Abstract
Given the aftermaths of Hurricane Katrina in New Orleans and the surrounding parishes, the need to predict tropical storms and hurricanes has increased multifold. Accurate predictions may assist decision makers to implement appropriate evacuation plans necessary for the wellbeing of citizens. Coastal modelling is a key tool used to foresee hurricanes, storm surge and flooding. Of the many models implemented, ADCIRC, a project of the University of North Carolina, seems to hold much promise. Our objectives in this study were multifold. First, we ascertained the current limitations of the ADCIRC model. We achieved this goal through a variety of means including ADCIRC benchmarks on single and multiple clusters as well as a gprof profile of ADCIRC on a single cluster. Then, based on the information obtained from these efforts, we suggested appropriate improvements to optimise ADCIRC. With the achievement of these objectives, we hope ADCIRC will become a vital tool in developing accurate evacuation plans under duress that will aid in saving a countless number of lives.
| Original language | English |
|---|---|
| Pages (from-to) | 949-954 |
| Number of pages | 6 |
| Journal | Journal of Coastal Research |
| Issue number | SPEC. ISSUE 50 |
| State | Published - Dec 2007 |
Keywords
- ADCIRC
- Coastal modelling
- Tightly coupled applications
Fingerprint
Dive into the research topics of 'Towards a faster and improved ADCIRC (Advanced Multi-Dimensional CIRCulation) model'. Together they form a unique fingerprint.Cite this
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver