[Paper Review] Lifestyle Pattern Analysis Unveils Recovery Trajectories of Communities Impacted by Disasters

Natalie Coleman, Chenyue Liu+2 more|arXiv (Cornell University)|Jul 7, 2022

Disaster Management and Resilience4 citations

TL;DR

This study uses privacy-enhanced location intelligence data to analyze lifestyle recovery trajectories in Harris County, Texas, after Hurricane Harvey. By applying k-means and secondary clustering to home census block group visits to points of interest, it identifies four lifestyle patterns and three distinct recovery trajectories, revealing heterogeneous recovery rates even within similar demographic and lifestyle groups, with implications for disaster resilience planning and monitoring.

ABSTRACT

The return of normalcy to the population's lifestyle is a critical recovery milestone in the aftermath of disasters, and delayed lifestyle recovery could lead to significant well-being impacts. Lifestyle recovery captures the collective effects of population activities and the restoration of infrastructure and business services. This study uses a novel approach to leverage privacy-enhanced location intelligence data to characterize distinctive lifestyle patterns and to unveil recovery trajectories after a disaster in the context of 2017 Hurricane Harvey in Harris County, Texas. The analysis integrates multiple data sources to record the number of visits from home census block groups (CBGs) to different points of interest during the baseline period and disruptive period. First, primary clustering using k-means characterized four distinct essential and non-essential lifestyle patterns. Then, secondary clustering characterized the impact of the hurricane into three recovery trajectories based on the severity of maximum disruption and duration of recovery. The results reveal multiple recovery trajectories and durations within each lifestyle cluster, which imply differential recovery rates among similar lifestyle and demographic groups. The findings offer a twofold theoretical significance: (1) lifestyle recovery is a critical milestone that needs to be examined, quantified, and monitored in the aftermath of disasters; (2) the spatial structures of cities formed by human mobility and distribution of facilities and extends the spatial reach of flood impacts on population lifestyles. The analysis and findings also provide novel data-driven insights for public official and emergency managers to examine, measure, and monitor a critical milestone in community recovery trajectory based on the return of lifestyles to normalcy.

Motivation & Objective

• To examine lifestyle recovery as a critical yet undermeasured milestone in post-disaster community recovery.
• To quantify and monitor the return of population activities to normalcy after disasters using data-driven methods.
• To understand how spatial structures of cities and facility distribution influence the reach and duration of disaster impacts on lifestyles.
• To provide emergency managers and policymakers with actionable, data-informed insights into recovery progress based on lifestyle patterns.

Proposed method

• Utilizes privacy-enhanced location intelligence data from home census block groups (CBGs) to track visits to points of interest (POIs) during baseline and disruptive periods.
• Applies primary k-means clustering to classify four distinct lifestyle patterns based on visitation frequency to essential and non-essential POIs.
• Employs secondary clustering to categorize recovery trajectories based on maximum disruption severity and recovery duration.
• Integrates multiple data sources to map spatial and temporal dynamics of lifestyle recovery across different CBGs.
• Uses normalized visitation patterns to assess deviations from baseline behavior and recovery progression.
• Validates recovery trajectories by comparing clustering outcomes with observed disruption severity and demographic characteristics.

Experimental results

Research questions

• RQ1 How do distinct lifestyle patterns manifest in urban populations before and after a major disaster?
• RQ2 What are the dominant recovery trajectories in communities affected by Hurricane Harvey in Harris County, Texas?
• RQ3 To what extent do similar demographic and lifestyle groups exhibit divergent recovery rates post-disaster?
• RQ4 How do spatial structures of cities and facility distribution influence the geographic spread of lifestyle disruption?
• RQ5 Can lifestyle recovery be quantified and monitored as a reliable proxy for broader community resilience?

Key findings

• Four primary lifestyle patterns—essential, non-essential, mixed, and low-activity—were identified through k-means clustering of POI visitation data.
• Three distinct recovery trajectories emerged from secondary clustering: rapid, moderate, and prolonged, based on disruption severity and duration.
• Even within the same lifestyle cluster, communities exhibited significant variation in recovery speed and duration, indicating heterogeneous recovery dynamics.
• The study revealed that recovery trajectories are not uniformly distributed across demographic or spatial groups, challenging assumptions of homogenous recovery.
• The spatial structure of cities, particularly the distribution of facilities, extends the reach of flood impacts beyond physical inundation zones.
• Lifestyle recovery is a critical, measurable milestone that correlates with broader community well-being and should be monitored in disaster response and planning.