P2.50
Large-scale scope of damage prediction model for landfalling hurricanes
PAPER WITHDRAWN
Joseph Spain, ImpactWeather, Inc., Houston, TX
When a hurricane threat materializes, many organizations require information about the potential scope of damage that could result, in monetary terms. Such information would enable stakeholders, such as insurers and disaster response officials, to fit response personnel and resources to the public and private sector needs that emerge during the aftermath. In response to this need, the development of a scope of damage prediction model to forecast a range of likely damage, in terms of U.S. dollar losses, is progressing. The model's methodology employs a multiple polynomial regression technique using four explanatory variables, including the size and intensity components of the Hurricane Severity Index (HSI). HSI is an enhanced hurricane rating system that defines the strength and destructive capability of hurricanes. The two additional explanatory variables are maximum inundation associated with the storm surge and vulnerability (exposure of wealth). The vulnerability parameter is a proxy for wealth exposure, derived from population and per capita income at the county/parish level. The scope of damage prediction model leverages the relationships that exist between the explanatory variables and the predictand (normalized damage) to forecast future damage. The output from the model would enable a variety of users, such as insurers and other disaster response agencies, to manage a realistic volume of agents, personnel, and resources to cope with the expected aftermath of a hurricane during the days before landfall. Consequently, insurers and disaster officials could improve their efficiency and cost-effectiveness when preparing to meet the needs of policyholders and citizens.
Poster Session 2, Posters: Tropical Cyclone Modeling, Convection, Tropical Cyclone Structure, Intraseasonal Variability, T-PARC, TCS-08, Air-Sea Interaction, Convectively Coupled Waves, Tropical Cyclone Observations, Climate Change, Probabilistic Forecasting
Thursday, 13 May 2010, 3:30 PM-5:00 PM, Arizona Ballroom 7
Previous paper Next paper