Neglecting the temporal variations in population distribution can lead to significant discrepancies in exposure estimations for disaster management, especially in the face of increasing natural hazards due to climate change. Effective disaster management necessitates a nuanced understanding of how the urban environment influences the temporal variations in population distribution. This study addresses this knowledge gap by investigating the relationship between the spatial patterns of urban elements and daytime-nighttime population differences across eight European cities. The study reveals a substantial association between urban form indicators and daytime-nighttime population differences. Although the findings suggest that there is no one-size-fits-all set of indicators for different cities, ‘closeness centrality’, which measures the accessibility of a specific location within the overall street network, is identified as a key proxy for daytime-nighttime population differences across all cities analysed, which can be further linked to the accuracy of hazard exposure estimation. These findings can contribute to enhancing urban resilience by offering insights into spatio-temporal population dynamics and considering their implications for disaster management.