We study hurricane risk on the U.S. Gulf Coast during 1950‐2005, estimating the wind damage and storm surge from every hurricane in this extended period. Wind damage is estimated from the known path and wind speeds of individual storms and calibrated to fit actual damage reports for a sample of Gulf Coast storms. Storm surge is estimated using the model developed by . These models provide the first comprehensive overview of the hurricane storm hazard as it has been experienced over a fifty‐six‐year period. We link the estimated damage with information on the population and specific socio‐demographic components of the population (by age, race, and poverty status). Results show that white, young adult, and nonpoor populations have shifted over time away from zones with higher risk of wind damage, while more vulnerable population groups–the elderly, African Americans, and poor—have moved in the opposite direction. All groups have moved away from areas with high risk of storm surge since 1970. But in this case, perhaps because living near the water is still perceived as an amenity, those at highest risk are whites, elderly, and nonpoor households. Here exposure represents a trade‐off between the risk and the amenity.

我们研究了1950-2005年美国墨西哥湾沿岸的飓风风险，估算了在此较长时期内每种飓风造成的风灾和风暴潮。根据单个风暴的已知路径和风速估算风害，并根据墨西哥湾沿岸风暴样本的实际损坏报告进行校准。风暴潮的估算是使用所开发的模型。这些模型提供了对飓风风暴危害的第一个全面概述，因为它已经经历了56年。我们将估计的破坏与人口信息以及人口的特定社会人口统计学组成部分（按年龄，种族和贫困状况）联系起来。结果表明，随着年龄的增长，白人，年轻人和非贫困人口逐渐远离风灾风险较高的地区，而脆弱人群（老年人，非裔美国人和穷人朝相反的方向发展。自1970年以来，所有群体都远离了发生风暴潮的高风险地区。但是，在这种情况下，也许是因为仍然认为生活在水边仍然是一种便利，所以高风险人群是白人，老年人和贫困家庭。这里的暴露代表了风险与便利之间的权衡。