CeADAR Leverages AI for Flood Early Warning System in Irish Communities

Utilizing the power of artificial intelligence (AI) paired with satellite technology, CeADAR, Ireland's centre for Applied AI, is innovating an early warning system aimed at communities potentially facing severe flooding. This initiative is a response to the increasing risk of flood events exacerbated by climate change and seeks to arm vulnerable areas with timely alerts.

Studied Areas and Satellite Data Utilization

Research under the project has encompassed several locations, including Carrick-on-Shannon in County Leitrim, Middleton in County Cork, Athlone in County Westmeath, and Limerick City. Researchers gathered historical flood data through the European Space Agency's Sentinel-1 satellite, creating flood extent maps that serve as training input for an AI model. This model predicts potential flooding with a remarkable accuracy of up to approximately 20 meters.

Project CAMEO and Flood Prediction Model

The flood prediction model is a piece of the larger CAMEO project, which has secured €9 million in funding led by University College Dublin. CAMEO's wider goal is to boost an Earth Observations (EO) services sector within Ireland, investigating the role of EO data in climate, agriculture, and marine studies.

Combatting Climate Change-Induced Flooding

Recent significant flooding in Ireland, following storms such as Ciarán, Debbie, and Babet, has highlighted the growing challenge. Climate change adds a concerning dimension to this issue, with forecasts predicting an upsurge in intense winter rainfall and unprecedented flooding in both historically vulnerable and new regions.

Economic Impact and Damage Control

The Irish Fiscal Advisory Council has sounded alarms on the financial toll of extreme flooding events, which could reach €500 million annually by decade's end. Dr. Omid Memarian Sorkhabi, the post-doctoral researcher spearheading the model's development at CeADAR, stands at the frontline of this innovative defense. He successfully monitored Middleton's flooding in real-time during Storm Babet, demonstrating the model's potential to reduce future damage through improved prediction accuracy.

Potential and Accuracy of the Early Warning System

Dr. Oisín Boydell, Director of Applied Research at CeADAR, emphasized the project's significance for flood-prone communities. Unlike traditional flood predictions, this data-driven approach promises unparalleled precision. The accessibility of high-resolution data affords authorities crucial lead time for mitigative actions, potentially safeguarding lives, property, and the economy against future flooding catastrophes.

AI, flooding, satellite