Air pollution remains a significant global environmental and public health challenge, with nitrogen dioxide (NO2) playing a critical role due to its adverse effects on human health and atmospheric chemistry. This dissertation investigates NO2 pollution dynamics in Poland, employing cutting-edge satellite observations and advanced methodologies to address three key research objectives: (i) assessing the impact of the COVID-19 lockdown on NO2 levels, (ii) improving the conversion of satellite-derived NO2 data to surface concentrations, and (iii) quantifying nitrogen oxide (NOx) emissions from power plants.
The study utilized high-resolution satellite data from the TROPOspheric Monitoring Instrument (TROPOMI) and Ozone Monitoring Instrument (OMI). Chapter 2 revealed that the COVID-19 lockdown resulted in only minimal NO2 reductions in Poland, with observed changes falling within typical year-to-year variability, highlighting the limited impact of temporary events on air quality. Chapter 3 focuses on converting satellite-derived NO2 column retrievals into surface concentrations. This demonstrates the potential of machine learning models supported with meteorological and land-use variables to enhance estimation accuracy. Chapter 4 applied the Exponentially Modified Gaussian (EMG) model to TROPOMI retrievals, successfully quantifying NOx emissions from Poland’s Bełchatów power plant and showcasing the utility of satellite data for emission inventories and regulatory compliance. This research underscores the importance of long-term policy interventions for sustainable air quality improvements, the potential of machine learning in advancing satellite applications, and the essential role of satellite data in monitoring industrial emissions. The findings contribute to air quality research and provide valuable tools for policymakers aiming to address air pollution effectively.