[论文解读] Equitable Community Resilience: The Case of Winter Storm Uri in Texas

Ali Nejat, Laura Solitare+2 more|arXiv (Cornell University)|Jan 17, 2022

Disaster Management and Resilience被引用 10

一句话总结

本研究利用县一级停电数据和通过计算机视觉分析的卫星图像，调查了德克萨斯州冬季风暴“艾瑞”后电网韧性的公平性问题，揭示了种族与收入差异带来的影响。研究发现，非西班牙裔白人人口比例和中位收入与停电比例呈显著负相关，而语言隔离和依赖公共交通的人口比例则与停电比例呈正相关，建议通过电网现代化和分布式可再生能源实现更公平的韧性提升。

ABSTRACT

Community resilience in the face of natural hazards relies on a community's potential to bounce back. A failure to integrate equity into resilience considerations results in unequal recovery and disproportionate impacts on vulnerable populations, which has long been a concern in the United States. This research investigated aspects of equity related to community resilience in the aftermath of Winter Storm Uri in Texas which led to extended power outages for more than 4 million households. County level outage and recovery data was analyzed to explore potential significant links between various county attributes and their share of the outages during the recovery and restoration phases. Next, satellite imagery was used to examine data at a much higher geographical resolution focusing on census tracts in the city of Houston. The goal was to use computer vision to extract the extent of outages within census tracts and investigate their linkages to census tracts attributes. Results from various statistical procedures revealed statistically significant negative associations between counties' percentage of non-Hispanic whites and median household income with the ratio of outages. Additionally, at census tract level, variables including percentages of linguistically isolated population and public transport users exhibited positive associations with the group of census tracts that were affected by the outage as detected by computer vision analysis. Informed by these results, engineering solutions such as the applicability of grid modernization technologies, together with distributed and renewable energy resources, when controlled for the region's topographical characteristics, are proposed to enhance equitable power grid resiliency in the face of natural hazards.

研究动机与目标

• 考察社会与人口因素如何影响德克萨斯州冬季风暴‘艾瑞’期间的停电分布与恢复情况。
• 通过分析县一级和普查区一级的停电模式，识别电网韧性中的系统性不公平现象。
• 利用高分辨率卫星图像评估社会经济与语言脆弱性在长期停电中的作用。
• 提出工程解决方案（如电网现代化和分布式可再生能源），以增强公平性韧性。
• 将地形和社区层面的属性整合到未来极端天气事件的韧性规划中。

提出的方法

• 分析冬季风暴‘艾瑞’期间的县一级停电与恢复数据，评估其与人口统计与社会经济变量的相关性。
• 应用计算机视觉技术处理夜间卫星光影像，估算休斯顿市普查区级别的停电范围。
• 使用统计模型测试普查区属性（如种族、收入、语言隔离）与停电严重程度之间的关联。
• 采用多元回归分析，在控制区域地形因素的前提下，分离特定人口因素对停电比例的影响。
• 通过不同社区类型停电持续性的时空分析验证研究发现。
• 提出工程干预措施，如分布式能源资源和电网现代化，其设计充分考虑区域地形限制。

实验结果

研究问题

• RQ1 在冬季风暴‘艾瑞’期间，人口统计与社会经济特征如何与停电的持续时间和范围相关？
• RQ2 语言隔离人口比例较高的普查区在多大程度上经历了更长的停电时间？
• RQ3 中位家庭收入与受影响县停电比例之间存在何种关系？
• RQ4 非西班牙裔白人人口比例与德克萨斯州各县停电韧性之间有何关联？
• RQ5 可设计哪些工程解决方案以在极端天气条件下提升电网的公平性韧性？

主要发现

• 在恢复阶段，非西班牙裔白人居民比例较高且中位家庭收入较高的县，停电比例显著较低。
• 在普查区层面，语言隔离人口比例较高的区域，其停电严重程度经计算机视觉检测后显示为正相关。
• 公共交通依赖人口比例较高的普查区也表现出与长期停电的正相关性。
• 统计模型显示，种族与收入等人口因素是停电韧性的显著预测因子，其影响超越了地理或地形变量。
• 基于卫星图像推导的停电模式证实，弱势社区在风暴过后受到不成比例的影响。
• 本研究支持将分布式和可再生能源资源作为提升脆弱社区公平性电网韧性的关键组成部分。