AI-Powered Flood Assessment Tool Developed By MBZUAI In Response To Gulf Weather Events

In the wake of record-breaking weather conditions experienced across the Gulf region on April 16, the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) has embarked on a pioneering project. Under the guidance of Dr. Salman Khan, Associate Professor of Computer Vision, a dedicated team of researchers and students is leveraging artificial intelligence (AI) and computer vision to develop an automated satellite data analysis prototype aimed at enhancing flood assessment capabilities.

This innovative tool utilizes spatial satellite data to conduct a comparative analysis of remotely sensed imagery before and after severe weather events. By focusing on three localized case studies - the Palm in Dubai, the Musaffah area in Abu Dhabi, and the AlBuraimi region in Oman - the team aims to offer a valuable change detection resource for local authorities and municipalities. This tool is designed to facilitate rapid assessment of heavy rain impacts, identifying critical infrastructure and areas most at risk due to water accumulation.

The prototype relies on AI models and freely accessible data sources, including Sentinel-2 imagery with a 10-meter resolution and OpenStreetMap. This approach not only ensures the authenticity of geographic information but also underscores the community's role in contributing to data accuracy. Dr. Khan highlighted the prototype's potential by revealing findings from the Dubai case study, where 140 kilometers of roads were affected by rain, impacting vital facilities such as cafes, pharmacies, malls, and educational institutions.

Dr. Khan expressed optimism that such automated tools could significantly aid local authorities in pinpointing critical areas needing attention post-extreme weather events. Furthermore, he emphasized the potential for these case studies to expand into broader analyses, including mapping recovery timelines and forecasting future urban area impacts through historical data analysis.

MBZUAI's collaboration with IBM on AI-enabled solutions for detecting urban heat islands represents another step towards mitigating adverse weather effects. By identifying areas prone to excessive heat and analyzing contributing factors, this solution aims to assist city planners and residents in creating more livable urban environments amidst unpredictable weather patterns.

Looking ahead, Dr. Khan shared MBZUAI's ambitions to broaden their flood study across the Gulf region and collaborate with government bodies, industry organizations, and local developers like Aldar and Emaar. The goal is to refine and enhance the tool with additional data parameters for more accurate assessments. This initiative underscores MBZUAI's commitment to leveraging AI for societal benefits, particularly in making communities more resilient against extreme weather events and aiding in long-term urban planning and policy development.

By connecting with UAE government authorities such as the Environment Agency – Abu Dhabi (EAD) and Department of Municipalities and Transport (DMT), as well as engaging with local developers and the UAE flood assessment committee, MBZUAI aims to demonstrate the value of their findings. This collaborative effort highlights the potential of AI in addressing climate change-related challenges, offering hope for more effective management of extreme weather events through technological innovation.

With inputs from WAM