A topological approach in analyzing the shifts in air pollutants' dynamics pre- and post-COVID-19 lockdown era

The COVID-19 pandemic, which intensified in early 2020, disrupted global activities and significantly impacted human behavior, leading to changes in air pollutant behavior that were ambiguous. These shifts introduced complex variations in urban air quality. This study aims to investigate whether air pollutants' behaviors post-COVID-19 lockdown have reverted to pre-lockdown patterns or if the lockdown-induced changes have persisted. Daily concentrations of five key pollutants, PM₁₀, PM_2.5, NO₂, CO, and O₃, recorded at 500 monitoring stations across the United States between 2017 and 2023 were analyzed. Utilizing the Mapper algorithm, a topological data analysis (TDA) method, yearly topological networks were constructed to visualize pollutant dynamics. The structure of these networks was further examined using degree distribution line plots and network density metrics. Results revealed widespread disruptions in pollutant patterns during the 2020 lockdown across all pollutants. While most pollutants exhibited varying degrees of reversion to pre-lockdown behavior in subsequent years, O₃ displayed a distinct and persistent new structural pattern, and CO demonstrated only subtle changes. These findings highlight the importance of the qualitative TDA approach in uncovering hidden patterns in complex air quality data, complementing traditional statistical methods and offering insight into the long-term impacts of COVID-19 lockdown on air pollutant behavior, potentially informing future environmental policies.