Ordnance Survey Collaborates with Snowflake to Develop AI-Powered Flood Readiness Model
Ordnance Survey (OS) has partnered with Snowflake to create an innovative AI-powered data model called Intelligent Flood Readiness. This model is designed to assess the impact of flooding and has revealed that 1.2 million people in England are at risk of falling outside of existing flood protection measures, with many of them residing in some of the most deprived areas of the country.
The primary purpose of this model is to assist policymakers in analyzing flood risk and remediation strategies with more detailed and granular data. By utilizing the model, policymakers can identify specific areas, such as period properties with basements, and apply appropriate protocols to mitigate flood risks. The model also enables policymakers to evaluate clusters of vulnerability that may span across different administrative boundaries, such as local authorities or flood risk management plan zones.
This advanced model combines OS’s building data with various government datasets and current flood risk management plans. By integrating six critical data streams into a shared “structural intelligence” layer, the model provides a comprehensive analysis of flood vulnerability.
Through the initial analysis, which involved combining OS’s building datasets with the Indices of Deprivation in England, the model was able to identify intersections between physical vulnerability (building height and type) and social risk. This information was then overlaid with Environment Agency flood data, as well as an AI-driven analysis of statutory flood risk management plan documents.
The model estimates that up to 68% of the identified buildings are highly vulnerable to flooding, particularly those located in deprived areas. Additionally, it was found that 84% of these vulnerable buildings were constructed before flood risk considerations were integrated into planning permissions after 2001.
The model identified Yorkshire and the Humber region as having some of the highest concentrations of vulnerable properties. Furthermore, the analysis revealed that a significant percentage of at-risk premises were built before 1959, emphasizing the dynamic nature of England’s natural and built environment.
Tim Chilton, managing geospatial consultant at OS, expressed excitement about the collaboration with Snowflake, highlighting how the model can help local authorities in flood management and planning. By leveraging geospatial data and Snowflake’s technology, the model provides valuable insights for prioritizing flood defences and resource allocation.
The data also indicated that 85% of at-risk buildings are vulnerable to surface water flooding, underscoring the need for investments in surface water infrastructure such as improved drainage systems.
Fawad Qureshi, global field chief technology officer at Snowflake, emphasized the importance of data in informed decision-making and how technology can facilitate collaboration and data-driven insights for better decision-making.
Overall, the collaboration between Ordnance Survey and Snowflake has resulted in a powerful tool that can help local governments shape policies, allocate resources effectively, and protect communities from the impacts of flooding.
