Air Pollution Trends and Predictive Modeling for Three Cities with Different Characteristics Using Sentinel-5 Satellite Data and Deep Learning

Air pollution is a major issue with serious risks to human health and the environment. This study investigates air pollution concentrations in three cities with distinct characteristics: a city with high industrial activities, a city with high population and urbanization, and an agricultural city. The air pollution data was collected using the Sentinel-5P satellite and Google Earth Engine to apply descriptive analysis and comparison of two years, 2022 and 2023. The cities in Saudi Arabia were Al Riyadh (high population), Al Jubail (industrial), and Najran (agricultural). The selected pollutants were SO₂, NO₂, CO, O₃, and HCHO. In addition, the study investigates the variations observed in all the pollutants during the months of the year, the correlations between the pollutants, and the correlation between NO₂ and the meteorological data. Based on the findings, Al Jubail has the highest level of all the pollutants during the two years, except for NO₂, which has the highest level in Al Riyadh, which has witnessed notable urbanization development recently. Moreover, this study developed a forecasting model of the concentration of NO₂ based on the weather data and the previous values of NO₂ using Long Short-Term Memory (LSTM) and Time2Vec. The modeling proved that any model that is trained on data collected from a specific city is not suitable to predict the pollution level in another city and for another pollutant, as the three cities have different correlations to the pollutants and the weather data. The forecasting models are useful to enhance air quality monitoring and forecasting capabilities and support the implementation of proactive strategies to mitigate air pollution. The results of this study contribute to ongoing efforts to understand the dynamics of air pollution based on the city's characteristics and the period of the year.