Assessing heat vulnerability in Philadelphia using geographically weighted principal component analysis (GWPCA): A geospatial big data-driven approach

Abstract

The impact of climate change, specifically more intense heat waves, has increased concerns about heat vulnerability, particularly among high-risk populations. This research utilizes multi-source geospatial big data and employs Geographically Weighted Principal Component Analysis (GWPCA) as well as Global Principal Component Analysis (GPCA) to analyze heat vulnerability in Philadelphia. Using GPCA, four key components are identified Sensitivity, Adaptive Capacity, proxy for Sensitivity, and Exposure, respectively. The subsequent GWPCA analysis reveals localized vulnerability differences, showing distinct patterns across the city. Notably, Sensitivity factors are prominent in the western and southwestern regions, whereas Exposure is dominant in the central and southern parts. This study underscores the significance of considering spatial heterogeneity when assessing heat vulnerability. It also highlights the potential of GWPCA to capture subtle disparities within specific areas and proposes targeted strategies to reduce heat vulnerability in affected communities. Therefore, the incorporation of an advanced spatial analysis model enables a comprehensive understanding of heat vulnerability in complex urban environments. This progress is crucial in enhancing resilience and adaptation to evolving climate conditions.

Keywords

• Heat vulnerability
• PCA
• GWPCA
• Big data
• Spatial heterogeneity