Air pollution poses a significant environmental and public health concern, particularly in high-traffic areas such as airports. Dhaka, one of the most polluted cities in the world, faces significant air pollution challenges, particularly in high-traffic zones like airports. This study examines the spatiotemporal variations in key atmospheric pollutants – nitrogen dioxide (NO2), sulphur dioxide (SO2), carbon monoxide (CO), and ozone (O3) – over Hazrat Shahjalal International Airport from 2019 to 2024 using Sentinel-5P satellite data via the Google Earth Engine platform. Satellite-derived TROPOMI observations were validated against ground-based air quality measurements, revealing strong temporal coherence and confirming their reliability. The analysis revealed strong seasonal patterns, with higher NO2 and SO2 levels during winter due to meteorological conditions and anthropogenic emissions, and decreased levels during the monsoon owing to enhanced atmospheric dispersion and wet deposition. Conversely, CO exhibited relatively stable annual trends, while O3 column densities increased over the study period. The impact of the COVID-19 lockdown in 2020 was evidenced by a temporary decline in pollutant levels. Time series forecasting models, including SARIMA, Holt–Winters, and stacked LSTM, were applied to predict 2024 column densities. Stacked LSTM model provided the most accurate predictions for NO2 and O3 (R2 = 0.776 and 0.635, respectively), while Holt-Winters performed best for CO (R2 = 0.877). These findings highlight the effectiveness of satellite-based remote sensing for urban air quality monitoring and the complementary strengths of statistical and machine learning models in forecasting pollutant trends. A comparative analysis with greater Dhaka suggests that Hazrat Shahjalal International significantly contributes to localized air pollution, underscoring the need for targeted mitigation strategies in such high-emission zones.