High Resolution Atmospheric Models Key to Storm Surge Predictions

For millions of people living and working in coastal zones, hurricanes and their storm surges are costly and deadly. Simulating storm surges is challenging. Surges are sensitive to a number of factors, including atmospheric wind speeds and pressures. Scientists found that accurate models of storm surges could be built by taking advantage of high-resolution regional simulations of tropical hurricanes.

Using high-resolution regional atmospheric models to improve tropical cyclone simulations may in turn yield more skillful predictions of storm surges in the ocean below. The result? The findings could help better predict and mitigate flood risks in low-lying coastal regions.

Simulating tropical cyclone winds is challenging because they are sensitive to moist atmospheric processes such as cloud microphysics and convection, which are notoriously difficult to capture using models. In a recent study, researchers evaluated the uncertainty in simulating hurricane winds and pressures using high-resolution regional models that better represent these processes. To gain insights on modeling uncertainty, they evaluated an ensemble of regional simulations with different representations of clouds and convection against observed atmospheric data for Hurricane Katrina. They also used simulated winds and pressures from the high-resolution atmospheric model to predict Hurricane Katrina’s storm surge in the northern Gulf of Mexico. They evaluated the storm surge simulations using high-water marks collected by the Federal Emergency Management Agency along the Alabama, Mississippi, and Louisiana coasts. Results showed that regional simulations of Hurricane Katrina are sensitive to the atmospheric model’s approximations of both convection and cloud microphysical processes. These processes are linked to hurricane development and intensification. When the storm surge model was driven by the most skillful result from the ensemble of atmospheric model simulations, storm surge error statistics were comparable to simulations driven by observed winds and pressure. This finding demonstrates the value of simulating storm surge and inundation using high-resolution regional simulations of hurricane winds and pressure. The results could also help improve understanding of coastal flooding vulnerabilities and risks.

This study was funded by the U.S. Department of Energy, Office of Science, Office of Biological and Environmental Research as part of the Integrated Assessment Research program.