Project Summary
This research aims to improve the accuracy of hurricane intensity and flood predictions by refining atmospheric models to better represent turbulence diffusion. Led by Ph.D. candidate Md Murad Hossain Khondaker and assistant professor Mostafa Momen at the University of Houston, the study addresses a key issue in current weather models: the overestimation of kinetic energy diffusion, which leads to weaker intensity forecasts and underestimated hurricane impacts.
Methods
Using high-performance computing resources—including the Bridges-2 supercomputer at the Pittsburgh Supercomputing Center—researchers ran large-scale simulations to test their hypothesis. By adjusting turbulence parameters, they achieved up to a 40% improvement in hurricane intensity forecasts and a 34% enhancement in flood predictions. Their findings also revealed that more intense hurricanes do not necessarily produce more rainfall overall but instead generate more localized and extreme precipitation, increasing flood risks in urban areas.
Goals
The study, published in the Journal of Hydrometeorology, underscores the importance of refining hurricane models to enhance disaster preparedness. Future research will focus on improving the prediction of storm surge and compound flooding, as well as developing advanced turbulence diffusion adjustments to further enhance forecasting accuracy and mitigate hurricane-related risks.