Neural Network Predictions of Peak Storm Tides due to Tropical Cyclones

Storm-driven flooding is a hazard for coastal communities. Process-based models can predict the combined effects of tides, winds, and flooding due to tropical cyclones, including in real-time, but often with restrictions due to a model’s runtime. Researchers have developed neural networks (NN), trained on libraries of storm surge simulations, to predict flooding in seconds. However, previous NNs ignored interactions with astronomical tides, limited to storms of specific durations, and trained for extreme conditions.

In this study, a NN is developed to predict peak values for storm tides (storm surge and tides) at nine stations along the North Carolina coast. For training, a library of storm-tides was developed via process-based model simulations of 1,813 synthetic storms based on historical data in the north Atlantic Ocean, but with a specific focus on North Carolina, and then augmented by a factor of 50 via combinations with random tides. Unlike previous NN, this approach incorporates the astronomical tides in the training and uses data augmentation techniques for enhanced generalization. The NN performs well, with root-mean-square errors of about 6 cm and mean bias errors for the extreme storms of about 5 cm. For probabilistic predictions of historical storms, the model can predict for 100 ensemble members in 1 sec, and the ranges of peak storm tides are close to their true values.

TA Cuevas López, BJ Tucker, JC Dietrich, DL Anderson, E Lobaton, JS Mariegaard (2025). “Neural Network Predictions of Peak Storm Tides due to Tropical Cyclones.” Ocean Modelling, available online, DOI: 10.1016/j.ocemod.2025.102588.

Brandon wins Outstanding Senior Award for Scholarly Achievement

Undergraduate student Brandon Tucker won an Outstanding Senior Award for Scholarly Achievement, which recognizes exceptional academic performance including participation in undergraduate research. Brandon was among four outstanding seniors recognized by our department.

[H]he continued undergraduate research as part of the Coastal and Computational Hydraulics Team with Associate Professor Casey Dietrich and graduate student Tomás [Cuevas] López. The project was related to predictions of coastal flooding due to hurricanes.

“I helped run more than 1 million CPU hours of hurricane models to train our machine-learning model called Concorde, which can predict storm-driven flooding in seconds,” Tucker explained. “I also created detailed Python examples for Kalpana, an intermediary model used by researchers across the country.”

“This was a lot of work,” Dietrich emphasized. “Each hurricane simulation can take several hours on a parallel computing cluster and generate gigabytes of data, and so it took about two months to complete the simulations. It would have taken much longer without Brandon’s help and creativity. He wrote scripts to automate the process to submit, monitor, and archive the simulations, and he contributed to a post-processing visualization script. His documentation and examples are now shared widely with all users of the software. Brandon is strong at the technical skills of computing and programming, but he also sees the larger picture and looks for ways to contribute.”

After graduation, Brandon will pursue a Master of Civil Engineering at CCEE with a focus on transportation systems, while also completing a Graduate Certification in City Design from the College of Design.

Congratulations to Brandon!

Conferences: Fall 2023

Conference: USNCCM 17

Conference: ADCIRC Users Meeting 2023