Major cities worldwide are increasingly experiencing intense heatwaves, which disproportionately affect vulnerable populations. This suggests the urgent need for focused attention and mitigation strategies to protect those most at risk from the impacts of heatwaves. The presence of vulnerable groups in buildings with varying thermal conditions can result in different outcomes during extreme heat events, making it crucial to evaluate the thermal resilience of buildings in conjunction with social vulnerability. This research advances the understanding of thermal vulnerability by integrating building thermal characteristics with social vulnerability. While previous studies have predominantly focused on socio-ethnic and economic dimensions of vulnerability, this study specifically investigates the relationship between social vulnerability and thermal resilience, with a focus on building thermal parameters. Datasets from Zillow and ResStock are used to capture the thermal properties of buildings, while social vulnerability data is obtained from the Center for Disease Control and Prevention. A Gaussian Naïve Bayes (GNB) model is applied to enhance building envelope attributes from Zillow, leveraging detailed information from ResStock data. Additionally, a Kolmogorov-Smirnov (KS) test is conducted to analyze diverse thermal attributes, such as age of buildings, room size, number of stories, building area, envelope characteristics, systems, infiltration, across differing levels of social vulnerability (high and low). Specifically, Philadelphia is selected as the analysis focus. The analysis reveals significant associations between thermal resilience and social vulnerability. Groups with higher social vulnerability are more likely to be exposed to elevated levels of thermal vulnerability compared to less vulnerable groups. This highlights the necessity for tailored interventions aimed at promoting the development of more resilient and equitable buildings, capable of reducing the impacts of heatwaves on vulnerable communities.
Read full abstract