Process-Based and Data-Based Storm Surge Models for Rhode Island Coastal Flooding within the STORMTOOLS Framework
Here, we present two approaches for storm surge forecasting in coastal areas of Rhode Island: A regional ADCIRC model, and artificial intelligence (AI). Recent numerical modeling results published by north atlantic coast comprehensive study (NACSS) were employed as the basis. Using a downscaling app...
| Published in: | Coastal Structures and Solutions to Coastal Disasters 2015 |
|---|---|
| Main Authors: | , |
| Format: | Text |
| Language: | unknown |
| Published: |
DigitalCommons@URI
2015
|
| Subjects: | |
| Online Access: | https://digitalcommons.uri.edu/oce_facpubs/235 https://doi.org/10.1061/9780784480304.028 |
| Summary: | Here, we present two approaches for storm surge forecasting in coastal areas of Rhode Island: A regional ADCIRC model, and artificial intelligence (AI). Recent numerical modeling results published by north atlantic coast comprehensive study (NACSS) were employed as the basis. Using a downscaling approach, a high resolution ADCIRC hydrodynamic model was developed and interfaced with the NACCS model along the open boundaries. Although this model could effectively predict storm surges for the past historical/synthetic storms, it was numerically very expensive to provide boundary information for any storm surge forecasting scenario. To address this issue, an efficient AI data-based model was developed. The AI model predicts the storm surge using tropical storm parameters (i.e. central pressure, radius to maximum winds, forward velocity, and storm track). The AI model was validated using a set of randomly selected synthetic storms as well as real extreme storms in this region, and the performance was found satisfactory. |
|---|