Comparative Assessment of Linear Regression and Machine Learning for Analyzing the Spatial Distribution of Ground-level NO2 Concentrations: A Case Study for Seoul, Korea

Abstract

Atmospheric nitrogen dioxide (NO2) is mainly caused by anthropogenic emissions. It contributes to the formation of secondary pollutants and ozone through chemical reactions, and adversely affects human health. Although ground stations to monitor NO2 concentrations in real time are operated in Korea, they have a limitation that it is difficult to analyze the spatial distribution of NO2 concentrations, especially over the areas with no stations. Therefore, thisstudy conducted a comparative experiment ofspatial interpolation of NO2 concentrations based on two linear-regression methods(i.e., multi linear regression (MLR), and regression kriging (RK)), and two machine learning approaches(i.e., random forest (RF), and support vector regression (SVR)) for the year of 2020. Four approaches were compared using leave-one-out-cross validation (LOOCV). The daily LOOCV resultsshowed that MLR, RK, and SVR produced the average daily index of agreement (IOA) of 0.57, which was higher than that of RF (0.50). The average daily normalized root mean square error of RK was 0.9483%, which was slightly lower than those of the other models. MLR, RK and SVR showed similar seasonal distribution patterns, and the dynamic range of the resultant NO2 concentrationsfrom these three models was similar while that from RF was relatively small. The multivariate linear regression approaches are expected to be a promising method for spatial interpolation of ground-level NO2 concentrations and other parameters in urban areas.