Air Pollution Assessment and Computational Air Quality Modeling in North Kuwait

Mon, 08 Jan 2024 00:00:00 +0000

The thesis titled “Air Pollution Assessment and Computational Air Quality Modeling in North Kuwait” by Ahmed Ewis focuses on evaluating air quality in the urban areas of Al Jahra and Saad Al Abdullah in Kuwait. It examines hourly pollution levels of nitrogen dioxide (NO2) and ozone (O3), using advanced statistical methods in R programming language to analyze their relationship and exceedances against the Kuwait Environmental Public Authority (KEPA) standard. The study employs various machine learning and deep learning models, including boosting techniques like CatBoost, to accurately predict hourly pollutant concentrations. The research findings highlight the significant seasonal variations in NO2 and O3 concentrations and the robust negative correlation between them. The study emphasizes the importance of monitoring and improving air quality in Kuwait, offering insights for future research and policy-making in the field of air quality assessment.

In conclusion, this thesis this sheds light on the crucial issue of air quality in Kuwait and provides a comprehensive evaluation for the hourly data of the air quality in the urban localities of Al Jahra and Saad Al Abdullah in Kuwait. The research aimed to identify the hourly pollution levels of nitrogen dioxide (NO2) and ozone (O3) and establish correlations between them. The study found a clear seasonal pattern in the concentration levels of NO2 and O3, with higher NO2 concentrations in the winter and higher O3 concentrations in the summer, this study provides insights into the extent and nature of the problem. The use of advanced statistical methods, including machine learning and deep learning techniques, provided insights into the relationship between NO2 and O3. The results indicated a strong negative correlation between the two pollutants, with the CatBoost algorithm demonstrating the highest level of accuracy for this type of data. These findings provide a basis for future research in the field and allow for better forecasting and decision making by the authorities. This thesis serves as a valuable contribution to the field of air quality assessment and underscores the importance of monitoring and improving air quality in Kuwait.

For future recommendations, experts advise collecting more data and retraining the model to capture sudden changes in behavior, to ensure that the model can capture the most recent patterns and variations in the data. This will ensure that the model remains up-to-date and accurate, providing more valuable insights and information for monitoring and improving air quality in Kuwait and beyond. However, it is important to note that data drift, where the underlying statistical properties of the data change over time, can affect the accuracy of the model. To avoid data drift and ensure the sustainability of the model, it is recommended to regularly collect more data and retrain the model to capture sudden changes in behavior. This will help in maintaining the model’s accuracy and relevance, and ensure that it continues to provide valuable insights into the relationship between NO2 and O3 and their impact on air quality.

Test on streamlit webapp: https://ahmedewis-ozone-predictor-by-catboost-cat-api-1x573m.streamlit.app/