Development of an Optimized Hurricane Storm Surge-Wave Model

Active Research Projects

Development of an Optimized Hurricane Storm Surge–Wave Model for the Northern Gulf of Mexico for use with the ADCIRC Surge Guidance System

Project Sponsors: Department of Homeland Security Coastal Resilience Center

PI: Scott C. Hagen, LSU CCE/CCT, Director LSU CCR

Co-PI: Stephen C. Medeiros, University of Central Florida

Collaborating Partners: Matthew V. Bilskie, LSU CCR

Project Summary This project will advance state-of-the-art model development by introducing novel terrain analysis techniques and lidar-based surface roughness parameterization at the regional scale. These advanced techniques will also be used to develop intelligent, stable, and semi-automated mesh de-refinement methods for optimizing a research grade (i.e., high resolution) storm surge model to reduce computational time to the point where it can be run within reasonable real-time forecast time frames (e.g., ~1-2 hrs). We will use a protocol based on emphasizing hydraulically significant embankment or valley features to optimize
a research grade model of the MS, AL, and FL Panhandle. Since the purpose of ASGS is the provision of real-time hazard guidance, we will emphasize the accurate capture of the timing and magnitude maximum water levels. This will be achieved by employing mesh
development techniques such as: running preliminary simulations to define active floodplain and removing unnecessary elements (relevant because the research grade model was developed to accommodate up to two meters of sea level rise); employing accelerated
element relaxation moving outward from significant vertical features; and enforcing stricter criteria for vertical feature inclusion (especially for channels). Objective error metrics will be used to assess model performance. The final outcome/deliverable will be
an accurate, optimized hurricane storm surge model of the northern Gulf of Mexico (MS, AL, & FL Panhandle) that is suitable for use with the ASGS including improved surface roughness parameterization from our lidar-based technique. In addition, this high resolution
ADCIRC+SWAN model will serve as a benchmark for validating future versions that may incorporate less resolution or smaller regional focus. 

Click here for additional information.