A tale of two coasts: Unveiling U.S. Gulf and Atlantic coastal cities at high flood risk

Flood impacts in coastal cities can be mitigated through proactive risk
management, which requires comprehensive risk assessment. This study
develops a data-driven risk assessment framework to identify high-risk
coastal cities, estimate exposed populations and their infrastructures,
and reveal underlying flood risk factors under dual scenarios—General
Flood Damage (GFD) and Extreme Flood Damage (EFD)—along the U.S. Gulf and
Atlantic Coasts (USGAC). Using historical flood damage data and 16 factors
representing hazard, exposure, and vulnerability, three machine learning
models—Support Vector Machine, Random Forest, and Multi-Layer
Perceptron—are adopted to assess flood risk. Results indicate that under
GFD, 1.14 % of the area with 16.67 % of the populations are at very high
risk, while under EFD, 0.53 % of the area with 4.08 % of the population
faces very high risk. Eight coastal cities are identified as high-risk.
New York City has the largest population at risk (GFD: 4.75 M, EFD: 4.40
M), while New Orleans has the highest relative exposure (~99 % under
both). Low elevation is the most influential factor for GFD, and high
drainage density is the dominant factor for EFD. This scalable framework
offers actionable insights for policymakers to reduce flood risk.