Refining Flood Risk Predictions in Hawaiʻi with Generative Machine Learning
Project Overview
As climate change intensifies, accurately forecasting regional flood, drought, and heat risks is critical for planning, but current global climate models lack the resolution needed for useful local predictions. Researchers supported by this Pacific Islands CASC project will use generative artificial intelligence (AI) to create high-resolution precipitation maps and flood risk estimates for the Hawaiian Islands, equipping local decision-makers with better tools to plan for climate change impacts.
Project Summary
As climate change increasingly affects ecosystems, economies, and communities worldwide, effective mitigation and adaptation strategies are more critical than ever. A key element in making these efforts successful is the ability to accurately forecast the future impacts of climate change. While global climate models (GCMs) offer climate projections, their coarse spatial resolution does not allow regional characteristics to be captured accurately, which are details crucial for managing risks related to climate hazards like floods, droughts, and heatwaves.
Dynamic downscaling, which uses regional climate models, offers more detailed projections but is resource-intensive and cannot produce the ensemble (wide range) of projections that GCMs can, limiting its usefulness for risk assessment and uncertainty analysis. A more efficient alternative is statistical downscaling, which generates high-resolution climate projections with less computational effort. Recent breakthroughs in generative artificial intelligence (AI) with neural networks offer new possibilities for statistical downscaling. This project will use these generative AI advancements to create high-resolution (1km) daily and sub-daily precipitation maps for the Hawaiian Islands and use these maps to quantify the risk of flooding events.
By improving flood risk estimates, this project will provide local decision-makers with better tools to plan for the impacts of climate change.
- Source: USGS Sciencebase (id: 667da659d34e67892486527f)
Project Overview
As climate change intensifies, accurately forecasting regional flood, drought, and heat risks is critical for planning, but current global climate models lack the resolution needed for useful local predictions. Researchers supported by this Pacific Islands CASC project will use generative artificial intelligence (AI) to create high-resolution precipitation maps and flood risk estimates for the Hawaiian Islands, equipping local decision-makers with better tools to plan for climate change impacts.
Project Summary
As climate change increasingly affects ecosystems, economies, and communities worldwide, effective mitigation and adaptation strategies are more critical than ever. A key element in making these efforts successful is the ability to accurately forecast the future impacts of climate change. While global climate models (GCMs) offer climate projections, their coarse spatial resolution does not allow regional characteristics to be captured accurately, which are details crucial for managing risks related to climate hazards like floods, droughts, and heatwaves.
Dynamic downscaling, which uses regional climate models, offers more detailed projections but is resource-intensive and cannot produce the ensemble (wide range) of projections that GCMs can, limiting its usefulness for risk assessment and uncertainty analysis. A more efficient alternative is statistical downscaling, which generates high-resolution climate projections with less computational effort. Recent breakthroughs in generative artificial intelligence (AI) with neural networks offer new possibilities for statistical downscaling. This project will use these generative AI advancements to create high-resolution (1km) daily and sub-daily precipitation maps for the Hawaiian Islands and use these maps to quantify the risk of flooding events.
By improving flood risk estimates, this project will provide local decision-makers with better tools to plan for the impacts of climate change.
- Source: USGS Sciencebase (id: 667da659d34e67892486527f)