Exploring the Role of AI in Disaster Risk Reduction: APDIM leads a learning lab at APMCDRR 2024

On 15 October 2024, the Learning Lab titled “AI Innovations in Disaster Risk Reduction and Management: Enhancing Systemic Risk Analysis for a Resilient Future” was held at the Asia-Pacific Ministerial Conference for Disaster Risk Reduction (APMCDRR) in Manila, Philippines.

The event was organized by the Asian and Pacific Centre for the Development of Disaster Information Management (APDIM), in collaboration with the International Telecommunication Union (ITU), Fraunhofer Institute for Telecommunications, and the University of Padua.

Moderated by Letizia Rossano, Director of APDIM, and Lorenzo Nava, a researcher from the University of Padua, the session set the stage for discussing Artificial Intelligence’s transformative potential in Disaster Risk Reduction and Management (DRRM). 

Dr. Joel Joseph S. Marciano Jr., Director General of the Philippine Space Agency (PhilSA) delivered the keynote address, highlighting the critical role of AI in enhancing space technology for DRRM and its potential in advancing environmental monitoring capabilities. He showcased examples such as Flood Extent Mapping and National Mangrove Maps to demonstrate AI’s potential in this space, and called for a robust framework that fosters both risk mitigation and innovation.

Following the keynote, participants were invited to engage in an interactive session, sharing insights on utilizing AI in DRR.

The discussion then shifted to key issues of data annotation and addressing biases in AI models. Noel Jerome Borlogan, from PhilSA, demonstrated a real-world application of AI in disaster risk management and underscored the importance of accurate data annotation in training AI models to effectively identify flood-prone areas. Alan Marie Coy, from the UK Met Office, followed with a presentation, highlighting the necessity of being mindful of data biases that can distort AI predictions. She shared specific instances where biases in weather data can lead to inaccurate forecasts.

The experts also explored other implications of AI in disaster management. Dr. Yifang Ban, Professor and Director of the Division of Geoinformatics at the KTH Royal Institute of Technology stressed the need for transparency and model evaluations for accurate AI-driven disaster predictions, using wildfire management as a case study. Madhab Uprety, Senior Technical Adviser and Asia Pacific Focal Point at Red Cross Red Crescent Climate Centre then highlighted the human-centered design approach, emphasizing the importance of understanding human needs in the development of AI tools to ensure trustworthiness and effectiveness. 

The closing remarks were delivered by Margareta Wahlström, a distinguished DRR expert and the former Special Representative of the Secretary-General. The Learning Lab concluded with her calls for collaboration between scientists, policymakers, and communities to harness the potential of AI for disaster risk reduction and ensuring that technological advancements benefit all. 

The one-and-a-half-hour event was held at the Philippine International Convention Center in Manila. 

Dr. Yifang Ban, Mr. Lorenzo Nava, Ms. Margareta Wahlstrom, Ms. Letizia Rossano, Dr. Joel Joseph S. Marciano Jr., Mr. Noel Jerome Borlongan, Ms. Helen Marie Caughey, and Mr. Madhab Uprety.