A Data Driven ARIMA Study of Air Quality Trends and Forecasts of an urban agglomeration

Abstract

The increasing level of pollutants in ambient air has deteriorated the air quality of Ghaziabad at an alarming rate. It is important for the common people, allergist and patients with respiratory symptoms, to have a current knowledge of adverse respiratory effects that can occur from air pollution and the potential health effects and how they might affect their health. This brought us to focus our study on air quality in Ghaziabad region. The prediction of future air quality has been carried out by analysing the AQI daily city level using data analysis techniques. Detailed analysis from 2018-2024 of AQI daily city level has been analysed in this paper by taking year 2020 as base, the year of lock down. Study shows a decline in annual averages AQI in post covid years, reflecting an improvement due to mitigation measures. In addition to descriptive and regression-based trend analysis, an AutoRegressive Integrated Moving Average (ARIMA) model was employed to capture temporal dependencies and seasonal variations in AQI values. The ARIMA-based time-series analysis enables short-term forecasting and provides a more robust understanding of air quality dynamics. The study reveals a declining trend in annual average AQI levels in the post-COVID years, indicating an overall improvement in air quality, likely attributable to mitigation measures and regulatory interventions.