Climate Justice Implications of Natech Disasters: Excess Contaminant Releases during Hurricanes on the Texas Gulf Coast

• Published in: Environmental science & technology 2024-08, Vol.58 (32), p.14180-14192
• Main Authors: Berberian, Alique G., Morello-Frosch, Rachel, Karasaki, Seigi, Cushing, Lara J.
• Format: Article
• Language: English
• Subjects: Climate Change; Climate justice; Contaminants; Cyclonic Storms; Disasters; Emissions; Energy and Climate; Extreme weather; Humans; Hurricanes; Land pollution; Low income areas; Low income groups; Neighborhoods; Petrochemicals; Population density; Poverty; Refineries; Regression analysis; Regression models; Residential areas; Rural populations; Texas
• ISSN: 0013-936X; 1520-5851; 1520-5851
• DOI: 10.1021/acs.est.3c10797

Extreme weather events are becoming more severe due to climate change, increasing the risk of contaminant releases from hazardous sites disproportionately located in low-income communities of color. We evaluated contaminant releases during Hurricanes Rita, Ike, and Harvey in Texas and used regression models to estimate associations between neighborhood racial/ethnic composition and residential proximity to hurricane-related contaminant releases. Two-to-three times as many excess releases were reported during hurricanes compared to business-as-usual periods. Petrochemical manufacturing and refineries were responsible for most air emissions events. Multivariable models revealed sociodemographic disparities in likelihood of releases; compared to neighborhoods near regulated facilities without a release, a one-percent increase in Hispanic residents was associated with a 5 and 10% increase in the likelihood of an air emissions event downwind and within 2 km during Hurricanes Rita and Ike (odds ratio and 95% credible interval= 1.05 [1.00, 1.13], combined model) and Harvey (1.10 [1.00, 1.23]), respectively. Higher percentages of renters (1.07 [1.03, 1.11], combined Rita and Ike model) and rates of poverty (1.06 [1.01, 1.12], Harvey model) were associated with a higher likelihood of a release to land or water, while the percentage of Black residents (0.94 [0.89, 1.00], Harvey model) was associated with a slightly lower likelihood. Population density was consistently associated with a decreased likelihood of a contaminant release to air, land, or water. Our findings highlight social inequalities in the risks posed by natural–technological disasters that disproportionately impact Hispanic, renter, low-income, and rural populations.