Predicting plateau atmospheric ozone concentrations by a machine learning approach: A case study of a typical city on the southwestern plateau of China.
Journal:
Environmental pollution (Barking, Essex : 1987)
PMID:
39368623
Abstract
Atmospheric ozone (O) has been placed on the priority control pollutant list in China's 14th Five-Year Plan. Due to their unique meteorological conditions, plateau regions contain high concentrations of atmospheric O. However, traditional experimental methods for determining O concentrations using automatic monitoring stations cannot predict O trends. In this study, two machine learning models (a nonlinear auto-regressive model with external inputs (NARX) and a temporal convolution network (TCN)) were developed to predict O concentrations in a plateau area in the Kunming region by considering the effects of meteorological parameters, air quality parameters, and volatile organic compounds (VOCs). The plateau O prediction accuracy of the machine learning models was found to be much higher than those of numerical models that served as a comparison. The O values predicted by the machine learning models closely matched the actual monitoring data. The temporal distribution of plateau O displayed a high all-day peak from February to May. A correlation analysis between O concentrations and feature parameters demonstrated that humidity is the feature with the highest absolute correlation (-0.72), and was negatively correlated with O concentrations during all test periods. VOCs and temperatures were also found to have high positive correlation coefficients with O during periods of significant O pollution. After negating the effects of meteorological parameters, the predicted O concentrations decreased significantly, whereas they increased in the absence of NO. Although individual VOCs were found to greatly affect the O concentration, the total VOC (TVOC) concentration had a relatively small effect. The proposed machine learning model was demonstrated to predict plateau O concentrations and distinguish how different features affect O variations.