Speaker
Description
Abstract:
Ground-level zone (O₃) is a key air pollutant influenced by air pollutants and meteorological conditions. Understanding the causal relationships between ground-level ozone and (i) meteorological factors and (ii) other air pollutants is critical for two reasons: 1) accurately predicting and forecasting and 2) air quality management. This research investigates the causality and time-lagged effects of air pollutants and meteorological factors on ozone formation and depletion using Granger causality tests, cross-correlation analysis, multiple linear regression (MLR), deep NARMAX model, and structural equation modelling (SEM). A time-series dataset containing hourly meteorological and ozone concentration data over multiple years is analysed to determine key drivers of ozone fluctuations. The dataset was downloaded from the Romanian Environmental Agency’s website (https://calitateaer.ro/) from four air quality monitoring stations from Dolj County. It spans five years (January 1, 2020 – December 31, 2024). The results provide insights into the delayed effects of air pollutants and meteorological conditions on ozone pollution and help develop more effective predictive models for air quality management.
Keywords: ground-level ozone, causal analysis, meteorological variables, time-series analysis, Granger causality, cross-correlation, structural equation modelling, air pollution, forecasting.
References:
[1] El Mghouchi Y, Udristioiu MT, Yildizhan H, Brancus M. Forecasting ground-level ozone and fine particulate matter concentrations at Craiova city using a meta-hybrid deep learning model. Urban Climate 2024;57:102099.
[2] Udristioiu MT, EL Mghouchi Y, Yildizhan H. Prediction, modelling, and forecasting of PM and AQI using hybrid machine learning. Journal of Cleaner Production 2023;421:138496.
