect model (MEM) to estimate the ground-level NO2. By using the data of tropospheric NO2 in China from January 1, 2014 to June 30, 2020 and other multivariate auxiliary data such as meteorological elements and terrain elevation, the reliability of daily ground-level NO2 in typical populated areas of China estimated by the MEM was evaluated. The average of monthly R2 of 10fold CV in each study area during 2014-2020 is greater than 0.60 and the proportion of R2 greater than 0.7 is about 71%, suggesting the reliability of MEM. It is found that the ground-level NO2 distribution characteristics of each study area are more distinct, and the influential factors are also different. In addition, associated with the air quality control policies and emission reduction measures in various regions, the ground-level NO2 in each study area has shown an overall downward trend during 2014-2019. The uncertainty of daily-scale meteorological elements and boundary layer conditions can lead to varying degrees of deviations in daily-scale predictions of ground-level NO2. Validation with the station NO2 observations demonstrates that the ground-level NO2 prediction at seasonal time scale (R2 = 0.81, RMSE = 3.86 mu g/m3) performs better than those at time scales of daily and monthly (R2 = 0.65, and 0.75, RMSE = 7.92, and 6.24 mu g/m3). Therefore, the method of averaging can be used to improve the accuracy of ground-level NO2 predictions on individual dates. In summary, this study shows that MEM is a promising ground-level NO2 modeling method, and is effective for air pollution mapping in a large geographic region.

Estimating Ground-Level NO2 Concentrations Using Machine Learning Exclusively with Remote Sensing and ERA5 Data: The Mexico City Case Study

Estimating Ground-Level PM 10 in a Chinese City by Combining Satellite Data, Meteorological Information and a Land Use Regression Model

Satellite-based estimation of NO2 concentrations using a machine-learning model: a case study on Rio Grande do Sul, Brazil

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

Inferring Surface NO2 Over Western Europe: A Machine Learning Approach With Uncertainty Quantification

Estimation of ground-level NO2 and its spatiotemporal variations in China using GEMS measurements and a nested machine learning model

Improving machine-learned surface NO 2 concentration mapping models with domain knowledge from data science perspective

Retrieving Ground-Level PM2.5 Concentrations in China (2013–2021) with a Numerical Model-Informed Testbed to Mitigate Sample Imbalance-Induced Biases

Estimation of Daily NO2 with Explainable Machine Learning Model in China, 2007–2020

Estimation of daily NO 2 with explainable machine learning model in China, 2007–2020

Estimation of Ground NO2 Measurements from Sentinel-5P Tropospheric Data through Categorical Boosting

Estimation of NO2 emission strengths over Riyadh and Madrid from space from a combination of wind-assigned anomalies and a machine learning technique

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

Interpretation of Atmospheric CO2 Measurements in Mexico City

Interpretation of Atmospheric CO2&#160;Measurements in Mexico City

Estimating Nitrogen Oxides Emissions at City Scale in China with a Nightlight Remote Sensing Model

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

Estimation of Daily Ground Level Air Pollution in Italian Municipalities with Machine Learning Models Using Sentinel-5P and ERA5 Data

Assessing NO2 Exposures with High Spatiotemporal Resolution Across the Contiguous United States Using Ensemble Model

Ground-level NO2 Concentration Estimation Based on OMI Tropospheric NO2 and Its Spatiotemporal Characteristics in Typical Regions of China

Estimation of surface-level NO2 and O3 concentrations using TROPOMI data and machine learning over East Asia