Abstract
A significant amount of hurricane losses are caused by rainfall-induced freshwater flooding, and hurricane precipitation is expected to increase in the warming climate. However, the climate change impact on the hurricane rainfall-induced flood risk has not been investigated well yet, which leads to the potential increase in the flood risk in the future being overlooked. Investigation of future hurricane rainfall-induced flood risk can be very time-consuming because of the high resolution of the models for climate-dependent hazard simulation and regional loss assessment. Therefore, this study developed a statistical model for future hurricane freshwater flood loss assessment considering climate change scenarios by utilizing artificial neural networks (ANNs) and gradient boosting. Future hurricane freshwater flood risks to residential buildings in the southeastern US coastal states were investigated with the model. It was found that the hurricane rainfall-induced freshwater flood loss can increase by 16% and 30% from 2020 to 2030 and from 2090 to 2100, respectively. The investigation suggests the need to consider climate change impact on future flood risk management practices.
Original language | English (US) |
---|---|
Article number | 04020061 |
Journal | Natural Hazards Review |
Volume | 22 |
Issue number | 2 |
DOIs | |
State | Published - May 1 2021 |
Keywords
- Artificial neural network
- Climate change
- Freshwater flooding
- Hurricane
- Residential buildings
- Risk assessment
ASJC Scopus subject areas
- Civil and Structural Engineering
- Environmental Science(all)
- Social Sciences(all)