ACCESS
Research Article
Volume 1, Article ID: 2025.0020
Maria Iruj
maria.iruj@gmail.com
Danish Mustafa Khan
dmustafa@gmail.com
Saima Yaqoob
engg_saima@hotmail.com
Zunaira Iqbal
zunairadanishkhan@gmail.com
1 Mechanical and Industrial Engineering Department, Abu Dhabi University, Abu Dhabi, United Arab Emirates
2 University of Hull, Kingston Upon Hull, HU6 7RX, UK
3 Department of Industrial and Manufacturing Engineering, NED University of Engineering and Technology, Karachi, Pakistan
4 PGD Sustainable Engineering, NED Academy, NED University of Engineering and Technology: Karachi, Pakistan
* Author to whom correspondence should be addressed
Received: 22 May 2025 Accepted: 28 Dec 2025 Available Online: 29 Dec 2025
This research develops a predictive AI model to monitor and forecast air quality in Abu Dhabi using publicly available, satellite-based environmental datasets. The study uses datasets from NASA's MODIS, Copernicus Atmosphere Monitoring Service (CAMS), and OpenAQ, alongside meteorological data from the UAE’s National Center of Meteorology. Supervised learning techniques, including Random Forest and LSTM neural networks, are applied to analyze PM2.5, PM10, NO₂, and CO concentration trends with temperature, humidity, wind patterns, and urban development indices. The impact of seasonal events such as sandstorms and traffic emissions on air quality are also discussed in this study. The novelty of this work lies in bridging the gap of sparse sensor networks, by adopting a real-time hybrid machine learning model that accurately forecasts Abu Dhabi air quality using only satellite and key meteorological data.
Disclaimer : This is not the final version of the article. Changes may occur when the manuscript is published in its final format.