Specifying Evacuation Return and Home-switch Stability During Short-term Disaster Recovery Using Location-based Data

TL;DR

This study uses location data to analyze evacuation return and home-switch stability during disaster recovery, revealing disparities among subpopulations and complex recovery patterns to inform better disaster management.

Contribution

It introduces a data-driven approach to specify and analyze critical recovery milestones and disparities using privacy-preserving location data during Hurricane Harvey.

Findings

Shorter evacuation return in flooded areas may indicate barriers to evacuation.

Lower-income residents show less stable home-switch patterns.

Recovery patterns follow a two-phase return process.

Abstract

The objectives of this study are: (1) to specify evacuation return and home-switch stability as two critical milestones of short-term recovery during and in the aftermath of disasters; and (2) to understand the disparities among subpopulations in the duration of these critical recovery milestones. Using privacy-preserving fine-resolution location-based data, we examine evacuation return and home move-out rates in Harris County, Texas in the context of the 2017 Hurricane Harvey. For each of the two critical recovery milestones, the results reveal the areas with short- and long-return durations and enable evaluating disparities in evacuation return and home-switch stability patterns. In fact, a shorter duration of critical recovery milestone indicators in flooded areas is not necessarily a positive indication. Shorter evacuation return could be due to barriers to evacuation and shorter home move-out rate return for lower-income residents is associated with living in rental homes. In addition, skewed and non-uniform recovery patterns for both the evacuation return and home-switch stability were observed in all subpopulation groups. All return patterns show a two-phase return progress pattern. The findings could inform disaster managers and public officials to perform recovery monitoring and resource allocation in a more proactive, data-driven, and equitable manner.