The fifth of the neighborhood blocks in the continental United States most vulnerable to natural disasters account for a quarter of the risk in the lower 48 states, according to a detailed vulnerability assessment.
Leaders in data-driven risk modeling, researchers at The University of Alabama have applied advanced data analysis and machine learning to more than 100 factors that influence vulnerability to natural hazards for about 11 million people. blocks by the United States Census Bureau, finding significant differences may exist between neighboring blocks.
The results were published in the journal Communication in Nature is the first to map the exposure of the continental US to impacts from natural hazards at the block level, overcoming the limitations of previous inference efforts and assessing the vulnerability of larger scales to mask local variability – again.
The detailed data, called the Social, Economic, Infrastructure Vulnerability Index, has the potential to play a key role in effective risk management and will provide important information for improved risk-informed decision-making.
“The SEIV Index looks for communities and neighborhoods that have an immediate need to consider the implementation of risk reduction measures and bottom-up methods for adaptation,” said Dr. Hamid Moradkhani, the Alton N. Scott professor of Civil and Environmental Engineering and director of the UA Center for Complex Hydrosystems Research. “Such a good resolution assessment helps determine the relative urgency of action and resource allocation for improved stability.”
With natural hazards becoming more frequent and severe, the UA team of researchers wanted to conduct a comprehensive vulnerability assessment across the country that takes into account a wide range of factors that influence risk. Researchers also want to reduce the subjectivity of the vulnerability assessment.
In conducting an analysis of reported property damage costs, UA researchers used machine learning to evaluate variables that contribute to an area’s vulnerability to natural disasters. The allocation at the block level combines socio-economic factors with infrastructural characteristics of communities such as distance to the nearest emergency facilities and number of buildings. The approach overcomes extensive, national-level efforts to understand similar problems, but misses the unequal vulnerability between neighboring blocs in the same disaster.
“Natural hazards disproportionately affect communities around the Conterminous United States,” Moradkhani said.
Among the top 10 states with the highest percentage of blocks with high vulnerability to natural disasters are the two states in the top 10 of the largest gross domestic product, a measure of the economic output of a state. Minnesota with 82% of its Census blocks having high or very high SEIV indices, second highest among the 48 states in the study, and Ohio with 76% of its Census blocks with very weak, third highest, both in the top 10 for state GDP. .
These two examples show that aggregate information can mask inequality and mask local vulnerability to natural disasters, Moradkhani said.
“Our main motivation is to create a new vulnerability index that covers the entire continental US and provides high-resolution information at the block level with consideration of socio-economic and infrastructure vulnerability metrics,” he said. “It provides an objective approach to inform the public, emergency managers and stakeholders to develop adaptive capacities and prepare more resilient communities in their jurisdictions.”
Farnaz Yarveysi et al, Block-level vulnerability assessment reveals disproportionate impacts of natural hazards across the United States, Communication in Nature (2023). DOI: 10.1038/s41467-023-39853-z
Awarded by the University of Alabama at Tuscaloosa
Citation: Vulnerability to natural hazards shows uneven risk across US: New data analysis (2023, July 18) Retrieved 18 July 2023 from https://phys.org/news/2023-07- natural-hazard-vulnerability-disproportionate-analysis.html
This document is subject to copyright. Except for any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. Content is provided for informational purposes only.