To help Texas emergency managers develop better storm simulations, the National Science Foundation donated $3 million to the University’s Institute for Computational Engineering and Sciences.
Clint Dawson, head of the ICES Computational Hydraulics Group, started research about the storm surge problem in the late ’90s. He said the grant is for a joint project between UT, Louisiana State University, The University of Notre Dame and the University of North Carolina-Chapel Hill.
He said the project started with the University of Notre Dame, which developed a storm surge model based on the ADCIRC system for southern Louisiana. According to Dawson, the ADCIRC system is a computer model for simulating how hurricanes cause water to flood coastal regions.
“The model is used as a forecasting tool for emergency managers as storms approach land to help with evacuations,” Dawson said. “One of the goals of this grant is to make the code more computationally efficient and robust as a storm surge forecasting tool.”
Dawson said researchers have extensively used the model to study past hurricanes and to forecast current hurricanes as they approach land. He said the application would be the same with the grant, but the focus is to extend the code in several ways by developing new algorithms and implementing new computer science tools, which will help the code adapt to new computer architectures.
“The main focus is to keep the code viable into the future,” Dawson said.
The grant will help form STORM, a new version of ADCIRC that’s designed to perform more efficiently across a variety of computer hardware architectures, according to Dawson. He said one of the goals for STORM is to work twice as fast as ADCIRC, enabling storm predictions to be made within an hour of receiving data inputs.
Dawson said turning ADCIRC into STORM will help in the understanding of the history and composition of the original code, which could aid in creating other programs in general.
“We have to study how to mitigate storm surge and how to effectively evacuate during a storm,” Dawson said. “These problems are too complicated to study only in a laboratory setting, and computer stimulations must be used. Hopefully, this grant will make a difference to future stimulation capabilities for studying these problems.”