ERI: Advanced Forecasting of Coastal Flooding and Extreme Rainfall

Coastal communities across the southeastern United States face increasing risks from flooding, heavy rainfall, and rising sea levels. These hazards threaten public safety, transportation systems, critical infrastructure, economic activity, and natural resources. South Carolina is particularly vulnerable because of its low-lying coastline, rapidly growing population, and dependence on coastal industries. To protect infrastructure and safety, communities need more reliable information to prepare for and respond to extreme weather events. This Engineering Research Initiation (ERI) award supports research that will improve the ability to anticipate periods of unusually heavy rainfall and elevated sea levels by identifying the large-scale patterns that influence these events and incorporating that knowledge into advanced forecasting tools. The resulting forecasting tools can support emergency managers, planners, public officials, and coastal communities in preparing for hazardous weather events and strengthening climate resilience. By improving climate forecasting and supporting evidence-based planning, the project serves the national interest by helping protect lives, property, infrastructure, and economic resources in vulnerable coastal regions. The project will also strengthen the future workforce by providing research, training, and educational opportunities for undergraduate students in science, technology, engineering, and mathematics. This project will investigate relationships between large-scale oceanic and atmospheric variability and regional extreme precipitation and sea-level variability in South Carolina. Historical climate observations will be analyzed to identify the dominant atmospheric and oceanic patterns associated with hydroclimatic variability and to quantify the climate processes that influence regional predictability. These insights will be incorporated into a hybrid forecasting framework that combines physics-informed machine learning, deep learning, and hybrid quantum-classical machine learning models. Forecast performance will be evaluated using established forecasting skill metrics and benchmarked against conventional statistical approaches. The project is expected to advance understanding of climate teleconnections, subseasonal predictability, and computational methods for hydroclimatic forecasting. The project will produce forecasting models, climate prediction maps, open datasets, software resources, peer-reviewed publications, and educational materials that support climate adaptation and resilience planning across the southeastern United States. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.