Title: Fine-tuned regression and statistical assessment of India's air quality during COVID-19 disease

Authors: Kanu Goel; Harsh Bansal; Shivangi Sharma; Shefali Arora Chouhan

Addresses: Department of Computer Science and Engineering, Punjab Engineering College (Deemed to be University), Chandigarh, 160012, India ' Department of Computer Science and Engineering, Amity University, Punjab, Mohali, 140306, India ' Department of Computer Science and Engineering, Amity University, Punjab, Mohali, 140306, India ' Department of Computer Science and Engineering, National Institute of Technology, Jalandhar, 144011, India

Abstract: The COVID-19 lockdown in India substantially influenced air pollution levels, with the closure of factories and businesses leading to a significant reduction in key pollutants such as dust and carbon dioxide. This research explores the impact of the lockdown on air quality by analysing the air quality index (AQI) data from various metropolitan cities, focusing on the lockdown months. Utilising conventional statistical analysis and a seasonal autoregressive integrated moving average (SARIMA)model trained on pre-lockdown data, the study aims to assess the applicability of machine learning, specifically time series forecasting, in evaluating extraordinary events like the lockdown on air quality. Results indicate a 45.22% reduction in pollution levels for Delhi using statistical averaging and a 37.2% difference between SARIMA model predictions and actual measurements. These findings emphasise the potential of machine learning in modelling and predicting the outcomes of pollution mitigation measures, providing crucial insights for policymakers and environmental stakeholders.

Keywords: AQI; air quality index; COVID-19; statistical modelling; machine learning; SARIMAX model; ARIMA model.

DOI: 10.1504/IJEP.2023.139849

International Journal of Environment and Pollution, 2023 Vol.73 No.1/2/3/4, pp.84 - 106

Received: 05 Jul 2023
Accepted: 26 Feb 2024
Published online: 08 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article